Skip to main content

Full text of "Cotton spinning"

See other formats

iiBWiiiinnniiwrf MiwwwiMHiiiiiiiiiiinj uii 


Established 1790. 

Telejrraphic Address : "DOBSONS, BOLTON." 

ABC and Lieber Codes. 

National Telephone No. 60 1. 




Roller and Saw Gins 
Hopper Bale Brea! 

Mixing Lattices. 
Hopper Feeders. 
Vertical and H< 

Carding- Engines. 
Improved Grinding M 
Improved Grinding R 
Stripping and Eu 

Sliver Lap Machines. 
Derby Doublers. 
Draw and Lap 


Machinery for \Ai 


Tools, Si 










Driving of 



Frames a 


Telegrams : 


Ask for 


Pamphlet : 

== "Tlie 
^ Transmission 
of Power 
by Ropes." 

New Edition 


Place your 
before us. 

Telephone : 



Wm. KENYON & Sons, Ltd. 

DUKINFIELD, Cheshire, England. 

Branches— LONDON OFFICE : Wm. Kenyon & Sons, Ltd., 

Cablerie du Nord, Armentieres, France. 96-97 Finsbury Pavement, E.G. 

Woodhouse Bros., Preston, Lanes., Eng. CANADIAN AGENTS : Dodge Manufacturing 

John Ruscoe & Co., Ltd., Hyde, Ches., Eng. Co. of Canada, Ltd., Toronto. 

U.S.A. AGENTS: Dodge Manufacturing Co. , Mishawaka, Indiana. 

Crotvn d>vo. lOs. net. 





Introduction. Chap. I. Cotton. II. Bales. III. Mixing, 
IV. Bale Breakers, etc. V. Hopper Feeder. VI. Openers 
and Scutchers. VII. Carding. VIII. Drawing Frames. IX. 
Comber and Preparing Machines. X. Bobbin and Flyer 
Frames: Speed F'rames. XI. Self-acting Mule. XII. Ring 
Frames. XIII. Testing. XIV. Miscellaneous. Index. 

The above contents will indicate the scope of the book. Each 
section is treated very fully in every practical detail. Very complete 
conditions are given for each process in regard to the most efficient 
and economical working, with hanks, speeds, drafts, settings, etc., 
for all ranges of counts and cottons. 

Many mills are working under conditions that are far from being 
economical or efficient. This may be due to careless supervision, 
or too strict an adherence to old methods, or a failure to appreciate 
the great importance of the essential details, or in many cases to a 
lack of a clear understanding of how to obtain the best and most 
out of each machine with a view to reducing labour, increasing pro- 
duction, eliminating machines and attaining a high standard in the 
ultimate yarn. 

The book deals with these practical factors in mill management 
and has been written solely for practical men. 

W" SCOTT TAGCART, m.i.mech.e. 

Consulting Engineer. 

22 Bridge Street, also East Bank, 

Manchester. Doffcocker, 

tele.: 3815 central, bolton. 






author of 








Highest Awards of Merit at International and other Exhibitions 
for Textile Machinery, from London, 1851, to Ghent, 1913. 

Codes : 
Telephone: No. 1826. Telegrams: ''PLATTS, OLDHAM." 

A. 1, A.B.C. 4th, 5th, and 6th Editions, Western Union,' Bentleys. 












author of 

'cotton mill management,' 'cotton spinning calculations' 

'cotton machinery sketches,' 'quadrant and shaper of the s.a. mule' 

'textile mechanics,' etc. 

late examiner in cotton spinning to the city and guilds of LONDON institute 









First Edition 189S. Second Edition 1902 

Reprinted 1007. Third Edition 1911 

Fourth Edition 19] 6 

Fifth Edition 1920. Reprinted 1921 1925 



Several corrections have been made in the body of the 
book, and some important additions made which will 
be found in the Appendix. These additions include 
practically a full description of a self-acting mule that is 
used extensively for the productions of fine numbers. 
Interesting details of another type of mule are also added. 
A very complete set of gearing illustrations of the chief 
types of self-acting mules, together with full calculations 
of each, will be found in Cotton Spinning Calculations 

recently published. 

W; S. T, 

Bolton, 1910. 


Ix the two previous volumes the preparing processes in 
cotton spinning have been fully treated. In this volume 
the subject of spinning and the preparation of yarns is 
treated in an equally exhaustive manner, vath, I trust, an 
avoidance of some defects that existed in the earlier books. 

In a work of this kind, which covers so much ground 
and deals with many features wpon which other men have 
written, it is perhaps unnecessary to suggest that originality 
is a difficult matter to ol)tain in one's treatment of the 
subject. In my efforts to do so I may have, occasionally 
but unconsciously, adopted similar methods of other writers ; 
when such has been pointed out to me I have tried, and I 
hope successfully, to prevent this being an ofTence, and I 
sincerely trust that readers and writers alike will find in 
my efforts a desire to simply present the study of cotton 
spinning in an interesting and instructive manner, so that 
it may prove of value to those whose Avell-being is dependent 
on the success of that part of the industry it represents. 

The various parts of the subject have been treated in 


such a manner tliat the young student may with great 
benefit to himself use them as a text-book, whilst the 
older reader will undoubtedly find in them much to 
interest him and develop a desire for a fuller knowledge 
and a more perfect grasp of the principles underlying 
many of the processes and much of the mechanism of 
cotton-spinning machinery. 

Completeness is impossible, and defects must exist in 
such a work as this ; but publishers, printer, and writer have 
done all they could to render it of more than ordinary 
value to those interested, and suggestions, corrections, and 
advice to make the books more serviceable will be fully 

I beg to thank several machine firms Avho have generously 
helped me by supplying me with sketches of parts of their 
machines, which it Avould otherwise have been difficult 
for me to obtain, and the Textile Mercury is specially 
deserving of recognition for the excellent reproductions 
of my drawings. 


Bolton, 1898. 


Owing to the size of the book, it has not been considered 
advisable to inchide all the smaller improvements recently 
made to the machinery dealt with in this volume. Addi- 
tions and corrections, however, have been made so as to 
bring the book up to date and render it useful to the 
practical man and to the student. 




Theory OF Spinxixg . . . . r ^ ,. . 1 

Mechaxism and AVokkixg of the Mule , c , 24 

The Rixg Spixxixg Fiiame ..... . , 278 

Bobbin' Winding Frame . . c - , , . 337 


Doi'BLING ..,.. = ,= . 35b 

Yarn Preparing Machines ...... 372 


JIii.L Planning , > 385 





Humidity ,,..,.,.., 399 


Useful Information ...... . 408 

APPENDIX L ,.,,,«„.. 421 

APPENDIX II. . . . , . , -. . 454 

INDEX ........... 483 






Diagram illustrating the Cause of Twists going to the 

Tliinnest Parts of Yarn during Spinning 
Diagram illustrating the Arrangement of the Fibres in Yarn 
Cross Section of Yarn showing Position of the Fibres 
Diagram of the Twisting Action in the Mule 

)) 55 5) " • 

,, ,, ,, and the Effect 

of an Inclined Spindle ...... 

} Diagrams showing Difference between a Vertical and an 
Inclined Si)indle ....... 

Plan of a Pair of Mules ...... 

Section of a Mule 

Plan "Mew of the Gearing of a Mule .... 

Various Arrangements of Mule Creels 

>> )> >j ... 

View of the Back of Headstock ..... 

Driving of the Mule, End and Side View . 

Gearing showing Driving of Front Roller and Back Shaft 

Plan View showing all the Mule Scrolls . 

Back Shaft drawing the Carriage out 

Out End Back Shaft Scroll moving Carriage 

Drawing-up Scroll ....... 

Check Scroll ........ 

S(|uaring Band under the Carriage ... 

I Method of Constructing a Scroll . „ . . 












Section of Mule showing Driving of Spindles, etc. 

Back View of Headstock showing Driving 

Spindle or Tin Drum Driving .... 

Xew Method of Driving Spindles and Front llolkr 

simultaneously ..... 
Section of Rim Shaft and Pulleys 
Duplex Driving and Drawing-up Arrangement 
Drawing-up hy Strap for Fine Spinning 
General View of ilechanism of Cam Shaft !Mulc 
End View showing Cam Shaft driven from Rim Shaft 
Cam Shaft when placed below the Long Lever 
Cam for iloving the Strap Fork 
Twist Latch Lever, Backing-off and Strap Fork Ar 

ment in Cam Shaft ilule .... 

-Operating the Cone Clutch on the Cam Shaft . 

Operating the Front Roller and Back Shaft 
Mechanism for Operating Back Shaft and Drawiug u^ 

Clutch in Cam Shaft Mule. . 
Holding out Catch . . . . 
Backing-off and Drawing-up Mechanism 
Strap-relieving Motion 

-Details of same .... 

j- Position of Faller "Wii'es for Diflerent Stages of the Cop 

~\ Position of Sickles and "Wires for the Inward and Outward 
/ Run of Carriage 

Mechanism of Backing-off Chain-tightening Motion 

A'iew of Spindle and Cop .... 

Diagram of Cop showing Layers and Crossiuo 

Diagram showing Curves of Variation of the 
Spindle for Winding the Cop Bottom 

Diagram of Cop 

Gearing Plan of Mule-gearing 


.-Diagrams illustrating Action of the Quadrant 
.-Diagrams illustrating Exidanation of Quadrant 



























Diagraiu.s ilUistratiiii,' E.\"]ilaiiati(in of (^Munlraiit 

Diagram of Variation of Initial Speed of Spindle 
Diagram of Rate of Movement of Nut up the Quadrant 
General View of Quadrant and its Connections . 
AVinding Drum and its Connection to the Tin Roller 
Long Shaper and its Connections .... 
Diagram illustrating the Shaper .... 

Front, Middle, and Back Plates of Shaper. 

Diagram of Long Siia2>er explaining Curvature of Long 


Diagram of Long Shaper Inclined Guide Bracket 
Diagrams of Defective Cops ...... 

Diagrams of Shaper indicating Remedies for Defective Cops 

-Faller Weighting and Easing ]\lotion .... 

Diagram of Cop, etc., illustrating Principle of Xobing ilotion 

Diagram illustrating Principle of Xosing ^lotion 

Xose Peg Arrangement ..... 
Automatic Nosing ^lotion ^v(lrked from Fallers . 
Nose Peg Arrangement ..... 
Automatic Nosing Motion worked from Shaper . 

Governor or Strapjjing Jlotioii .... 
>• ,, ,, Another ^Method 

f >> >> >i 
















no. FACE 

106. Curve showing Rate of the Movement of the Xut up the 

Screw of the Quadraut 20") 

107. General View of the Long Lever Mule .... 207 

108. Mechanism for Producing the Changes in the Long Lever 

Mules 210 

109. Mechanism for Producing the Changes in the Long Lever 

Mules showing Rim Shaft Drawing-up and Backing-otf 
Arrangements 21 i 

110. Diagram of Long Lever showing Positions after Changes . 213 

' "Twist Latch Lever and Strap-relieving Motions . , 21/ 

113. Backing-ofF Chain and its Connections showing its Tighten- 

ing Motion ......... 221 

114. Backing-otf Arrangement . ...... 225 

115. Double Speed Driving 229 

116. "Winding Motion for Fine Spinning 231 

117. Plan of Gearing of Fine Spinning Mule .... 233 

,„' 1 Group of Motions illustrating Method of obtaiuin" "Gain," I 
Z^ r "Ratch," Roller Motion whilst Twisting at the Head, Y 
'I and Roller Motion whilst "Winding . . . .1 

12'? ~\ „ , . 

JBackiug-otl Arrangement . ...... 241 

124. Section of Rollers and Stand showing Weighting, etc. . 242 

125. Diagram showing Method of calculating Pressure on 

Rollers 242 

126. Gearing of Rollers 242 

127. Diameters and Spaces of the Rollers in a Mill for Japanese 

Cotton 245 

128. Diameters and Sjiaces of the Rollers in a ^lill for Chinese 

Cotton 245 

129. Diameters and Spaces of the Rollers in a Mill for Indian 

Cotton 246 

130. Diameters and Spaces of the Rollers in a 31111 for American 

Cotton 247 

131. Diameters and Spaces of the Rollers in a ilill for Egyptian 

Cotton . . . 248 

132. Diameters and Spaces of the Rollers in a ilill for Egyptian 

Cotton 249 

133. 'I 

' /-Mechanism of another Form of Long Lever Mule . . 250 

134. ] 

ISo. General A'iew of Ditto and showing Double Soeed Driving . 252 











lAnti-snailiiig Motion . . . » o » . 

Anti-snailing Motion. Another Method .... 
Diagram illustrating the Change in the Inclination of the 

Yarn as the Carriage ti'avels out . 
Diagram showing Horse-power of iMule 

Gearing Plan of Mule ..... 

Half Section and Half Elevation of Ring Frame 
Rope Driving for both Tin Rollers . 
Section of Roller and Stands showing "Weighting 

>> J J >> )) 

Diagram explaining the Reason for Inclined Roller Stands 
Diagrams explaining the Weighting of Roller Stands 
Section showing Rollers, Thread-guide, and Spindle , 

j Thread Boards and their Lifting Arrangement . 

Section showing Poker, Ring Plate, and Ring . 

J , of Ring 

,, Douhle Ring ..... 

,, Ring and Traveller .... 

Building Motion ...... 

Diagram of Ring Bobbin ..... 

Building Motion and its Connections to tlie Pokers 

I Diagrams explaining how the Traveller puts the Tw 
I the Yarn ....... 

ist in 

Diagram showing Ballooning .... 
Diagram illustrating the Forces affecting tlu; Travdl 

!- Diagrams illustrating Minimum Sizes of Bobbins 

Sections of Self-contained Spindles . . . . 

,, Rabbeth Spindle . . . . . 

,, Booth-Sawyer Spindle . . . . 

,, Dobson-Marsh Spindle . . . . 
,, Five typical Self-contained Spindles 

,, Oil Cup Sjiindle . . . . , 

Catch for holding tlie S2)indlc down . 














Seition of Spindle and Tirn Bobbin . 
Gearing ol" King Frame . . . . 

Section of Bobbin Winding Franie . 

,; Quick Traverse Winding Frame 

Traverse Motion of Quick Traverse AVinding F 

j- Section of Quick Traverse Winding Frame 

,, Clearer Winding Frame . 

Gearing of Doubler Frame . 
Section of Doubler Frame . 
Creel and Trouglis of Doubler Frame . 

Trough of English System of Doubler 

, , Scotch , , 

Section of Doubler Spindle . 
Knee Brake for Doubler Spindle 
Roller Stop Motion for Doubler 
Roller and Spindle Stop Motion for Doubler 
Ring and Traveller of Doubling Frame 

Diagram of Tv;ist in Doubling Two or more Ends into One 

Rope Driving in Ring Spinning and Doubling Frames 

Section of Reel showing Dotling ilotion 
Side View of Reel .... 
Old Form of Doffing ilotion 
Coleby's Reel ..... 

]• Sections and Gearing of same .... 

Gassing Frame ..,..,. 
Bundling Press ...... 

Plan of a Card Room for a Mill of 80,648 Spindles 
Plan of 4th Spinning Room ,, ,, ,, 
Plan of Card Room Machinery .... 

Plan of Preparing Machinery for Combed Yarn 
Plan of Card Room of an Indian Mill 
Plan of a Spinning and Weaving Mill 
Hygrophant ...... 

Improvements in Long Lever j\hile . 
Short Shaper . ..... 








Copjiing ^Motion and Sliort Sluiper . 
Backing-otf Motion, etc. .... 

Setting-on and Drawing-up Motions . 
Backing-off Motion ..... 

Roller-delivery and Twist Motions . 
Section of Donble Rim Shaft for Double Speed 
Brake ]\rotion ...... 

Roller-delivery Motion .... 

Roller Motion Click Wheel 
Setting-on and Drawing-up Motions . 

,, „ ,, Details of Fig. 2J0 . 

Dra-wing-out, Ratcliing, Roller, Backing-oflf, etc., Motions 
Assistant Winding i\Iotion 
Gearing Plan of Special Fine Mule . 
Jacking Motion .... 

Strap Relieving Motion 
Twist Motion on Tin Roller 
Backing-ofF Motion . . „ . 
Gearing Plan of Mule . , , 

/■Drawings of Single and Twofold Yarns . 

Section of Horizontal Quick Traverse Gassing I' 
/"Section of A'ertical Gassing Frame , 
I Section of Split Drum Traverse Gassing Fi'an 
/"Section of Upright Spindle Winding Frame 
I Bottle-shaped Winding Bobbin 
Section of Quick Traverse Winding Frame 
Section of Ball Clearer Drag 
Gearing and Cam of Winding Frame 
Section of Reel ; from Cheeses and Bobbins 
Disposition of Fibres .... 

Passage of Cotton between Cages and Calender Rollers 






Photograph of Cotton Bolls .... Fro7itispiece 

no. PAGE 

1. Map of the Cotton Gro^ying Countries of the World . , 3 

2. Enlarged Diagram of Cotton Fibre, showing Ripe, Unripe, 

Over-ripe, and irregularly Twisted Fibres, together with 

Transverse Sections ....... 21 

3. Diagram showing the Degree of Irregularity in the Direc- 

tion of Twist, and the Cotton Fibre .... 24 

4. Section and Plan of the " Knife Roller " Gin, Double Action 35 

5. Diagram showing effect of Knives in " Knife Roller " Gin . 36 

6. Enlarged Section of the Ginning Organs of " Knife Roller" 

Gin 36 

7. Section through a Single Action " Macarthy " Gin . . 39 

8. Relative Positions of the Ginning Organs in "Macarthy" Gin 40 

9. Section f^hrough a Double Action " Macarthy " Gin . . 42 

10. ,, "Saw" Gin with Lattice Feed, and 
Condenser ......... 43 

11. Bars of the "Saw" Gin ....... 44 

11a. Section of " Saw " Gin with Double Row of Saws . . 45 

llu. ,, ,, showing Inner and Outer Breast . 46 

Foot-Roller Gin .49 

Simple Churka Gin ........ 49 

12. Section through Bale Breaker witli Four Lines of Rollers . 55 

13. ,, Pedal Bale Breaker ..... 57 

14. ,, Porcupine Bale Breaker . . . .58 

15. Hopper Bale Breaker with Dust Extractor .... 60 

16. „ „ ,, .... 61 



17. Hopper Bale Breaker with Dust Extractor. 
















ilixing Room, with Lattice Arrangement and Bale Breaker 
Mixing Room, with Lattice, Bale Breaker, Hopper Feeder, 

Porcuiiine Opener, and Trunks to Opener . , 
Combined ilachine formed by Coupling Hopper Bale 

Breaker, Hojiper Feeder, Double Buckley Opener, Beater, 

and Lap End ......... 

Plan and Elevation showing Hopper Bale Breaker, Hopper 

Feeder, Small Porcupine Opener, Crighton's Opener, and 

Exhaust Opener, all coupled together . . . . 
Plan and Elevation showing Hopper Bale Breaker, Hopper 

Feeder, Small Porcupine Opener, Crighton's Opener, and 

Exhaust Opener, all coupled together 
Section of an Automatic Hopper Feeder 

Diagram showing Plans and Relative Positions of Hopper 
Feeder, Opener, and Scutcher 

. Section through Hopper Feeder 

,, a Vertical Beater Opener 

,, a Small Porcu]iine Opener 

,, Footstep Bearing of A'ertical Opener 

,1 ), )) ), 

,, Double Vertical Opener . 

., Vertical Ojjeuer with Horizontal Beat 

and Lap Part .... 
,, Horizontal Conical Beater Opener . 

,, Large Porcupine Opener (Single) with 

Hopper Feeder 

,, Large Porcupine Opener (Double) with 

Hopper Feeder . . . . . 
,; Tiic Buckley Ojiener (Single) . 













Section througli tlie Buckley Opener (Double) with Hopper 
Feeder ...... 

,, Horizontal Exhaust Opener with Small 

Porcupine Feeder .... 

,, Single Scutcher, Doubling from Four Laps 





42. ,, ,, ,, Three Laps 

43. Diagram showing the arrangement of Doubling from Laps 

44. Longitudinal Section through Pedal Roller and Pedals 

45. Diagram explanatory of the Curves and Cone Drums 

46. ,, showing method of forming Cone Drums 
50. Arrangement of Bolls and Boll Rail for reducing Friction . 

^1- j> >) )i )> 

52. Link and Lever Arrangement for Regulator Motion 

53. "Wire and Lever Arrangement for Regulator Motion 

54. Link and Lever Arrangement for Regulator Motion 

55. Section through the Feed Part of Scutcher, showing Cotton 

struck from the Pedal Xose 

56. Section through the Feed Part of Scutcher, showing Cotton 

struck from Feed Rollers 
Adjustable Beater Bars in Scutcher 

I Section, End View, and Plan of Feed Regulating ilotion 
r of Openers and Scutchers ...... 



Sections of Feed Rollers and Pedal 

Section showing Stripping Plate, Beater Bars, etc. 
,, Adjustable Beater Bars, etc. 

Elevation and Plan of Double Scutcher 
62a. Section showing Beater and Beater Bars, etc. . 
62b. Teacher's Patent Pedal . 

63. Diagram of Three-Bladed Beater 

64. ,, Two-Bladed Beater . 

65. Section through a Combing Beater 

66. Lap End of Scutcher with Cages, etc. 








' [Metliod of Weighting Calender Eollers of Lap End . 

67. Stop Motion for Full Laps 





Diagram of Two Wheels in Gear 
,, a Train of Wheels . 

Elevation of the Gearing of a Scutcher 
Plan of the Gearing of a Scutcher 
Diagram of Dust Flues and Chimney 




Section through Roller and Clearer Card 
Enlarged View of Roller and Clearer 
Section through the Revolving Flat Car^l 
Feed Roller AiTangement in Card 
Dish Feed Arrangement in Card 
Diagram of Cotton after the passage throu 

the Taker-in ..... 
Dish Feeds for various classes of Cotton 
Section through Dish Feed, Mote Knive 

Undercasing ..... 

rh the Tei 

Taker-in, am 

Diagrams of the Action of the Taker-in Teeth 

,, Card Setting Gauges 

Section of Feed Arrangement, etc. 

,, through Taker-in and Cylinder . 
Enlarged Section of Taker-in and Cylinder 
Section of Card Filleting .... 
Open-Set Card Wire .... 

Twill-Set Card Wire .... 

Rib-Set Card Wire 

Diagrams of the Angles of Carrl Wire 

Section showing Flats entering upon the Cylinder 

th of 




Relative Positions of Flats and Cylinder . 
Diagram ex[)laiiatory of elFect of Grinding 
Card Flexible Bend, Five Setting Points , 
,, ,, Single Setting Points 

Diagram explanatory of Fig. 104 

Card Flexible Bend, Five Setting Points . 

Card Bend with Steel Bands 

,, Flexible Bend, Single Setting Point 
Diagram explanatory of Fig. 110 
Card Flexible Bend, Single Setting Point 
Section throngh Flexible Framing and Cylinder 
Card Flexible Bend, Single Setting Point 
,, ,, Five Setting Points . 

Section of Fig. 115, showing Adjustment, etc 
Adjustable Card Centre 

Section through Doffer and Cylinder 

Section through Coiler with Details 

'Back Stripping Comb 

Sections of Card "Wires 

Flat Grinding Arrangement with Details 












137. Flat Grinding Arrangement with Details 

138. Diagrams exjjlanatory of Fig. 137 

139. Section of Horsfall Grinding Roller 

140. Doffer Driving .... 

141. Section of a Comb Box 

142. Slow Motion for Doffer 

143. Card Feed Roller Weighting . 

144. Diagram of Card "Web 

145. ,, ,, . . 

146. Elevation of the Gearing of Card 

147. Plan View of the Gearing of Card 

148. Diagram of Prices of Standard Grades of Cotton 

149. Double Roller "Macarthy" Gin 

150. Hopper Bale Breaker. Dobson and Barlow 

151. Small Porcupine Opener. Platts 

152. Hopper Feeder .... 

153. Exhaust Opener. Feed Part Section 

154. Travelling Lattice in Dust Trunk 

155. Buckley Opener. Taylor Lang 

156. ,, ,, Single for Four Laps. Taylor La 

157. ,, ,, Howard and Bullough 

158. „ ,, Lap End 

159. Pressure Gauge for Air Pressures 

160. Lattice under Dust Grids. Howard and Bullough 

161. Pneumatic Delivery of Cotton. Dobson and Barlow 

162. Details of do 

163. Pneumatic Delivery of Cotton. Another Method 

164. Detail of do 

165. Diagram of Lengths of Cotton Fibres 

166. ,, ,, ,, ,, and "Waste 

167. ,, showing L-regularities in Scutcher Laps 

168. ,, of a Perfect Lap . 

169. ,, of an Irregular Lap 

170. ,, of Pedal Roller and Pedal Ends, showing Feeding 

171. ,, of Scutcher and Opener Cone Drums 



172. Diagram of Scutcher and Opener Cone Drums . . . 296 

173. ,, ,, ,,,,,, . , . 298 

174. showing Irregularities of Card Sliver . . . 300 

175. „ ... ,, ,, „ ... 301 

176. Mote Knives and Undercasings of Card . , , . 302 

177. Flat Grinding Apparatus. Dobson and Barlow , , 303 

178. Doffer Slowering Motion. Howard and Bullough 304 

179. A 

180. Ivoulten Opener . . - 306 

181. J 


1. Section of Draw-Frame 

2. Tandem System of Draw-Frames 

3. Alternate Sj'stem of Draw-Frames 

4. Zigzag System of Draw-Frames 

6 J 

Weighting of Rollers in Draw-Frame 

7. Solid and Loose Boss Rollers 


I Diameters and Spaces of Draw-Frame Rollers for variou; 
I classes of Cotton 

Diagram showing effect of Doubling and Drawi 

Diagram illustrating Draft in Draw-Frame 

Front and Back Stop jMotiou 

Details of Stop Motion in Draw-Frame 

Front and Back Stop Motions in Draw-Frame 

Electric Stop Motion 

Patent Revolving Top Clearer 

Ermen's Top Clearer 

Colling's ,, ,, 

Full Can Stop Motion 

Section of Draw-Frame 

Asa Lees 

Dobson and Barlow 

Gearing of Draw-Frame 

■Driving of Rollers in Draw-Frame 




14, 15 




27. Diagram of Roller Gearing in Draw-Frame 

28. ,, ,, ,, . . . 
28a. Draw and Lap Macliine. Dob:-on and Barlow 
28b. ,, ,, ,, ,, . . 
28c. „ „ ■ „ „ . . 
28d. „ „ ,, ,, . . 
2Se. Gearing of Ribbon Lap Macliine .... 

29. Section through Comber (Duplex). Dobsou and Barlow 

30. Star Feed Wheel 

31. Section through Comber and Nipper Cam 

32 1 

' j-Two Arrangements of tlie Xippers .... 

34. Quadrant Cam and Quadrant Feed in Comber . 

35. Roller or Quadrant Cam showing Cycle of Actions . 

36. Side View, Quadrant, Quadrant Cam, and Clutch Cam 

37. Notch Wheel Feed Motion in Comber 

39. Detaching Roller Meclianisni . 

40. Section of Single Nip Comber . 

41. ,, Double Nip Comber 
42. -^ 


(-Diagrams explaining the Combing Action 


51. Diagrams explaining Action of Nasmith's Comber 

52. Detail of Nasmith's Comber .... 

56. [Gauges for Nasmitli's Comber 

Section through Nasmitli's Comber 
Gearing Plan of Nasmith's Comber 
Detail of Nasmith's Comber 

^ Details of Nasmith's Comber 



58. Stop Motions ou Comber. Hetliciington . 

59. ,, ,, ,, . . 

60. Whitin Comber. Howard and BuUough . 
60a. ^ 

60b. -Diagrams explaining Action of Whitin Comber 
60c. } 

61. Section of Comber ...... 

62. Gearing Plan of Comber ..... 

63. Section through Flj'-Frame .... 

64. Plan of the Spindle Rail 

65. Section through the Rollers and Stands . 

66. Cap Bar 

^1. Rollers and Stand in Fly-Frame 

68. Diameters and Spaces of Rollers in Fly-Frame 


Fly and Bobbin with Driving 


71. Spindle Footstep Bearing 

72. ~i Diagrams explaining the Action of 

73. J Presser ...... 

74. Flyer Legs with Straight and Curved Slot 

75. Driving the Bobbins and Spindles 

76. Diagrams explaining Winding in the Fly 

77. Diagi'am explaining "Flyer Leading" 

78. ,, ,, " Bobbin Leading "' 
79. 1 Diagi'ams explaining Variations of S[)ee 

80. J during "Winding .... 

81. Gearing of Fly-Frame 

82. I . 

> Diagrams explaining the Curves of the Cone Drums 
83. ) 


. 1C6 

. 107 
. 108 

. 109 

. Ill 



. 134 
the Flyer and 

136, 137 



J of the Bobbin 

151, 153 

84. "I Diagrams explaining the Constrii 

85. J Drums 

86. Epicyclic Train of "Wheels 

87. ,, ,, ,, 

88. ,, ;, ,j 


ction of the Cone 













Differential Motion (Sun and Planet) 
Section tlirough Patent Differential Motion 

Diagram of Fly-Frame Full Bobbin 

Gearing of Fly-Frame ..... 

Building or Traverse Motion in Fly-Frames . 

Details of Traverse Motion in Fly-Frames 

Diagram explaining Traverse Motion in Fly-Frames 

Building or Traverse Motion .... 

Improved Methods of Driving the Bobbins 
Ordinary Method of Driving the Bobbins 
Gearing of Fly-Frame .... 

Bobbins and Skewer'- 




In the course of the two preceding volumes, it has been 
considered necessary several times to examine into the 
principles underlying the various operations of the different 
machines, and the etfect these machines have upon the 
cotton passing through them. The remarks already made 
may now be extended into a more detailed examination ; 
and as they will serve the purpose of an introduction to a 
description of the Self-acting Mule, on which the final 
operation of spinning — that of making the cotton into 
yarn — is performed, the conclusions arrived at will 
materially assist in making some of the operations of 
that machine more readily understood. 

The ideal state of the cotton, to obtain which every 
effort has been made, may be su.mmed up as follows : — 
Absolute cleanliness ; equality in length of fibre ; perfect 
parallelisation of the fibres ; a disposition of the fibres 
among themselves such as to ensure the strongest result: 
uniformity throughout the length in diameter, weight, and 
strength ; and a round solid yarn as the ultimate result of 
the whole series of operations. 

Sufficient has been said already to show that the first 
three conditions have been almost satisfactorily attained, 

VOL. Ill B H 


especially when the cotton has been combed ; otherwise it 
cannot be assumed that the parallelisation of the fibres is 
so nearly perfect as is generally supposed. It will, there- 
fore, be necessary to confine the examination to the last 
three ideal conditions of a perfect yarn ; and to see 
whether they are possible of attainment, and how far 
present machinery is capable of achieving them. 

Uniformity of the Yarn. — In the first place the ques- 
tion of the uniformity of yarn will be considered. Uni- 
formity applied to yarn generally means that the yarn is 
uniform in diameter and in weight ; the uniformity of 
strength is not as a rule the occasion of so much observation 
as the other two, provided a very long length doubled 
many times answers to a satisfactory test for breaking 
weights (this point will be dealt with later on). 

Regularity of Diameter. — There are several ways of 
testing yarn as to its uniformity of diameter, the common 
one being to wrap a certain number of lengths side by side 
on a black slip of wood or cardboard. The contrast of colour 
thus aftbrded gives a very good idea of inequalities, and a 
judgment based on such an examination is generally con- 
sidered sufficient for practical purposes. It is, however, at 
best only a crude method, and when what are considered good 
results are passed under the microscope with low power, 
the great diff"erence between adjacent diameters is instantly 
recognised. The fault of the " sight " method of judging 
yarn is due to our inability to see small diff"erences in 
small diameters. "We give an example : — 

Suppose a 20's yarn is yl^ inch in diameter, this we 
can see is a very small dimension ; if a thinner one is taken, 
say j-^ inch diameter, and a thicker one, say -^^^ inch 
diameter, the difference between the three yarns is so little 
triat it would require an unusually good eyesight to detect 


it. When, however, such a yarn is passed under a micro- 
scope and magnified say 100 times, the y^^ inch would 
become of such a size as to show clearly any difference of 
other diameters when compared therewith ; if y^Q of an 
inch became enlarged to one inch, the two other dimensions 
would become -j^- and ^^ of an inch respectively — the 
difference in each case being J^ inch. Such a difference 
is really enormous and represents a large percentage of 
variation, and yet it is one that Avould not readily be 
noticed by the ordinary testing method, simply because 
of the eye's failure to judge of such small differences. If 
a finer yarn is taken, say 60's, its diameter of say -^^^q inch 
would render it even more difficult to discern variations 
of diameters unless they were unusually large. We thus 
see that the usual method of judging yarns is by no means 
perfect ; it evidently satisfies ordinary requirements of 
trade, but we ought not to ignore the fact that very 
unequal yarn is still made in spite of all that has been 
done to perfect the machiner}' for making it. Combed 
yarns among the higher numbers display almost as great 
an inequality of diameters as the low numbers do, mainly 
because of the fact mentioned above ; but combed 60's com- 
pared ■\\dth ordinary 60's is much superior, although, as 
already noted, a very large percentage of A^ariation exists. 

Another way of rendering very apparent the variation 
that exists in the diameter of yarns is to double together 
two rovings of the same hanks, and the same cotton, but 
one of them dyed, the other white, or of a contrasting 
colour. Each roving b}' itself will probably show very 
little variation ; but when doubled, the mere fact of 
twisting will bring out everj' thin and thick place in a 
remarkable manner. The writer had recently a striking 
object-lesson on this point, while in a spinning mill on 


the Continent. The specialty of the mill in question is 
coloured and mixed yarns made from doubled rovings. 
A cop formed of double roving, one white and one black, 
at the mule, while generally even in a^jpearance at the 
first glance, was in reality one whole length of irregu- 
larities. These were made apparent by the distinct 
character of the twists, which could easily be seen, owing 
to the contrast in colour of the two rovings ; the twists 
lay very close together in places, drawing the yarns tightly 
together and making a thin hard place ; at others they 
were correspondingly separated, and at these spots a thick 
fuzzy place was formed. Such irregularities existed and 
followed each other in varied lengths from ^ to 1-^ inches 
throughout the cop. At first the suggestion was made 
that the dyed roving was perhaps the chief offender; but 
when two dyed rovings were used, similar results followed, 
and an examination of the rovings only showed that they 
were good average results of "good middling" cotton 
obtained after passing through modern preparing machinery. 
The same two rovings put through a ring frame gave a cop 
that was scarcely distinguishable from that of the mule so 
far as the marked character of the variations was to be 
seen. A strange thing about it was that when double 
rovings of white, or two of the same colour were used, the 
yarn was remarkably good and even in appearance ; but no 
sooner were the twists made apparent by a contrast of 
colour than the unreliability of one's judgment by sight 
was immediately emphasised. 

Regularity of Length and Weight. — In close con- 
nection with the uniformity of diameter is that of 1-ength. 
Owing to the universal use of the wrap reel and scales, any 
variation in this direction is quickly noted, and the judgment 
lias little if anything to do with the decision. But even with 


tlie Aviiip reel it is only average results that are dealt in ; long 
lengths are always taken, varying from Il^O to 840 yards, 
and the weights of the same lengths from different cops are 
compared. This rough method, however, fails to show 
whether the yarn is uniform, for if fifty cops can be taken 
from different i)arts of the same mule, wide variations will 
be noted in their weighings. Such variations, however, Avill 
be intensified if a number of wrappings be taken from the 
same cop and carefull}^ compared. Diflferences like the one 
just suggested are of a distinct practical character, and being 
very well known are always allowed for ; but if the examin- 
ation be continued by splitting up say 840 yards into pieces 
of 10 yards each, or even less, and weighing them, the same 
average result for the whole length will be given, but the 
individual weighing will vary to an extent that is astonish- 
ing. It is a difficult matter to say how it happeiis that this 
state of things exists ; it is probably due to errors in the 
previous machines, })rincipally in the card and scutcher, and 
the reason for its non-detection at these machines is the too 
great reliance that is })laced on average weighing in the bulk, 
and the fact that a slight variation under such conditions 
is not considered of importance for practical purposes. 

To show what is meant, let it be supposed a scutcher 
makes laps that vary only within \ lb. in a lap of 32 lb. ; 
this would be a very good result indeed, and if it represented 
the actual variation of the laps, there w^ould be an luiusual 
degree of uniformity in the yarn. But when we consider 
that a difference of \ lb. in a 32 lb. lap causes a variation 
of a single hank at 60's, it will be readil}^ understood that 
uniformity of scutcher laps in the bulk is not a good 
foundation on which to base anticipations of uniform yarn. 
By taking periodically very short lengths of the lap, and 
weighing them, a much better idea of the variation would 


be arrived at, and means could then be taken to ensure 
more uniform results. Practical tests in this direction of 
weighing short lengths of what seemed to be a good lap, 
have shown variations of as much as 25 per cent. It is 
therefore not surprising to find that yarn is not uniform ; 
to a large extent variations will always exist, but much 
could be done to remedy them if a correct judgment were 
formed by individual observation instead of depending so 
much on large average results. 

Although irregularity of diameter is such a noticeable 
feature, it by no means follows that it corresponds to the 
variations in weight, except in the case of sliver and 
rovings ; in yarns the twist put in has an all-poAverful 
influence in affecting the diameter. There is no doubt 
from even a casual observation that variations exist, but 
they are so distinctl}^ brought to view by means of the 
twist put in the yarn, that a little consideration of this 
feature will not be out of place. 

Twist and Weft. — The object of twisting has already 
been explained. P'rom the fact that the twist can be put 
in the yarn in two directions, the terms "twist and weft 
way " are general. The term weft, however, is not applied 
so much to the direction of the twist as to its condition. 
It implies less twist and a softer yarn, and as a rule weft 
yarn is made from cotton that gives a soft and more jjliable 
effect. Twist is as a rule formed by turning the spindle in 
the same direction as that in which the hands of a clock 
turn, and it gives to the yarn a spiral twist, corresponding 
to that seen on a right-handed screw. Weft has its twist 
})ut in generally in the opposite direction. It does not 
always follow, however, that the direction of the twist gives 
the yarn its character of twist and weft. 

Effect of Twist. — The tendency of the twists lo fly to 


the thin places in the yarn is a well-observed fact, and 
several suggestions have been made as to its cause, the chief 
one being the greater difhculty of twisting a thick place than 
a thin one. Whether the thick place be caused through a 
larger number of fibres existing at the place, or through the 
fibres being coarser, it is highly probable that the above 
reason is the correct one ; and if so, it resolves itself into 
a purely mechanical fact that the twists should fly to the 
thinnest places of the yarn. 

The following illustrations will serve to make this point 
clear, and every reader can readily convince himself of its 

truth. Take three lengths of narrow tape (or even slips of 
paper), cut to the shapes shown in Fig. 1 at A, B, and C. 
A is a uniform narrow slip, and it has been twisted one 
complete turn : a perfectly uniform twist is the result, 
because the resistance to twisting is the same throughout 
the strip. If a second slip be taken, wider at one end than 
at the other, as at B, the complete turn does not give a 
uniform result, the wide end of the slip being more difficult 
to turn, and as a consequence the twist is confined to the 
narrow end. By making the slip wide in the middle and 
thin at the ends, as at C, we have a similar effect ; but in 
this case the thick portion, while it has evidently turned 
and transferred the twists from one thin end to the other, 


has failed to be twisted itself, the naiTOwer ends only 
receiving the twists. This is very conclusive evidence of 
the eftect of a thick or thin place in the yarn, and, as can 
be seen, it is one of a purel)^ mechanical nature. If a 
similar problem were presented in regard to wire, or a 
shaft, its solution would be found at once, and definitely, 
on the above lines ; and the fact that yarn is not so 
homogeneous as iron does not interfere very materially 
with the reasoning ; it only prevents a definite conclusion 
as to the amount of the result being arrived at. 

As will be seen a little later, the peculiar action of the 
mule — and it is one of its chief advantages — has a beneficial 
effect in modifying the extreme result of twist ; neverthe- 
less, it is always considerable, and the only reniedj^ is to be 
sought in more uniform results in the preparing processes. 

Strength of Yarn. — The strength of yarn depends upon 
two principal factors, namely — the kind of cotton, and the 
arrangement of the fibres among themselves. The strongest- 
fibred cotton does not make the strongest yarn : firstly, 
because it is shorter, and therefore not capable of being 
bound into as strong a yarn as the longer but weaker fibres; 
and secondly, Ijecause its greater diameter does not allow 
of as many fibres in the cross section of the yarn as is the 
case when finer fibres are used ; the percentage of extra 
fibres in such a case is greater than the percentage of weak- 
ness in the individual fibre : consequently, if, say, 30's be 
spun out of Indian and Sea Island cottons, the weaker Sea 
Island fibre would make the stronger yarn — for the two 
reasons given above. 

Arrangement of the Fibres in the Yarn. — The 

disposition of the fibres in the yarn is rather an important 
matter, and it is qi;ite obvious that — other things being 
equal-— the strongest yarn is that which has its fibres 


arranged to the best advantage in respect to one another. 
The actual arrangement of fibres in yarn is of course 
practically iinknown, but we may reasonably argue from 
some of the known facts, and conjecture. For instance, 
cotton that retains 15 to 20 per cent of its shorter fibres 
is clearly bound to produce weaker j^arn than if those 
fibres were removed by the combing process ; and in the 
same way it is reasonable to suppose that the haphazard 
arrangement of the fibres taken from the doffer must yield 
poorer results as to strength than the ordered condition of 
the fibres after passing through the comber. Both sets of 
fibres, however, are modified as to their ai-rangement in 
the subsequent processes, and it is most probable that the 
former is greatly improved, whilst the latter loses some- 
what of its advantages. Nothing definite is known, 
however, and this opinion is only expressed after a careful 
examination of the drawing process, as seen in such 
machines as are used for jute ajid flax, where the operation 
— owing to the long length of fibre — is easily seen, and its 
action readily followed. 

In order to demonstrate what might be considered an 
ideal state in the disposition of the fibres, it Avill be 
necessary to make use of the diagrammatic method, similar 
to that used b}'^ Mr. Nasmyth in his book on Cotton 
Spimmig. The reasoning and conclusions arrived at, 
however, are different, the similarity being simpl}' in the 
diagrams. Fig. 2 shows several possible arrangements 
of the fibres ; but it must be thoroughly understood that 
none of them are prol^able, the actual conditions most 
likely partaking of a combination of all of them. At A an 
arrangement is shown which gives a perfectly uniform 
thickness of yarn ; but it is absolutely without strength, 
for the obvious leason tiiat the fibres are simply end-on- 

lo C OTTO A' SPINNING chap. 

end, and are not bound together in any way. It may be 
taken as rej^resenting one extreme in any combination 
that may take place, and the probability of its happening 
to a certain degree, if only a small one, introduces a 
possible cause of the well-known weakness of yarn com- 
pared with the strength of the individual fibres. At B the 


»-l--i 2 




*- — I— 5^ 

A 3 


Fig. 2. 

fibrss are shown with a short overlap ; when twisted 
together a certain strength would be obtained, but it 
would clearly be only of a slight character, and it is highly 
probable that its weakness would lie in the slipping of the 
fibres over one another owing to the insufficient lap. A 
more serious evil, however, is seen in the unevenness of 
the yarn that would be made ; at 1 the thickness is tliat 


of twelve fibres, -while at 2 only six fibres are twisted. 
This arrangement most certainly exists in yarn, and is 
the cause of unevenness. The Avell-known action of the 
comber arranges the fibres on this plan, but of course Avith 
a much gi'eater overlap. At C is shown a modified form 
of B, in which the fibres produce a uniform thread, and 
equally as strong. It is an unknown point Avhat propor- 
tion of the length of the fibres ought to be twisted in 
order that the weight to cause rupture should just equal 
that necessary to produce slippage. If one-eleventh of the 
length be suihcient to resist slippage when a number of 
fibres (say twelve) are twisted together, the arrangement 
shown at C would be the strongest possible one. At 2 a 
section is given of the weakest place, and yet it is only 
9 per cent less than the theoretical value of all the fibres ; 
at 3 the full value is obtained, but since the strength of 
the yarn is that of its weakest spot, rupture would take 
place probalily at 2. At D the ari-angement is one in 
which the greatest possible adhesion is given to the mass 
of fibres in the yarn when twisted, the possibility of slip- 
page being reduced to a minimum, and from this point of 
view it has an advantage over C ; it is also uniform, but 
a glance at the diagram will show that the weakest spot 
of such a combination of fibres contains only six fibres 
(see 2) and is therefore 50 per cent weaker than the 
strongest place (as at 3), which has twelve fibres in cross 
section. Next to A, D is the poorest combination that 
can be given to the fibres, on the assumption, of course, 
that we are treating of fibres all of which are of equal 
length, and that half the length of the fibre is requisite for 
twist in oi'der to equal the breaking weight. It is the 
opinion of the writer that a much smaller proportion of 
length is sufficient, and of course the smaller it is the 


stronger is the yarn ; the object of attainment seems to be 
to lay the fibres in such a way as to break as much as 
])ossible the joint caused by the ends coming together. 
Such an assumption as that mentioned above is, however, 
far from being correct in practice. In the best combed 
cotton a hirge percentage of variation exists, and this 
means that the overlapping of the fibres follows no strict 
law ; moreover, when we know that a very large draft is 
given to the sliver after passing the comber, before it is 
made into yarn, it is clearly impossible to suppose that 
the apparent regularity with which the comber does its 
work results in the fibres of the yarn being arranged as at 
C or D. If a single end of combed sliver, with its fibres 
arranged as at C (which is quite possible), be made into 
yarn, the nearest approach to the aggregate strength of 
its component fibres will be obtained, but when several 
slivers are doubled, the overlappings of the filjres in the 
different slivers do not correspond, and a condition is 
produced which prevents dependence on the original 
arrangement. We may, however, conclude that a stronger 
yarn will be made, both from the greater uniformit}^ in 
the length of the fibres as well as from their better dis- 
position, which is a source of strength when twisted. 

The above remarks will have prepared the reader for 
the conclusion that the strength of yarn is a very variable 
factor ; that the disposition of the fil)res folloAvs no fixed 
arrangement; that it is impossible to arrange them in a 
manner to obtain more than a relatively small percentage 
of the strength of the individual fibres ; and that the 
probable arrangement of fibres is a mixture of those shown 
in Fig. 2. 

Rotundity of Yarn.— In ct)nsidering the question of 
the rotunditv of the yarn after it has been twisted, it ought 


to be remembered that it is not simply one of a numl)cr of 
objects sought for in the making of good yarn : it really 
represents the sum and substance ci all of them combined. 
Granted that ideal conditions in material and processes 
existed, perfectly round yarn would be the natural result ; 
but in, the absence of ideal conditions, round, or rather 
sectionally round, yarn is still possible. 

The roving as it passes between the rollers is compressed 
into a thin flat ribbon of fibres, and on issuing from them 
is immediately twisted into a strand in which all the fibres 
are more or less bound together. Considering the number 
of twists given to the yarn it is natural to expect a 
cylindrical form as a result. The only thing that interferes 
with this conclusion is the homogeneity of the fibres as a 
whole, and it is upon this feature that the question de- 
pends. Roundness is the result of twisting. If the yarn 
were homogeneous throughout its length it Avould have a 
circular appearance in a sectional elevation, but this 
rotundity would not necessarily be perfectly cylindrical, 
because, as we have already pointed out, the sliver from 
which the yarn is spun is unequal, therefore there Avould 
exist different diameters at various points. In spite of 
this a sectional view would give a circle. It is quite 
obvious that thick places, whether containing more fibres 
in cross section, or the same number of fibres each of a 
greater diameter, can be made round, just as readily as 
in the case of a small number of fibres. The fact that 
yarn is far from being round must be sought for on the 
assumption that any given section of it is not homogeneous, 
which assumption can be easily verified l)y the microscope. 
Suppose that in the thin ribbon of fibres which issues from 
the front rollers there are two or three fibres slightly 
thicker than the rest, the presence of those fibres will 


cause that particular part to offer a greater resistance to 
twisting than that of the weaker and thinner fibres, and as 
a consequence an irregular shape will ]je produced. Xow 
it is fully well known that thick and thin fibres exist 
throughout the best of cotton. In some classes this is 
more so than others, and it is the fact that a few of these 
thicker or even utuisuall}" thin fibres can be found in the 
cross section of any yarn that causes the irregular shape it 
is found to possess. In regard to the round form of section 
of yarn and of fibres, it is as well to observe that it may 
have two distinct meanings. It may mean that the cross 
section itself is round or that the general view from the cross 
section is round. These are two very widely different 

Elasticity. — Elasticity is all-important in the character- 
istics of yarn, and this to a greater or less extent exists in 
all textile fibres. It may be defined as a property Avhich 
enables a substance to be distorted to a certain extent and 
yet to return to its original condition without having 
suffered injury. If all the fibres in Fig. 3 were packed 
closely together, there would be very little elasticity, because 
the fibres have no room in v/hich to yield ; the j'arn cannot 
lengthen unless the diameter becomes smaller at the same 
time, so that if the smallest diameter is obtained by close 
packing, the yarn ceases to have elasticity in the sense 
understood in cotton spinning. The drawing shows that 
the fibres are not arranged in any close order, and, as a 
consequence, if the yarn is stretched slightly, the diameter 
is reduced ; the fibres come together, and in doing so cause 
a lengthening to take place. A yielding of this kind 
naturally relieves the j-arn of any shock that may come 
upon it and thus prevents rupture. At the same time 
the fibres themselves, in the aggregate, possess sufficient 



elasticity to cause theiu to spring back into their original 
position when the pressure is removed from the yarn. 

Whilst recognising that elasticity and strength are 
not convertible terms, it must be understood that they 
are entirely dependent upon each other. The maximum 
strength of any given yarn depends upon a certain degree 
of elasticity, and this in its turn depends upon the char- 
acter and number of the twists put into the yarn. Confin- 
ing our attention to mule yarn it will be seen that a less 
number of twists than what is considered normal will 


Fio. 3. 

increase the liability to lengthen when pressure is applied, 
but such a reduction in twist will weaken the yarn, and, 
therefore, a considerably less pressure will cause rupture 
or slippage. Consequently nothing is gained by this 
procedure in the way of strength. It happens, however, 
that strength is not the all-important factor in some classes 
of yarn. A yielding thread is often desired to be used 
in material or for pur2:)oses where it is not subjected to 
forces that will cause rupture, so that we find large 
quantities manufactured to serve such special conditions. 

On the other hand, an unusual degree of hardness in 
the yarn is sometimes desired, and in such a case elasticity 


is sacrificed, and extai twists put in the yarn. It must 
be borne in mind, though, that extra twist means additional 
strains on the fibres, and these naturally are a source of 
weakness ; but since circumstances demand hard twisted 
yarn, it is necessary to make it. 

In further consideration of the subject it will be noted 
that between the two cases mentioned above it is possible 
to obtain a yarn with a maximum strength combined with 
such a degree of elasticity as to satisfy the best conditions 
of the two factors. Exactly at the moment when rupture 
takes place the yarn should cease to stretch, and, simul- 
taneously with this, slippage of the fibres over each other 
ought to begin. Under these circumstances a standard yarn 
would be produced. It need scarcely be remarked that our 
present knowledge absolutely prevents such a high degree 
of excellence in the making of yarn, partly from the fact 
that the cottoft fibre is an ever-varying element, and also 
that little, if anything, has been done in the way of in- 
vestigation into the best means of obtaining a basis upon 
which to work. 

The following table, taken from The Textile Mercury. 
will give some idea of the elasticity of yarn : — 

For Nos. 20 to 30 
„ „ 30 to 40 
„ „ 40 to 60 
„ „ 60 to 80 
„ „ 80 to 120 
„ „ 120 to 140 
„ 140 to 170 

4'5 to 5 per cent 

4-0 to 4-5 „ „ 

3-8 to 4-0 „ „ 

3-5 to 3-8 „ „ 

3-0 to 3-5 „ „ 

2'5 to 3'0 „ „ 

2-0 to 2-5 „ „ 

In measuring the diameter of yarn it is often over- 
looked that a maximum diameter and minimum diameter 
may exist at the same part of any given section, and yet 
if this were used for a basis upon which to obtain an 


average diameter, absurd results would follow. To use 
the example of a twist drill, it is })alpably iucorrect to 
estimate its diameter from the average of its least aud 
greatest diameters. Paradoxical as it may seem, its 
average diameter is certainly its greatest diameter. This 
comes about because the greatest diameter is uniform. 
In yarn the greatest diameter is not uniform, consequently 
the average diameter in such a case must be obtained from 
a large number of measurements of the larger diameters 
obtained from sections in which the least dimensions at 
those points can also be observed. 

Rule for the Diameter of Yarn.— -/^7-- = dia. in 

inches. This rule is, of course, based on finding the volume 
of a certain Aveight and length of yarn and then calculating 
the diameter. 

The Principle of the Spinning Action in the Mule. 
— In the mule, as in all spinning machines, the characteristic 
action is that employed for putting the twist into the roving. 
" Spinning " is the general name applied to this action Avhen 
the amount of twist is in excess of that required to strengthen 
the roving so as to enable it to be taken from one process 
to another : in other Avords, spinning transforms the loose 
fibrous roving into the finished yarn. There are several 
important methods of performing this operation, Avhich Avill 
receive attention sul)sequently. The one noAV to be dealt 
Avith is that applicable to the mule. 

It is an exceedingly simple operation in itself, but, as 
Avill be seen later, the mechanism necessary to perform it 
automatical!}^, and the actions associated therewith, are ot 
a very complicated character. It Avill therefore be ad- 
visable to explain first the principle underlying the action 
of tAvisting, and afterAvards to deal Avith the various features 
connected with and dependent upon it. 

A'OL. Ill C 


In effect, the twists are put into mule yarn In' first 
winding it upon a thin steel spindle, and then drawing it 
off from the end. This results in giving one twist for 



Fio. 4. 

every turn the yarn has heen Avound round the spindle. 
The accompanying sketch fully explains the action. At 
A, Fig. 4, a spindle is shown with yarn wound round it 
a number of times. If the end at 1 be drawn off, the 
portion previously on the spindle will appear as at C, the 

Fig. 5. 

number of twists corresponding to the number of the turns 
of the yarn at A. That this is so can readily be seen by 
inspection of the sketch at B, which is exactly like A, but 
with the spindle removed ; if B is straightened it will 



appear twisted as at C. This exaniplo serves to demon- 
strate the effect of drawing yarn from the end of a spindle 
after it has been wound thereon. In the mule this action 
is taken advantage of, but in an improved and modified 
form. In the first place a method is adopted of ■winding 
the yarn on the spindle and unwinding it in such a 
manner as to obtain a continuous action for a long length 

Fig. 6. 

of yarn. To a casual observer it appears as a single 
operation, but in reality it is composed of two distinct 
actions. This is absolutely necessary if the yarn is to be 
twisted ; it must first l>e wound on the spindle and after- 
wards drawn off, as was shown in Fig. 4. On refei'ence 
to Fig. 5, a spindle B is placed at i-ight angles to the 
source A from which the yarn is deliveied ; if I> is revolved, 
it is clear that the _yarii will be wound on at C oidy, and 
Avhen it is desired to t^vist it by drawing it off from the 


end of B, A must be removed to A^. This, of course, is 
impracticable, but the same effect is obtained by 2)lacing 
the point of delivery A above the point at which the yarn 
passes to the spindle, as in Fig. 6 ; in this position the 
yarn is not wound on at right angles to B, bat by virtue 
of its inclination to the spindle its tendency is to assume a 
position at right angles ; in doing this it naturally rises up 
the spindle in a series of spiral turns, each turn bringing it 
more into the desired position, which would be at D if the 
spindle Avere sufficiently long. 

It is in connection with this feature that the character- 
istic of mule-spinning is seen. If the end of the spindle B 
is arranged to be below the point D, there will be no 
interference with the tendency of the yarn to rise to that 
point as the spindle revolves, and consequently when the 
end of the spindle, at 9, is reached, the yarn continues its 
upward course, and naturally slips off the end and insta.ntly 
drops to 8. This is equal to having one turn off the 
spindle-point, and that turn of course puts one twist in 
the yarn between the spindle and A. As the spindle 
continues its revolution another turn is wound on from 
8 to 9, by virtue of the tendency to reach D, and another 
slippage over the sjDindle-point takes place. This goes on 
until the desired number of twists have been put in, after 
which another operation comes into action. (An interest- 
ing experiment to illustrate this explanation can be made 
by winding a thin narrow tape on the spindle and noticing 
the effect as it winds itself up the spindle and slips over 
at the end.) 

It has just been stated that the 3'arn must not be 
allowed to pass to the spindle during the twisting process 
at right angles to the axis. To prevent this, the nip of the 
front roller at A is placed a])ove the spindle-point, and still 


further to improve matters, as Avell as to prevent the 
vertical distance between the points heing unduly large, 
the spindle itself is inclined. 

So far it has been assumed for the purpose of explana- 
tion that the twists are put into a fixed length between A 
and 9 on E, but this is only partially true. The spindles 
during the twisting operation are caused to move slowly 
away from the front rollers, which at the same time 
revolve and deliver almost sufficient roving to compensate 
for this movement. As the spinning continues while the 
spindles move from P to P^ (Fig. 8), the full length of the 
yarn between the points has the twists comparatively well 
distributed. As an aid to this distribution of the twists, 
the vibratory motion given to the yarn as it slips over the 
spindle-point is rather important; the shaking which it 
receives in this Avay causes the twists to assume a perfectly 
natural position in the yarn, instead of being instantly 
fixed at the point where the twist Avas given. A further 
and highly characteristic feature is also to be observed as 
the movement of the spindles takes place. The slight 
excess of the traversing movement of the spindle over the 
amount of roving given out by the front rollers causes a 
little stretching to take place in the yarn ; the tension to 
which it is in this Avay suljjected causes the thicker and 
softer portions to be drawn out, and, as already explained, 
this tends to equalise the twist, which would otherwise 
leave the thicker parts with a less proportion of twists 
than the thinner portions receive. 

Inclination of Spindle. — From Fig. 8 the influence 
of the inclination of the spindle can also be observed. If 
the spindle were vertical, as in Fig. 7, its inclination with 
the yarn near the rollers at A would probably be enough 
for spinning easily ; but when it reaches its extreme out 


ward position, at B, tlie angle has been considerably reduced 
■ — to almost a right angle — and the slippage of the yarn over 
the point would not he so easily performed — especially con- 
sidering the vibratory motion of the yarn which might 
readily cause the two to be momentarily at right angles, in 
which case spinning would cease and the ends would break. 
By inclining the spindles, suitable conditions exist through- 
out the traverse of the spindles, and although the angle 


is reduced it is still considerable, and the positions of 
the points at right angles to the spindle in the extreme 
positions at B and C are so high above the end of the 
spindle that there is no danger of non-slippage of the yarn. 
The Taper of the Spindle. — The taper of the spindle,' 
as shown in Fig. 6, is due ])artly to the fact that this 
form enables the cop to be readily withdrawn, but primarily 
because as fine a point as is consistent with rigidity is 
necessar}'' in order to get the best result in the slipi)age of 
the yarn over the end. If the end be thick, slippage 


would be bound to take place ; but it Mill be seen that the 
one turn unwrapped from a large diameter would cause a 
slackness that would be inconvenient in several Avays : the 
slackened yarn might run into snarls, or disturl) the tiu'ns 
that are on the spindle just below the point, and thus intro- 
duce variations that would destroy the value of the lesult ; 
excessive vibration might also be easily caused ; a quarter 
of an inch diameter of spindle gives three quarters of an 
inch of yarn in one turn, and this being set free at the rate 
of 5000 to 10,000 times a minute is not likely to prove 
beneficial, consequently the ])oint is made much thinner 
than the bodv, and for A'ery fine work it is frequently only 
a little over one-sixteenth of an inch in diameter. 



General Arrangement. — Before giving a description 
of the mechanism of the self-acting mule, it will be advis- 
able to briefly point out the disposition of the various parts 








1 C 

^ 1 


1— 1 



1 C 










which go to malce up the complete machine. For this 
purpose a sketch plan is given in Fig. 9, which is some- 
what similar to the diagrammatic re])resentation usually 
shown on mill plans. Its main features consist of the 
headstock A, which contains practically the whole of the 
mechanism, and from which point the machine is driven : 
extending for some distance on either side of the headstock 
is a strong wooder structure C, called the carriage, which 



canies the spindles, fallcr rods D, etc. The creel E and 
rollers F are arranged ])arallel to the carriage and extend 
in a similar manner on each side of the headstock A. The 
ends of the machine are terminated by a frame B firmly 
bolted to the floor. The accompanying illustration, Fig. 
10, will noAv enable a general description of the mule to 
be given. It represents a section through the essential 
parts of the machine, and from it an outline of its action 
can be obtained. 

The bobbins A are taken from the last passage of fly- 
frames and placed in the creel at the back of the mule ; 
from here the rovings are guided over wires and passed 
through three lines of rollers which are arranged to give 
it a suitable draft. From the front rollers it is now led on 
to the spindles, and after receiving the requisite amount of 
twist it is wound on in the form of a cop. 

The headstock is a strong framework consisting of two 
frames similar to that shown in Fig. 11. The two 
portions are firmly connected by cross pieces, and within 
the rectangular structure thus formed the mechanism is 
placed. This mechanism is of a very complicated character, 
and in the descriptions of the various actions that take 
place during a cycle of operations repetition will be un- 
avoidable and in many cases necessary. This is rendered 
more so by the fact that most of the actions are directlj' 
connected, or depend upon each other for their performance 
and in several instances are working simultaneously. 

When the carriage commences the twisting operation it 
is brought as close to the rollers as possible, the spindles 
occupying the position shown at L ; this distance is usually 
from 3 to 5 inches. As already explained, the twisting 
continues by causing the spindles to revolve at a rapid 
^atc, and at the same time moving them gradually away in 




tlie direction of tlic arrow, until they arrive at ]\I ; when 
this position is reached the spindles cease twisting, an 

action called "hacking off" comes into plaj', and im- 
mediately following this the carriage begins its return 
journey to the rollers ; whilst this is heing performed the. 


yarn which Avas twisted during the outward run is wound 
on the spindle. 

The distance traversed by the carriage from L to M is 
termed the " stretch " ; its length varies for different 
purposes, ranging from 48 up to as liigh as 68 inches, the 
most usual length, however, being about 64 inches. A 
" draw " is generally understood to mean one complete 
action, i.r. from the commencement of s})inning Avhen the 
carriage is at L to its return to the same position after the 
"outward run" and the "run in." If a mule works, say, 
four draws in one minute, it means that the carriage has 
started from L and returned to it four times in the course 
of one minute ; in other words, the machine has gone 
through the whole of its actions four times in sixty 

The carriage is mounted on a series of bearings X sup- 
ported by bowls I and H ; they are placed at suitable 
intervals along the length of the carriage and run on iron 
rails P. The spindles are driven by the tin cylinder F 
carried Ijy the carriage. This arrangement, however, only 
drives the spindles whilst twisting ; when winding, or 
building the cop takes place, they receive a special motion. 
The cop is formed through the medium of the wire T carried 
by a lever, centred on the copping faller K, and during 
the building, tension is maintained in the yarn by the wire 
S carried by a similar lever, but which is coiniected to 
the shaft J and called the counter-faller. All the above- 
mentioned features are carried by the carriage and will be 
dealt with subsequently in detail, and fully illustrated. 

As will be observed from Fig. 9, mules are Avorked 
in pairs, arranged so that they can be attended to by one 
set of Avorkers. The spindles of each mule approach each 
other ill their outAvard run to Avithin such a distance as Avill 


permit of freedom of movement for tlie workers, wlio in 
the course of their duties pass to and fro along the passage 
between the faller rods D of each mule. 

In order to convey an idea to the reader of the mechanism 
of the headstock, or at least the general features of it, 
in plan view, an illustration is given in Fig. 11. The 
principal driving of the machine takes place through the 
pulleys H, G, driven from a counter shaft above ; from 
the same counter shaft is driven the "drawing-up" pulley 
a by means of a band. The spindles v are driven by the 
rim pulley D through t, and the tin cylinder u ; the carriage 
is actuated by means of strong bands through the scrolls 2, 
3, 4, and 5. The gearing for the driving of the rollers can 
be readily traced from the wheel J on the rim shaft. The 
turning of the spindles for " winding " during the inward 
run of the carriage is produced by means of the quadrant, 
a chain from which passes over the winding drum and 
transfers the motion through the wheels z and x to the tin 
cylinder and on to the spindles. A reference to this 
drawing in connection with the further descriptions that 
will be given, will be of great assistance in exi^laining 
much that might otherwise appear vague. 

The Creel. — Although the arrangement of the creel is 
not of much importance as a detail of a machine, yet it 
ought to be noticed, especially in connection with the mule. 
There are obvious advantages to be gained by giving to the 
bobbins a disposition that will economise space, and save 
time in filling the creel and in keeping, them at a suitable 
height adapted to the workers who attend tin's feature of 
the machine. 

Figs. 12 and 13 illustrate a variety of methods of form- 
ing the creel, and plan views are also shown. In Fig. 12 
the usual Bolton system is given ; single rows of i-ails are 




employed, -which saves space, but it necessitates half and 
full bobl)ins -with douljle rovings. Alternate arrangements 
for all full l:)oljbins, with four heights, are shown, in which 
two rails are used, and also an arrangement with a broad 
single rail, the bobbins being arranged in zig-zag order. 
Three heights of bobbins for single rovings are illustrated. 

Fio. 14. 

An exceptional method is illustrated Avhich is only adopted 
when circumstances prevent the application of the other 

The creel itself is built up on a series of upright rods, 
firmly fastened to the spring pieces which carry the roller 

Driving" the Mule. — Owing to the fact that the two 
headstocks of a pair of mules are always placed out of the 
centre of their respective lengths (see Fig. 9), the driving 




belt is often so much inclined us to necessitate a slight 
alteration in the arrangement of the end of the creel at the 
headstock. Such an alteration is shown in Fig. 14. It 
would clearl}^ ])e impossible to have B straight up, as at A, 
on account of the dri\dng belt C ; therefore a method 
similar to tliat illustrated is usually adopted. 

Fig. 15. 

A general idea of the main driving of the mule can be 
obtained from Fig. 15. The end view is taken from the 
back of the machine, and shows all the shafts ^n section. 
It will be seen that the line shaft or main driving shaft 
is at right angles to the direction of the carriage length. 
In all new mills this shaft runs from end to end, and is 
driven direct from the engine ; the various counter shafts 
for each machine are independent of each other, l)ut all 

VOL. Ill D 


are driven from the line shaft in a manner similar to that 
shown in the drawing. The pulley A drives B on the 
counter shaft; a separate pulley C on the counter shaft 
drives D on the rim shaft. The driving pulley A is 
made double the width of the belt, so that to stop the 
mule all that is necessary is to move the strap on the 
loose pulley at B ; this completely stops the whole machine. 
Owing to the alternate motions of the mule, it is necessary 
to continue the working of some parts whilst others are 
stopped ; this is effected partly by means of a fast and 
loose pulley on the rim shaft, and also by the employment 
of clutch cones and wheels that at^e put. into and out of 
gear at their correct times by other parts of the moving 

It has been remarked that the principal driving of the 
self-actor is performed through the driving belt from C to 
D. Formerly this belt supplied the entire machine with 
its motion, but within the last few years an important 
change has taken place by transferring a portion of its strain 
to a supplementary driving arrangement by means of a band. 
This is shown in the sketch. Fig. 15. E is a band pulley 
on the counter shaft, and drives the pulley F on the 
drawing-up shaft. Its principal function is to produce the 
inward run or drawing-up of the carriage ; several other 
important actions are effected also by it, which ensure a 
more perfect working of the machine than in the old 
system, where the whole work was thrown on the driving 
belt. The liability of the drawing-up band to stretch is 
compensated for by means of a tightening pulley G, which 
ensures a regular tension. The above general description 
is given so that the more detailed descriptions of each action 
which follow will be better understood, and the illustrations 
also will be extremely useful for reference, as it is clearly 


impossible in illustrating such a complicated machine to 
show more than one or two motions in a single sketch. 

Although Fig. 15 shows the line shaft at right angles 
to the carriage, and thus brings the rim pulley at the back 
of the headstock, it ought to be remarked that this is not 
invariably the practice. It sometimes happens that, owing 
to the formation of the mill or the necessity for having 
the shafting fixed in a certain position, the line shaft is 
placed parallel to the length of the carriage. When such 
is the case, the rim pulley is arranged at the side of the 
headstock, and by very little re-arrangement of gearing all 
the other motions work in the same way as when the rim 
is at the back. 

Movement of the Carriag'e. — We will now consider 
the question of how the carriage is moved during its outward 
and inward run. The remarks previously made will have 
demonstrated that there are two distinct actions, namely, 
spinning and winding — spinning when going out and wind- 
ing when coming in — and for each of these the motion of the 
carriage undergoes a change of speed. It is perhaps neces- 
sary to explain the reasons for such a change of speed. The 
motion of the carriage during the operation of twisting is 
clearly dependent upon the numl^er of twists required to be 
put in a given length of the yarn ; the quicker the twists can 
be put in, consistent with the character of the cotton and the 
perfect working of the automatic actions associated with it, 
Avill provide a foundation in olitaining the speed of spindle ; 
and this speed, when decided upon, regulates the speed 
of the carriage. From these considerations it is an easy 
matter to reason in a general way that the lower the 
counts spun the quicker the speed of the spindle ; and, as 
lower counts have less twist than the higher counts, it 
follows that the speed of the carriage is quicker for low 


counts than for high counts. It is also not difficult to 
understand from what has been already said that the 
twisting operation is necessarily slow. When, however, 
the spinning is completed, and winding on begins, there is 
nothing to prevent as quick a return as possible to the 
roller beam. We therefore find a wide difference between 
the two motions of the carriage, and moreover they are 
performed by two distinct actions of the mechanism. 

To convey an idea of the difference of the time, an 
exam})le is given as follows : — Suppose a mule is found to 
complete its whole cycle of operations three times over 
in 54 seconds, this would give 18 seconds for each draw, 
i.e. an outward and inward run. Of this 18 seconds there 
would be about 4| seconds in which the mule would back 
off and run in, thus leaving 13i seconds for the outward 
run during which spinning is taking place. In this time 
the carriage has travelled 64 inches, and, in order to put 
the right number of twists in the yarn, the spindles must 
run at the rate of 9000 revolutions per minute without 
allowing for slippage of the bands. This gives us a good 
conception of the comparative sj)eeds of the chief working 
parts, so we can now proceed to examine the methods 
adopted for obtaining them. 

In order to fully appreciate the methods adopted in 
moving the carriage, it is as well to thoroughly understand 
the reasons for their adoption. In the first place the 
carriage is very long, and consequently heavy ; if it 
contains 1000 spindles of If inch gauge its length will 
probably be about 120 feet. To move this long heavy 
mass, Avhich includes the faller rods and all their connec- 
tions, the spindles, tin drums, square, the framework of 
the carriage and its bowls, etc. etc., is of itself a difficult 
matter ; but when this heavy mass keeps stopping and 


starting, it is still more difficult to regulate its movements 
so that it may commence smoothly, and also finish without 
any abruptness. The problem is solved, however, by the 
introduction of Avhat are technically called "scrolls." 
These are a kind of drum in the form of a spiral, and of 
sufficient length to wind on the requisite amount of band 
for the stretch. The small diameter with which they 
commence enables a very slow motion to be given to the 
carriage on the commencement and finish of its stretch, 
whilst the intermediate portions of its movement are much 
quicker; abruptness of actions and its consequent strains 
are by this means avoided. The above remarks are 
general to the two movements of the carriage, but are 
specially applicable to the inward run. During the 
outward run the carriage moves very slowly, but the 
inward run being much quicker, both the commencement 
and finish are made as slow as possible. 

The outward run is obtained direct from the front roller 
through a train of wheels to the back shaft. Fig. 16 
illustrates this connection ; it is an enlarged view of a 
portion of the general gearing plan given in Fig. 11. The 
motion in the first place is received from the rim shaft 
through the wheel J ; from here it passes through the 
compound carrier K L, and to the back change wheel or 
speed wheel C. A bevel wheel E. conveys the motion to 
the front roller bevel S. Connected to S by means of a 
clutch-box is a wheel T, and from this wheel through the 
wheels 0, E, P, and Q, the back shaft is driven. The 
speed of the carriage is of course directly related to that of 
the front roller ; any required change between the two 
speeds is readily obtained by changing the pinion P, and a 
further change, in which both front roller and carriage will 
be altered in speed, can be made through the Avheel C, and 



sometimes by changing L and K. These speeds and the 
calculations connected with them will be dealt with under 
the head of " Calculations " when we reach that part of the 

Fig. 16. 

The arrangement for taking the carriage out by means 
of the back shaft is shown in the three illustrations : 
Figs. 17, 18, and 19. Bands j)assing over and around 
the scrolls are fastened to the carriage either at the back, 
front, or ends ; the revolution of the shaft acting through 
the bands draws the carriage either outwards or inwards, as 



the case may be, this of course depending on the direction 
of the rotation of the shaft. 

As a rule there are five scrolls in the l>ack shaft ; one, 
A, is connected by band to a large scroll 2 (see Fig. 17) 
on the scroll shaft. B and B are placed each about half- 
way between the headstock and the ends of the machine. 
One is also placed at each end, as at F and F. The 
method of connecting the bands to the carriage is sho^^'n 
in Fig. 17; but to make it more clear, drawings ai'e given 
in Figs. 18 and 19, which show the attachment very 

distinctly. In Fig. 18 the scroll B is represented as 
drawing the carriage out ; this it does by means of the 
band F, which passes under the carriage, over a guide 
pulley D at the front of the mule, and from here is fastened 
to the carriage at E; its motion in the direction of the 
arrow draws the mule out. The same drawing also shows 
that if the direction of motion of the scroll B is changed, 
the carriage can be drawn in through the band G, whicli is 
also fastened to the carriage. It must clearl}'^ be under- 
stood, however, that the motion of the back shaft for 
performing the " outward run " is obtained directly from 
the front roller, and the mo\ ement it gives to the carriage 



is a very regular one, except at its commencement, when 
the band is working on the small diameter of the scroll 
l>art at H, Fig. 18. The mule at this point is close to the 

Fig 18. 

roller beam, and stationary, and consequently the movement 
of the heavy mass must be brought about slowly. This is 
effected by making a short S})iral at H for about half a 
revolution before attainina; a maximum diameter at B. 


Fig. 19. 

When this s])iral portion of the drum is passed the 
remainder of the stretch is performed at a luiiform speed 
by the straight portion of the drums. 

The ends of the cai'iiage are moved in the manner shown 


in Fig. 19. B is the scroll, corresponding to F in Fig. 
17. One of the bands H passes from B over a carrier 
pulley C, and a loose stud D, and is fastened to the carriage 
end at F ; the other band, J, simply passes over D, and is 
then fastened at E. The revolution of B in either direction 
will produce a similar movement of the carriage. The 
direction shown by the arrows is the "outward run" 
during the spinning process. 

When the carriage has reached the end of its outward 
run, an action called " backing-oif " takes place, and im- 
mediately afterwards the inward run commences. As 
already described, this inward run is performed very quickly. 
The comaection of the front roller with the back shaft is 
broken by disengaging the clutch. Fig. 16, which leaves 
the back shaft free to be driven from another source, namely, 
the scroll shaft. 

The scroll shaft is driven through bevel wheels from 
the drawing-up shaft, see Fig. 11. On it are keyed four 
large scrolls, three of which are used in drawing the carriage 
in (Fig. 17), Nos. 3 and 5 are directly connected to the 
carriage to serve this purpose, while No. 2 is connected to 
the back shaft by a band on the scroll A. The whole back 
shaft is thus utilised for the inward run as well as for the 
outward run, its direction of revolution of course being 
reversed to enable the latter operation to be performed. 

The fourth scroll, called the check scroll, is introduced 
in order, as its name implies, to check any irregularities of 
movement that may be caused through the varying and 
quick motion of the carriage during the inward run. Its 
effect will be better understood l)y comparing its position 
and action with the drawing-up scrolls 3 and 5. In Fig. 
20 the scrolls Nos. 3 and 5 are shown attached to the 
carriage, being represented as drawing it in. When the 



band is on the smallest diameter the speed is slow, but on 
the large diameter it is quick, and attains its maximum 
speed on the largest diameter, and then begins to decrease. 
It is, however, quite possible that after the carriage has 
attained its quickest speed its momentum will compel it to 
continue at a slightly greater speed, for a moment or so, 
than that of the scroll. This is a contingency that must 

Fig. 20. 

No.3 & 5 

■f///^^////^^/^/y/^//y'///'//// ^. 


Fig. 21. 

be avoided, as it might lead to disastrous results. The 
check scroll is therefore arranged to efiect this, and Fig. 
21 illustrates the arrangement. It will be noticed that 
its position on the shaft is opposite to that of the other 
scrolls, and that its band leads off" from its lower side, and, 
passing Tuiderneath the carriage, is carried over a guide 
pulley G and connected to the front of the carriage. Now 
it is quite clear that any tendency of the carriage to over- 
run the scrolls 3 and 5 will be counteracted by scroll No. 4, 


because overrunning would result in tightening the Ijand 
of the check scroll : in other words, this scroll serves the 
purpose of a drag on the carriage the moment it varies 
from the speed of the drawing-up scrolls. 

There is a very important feature in connection with 
the various scroll bands that have been mentioned, which 
ought not to be overlooked. It will be observed that one 
very essential condition of the successful working of the 
mule is the necessity for maintaining the carriage perfectly 
parallel to the rollers. To maintain this requires in the 
first place a strong carriage to resist flexure, and the faller 
rods must be strong also as an aid to this condition \ but 
the most important feature is the connection of the various 
bands to the carriage. Bands are at the best an uncertain 
element, so everything must be done in choosing only the 
very best bands and compensating in every way their 
tendency to stretch and to take up the extra length they 
acquire through the strain to which they are subjected. 

The attachment of the bands to the carriage becomes 
therefore a very important factor in good Avork. Special 
ratchet arrangements are applied at the vaiious points, so 
that an adjustment as fine as experienced judgment will 
allow can be attained. Frequent adjustment is necessary, 
for the bands are very irregular in their stretching qualities, 
and it is a serious matter if the carriage be allowed to vary 
from a straight line during its traverse. The yarn coming 
from the rollers will in such a case be irregularly stretched 
or drawn during the outward run, and on the inward run 
its winding on the spindle will consequently be unequal at 
various parts of the mule. When the carriage finishes its 
inward run, it ought to do this simultaneously throughout 
its whole length, coming against all the back-stops at the 
same moment with a smooth silent finish, and not abruptly. 


occasioning noise and shock, which would result in faulty 
yarn in the form of snarls or broken ends. 

The extra long mules now made render close attention 
to the bands imperative. The carriage as now made is 
constructed on lines that reduce its flexure to a minimum ; 
at the same time it is sometimes mounted on bowls that 
work on friction rollers, and the same feature is introduced 
in some cases for the faller I'ods and even for the back 
shaft. Everything, in fact, is done to prevent torsion and to 
preserve a perfectly straight line through the centre of the 
spindles, and also to maintain this line absolutely parallel 
with the front roller throughout the traverse of the carriage. 

The adjustment of the bands just described is generally 
termed "squaring the mule," but "squaring band" is a 
name that is given to a special band which is used to 
obtain the movement of the end of the carriage equal to 
that of the middle part or square. It is illustrated in Fig. 
17, but a detailed reference will be made to the diagram 
Fig. 22. Half the length of the carriage is shown, each 
half having its own bands, and the band is placed under- 
neath it. Two bands are used, L and M. The band L is 
fixed at one end at a suitable spot E, and passes round the 
pulley C and D, the other end being fastened at F. A 
similar thing is done with the band M, but in the reverse 
order. Both bands are used for the same purpose, L for 
the outward run and M for the inward run ; so we refer to 
L in the explanation. If the carriage be draAvn outwards 
in the direction of the full arrow, a tension will exist in 
the band L as if it were being stretched in the direction 
shown. Now since the band passes from E to F, the same 
tension will exist in the band throughout its length, and by 
following it through Ave shall find that an effect is produced 
as if some force were pulling the band at F, in the direction 



shown. This has clearly the effect of jnilling the eiul of 
the carriage out in the same direction as the middle, and 
with an equal force. The squaring band is therefore an im- 
portant element in the "squaring "of the mule ; but, like 
the other bands, it is necessarj' to keep a constant watch to 
see that it does not become defective for want of adjustment. 
It will be interesting at this stage to devote a few words 
to a description of a "drawing-up scroll." The essential 
conditions to be fulfilled by such a scroll are — as slow a 
movement as possible at the commencement of the inward 

•e H 

. A 




Fio. 22. 

:q Fi 

run, and a similar finish when the carriage reaches the back 
stops ; the movement of the carriage between these two 
positions depends upon the number of revolutions given to 
the scroll shaft, and the length of the stretch. These facts, 
of course, decide the maximum diameter of the scroll, and 
the maximum diameter, in its turn, decides the intermediate 
speeds between the start and finish of the "run-in." 

It is umiecessar}' to exj)laiu the method of obtaining the 
size of a scroll that will serve for any given stretch ; but 
Ave will suppose the scroll has been designed, and that 2i 
revolutions of the scroll shaft are sufficient for the pur])0se. 
This means that durini' the inward run the scroll must 


make 2| revolutions, and in doing so must wind on the 
band by which the carriage is drawn in. In order to 
obtain the commencing slow movement it is necessary to 
commence winding on a small diameter, as at B, Fig. 23 ; 
the diameter is then gradually increased by making the 
drum of a spiral form, until the largest diameter is obtained 
at C, where naturally the greatest speed is given to the 
carriage, which on examination of the diagram is found to 
be halfway in the stretch ; from this point a reduction in 
speed takes place by a corresponding curve to the first half 
of the scroll, and it finishes on the same diameter as that 
on which it commenced. 

To show the variation in the speed given to the carriage 
during its run, two portions of the scroll have been marked 
off. At B G a length is shown which represents the 
amount of band wound on during the first quarter of a 
second of the run in, while at the middle of the stretch the 
length wound on during the same time is shown at F E. 
The intermediate lengths could be easily shown in the same 
manner, but a better method is given in Fig. 24. The 
movement of the carriage for each quarter of a second is 
there shown ; starting at A it would move to B in the first 
quarter of a second ; each successive qiiarter Avould find 
the carriage at C, D, E, etc., until it had completed its 
journey at K This diagram shows very distinctly the 
varying movement of the carriage ; to those, however, who 
are interested in the matter, the diagrams in Figs. 25 and 
26 will convey a much clearer idea of how the movement 
of the carriage is controlled. In Fig. 25 the straight 
lines D, C, B, show the development of the curve of the 
scroll, and the fact that straight lines represent such a 
development tells us that the carriage starting at B has a 
regularly increasing movement given to it until it reaches its 



greatest speed at C, from which point it at once begins to 
decrease to I). The point to ohserve in this diagram is 
tliat the change of speed, whether at the start, middle, or 
end, commences at once. Some authorities condemn tliis 
method and find much better results given by forming the 
scroll so as to give a movement as represented in Fig. 26. 
Here, instead of commencing to increase regularly, the 
initial slow movement of the carriage is continued a little 

Fig. 23. 

1 2 3 4 5 6 7 

9 10 11 12 13 



Fig. 24. 


longer, and then gradually increased to a regular acceleration 
until near the maximum at C. Here the sj^eed is maintained 
a moment or two longer, and then a more gradual reduction 
is made to the decreasing speed than in the case of Fig. 26 
until D is reached. 

In Fig. 17 was shown a system of arranging the scrolls 
which up to a few years ago was generally followed. At 
the present time, however, one or two important firms 
have arranged their systems on a slightly different plan. 
Instead of the scrolls Nos. 3 and 5 being placed so far 


apart and independent of each other, they are brought 
closer together, and one band only is used for the two. 
This band, instead of being fastened to the usual ratchet- 
tightening arrangement on the square, goes from one scroll 
and passes round a horizontal carrier pulley, or round 
fixing, on the square, and from there back to the other 
scroll. The object of this is to obtain exactly the same 
tension in the band of each scroll, AVe have seen how 
important a matter this uniformity of tension is, and it 

will be admitted that this method is an excellent one for 
attaining it. Of course it is necessary to keep the carrier 
pulley itself adjusted as the band becomes slack. Although 
having obvious advantages, it is open to question whether 
this method is superior to that illustrated. The pull, 
taking place at what is practically one point, is bound to 
be inferior in effect to that of a pull at two points as far 
apart as possible on the rigid part of the carriage called 
the square. Unequal wear that may take place in the 
band will lead to more waste and loss of time than in the 
old method, and the new one is under a distinct dis- 



advantage when, as sometimes happens, the band breaks 
and the breakage is not immediately noticed ; serious 
results in such case would certainly follow. Under the 
old system, when one band breaks the other band will 
prevent any mishap occurring until the minder discovers 
it and effects a remedy. 

Driving the Spindles. — Fig. 27 illustrates the method 

Fig. 2S. 

adopted in driving the spindles. They are driven from the 
rim shaft through a large band pulley D ; the band passes 
down behind the headstock over a fixed back carrier pulley, 
and on to another carrier pulley E ; from this it passes 
round a band pulley B on the tin drum shaft through 
which the spindles are driven. Continuing, it goes forward 
to the front of the headstock and over a carrier pulley F, 
by which it is guided on its return journey, and passing 
over another back carrier reaches the rim pulley. The 
VOL. Ill F. 


back view of the mule is given iu Fig. 28, and the 
positions of the back carrier pulleys E and S show the rim 
band guided in a direction at right angles to that in which 
it leaves the rim pulley. 

The revolution of the rim shaft in the illustration is in 
the direction shown by the arrow, but it is not necessarily 
so in all makes of mules ; some have the rim running the 
opposite way, and with an arrangement of the driving of 
the tin cylinder as represented in Fig. 29. There is 
practically no difference between the two methods, the 
wear, strain, and length being about the same in each case. 

Although only a single grooved band pulley is shown 
in the sketch, this has merely been done to simplify the 
drawing. On the mule tAvo or three grooved pulleys are 
used, and the band is consequently twice or three times 
the length represented in thq sketch. A long length of 
rope, such as this, is subject to a considerable amount of 
stretching ; and, especially when it is new, some attention 
must be given to it to keep it at a uniform tension. The 
carrier pulley F is fixed in a slide, which can be readily 
adjusted to compensate for any stretching that may take 
place. Unless the rim band be kept well to the grooves 
of the rim and the tin cylinder pulleys, considerable 
slippage is likely to occur ; even under the best conditions 
some slippage is unavoidable, but neglect in keeping the 
band tight leads to very serious faults in the yarn. Every 
care should therefore be taken in attending to this feature 
of the mule. The tin drum or cylinder, extending the 
full length of the carriages, drives each spindle by means 
of a short length of banding, Avhich passes round a small 
pulley on the spindle, called a Avharve. The direction of 
rotation of the spindle can be varied by a change in the 
crossing of the band from the cylinder ; a change in its 


speed is brought about by changing the rim pulley D and 
replacing it b}' a larger or smaller as the case requires, 
the end of the rim shaft being arranged so that this may 
be quickly effected. 

An interesting point to observe in the two drawings, 
Figs. 27 and 29, is the effect of the movement of the 
carriage on the band. In both cases the band, when 
leading on and off, is running in the same direction as the 
carriage. To get an accurate idea of the revolution of 
the cylinder this must therefore be taken into account, for 
there is clearly a loss, which amounts to from 1| to 2 per 
cent in ordinary numbers. This loss of speed must not 

Fig. 29. 

be confused with slippage, because it is due to an entirely 
different cause. As a rule, however, it is included in the 
term "slippage," such term including the difference between 
the calculated number of revolutions and the actual number. 
One chief reason for the employment of a three-grooved 
rim pulley is the desire to reduce slippage to a minimum, 
even when a slightly increased power is the result ; and 
at the present time for good work the three-grooved pulley 
has become general. 

Another interesting feature displayed by the spindles 
is the relative slowness by which they attain their speed 
on commencing the outward run. Theoretically the 
spindles are supposed to commence running immediately 


the carriage starts on its outward run, and at the same 
instant the roller also commences to turn. Careful obser- 
vations extending over a large number of mules show that, 
starting from the beam, the spindles do not attain their 
maximum speed until the carriage has moved 10 to 30 
inches away from its starting point. This accounts for 
much of the irregularities of twist, counts, and other con- 
ditions of mule yarn which aiiect its quality, and it ought 
certainly to be taken into consideration more than appears 
to be done in estimating twist, etc. 

The explanation in a general way is that it is due to 
the enormously high percentage of power required to 
start the mule carriage and the spindles on the outward 
run. In a 1000-spindle mule the power required during 
the first half-second rises as high as 25 h.p., and this is 
developed immediately the strap goes on the fast pulley. 
Such a high power is undoubtedly due to the resistance 
of the carriage and spindles, both being at rest at the 
time. They yield graduallj-, and in doing so a large 
percentage of slippage must take place, especially on the 
rim band, and the spindle bands, and also on the driving 
belt. The carriage itself loses nothing, because slippage 
is almost impossible in its case, but it adds to the general 
disarrangement of the relative movements of itself and the 
rollers and spindles for the first second or so of the run out. 

A very ingenious method of trying to overcome the 
difficulty just mentioned, that of starting the spindles at 
their full speed, has been introduced by a well-known firm 
of machine makers. It is illustrated in the accompanying 
sketch, Fig. 30. The variation in the relati-\'e motions 
has been overcome by what is practically driving the front 
roller and carriage from the tin cylinder. The tin cylinder 
is driven in the usual way, but by a special arrangement 



the same band transmits its motion to the front roller, 
which is therefore not driven in the direct manner by 
gearing, as is usual in other mules. 

On reference to the drawing it will be seen that the 
driving pulleys E are mounted on a hollow shaft B, to 
which the rim pulley also is fixed. Within the shaft 
B another shaft A is placed, carrying at one end a band 
pulley H, and at the other end a wheel I from which the 

Fin. 30. 

front roller is driven. Now as the outward run commences, 
the rim pulley G will be driven. Its band will drive the 
tin cylinder in the manner shown, and on returning to the 
back of the headstock is passed round the band pulley H 
on the inner shaft A A, Avhich it therefore drives at a speed 
equal to E E, but minus any slippage that has occurred in 
the rim pulley G. The wheel I on the shaft A drives the 
front rollers, and from the front rollers the back shaft is 
driven in the usual manner by the train of wheels shown. 

We are now in a position to see the peculiarity of this 
motion and also its advantas:es. The usual method is to 


drive the front roller and back shaft direct from the rim 
shaft : consequently little or no slip occurs ; but since the 
spindles are driven by band, a large percentage of slipjjage 
occurs, especially as the carriage starts out from the roller 
beam, and inequalities of twist of rather a serious character 
are therefore introduced. To neutralise these as much as 
possible the direct method of driving the carriage is dis- 
pensed with, and both spindles and carriage are driven by 
the rim band ; any slippage that takes place in the band 
will now affect each motion, and if the spindles start slowly 
the carriage will also do the same, and in this way prevent 
any inequality of twist that would otherwise occur. It 
must be clearly understood that the "initial" slippage of 
the bands of the mule, which is very great, must not be 
confounded with what might be termed a " general " 
slippage, which must always exist throughout the travel 
of the carriage, and which is sometimes estimated to be 
as high as 5 per cent of the speed of the spindles. No 
band can be kept at such a tension, and in perfect contact 
with its pulleys, to an extent that would actually prevent 
slippage, so something must always be allowed for this 
when dealing -with calculated speeds where bands are 

The Rim Shaft. — Before proceeding further in our 
description, it will be an advantage to illustrate and describe 
those parts of the mule from which the actions already 
mentioned receive their motion. The rim shaft is naturally 
the first point to which attention must be directed, and in 
order to show clearly the disposition of the driving pulleys, 
an illustration is given in Fig. 31, Avhich represents in 
section this important feature. Reference may also be 
made to the sketch, Avhich shows a section through the 
duplex system of driving. 



The rim shaft is generally carried by two bearings, G G, 
which form part of the general framing of the machine. 
On the shaft between these bearings are placed the main 
driving pulleys. They consist of fast and loose pulleys 
B and C ; the fast pulley B is keyed to the shaft, and 
throu'di it the mule receives its chief movements. One 

Fig. 31. 

■ :' scroll"-.. ■■ 

'. •■ SHAFT. .•■ . 

edge of this pulley is extended, and formed with a conical 
surface, upon which is riveted a layer of leather, T ; a 
large Avheel, A, called the backing-off cone wheel, also has 
its outer rim extended and its interior side recessed out 
in a conical form for the reception of the conical part of 
the fast pulley. The large wheel A is not keyed to the 


rim shaft, simply riding loose upon it ; but by means of 
a fork, fitting in the grooved part of the boss at E, and 
levers, the backing-ofF cone wheel can be moved into or 
out of contact with the fast pulley. The loose pulley, to 
which the strap is moved when certain actions are at rest, 
rides loose upon a bush, as shown in the drawing. This 
bush may be either a separate piece or be formed as part 
of the brass bearing which fits in the framing and carries 
the rim shaft. The end of the rim shaft upon which the 
rim pulley is bolted is specially prepared to receive the 
rim and to effect a speedy change when a larger or smaller 
pulley is necessary ; this detail is fully shown in the sketch. 
The place to which the pulley is bolted is, in some mules, 
forged on the shaft and case-hardened, by which means 
the possibility of breakage, owing to the sudden strains to 
which it is subjected, is reduced to a minimum ; at the 
same time, the fact of its forming part of the shaft and 
being turned and finished therewith ensures more perfect 
running, and far smoother driving of the spindles. The 
three-grooved rim pulley is illustrated, as this form is now 
generally used, and is recognised as the best for driving 
purposes; through it the band, which is much longer in 
consequence, maintains a better grip in the grooves, and 
therefore reduces slippage. 

Drawing-up and Backing-off, etc. — There are practi- 
cally two systems in vogue on mules at the present time 
in regard to the "drawing-up" and "backing-ofF" arrange- 
ments. Owing to the great advantages that have been 
found to result from the "drawing-up" by means of a 
separate driving by band or strap, the older form is gi'adu- 
ally becoming obsolete ; but as a very large number of 
mules are working under the old conditions, a brief sketch 
of the arrangement will be given. As a preliminary, it 


must be clearly understood that the loose pulley on the 
mules is not used as a means to stop the mule : this is 
effected in the counter shaft ; therefore the word " loose " 
is only used in a local sense. In the action about to be 
described, the loose pulley performs very important func- 
tions, to which reference Avill now be made. The drawing, 
Fig. 31, can be used to aid the description of the older 
form of "drawing-up" and backing-oft'," the special j^arts 
relating to it being shown in dotted lines. 

When the strap is on the fast pulley B, the backing-off 
cone Avheel A is out of contact with it, and therefore free 
on the shaft. Under the circumstances, all that is fixed 
on the rim shaft will revolve. Two important actions now 
commence, viz. — The turning of the spindles through the 
rim pulley F, which constitutes the spinning process ; and 
the revolution of the rollers and outward movement of 
the carriage, which is effected through the wheel W fixed 
on the rim shaft. These actions continue as long as the 
strap remains on the fast pulley, but after the carriage 
has moved what may be considered the necessary distance, 
say 64 inches, its own movement, acting through levers, 
brings about what are technically called "changes": i.e. 
certain actions are made to cease and others come into 
operation. These " changes " will be described in detail ; 
for the present purpose it is sufficient to mention that one 
of the changes causes the strap to be moved from the fast to 
the loose pulley C, which has the effect of stopping the rim 
shaft, and therefore the spindles, the rollers and the carriage. 

During the time the strap is on the fast pulley, a small 
portion of its breadth is working on the loose pulley, and 
causing it to revolve. This movement is sufficient to make 
the wheel A revolve, because the loose pulley has on its 
boss a wheel K, through which the "backing-off" cone 


friction A can be driven. Gearing into A is a wheel on 
the cam shaft (not shown in the sketch) ; this latter shaft 
has a cone clutch driving arrangement, which is put into 
and out of gear by the carriage. At the termination of 
the outward run, one of the "changes" produced by the 
levers referred to above, puts the cone clutch on the cam 
shaft into gear, and enables the movement of the loose 
pulley to turn the cam shaft and by this means to put the 
roller and back shaft catch boxes out of gear, and thus stop 
the carriage, etc. Immediately the carriage and spindles 
have ceased working, the strap being on the loose pulley 
C, two other important actions are brought into play. 
One is called the "backing-ofF," its object being to cause 
the spindles to revolve a few turns in the opposite direction 
to that in which they revolved when spinning. This un- 
winds the yarn on the spindle, which is coiled between 
the cop and the spindle point. The action is brought 
about by certain levers forcing the backing-off wheel A 
into contact with the conical part of the fast pulley. As 
A is being driven at the time through the wheels K, L, 
M, and J, and in the contrary direction to the driving 
strap, it commences to turn the rim shaft in the opposite 
direction, and so gives the desired movement to the 
spindles. The other action is the "drawing-up" of the 
carriage during the inward run. This, as already stated, 
is the duty of the scroll shaft ; it receives the motion 
enabling it to do this through the wheel K, on the loose 
pulley, acting through the wheels M, N, 0, Q, and li. 

It will be seen that all the movements referred to in 
this description are connected wath one another almost 
directly ; it is only by the careful adjustment in putting 
cone clutches in and out of gear that it is possible to bring 
about the several operations that have just been described, 






O ' 







11 ' r^y^ MODERN MULE 59 

and in order to give a clearer idea of these complicated 
actions the following table may prove useful : — 

When the strap is 011 the fast pulley, during the ontwaid run : — 

The spindles are revolving. 

The rollers are delivering roving. 

The carriage is making its ontward run. 

The "backiiig-oH'" eone friction is out of gear. 

The "drawing-up" friction is out of gear. 

The " backingoff" cone wheel A and the cone dish P on the 
upright scioll shaft are revolving, because a portion of the 
strap is on the loose pulley C which drives tiiem through K. 

Tlie cone clutch on the cam shaft is ont of gear. 

When the carriage reaches the end of the stretch, changes take 
place which have the effect of : — 

/•Putting the cone clutch on cam shaft in gear. 
Moving the strap on to the loose pulley C. 
Stopping the spindles. 
Putting the "backing-off " friction into gear with the fast 

pullt'y, and causing "backing-otf." 
Stopping the carriage and back sha(t. 
"Stopping the rollers. 

When "backing-off"" has finished, 
The cone clutch at P is put into gear, and the scroll shaft draws 
in the carriage. 

The brief analysis just given does not by any means 
exhaust the actions of the mole during the period de- 
scribed ; it merely presents in a concise form the chief 
points of the description already given ; and much of it 
will of necessity be recapitulated as the mechanism is 
dealt with which is used to bring about the various 
" changes " referred to. 

The modern form of the "drawing-up" and the 
" backing-ofF " can now be presented, and with this object 
Fig. 32 has been prepared, showing it fully in detail. It 
must be understood that other types of machine differ in 
the general disposition of the parts from that shown, but 
since the object is the same in each, one description "will 
suffice. Advanta2;e has also been taken in this sketch to 


illustrate what is now becoming a very usual practice in 
the driving of the mule, namely — two sets of fast and 
loose pulleys under the name of " duplex " driving. These 
pulleys are shown at H and G. Instead of a 5 -inch 
strap being used, as seen in Fig. 32, working on a wide 
pulley, two narrow ones are now employed, generally 
each about 2:^- inches wide, woi'king on a similarly reduced 
width of pulley ; the direct object of the arrangement is to 
obtain a quicker change than is possible with a wide belt, 
and although special means are taken in most mules to 
assist the strap in moving from one pulle}^ to the other, 
there must always be some little delay in doing it. The 
adoption of the "duplex" system results in a distinct 
saving of time, and although assistance in the form of a 
strap-relieving motion is not so necessary as before, it is 
still often employed, and usefully so, in helping to obtain 
the change in as short a time as possible. Slight objections 
are raised by some against the arrangement ; such as the 
possibility of unequal tension in the two belts, Avhich 
would throw most of the driving on to one strap and so 
cause breakages and also damages to the machine through 
entanglements, etc. These objections are of a practical 
character, which experience only can decide ; but so far 
nothing has happened to prevent their very extensive 
adoption, and a large proportion of mules now made have 
the " duplex " arrangement applied to them. In order to 
obtain a clearer idea of the disposition of the driving in 
Fig. 32, reference ought to be made to a sketch already 
given in Fig. 11, where a full plan view is represented, the 
lettering in each, with few exceptions, being the same. A 
is the rim shaft containing the driving pulleys H and G ; 
the fast pulley, as in the last example. Fig. 32, has a 
conical extension covered with leather for the purpose of 



forming a cone clutch with a corresponding recessed 
portion of the backing-off cone wheel D riding loose on the 

H. &. H. &. 

Fig. 3a 

rim shaft. Situated on one side of the rim shaft is an 
extra shaft B, on which is keyed a band pulley " a," 


through which the "drawing-up" is effected; "a" is 
independently driven from the same counter shaft that 
drives the rim shaft, but its speed of course can be 
regulated to any extent required. On the other end 
of the shaft B is keyed a small pinion " c," which gears 
into the backing-off wheel " d " and so drives it ; so long as 
the driving pulley and backing-ofF cone are not in contact, 
the revolution of "d" serves no purpose, but directly the 
carriage ceases its outward run, "backing-off" must be 
performed by reversing the spindles ; this is done, as in 
the previous case, by putting " d " and G into contact with 
each other, through the lever E centred at C, and so 
causing the fast pulley to be driven and consequently the 
rim shaft. It is only a momentary action, as will be seen 
when the subject is treated more in detail ; it is mentioned 
here merely to show how it is effected by means of the 
separate driving through the pulley "a." On the side 
shaft B, close to the band pulley, is fixed a bevel "e," 
gearing into a large bevel "f " on an upright shaft W. 

At the lower end of this shaft is the cone clutch and 
bevel necessary for driving the scroll shaft for the purpose 
of "drawing-up" the carriage during the inward run. 
The feature is given in detail so that it can easily be 
understood. Fixed on the shaft is a conical pulley T, on 
whose outer surface is firmly riveted a layer of leather ; on 
its under side is fixed a bevel wheel "g," which gears into 
a large bevel "h" on the scroll shaft. Sliding on the 
shaft W and covering up the lower cone pulley T is a 
conical dish R, which rides loose upon the shaft ; for the 
purpose of driving R the upright shaft is specially prepared 
by having forged on to it a plate S to which is fastened two 
pins U ; these pins fit in holes in the cone dish R, and as 
the shaft revolves they carry the dish round with it. 


During the outward rmi of the carriage the cone 
clutches R and T are out of gear, but immediately the 
run out is finished and the necessary change made, the 
cone clutch comes into gear ; the scroll shaft is then 
directly driven from the band pulley "a" and the "draw- 
ing-up " commences and continues until the arrival of the 
carriage at the stops puts the cone clutch again out of 
gear. The drawing, Fig. 32, shows one method adopted 
for putting the cone clutch in and out of gear. The 
upper part of the cone dish R is prepared with a recessed 
boss for the reception of a forked lever P carrying studs 
fitting in the recess. The lever P is centred at Q and its 
other end is connected by means of an adjustable link N 
■with one end M of the drawing-up lever K, centred on 
part of the headstock at L. The drawing-up lever hangs 
down and lies in the path of the carriage, so that towards 
the finish of the inward run, when the cone clutch is in 
gear, an adjusting screw on the square moves the lever 
K forward and through its connections N and P lifts the 
cone dish out of contact with the cone pulley T and so 
stops the scroll shaft. As the cone clutch must be kept 
out of contact during the outward run, special arrangements 
are provided to prevent K from returning to its original 
position when the carriage and the adjusting screw 
move away from it on their outward run ; the details 
of the action will, however, be treated subsequently. On 
the completion of the outw^ard run a " change " occurs 
which relicA^es the lever K, and a strong spring attached to 
the lever P at forces the cone dish R into contact with 
T and so causes the scroll shaft to revolve and draw up 
the carriage. 

Although the drawing-up cone friction is apparently a 
simple arrangement, and one capable of performing its 


work perhaps better than other methods yet tried, it has 
inherent faults which necessitate extreme care in using and 
setting it. Its whole action depends upon the friction 
between the external and internal conical surfaces ; one 
surface is covered with leather, and this must be of the 
very best quality, firmly and evenly fastened on the 
lower cone and turned in the lathe before applying it 
to the machine. In the form of the cones several points 
must be taken into consideration ; the chief are — Diameter, 
inclination and breadth of the surfaces in contact. In 
regard to the diameter it is clear that this depends upon 
the principle of leverages, and the economical use of power ; 
small cones require much more power, and as a consequence 
the extra power and strain leads to a greater tendency to 
slippage of the surfaces and their speedy destruction. 
Large diameters are therefore a necessity, and within the 
limits set by the work they perform it may be said that 
the larger the cones are the better. All makers try to 
keep them as large as possible, and though local circum- 
stances and individual opinions may cause one maker to 
have the diameter slightly larger than another, the 
difference is not now so great as to give more than a 
superficial advantage. Formerly much trouble was caused 
through diameters being too small and more especially 
when this was associated with a very narrow width. 

The width of the surfaces brought into contact is highly 
important. Although friction is said to be independent of 
surface, it must be considered that in the case of a cone 
friction the wedge action is really the vital principle, and 
as such the ordinary idea of friction must be set aside, 
because surface under such conditions plays a very 
important part ; the greater the surfaces bound together 
for the time being, the greater the force that can be 


transferred through them without yielding. In con- 
sequence of this, a large area of contact is obtained hy 
large diameters and wide surfaces, and the dilierence in 
Avork is only too easily seen by a comparison of the ■work 
of a modern mule and the old narrow frictions. 

The question of the inclination given to the conical 
surfaces is an extremely delicate matter. In the short 
space of probably three-eighths of an inch, the two cones 
must be brought together so firmly as to revolve as one 
and to convey in this condition force sufficient to bring 
the carriage in, and also when apart from each other to 
be perfectly free without the slightest tendency to touch. 
Instantaneous action is indispensable, and this must be 
effected with a minimum strain on the parts controlling it. 
If the angle is not sufficient, the smallest fraction of wear 
or permanent compression in the leather will prevent the 
grip of the surfaces, and even if the adjustment of the 
levers allow of a grip being obtained, the difficulty of 
separating the two cones, when once Avedged together in 
consequence of a too slight taper, is so great that the 
strain is sure to result in frequent and considerable damage 
both to the machine and the yarn. On the other hand, if 
the angle is too great, the wedge action loses its power and 
the grip is not sufficient to draw up the carriage without 
an amount of slippage which practically destroys the 
value of the yarn that is being spun. It will be seen, 
therefore, that a strong element of success in the working 
of the mule depends upon perfect conditions in the 
formation of the friction cone. 

Now, although a mule may be set to work with a 
friction cone practically perfect, its usefulness may be 
partially destroyed by carelessness in the setting of the 
parts that put it in and take it out of gear. Leather 

VOL. Ill F 


wears and is affected by the weather, so that constant 
attention must be given to it, to see that it is performing 
its work properly ; and as all mules contain adjusting 
points this ought to be an easy matter, if care is taken to 
attend to it. 

For fine spinning, say from 120's to 300's, some makers 
have found it an advantage to dispense with the drawing- 
up friction cone, and in the accompanying sketch is 
represented an arrangement of a very effective character 
adopted by one firm of machinists, who are noted for 
their attention to this class of work. Fig. 33 shows the 
chief points of the motion. The backing- off shaft G, 
instead of being driven by band, as in the last example, 
Fig. 32, has two pulleys on the end of the shaft. The 
di'awing-up motion is effected when the strap is on the 
loose pulley B ; it drives the scroll shaft through the 
usual bevel wheels C D and E F, the bevel C, of course, 
being fastened to the pulley B. The backing-off is dri^^en 
fi'om the fast pulley A through H and J, when the wheel 
J is put into gear at the proper moment with the fast 
pulley K on the rim shaft. The carriage, as in the 
p^e^^ous case, moves the strap on to the fast backing-off 
pulley A. It does this at the termination of the inward 
run, by the adjusting stud M coming in contact with the 
draAving-up lever X, and this lever's connection with the 
strap -fork lever E produces the change. Means are 
taken to keep the strap on the fast pulley A during the 
outward run (see description and illustrations of "long- 
lever " mule), so that at the right moment for backing-off it 
is instantl}^ performed b}' H driving J when J has been 
put into gear with the fast pulley K on the rim shaft. 
Means are also adopted to adjust the amount of strap on 
the drawing-up pulley B both by a stop-rod and screw. 



In this way the speed of the dra wing-up can be regulated to 

L K fri 

the amount considered necessary for the numbers being spun 


Changes on the "Cam Shaft" and the "Long- 
Lever" Mule. — It will be convenient at this point to 
desci'ibe how the various changes of action are produced, 
which give to the mule its characteristic motions. In doing 
this there will be an advantage in confining our attention 
to two well-known tj'pes of machines, known generally as 
the "cam-shaft mule" and the "long-lever mule." The 
first-named is so called because its actions depend chiefly 
upon certain important changes being brought about through 
the medium of cams ; Avhile the latter mule obtains similar 
effects almost directly through the regulated movements of 
a long lever. Both systems are good, and give excellent 
results for all classes of yarn, though thei'e is a tendency in 
some quarters to consider the long-lever principle more 
applicable to fine spinning than to the production of coarse 
numbers. Such, however, is not the case ; mules fitted up 
in either system give equally good results, whether for 
coarse or fine numbers. The application of the lever is 
becoming more general, on account of simplicity, easy ad- 
justment, and certainty of action. The cam system of 
course also possesses these attributes, and it must be under- 
stood that it is only in a comparative sense they are indi- 
cated here, but the fact that the best fine-spinning mules 
are almost always built on the long-lever system shows that 
its advantages are fully recognised. 

Cam-Shaft Mule. — As the cam-shaft mule is the one 
most generally known, this will be described first, and 
numerous illustrations will be used to illustrate the several 
features described. Fig. 34 presents a general view of 
the cam shaft as usually applied. In order to fully convey 
the idea of its working, a little recapitulation of what has 
already been said becomes necessary. The driving of the 
machine takes place through the pulleys on the rim shaft A. 


The backing-off cone wheel C is driven continuously, either 

by a wheel from the rim shaft or l)y the independent 


system of driving by band or belt, as shown in the sketch. 
The front roller is driven from the rim shaft, and from the 
front roller the back shaft receives its motion. 

When the carriage commences its ontward run, the 
strap is on the fast pulley, driving the spindles, the rollers, 
and the cari'iage. On arriving at its outermost position, 
changes must be effected which will stop all these actions, 
and it is through the medium of the cam shaft that the 
necessary "changes" are produced. These changes may 
be summarised as follows : — The carriage must be brought 
to rest ; the spindles must be stopped ; backing-off must 
take place ; the front rollers must cease to deliver the 
roving ; and the back shaft must be disconnected from 
the front roller so as to permit the scroll shaft to bring 
the carriage in. 

On reference to Fig. 34, the cam shaft B is shown 
alongside and parallel to the rim shaft. A wheel thereon, 
D, gears into the backing-ofF cone wheel C, and as C is 
always revolving, D, which rides loose on the cam shaft, 
will do the same. On one side of D is cast an internal 
cone dish F, into which can be made to fit a conical clutch 
G ; G is made to slide on the cam shaft by means of a 
float key, and it is kei)t out of gear with F by a lever 
N pressing against it. So long as the cone clutch is not 
put into gear, the Avheels D and F run loose on the shaft, 
but when, by the removal of the lever N, the spring at 
W forces G into contact with F, the cam shaft revolves 
and the desired changes can then take place. By the help 
of the drawing this action can be closely examined. A 
long lever on the inside of the headstock is centred at 
P ; at each end are fitted pins E and S ; on the carriage 
square an arrangement is made for carrj'ing two inclines 
V, T, and by the motion of the carriage these inclines 



come respectively into contact with the pins S and E, and 
depress that end of the lever acted upon. The movement 
of the long lever raises or lowers the link O^ whicli in its 
turn actuates the lever N, and in N we have the control- 
ling movement, Avhich puts in or takes out of gear the 
cone clutches G, F. In the drawing the cone clutches are 
out of gear, but directly they are brought into contact, the 
cam shaft revolves. On the cam shaft are placed several 
cams, which effect the necessary change ; these are shown 
at M, J, and one on the back of the cone clutch at G. 

Fio. 36. 

Their several actions will be considered in detail, as 
well as the special construction of the cam plate H, which 
controls the working of the cone clutch. A small end 
view in Fig. 35 of the two wheels C and D is given, 
which shows their relative positions to each other. 

Although the cone-clutch arrangement on the cam shaft 
is the one generally adopted, there are other makes of 
mule in which clutch wheels are employed in preference 
to the frictional grip. One of the best known is illustrated 
in Fig. 36, where a partial end view is also shown. In 
this case the cam shaft is placed below the long lever ; at 


each end of this long lever A is fastened an inclined bracket 
N. The carriage carries a bowl M, so disj^osed that it 
comes into contact Avith the inclined bracket and depresses 
that end of the lever. The movement of the lever A so 
produced lowers a specially constructed pendant plate C 
in such a way as to relieve the pressure of the spring at 
L, so that the two clutch wheels J, K are at once brought 
into contact. A view of the swing plate C is given in 
order to make it clear how this action is produced ; but 
first it must be understood that the cam shaft S is continu- 
ally revolving through a wheel thereon being in gear with 
the backing-oif cone wheel. The revolution of the cam 
shaft S carries round the half clutch wheel K, which is 
connected to the shaft by a float key ; the other half of 
the clutch wheel, J, is keyed to a loose shell T, whicli 
practically covers in a large part of the cam shaft. On the 
loose shell are fitted the various cams for producing the 
changes. These can only be driven when the two half 
clutches J and K are brought into gear. This is effected, 
as already stated, by the lowering of the swing plate C, 
in the following manner. The plate is arranged to fit 
loosely on the shaft, and also is made capable of rising and 
falling. On one of its faces, inclines are arranged directly 
opposite to each other, and these inclines, in a similar- 
manner to those described in the previous example, serve 
the purpose of keeping K out of contact with J. A pin 
H, passing through the body of J, connects the clutch K 
and the plate C, and as long as the pin is on the highest 
point of the incline at D the two clutches remain out of 
gear. In the position shown in the sketch the long lever 
is on the point of being depressed ; as it falls the plate C 
will be lowered, and will move out of the way of the pin 
H. Directly the pin is free from the incline the spring L, 



which is in compression, at once forces K into contact with 
J, and the shell T immediately commences to revolve. 
During this revolution the ])in H is carried round hy J, and 
is brought into the path of the incline ( r F, opposite to the 
one from which it has just been fixed ; as it travels up the 
incline it forces K out of gear with J, and the cam shell 
instantly stops. Only half a revolution has thus been 
given to the cam shell, this l)eing sufficient to produce the 
necessary changes. AVhen the cari'iage finishes its run-in, 
the other end of the lever is depressed, with the effect that 
the swing plate C is lifted up, and the pin H is relieved 
from the high point of the cam at F, and therefore permits 

Fio. 37. 

the clutch to gear again and 2:)erform another half revolu- 
tion of the cam shell T. 

Movement of the Strap-Fork. — The movement of 
the strap-fork from one pulley to another is effected through 
the cam M on the cam shaft. Its general position was shown 
in Fig. 34. Here a small plan view is given, in Fig. 37. 
Two positions are represented, showing the effect of the 
half revolution of the cam in moving the strap-fork from 
Z to U. Since the cam M is not double-acting, the strap- 
fork returns from U to Z l)y means of springs. By refer- 
ence to Figs. 34 and 3S it will be seen that although L is 
referred to as the strap-fork, it is not so in reality, but is 
simply a kind of buffer or relieving rod between the cam 


M and the strap-fork K itself. This feature is sufficiently 
important to warrant a more detailed description, which 
will now be given. 

From the general plan of the cam-shaft arrangement we 
can proceed to consider it more in detail. The first feature 
to attract attention is the method of moving the strap from 
the fast to the loose pulley. Although the cam M is 
nominally spoken of as performing this function, it does 
not do so in reality ; its chief duty is to move the 
strap from the loose to the fast pulley, and when the 
opposite effect is necessary the cam simply moves into 
a position which allows another action to bring about the 

Fig. 38 shows sufficient of the parts to make the 
description clear. It will be seen that the cam M actuates 
the strap-fork lever K, not in a direct manner, but through 
another lever L, upon which the bowl H is fastened. This 
lever is centred on a short fixed shaft at Y, and its lower 
end is extended at Z in the manner shown, for the attach- 
ment of strong springs a and h. The strap-fork itself is 
centred at /.•, and simply works free in this position ; it has 
a projection about the middle of its length, to Avhich a 
spring a is connected, and by means of this spring the two 
levers L and K are kept together. For instance, if the 
cam M moves the lever L in the direction of the fast 
pulley, the lower end of it at Z will be depressed, and will 
exert a strong pull on the spring «, which will be sufficient 
to draw forward the strap-fork lever K. This depression 
of the lower part of the lever L will also bring into tension 
a spring ft, which is attached to the framing of the machine, 
so that as long as the strap is on the fast pulley the spring 
h is in tension and always exerting its power to force the 
strap on to the loose pulley. It is prevented from doing 



so by reason of tbc lever L being locked in position by the 
twist lever Q, which is attuched to it. This lever serves 
the important purj)ose of regulating the time at which the 
strap can be moved ; the cam ]\I may make its half revolu- 


tion, but the strap-fork will not move on that account ; the 
movement can only take place when the twist lever Q is 
relieved. The action of this lever is as follows : — One end 
of it is attached to L ; at a certain part of its length a 
projection on it abuts against a stop R, which is fixed to 


the framing ; the other end is brought into such a position 
that it can be acted upon by a lever S, which revolves by 
virtue of its connection to a wheel T. This wheel is driven 
from a worm W on the end of the rim shaft, through the 
worm wheel V and the pinion U. Now, since the relief 
of the twist lever is seen from this arrangement to be 
controlled from the rim shaft, it will readily be understood 
that the number of revolutions of • the spindles is the 
dominant factor in deciding when the strap must be 
changed. It has already been shown that the movement 
of the carriage at the termination of its outward run brings 
about half a revolution of the cam M, and so leaves a clear 
space for the pin H to return and carry the strap on to the 
loose pulley ; but unless the spindles have received a 
sufficient number of turns during the run-out, it is not 
necessary that the two actions should be simultaneous ; it 
is frequently advantageous to continue to turn the spindles 
after the carriage has stopped — an action termed " twisting' 
at the head," and for this purpose the twist lever can be 
employed. This action is rendered very simple by the 
arrangement of the wheels ; by changing the pinion U, the 
revolutions of T, and consequently the lever S, can be 
adjusted to cause the strap-fork to move on to the loose 
pulley simultaneously with the half revolution of the cam 
M, or to follow it at whatever interval may be considered 
necessary. During the revolution of S it comes into 
contact with the end of the lever Q and depresses it ; this 
lowers it sufficiently to unlock it from the catch R, and 
when this happens the spring B pulls the strap-fork over 
on to the loose pulley. At the same time the spring X, 
attached to Q and also to the framing, is put into tension, 
so that when the cam M makes the next half revolution on 
the completion of the run-in of the carriage, the movement 


of the lever L pushes Q forward, and the spring draws it 
upwards and locks it again at R. 

Of course it is often unnecessary to have "twisting at 
the head," and therefore an arrangement such as that 
described can be accurately adjusted to give that result, or 
it may be dispensed with altogether, and the twist be 
regulated from the gearing which drives the front roller. 
If the arrangement in Fig. 38 is absent, the only effect 
will be to cause the strap-fork to move on to the loose 
pulley at the same time as the cam M makes its half 
revolution ; this is generally spoken of as " striking 
through " ; but when a definite number of twists per inch 
are required, which it is not considered advisable or 
possible to put in while the carriage runs out, then this 
arrangement supplies the deficiency between the time the 
carriage stops and the backing-off takes place. It may be 
remarked that the change of the strap on to the loose 
pulley is not confined to the method shown, and frequently 
an independent system, called the " strap-relieving motion," 
is employed, which will be described presently. 

Driving of the Cam Shaft.^ — The next point to be 
noticed on the cam shaft is the action of the long lever in 
bringing about the engagement and disengagement of the 
cone clutch on the cam shaft. One method has already 
been described in detail ; the one now under notice refers 
to that illustrated in Fig. 34. Two other sketches are 
now given. Figs. 39 and 40, in explanation of the point. 
Therein B is the cam shaft in section, and H is the plate 
on whose surface are formed two inclines, E F and G J. 
The inclines are at different distances from the centre of 
the shaft, and each one at its highest point, E and G, falls 
abruptly, while the other ends, F and J, fall gradually to 
the level surface of H. The lever N, centred at A, is con- 



nected to tlie long lever by the link ; as the long lever 
is actuated by the inclines on the carriage, N will oscillate 
and be alternately brought into the paths of the inclines. 
Two positions of the cam plate are represented ; the one in 
Fig. 39 shows the lever jST on the highest point of the inner 
incline, in which position the cone clutch is disengaged and 
the spring W (Fig. 34) is in compression. "When the end 
of the long lever opposite to is depressed, will be raised, 

Fig. 39. 

Fig. 40. 

and this movement takes the end C of the lever X away 
from the incline E F, so that the spring is free to force the 
cam plate forward and thus cause the two halves of the 
cone clutch to engage Avith each other and bring about 
the revolution of B, and consequently of the cam plate 
itself. Now it will be observed that, as H is pushed forward 
by the spring W, the surface of the cam plate is brought 
almost into contact with the end C of the lever, and 
therefore as the plate revolves it lies directly in the path 
of the opi)osite incline G J, which in passing tends to torce 


the lever N on one side. In this effort, however, the lever 
remains rigid, the spring "\V at the back of H yields 
instead, and the cam plate is thus forced back and so 
disengages the cone clutch to which it is attached. This 
naturally stops the motion of the cam shaft after it has 
made half a revolution, and it remains stopped until the 
end of the long lever depresses and moves the lever N 
from the highest point of the outer incline at G, and 
places it in a position nearer the centre to be acted upon 
bj' the inner incline during the next half revolution of the 
cam shaft. 

It will be seen that the action just described is one of 
the utmost importance in the operations of the mule. 
Absolute accuracy must be sought for in the adjustment of 
the levers so as to obtain the greatest eft'ect from the 
inclines. The jjrecise moment for putting the clutch in 
and out of gear depends on a number of points that require 
careful attention, most of which are associated with the 
cam plate and levers. The long lever itself must be acted 
upon at the right moment, and of course adjusting brackets 
are always provided to eft'ect this. The surfaces of the 
inclines are sul)ject to wear, and although they are invari- 
ably case-hardened, still it is not unusual to find sufficient 
wear taking place to necessitate care in attending to them. 
They are likewise made separate so as to be easily replaced 
when required ; and the same remarks can be applied to 
the end C of the lever N, upon which a great strain is 
thrown. The spring at the back of H must be strong 
enough to cause a thorough grip in the cone clutch, and 
attention must be paid to the two halves of the clutch to 
see that they are working correctly, and adjustment be 
made for any wear and tear that occurs. 

Attention may be now paid to the other cams on the 


shaft B. These are illustrated in Figs. 41 and 42, which 
represent the action of the cams in operating the front 
roller and the back shaft respectively. The front-roller 
cam G, Fig. 41, forms part of the back of the cone clutch 
on the cam shaft ; it is double-grooved, so that the clutch 
wheel at F can be put into and out of gear without springs. 
A lever D, centred at A, works in the groove, and its other 
end fits in a ring groove on the half clutch E. In the 
position shown the carriage is performing its outward run 
and is spinning ; therefore the clutch is in gear and the 
rollers are revolving. When the carriage stops, the cam 
shaft turns half a revolution, which 
brings the lever D from the highest point 
of the cam down to the lowest point 
directly opposite. This effects a separa- 
tion of the clutch at F, and the rollers 
cease revolving, remaining stationary 
until the run-in is completed, during 
which winding is going on. (It may 
be remarked that this latter statement, 
though correct so far as the cam CI and clutch F are con- 
cerned, has exceptions.) For certain classes of yarn of 
good quality and generally fine numbers, the rollers are 
subject to two other independent motions, namely, " wind- 
ing delivery motion " and " jacking delivery motion " ; these 
will be fully described subsequently. 

Outward Run of the Carriage. — The next cam to 
be noticed, J, serves the purjDose of independently discon- 
necting the back shaft, and so stopping the carriage at the 
finish of the outward run. 

As it performs other functions besides this one, the 
accompanying drawing. Fig. 42, taken from a recent 
machine, has been prepared, from which it will be seen 



that the putting of the Ixack-shaft clutch box into and out 
of gear is not the only action it performs. To understand 

the various movements, great care must be taken to follow 
the descrii)tion, and as it woidd be both difficult and in- 
VOL. Ill G 


convenient to show the different parts in their changing 
positions, the reader must picture to himself the chitches 
being put into and out of gear and the lifting and lowering 
of the levers. 

In the drawing, Fig. 42, A is the rim shaft, from which 
the cam B is revolved half a revolution, just previous to 
or on the termination of the outward and inward runs 
of the carriage. Working in the groove of the cam is a 
bowl C, carried by one end of a bell-crank lever centred 
at D ; the other end of this lever, at E, carries a link, 
whose lower end is slotted and slides on a pin fixed in 
the lever whose centre is at P. A projection on the bell- 
crank lever at V carries a set screw, capable of adjustment, 
Avhich bears against the horizontal arm of another lever 
l)ivoted at H. The vertical arm of this lever carries a 
bowl A, which Avorks in the groove on the back of the 
clutch wheel J ; the drawing represents the clutch in gear, 
with the bowl C on the loAver portion of the cam B. 
Confining ourselves for the moment to this action, it Avill 
readily be seen that the next half revolution of the cam 
will, acting through the stop-screAv at I, lift the horizontal 
arm of the lever G and put the clutch J out of gear, thus 
stopping the back shaft and consequently the carnage. 
The spring at L serves the purpose, as already observed, 
of keeping the two halves of the clutch in gear during 
the outward run of the carriage. 

Drawing-up. — We may now trace the action of B still 
further, for it performs the important functions of directly 
taking the drawing-up cone clutch out of gear and indirectly 
of putting it into gear ; this it does in the following manner : 
— A lever centred at P has on one end Q a forked jaw, which 
fits in the ringed groove of the upper half of the drawing- 
up cone clutch ; its other end M is connected by a strong 


spring to tlie horizontal arm of the lever G. Also in 
connection Avith the lever P is the slotted link F, which is 
pendant from E; a, direct connection is therefore obtained 
between the drawing-up cone and the cam B. As the 
cam B revolves, the end of the lever at E is raised ; this 
action lifts the link F, hut this has no effect on the lever 
at P, because of the slot at its lower end. At the same 
time the lever G is also raised, which puts a strong tension 
on the spring 0, this naturally exerts a powerful tendency 
to pull the end of the lever at M in an upward direction ; 
it cannot, however, effect this purpose at once, because the 
other end of the lever P is locked by the lever Y, and it 
is only at the moment of the finish of the run-out that the 
action of the carriage draws the lever Y on one side, and 
permits the tension in the spring to lift ^I upwards and 
force the end Q downwards, thus putting the cone clutch 
into gear. Directly this happens the bevel on the scroll 
shaft is driven, and the carriage is drawn in. Now it will 
be noticed that the fact of lifting up the lever at M brings 
the pin at N to the top of the slot in the lower part of the 
link F ; therefore, as the cam B makes half its revolution, 
when the carriage is on the point of finishing its iuAvard 
run, the end of the lever at E is depressed, and the link F 
presses downwards on the pin at N and lifts up the other 
end of the lever at Q, thus taking the clutch out of gear 
and stopping the scroll shaft. At the same time the 
lowering of the end M puts tension on the spring 0, 
which, together with the spring at L, forces the clutch-box 
at J into gear and enables the outward run to Ije made. 

Locking Arrangements. — It will be readily under- 
stood that, although the actions just described are simple in 
character and are obtained by means free of complication, 
they are so highly important and depend on such a delicate 


adjustment, both in regard to the time of their action as 
well as the extent of their moAement, that means must be 
taken to ensure accuracy and prevent any derangement 
happening through Avear or accident, to the machine. Those 
practically acquainted with machinery will understand this 
fully, and in the mule the precaution of what is called 
" locking " each motion to guard against irregularities is an 
absolute necessity. We have already seen that the lever Y 
locks the lever P in position until the carriage itself moves 
it away and allows the cone clutch to drop into gear. (This 
feature will be more fully described presently.) There is, 
in addition, an arrangement at the front of the headstock by 
which an effective control of the machine may be obtained, 
which serves the purpose of locking the carriage at the 
termination of its outward run and during the period of 
backing-ofF. An enlarged drawing of it is given in Fig. 43. 
The rod T is connected by the bell-crank lever S and the 
link R to the lever P ; any movement of P will therefore 
move the rod T. Now by locking T in any position it is 
possible to keep P from moving until the rod T is relieved. 
The locking is performed in the following manner : — A lever 
X, centred on a stud, is capable of beijig moved upAvard by 
the carriage as it is finishing its outward run. Tlie rod T 
at this moment is out as far as it Avill move, and a pro- 
jection on the lever X rests in the recessed part of the end 
piece F of the rod T, and keeps the rod from moving 
inAvard. It Avill be noticed on reference to Fig. 42 that 
until the rod T is relieved the cone clutch cannot fall into 
gear, so there have been tAvo agencies at work during the 
outward run to prevent the scrolls being driven from the 
draAving-xap cone clutch, namely : the lever Y and the stop- 
rod T. Directly, however, the carriage comes against the 
leA'er X, the projection at E is lifted up and the rod T is 



free to moA'C Avhen Y is released. Just previous to this, 
however, the carriage in moving outward has lifted up the 
catch Y, and a jirojectiou AY on the carriage passing under 
it allows Y to fall and thus locks the carriage. For about 
the next two seconds the carriage is stationary, and back- 
ing-oil' takes place ; when this action is completed the 
lever Y is moved aside and the cone clutch is freed from 

Fig. 43. 

restraint, so that the spring (Fig. 42) forces it into 
gear. In doing this the rod T is moved in the direction of 
the carriage, as seen in Fig. 43, and at the same time 
the lever Y, through its connection with the rod, is lifted 
up and thus releases the carriage, which at once commences 
its inward nm. 

A lever A, centred at B, is devised to enal^le a stoppage 
of the carriage to Ijc made in anj' position. The drawing 
shows the position of the parts as tlie carriage is on the 

86 COTTON SPINNING chap, ii 

point of running in ; by pressing on A the projection at C 
will force the stop-rod T outwards, and lift the upper part 
of the cone clutch at Q out of gear, which at once stops 
the carriage. The lever A itself can he locked hy the 
catch G when required. On the other hand, wlien the 
projection E is engaged in the recess at F, it can be 
raised out of contact by the projection D of the lever A, 
thereby relieving the rod any time dui-ing the outward run. 

Backing'-oflf. — "Backing-ofF" is the next feature calling 
for attention. At this point, however, only a descrijjtion 
of the means adopted for obtaining it will be given. The 
reason and the effect will be dealt with fully Avhen the 
spindle and cop are treated ; the general idea of the action, 
already given, will meanwhile be sufficient to enable the 
reader to understand the following remarks : — 

In Fig. 44 a full view of the arrangement is shown. 
The chief object in view, it will be remembered, is to put 
into and take out of gear the large backing-off cone wheel 
with the fast pulley, in order to reverse the direction of 
revolution of the rim shaft, and consequently of the spindle. 
The duration of the movement is scarcely more than a 
fraction of a second, but its importance necessitates extreme 
accuracy and prom})tness of action. The cone wheel 
contains a ring groove, in which works a forked lever, 
centred at I. This fork is connected to a lever H, Avhose 
lower end fits loosely on a rod that runs along the side of 
the headstock. This backing-off rod has one end E coupled 
to a bell-crank lever, pivoted at 0, while the other end is 
joined to the upper i)art of tlie lever G, whose function it 
is to lock the drawing-up cone in position during the 
"backing-off." As the carriage moves out, the fast pulley 
and cone wheel are out of gear, as is also the drawing- 
up cone clutch. On the termination of the outward run 



each of these must Le put into gear — the first one the 
moment the carriage stops, and the second one on com- 
pletion of backing-off. In the carriage square is pivoted 
at B a specially shaped jaw lever, the mouth of the jaw 
having au inclined projection. As the carriage approaches 
the finish of the stretch, the inclined portion comes into 
contact with a bowl D carried by the bell-crank lever, so 
that this end of the lever is depressed and the backing-ofF 
rod is moved in the direction shown by the arrow. The 
eff"ect of this action is to move the fixed stop-washer K 
forward, and compress the spring Y, which bears against 
the lower end of the vertical lever at H ; the compi-ession 
of the spring exerts sufficient pressure to force the lever H 
forward and to bring about the necessary contact of the 
backing-off" cone wheel and the fast pulley. At the same 
time the end of the backing-off" rod moves the lever G 
outwards, and places the pin underneath the lever Z, thus 
preventing the upj^er cone clutch from falling into gear. 
Other actions have come into play as this occurs (as already 
described), by means of which the strap is moved on to the 
loose pulley, so that directly the cone wheel has turned 
through a portion of a revolution, it must be taken out of 
gear instantly, the moment before the scroll shaft commences 
to draw the carriage in. It is obvious that this necessary and 
rapid movement of release cannot be performed by allowing 
the carriage to move until the bell -crank lever is free; 
accordingly other actions are introduced to eff"ect it. It 
will be sufficient at this point merely to indicate, rather 
than describe, the means adopted, as a fuller description 
will be given later. 

On the faller X is fastened a lever, one end W of 
which is connected to a pendant arm, whose lower end N 
slides on a bowl rt, carried by a slide, which is moved up 


and down by tlie sluiper as tlie carriage moves in and out. 
As the carriage finishes its outward run, the position of 
the copping faller arm is ap])roximately that shown in the 
sketch. Directly, hoAvever, the tin roller reverses its 
direction of revolution for haching-oft", a small sci'oll (or 
"snail" as it is called) L winds on a chain, Avhich passes under 
a bowl C on the lever A and on to the faller lever at M ; 
this end of the lever will consequentl}' be depressed, and so 
draw up its other end W, and along with it the locking 
faller arm N. It will lift this latter so high that the lecess 
at X will come opposite the Ijowl 0, and its natural tendency 
would be to fall forward and rest there. This is, indeed, 
what actually occurs, but to render this a definite action 
the lever A is connected to the locking arm l)y a liidv P, 
and, in addition, the lever A itself is connected by a strong 
spring S with the opposite side of the square. As long as 
the locking arm X simply rests against the bowl a, the 
lever A will remain fixed in spite of the strong pull of the 
spring S ; but immediately the tin drum, through the snail 
L, draws ]\I downwards and the locking arm N upwards, 
X becomes free from the bowl rt, and the spring S draws 
forAvard with a quick action the lever A, together with the 
link P and the faller locking arm X. This movement of 
A is the one that releases the bowl D from the jaw ; for as 
A is suddenly shot backwards, the jaw itself lifts up and 
carries D Avith it. The upAvard movement of D draAvs the 
l)acking-ofF rod forAvaid, and in doing so the stop-Avasher J 
is brought against the lever H, and pulls the backing-off 
cone Avheel out of gear Avith the fast pulley. At the same 
time the lever G is also moA^ed forAvard, and its projecting 
pin is brought from under the end of the cone-clutch leA'cr, 
and permits it to fall into gear, Avhereupon the carriage at 
once commences its iuAvard run. 


In Fig. 38 a drawing is given sliowing lio\v the move- 
ment of the strap from the fast to the loose pulley is 
regulated from the revolution of the rim shaft, by means of 
the twist Avlieel. The strap can oidy be moved after the 
twist wheel has made a given number of revolutions, and 
b}^ relieving the twist lever allowing the strap-fork to be 
pulled over by a powerful spring. 

Strap-relieving Motion. — It is not always necessary 
to adopt this method of regulating the twist in the mule, 
and frerpiently, instead of employing it, a strap-relieving 
motion is used. An arrangement of this kind is shown in 
the drawing. Fig. 45. A few Avords as to the reason of its 
introduction are necessary before giving the description of its 
action. The application of a twist-wheel motion, as will be 
remembered, enables a very definite number of twists to be 
put into the yarn as the carriage runs out. If enough twists 
have not been put in by the time the carriage has finished 
the stretch, then, although the strap-fork cam has made its 
half revolution, the strap-fork cannot change until the twist 
Avheel relieves it, and until this occurs the sj^indles will 
continue to be driven and so put extra twist into the yarn. 
Technically this is called "twisting at the head "; but it will 
be observed that when the twist is sufficient, their revolu- 
tion must be instantly stopped. Such an action entails an 
enormous, though momentary, effort of those parts of the 
macliine Avhieh perform it. There is an excessive amount 
of friction set up in the cone clutch, and difficulty is 
experienced in moving the strap quickly on to the loose 
pulley. It is, therefore, found that for some classes of 
yarn, lower numbers especially, and in many cases according 
to the opinion or experience of the spinner, the advantages 
of the twist lever are not sufficient to outweigh the advan- 
tages of a strap-relieving motion. In the first place, 



therefore, this motion disphices the twist lever. In its 
stead we have tlie ai-rangcnient shown in Fig. 45. The 
carriage is moving outwarils, and, when Avithin a few inches 
of its finish, comes into contact, througli an adjustable stop 
A, with an inclined lever B, pivoted at C. The lever B 
carries a stud I), which fits a recessed part of E, so that 
the depression of B in the direction of the arrow draws the 
strap-relieving rod forward. This rod is attached at F to 
a pendant lever J, centred at K. On the stud or short 

Fig. 45. 

shaft K is fixed a lever L, whose upper end bears against 
the strap-fork rod. It will, therefore, be seen that the 
forward movement of the rod E will move the strap from 
the fast to the loose pullej', and it will do this gradually as 
the carriage is finishing the last four to eight inches of its 
outward run. By the time the carriage is at rest, and 
backing-off commences, the spindles have therefore lost a 
good proportion of their speed, and a great saving of power 
is cfl'ected, Avith its consequent reduction in strain, etc., in 
bringing them to rest and reversing them for backing-oflF. 
It must not be overlooked, however, that theie may l)e a 


" slight " loss of twist through slowing the spindles ; for 
although the rollers are affected in their speed in the same 
degree, the two are not driven in the same Avay, and it is 
possible for the proportion between them to be slightly 
disturbed. This cannot be regarded, however, as a dis- 
advantage, because it has no practical value, especiall}' in 
regard to the class of yarn it is used for. 

As the rod E is moved forward, a spring S, threaded 
upon it, is compressed by the stop-washer G pressing it 
against the fixed bracket H. In addition, a spring O is put 
into tension at the same time, so that when the carriage 
commences its inward run the rod gradually returns to its 
original position, and leaves the strap-fork on the loose pulley 
until the cam changes it at the completion of the inward run. 

A further point to observe is, that the backing-off lever 
P is locked by this motion, in a similar way to that adopted 
in the twist lever. In order to show this clearly, two 
detached and enlarged views are given in Figs. 46 and 47. 

In Fig. 46 the arrangement is shown in position for 
the strap on the fast pulley and the backing-off out of gear. 
The strap -relieving motion keeps the backing-off lever 
locked by connecting a short lever M to the shaft K, and 
in turn connecting M to a link E ; a projection T on E, 
bears against the backing-off lever P, and until this projec- 
tion is removed backing-off cannot take place. As the 
carriage acts upon the strap-relieving motion, the link li is 
drawn on one side (as shown in Fig. 47) and P is left free 
to put the backing-off cone clutch into gear with the fast 

It will be seen from the drawing that adjustments can 
be made in several positions, and these are necessary. 
Sometimes it is only desired to begin moving the strap 
4 inches before finishing the stretch, Avhile in some cases a 



gradual movement through as mncli as 10 inches can he 
ohtained. Means for oljtaiuing this range of action are 
therefore provided ; but when adjustments are made, an 
important point to be careful about is to see that the strap 
is clear of the fast pulley just as the carriage finishes its 

Object of Backing-off. — Before proceeding further it 
will be advisable to give a general idea of what is meant 
by "backing-olT," in order to explain certain irregularities 
which this action causes, and the mechanical methods 
adopted to compensate for them. 

The spinning operation during the run-out of the carriage 

Fig. 47. 

has already been f iilh^ explained and illustrated. Now one 
direct effect of this method of obtaining twist ia that the 
yarn must be taken from that jwrtion of the spindle on 
which the cop is being formed, and raised to the point : and, 
vice verm, when spinning is completed, the yarn must be 
taken from the point of the spindle and guided on the coj) 
in whatever part of the blade it happens to be. To make 
this quite clear, two diagrams are given. Figs. 48 and 49. 
Spinning is supposed to be taking place, as shown by the 
full lines. When winding takes place, the yarn must be 
taken from M and wound on the cop below X, and when 
that operation is completed it must be returned to the point 
of the spindle. Examination of the diagrams. Figs. 50 and 


51, will very clearly show what the effects of these two 
operations are. In the first place, a Avire C^ running the 
full length of the mule is provided, and over this the yarn 
is guided on to the spindle during the winding process. It 
is carried by an arm centre^ on a shaft A, called the 
" copping " faller ; this fuller rod is actuated by levers from 
the shaping or copping mechanism, and by this means the 
wire C^ guides the yarn on to the upper portion of the 
shaded part of the cop, and in doing so moves through the 
space between C and C\ When the carriage arrives " in " 
against the rollers, the yarn must be transferred from the 
point C^ to the point of the spindle, as in Fig. 50, and in 
effecting this we are brought into contact with one of the 
most interesting and characteristic features of the mule. 

If the yarn were led on to the cop direct from the rollers, 
it is clear that the act of lifting it from C^ to the toji of 
the spindle would cause the whole of the ends to break, 
because of the longer length of yarn required in this latter 
position. And again, we saw when treating of the spinning 
process that the peculiar action of twisting in the mule 
necessitates a certain number of windings of the yarn round 
the spindle up to the point before spinning can commence ; 
and this condition could not be fulfilled if the thread were 
guided direct on to the cop. To obtain each of these 
necessary elements, another wire at D is provided, carried 
b}^ an arm working from a shaft B called the "counter" 
faller. Over the wire D the yarn passes on to the wire C ; 
D is kept in such a position that the length of yarn from 
the cop to the rollers as it passes over the two wires is 
much more than the straight line between them, and conse- 
quently as the carriage gets in and before the spindles cease 
turning, the wire C rises up, and in doing so the spindle 
winds on the extra length in a series of turns, as seen in 



Fig. 51. At the same time the wire D is lowered out of 
contact with the yarn, and the thread is free to be twisted 
as the carriage goes out. 

When the outM'ard run is complete, these extra turns on 
the spindle must be unwound before the winding can take 
place, and as they have been wound on in the same direction 
as the twist, it is evident that the spindles must l)e reversed 

«• V ©A 

60 '®a' 

Fio. 4S. 

Fig. 49. 

to unwind them ; and also, since the unwinding means an 
additional length of yarn, something must be done to take 
up the extra length taken oft' the spindle. 

The reversing of the spindles, in order to unwind the 
yarn from the bare portion of the blade between the cop 
and the point, is the special function of the "backing-ofF" 
process, already described. The action of the wires in com- 
pensating for the extra length i^nwound will be described 



subsequently, the object at present being merely to point 
out the necessit}' of backing-ofF and how it is effected. 

Tightening the Backing-off Chain. — When the cop 

is in the early stages of its formation, the length of the bare 
spindle is considerable, and a good length of yarn requires 
to be unwound, as will be seen on reference to Fig. 49. 
As the cop gets larger, and gradually fills the spindle, the 
amount to be unwound comes less, until, at the finish, it is 
quite a small amount. Fig. 48. As we shall see presently, 

the diminished revolutions of the spindles on reversal, Avhich 
this necessitates, is very easily effected ; but another point 
arises which requires a very careful consideration. The 
mechanism which causes the " copping "-faller wire to move 
from A^ to A and the " counter "-faller wire to move from 
B^ to B acts quickly, and therefore there is a danger that 
the downward motion of A will be much quicker than the 
rate at which the spindles unwind the yarn from the point 
to the cop. For this reason the moA'ement of the copping- 
faller wire is, as it were, delayed, until the spindles 
commence to reverse, and Ity this means the likelihood oi 


breakage is avoided ; and if any slight slackness in the 
yarn results, the counter-faller wire has time to compensate 
for it. . As the cop enlarges, however, the delay in the 
movement of the copping wire, as the spindles reverse, 
becomes a disadvantage ; for there is less chance, owing 
to the shorter length of yarn to be unwound, of the 
wnre overtaking the yarn, and therefore there is less neces- 
sity for the slight slackness in the yarn caused through 
the spindles reversing before the wire begins to move. On 
the contrary, this slackness of the yarn, in consequence of 
the lateness of the action of the wire, results in the making 
of very bad cops and snarly yarn, Sevei^al ingenious 
methods have been adopted to overcome this difficulty. 
Their object is that, while permitting the faller Avire to be 
behindhand in its movement when the cop is begiiniing to 
be formed — because there is a distinct advantage in being 
so — it shall be so controlled that, at each layer added to 
the cop, its moment of action begins to approach that of 
the reversal of the spindle, until when the cop is finished 
the wire is brought to touch the yarn at the exact moment 
the spindles reverse. From this point, down to the cop, is 
so short a distance that there is no danger of the wire over- 
taking the yarn, and at the same time it maintains the thread 
at a tension that enables a perfectly'' solid coj) to be formed. 
Having explained the necessity for adopting some means 
of tightening the "backing-ofF" chain as the cop gradually 
enlarges, it remains to give an example of one method of 
doing it. For this purpose the drawing. Fig. 52, has been 
prepared. It is practically an enlarged view of a portion 
of Fig. 44, and, although showing a few variations in the 
arrangement and details, it can be used for reference in 
reading the remarks made when describing that drawing. 
As the carriage comes out, the various parts are in the 
VOL. Ill H 


positions s]ao^yn in the drawing. The open jaw of the lever 
K depresses the bell-crank lever X, and so puts the backing-' 
off cone clutch into gear with the fast pulley. In con- 
sequence of this, the tin cylinder Z reverses, and in addition 
to reversing the spindles in order to unwind the yarn from 
the bare part of the blade above the cop, it also winds on 
a portion of the chain L, and in doing so pulls down the 
faller arm C which is fastened to the copping faller A. The 
wire / is brought down by this action, and follows the yarn 
down the spindle as it is unwound ; the rate at which it 
does this is regulated by the scroll surface on which the 
chain L is wound. A slight slackness of the chain L during 
the earlier part of the cop is not of much consequence, as 
already explained, and therefore the wire / need not touch 
the yarn the exact moment it begins to unwind from the 
spindle. As, however, the cop enlarges, the action of /must 
be brought earlier into operation, and this necessitates the use 
of some arrangement similar to that shown in the di'awing. 
Attached to a kind of boss of the scroll M is a- chain L, 
whose other end is connected to a lever centred at N. As 
the carriage moves outward, one end R of this lever is so 
arranged that during the time the cop is having its first 
layers formed, it just comes into contact Avith an inclined 
plate S. This plate is connected to the shaper-plates by 
the rail T, and as these shaper-plates move during the 
building of the cop, the incline S is also moved, so that, 
instead of the end of lever at E just coming into contact 
with it the moment the carriage stops, the advance of the 
incline causes R to come into contact a little earlier after 
each layer is added. The effect of this is to cause the lever 
to yield and pull down the chain M, which in its turn 
moves the scroll on which the chain L is wound, and this 
action draws L tighter and gradually takes out the slack- 



ness, so that directly backing-off commences, the chain 
responds a little earlier after each draw, to the backward 
turning of the tin cylinder. In order to present these 
features of the self-actor as fully as possible, it will be 
necessary to give other examples of most of the arrange- 

FiG. 52. 

ments previously illustrated ; but it is also necessary in 

order to prevent complication to present the subject in a 

consecutive form as far as possible, and with this object in 

view it is advisable to proceed with an explanation of the 

building and winding mechanism, after which reference to 

and further descrijition will be given of other methods of 

performing the actions already so far described. 

The Mule Cop.^ — The only way to thoroughly under- 

^ See the author's book, Quadrant and Simper, for a more detailed 
description of the Mule Cop. 


stand the operation of building the cop and Avinding the 
yarn upon it is to make a complete examination of the cop 
itself, and from it to deduce the reasons for emploN'ing the 
special mechanism by which these results are obtained. In 
this way much of the description that follows will be less 
difficult to understand, and a better understanding of the 
problems will follow from the careful reasoning which it 
will involve. 

It has already been pointed out that the spindles are 
carried by a long wooden structure, called the carriage. 
The portion of the carriage which does this is shown in 
Fig. 53. The spindle is supported at two points B and C, 
and the wharve is placed between them, its position being 
nearer the upper or bolster-bearing C than the footstep- 
bearing B. Above the bolster-bearing there projects the 
part of the spindle upon which the cop is built, and it is to 
this feature that our chief attention will be given. An 
enlarged drawing of the cop F is shown in Fig. 54. Its 
general shape is that of a cylinder, with conical ends, one 
end having usually a longer taper than the opposite end. 
The reason for adopting this shape in making a cop is not 
far to seek, and may be summed up in the words, solidity, 
and facility in being unwound again. 

Let us now see how this peculiar shape is obtained, and 
ask ourselves various questions as to Avhat is necessary in 
fulfilling the conditions of its structure. To begin then, 
the yarn must be first wound on the surface of a steel 
spindle, say \ inch in diameter. Frequently this surface is 
slightly enlarged by using tubes as a foundation ; but for 
the present purpose it will be preferable to confine ourselves 
to the most usual course of Avinding the first layers on the 
bare spindle. That part of the blade on which the yarn is 
first wound is practically parallel, and we might almost say 


that the whole of the coj) bottom is wound on to what 
might be termed a perfect cylinder. Above the coj) bottom, 
however, the blade gradually tapers to the point T, where 
its diameter is made as small as possible consistent with 

Fig. 53. 

strength and with the yarn it is spinning. The reason for 
this has already been given in an earlier part of the book. 

As winding takes place, during the return of the carriage 
to the roller beam, it will be necessary to revolve the 
spindles constantly at such a speed that at each inward 
run they Avill Avind on the 64 inches of yarn that has been 
delivered by the rollers and twisted during the outward 
run of the carriage. Readers will understand that it is not 


desirable to complicate matters by mentioning the gain of 
carriage, etc. ; therefore, for our present purpose, the delivery 
of the rollers, of 64 inches, must be accepted as tentative, 
being only for the simplicity of illustration. The first 
layer of yarn Avill therefore consist of 64 inches, and it "will 
be wound upon a solid cylindrical surface. The length of 
that portion of the spindle upon which it is wound is, of 
course, arbitrary, but, as will be shown presently, it is 
made as short as possible, so that the layers are compact 
and close together ; |th of an inch, or 1 inch, is the usual 
length. Subsequent layers are added, and mechanism is 
employed which gradually causes the cop to assume the 
shape shoMu in the lower part of the diagram. Fig. 54. 
The first layer is represented at A B. From A each 
additional laj'er has its commencing point raised in such a 
manner that an inclined surface is produced along the line 
A E J. At the same time, the surface or " chase " upon 
which the yarn is laid is also lengthened ; this lengthening 
of tlie traverse or " chase " is shown by the lines E C and 
also J G. When a diameter has been obtained, as at J K, 
which is considered large enough, a cessation of some 
portion of the mechanism, and a slight modification of 
other portions, cause the commencing point of each layer 
to be raised, but this is done in such a manner that instead 
of giving a conical form, as it did from A to J, it begins 
to rise vertically, and in this way it continues to L, so that 
a cylindrical shape is given to the bod}- of the cop. 

It is clear that, no matter what diameter may be decided 
upon as large enough, the yarn must always finish M'inding 
on the spindle, so that the conical form is continued 
throughout the cop in the same condition practically as it 
had when the foundation A J G H K, or " cop bottom," as 
it is termed, Avas finished. 


It has already l)een remarked that the first layer on the 

spindle from A to B is wound on the hare spindle, and is 

practically a parallel layer. To do so it will be necessary 

to revolve the spindle a certain number of times — a number 

readily calculated. For instance, a \ inch spindle must 


turn = 81vV times to wind on 64 inches. Now 


when the next layer is added, it will begin on a larger 
diameter, represented by the extra layer of yarn ; but it 
will finish on the same diameter as the first layer did. It 
will readily be seen that the speed of the spindle, for the 
second layer, will require to be altered ; but this alteration 
must only take place at the commencement, for since the 
end diameter remains the same, so also must the speed. 
Succeeding layers increase the diameter of the cop at the 
bottom, but finish at the top with the same diameter, until 
we get to the full diameter, as at J K, and a long conical 
surface, as at J G, where the alteration in speed, in order 
to wind yarn on this surface, must vindergo a considerable 
variation from that necessary at the commencement. 

While the speed of spindle during the winding of the 
first layer was uniform, because of the cylindrical surface 
on which it was wound, the speed during the winding of 
the last layer, J G, must be ever varying, simply because 
the yarn is wound on varying diameters. The same length 
is wound on and in the same time as the first layer A B. 
To do this and at the same time maintain an equal tension 
on the yarn, it is clear that the speed of the spindle, when 
the yarn is j^assing on at J K, must be slow ; and, corre- 
spondingly, when it travels up the cone the diameter 
becomes less, and the speed increases until it reaches the 
smallest diameter at G H, and here we must have the 
quickest speed. One revolution of 1^ inch diameter at 


5 X 22 

J K will wind on ^ = 3'92 inches, -while one revolution 


of the small diameter, \ inch, at G H, will oidy wind on 


=-7854 inches: that is, the small end must revolve 


five times Cjuicker than the large end. This increase of 

speed must therefore be gradual, and of such a nature that 

it corresponds as nearly as possible to the gradual decrease 

of diameter. From this reasoning in regard to the last 

layer of the cop bottom, we can see that a variation of 

speed must exist in each layer after the first one, and the 

only difference is that the variation between the first speed 

and the last one is not so great, this, of course, depending 

on the relative sizes of the cop at its various points. For 

instance, when the cop is 1 inch diameter, the variation 

in speed Ijetween the bottom and the top is as 4 to 1, and 

so on for the different diameters. The gradual variation 

in speed during the winding of any single layer, as well as 

the variation of speed between the different layers, can 

easily be shown by means of a diagram, and this we shall 

proceed to show. 

On the assumption that the bare spindle is \ inch 
diameter, it has already been shown that a little over 80 
revolutions will be required to wind on the 64 inches of 
stretch ; and, moreover, since the first layer is wound on a 
parallel surface, the 80 revolutions must be made without 
variation in speed during the Avinding. 

After the first layer, a new set of conditions arises, and 
each successive layer afterwards necessitates a change in 
position from the previous one, and also a complete change 
of the variation of speed which was required for the last 
layer put on. 

When dealinsr with the buildiuG: of the bobbin on the 



Hy-frames, we sav/ that each hiyer required a different 
speed as the diameter increased ; the same necessity also 
arises in the case of the cop, for as the diameter eidarges 
from A to J K, Fig. 54, the speed of spindle must be 
altered, in order to wind on the yarn at this point in the 
same time as when wound on the bare spindle. But here 
the similarity ceases ; in the fly-frame bobbin a parallel 
form is built throughout, while in the cop a conical form is 

T *"■ VARlftTION InTsVEE^ of THE' ; 

B.jT*^ ■ -1 .; . J... J, ..J....... A.^rJN'?-LtA5X'<.sxqS*°iJM.' 

of : ■ ■ ; : ■ • f I i ; is r,uict. ; : • • 1 
"^ : j_ : > .j_. .:...:. ..!... ."..,:....•;.—;.. .!.....i.. ..■ \.. ' 

ol, J....:.. i...ji...J._.[..J.— .i--L-a-^;J — i,..l— i*-il.Je, 


Fig. 55. 

made, Avhich tapers from a larger diameter to a smaller 
one ; and, in addition, the proportion between the two 
diameters varies with each layer. This continued cliange 
of shape renders necessary a change of speed to suit each 
new set of conditions. 

In building the conical form of cop it Avill readily be 
seen that the speed of the spindle must vary, from being 
slow at the large diameter to quick at the small diameter, 
and that this condition must hold good from the first to 
the last layer. It must not, however, be assumed that. 


because the cop has a "straight"' taper, the variation in 
speed is a uniformly increasing one ; this will clearly be 
seen as each layer is carefully examined and its speed 
found As an aid to making this examination, the diagram 
in Fig. 55 has been prepared to show in a graphic form 
the variations of speed for different parts of the cop. 
(For the sake of simplicity the length of the " chase " is 
assumed to remain the same throughout the cop bottom ; 
this assumption makes no difference to the " character " of 
the curves, but to those who desire it, it is an easy matter 
to realise that the length of chase for A is 1 inch, and for 
G 2 inches, all the others, of course, lying between these 
extremes.) The horizontal lines of the diagram represent 
the speed of the spindle ; on the first line we can, therefore, 
mark off the number of revolutions that any given diameter 
will require in order to wind on. In this Avay we find that 
\ inch diameter requires 81 '5 revolutions ; y"*^ inch diameter 
commences to revolve at the rate of 65-2 revolutions ; and 
so on for the other diameters as shown in the table : — 

\ ill. diameter commences at the rate of 81 ".5 revs. 

-iz ill- !, ,. T> 6.5 "2 „ 

gin. „ ,, ,, 54-3 „ 

I in. „ „ „ 40-7.'. .. 

# in '27 "10 ., 

1 in. ., ., ,, 20-4 „ 

Uiii. ;, „ „ 16-3 „ 

These initial rates ol speed give us the starting-points 
of the curves. The other points are not difficult to obtain ; 
but first let us notice what character the curves must have, 
before drawing them. The line representing the speeds 
for the first layer will naturally be straight, as representing 
a uniform rate the full length of the chase ; this is drawn 
at A and shows the same speed throughout. Layers are 
added until the diameter becomes -^^ inch. Starting at 



65*2 revolutions, it finishes at the same rate of speed as 
the first layer, namely 81"5. It is readily seen that the 
slight difference in the end diameters necessitates a variation 
in speed, but not sufficient to show clearly the character of 
the variation, so the line B is almost straight, though it 
will be observed that the end of it takes an upward curve 
at a little quicker rate than at its commencement. 

To emphasise the characteristics, the larger diameter of 
1} inch Avill be taken as an example. Here 
we begin with a rate of 16 '3 revolutions, and i'^"? 

finish at 81 '6 revolutions. As the yarn travels 
upwards along the line G 7, Fig. 56, it will 
reach a point that is 1 inch diameter, and, con- 
tinuing, will pass the | and h inch diameters. 
The question is now — At what rate must the 
spindle I'un in order to wind on the yarn 
evenly, so as to maintain the same tension in it 
at these various diameters 1 This can readily 
be answered ; for we simply have to remem- 
ber Avhat was clearly explained in reference to 
the flyer bobbin (see Vol. II.), that the rate of 
speed must vary inversely as the diameter 
of the bobbin. For instance, if the spindle 
revolve at 16 '3 revolutions for 1| inch diameter, then at 1 
inch diameter it will run at ^- of 16-3 = 20'4 revolutions, and 
so on for the other speeds. A tal)le.will show this better : — 

1^ in. requires a rate of revohition of 16'S 

1 in. ,, ,, ,, JoflG-3 = 20-4 

fin. „ ,, ,, -V of 16-3 = 27-16 

. I'm. „ „ „ JjO- of 16 -3 = 40 -75 

Jin. „ ,, ,, |ofl6-3=81-5 

It is to be observed that the speeds in the above table 
vary inversely as the diameter, for, on comparing the sp(H;d 


at IJ inch and \ inch, we find that while the \ inch is one- 
fifth tlie diameter of 1^ inch, tlie speed is five times 
quicker ; and the other speeds follow the same proi^ortion. 
It only remains to add that from this consideration we 
recognise at once that the characteristic curve of the 
hyperbole will represent the true variation in the revolu- 
tion of the spindle while winding on a conical surface. 
Any diameter similarly treated will give the same charac- 
teristic features, so we are now in a position to represent 
graphically the information obtained from the table. 

By marking off on the line M points representing 
the 1, f, and \ inch diameters, and on the vertical lines 
measuring the number of revolutions corresponding to 
those diameters, we obtain points through which a curve 
may be drawn. This is shown at G, and from it we see at 
a glance the full character of the variation. Starting at 
16-3 revolutions a gradual increase takes place ; instead of 
being uniform, however, the increase occurs at an irregular 
rate, and as it approaches the smaller diameter it rises 
very rapidly, until it finishes on the bare spindle five times 
quicker than at its commencement. This irregular increase 
of speed must be thoroughly understood, for the principle 
of the " quadrant " entirely depends upon it ; and it must 
not be confounded with a uniform increase in speed, which 
would be represented by the dotted line joining the two 
ends of the curve G. §uch a variation differs greatly from 
what should be the real variation, as shown at G. If this 
increased but irregular acceleration of the speed of the 
spindle, as the yarn is wound from the base to the apex of 
the cone, be completely realised and comprehended, the 
understanding of the quadrant will be a comparatively 
easy matter. 

Thus far we have assumed the diameter of the spindle 


to be I inch, ])ut this refers only to the part on wliich the cop 
bottom is built. From this point to the end, it is tapered, 
and therefore each additional layer finishes on a smaller 
diameter, and consequently at a quicker speed. This prob- 
lem will be dealt with at a later stage, as will also the ques- 
tion of guiding the yarn on the cop as winding takes place. 

Another feature to be noticed in regard to the cop is 
the method of obtaining as solid and compact a form as 
possible. We have spoken of laying the yarn on the 
conical surface, from J to G, Fig. 54, but before it can be 
brought from above G to J it must pass over the conical 
surface. This is taken advantage of by causing the faller 
wire W to fall very quickly as the carriage commences its 
inward run, which has the effect of winding the yarn on 
the cop in several spiral turns, which binds together the 
layer below. On reaching M, the wire X commences its 
upward movement. It is this special movement that we 
have been considering, and it is this wliich is generally 
understood when " winding " is mentioned. 

The Mule Quadrant and its Action.^— Having 
given an explanation of what is required in regard to 
driving the spindles at a correct speed while building the 
cop, we proceed to examine and explain the means adopted 
to obtain it. A rough outline only of the mechanism will 
be given at this point ; fuller details will follow as we 
proceed with the examination of its action. 

As already described, the spinning or twisting process 

takes places as the carriage moves out and the spindles are 

driven from the rim shaft. During the drawing-up, the 

tin cylinder is disconnected from this source, and receives 

its motion for winding purposes from an adjacent drum to 

which it is geared. This will be observed on reference to 

1 See the author's Look, (Quadrant and Shapcr, for a more detailed 


Fig. 57 ; the tin cj-linder u is seen to be geared, by the 
wheel X and ^, to a drum, round which a chain is AA^ound. 

This chain is firmly fastened to the drum, and after passing 
round it several times it is connected to an oscillating arm 
called the "Quadrant," It is from this quadrant that the 


spindles receive their S2)eed, or rather they are controlled and 
regnlated by it as the cop passes through its various stages. 

A good idea of its position and proportions can he 
obtained from the drawing, Fig. 58. 

An enlarged view of the winding chain and drum is 
given in Figs. 59 and 60. One is a plan view, and shows 
the chain A passing round the drum B and connected to 
a hook D. If the hook is fixed, and a horizontal move- 
ment be given to the drum in the direction of the arrow E, 

Fig. 5S. 

the drum will be compelled to yield by turning on its 
centres ; this it will do by revolving in the direction of 
arrow F, and so unwinding some of the chain as the distance 
from its first position increases. In this apparently simple 
method of producing rotation there lies the germ of the 
mule (piadrant, and we shall try by reasoning, to follow 
out the course Avhich led Koberts to devise a mechanical 
arrangement that takes rank as one of the most remarkable 
and ingenious inventions of the last century. In passing, 
it may be as well to point out that readers are occasionally 


met with who look on the mule quadrant as the " differential 
motion " of the self-actor. It is scarcely possible for a 
reader, who has followed what has already been said, to 
labour under this impression, for it was emphatically shown 
when dealing with the fly-frame (see Vol. II.) that a 
diflferential motion is simply a convenient method of com- 
bining two distinct motions, through the medium of which 
a variation in one or the other can be effected. It possesses 
no variable element in itself, nor has it any part in either 
building or winding in the fly-frame. The variable motion 
of the bobbin in this later machine is entirely brought 
about by the cone drums, and the differential motion has 

Fig. 59. 

Fig. 60. 

nothing whatever to do with it, except as an arrangement 
of wheels which assists in transferring a variable motion 
already given. 

If the quadrant can be compared to anything, it is to 
the cone drums that it bears a resemblance — but only to 
the extent that they are both the direct means of giving a 
variable speed to whatever they drive. They do this, 
however, by such entirely different methods and principles, 
that a similarity exists only in the " name " of their purpose, 
namely — winding. Readers are therefore warned against 
falling into the error pointed out above, for it denotes 
failure in the attempt to understand the principle and 
purpose of either arrangement. 


In the following explanations of the various phases of 
the action of the quadrant, the illustrations are mainly of 
a diagrammatic character, the chain, quadrant, and cylinder 
being represented as simple lines, free from details. 

Although the explanation will be made as thorough and 
comprehensive as possible, and from it an almost complete 
understanding of the subject can be obtained, it must not 
on any account be considered a " theory " of the ciuadi'ant. 
It is rather a practical demonstration of the action of the 
quadrant drawn out to scale and shown in diagrams, a 
mere fringe of the theory being introduced in order to 
explain some of the results brought to light by these 
draA\angs. This is stated in order to prevent readers from 
falling into the error of ascribing to a brief exjilanation 
the term "theorj'." A theoretical consideration of the 
problem would be entirely out of place in these pages, 
chiefly because the subject requires a degree of knowledge 
for its comprehension Avhich is totally beyond the average 
reader. The practical view here given, is designed to give 
the required information in the simplest manner, and also 
to dislodge some of the peculiar ideas which many hold on 
the subject. 

Our first attempt will be confined to noticing the effect 
of the chain on the winding drum, when the point of its 
attachment is fixed, during the whole of the period of the 
run-in of the carriage. The accompanying series of dia- 
grams y^WS. illustrate the remarks. In Fig. 61 the chain 
is fastened at H, and the other end is Avound round the 
drum, which in its outermost position is shown at A. As 
the run-in takes place the drum will travel from A to G, 
and by dividing the stretch into equal parts, say six, Ave 
get seven difterent positions as occupied by the carriage 
whilst winding, these being shown at A, B, C, D, E, F,. 
VOL. Ill 1 

114 COTTON SPINNING chap, ii 

and G. Now, since the end of the chain is fixed at H, 
the motion from A to B will cause a certain length of 
the chain to be unwound from the drum, and, as before 
explained, this will cause the drum to revolve, the amount 
of the revolution of course depending upon the length of 
chain unwound. On account of the position of H in relation 
to the drum (which, it will be observed, is in the same 
horizontal line with the movement of the upper diameter 
of the drum), the chain unwound equals the distance 
moved by the carriage, and as each distance moved is 
exactly equal to the last, we get, for each of the divisions 
shown in the diagram, equal lengths of chain unwound. 
The chain unwound from A to B is equal to I J, and from 
F to G it is equal to N P, and so on for the other lengths, 
all of which are equal to each other. The movement of 
the carriage under these conditions clearly produces an 
equal rate of revolution in the winding drum in each 
division, and therefore a " uniform " rate of speed is 
obtained throughout the stretch. This equal horizontal 
movement of the drum, producing a uniform revolution, 
must be specially observed to depend on the position 
occupied by the fixed end of the chain at H. If this 
position is changed, another set of conditions arise which 
totally destroy all ideas of uniformity ; and to emphasise 
this important point an illustration will be given. Let it 
be supposed, as shown in Fig. 62, that the point of 
attachment is raised vertically over the position H^ to H ; 
the chain would then pass from H to the drum A, and its 
point of contact there, Avould be at I (the unused part of 
the chain is shown in dotted lines throughout). The drum 
moves equal horizontal distances, as in the upper figure, 
so we may readily compare the eff"ects of the two sets of 
conditions. In Fie;. 61 it was found that the length of 



chain unwound was exactly equal to the horizontal distance 
moved by the drum, but in Fig. 62 the chain unwound is 
very far short of the distance from A to B. This length 
is shown by a thick line at I^ to J, and a glance will show 
the great difference between the two. A further movement 
from B to C will cause another length of chain to be 
unwound, which is shown in thick lines from J^ to K; 
this is a greater length than was unwound during the 
first movement from A to B. We shall also find on follow- 
ing out the other movements of the drum that each 
successive length unwound is longer than the previous 
one, and when we come to the last one, from F to G, the 
length N^ P unwound is over twice that unwound during 
the first movement, from A to B. We thus find that by 
altering the position of attachment and making it fixed 
we destroy the uniform motion of Fig. 61, and obtain a 
gradually increasing and varying one in its place. 

At the first glance it might appear that these results 
would produce the variation in the speed of the spindle 
required in making the cop bottom ; and, as a matter of 
fact, in a limited sense, a conical cop could be built by 
this arrangement. The first layer would be Avound on a 
parallel spindle, when the chain was at H in Fig. 61, 
and by moving the point of support vertically the various 
layers of the cone would be added until the point H was 
reached in Fig. 62. 

This may be made much clearer in a diagram showing 
by means of a curve the relative variation of speed for 
each position. Fig. 63 has been prepared with this object. 
The upper line I to P represents uniformity of the motion 
in Fig. 61, and corresponds to a similar line in Fig. 55. 
The curved full line F to P^ represents the variation as 
produced in Fig. 62, and we can readily see that it has 


all the characteristics for giving the variable motion 
necessary for a conical form of cop. By comparing this 
curve, however, with the corresponding one in Fig. 55 
it will be immediately observed that, although the two are 
allied in character, they are the reverse of each other. In 
Fig. 55 the curves increase slowly at first, and finish 
rapidly. In Fig. 63 the opposite is the case ; we get a 
rapid increase at the beginning and a slow finish. In other 
woi'ds, Fig. 55 is the curve for building a conical form 
with the larger diameter at the bottom, while Fig. 63 is a 
curve of speeds for a cop "upside down," Avith the smallest 
diameter at the bottom. The dotted curve represents the 
variation required for the actual conditions of a mule cop, and 
Ave can clearly see a reverse order of their characteristics. 

Two lessons can be learnt from this illustration. The 
first is that a statement which makes out that with a fixed 
point of attachment for the chain, and equal horizontal 
moA'ements of the drum, a uniform motion is produced, 
is entirely wrong in principle ; the second, that statements 
in connection with the quadrant, Avhich point out that 
certain variable results in motion are produced, is not 
sufficient to explain, even from an elementary point of 
view, the principle underlying such an important piece of 
mechanism. A comparison is made in Fig. 64 between 
the total length of chain unAvound from Fig. 61 and Fig. 
62, and corresponding points in each length are connected 
by dotted lines to emphasise the difference betA\'een them. 
"We see that in addition to the variable motion of Fig. 62 
a shorter length of chain is used, and consequently the 
total revolution of the winding drum, and therefore the 
spindles, is less than in the case of Fig. 61, Avhich winds 
the first layer on the spindle. 

We shall noAV consider the question as it actually pre- 


sents itself in the mule. The point of attachment for the 
chain, instead of being fixed, is carried by an arm, which 
is made to oscillate round a fixed centre. The point of 
attachment at the commencement of the cop is as near 
this fixed centre of the lever as possible ; and as the cop 
enlarges, the nut to which the chain is hooked is raised up 
by a screw working Avithin the arm of the lever. The new 
positions of the point of attachment, in conjunction with 
the movement of the arm itself, brings about the required 
degree of variation in the unwinding of the chain, and 
therefore in the speed of the spindles. 

Examining the action of the quadrant in bringing about 
this result, let us first take the case when the point of 
attachment is near the fulcrum of the quadrant arm, Fig. 
65. The arm, centred at H, is caused to move in unison 
with the carriage, through a quarter of a circle. It is 
arranged to commence from a line which is a little back 
from a vertical through the centre H, probably about 15°, 
as at H J ; from here it moves through 90° to H Q. As 
the carriage moves from A to G, the quadrant moves 
through this quarter of a circle. An important feature 
must be noticed in this connection : during the earlier 
portion of the inward run of the carriage, the copping 
faller wire is depressed quickly, and lays some yarn on 
the cop in a few coarse-pitched spirals- — an ojjeration 
called " crossing " ; the carriage has moved a little distance, 
10 or 12 inches, before this operation is finished, and 
during this time the quadrant arm has also moved forward. 
When "crossing" is complete, the essential part of the 
winding commences, and it is for this feature that the 
quadrant serves its real purpose, and to which we are now 
drawing attention. Generally speaking, the quadrant arm 
is vertical when "crossing" is finished, and, relatively, 


the winding drum is in the position at B. The movement 
of the quadrant from J to K, and of the carriage from A 
to B, has nothing to do with the problem of winding, 
except that *' crossing " takes phice during this period. 
From B onwards, however, the si^inclles must be revolved 
to wind the yarn from a large diameter, which gradually 
tapers, until the bare spindle is reached. The movement 
of the carriage during winding is divided into five equal 
divisions, giving six positions of the drum ; by dividing 
the path of the quadrant nut into the same nimnber of 
equal divisions we get the position the nut occupies for 
each position of the drum, and we can then, by measure- 
ment or otherwise, find the lengths of chain unwound as 
the carriage moves in. These respective lengths are 
shown in thick lines from 3 to 4, from 5 to 6, from 7 to 8, 
from 9 to 10, and from 11 to 12. The difference between 
them is very slight indeed, and while theoretically they 
correspond to a conical surface, it is so little as to be almost 
imperceptible. During this movement of the carriage 
from B to G, the point of attachment of the chain has 
moved forward in the small arc of a circle from K to Q, 
and by doing this has prevented the unwinding of a little 
of the chain which would have been unwound if K had 
remained fixed. We get the first la^'er wound on the bare 
spindle during this period. As the layers are added, the 
nut is caused to travel up the screw of the quadrant until 
the cop bottom is complete, and its position at this point 
is shown at K in Fig. 66. The quadrant arm never varies 
in the angle it describes; so with the nut at K it still 
traverses the same angle, but as the circle is much larger, 
the length of the arc K ]\I Q, which the nut travels along, 
is much greater than K Q, in Fig. 60 ; consequently the 
amount of chain unwound is considerably less, because the 


nut moves in the same direction as the carriage to a 
greater extent than when the smaller arc of a circle in 
Fig. 66 is being traversed. 

When the quadrant arm is vertical the nut is at K, 
Fig. 66, and the chain passes from this point to the Avind- 
ing drum B, which it touches at 2. The length of chain 
between K and 2 is unused chain. As the carriage moves 
inwards to C, the quadrant travels from K to L ; and as 
this movement is almost a horizontal one, the difference 
between the lengths B C and K L represents nearly the 
amount of chain unwound from the drum. The amount 
unwound is shown by the thick line 3, 4 ; it is relatively 
a short portion of chain, and from it we see that the 
spindles are revolving slowly, because at this time the yarn 
is being Avound on the thick part of the cop bottom. By 
measuring off or calculating the length of chain unwound 
as the carriage traverses each of the divisions C to D, D to 
E, E to F, and F to G, w^e get for each of these movements 
respectively a length equal to each of the dark lines at 
5 to 6, 7 to 8, 9 to 10, and 11 to 12. These lines represent 
the amount of chain unwound, and it is clearly to be seen 
that the drum is revolved very slowly at first, and much 
quicker at the termination of the run-in. They represent 
very graphically the varying speed given to the spindles 
during the winding of the last layer on the cop bottom. 

In order to present the residts in the same way as those 
given for the speed of spindle in Fig. 55, a small 
diagram is given in Fig. 67, for the j^urpose of com- 
parison, so that an idea may be formed as to whether the 
quadrant turns the spindles at a correct speed for winding. 
It is generally assumed that the quadrant does wind 
correctly, and therefore we find writers dismissing the 
subject by pointing out a variation in certain lines, and 


saying this variation explains the action of the quadrant. 
We have warned readers against this kind of explanation, 
and it would scarcely be consistent for the Avriter in this 
case simply to point to the thick lines in Fig. 66, and 
say these represent the necessary variation in the speed of 
the spindle for building a conical cop. Fig. 67 is therefore 
prepared to show why an oscillating arm, as Ave have it in 
the quadrant, gives results in winding of an opposite char- 
acter to those produced by a fixed arm, and which approach 
most nearly to the actual conditions of speed required. 

In the diagram. Fig. 67, the upper dark line represents 
the variation in speed produced when the quadrant nut is 
in its lowest position, as in Fig. 6.5. It is practically 
straight, and from this fact we see that an almost uniform 
motion is given to the spindles during the winding of the 
first layer on the bare spindle. The lowest curved line 
shows the variation in the speed of the spindle as 
produced when the nut occupies the highest position on 
the quadrant arm, during which time the full conical form 
of the cop bottom is completed. The dark lines trans- 
ferred from Fig. 66 to Fig. 67 give the curve for the 
third position ; its character corresponds closely to the 
similar curve in Fig. 55. It Avill be noticed that it rises 
very sloAvly at first, and afterwards the acceleration is 
greatly increased. This is what we know ought to be the 
case for a conical form of cop, but a very important point 
must not be overlooked : it ought to be asked whether 
this curve is actually similar to the one required for the 
speed and spindle. If any variation exists, then the 
quadrant is not performing its work perfectly. It Avould 
be impracticable to enter into the question fully, so it must 
suffice to point out that the two curves do not correspond. 
The dotted curve shown in Fig. 67 represents approxi- 


mately the variation of speed the spindle oxujht to have, 
while the thicker cnrve underneath shows lis the speed it 
actually has given to it by the quadrant. There is a very 
perceptible difference between the two curves, and it 
represents a considerable percentage of variation, which 
extends throughout the "stretch." The quadrant is 
therefore by no means "perfect" in giving the correct 
speed for winding ; the difference just pointed out must 
be compensated for in some way, in order that proper 
winding can take place. Fortunately this can be effected 
very simply in the mechanism employed to put the yarn 
on the spindle, so that by means of the "shaper" the 
errors of winding, produced by the quadrant, are practically 
eliminated. In Fig. 65 a middle position of the nut has 
also been taken, and from it tlie second position curve in 
Fig. 67 has been drawn. 

Another method of showing the length of chain un- 
wound during each horizontal movement of the carriage 
is given in Fig. 68, A, B, and C representing the 1st, 2nd, 
and 3rd positions respectively ; we see, in the full parts of 
each line, the varying portions of the chain unwound 
The total length of chain used for turning the drum gets 
shorter as the cop builds, and from this we gather that the 
total number of revolutions made by the spindle becomes 
less and less as the cop bottom nears completion. 

It is the practice, sometimes, in explaining the action 
of the quadrant, to draw a diagram somewhat similar to 
Fig. 66, and to drop vertical lines from the points J, K, 
L, M, N, P, Q. The horizontal and varying distances 
between these lines, as at a b, b c, c d, d e, e f, andfg, are 
then considered to represent the variation in speed 
produced by the quadrant, because it is said the quadrant 
delivers chain, as it were, in these proportions to the flrum 


as the carriage moves in. It need scarcely be pointed out 
that such an explanation is entirely wrong, and the use of 
a pair of compasses in measuring the diagram will at once 
prove how totally at variance it is with the actual conditions. 
Another point in the explanation is the statement that 
the amount of chain delivered forward as the carriage runs 
in is equal to the hoi'izontal distance a to g. This can 
also be so easily tested and found to be wrong that it is 
strange the above explanation, with all its errors and the 
wrong conception of the principle of the quadrant, should 
be so persistently repeated. A point also to be carefully 
guarded against is that on no account must the movement 
of the quadrant from J to K be allowed to enter into the 
question of the building of the conical part of the cop. 

Having shown how the quadrant produces, approxi- 
mately, the necessary variation to the speed of the spindle, 
during winding on a conical surface, there remains another 
feature to be pointed OTit and explained. The description 
so far has been confined to demonstrating how the above 
variation from a large diameter to a smaller one is brought 
about. We shall now describe how the initial speed for 
each new layer is produced. Every fresh layer makes a 
new conical surface, and while the smallest diameter of the 
cone jiractically remains the same throughout the cop 
bottom, the base of the cone is continually enlarging ; and 
this necessitates a difterent initial or starting speed for 
each additional layer. For instance, the bare-spindle 
diameter Avill wind on 64 inches by revolving a little over 
80 revolutions during the run-in. (NoTE. — It has not 
been considered necessary in this remark or in the previous 
ones to subtract the amount of yarn used during crossing 
from that actually Avound on after crossing, as it makes no 
difference at all to the reasoning employed or the character 


of the curves deduced from it.) When the base is enlarged 
to \ inch diameter the initial speed must be at the rate of 
a little over 40 revolutions, and for f inch diameter a 
corresponding reduction in the initial speed is produced 
equal to about 27 revolutions. For 1 inch diameter the 
starting speed becomes a fraction over 20 revolutions, and 
on being enlarged to IJ inch diameter a slight reduction 
on this (to about 16 revolutions) is necessary. By incor- 
porating these results in a diagram. Fig. 69, a curve can 
be drawn which represents very distinctly how^ the starting 
speed for each new layer varies from a quick speed on the 
bare spindle to a slow speed on the 1 \ inch diameter. This 
variation in the initial speed, although not previously 
mentioned, can be clearly noticed in the diagram, Fig. 55, 
which shows the full variation for several parts of the <Top. 
The curves in that diagram, if transferred to Fig. 69, would 
start from the points A, B, C, D, and E, and would follow the 
directions shown by the lines F, G, H, J, and K. From Figs. 
67 and 68 the same information can also be deduced. 

It was made clear in describing Figs. 65 and 66 that 
the movement of the point of attachment for the chain, up 
the quadrant arm and away from its fulcrum, enabled us 
to obtain the desired condition of winding. The question 
arises — ^What position must the nut to which the chain is 
connected occupy, for the various layers as they are added, 
in order to wind correctly ? Only a relative answer can 
be given here to this question ; to deal Avith it fully Avould 
require a number of very carefully-drawn diagrams, or a 
complicated system of calculation, which would scarcely be 
of use, at present at any rate ; so we will simply give a 
practical example. 

It was seen in Fig. 69 that a very great reduction 
takes place in the initial speed of spindle during the time 



the first \ inch increase of diameter is added ; in fact, it 
falls to one-half. It was understood from Fig. 66 that 
the initial speed becomes slower as the nut travels up 
the quadrant. From these deductions, therefore, we can 
conclude that the first \ inch increase of diameter necessi- 


Fig. 69 

tates a considerable movement of the nut up the screw to 
correspond to the great reduction in speed of the spindle. 
Now let us notice the reduction of speed when the last 
\ inch is added, as from I) to E, Fig. 69. It is compara- 
tively little, and therefore, as ]>efore, we conclude that 
only a slight movement of the nut up the quadi*ant screw 
will produce the necessary change. Between the two 



extremes the movement of the nut gradually lessens, and 
at first sight it might be said that the curve in Fig. 69 if 
reversed would represent the rate of movement. This 
conclusion, however, would be wrong ; the curve gives us 
a " clue " to the rate of travel of the nut, l)ut it by no 
means represents the actual rate. 

In order to present to the reader an actual practical 

Fig. 70. 

illustration of the movement of the nut up the quadrant, the 
diagram. Fig. 70, has been prepared. It M'as taken under 
ordinary working conditions. A good minder was chosen, 
and was permitted to "govern" the quadrant just when he 
thought proper ; notice Avas taken of each movement of the 
nut and its amount, as Avell as the number of draws in the 
cop bottom, and the intervals between each movement. 



Number of draws showing at which number the- quad- 
rant was actuated. 













































































































































































































































































































The results Avhen drawn out in diagram form yielded 
the curve shown in Fig. 70 ; and it is striking as 


showing most clearly what is readily deduced from the 
foregoing descriptions. The vertical lines, Fig. 70, 
represent equal intervals of layers on the cop bottom, and 
the horizontal lines represent inches on the quadrant arm. 
The first few layers required a movement of the nut from 
A to B, about 1\ inches; the next few layers necessitated 
its moving from B to C, 2 inches only ; and the last lot of 
layers (equal to the first lot) required only a movement of 
a little over half-an-inch. The intermediate positions of 
the nut are shown on the last vertical line, and to those 
unable to understand the curve, this line will show how 
the nut moves less and less as the cop bottom increases in 

The straight dotted line joining A and X represents the 
uniform movement of the nut up the screw, and it is easy 
to compare the two lines and from them understand how 
the movement is quick at first and slow at the end. It is 
almost needless to add that a curve similar to Fig. 70 
would be given if the results were based on an investi- 
gation of the quadrant itself. AVhen the mule is not 
fitted with some automatic arrangement for actuating the 
quadrant screw, the "minder" attends to it himself, and 
it requires a considerable amount of skill and attention to 
so move the nut as to give good results. When dealing 
with the subject of automatic "governors" (or, as they are 
sometimes called, " strapping " motions) further reference 
will be made to this subject. 

To avoid complications, no attention has been paid to 
the effect which the tapered spindle has upon the question 
of winding. The diameter of the spindle Avhere the cop 
bottom finishes is larger than the part where the full cop 
is complete. To compensate for this taper, some addi- 
tional arrangement is necessary to help the quadrant ; the 
VOL. Ill K 


mechanism employed is usually termed a " nosing " motion, 
but, as it is generally actuated from the "shaper" or 
fallers, its consideration will he deferred until a full ex- 
amination of both these features has been made. 

In the accompanying sketch, Fig. 71, a view is given 
of the quadrant and its connections. Only the essential 
features are shown, the chief ones being drawn in full 
lines. The driving of the quadrant is obtained in an 
indirect manner, and is an example of a rectilinear motion 
producing a circular one. Previous pages have described 
how the carriage receives its inward motion through a large 
scroll on the back scroll shaft, and it will be remembered 
how the carriage by this means had an irregular move- 
ment given to it. Now it is quite clear, from the foregoing 
explanation of the quadrant's action, that the forward 
motion of the quadrant must correspond to the motion 
of the carriage ; therefore the irregularity of the one must 
be reproduced in the other. The best Avay to obtain this 
result is to drive the quadrant from the carriage itself, 
either directly, as shown in the drawing, or indirectly 
through the back shaft. 

On reference to Fig. 74 it will be noticed that a band 
is fastened on the carriage square at J, whence it passes 
towards the back of the headstock and over a loose pulley 
H, or in some cases over a pulley on the back shaft. From 
this point it returns to the front of the headstock, and 
after passing round the quadrant drum G several times, it 
is attached to the carriage square at K. In whichever 
direction the carriage moves, it will, by means of the band, 
drive the drum G. On one end of the shaft that carries 
the drum is keyed a small pinion F, which gears into the 
toothed portion of the quadrant E; the drum and wheel 
F are so proportioned that one complete draw causes the 

- *! 



quadrant to move backwards and forwards througli a light 
angle about the centre A. 

The screw B is carried in the hollow box part of the 
arm by bearings at each end ; its upper end at V has fitted 
to it a ratchet wheel, into the teeth of which a pawl 
engages, A handle D enables the screw to be turned in 
either direction, so that the nut C can be raised during the 
building of the cop bottom, or lowered to its starting point 
for the commencement of a new set of cops. The chain L 
passes from the qiiadrant nut on to the winding drum ]\I ; 
the end of this drum carries a large Avheel N, which gears 
into a smaller wheel P on the tin roller. In this way the 
spindles receive their motion as the chain is unwound from 
the winding drum. The precise action of this connection 
between the winding drum and the tin cylinder can now 
be explained ; a large view of the arrangement is shown 
in Fig. 72. 

Winding Drum and Tin Roller. — AYe have already 
described how the spindles receive a quick speed from 
the rim shaft during the spinning process, and a drawing 
"was given in Fig. 27 illustrating the driving arrangement. 
When winding takes place, with its comparative slow 
speed, some method must be adopted to disconnect the tin 
cylinder from the rim shaft driving, so that the spindles 
can be driven independently from the winding drum. 

The sketch, Fig. 72, fully explains the means adopted. 
The chain L passes round the winding drum M ; the end 
wheel N from this receives its motion, which it transfers to 
the wheel P, which runs loose on the tin roller. This tin 
roller wheel P is formed with a disc Q fixed on its boss, or 
cast in one piece with it ; on the disc is fastened a stud, 
which carries a catch or click C ; this catch can, when occasion 
requires it, be put into gear with a ratchet wheel A, which 



is keyed on the tin roller shaft. This occurs when winding 
takes place, so tliat the revolution of N, brought about 
by the unwinding of the chain from the drum, causes the 
tin roller to be driven through the catch C and ratchet 
wheel A. 

Details are given in the sketch, which enable the action 
to be easily understood. When the Avinding drum receives 
motion, immediately the carriage commences its inward 

run, the disc will begin to revolve, and in doing so will 
cause the click C to engage with the teeth of A, and so 
rotate the tin roller. This is the chief element of the 
arrangement, but there are two considerations to be taken 
into account : in the first place, the click C must be kept 
out of contact with the teeth of the click wheel A during 
spinning ; both noise and damage would result if this were 
not done ; secondly, we must recognise how important it 
is that winding begins immediately the carriage starts in. 
In order that this shall happen, the click C must be in a 


position to engage instantly with the teeth of A; other- 
wise a slight interval will elapse ; for instance, when the 
click is just on the jDoint of one tooth, it must pass over 
the pitch before engaging with the next. In this time, 
although it is slight, the carriage will have moved, and as 
no winding has taken place the yarn becomes slack and 
snarls are formed. To obtain both of the above necessary 
conditions a pendant lever D hangs loosely from the tin 
roller shaft ; and on a groove in its boss there is i^laced a 
spring B, so shaped as to grip the boss firmly ; a leather 
lining on each side of the spring gives the necessary 
resistance in the form of friction to the free movement of 
the spring. One end of the spring is extended, and fits 
in a slot made in a projection of the click C. Now let it 
be supposed that the carriage commences its inward run ; 
the rod D hangs vertical, and its lower end is in contact 
with a stop fixed on the rod E (the purposes of this rod 
have already been fully explained in connection with Figs. 
42 and 43). When backing- oft' is finished the rod E 
shoots back, and the stop F moves the rod D on one side, 
as shown, for example, l)y dotted lines, from D to H. 
This oscillation of D carries with it the spring B, which 
clips its boss ; the spring in its turn acts on the click C and 
forces its other end into engagement with the teeth of the 
ratchet wheel A ; winding can therefore take place instantly, 
because the click being already in contact with a tooth can 
at once commence to turn the tin roller shaft. 

When the inward run is completed, the strain on the 
click is taken off, the disc begins to revolve in the opposite 
direction, because the winding drum must wind on the 
chain during the outward run ; and the ratchet wheel A con- 
tinues, after a moment's stoppage, to revolve at a high rate, 
inasmuch as twisting is now in progress and the outward 


run has commenced. The click is therefore inoperative 
during the run-out and backing-ofF, and only comes into 
action when the run-in commences. A spring E ensures 
the clip being kept out of contact with the teeth of the 
ratchet Avheel, and brings the rod D back into position 
after it has been acted upon by the stop on the rod E. 

Shaper or Building- Motion. — At this stage we can 
enter upon the discussion and description of the building of 
the cop, a subject so closely allied to the "winding" process 
that neither can be perfectly luiderstood without reference 
to the other. In the analysis of the quadrant it was shown 
how the spindles were made to revolve whilst winding on 
the yarn, so that in spite of a continual variation of con- 
ditions the same length Avas practically wound on during 
each inward run. 

In treating of the building operation it will be shown 
how the yarn is guided on to the spindle in a manner 
suitable to the revolutions produced by the quadrant. 
In considering this question the important point must not 
be overlooked that the quadrant, in its fundamental action 
and jDrinciple, is, Avithin narrow limits, practically an unad- 
justable piece of mechanism. The building of the cop 
therefore must be performed in entire accordance with the 
conditions set up by the action of the quadrant, as far as 
the conical form is concerned. 

On referring back to Fig. 54 it was pointed out that the 
cop underwent several changes in form from the first layer 
put on the bare spindle to the full size of the cop bottom. 
It is the duty of the building motion to guide the yarn on 
the spindle in such a manner as to produce as perfect a 
shape as possible during these various changes of form. 
The subject is not by any means a simple one, and to those 
having charge of the mule, it proves to be one of the most 

136 COTTON SPINNING chap, ii 

delicate and intricate features of the machine, requiring, 
when necessity arises, a considerable degree of skill and 
attention in its management and manipulation. 

A general description of the meclianism will first be 
given, in which its essential features will be pointed out 
and the principal effects outlined. A detailed examination 
will then be made, Avhich -sWll probably be of great practical 
use to those most interested in the subject. 

Fig, 73 presents all that is necessary at this stage to 
enable the action to be understood. The yarn coming 
from the front rollers is guided on to the spindle by passing 
over a wire V. This wire is carried by a series of levers 
or sickles fixed to the copping faller rod W. To give 
motion to the copping wire, this faller rod nuist be actuated 
from the building motion, and it will partake of a species 
of oscillating action, producing an upward and downward 
movement of the wire. A special lever 0, usually called 
a "sector," is fixed to the copping faller W; one end 
carries the wire Y, Avhile the other is hinged at IT to a 
pendant arm T, called the " faller leg." The lower end of 
this faller leg rests upon a stud R, which carries a bowl P. 
The bowl, during the travel of the carriage, runs ujDon the 
upper edge of a specially formed rail, called the " copping " 
or "shaper" rail. It will readily be seen that if this rail 
is inclined in any Avay, the bowl will rise and fall as the 
carriage moves in or out. During the outward run, the 
faller leg is quite free from the slide that carries the bowl ; 
but, as already explained, the backing-off brings about a 
change, which causes the leg to be raised, until a recess on 
it passes over the upper end of the slide at Q, and in this 
position it rests uijon the slide Q, and is said to be locked. 
AVhen the inward run commences, the bowl P occupies the 
first position, as shown on the sketch, and as it passes over 



the surface of the rail at B it rises up until it reaches the 
second position at Y. This of course lifts up the Avhole 
of the faller leg T, and causes the wire at V to be corre- 
spondingly depressed ; " crossing " takes jDlace during this 
period, and a few coarse spirals of yarn are wound on the 
spindle. As the bowl passes from 2 to 3, it descends the 
inclined rail A, and the faller leg falls. This causes the 
wire V to be raised, and while this takes place the yarn 
is being guided in close S2)irals upon the cop. At the 
end of the inward run the faller leg is freed from its con- 
nection with the slide Q E,, so that the wire is raised out 
of contact with the yarn, and the outward journey of the 
carriage is made without the shaper in any way affecting 
the copping faller. 

Crossing.- — The preceding remarks will have conveyed 
the essential idea of the builder motion. We can now go a 
step further, and point out the diftcrence between the two 
inclines on the main copping rail. The earlier portion on 
which the bowl travels as the carriage goes in, is short com- 
pared with the later portion, although the vertical height 
through which the faller wire passes is pi'actically the same ; 
this means, of course, that the wire falls, and therefore jiuts 
the yarn on the cop much more quickly during the down- 
ward movement than during the upward movement. In 
doing so, the result is that the yarn is laid in a coarse series 
of spirals one way, and in a closely arranged series the other 
way, thus producing a foundation that is strong and not 
easily unravelled through carelessness. The proportion of 
length that the shorter incline bears to the longer one does 
not, however, give any idea of the respective rates of motion 
of the downward or upward movement of the wire ; this 
can only be obtained by a careful consideration of the speed 
of the carriage during the traverse over the two parts of 


the shaper rail. By suoli observation it will he found that 
the dowHAvard motion is performed considerably quicker, 
compared with the i;p\vard movement, than a comparison 
of the two lengths would lead one to expect. This point 
is mentioned because there is a tendency to use simply the 
length of the inclines for a comparison of the two move- 
ments, thus conveying an altogether erroneous impression. 

The word " crossing " is generally used to describe the 
quick downward movement of the wire ; and when it is 
completed the bowl Y is on the point Avhere the two inclines 
meet. When in this position it is the usual practice to 
have the quadrant screw vertical ; but of course other con- 
siderations of a practical character may lead to a variation 
from this practice, and it may remain a factor to be regulated 
according to the requirements of the machine ; therefore 
no hard-and-fast rule can be laid down in regard thereto. 

AVe will now refer again to Fig. 54. The drawing shows 
what other conditions must be fulfilled by the shaper 
mechanism in order to Ijuild a cop. In the first place it 
Avill be noticed that the cop commences with a short layer 
at A B, and each layer afterwards is made longer, as at J, 
Ct. The shaper must therefore be adapted to produce this 
result ; in other words, the inclines of the rail must be 
altered in such a way, that for each inward run, the vertical 
height between the highest point of the rail and the lowest 
must be made to increase. 

Again, it will be noticed that in addition to the increased 
length of the traverse of the faller wire (" chase " it is 
generally called), the finishing point of the downward 
movement, and consequently the starting point of the 
upward moA'ement, is not quite so low after each lajer. 
This is quite a small difference when each layer is considered 
by itself; but taking the cop as a whole it results in a 


conical end lieing formed on the lower part of tlie cop as 
at A, E, J. The shaper must be arranged to produce this 
result, for the strength of the cop depends ecjually upon 
this lower end being well formed as upon the upper conical 
portion. On reference to Fig. 73, the complete arrange- 
ment is shown by which the above conditions can be fulfilled. 
The shaper rail is represented as resting, by means of pins 
or small bowls at E and K, upon short inclined surfaces ; 
these are termed front and back inclines, and are connected 
by a rod N, so that any movement is produced equally in 
each. It will readily be understood that if the front and 
back inclines are moved, the ends of the shaper rail will be 
raised or lowered as the case may be, and Avill thus alter 
the position of the path along which the bowl travels. This 
will cause the yarn to be put on the spindle in a new position. 
To have this position correct, it is necessary to move the 
inclines in a special way, and also to have the inclines so 
formed as to give the required shape. The movement of 
the inclines is brought about by means of a screw L work- 
ing in a nut carried by the front plate F. Supports from 
the floor fixing prevent the screw having any horizontal 
movement, so that any motion given to it produces a move- 
ment in the front plate, and this is transferred to the back 
plate through the connecting rod N. The screw is actuated 
from the carriage during each draw, through a ratchet 
Avheel M fixed on one end of it. This Avheel plays an 
important part in the working of the shadier, and attention 
will more fully be drawn to it at a later stage. 

It may then be briefly stated that the inclines have their 
surfaces F and K, as a rule, unequally inclined to one another 
to produce variations in the chase ; the earlier, or higher, 
portions differ in form from each other, in order to produce 
the bottom conical surface called coning ; the inclines are 




^-k M 




moved, in ordei' to lower bodily the whole of tlie rail and 

thus make the cop longer ; the shorter 

incline B on the copping rail is made 

loose, so that it can be guided on the 

short incline D in such a manner that 

the faller leg can be locked always 

about the same distance from the 

" nose " of the cop, no matter whether 

the cop is short or long. 

The above brief statements will now 
be enlarged upon, and by a series of 
diagrams it is hoped to make the matter 
perfectly clear to the reader. In the 
first place, Avhile dealing Avith the prin- 
ciples, it will not be necessary to make 
the diagrams proportionate to actual 
condition ; and in the next place, it Avill 
be assumed for a short time that the 
tAvo inclined surfaces of the copping rail 
are in one piece. 

Although all the requirements for 
building the cop are carried out in an 
apparently very simple manner, yet the 
difficulty experienced by most people in 
thoroughly understanding the mechan- 
ism, proves the necessity of a little 
more than the usual description being 
given. The motion will, therefore, be i<_z_i 
analysed as completely as possible, con- ' 

sistently with the object of this book. / 1 I 

In the first place, let it be noted / | ; 

how the bodily movements of the back 
and front inclines alter the length of the cop. Tliis is 

142 COTTON SPINNING chap, ii 

illustrated in the diagram, Fig. 74. For the present 
Ave will assume the shaper rail A B C to be in one piece, 
and the front and back inclines to be equal to each 
other and quite straight. According to the diagram, the 
length of the layer of yarn put on the cop would be pro- 
portionate to the vertical distance P between A and B, or 
to N between C and B. If the inclines D and E are now 
moved forward to J and K, a distance equal to M and L, 
the shaper rail ABC Avill fall bodily to the position 
F G H ; and since the front and back inclines are equal, every 
point of the shaper rail will fall an equal amount — which 
in the drawing is shown at Q. The effect of this "lower- 
ing" of the rail is to "raise" the faller Avire so that the 
yarn is wound on to a higher part of the cop and thus 
lengthens it. From this diagram it is an easy matter to 
understand the general principle underlying the method of 
lengthening a cop ; and as the cop is built up by additions 
to its length, the principle remains the foundation of the 
mechanism ; but owing to the necessity of increasing the 
length in a special manner, variations must be made in the 
arrangement in order to fulfil the required conditions. 

A drawing is given of a portion of a cop in Fig. 75. 
From it we shall quickly see what the building motion 
must do in placing the four laj'^ers of yarn shoAvn in the 
diagram. The first layer F G is wound on the bare spindle, 
and is a comparatively short one ; other layers are added 
until the layer K H is reached ; and here we notice that it 
is longer than the first layer. It is, however, from the last 
layer C J of the cop bottom that Ave shall be a1)le to observe 
the changes that have been eff"ected in the form of the cop 
and the layers. In the first place, C J is considerably 
longer than F G, though it is Avell to bear in mind that in 
spite of this there is practically the same length of yarn 





wound on in both cases. In the next place, Ave notice that 
the point on the spindle where the downward movement of 
the faller wire, or "crossing," commences, has been raised 
from G to J ; and at the same time we observe that the 
point for commencing the upward motion of the faller wire 
has been raised from F to C. A comparison of these two 
lengths will show that the finishing point of the "building" 
layer has risen at a quicker rate than the commencing point, 
and consequently the length E C is much less than G J. 
It is this fact that causes the layer to be lengthened ; there- 
fore in arranging the shaper mechanism, one of the chief 
considerations is to so adapt the shaper rail that the point 
on Avhich the faller leg bowl rests, when the "building" 
layer commences, shall not be displaced to the same extent 
as the point which represents the finish of the layer. This 
opens l^23 a very interesting question, and it Avill be profit- 
able to fully discuss it. Fig. 76 represents in a diagram- 
matic form the simplest arrangement ; A B C is the ivail all 
in one piece ; both ends A and C rest, as in the previous 
case, upon inclines, but these inclines, instead of being equal 
to one another, are made with different angles : for instance, 
the front incline D is more horizontal than the back incline 
E ; Avhen, therefore, the inclines are moved forward to J 
and K, the rail ABC will be lowered ; but owing to the 
difference in the inclines, the ends A and C will not fall to 
the same extent, and the angles of the two portions of the 
shaper rail will consequently become changed, and a varia- 
tion Avill be introduced in the length of the layer of yarn 
put on the spindle. The diagram, Fig. 76, shows the 
extent of the alteration occasioned by making the front 
and back inclines of different inclinations. The vertical 
distance between A B or C and B is equal to N ; this, we 
will presume, gives the first layer, as at F G, Fig. 75. If 


the inclines are now moved a distance equal to L and M, 
the end A will fall to F, B will fall to G, and the end C 
will fall to H. A comparison of these distances as marked 
at X, Q, and T respectively, shows that the end C has been 
lowered much more than B or A. Now, since the point B 
represents the beginning of the upward or "Imilding" layer 
(for instance, from F to G, K to H, or C to J), the lowering 
of B to G will be shown on the cop by the change of the 
commencing point from F to C ; and also, since C represents 
the finish of the same layer, we find the termination of the 
layer is much higher up the spindle at J than when the 
shaper rail occupied its first position, which gave the 
terminating point at G. The distance G to J produced 
b)^ the lowering of the rail from C to H is much greater 
than the distance E to C, which is brought about by the 
lowering of the point B on the rail to G. It will be 
observed that only sufficient of the front and back inclines 
D and E have been used to make the cop bottom ; and the 
variation in their inclination has had the eff'ect of raising 
the point F to C (Fig. 75) and so forming a conical end 
on the bottom of the cop ; it has also had the effect of 
lengthening the layer as the cop bottom enlarged. The 
operation just described is generally termed " coning " ; and 
it must be carefully noted and understood that the chief 
essential in producing it is in the difference of the inclina- 
tions of those portions of the front and back inclines on 
which the shaper rail rests while the cop bottom is being 

In connection with the diagram, Fig. 76, it will l)e 
noticed that A and C are in the same horizontal line. 
This means that the yarn in Fig. 75 commences at G, and 
comes down to F, and back again, exactly the same distance, 
to G. It is very desirable that the same thing should occur 

VOL. Ill L 


in the last layer also ; but according to tlie diagram this is 
impossible, for it will be observed that, in consequence of 
the end A not having fallen to the same extent as C, the 
beginning of the downward " crossing " movement does not 
correspond with the finish of the upward " building " move- 
ment. The result is that crossing commences below the 
actual nose of the cop, and since the position of the bowl 
and its carrier, which travels along the shaper rail, regulates 
the locking of the faller leg, Ave get the locking operation 
performed a little later than is desirable. 

The method now adopted of overcoming this fault of the 
shaper rail being in one piece, is to make the front short 
incline loose, so that its inclination can be so regulated to 
give both the crossing and building layers the same exact- 
ness, in order that locking shall take place always at or 
near the nose of the cop. 

Fig. 77 has been prepared to illustrate graphically the 
essential features of the shaper with loose front inclines. 
The shaper rail is A B C ; the part A B, instead of being 
in one piece Avith B C, is loose, and hinged at B, so that it 
can alter its inclination irrespecti^'e of the inclination of 
B C ; by this means it is possible to lower the point A to 
the same extent as the point C, and in this way the locking 
of the faller at the termination of the backing-off will 
always take place at the nose, or, in other words, at the 
highest point of each layer of yarn. 

Let us now carefully examine the arrangement as shown 
in Fig. 77, and see how l>y its means Ave can build the cop 
bottom. To sum up the conditions : it is necessary that 
B C should be capable of altering its inclination ; also it 
must be so arranged that the point B Avill fall to a less 
extent than the end C ; and at the same time the por- 
tion of the rail at A B must be so arramred that the 


end A will be on about the same horizontal level as the 
end C. 

In order to fulfil the first condition, the long portion of 
the shaper rail C B is lengthened out to N, and this end 
rests upon the front incline Q. This incline, so far as the 
cop bottom is concerned (and it is this part of the cop with 
which we are now dealing), has a different inclination from 
the back incline E, so that any movement of the two 
inclines, for instance to S and K, Avill lower the rail C B to 
G H. This differential lowering fulfils the conditions for 
lengthening the chase of the cop bottom, as was shown in 
connection with Fig. 76, and by its means the bottom 
conical end of the cop is obtained. 

The second condition is fulfilled by the end A of the 
loose incline A B resting on a separate incline D, which 
moves forward to the same degree as the back and front 
inclines. The short loose incline is therefore lowered quite 
independently of the position of B, and we can, by making 
a suitable profile on D, lower the end A to any required 
extent necessary for locking the faller leg correctly. 

From the sketch it will be an easy matter to make a 
comparison of the various distances moved ; A and C are 
seen to be on the same horizontal distance, and in this 
position the first layer is put on the spindle. When the 
rail is lowered, F and H are still on the same level, and the 
last layer C J, Fig. 75, is put on in this position. The 
dra^ving will also show that the point B has been lowered 
much less than the ends of the shaper rail, and so we con- 
clude that while the " chase " has been lengthened the 
ascent of the point F has not taken place at the same rate. 

When the cop bottom is finished, all the inclines, Q, I), 
and E, partake almost of the same inclination, and contiiuie 
to build the cop to whatever length is required. The next 


sketch will show the three inclines in such a manner that 
they can easily be compared, and a few remarks will be 
made on the character of their profile. 

It must be understood clearly that in the preceding 
descriptions the principle only of the copping motion has 
been dealt with, and for that purpose only those portions 
of the front, middle, and back plates have been used to 
illustrate in diagram form the building of the cop bottom. 
The inclinations of the three plates were also denoted by 
straight lines. In practice, however, it is found beneficial 
(and indeed necessary) to avoid the sharp angle where the 
cop bottom finished ; and moreover the bottom conical part 
is very seldom straight, but slightly convex in outline. 
These considerations necessitate the use of curved inclines 
specially shaped to produce the desired result. The 
inclinations of these curves, however, differ from one 
another, so that our explanation is not affected. 

In order to show actual conditions, and so that a 
comparison can be made of the inclinations, the front, 
middle, and back plates, taken from a mule, are shown in 
the drawing, Fig. 78. Comparing the front and back 
plates, we find a great difference in their inclination at 
those parts used during the building of the cop bottom, the 
reason for which has been so fully explained, that there 
ought to be no difficulty in thoroughly understanding it. 
In regard to the middle plate, the curve up to F is almost 
similar to that of the back plate, the reason for which is 
almost obvious, since it has been shown that the end of 
the loose front incline of the rail must fall almost in the 
same degree as the back end of the shaper rail. 

There remain yet the other and longer portions of the 
plates to be mentioned. These portions B C, D E, and 
F G are used for building the remainder of the cop after 



the bottom is completed. They are, as a rule, perfectly 
straight in profile, but circumstances may possibly arise 
necessitating a very slight variation from the straight line. 
The inclinations of the middle and back plates are seen to 
be practically alike, but a difference between the front and 
back plates requires some explanation. We have seen that 
a long chase is being made when the cop bottom is complete. 
This may be carried through to the finish of the cop, but 

frequently it is caused to shorten, the object being to gain, 
compactness and weight. The difference of inclination of 
the two plates brings about this result, as was pointed out 
in diagram. Fig. 75. In that illustration, however, the 
difference caused a lengthening of the chase ; but in Fig. 78 
the opposite effect is produced, because the back plate has 
less inclination than the front one. 

Long" Incline of Shaper Rail. — So far, it has been 
assumed that the long incline of the shaper rail is an 
inclined surface on which the shaper bowl travels to and 
fro ; but a little consideration will show that this incline 


must be made a special shape in order to lower the bowl in 
such a manner that the faller wire guides the yarn on the 
cop in a series of regularly spaced spirals. This point 
will now be examined. It has already been shoAvn that 
each revolution of the spindle, during winding, winds on 
unequal lengths of yarn. We also know that each length 
of yarn wound on represents the distance moved by the 
carriage. A further condition known is, that the faller 
wire must rise equal distances for each revolution of the 
spindle, this being necessary if the spirals of yarn are to 
be spaced equally. 

Having these facts as our guide, it is an easy matter to 
find the outline or shape of the rail required. A cop has 
been carefully unwound and the length of each turn of 
yarn on the cop measured. These various lengths have 
been marked out on the line 1 to 22 in Fig. 79, so that in 
reading the drawing it must be understood that as the 
carriage moves from 1 to 2 the spindle has revolved once 
and wound on a length of yarn ec[ual to the distance moved 
by the carriage, viz. from 1 to 2. The same thing occurs 
in the carriage moving from 2 to 3, and so on through all 
the positions marked up to 22, when the stretch is com- 
pleted. It is to be noted that 21 revolutions of the spindle 
have been made and that each revolution has Avound on a 
less length than the preceding one, so that a long length 
has been wound on in the first revolution and a short 
length in the last revolution. But it is readily seen that 
although all the lengths wound on are unequal, we still 
know that they differ from each other by a practically 
equal amount. 

Now that these lengths have been measured off on 
Fig. 79, the long incline of the rail is drawn in, and 
the distance between the highest j)oint or shoulder and 



the lowest point is divided up into eriual divisions corre- 
sponding to the number of revohitions of the spindle 
during winding. (This can be calculated, but it is much 
easier to count the turns by using a well-made cop.) This 
equal division is done because of the equal pitch of the 
spirals. By drawing vertical lines through the carriage 
positions and horizontal lines through the shaper bowl 
positions we obtain intersections through which a curve 
can be drawn, and this curve is the shape required on the 
long incline of the rail in order to move the faller wire in 
such a way as to guide the spirals of yarn on the conical 
surface at equal distances apart. The curve on the rail 
is parabolic in character, and it clearly cannot be correct 
except for a certain layer or a series of layers all of Avhich 
are equal. The layers in the length of the cop after the 
cop bottom is finished are the most numerous, and therefore 
the rail is made to suit these layers. The cop- bottom 
layers are not so regular in their spacing, and, moreover, 
it so happens that the action of the quadrant not being 
absolutel}' perfect, the two qualities counteract each other 
somewhat and thus enable a fairly perfecth^-shaped cop to 
be made. 

It must also clearly be understood that the question of 
absolutely equal spacing of the coils on the cop is assimied 
to some extent. Careful measurements of cops show 
differences from different types and makes of mule, and 
when it is remembered that practically all shapers as 
at present used are the result of cutting, carving, and 
filing, such differences taust be expected and allowed 
for ; they are compromises in adjustment both for the 
quadrant and the curved edge of the long incline of the 
shaper plate. 

Another feature of the shaper motion that calls for 


some explanation is the \;se of the small inclined floor 
bracket X, as shown in Fig. 72. This bracket is generally 
known as the " steadying " bi-acket. A diagram has been 
prepared in Fig. 80 in order that its action may be 
explained, but in this connection it must be observed that 
the reasoning applies only to the old form of shaper. 
Loose incline shapers do not require an inclined steady 
bracket, and although they are generally used the inclines 
are formed to compensate for them. If a vertical slotted 
bracket was used it would simply mean altering the shaper 
ratchet wheel. The rail ABC rests on the two inclines G 
and J, and a pin X in the long rail fits in an inclined groove 
of the bracket L. As the inclines are moved to H and K 
the rail is lowered, but instead of falling vertically the 
groove of L causes it to fall in the direction of the incline 
and to take up the position shown by dotted lines at D, E, 
and F. The chief effect of this loAvering of the rail is to 
give a horizontal movement to it, so that the relative 
amounts of the short and long inclines of the rail are 
altered, and the highest point of the rail is moved inwards 
to the extent shown at T. Several reasons are assigned 
for this action. The principal one may readily be under- 
stood when Ave point out that a longer time is taken up in 
crossing and a shorter time in winding; the yarn is 
consequently relieved of considerable strain as the cop 
lengthens, because since backing-ofF unwinds less and less off 
the bare spindle as the cop lengthens, it is advisable to do 
the crossing a little more gradually, and this can be done by 
taking more time to do it in. "Cros.sing" in any case 
induces an unusual strain in the yarn ; but in the earlier 
stages of building there is so much yarn over and above 
the actual length of the stretch that the strain is not of 
any moment. As this surplus yarn becomes less, it is 


almost necessary to adopt a relieving action to prevent 
the rapid winding, which takes place during crossing, 
from breaking the yarn ; and this is obtained by lengthen- 
ing the time during which the crossing is performed. 
Of course a guide bracket of some kind is necessary to 
guide the rail in its descent, and for this purpose a straight 
bracket would be effective. The inclined bracket, however, 
serves another purpose in giving greater stability to the 
rail ; for, since its inclination is opposite to the inclined 
plates on which the rails rest, it prevents vibration and 
keeps the shaper steady ; hence its name. 

Shortening the long part of the rail has the effect of 
winding a little more tightly, and thus helping in com- 
pensating for the diminishing diameter of the spindle. 
The positions of the faller-leg slide bowl are shown at M, 
N, and Q ; and as the start and finish are always the same 
throughout the cop they begin and end at the same place, 
Q and S, at the finish; only the position at the highest 
point of the rail is altered from N to R ; but as previously 
observed, this alters the relative lengths of the short and 
long portions of the rail. 

Defective cops and their remedies. — To the 

practical reader the foregoing description of the shaper, 
and the explanation of its principle, will be of great 
assistance as an aid in solving many of the problems 
associated with the formation of the cop. It is in connec- 
tion with the shaping mechanism that perhaps the greatest 
amount of intelligence and skill is required in managing 
the self-acting mule of to-day ; and, as good and bad results 
of its working concentrate themselves to a large extent 
upon the shape of the cop, the subject of defects in its 
formation may be made the occasion for investigation. A 
very great number of imperfections arise in the shape of 


the cop either locall}^ or generally. To enumerate them 
all would luiduly extend the limits of this book, but the 
importance of the subject necessitates that some attention 
should be given to the most characteristic faults that arise 
in connection with the cop. 

Badly-formed cops are not always the result of a faulty 
shaper, so that when a decision has to be given upon the 
cause of any particular irregularity in the cop's condition 
or shape, a very careful investigation into the matter is 
required before fixing on the exact point for correction. 
In the following notes, therefore, it must be understood 
that the remedies pointed out are only suggestions of what 
may be the cause. We shall restrict our attention first to 
defective cops x'esulting from other causes than those directly 
connected with the shaper rail or its inclines. 

(1) Cops, instead of being perfectly parallel, may be 
very ridgy on their surface. Several causes are capable 
of producing this result. For instance, the bowl that runs 
along the shaper rail may be worn flat in one or more 
places ; it may be loose on its stud ; or it may be badly 
mounted and work on its edge instead of its full width. 
The bowl also on Avhich the faller leg locks, if faulty in the 
same way as the rail bowl, Avill cause ridges. The fault 
can be corrected by returning the bowl and replacing the 
studs, or so mounting the bowl that it runs level on the 
rail. The ridges may be produced by the copping faller 
rod not working smoothly in its bearings through ha'sing 
play in the faller stands, especially in those near the head- 
stock. The shaper screw may not be perfectly true, and 
its irregular movement will give unequal advances to the 
inclines ; a good screw will remedy this fault. If the collar 
which fits the screw binds, it may also cause a ridgy 
appearance. One frequent cause of ridginess is due to 


the tumbler not taking the teeth of the ratchet wheel 
regularly ; this gives an irregular movement to the inclines, 
and the irregularity is reproduced on the cop. Other 
causes of ridgy cops may be found in a loose backing-off 
sector ; in a carriage not being firmly fixed in the square ; 
and occasionally, through not gearing the c^uadrant pinion 
deep enough, ridges have been produced. 

(2) Cops may be longer at the outer ends of the carriage 
than those nearer the headstock. Weakness in the faller 
shafts is the chief cause of this defect ; and in some cases, 
if the weights are too heavy and placed too near the 
outer end of the carriage, the same fault arises. Faller 
shafts must be strong enough to resist torsion, and the 
weighting must be arranged to obtain a uniform strain 

(3) Cops may be soft throughout the mule. It is Cjuite 
possible that cops should be made soft, especially if cotton 
of a poor quality is being used ; such cotton cannot resist 
breakage so well as better cotton, and accordingly the 
weighting of the under faller must be much less in order 
to prevent breakages when backing-off takes place or 
winding. Cops become soft, however, when such a con- 
dition is neither necessary nor desirable, and it may arise 
from the following causes : — 

Winding may be badly performed, that is, the quadrant 
may turn the spindles too slowly. To remedy this the 
quadrant must be put back a little. It is a frequent 
practice to put the quadrant forward so as to obtain easy 
winding and avoid breakages ; but it results in a soft cop. 

If the driving strap touches the fast pulley during the 
run-in, the result will be soft cops ; see, therefore, that the 
strap when on the loose pulley is quite clear of the fast 


Sometimes, after years of ■work, the highest point of 
the shaper rail, where the two inclines meet, becomes worn 
and flat; this makes the shoulder of the cop larger and 
softer throughout its length. 

Faller rods sticking in the stands is an occasional cause 
for soft cops ; and sometimes a difference in level between 
the slips on each side of the headstock has produced the 
same result. 

If, during backing-off, the tin roller slips a little on the 
shaft we get soft co])s. 

"When the softness appears in the cop after a certain 
length has been made all right, and the mule has previously 
made a good parallel cop, it shows that the nosing has not 
been carefully performed, or has even l)een neglected. To 
cover this kind of carelessness the winding is slackened a 
little, which causes larger shoulders to be made and corre- 
spondingly softer cops. 

(4) Cops may be soft close by the headstock only. In 
such a case, the drawing-up scrolls may be too large for 
the length of stretch. Anything wrong with the scrolls, 
such as being loose on the shaft, wrongly set, or not in 
right position, will give to the carriage during the run-in 
an irregular movement, and so cause softer cops at one 
part than another. If the back shaft is too weak we get 
a similar result. 

(5) When thick and soft noses are made, the following 
may be sought for as the cause : — 

Delaying tlie use of the nose peg or nosing motion. 

The under faller being depressed too soon before the 
copping faller has been unlocked. 

When the backing-off chain is too slack wc get a frequent 
cause of soft noses, and correspondingly, if it acts too 
quickly. A defect in the shaper rail which causes a hollow 


at the upper end of the chase produces slack yarn at the 
termination of winding, and in this way either snarls are 
made or soft noses. 

We can now deal with badly-shaped cops whose faults 
may be traced directly to the shaper. A few typical cases 
will be examined, and remedies suggested whereby a 
correct form may be obtained. 

In the first place, a thorough examination must be made 
to see that the shaper mechanism is in perfect Avorking 
order ; the studs firmly fixed, the bowls quite round and 
set without " winding " ; the faller sickles, especially the 
faller sector, connected to the faller leg, must be securely 
fastened on the faller rods ; no dirt or waste ought to lie 
about the copping motion to prevent the free working 
along the slides of the inclines ; and the faller rods ought 
to be well supported in their stands, and perfectly free to 

Before flying to the shaper for a remedy, one must 
be quite sure that the fault does not arise from some 
other defect in the machine. For instance, the motions 
may not be acting in unison ; a faller wire in its move- 
ment may catch some chain or bracket ; or weights may 
be touching the floor or some fixing, and thus jDroducing 
irregularities in tension. These and a number of other 
apparently small matters all have some influence in aff'ecting 
the building of the cop, and no improvement can then be 
effected through the shaper. 

In the accompanying sketch, Fig. 81, a few well-recog- 
nised faults in cops are shown. The one marked A may 
be taken as a standard by which the others may be com- 
pared. B represents a cop ridgy in its body part instead 
of being pei'fectly parallel. All the other cops show in 
their full lines variations from the dotted lines, which 



should give a, cop similar to A ; Ave can in this way sec 
whether too much or too little yarn has been placed on the 
parts that are faulty. Of course it will be understood that 
a combination of the faults illustrated may be found in 
a cop at one time : for instance, a ridgy cop can have a 
hollow bottom and a soft nose. 

Fig. 81. 

In suggesting alteration in the shaper for remedying 
the defects illustrated, a warning must be given that the 
utmost care possible is absolutely necessary in making the 
correction by filing the plates. It should always be done 
gradually, and in order to gauge the alteration accurately 
a template or exact cojiy of the plate should be made 


previous to the filing ; and its position should be definitely 
noted, otherwise the labour of hours may be wasted and 
the work required to be begun again. A few words of 
general purport, but nevertheless important, will probably 
fix the general principles in mind and serve as a guide in 

The bottom cone of the cop is formed by the earlier 
portions of the front and back inclines, namely, the upper 
short flat portion of the front incline and the short curved 
portion of the back incline ; in most machines marks are 
put on the plates showing the starting point. Faults in 
the lower conical portion of the coj) may be corrected by 
attention to these parts of the plates. 

The long parallel body of the cop is formed by the 
straight portions of the front and back inclines ; remedies 
for defects that appear in the body may be sought for at 
these points of the plates. 

The upper cone, or chase of the cop, is obtained from 
the shaper rail, to which attention must be directed Avhen 
faults appear in that part of the cop. 

Desired alterations in length of chase, length of the 
body, or length of the bottom cone, can be obtained by 
raising or lowering the plates that support the rail. 

A bulging or "lump'' on any part of the cop is due to 
a "hollow" in the rail or plates, and its location is easy, 
by observing the position of the bowl on the rail as the 
faller wire is guiding the yarn on the lump. Correspond- 
ingly, a hollow place on the cop is due to the rail or plates 
being too high or lumpy at the point ; its position can be 
found as above. 

Let us now trace out the alterations required to correct 
the faults in the cops illustrated in Fig. 81. 

(B) Kidgy Cop. — This fault is due (if not to the causes 


already given) to a similar condition of the straight portions 
of the front or back inclines, or both. Examine their 
profile and see Avhether it is irregular ; if so, file carefully 
until a good profile is obtained. When, instead of a 
number of ridges, there is only one or two, locate the 
spots as directed above, and file until a straight body is 
given. Be careful to note whether a series of irregularities 
on the body of the cop are ridges or hollows. This is 
important, for as in the sketch B the bottom is finished 
the correct diameter, so the irregularities at R are really 
hollows. By filing the plate to correct the hollows the 
coning portion would be slightly shortened, and conse- 
quently the next cop would be thinner in diameter. To 
keep the diameter correct, therefore, we must take a parallel 
filing off the coning part of the incline, so as to have the 
same length as before. 

Again referring to Fig. 81 : two common faults are 
represented at C and D. The narrow thin form s in C is 
due to a too quick fall of the rail on the coning parts of 
the inclines ; while at T in D it is thicker than is desirable, 
and is caused by too little or slow a fall of the rail on the 
coning parts. A series of diagrams in Fig. 82 will enable 
us to point out the necessaiy changes to be made in order 
to correct these faults. To thicken out the bottom at S, 
Fig. 81, to fill up to the dotted line, the rail must fall 
slower, and to do this we must not start so high up ; the 
beginning must be brought a little nearer to the finish (see 
No. 1, Fig. 82). If the full lines represented the starting 
points for the cop C, an alteration to the dotted lines would 
cause a slower descent of the rail (because the fall is not 
so steep), and produce a thickening of the bottom cone. 
At the same time as this is done we reduce the A'ertical 
height through which the rail falls Avhile on the coning 
VOL. Ill M 


part, and consequently shorten the bottom cone. As a 
rule this is not an objection, as a hollow bottom is generally 
associated with a long one, and so both faults are corrected. 
In regard to D, the thinning of the bottom is brought about 
by adopting an opposite course to that suggested for the 
specimen C ; the coning is therefore lengthened a little by 
starting a little higher up the plates. 

In the specimens marked E and F we also find a state 
that is occasionally troublesome ; E, it Avill be seen, is too 
short in the bottom cone, while F is equally too long. To 
a certain extent they are only a variation of the effects 
noticed in the copa C and D ; this fact, however, will be 
treated of a little later. 

In the first place, to deal with E : the shoulder requires 
to be raised, and to do this it is necessary to have a greater 
fall of the rail between the start and finish of the coning 
parts. This is effected by filing off a portion of the plate 
at the finishing point of the coning and going up to nothing 
at the starting point. (See No. 2, in Fig. 82, as marked 
by dotted lines.) 

In the cop F the length from the bottom end of the cop 
to the shoulder requires to be shortened, consequently we 
adopt an opposite course to the above ; for instance, the 
fall in the coning part must be lessened, and to do this the 
filing must start at nothing on the finishing point of the 
coning part and finish at the necessary depth at the starting 
point. No. 3 in Fig. 82 will illustrate this. 

In the above remarks the suggestions made are probably 
the most delicate matters that can be found in the whole 
range of cotton spinning. The operation of filing a plate 
on the coning part is one that requires the utmost care, 
and generally another fault appears whilst correcting the 
first one. The few remarks already made will indicate 



to the thoughtful reader in what way this arises. It was 
shown in regard to the specimen C that thickening the 
bottom meant also shortening it ; and in making D thinner 
Something of the kind occurs when 

we also lengthened it. 


Fig. 82. 

dealing with the faults at E and 1). To lengthen the cone 
part at E by filing the plates as suggested, we naturally 
lessen the vertical distance between the start and finish ; 
but at the same time Ave thin the bottom, and care is 
required that by doing so we do not exceed the jiroportions 
wanted. The sketch shows how necessary it is to thin as 


well as to lengthen it ; but great care is absolutely necessary 
in order to maintain a good shape even when the correct 
length is obtained. To shorten the cone at F we adopt a 
course of filing that is not nearly so difficult as at E, but 
the thickening effect on the cop is inevitable, although not 
relatively so great as at E. The chief difficulty lies in the 
connection of the coning parts with the straight inclined 
parts of the plates ; and it is sometimes necessary, to keep 
the coning parts their original lengths (although the vertical 
fall has been reduced, or rice versii), to file a parallel strip 
off" the full length of the straight 2:)art. 

While the diagrams may help in conveying an idea of 
the parts to be filed under certain circumstances, it is only 
by practical experience that the amount, and the foi'm the 
filing ought to take, can be decided. 

To continue the specimens, we will consider faults that 
may arise in connection with the nose of the cop. Four 
faulty cops are shown in G, H, J, and K. In G a hollow 
cone is made, and this at once suggests that the long rail is 
too high in the middle ; a correction can be made by filing 
the rail flatter. In respect to H, where the chase is round, 
a remedy is almost invariably found by making the rail 
Avith less fall between the highest point and the lowest ; or 
the highest point may be lowered by means of the adjusting 
screw which is connected with the bowl on the front plate; 
then file a little off" the outer end of the rail, and continue 
to nothing about the middle of the rail, or even further if 
it is found necessary. If the hollow is not a general one, 
as shown in the sketch, but only local, then locate the spot 
on the rail which produces it, and file on either side of it. 

Specimens J and K are easily corrected by simply 
altering the vertical height between the highest and lowest 
points of the rail. 


"When the cops are not formed parallel, they may either 
get smaller in diameter as the cop builds, as at L ; or larger 
in diameter, as at M. In either case carefully examine the 
faller sector ; as a rule the centre of the stud connecting 
the sector to the faller leg is in a line with or opposite to 
the centre of the faller rod when the faller wire is in the 
centre of the spindle blade. If the wire is higher than the 
centre of spindle blade the cop will become smaller in 
diameter as it builds ; and if lower, the cop will be formed 
with an enlarging diameter. Granted that the sector is in 
the correct position, a remedy may be suggested by filing 
the plates for L as shown in No. 5, and for M as at No. 4 in 
Fig. 82. 

In all these diagrams the amount of filing is exaggerated 
in order that the " direction " might be distinctly shown ; 
in almost all cases very little filing will be required, but 
whatever is done must be done carefully. 

A brief description, with illustrations, of the faller wires 
during the inward and outAvard run of tl^e carriage was 
given in an earlier portion of these pages, and in the 
preceding notes on the shaper, the position and movement 
of the winding faller wire have been fully discussed ; it 
therefore remains to direct a little attention to the other 
action and methods of controlling the counter, or under 
faller wire. 

Weighting the Fallers. — In different districts various 
names are given to the same features of the mule, so care 
must be taken in reading descriptions to follow out the 
references to the illustrations. In the accompanying 
draAvings, Figs. 83, 84, and 85, the counter or under 
faller rod is shown at A, while B represents the copping 
or winding faller rod. The wires carried by these two 
rods are marked H and J respectivel3\ As the carriage 


comes out, and the spinning process is going on, these 
two wires are inoperative, both occupying positions, as 
already ilkistrated, close to the spindle point, but per- 
fectly clear of the yarn being twisted. They are practi- 
cally locked during the whole of the run-out. When 
the carriage comes to the end of the stretch, backing-off 
takes place, and, as the yarn is unwound from the bare 
part of the spindle, the winding faller wire J comes down 
in position ready for winding, while the counter faller wire 
H rises in order to take up the surplus yarn that has been 
unwound. In doing this it makes the yarn taut between 
the rollers and the cop. So far we have simply indicated 
that the wire is carried by a series of sickles G from the 
counter faller rod A, and they being all on one side, the 
tendency is rather for the wire H to fall rather than rise. 
In order, therefore, to cause H to rise and put tension in 
the yarn, the sickles and wire must be balanced on the 
opposite side of the rod A. To do this a sector C is keyed 
to the faller rod, and to it is attached a chain, which is 
hooked at its lower end to a weighted lever E centred at 
F. There are several of these chains and levers in the 
length of the mule, and their direct effect is to lift up the 
wire H as backing-oif proceeds. 

Now it will readily be seen that this arrangement is a 
very important feature ; on the careful balancing effect of 
the weighted lever E depends the amount of tension in the 
yarn during the winding, for since A is free to oscillate in 
its bearings, it is quite possible that H might be forced too 
high and severely strain or even break the whole of the 
"ends," as the yarn is sometimes termed; on the other 
hand, the weight may be insufficient, and so produce a 
slackness that would result in the j^arn running into snarls, 
and, in addition to the bad yarn so produced, making a soft 



and misshaped cop. To carefully adjust the balance, 
arrangements are made on E for applying additional 
weights "W until the required tension is obtained. The 
skill of the minder is shown in his ability to gauge the 
tension, and while for the stronger yarns great care is 
necessary, it becomes a very delicate operation when the 
finer qualities are being spun. The character of the cop, so 
far as its density is concerned, is regulated b}' the tension. 

Fig. S3. 

Fig. S4. 

Fig. S.J 

and it is an easy matter to make it too hard or too soft by 
carelessness. We haA'e already pointed out how essential 
it is that the levers must be perfectly free from interference 
whilst they are in action, and also hoAv it may be necessary, 
in order to neutralise tension, to weight the levers more, 
nearer the headstock than at the outer ends of the carriage. 
Directly associated Avith the arrangement just described 
is the device illustrated in Figs. 84 and 85. They are 
shown in separate drawings in order to avoid complication. 
As the carriage runs in, the copping faller is guiding yarn 


on the spindle by virtue of its connection to the shaper. It 
is doing this in opposition to a strong spring D, Fig. 84, 
which is attached to the winding faller B by a strip of 
leather C and at its lower end to a fixed bracket E. Tlie 
only effect of the spring at this point is to keep the shaper 
bowl pressed against the rail ; when, however, the run-in 
is complete and the faller leg is unlocked, the tension in the 
spring D instantly causes the winding faller wire to move 
upwards a little above the spindle point. To prevent the 
wire being pulled up too far, a bracket S in the rod. Fig. 
85, has a projection which, when the winding faller reaches 
its correct position, comes on the top of the counter faller 
rod and so prevents any further movement. The action of 
the spring D must be of a very definite character and 
strong enough to overcome, during the outward run, the 
weighting of the counter faller. 

This last remark will be understood on reference to Fig. 
85. It was remarked a short time ago that the two faller 
wires occupied certain positions during the outward run of 
the carriage. These positions are regulated by an indirect 
connection of the faller rods to each other. On the copping 
faller rod the bracket S has attached to it at T a link U, 
which is connected by a rod V to a projection on the 
weighted lever E. During the inward run, the rod V is 
adjusted to be free from the influence of the weight by so 
arranging it that it is able to pass freely through the hole 
in the projection X ; Avhen, however, the spring D pulls the 
copping faller wire upwards, on the completion of the run- 
in, it also pulls tlie rod V in the same direction, and in 
doing so, the nuts Z on the lower end of the rod come into 
contact with the projection on the lever, and so lift the 
lever also upwards, and consequently relieve the counter 
faller of their weight. This lifting of the lever E is a 


definite amount, because the spring D can only pull E 
upward until the projection on S rests on the counter faller 
A. The counter faller wire, being free from the balancing 
effect of the levers E, naturally falls by gravity until the 
chain D, Fig. 83, is again taut, and when this occurs the 
two wires occupy their correct position for the outward 
run, and the spring D must be strong enough to maintain 
them in this position so long as spinning is in progress. 

Easing Motions. — The dotted portion of the drawing 
(Fig. 83) belongs to a class of mechanism called easing 
motions. The present illustration is given here for the 
purpose .of explaining the necessity and effect of such a 

It has been explained how the weighted levers E are 
partially supported from the copping faller rod B during the 
outward run of the carriage. When, however, the traverse 
is finished, a " change " takes place for the purpose of 
backing-off, and at this moment of changing, the full weight 
— or, rather, effect — of the weighted levers E woT\ld be 
thrown suddenly on to the counter faller A. Such an 
action is not necessary nor desirable ; rather the reverse, 
for it is easy to understand that the free yarn unwound 
from the bare spindle during backing-off requires to be 
taken up gradually and gently. To bring this about, an 
"easing" device is applied, whereby the weighted levers 
are partially freed from the influence of the weighted levers 
E. On the opposite side of the faller rod A to that in 
which the sector C is fixed, is fastened a lever K, to which 
is attached a spring M, by means of an adjusting screw L ; 
the other end of the spring is hooked to one arm of an L 
lever carried by the bracket P. The L lever has its other 
arm Q in such a position that just before the carriage arrives 
out, it comes into contact with an inclined floor bracket It, 


and this, preventing Q from going further forward, de- 
presses the other arm N, and so puts tension into the spring 
M. This tension in M, acting in the opposite direction to 
the pull of the weighted lever E, neutralises a great part 
of the weight, and prevents the shock that would otherwise 
come upon the counter faller wire ; in other words, it eases 
the faller rod considerably, and allows it to move upward 
in a much more gentle manner as backing off proceeds. 
When backing-ofF is complete, and the carriage runs in, the 
arm Q of the lever moves gradually out of contact with the 
inclined bracket R, and so, by destroying the tension in the 
spring M, permits the full effect of the weighted lever E to 
fall upon the counter faller wire, and so maintains the 
tension in the yarn. 

In passing, it may be observed that several of the 
weighted levers are placed in the full length of the mule, 
but the easing motion is only applied to two or three of 
them, a little discrimination being necessary in setting and 
disposing them in order to obtain the best results. Means 
are supplied for the necessary adjustments both for tension 
and position, the screw L supplying the former, and a 
regulation of the bracket R the latter. 

The Effect of a Tapered Spindle on the Winding. 
— Mention has been made of the necessity for tapering the 
spindle blade towards its point, and that this fact had an 
important bearing upon the problem of winding after the 
cop bottom was finished. AVe will now examine this 
question as fully as possible, so that, in considering the 
mechanical methods adopted for compensating for the taper, 
we shall be able the better to judge of their adaptability 
for the purpose. It will be advisable to recapitulate a little, 
so we will begin by showing the necessary winding effect 
required for building the "chase" of the cop at the point 



when the cop bottom is finished. Figs. 86 and 87 will 
illustrate the explanation. In Fig. 86 the chase of the cop 
at A D is shown soon after the cop bottom is complete. 
At this time Ave will assume that the full diameter of the 
cop is 1 \ inch and the diameter of the small end of the cop 
on the bare spindle at D is -^^ inch. To wind correctly 
this conical surface A D, the speed of the spindle must 
increase in the ratio denoted by the full-lined hyperbolic 

,- ^..^.^ 

NT2 UE- . 
5 MROE ; 
.TE-. ; 1 

1 ;' 
c- ". ' 










SPEED m^^E^ 

WING, the: ijlFFE' 
THE 'cOPSO-Vtom' 
'e "COT^ i^ 06MPU 




«M. «" ' 

f ""j" """ 




Fig. 86. 

— l.E,NG,TH oc CHASE:- 

FiG. ST. 

curve, as shown in the diagram, Fig. 87. For instance, 
suppose one revolution winds on a certain length of yarn 
at A, then the " rate " of spindle speed must be such that 
the same length could be wound on at B, C, and D, and in 
order to do this the proportionate speed at these points 
must be increased to IJ, 2, and 4 times respectively the 
speed at A. It is a very simple matter to obtain these 
numbers, for, knowing the diameters at the points A, B, C, 
and D, the proportion each of them bears to the diameter 
of A Avill give us the proportionate speed. 


Suppose A is 1:^ inch diameter and its speed 1, 
then B is \% inch diameter and its speed is -\ =\ x 11 =1^> 


and C is f inch ,, ,, ~t~^ ^ f =2, 

and D is y°5- inch ,, ,, -^ = JxJ^ = 4. 

These numbers, measured by any convenient scale along 
the speed line A 10 and horizontal lines drawn through 
them to meet perpendicular lines erected upon the chase 
line A E, as at B, C^ and D, Avill give points through Avhich 
the curve A^ D^ can be drawn. The line A D is drawn any 
length to represent the chase, and the points B and C are 
marked off equi-distant from each other in the same way 
as B and C on the chase line in Fig. 86. As the cop 
builds from the cop bottom upwards, the body remains the 
same diameter, namely, 1^ inch, but the nose of the cop 
gradually becomes smaller until near the end of the spindle, 
Avhen it is wound on a diameter of probably \ inch. In 
consequence of this reduction in the terminal diameter of 
the conical chase, a new set of conditions are introduced, 
which have an important bearing on the problem of wind- 
ing. For the moment, Ave will assume that the chase A D 
lengthens until it meets the smaller diameter, as at A E ; 
if this occurred, all that would be required would be a con- 
tinuation of the accelerated speed which formed the poi tion 
A D, and in the curve in Fig. 87 the speed curve would 
be lengthened as shown hx the dotted line D^ E\ In such 
a case one might reasonably say that a " nosing " motion 
would be a device for accelerating the speed of the spindle 
as the yarn Avas being Avound on the nose of the cop ; indeed, 
the word "nosing" is derived from such an idea, and it is 
perpetrated partly because of the })revailing idea that such 
a method of reasoning is fairly correct, and partly because 


the irregularities of Aviiiding show themselves more at the 
nose than in other parts of the coj). But we know that 
instead of the chase lengthening towards the spindle end, 
it is generally kept either the same length throughout, or is 
made a little shorter. Let us see what effect is produced 
when the chase is kept the same length. At F in Fig. 86 
is shown a part of the chase ; for convenience and com- 
parison, the diameter at F is brought down to L, and we 
then see and are able to compare the two chases A I) and 
A L, each being the same length. Starting from the same 
diameter at A, the initial speed in each will be the same, but 
there is a great difference between the diameters at D and L 
in the proportion of ^^ to 1, and this reduction is one that, 
starting at A, works down to L. Such a reduction in the 
conicity of the chase means that there oMylit to be a pro- 
portional increase in the speed of the spindle througlioxd the 
chase, and not, as in the last assumption, only at the nose. 
By drawing oiit the speed curve for the new chase A L, and 
plotting it as in the dotted curve in Fig 87, we can form 
a very clear idea of the difference that should exist between 
the speed of the spindle when the cop hnttom is comj^leted 
and when the cop is completed. The speed when winding 
at D is only four times what it is at A, whilst when wind- 
ing at F or L it is ten times greater than at A, and for 
proper winding this speed must be attained through a 
gradual acceleration starting from the bottom of the cop 
chase at A until L is reached. To show the great difference 
this makes, the corresponding divisions on the chase at B, 
C, and D are marked on the chase A L at H, J, and K ; 
this means that, suppose the spindle is making tAvice as 
many revolutions when winding the yarn on at C as at A, 
it must make on the new chase twice as many revolutions 
at J as at A. Each of the other divisions can be interpreted 

174 COTTOy SPINXING chap. 

in the same way, and it will help greatly in comprehending 
the vital importance of effecting a change in the rate of 
winding as the cop lengthens, and that this cliange should 
commence at the bottom of the chase. 

In this analysis the two extremes have been taken as 
illustrations, but during each added layer the variation 
throughout the chase should take place corresponding to 
the smaller diameter on which the chase finishes. The fact 
that the quadrant is imperfect in not giving the correct 
curve of speeds, as in Fig. 87, does not in the least interfere 
with the reasoning employed, for the difference between 
the two chases remains proportionately the same. 

Nosing Motions. — Various means are adopted to 
obtain the desired change of speed to compensate for the 
taper of the spindle, all of Avhich depend more or less 
upon two principles of action. In one case, the effect is 
produced by moving the winding chain out of a straight 
line dui'ing the run-in of the carriage, thereby unwinding 
more chain from the winding drum. In the other method, 
the winding chain is shortened, and at the same time it 
is arranged that the shortening effect causes a scroll 
portion of the winding drum to come into action ; and 
the act of the chain working on a smaller diameter pro- 
duces an increased speed in the spindles. 

Both systems Avill be briefl}^ examined in diagram before 
describing the actual mechanism used. In Fig. 88 the 
quadrant and its connections with the spindle are shown ; 
the chain is represented as perfectly straight between the 
screw H and the winding drum B, and it will work in this 
position while the cop bottom is being formed. To obtain 
an increased speed of spindle as the cop lengthens, the 
chain is, by suitable means, gradually depressed during the 
run-in, its jjosition under these circumstances being shown 



by the dotted lines. The fact of moving the chain from D 
to C unwinds a portion of it from the drum B, and as it is 
done gradually the spindles are in the same degree increased 
in speed. Fig. 89 will give a better idea of the action. 
AVhen the quadrant occupies the position at A B, the chain 
jmsses to tlie drum from B to F in a straiglit line. When 
the winding reaches the second position, the quadrant is at 
A G, and the chain during the interval has been depressed 

slightly from the straight line G H to the extent shown at 
N G. In the third position the depression of the chain 
has been increased to J P from the straight line I) K ; and 
in the last position it has been still further increased, as 
indicated at L Q. The depression of the chain has there- 
fore been gradual, and in the right direction. It is un- 
necessary at this point to inquire into the cpiestion of the 
exact amount of depression required, it being sufKcient to 
show that for practical pi;rposes a near approximation is 
obtained. AVhen the cop has commenced to form np the 
tapered part of the spindle, the depression of the chain is 


onl}' of the slightest chcaracter ; but it increases gradually 
as each layer is added. 

In the second method of increasing the speed of the 
spindle, the chain passes over a small bowl B on the quadrant 
arm, Figs. 90 and 91, and on to another bowl C, which 
is actuated so that as the cop builds it can wind on a little 
of the chain after each draAv. This action, of course, 
shortens the winding chain ; but so long as the chain was 
Avound on a cylindrical Avinding drum, the shortening would 
have no effect Avhatever on the speed of the spindles. In 
combination Avith the shortening, therefore, the Avinding 
drum F is made Avith a scroll end, so that Avhile the chain 
is wound round C on the cjuadrant, it pulls over the wind- 
ing drum and changes the finishing point from D to E. 
When the cop bottom is completed, the Avinding chain 
finishes Avinding at D ; by winding the chain on at C each 
draAv, the drum F Avill be pulled round, and Avhen the cop 
is complete, the Avinding of the chase finishes at E. Conse- 
quently, during a portion of the AAnnding, the chain has 
been unAA'ound fi'om a gradually reduced diameter, and this 
chain unwound from the smaller diameter results in an in- 
creasing speed of spindles as compared Avith the unAvinding 
of the same length of chain from a larger diameter on the 
Avinding drum. 

Nosing motions assume a A'ariety of forms, the majority 
of Avhich depend upon an action which depresses the chain. 
As a rule, each machine-maker has some automatic motion 
which is applied specially to the machine he makes ; but a 
type common to all mules, and which is still extensively 
used, is that knoAvn as the "nose peg." As it is not 
automatic in its action, a certain amount of skill is required 
on the part of the person in charge of the mule. Fig. 92 
illustrates its application to the quadrant. Its essential 



feature consists of a slotted bracket projecting from tlie 
upper end of the quadrant arm ; it may be curved, as 
shown, or straight out, and in some cases it is disposed 
angularly to the arm, partly with the idea of gaining 

% ! ■ 
\ 1/ 






Fig. S9. 



•mw«& xfliut* 

iSH ofC01= 

Fig 91. 

strength and partly to suit the requirements of winding. 
In the slot of the bracket is placed a stud C, having a 
Avinged luit, Avhich enables it to be readily adjusted and 
fastened in the position suitalile for the increasing length 
of the cop. As shown in the diagram, the carriage has 
commenced its inward run and the yarn is being laid 
VOL. Ill N 


during tlie downward movenaent of tlie faller wire. The 
carriage continues to move in, and as the quadrant follows, 
a point is reached when the stud or nose peg C comes into 
contact with the Avinding chain ; further movement of the 
quadrant then causes C to press on the chain -and move it 
out of a straight line, so that when the carriage arrives in, 
the chain and quadrant occupy the position shown in the 
dotted portion of the sketch. The extra chain unwound 
from the winding drum by this depression of the chain 
produces an acceleration in the speed of the spindles. 

As already pointed out, it is not necessary to use the 
nose peg until the cop bottom is complete. When this 
occurs the peg C must be set in such a position that it only 
slightly depresses the chain during the few succeeding 
draws. During the lengthening of the cop the nosing peg 
must be moved a little further away from the quadrant 
arm every few draws, and in this way the depression of 
the chain is increased in the ratio considered correct by the 
minder, the necessary movement being one entirely depend- 
ent on his judgment. It is probably on this account that 
the nose peg is still so generally employed, its convenience 
and the absence of complication being its great features. 

It has, however, some very serious faults, its chief one 
being obvious if the previous explanation has been carefully 
followed. We have shown that to be theoretically correct 
a nosing motion ought to commence to act directly the 
chase is begun, its action being extremely gentle and 
gradual at first. In the nose peg this is a condition 
practically impossible. In the first place, to refer to the 
sketch, the peg C would not come into action until the 
quadrant arm occupied the position as indicated by the 
dotted line H G, so that the acceleration of the spindles 
would only take place as the quadrant moved from H G to 


H D ; in othei- words, only the nose of the cop would be 
affected by the extra speed. This is apparently contra- 
dictory to Avhat reason would lead one to require of the 
motion ; but an important consideration, when pointed out, 
will explain why such an anomalous action gives good 
results. The j^arn when wound on the thicker parts of the 
chase is put on a comparatively soft and yielding founda- 
tion, but as it neai's the nose the foundation becomes more 
solid, and therefore the tension put in in consequence of all 
the acceleration being thrown into that part, simply causes 

Fig. 92. 

the yarn to wind tighter without enabling it to influence 
the shape of the chase. This difference in the character of 
the foundation even renders it in many cases advisable 
from a practical point of view not to cause acceleration to 
commence until past the middle of the chase. The fault of 
the nose peg arises from the fact that when it begins to 
depress the chain it depresses it too quickly at first, and not 
in the correct ratio. This often leads to spoiled yarn, partly 
through being strained and partly in snarls, and carelessness 
in moving the peg only after long intervals increases its 
inherent fault. To get the best results the peg C must be 


moved regularly and often during a set, and this means 
that a slight movement often repeated prevents any great 
increase of sjieed being given to the spindles, and the 
minder is able to better gauge the tension of the yarn as 
the nose is wound, and so prevent its being strained, or. on 
the other hand, soft noses being made. 

The " nosing peg " system, it will be noticed, depends 
for its success entirely upon the skill of the minder. Many 
attemjDts have been made to eliminate this factor, so as to 
obtain an automatic motion, but the problem is one that 
contains several conditions Avhich are so variable and un- 
certain that the success which some of them attain can only 
be described as comparative or local, and as due rather to 
the additional skill of the minders, who, after a reasonable 
experience of them, can adapt them to the special character 
of their machine and the work they perform. 

To make a nosing motion automatic, it must be actuated 
directly or indirectly from the copping faller rod. The cop 
lengthens as the faller wire rises and places the yarn on a 
smaller diameter of the spindle; it is to compensate for 
this that the nosing motion is necessary. Some motions 
are therefore worked directly from the faller rod ; but since 
the faller rod receives its movement from the shaper, it 
amounts to the same thing to use this feature of the mule 
for operating the aiitomatic mechanism, and in one or two 
cases the oscillation of the quadrant has been taken 
advantage of to obtain a regulation. This latter method is, 
however, bad in principle, though occasionally it works 
well ; more will l)e said on this point when dealing with 
governor motions. 

The first illustration is taken from a well-known source, 
and, as will be seen from the sketch Fig. 9.3, it is 
actuated from the faller rod. A bracket X containing the 



mechanism is fixed to the quadrant screw box. An arm K 
is centred on a stud L, and its other end carries two catches, 
or pawls, which a spring presses into the teeth of a portion 
of a circuhxr rack struck from L as a centre. A projection 
J of the arm K carries a hook to which is attached a chain 
M ; this chain first passes over a boAvl II and from this 
point goes forward and is fixed to the Avinding chain at G. 
We can now see that, according to the position of the arm 
K, the chain M can be so an-anged that it Avill have no 
pulling eff'ect on the winding chain, but by raising K we 

Fig. 03. 

practically shorten the nosing motion chain and consequently 
the forward oscillation of the quadrant causes the Avauding 
chain to be pulled out of its cotirse, as shoAvn in the draAA'ing, 
The higher the position of the arm K on the toothed portion 
of the bracket IST, the more is the AA-inding chain depressed. 
It is therefore an matter to use this motion just as 
one Avould use the nose peg Avithout using its automatic 
features. In many cases this is an adA^antage, because yarn 
is a material that requires considerable humouring in some 
of its conditions, and means ought ahvays to be proA'ided 
for helping or retarding an intended automatic action such 
as a nosing motion. 


The following means are provided for the automatic 
working. A projection on the arm K carries a stud P, on 
which is swivelled an incline Q V; this incline is in contact 
with the arm K through an adjusting screw, so that if the 
incline is lifted up it will also raise the arm K. On the 
copping fuller rod S is a lever carrying one end of a rod U, 
which is guided in a bracket fixed to the square. The rod 
carries a finger W, which can be readily adjv;sted in position. 
As the carriage moves out, the finger naturally occupies a 
high position, and so comes into contact with the lower 
portion of the incline V, which is arranged to swivel out of 
the way ; but directly the backing-ofF is finished, the finger 
has fallen much lower, so that as the carriage moves in, the 
projection X of the finger "W comes into contact with the 
back of the incline and lifts up the finger a little. This 
lifting up of the arm K through the action of the finger AV 
upon the incline Q V each draw, gradually shortens the 
chain M and gives the necessary increased acceleration to 
the winding drum F. Practically it is almost impossible 
to raise the arm K each draw, because owing to the large 
number of laj^ers in the cop it would be necessary to have 
in the short portion of the circular rack such a number of 
fine teeth as to make the motion unworkable ; a reasonable 
number of teeth are cut and a double effect is produced by 
having two catches, and by this means a permanent lift is 
produced when the faller wire has been raised high enough 
to cause a pawl to catch in every half tooth. To render 
the motion more efficient, the bowl H is made of a cam 
shape to suit the conditions, as near as possible, as laid 
down in a previous description. After doffing, the catches 
are released from the rack and the arm lowered to its starting 
point. It may be remarked, that if the arm is purposely 
or inadvertently lowered during the building of the cop 



the next draw sinii)ly lifts it into its correct position 

A nose peg in very common use is of the form shown in 
Fig. 94 ; by comparing this with the drawing, Fig. 92, 
it will 1)6 seen to vary from it in the direction of the slot 
along which the peg D is moved as the cop builds. Much 
more care is required in moving the peg along a slot which 

Fig. 94. 

Fig. 95. 

lies parallel or is only slightly inclined to the cjuadrant arm 
than when the slot is more nearly at right angles to it. 
The regulation is obtained in a much shorter length of slot, 
and consequ^tly each movement requires to be very little, 
and well judged, otherwise too much nosing will be obtained. 
The slot is frequently made in the form of a curve drawn 
out empirically, rather than upon any fundamental reason, 
and so long as the movement of the peg depends upon the 


judgment of the minder this is a matter of little moment, 
and in any case variation from what might be considered a 
correct form of curve will not interfere greatly with the 
efficacy of the motion. 

Fig. 95 is a motion used by a well-known firm, who 
employ it extensively, and who find that it gives very good 
results in practice. It might be termed an automatic 
method of moving the peg along the slot in Fig. 94. In- 
stead of a slotted bracket, a lever C is used, centred on a 
bracket B, fastened to the upper end of the quadrant arm. 
This bracket contains a portion of a circular rack F, into 
which engage two pawls or catches E, carried by the lever 
C ; a spring G keeps the catches engaged. The lever C is 
set so as to begin to depress the winding chain directly the 
tapered portion of the spindle is reached. Afterwards it is 
automatically lowered by a chain connection to the shaper. 
It has already been observed that the regular movement of 
the shaper is sometimes taken advantage of in actuating the 
nosing motion, and in this case a bracket K is fixed on the 
rod which connects the front and back shaper-plates. To 
this is attached a chain J, and in order to pass it to the 
other side of the headstock a bell-crank lever is used 
centred on a floor fixing N ; the chain, after being guided 
through the eye of the bracket M, is taken upwards and 
attached to the nosing lever at L. By this means, each 
movement of the shaper draws the lever C downwards, 
and causes the end of it D to come into contact with the 
winding chain sooner each draw. The dotted lines show 
one position of the motion when it is acting upon the 
winding chain. 

The next example of a nosing motion is a j)ractical 
illustration of the principle explained in connection with 
Fig. 90. It is of a much more complicated character than 


any of the motions previously given, Init apart from this 
disadvantage it works sufficiently well for the purpose. It 
belongs to that class of motion Avhich is actuated from the 
shadier mechanism, so that, like all automatic nosing 
motions, it works in an uncompromising manner and 
entirely independent of au}'^ peculiar characteristics of the 
yarn or formation of cop. The following description will 
disclose the salient features of its action. 

The shaper-plate M, Fig. 9G, is moved forward by the 
screw L ; attached to this part of the shaper is a bracket 
J, to which is connected a chain F. The chain is guided 
over the back surface of a hanging lever centred at its 
upper end on a stud Gr ; from here it passes round the 
lower end of the quadrant screw box and on in an upward 
direction to a curved lever E. This lever, as well as a 
ratchet Avheel D, is fixed on the end of a stud, which 
carries a small boss or scroll, to Avhich the winding chain is 
fastened. The winding chain before passing to the wind- 
ing drum E. is taken over a guide bowl C, so that it is 
at this point that the winding chain must be considered to 
be attached. The bracket B, which carries the whole of 
the arrangement at C, D, and F, corresponds to the quad- 
rant nut of the ordinary motions, and up to the point when 
the cop bottom is finished, B is caused to travel up the screw 
in the usual way for forming it. AVhen the nut B occupies 
its lowest position on the screw, the chain F is slack, but it 
is drawn tighter as the nut is worked upwards ; during 
this period there is very little effect produced in shortening 
the winding chain, but when the cop bottom is finished, 
every additional movement of the shaper-plate M takes up 
the chain F and pulls over the lever E, so that the winding 
chain is "wound up at a quicker rate on the scroll at I). It 
will be noticed, however, that the mere movement of the 


screw is not depended upon for pulling over the lever E. 
The extremely small amount of motion given to the bracket 
J by the screw L is only caj^able of moving the lever E 
after several draws, because E can only take up another 
position after sufficient movement of the bracket J enables 
the catch to escape half a tooth in the ratchet wheel D. 
Until this occurs there would be a great strain on the chain 
F if the hanging lever G was fixed. To relieve this strain 
the lever is made pendent from G, so that during the in- 
ward run of the carriage the chain F is free from strain. 
Daring the outward run, however, an inclined bracket P 
on one of the arms of the quadrant comes in contact with 
a bowl Q carried by the hanging lever and presses it 
backwards, thus tightening the chain F and pulling over 
the lever E. In Fig. 97 the other extreme position is 
shown for the shaper-plate M and the lever E, and from it 
we see the shortening effect produced on the winding 

As before observed, this shortening of the winding chain 
would have no effect on the problem if the winding action 
depended on the ordinary straight winding drum. To 
obtain the necessary variation, therefore, the drum R is 
made as a scroll for a portion of its length, and as the 
chain is shortened the drum is pulled round so that it is 
brought into action sooner, and the direct effect is to cause 
the chain to finish unwinding from a smaller portion of 
this scroll part after each movement of E. 

Sufficient examples have now been given to convey 
a good general idea of the various methods adopted for 
compensating for the taper of the spindle, and those who 
have carefully followed the descriptions will scarcely have 
failed to notice that in no case does a motion follow out 
the conclusions arrived at when the theory of the action 


was described. lu the first place, tlie arrangements start 

too late ; secondl}', tlieir starting point varies for each 


layer instead of remaining constant throughout tlie set ; 
thirdly, no attempt is made to vary the action of the 
motion in conformity ^yith the actual conditions of winding ; 
and, lastly, there is a serious disadvantage in the absence of 
means for the motions to adapt themselves to the inherent 
irregularities and characteristics of the yarn being spun. 

Note. — Attention is called to a slight error in the drawings, 
Figs. 96 and 97. The ratchet wheel D is shown with its 
teeth ill the wrong direction. 

Governor Motions. — Governor or strapping motions 
are the names usually given to the apjiliances which auto- 
matically regulate the position of the nut upon the quadrant 
during the building of the cop bottom. A variety of means 
have been employed for performing the operation, but very 
few have Ijeen found to stand the test of practical exjDeri- 
ence, and on English machines at least the arrangements 
are confined within very narrow limits, the methods 
varying only in details of construction and the time of 

The subject is not without interest, and affords oppor- 
tunities for diverse opinions as to the correct mode of 
action, so there will be some advantage in considering the 
matter in detail. It will be assumed that the ntit on the 
quadrant is in the correct position for catising the bare 
spindle to turn the correct number of times, during the 
inward traverse of the carriage, for Avinding on the length 
of yarn in the stretch. So far as Ave are at present con- 
cerned, this first layer is supposed to be wound on Avith a 
regular tension, and there is nothing to sttggest that the 
position of the nut on the qtiadrant should be altered from 
the start to the finish of the layer — that is, during the 
whole of the inward run the nixt must remain in the same 


position. Now before the next layer can be put on the 
cop bottom, it will be necessary to raise the quadrant nut 
in order to compensate for the increased diameter of spindle, 
and we can see clearly that the time to do this must be 
during the outward run of the carriage, so that the nut is 
in the right position for "starting" the next layer at the 
correct speed. 

From this reasoning it might be concluded that an 
automatic motion should be so arranged that after each 
layer is added it moves the nut up the quadrant. Practical 
considerations, which will be mentioned subsequently, pre- 
vent this conclusion from being accepted as fundamental, 
though in essentials it ought to be the foundation of a 
governor motion. As it is, Ave find that the subject is 
treated more as a question of 02)inion, and naturally there 
is an absence of unanimity in regard to it. When dealing 
with the quadrant it Avas pointed out that the rate at 
which the nut must be moved up the screw was a varying 
one, gradually decreasing from a quick movement at the 
beginning. In a goA^ernor motion this must be taken into 
account, and if Ave depended on pure reasoning Ave should 
expect that every layer Avould require its share of move- 
ment. In addition to these considerations, it must be 
remembered that the yarn itself is an ever-varying factor, 
and that there are inherent peculiarities in the cops, AA'hile 
the faulty or imjjerfect character of the connections to the 
faller rods must be taken into account, for they modify to 
a large extent the presumed ideal of a goA^ernor motion. 
A perfect goA'ernor motion might be summed up as possess- 
ing the folloAving jioints : — 

(1) To give a movement to the qiiadrant nut for each 
layer added to the cop bottom. 

(2) To give a " correct " decreasing movement each draAV. 


(3) To compensate for peculiarities of cotton, yarn, 
cops, or connecting motions. 

(4) To actuate the quadrant screw after one laj^er and 
before the commencement of the next one. 

To the practical reader it need scarce)}' be pointed out 
that these conditions are never fulfilled, and it might 
almost be added that the difficulties in the way of fulfilment 
have hitherto prevented any success being attained when 
the attempt has been made. 

Instead, therefore, of the governor motions working 
under ideal conditions, we generally find them entirely 
under the control of the yarn itself, and actuated either 
before or after the run-in of the carriage. Simplicity and 
convenience are the deciding factors in the case, and while 
good average results are obtained, the erratic and faulty 
character of many motions leaves much to be desired in 
the direction of a governor motion founded on correct 

By permitting the yarn to actuate its own Avinding we 
practically combine the first, second, and third conditions 
enumerated above, and by so doing take advantage of the 
tension-regulating action of the counter faller. The yarn 
unwound during backing-off is taken up by the counter- 
faller wire, and as winding proceeds a certain tension is 
maintained b}^ means of the weighting arrangement already 
described. It is easy to see that if through any cause the 
tension is lessened or increased, the wire Avill yield, and we 
can also understand that one of the chief causes of any 
variation in the tension of the yarn will be irregularity in 
the winding. On this effect the action of a governor 
motion is generally based. For instance, suppose the first 
layer has been put on the spindle correctly, the next layer 
will naturally require a slower speed of spindle,, and to do 


this the luit must be moved u}) the quadrant screw. 
Instead of doing this in anticipation, most mules commence 
to wind the next layer with the screw in the same position 
as for the first layer. The almost immediate efi'ect is that 
the larger diameter winds on too much yarn, and naturally 
puts so much tension in the yarn that the faller wire is 
pulled down. This, as will he shown subsequently, brings 
about, through suitable mechanism, a change in the position 
of the nut, which gives the required speed to the spindle. 
In such an action as this we get the third condition 
incorporated with the first two, and the yarn, as it is being 
wound, is relied upon to do all the regulating required. 
It will be noticed, however, that a serious evil is introduced 
in the great increase of tension that is put into the yarn 
at the commencement of winding, and this is especially 
noticeable at the commencement of the cop and in low 
and medium numbers ; the fact that it is a progressively 
decreasing one, helps to neutralise it considerably, and 
possibly on this account, together with the presence of 
some personally adjustable feature of the motion, maintains 
such a mode of action as a base of those arrangements 
which are most successful. 

A great difference of opinion exists in regard to the 
time when the nut ought to be moved upwards. The 
writer's opinion has been expressed above so far as the 
principle of action is concerned, but actuating the governor 
motion during the run-in has its advantages ; for instance, 
carelessness is more easily and quickly corrected l)y this 
system, and, moreover, insufficient governing during one 
draw will be corrected during Avinding in the next. Apart 
from the features common to both methods, it may fairly 
be taken for granted that the more uniform tension and 
preparedness for the next draw in the regulation during 


the outward run will equalise the practical advantages of 
the regulation during the inward run. In either case the 
class of cotton and quality of yarn must decide the 
question from a practical point of view, but it cannot 
be too strongly impressed upon the reader that the best 
resixlts can only be attained by keeping as closely as 
possible to the conditions laid down for a perfect motion, 
and for the best quality and finer yarns it is almost 
necessary that the last condition should be folloAved. 

The following examples of " band " governing motions 
may be taken as typical of the kind which find most 
favour. They are termed " band " motions because a band 
is used to give motion to the quadrant screw. On the 
lower end of the quadrant screw is fixed a bevel wheel, 
which gears into another bevel cast or fixed on a band 
pulley, which rides loose on the quadrant shaft A, Fig. 
98. An endless band is passed round this pulley and 
guided over a series of guide pulleys B, C, D, E, and F. 
Three of the guide pulleys, B, C, and E, are carried on fixed 
studs, but the other two, D and F, are carried on studs 
fixed to the carriage, so that as the carriage moves the 
pulleys travel backwards and forwards. The only effect 
this disposition of the pulleys has is to set up a certain 
amount of friction in the band, but since D and F are 
free to revolve on their studs, the friction is relieved by 
their motion, and the band remains unaltered in position. 
In order to produce some effect of the pulley on A, it will 
be necessary to grip or hold the band in some way, so that 
the movement of the carriage Avill draw it along. A 
variety of methods are adopted for doing this, several of 
which will be shown. Referring to Fig. 98, it will be 
seen that a lever or arm is fixed on each faller rod ; one 
end of a chain is connected to the arm K on the copping 


faller rod T, aiul after passing round a pulley II is attached 

to au adjusting screw carried by the arm L. The pulley 
VOL. Ill O 


H is supported by the end of a lever X centred on the 
carriage at J, and a projection on this lever is arranged so 
that it can be lowered into the path of a revolving toothed 
disc G fixed on the guide pulley F. 

As the yarn is being wound on the cop bottom it jDasses 
over the faller wires L and M. As the wire L guides the 
yarn on the spindle, it of course moves, and naturally the 
arm K does the same, but this has very little eftect on the 
chain, the lever being arranged in position so that it is 
passing along the upper part of the circle it describes ; H, 
therefore, is affected very little by this movement of the 
faller wire. In the case of the counter-faller wire M it is 
different ; the position of M depends upon the tension of 
the yarn as regulated by the faller weights. Therefore, 
directly a larger diameter of the cop or other circumstances 
cause the spindles to wind on too quickly, the tension is in- 
creased and the faller wire is pulled down, say, to N. The 
lowering of the wire ]\I to N gives a similar movement to 
the arm L, and this immediately causes the end of the lever 
X to drop, and the projection falling into the path of the 
disc G prevents the rotation of the pulley F ; this sets up 
sufficient friction in the rope to hold it so that the carriage 
takes the band forward and produces a movement in the 
pulley A in the opposite direction to that shown by the 
arrows. The revolution of A will continue to move the 
nut up the screw so long as the pulley F is held by the 
lever X ; but since the nut in its higher position on the 
screw will revolve the spindles more slowly, the tension 
will be quickly relieved and the wire M will rise to its 
normal position and lift the lever X out of contact with 
the disc G and so permit F to revolve freely. This action 
takes place just as often as the yarn becomes sufficiently 
tisht to draw down the wire M low enough to let the 


projection fall upon G. This occurs very frequently 
during the early part of the cop bottom, but at much 
longer intervals towards the finish. 

It sometimes happens that the nut has not been moved 
high enough for a certain layer, and in such a case we 
should find that the tension at the commencement of the 
following run-in would cause the lever X to at once fall 
into contact with the disc G, and so complete the raising 
of the nut. 

According to the quality of cotton or yarn, it is 
absolutely necessary to arrange for some means of ad- 
justment either for modifying or increasing its sensitive- 
ness ; a regulating screw is therefore provided on L, which 
enables this to be done, and it is also used to lift the lever 
X out of position, so that, after the cop bottom is finished, 
any incidental irregularity of the yarn Avill not give a 
higher permanent position to the quadrant nut ; this action 
must be left to the judgment of the minder. 

No arrangement is made for any reduction in the 
tension of the jarn, because such a condition is scarcely 
possible, and indeed evei'ything is done to prevent any 
lowering of the nut, a catch wheel being generally 
provided on the top of the screw box. The too easy 
movement of the screw is also prevented by means of a 
strong friction brake either on the top or on the pulley on 
the shaft A. Carelessness in allowing these brakes to 
become inoperative has frequently led to bad work and 
breakdown of ends. 

The illustration. Fig. 98, also shows an arrangement for 
winding back the winding chain during the run-out of the 
carriage. A band is fastened to one end of the winding 
drum, and its other end is attached to a Aveight AV after 
passing over the guide pulleys Q and E, carried by an 


upright rod. As the carriage makes its inward run the 
weight W is lifted up to near the top of the rod, so that 
during the outward run it falls, and in so doing turns the 
drum and winds on the chain. 

Fig. 99 presents us with another arrangement of band 
governing motions. It differs from the previous motion 
only in the method of holding the band in order to give 
motion to the quadrant screw. On the faller rods M and 
L are fixed the levers K and J, and to these are attached 
an endless chain which passes over a solid loose bowl H 
carried by one end of a lever centred on a stud at G. The 
other end T of the lever is so arranged that Avhen the 
tension of the yarn pulls the faller wire down, say, from P 
to Q, the weighted end will press T against a projection F 
on a bracket bolted to the front of the "square." The 
governor band passes through slots Avhich keep it always 
in front of the projection F, so that when the chain 
permits H to fall, the band is forced against F, and the 
pressure is sufficient to hold it fast while the carriage 
carries it forward in the direction of the arrows, to give 
motion to the screw. As in Fig. 98, the action of the 
lever K on the copping faller gradually lifts H a little 
higlier, and when the cop bottom is finished it will have 
been raised high enough to prevent it coming into action 
again unless an unusual amount of tension depresses the 
faller wire. Fig. 100 gives sufficient of a side view to 
enable the motion to be readily understood. 

Tlie front and side elevation of an interesting motion 
are given in Figs. 101 and 102, and although it now 
belongs to a numerous class of movements which have been 
found wanting, it has features which give it importance 
from a mechanical point of view. 

On reference to the drawings, there is a small pinion B 


cast to the back of the usual bevel C. A rack A is 

arranged to gear into B, and the interesting feature of the 

1 98 COTTON SPINNING chap, ii 

motion lies in the method adopted for regulating the 
number of teeth in the rack A to gear with B for each 
layer. It will be noticed that the rack A rests ujDon a 
sliding plate D by means of inclined projections, and that 
D rests upon a slide Q, one end of which is fastened to a 
rod E, while the other end slides upon a fixed box-like 
bracket which is firmly fastened to the headstock or floor. 
The rod E is supported by this bracket and floor fixings as 
shown. On it is also fastened a swivel catch R, which cau 
be acted upon by a drop pendant N connected by chains 
to the usual connections K and L on the faller rods. 
Variations of tension in the yarn during the outward run 
will cause N to drop ; in doing so it comes against the 
catch E, and carries the rod E forward, and naturally also 
the slide Q with its small slide D and the rack A. As Q 
is pulled along a projection on the end of the slide, D 
comes into contact with a jji'ojection on the box bracket, 
and its further movement is stopped, but the rack continues 
its forward movement with Q. Previous to this the 
inclined projections on the under side of A have been in 
corresponding slots in the slide D, so that when D stops 
moving A is compelled, by means of its inclined projections, 
to slide up and occupy the position shown in the drawing ; 
it will thus be seen that the rack A has been out of gear 
with the small wheel B during the inward run, and more- 
over the increased tension in the yarn has simply raised 
the rack up in such a position ready for the carriage, 
during the outward run, to push the slide Q, and cause A 
to gear with B, and so turn the screw. 

It will easily be seen that the full length of A would be 
employed each time the motion worked, unless an arrange- 
ment was made for regulating the number of teeth to be 
used to suit the size of the cop bottom. This is done by 


200 COTTON SPINNING chap, li 

means of a screw F, on which is threaded a stojD-washer H. 
On the slide D is swivelled a catch I which comes against 
H when the carriage pushes Q forward, and thus stops D 
from further movement. However, A continues with Q to 
move forward, and directly its projections come to the 
slots in D it falls down out of gear with D and finishes its 
forward movement out of gear Avith B. A ratchet wheel 
G on the end of the screw F is actuated by the end of the 
rod E, and H is moved along the screw so that the rack 
may, at the commencement of the cop bottom, use its full 
length in driving B ; but as H moves forward, the stoj^ on 
D, coming in contact with it sooner, causes A to drop out 
of gear with B earlier, and thus reduces gradually the 
number of teeth capable of driving the spur wheel. This 
continues until, when the cop bottohi is complete, the rack 
will fall down before any of its teeth can touch B. The 
screw F is variable in its pitch in the proportion necessary 
for the shape of the cop bottom, and the whole motion is 
of a character to fulfil almost all the conditions required of 
a successful motion. The great objection lies in the fact 
that the slightest carelessness will result in a derangement 
or even a breakdown of some part of the mechanism, and 
it requires such a careful adjustment that it has at last 
been discarded in favour of motions with less mechanical 
difficulties in their application. 

Our next example. Fig. 103, is very similar to the one 
given in Fig. 100; its main jDoint of difference consists in 
so arranging the faller connections that instead of a lever 
being allowed to fall when the faller wire is pulled down, 
it is drawn up and an extension of it made to bear 
against the governor band. Reference to the drawings 
will make this clear ; the faller levers D and C are both 
jjlaced on the opposite side of the faller rods A and B to 



the previous example, so that their movements lift the 
lever K instead of permitting it to drop. When the bowl 
J is lifted, a projection P on the lever K is brought against 
the governor band N, which passes through a recessed 
portion of the bracket M; in this way the pressure is 
sufficient to hold the band, and as it is connected to the 
quadrant pulley in the same way as in Fig. 103, it naturally 
gives motion to the screw. 

In order to regulate the pressure put on the band at 
]Sr, a spring G is connected to the leather band F and the 
chain H ; any excessive movement of the faller Avire W, 
therefore, simply stretches the spring G. 

As the cop bottom enlarges, the sector C is lowered, 
and this lowers the lever K further away from the band N, 
until at last it is low enough to remain out of action during 
the building of the body of the cop. For the same reason 
it is sometimes found advisable to have the projection P 
as near as possible to the band N when the cop is com- 
mencing, and to effect this a bowl Q is arranged on the 
sector C, which the leather band F passes over ; a sensitive 
action is thus obtained for the first layers, but afterwards 
such a degree of sensitiveness is not so necessary, and Q 
therefore works clear of F. The usual adjusting screw E 
is provided, and in addition slots at D and at Q enable a 
high degree of exactness to be obtained in setting the 
motion. In practice, this motion has been found to be 
unusually successful. 

Figs. 104 and 105 present us with another form of rack 
governor motion which has been found particularly suitable 
for fine spinning mules. To the usual faller connections 
D and is connected a chain F Avhich passes over a carrier 
bowl carried by a small frame G, which in its turn is 
hooked on to a drop pendant H ; this slides in a bracket 


J fixed to the carriage square, and at its lower end is a 
swivel piece K. When the faller wire is pulled down, the 
lever D is lowered, and the drop pendant H K falls into 
contact with a rack L, and the movement of the carriage 
takes the rack forward on its slide M in the direction of 
the arrow. To L is connected a rod JST, whose other end 
is screwed to a rack P, which gears with a small pinion R 
mounted loose upon the shaft X (see Fig. 105). The 
wheel R is arranged to drive the hevel S, and therefore 
the quadrant screw through the catch-box W ; consequently 
the forward movement of the rack P can be made to give 
motion to the nut Z. The return or outward movement 
of the carriage, by means of a finger bolted to the square, 
pushes the rack P back without operating the screw. In 
place of this, the catch-box may be arranged to be in- 
operative when the rack P is moved by the inward run 
of the carriage, and during the outward run to act upon 
the quadrant. The rack L is made long enough to enable 
the full length of the rack P to be used when such is 
necessary, as in the earlier layers, and also to use small 
portions when the cop bottom is getting finished or when 
the tension is only slightly altered. 

Sufficient examples of governor motions have now been 
given to show how near to self-acting the winding operation 
has been brought. At the same time the observant reader 
will have noticed the disadvantages associated with the 
various automatic arrangements des(rribed, and to any one 
with a practical knowledge of the subject, such disadvantages 
are almost considered inherent, and in most cases prove a 
source of difficulty wlien a motion is first applied. 

It has already been explained how necessary it Avas 
to move the nut up the quadrant sci-ew at a gradually 
diminished rate. When the action is performed by the 


minder lie finds it necessary to turn the screw several times 
for the earlier layers and only occasionally for the last 
layers. This calls into play a certain amount of care and 
judgment, which can be modified by making the screw 
with a varying pitch, as at B, Fig. 106. By this means 
a single revolution of the screw will move the nut a good 
distance upwards when the cop bottom is started, Avhile a 
similar turn when the cop bottom is complete only moves 
the nut a short distance. The example given at B is taken 
from an actual screw as used by a well-known firm of 
machinists. The rate at which the nut would travel up 
the screw is shown in full lines in the diagram, and from 
it we get a clear idea of the diminishing rate of its upward 
movement. Such a screw as this has been in use almost 
from the first introduction of the quadrant, and, probably 
from practical motives, it has remained until the present 
time ; in order, however, to prevent misconception, it is 
as well to point out tliat tlie screw B only goes a little 
way towards giving the nut its correct but varying move- 
ment for a " uniform " turning of tlie screw, A quadrant 
screw to do this ought to be made as shown at A, yielding 
a curve as dotted on the diagram. This curve represents 
the true movement of the nut up the quadrant, and its 
variation from the curve of B is considerable. The vertical 
lines may be taken as representing complete turns of the 
screw, so that in the case of B ths first turn would lift 
the nut from I to E, while for screw A the nut would be 
lifted from I to F, more than twice the distance. A similar 
difference, but in the opposite direction, is noticeable at 
the upper end of the screw, B having a much quicker 
movement than A for each turn. As a comparison, a 
uniform screw is shown at C, having an equal number of 
threads, as A and B. Its rate is naturally represented as 



a straight line in the diagram, and Ave see Aery clearly its 
difference from the A'ariable pitch. 

In the application of a governor motion, some firms 
have discarded the A^ariahle sorcAv, and rely upon the motion 
giving to the nut its correct variable moA'ement; but Ave 
see that although the tension of the yarn is likely to be 
increased above the required amount early in the iuAvard 
run Avhen the cop bottom starts, and late in the run-in as 

Pig. 103. 

it finishes, it may easily happen that in the former case it 
acts on the band or rack too late to move the nut high 
enough, and in the latter case too early, and so giA^es the 
nut too much movement ; A\'e therefore get the action 
spread CA-er tAvo draAvs for the first case, and there is no 
remed}', outside the minder, for the second case. A correc- 
tion for this is found by some makers in still retaining the 
variable screA\', and its difference from the correct form 
may be looked on as a convenient compromise betAveen 
the two extremes at A and C. 

2o6 COTTON SPINNING chap, ii 

Long-lever Mule. — The descriptions of the general 
actions of the mule have so far been confined to the cam- 
shaft principle of working the changes, but, as already 
stated, this is not the only method. The "long-lever" 
mule, as it is aptly termed, to distinguish it from the 
" cam-shaft " mule, is one that receives a veiy extensive 
application, and its range of Avork from the lowest to the 
highest numbers gives to its mechanism an unusual and 
important interest. Because the long-lever mule is made 
by firms who have a very high reputation for machinery 
adapted for good quality and high numbers, it is some- 
times thought that it is only suitable for such purposes ; 
this is a mistake, and it is desirable to emphasise the fact 
that its working is equally satisfactory on the lowest counts. 

A general outline description of the long lever and its 
action will now be given, and reference will be made to 
the drawing. Fig. 107. This sketch embodies the principal 
features of the mechanism, but it must be looked on rather 
as a key diagram than as an illustration of detail. An 
attempt will be made, as when dealing with the cam shaft, 
to describe and illustrate all features essential to a clear 
understanding of various actions. 

On reference to Fig. 107, the long lever A B is centred 
on a stud C fixed in the framing of the headstock. By 
giving movement to the ends of this lever we can bring 
about changes in the working of the mule which permit 
spinning, backing-ofF, winding, and drawing-up to be per- 
formed. In spinning, the strap is on the fast pulley W, 
and this both turns the spindles and takes the carriage 
out. While this is going on the strap must be kept on W, 
and the backing-ofF wheel V must be kept out of contact 
with the leather cone on W. This latter effect is produced 
by a stud E on the long lever coming against the lever, 



which puts V in and out of gear with W ; in the position 
shown it is impossible for V to be moved. On the other 
hand, the long lever must be locked in this position, so we 
find that at the outer end of the long lever a stud D is 
held by a catch G on a bell-cranked lever centred at H. 
As the carriage comes out, the c[uadrant drum shaft S is 
giving motion to a wheel R, upon Avhose face is a special 
cam groove. A slide U, the lower end of Avhich carries 
a heavy weight, is raised by the cam groove, and the end 
A of the long lever is free so far as this weight is concerned, 
but the upward movement of the slide U puts a spring 
into tension, which is attached to a lever T, upon which 
the long lever rests. There is, therefore, a force pulling 
A upwards, which is resisted by the catch at Gr. When 
the carriage arrives out, a bowl on the square comes into 
contact with the incline J and releases the catch G, the 
end A instantly moving upwards, and the other end B 
falling. The stud E, coming opposite a recess on the 
backing-ofF lever F, permits a spring to pull F forAvard and 
so puts the wheel V into contact with W. 

The backing-ofF action completed, a bowl on the square 
is moved forward and lifts an incline L, carried by another 
catch lever K centred on H. The projecting catch on K 
has prevented the stud E falling further than the recess 
in F, but now that K is released the stud E is forced 
further downwards and in this movement takes the lever 
F out of its way and consecpiently V out of contact 
with W. 

Matters remain in this condition during the drawing-up 
or run-in, but as this nears completion the carriage comes 
into contact with a finger on a rod N, which is attached to 
a lower portion M of the lever K L ; by moving O forward, 
the catch at K is taken from under the stud D and the 


end of the long lever suddenly falls under the influence 
of the weight and again assumes the position shown in the 

The other end of the lever is bolted to a piece Y, which 
carries a stud acting upon a lever Z, whose fulcrum is at 
7}. Special forms of incline slots in Z permit the stud on 
Y to be inoperative until drawing-up commences ; the fall 
of the weight, therefore, puts a catch box into gear by 
means of the lever Z, and enables the drawing-up to be 
effected. The movement of E in an upward direction 
takes the catch box out of gear and leaves the back shaft 
stopped during backing-off. 

A detailed description and enlarged drawings of the 
long-lever mule Avill now be given. The principal oj^era- 
tions performed are spinning, backing-off, and drawing-up 
and winding : all these actions are directly Avorked from 
the long lever, and, as in the cam shaft, the cycle of 
movements for producing them is termed the " changes " ; 
discussing them in their order, we will take spinning for 
first consideration. 

Spinning.^On reference to the drawings, Figs. 108, 
109, and 110, it may be pointed out that they represent the 
position of the mechanism during the operation of spinning. 
In Fig. 109 duplex driving is shown, and the strap is sup- 
posed to be on the two fast pulleys. Under this condition 
the backing-off cone wheel G^ is out of contact with the 
backing-off cone on the pulley H\ so that the continuous 
and independent driving by band of the drawing-up pulley 
U^ has no effect on the rim shaft. The rim shaft being 
driven, we have the motion transmitted through the rim 
pulley to the spindles. At the same time the rim shaft 
drives the front roller (see Fig. IG), and the front roller 
drives the back shaft through the wheels T, 0, E, P, and 
VOL. Ill P 


Q ; from the back shaft, motion is given to the carriage 
during its outward run. 

While these actions are going on, the scroll shaft T^ is 
rendered inoperative by keeping the drawing-up cone 

Fig. lOS. 

clutch Q^ out of gear with the cone E^, so that although 
the drawing-up pulley U^ is being driven, it has no effect 
on the scroll shaft. The end of the long lever iu Fig. 109 
has therefore three very important functions to perform 
while spinning is taking place. First, it must keep the 


backing-off cone wheel G^ out of gear -with the cone on 
H^ ; secondly, it must keep the drawing-up cone clutch Q^ 
out of gear ; and, thirdly, it must keep tlie clutch wheel 
on the hack shaft (Fig. 16) in gear. 

Fio. 109. 

The first object is performed in the following manner : — 
The backing-ofF lever D^ (see also Fig. 110) is pivoted on 
a shaft E^ ; its upper end F^ is forked to fit into a grooved 
boss on the backing-off wheel G^, while its lower end bears 

212 COTTON SPINNING chap, i! 

against a stud A^ carrierl by the long lever ; in this position 
the lever D^ is locked, so that it is impossible for the wheel 
G^ to go into contact with the pulley H\ The second 
object is effected as follows : — A lever N^ is carried from 
a stud at P\ and the lever is forked and fits a groove 
on the upper part of the cone dish Q^. The other end of 
the lever X^ is connected by a link at M^ to the lever I-^ 
centred on a stud K^ (see also Fig. 32, but the reference 
letters are not the same). A stud B^ on the long lever 
bears against a prepared part of the lever I^ as shown, and 
so long as the stud occupies this position the two halves 
of the cone clutch Q are prevented from going into contact, 
and the dra wing-up pulley U ^ cannot drive the scroll bevel 
T^. The third effect is produced by a stud on the end of 
the long lever at C working in a. groove on one extremity 
of a lever, the other extremity of which is forked and fits 
the groove on the clutch wheel on the back shaft, as shown 
in Fig. 16. 

The corresponding position of the outer end of the long 
lever during spinning is shown in Fig. 108. It occupies 
its lowest point, and in this position it is held by a stud 
M, coming under the catch of the L lever centred at P ; 
the long lever is therefore locked during the whole of the 
outward run of the carriage. As already explained, this 
movement of the carriage gives rotation to the shaft A for 
the purpose of working the quadrant. Its motion is taken 
advantage of to drive by means of a pinion the lifting 
wheel ; a stud C is carried round by this wheel, and in 
the course of its revolution it comes against the underside 
of a projection D on the drop weight lever K, and so 
raises it. A projection on K at E carries one end of a 
spring F, the other end of which is connected to a projec- 
tion G of a specially-formed lever H, whose centre is at J. 



The upper surface of H bears against tlie long lever, and 
the tension put into the spring F, as a consequence of 
E D K being lifted by the stud C, tends to force the 
long lever upwards ; so long, however, as the stud M is 
held by the L lever the tension in the spring has no 

Backing-ofF. — ^The carriage at last reaches its outer- 
most position ; at this moment a stud Y, on the end of 
a lever centred on the square, comes against the inclined 
tappet R on the L lever and lifts it up, thus freeing the 
stud M from its catch. Directly this occurs, the tension 
in tlie spring F instantly forces the long lever upwards, 
but it can only ascend a short distance, because, although 
L has been moved out of the Avay, the lever T has not 
been touched, so that the projection on T acts as a stop 
to the further upward movement of M (see Fig. 110). By 
referring now to Fig. 109, we shall see what effect this 
movement of the long lever has upon the levers D^ and I\ 
We already thoroughly understand that when the carriage 
has completed its outward run, the spinning process is over, 
and so the spindles must first be stopped and then immedi- 
ately reversed for backing-off. The ascent of M in Fig. 
108 means the descent of A^ in Fig. 109, and it Avill occupy 
the position of the middle dotted circle, or as shown at 
2 in Fig. 110. A strong spring in tension (not shown in 
the sketch) immediately pulls the lever D'^ forward, a recess 
cut in the face of the lever permitting this to be done. 
This at once puts the wheel G^ in gear with the fast pulley 
H^, and as the straps have been moved on to the loose 
pulleys, the drawing-up pulley U^ is enabled to drive the 
rim shaft through the pinion which gears into G^ (this is 
shown clearly in Fig. 57). The direction in which it is 
driven is also in the opposite direction to that in Avhich 


the straps drive when they are on the fast pulleys; the 
spindles are therefore reversed. 

It Mill be noticed that the drawing-np cone Q must 
still be kept out of gear dixring this backing-otf action ; 
for this purpose the lever I^ is so arranged that on the 
descent of the stud B^ it simph' comes on to a lower 
portion of the straight surface of I^ and produces no effect 
on the lever itself. 

The act of reversing the rotation of the rim shaft eff"ects, 
through special mechanism which will be described subse- 
quently, a movement in the fallers, and one effect of this 
movement is to lift ujj the end of the lever which carries 
the bowl Y in Fig. 108. The lifting of Y brings it against 
the tappet carried by the T catch lever, and of course this 
moves T on one side, freeing the stud M, so that the tension 
in the spring F forces the long lever still further upwards, 
and as it moves upward it passes from the lower side 
of the projection'on T to the upper side, where it rests 
(Fig. 110). 

Drawing'-up. — Tliis second change of the long lever 
causes the end in Fig. 109 to fall to its lowest point (see 
also 3 in Fig. 110). Its effect on the lever D^ is to force it 
backward as the stud A^ moves out of the recess, and this 
necessarily takes the cone wheel G^ out of contact with H^, 
and so stops the spindles. At the same time, the stud B^ 
falls clear to the lever I\ and a strong spring J immediately 
pulls the lever forward ; this action, through the link and 
the forked lever X\ forces the cone dish Q^ into gear with 
the cone clutch, and so permits the drawing-up pulley U^ 
to drive the scroll shaft and cause the inward run of the 

A locking arrangement is provided for the carriage on 
completing its outward run, in the form of a holding-out 


catch W. A stop O on the square comes against an 
incline "\V, lifts it, and passes under, so that the incline 
falls back and locks the carriage in position ; this latch 
must be lifted before the inward run can take place. Con- 
nected to the incline is a link, the slotted upper end of 
which fits a pin X on the long lever. This pin X moves 
clear in the slot of the link during the change from spinning 
to backing-ofF, but when backing-ofF is complete and the 
long lever makes its second upward movement, the pin X 
comes against the top of the slot and lifts the link, which 
raises the catch W out of the way of the stop O, and sets 
the carriage free to make its inward run. 

During the run-in the straps are on the loose pulleys ; 
winding is taking place, and the long lever is locked in 
position by the T latch lever (as shown in Fig. 110); the 
stud M occupies its highest position, and the studs A^ and 
B^ occupy their lowest positions ; the stud C on the lifting 
wheel is clear of the projection D, and therefore the full 
effect of the heavy weight on the drop lever K comes on 
the end of the long lever. So long, however, as the stud 
M is supported by the T lever, the weight is inoperative. 

As the carriage completes its inward run, it comes 
against the stop X\ fixed on a rod which is connected to 
an extension of the T latch lever at Q. The forward 
movement given to the rod pulls the lever T on one side 
and permits the full effect of the weight to come on the 
long lever and to pull it down in one movement to its 
lowest position. The studs A^ and B^ in Fig. 109 move 
up to their highest points, and in doing so assume the 
positions shown in the drawings ready for spinning. The 
upward movement of the stud B^ is not allowed to move 
the lever I^ on one side ; this is effected by a stud on the 
carriage (see Fig. 32), which comes against the lower end 



of the lever and lifts the clr;iwiiig-up cone clutch completely 
out of gear. 

The drawings have been made as complete as possiljle 
to enahle the descriptions to be clearly understood, but 
with this object in view several details have been ke})t out, 
such as the backing-off motion, the chain-tightening motion, 

Fig. 112. 

Fi.;. 111. 

the strap-fork arrangement, etc. These, however, will l)e 
full}' dealt with. 

Chang-ing Strap from Fast to Loose Pulley. 
Strap-relieving Motion. Hastening- Motion. — There 
are several methods of changing the strap from the loose to 
the fast pulleys, and vice versa. One of these is illustrated 
in Figs. Ill and 112. The duplex system of driving is 
shown. As the carriage moves out, the strap is on the 


fast pulley ; as it arrives within a few inches of the 
finish of the stretch, a stud W (Fig. 112) on the carriage 
comes into contact with a pendant lever T, centred on the 
framing ; this lever is moved forward, and A, a projection 
thereon, presses against a stop-washer fastened on the rod 
Q and moves the rod also forward. Attached to the rod 
at the outer end is a spring S, whose other end is fixed to 
the framing; the other end of the rod Q (Fig. Ill) passes 
through a slot in the lower part of a lever P, which is 
fastened on the strajD-fork shaft K. The forward move- 
ment of the rod causes Q simply to move freely in the slot 
of P ; but a spring R, attached to the rod and to the lever 
P, is put into tension ; and with this tension existing in 
the spring R there is a strong force tending to move P 
forward and put the strap from the fast to the loose pulley. 
This action would of course directly occur under some 
circumstances, but frequently an arrangement is provided 
whereby spinning continues until the necessary amount of 
twist has been put in the yarn. Until this occurs the 
strap-fork is locked by means of the twist latch lever H, 
which is attached to the strap-fork at J and a projection at 
M fitting over a portion of the framing at L, where it is 
locked. This lever is set free in the following manner : — 
A screw is formed on the end of the rim shaft A, into 
which gears a worm wheel called the "twist wheel" B. 
Through the gearing C, D, and E a short shaft is driven, 
whose end carries a tumbler F. This tumbler, though free 
on the shaft of E, is, through a pin, capable of being carried 
roimd. As it revolves, it comes against a projection G 
fastened on the upper part of the twist latch lever, and 
lifts it until the projection M rises clear of the catch L ; 
directly this happens, the tension in the spring R pulls the 
strap-fork over, and changes the strap from the fast to the 


loose pulley. Backing-oflf then takes place, and afterwards 
the carriage is drawn in by the drawing-up band. 

AVhen the lever H is freed from the catch L on one side, 
the spring E, pulls the strap-fork over, and with it the 
twist latch lever, so that this lever passes over the top of L, 
falls down on the other side, and again becomes locked ; the 
strap-fork therefore cannot be moved from the loose to the 
fast pulleys until H is again set free. Now it will be 
noticed that the tension put into the spring S by the 
carriage moving T forward is not affected when the spring 
E. acts on the strap-fork; Q makes no movement at the 
moment the strap changes ; P is simply pulled over, and 
now abuts against the nut on rod Q, the tension in the 
spring S remaining. Although this tension has a tendency 
to move the strap back to the fast pulley, it cannot do so, 
because it causes the twist latch lever to press against the 
projection L on the side opposite to the position it occujDied 
when the carriage was going out. The illustration, Fig. 
Ill shows the position during the run-in of the carriage. 

On the faller rods a small bracket is loosely fitted, 
carrj'ing a screw O on which is fitted a tumbler N. The 
use of the screw enables the position of N to be carefully 
regidated according to the circumstances of the case, and 
moreover N is so arranged that it can easily be turned 
over so as to avoid coming into contact with H. In the 
position shown, the carriage is moving in, and naturally N 
will come into contact Avith H and lift it ; tliis frees the 
twist latch lever from L, and permits the tension in 8 to 
pull the rod Q backwards ; the nut on Q being against P, 
forces P backwards, and so removes the strap from the 
loose to the fast pulley. By making N so that it can be 
moved on one side, the mule is enabled to be stopped when 
it completes its inward run, because it prevents the strop 


from being put on to the fast pulley so long as the twist 
lever H is locked on the catch; by turning N over, the 
lever H is untouched when the carriage gets in, and as the 
strap is not changed the mule stops. 

Adjustment is provided in every possible direction in 
order to obtain perfect harmony in the working of the 
several actions; while the inclined under-surface of the 
lever T permits a gradual movement of the strap from fast 
to loose pulley to be effected. 

The special arrangement shown in Fig. 112 is generally 
called a " strap-relieving motion," and the arrangement on 
the fallers at N may be designated a " hastening motion." 
Backing-off Chain and Faller Sector.- — The effect 
of the backing-off action on the copping faller has already 
been thoroughly explained ; it is therefore only necessary 
to describe and illustrate the method adopted for this 
purpose in the mule under discussion. 

"When the backing-off cone wheel is put into contact with 
the fast pulley (while the strap is still on the loose pulley) 
the rim shaft is driven in the opposite direction by the 
small pinion on the drawing- up shaft (Fig. 57). This 
reverses the direction of the revolution of the tin roller 
and of the spindles. A chain M, Fig. 113, is attached to 
a snail or small scroll on the tin-drum shaft, and passes on 
to a pulley K carried by a slide l)ar J. The pullej- K may 
be either single or compound, according to the movement 
required ; in this case a compound one is shown, to one of 
which pair of pulleys another chain is attached, its other 
end being hooked on to the faller sector fixed on the 
counter faller A. 

Through this faller sector on A, the movement of the 
faller ware, as it lays the yarn on the spindle, is regulated. 
(For a full description of this, see p. 93 t^ sai.) When 


the carriage has almost finished the run-in, the lower 
projecting end G of the faller leg C comes against a floor 
bracket H, and the slight further movement of the carriage 
forces the faller leg at D from the position it occupies on 
the slide E, as the winding proceeds. Directly D is free 
from E it falls down into the position shown in the draw- 

FiG. 113. 

ing, the descent being made sometimes more certain by a 
spring (not shown in the drawing) attached to C and the 
carriage end. As C is forced on one side by G coming 
into contact with H, it pushes forward a slide bar J, by 
virtue of a projecting stud on J being in contact with C. 
The forward movement of J puts tension into the spring 
attached to J at T and to a fixed bracket at S. This state 


of things, with the positions shown in Fig. 317, continues 
during the whole of the outward run. While backing-off, 
the reversal of the tin drum jduIIs down the faller sector 
through the chain M, and this also tends to pull back- 
wards the slide bar J because the pulleys at K are carried 
by it. The pulling down of the sector raises the faller leg, 
and at last it is lifted sufficiently high to allow the ledge 
at D to slip over the projection at E. The tension in the 
spring and the pull of the chain M cause the faller leg to 
shoot instantly over E, whereupon the faller becomes locked 
and ready to be actuated from the shaper through the 
bowl F. As the slide bar J shoots backwards, a stud Q 
thereon comes against an inclined part R of a lever centred 
on a bracket at P. Its other end carries a bowl IST, so 
that, directly backing-off is completed by the locking of 
the fallers, the almost simultaneous raising of the bowl 
N forces upwards the latch lever T, which releases the 
long lever and brings about the change for drawing-up (see 
Fig. 108). 

Backing-off Chain -tightening Motion. — The ar- 
rangement for tightening the backing-off chain is also shown 
in Fig. 113. A lever Y, centred at V, has a chain X attached 
to one end W. The other end of the chain is fixed to 
a small pulley on the tin-roller shaft, mounted in such 
a way that any pull on the chain X will give a movement 
to the snail round which the chain M is wound. A little 
tightening movement of the chain M is required at first, 
so the lever Y is arranged to just come into contact with 
the incline Z, carried by the shaper rod. 

A^or;?.— The incline Z is really at the outer end of the head- 
stock ; it is j)laced in the position shown in the drawing simply 
for clearness. 


As the cop builds, the necessity arises for having the 
backing-oft" chain ]\I tight, so that since Z moves forward 
with the shaper, Y is brought into contact with the incline 
earlier each draw, and in this way a little more of the 
chain M is wound on the scroll previous to backing-off, so 
that at last w^e get a practically tight chain, which is 
capable of acting immediatel)'' on the faller sector. 

Backing-off Motion. — There are one or two very im- 
portant variations of the mechanism shown in Fig. 108, the 
improvements primarily consisting of methods intended to 
quicken the backing-off action and render it more certain. 
One of these A^ariations is shown in the accompanying 
drawing. Fig. 114. 

The carriage is moving outwards ; the straps are on the 
fast pulleys; the backing- off and the drawing-up cone 
frictions are out of gear. As the carriage is completing its 
run-out, a stud or bowl at M, carried by a lever centred at 
N, comes in contact Avith a bracket L on a long rectangular 
backing-off rod ; the rod is moved forward, and as a 
consequence the studs B and C carried by it are moved, so 
that B comes under the end of the lever E, and C is moved 
out of contact Avith the backing-off lever D ; a spring K 
attached to the rod and to D is also put into tension. The 
force exerted by the spring K cannot, however, pull the 
backing-off cone clutch into gear (Avhich is its intention), 
because a projecting arm J on the strap-fork has a stop h 
Avhich prevents the lever D from moving. 

The result of the stud B coming under the end of the 
lever E is to prevent the drawing-up cone clutch from 
going into gear until such time as is necessary. Both cone 
clutches are therefore locked during the time the stud M 
is moving forAvard the bracket L and its rod. It Avill also 
be noticed that the long lever, by means of its stud J, is 


keeping the lever H from permitting the drawing-up clutch 
to go into gear. Wliile the stud ^I is still moving L 
forward, a stud on the carriage comes against the incline 
on the T lever, and lifts it ; this at once releases the long 
lever from the catch V on the T lever, and sets the lever 
H free from the stud J, so that now the spring g, which is 
in tension, exerts its full pressure to pull the lever E 
downward ; as long, however, as the stud B is under the 
end of E, the cone clutch remains out of gear. After a 
short interval (depending on the number of twists put in 
at the end of the stretch when the carriage is stopped) the 
twist latch lever is released (as already described), and the 
strap-fork is moved on to the loose pulley, and its projecting 
arm J being raised, sets free the backing-ofF lever D, and 
permits the spring K to pull D forward and so put the 
cone friction into gear. The actual backing-ofF action now 
commences; the reversal of the tin drum Avinds on the 
chain, pulls down the faller sector, and lifts the faller leg. 
The chain passes over a pulley S carried by the lever 
centred at X, so that its pull tends to draw the faller leg 
forward through the connecting link P. A spring E, in 
tension, also tends to pull forward the lever X M and 
consequently the faller leg. The gradual rising of the 
faller leg, as the chain is wound on the tin drum, at last 
brings the recess U opposite the slide Q, which rests on 
the shaper. Immediately this occurs the combined pull of 
the spring R and the chain causes the faller leg to shoot 
forward over Q, and the lever X JNI is drawn backwards. 
Four actions simultaneously occur in consequence of this 
movement of the lever X Vi. First : ]\I is taken out of 
contact with L, which permits the spring K to pull the 
backing-olf rod backwards. Second : a lever 0, working 
on the same centre X as the lever X M, is lifted, and^ 

VOL. Ill 


coming into contact with the incline /*, lifts it, and so frees 
the carriage which has been locked by the recess at Z 
fitting over the projection e. Tliird : the baching-off cone 
clutch is taken out of gear by the stud C coming against D 
and moving it backwards. Fourth : the faller leg, through 
being pulled over the slide Q, puts the copping faller in 
direct connection with the shaper. 

The first action, in moving the stud B out of contact 
with the end E of the lever F E, at once permits the spring 
g to pidl it downward, and so puts the drawing-uj) cone 
clutch into gear, Avhich action causes the carriage to be 
drawn in. At the same time the stud C, coming against 
D at the moment M releases L, moves D backwards and 
takes the backing-ofF cone clutch out of gear. 

The carriage now makes its inward run ; the stud on the 
carriage comes against II and lifts the drawing-up cone clutch 
out of gear, and so stops the carriage. At the same time 
the finger d is moved for^vard, and this releases, through the 
rod c, the long lever from the catch at W. This brings the 
stud J into the position shown in the drawing, and prevents 
the cone clutch from falling into gear again ; it also puts 
the catch box on the back shaft into gear, and so permits 
the front rollers to bring the carriage out. Simultaneously 
the incline on the faller rod releases the twist latch lever, 
and so changes the straps from the loose to the fast pulleys. 
When these actions are all finished their respective mechan- 
isms occupy the positions shown in the illustration. Fig. 114. 

Fine Spinning" Details. — A number of important 
details of the self-actor are only used when the machine is 
employed in spinning fine numbers. This discrimination 
between fine and coarse counts of yarn arises from causes 
that are not entirely obvious ; indeed, as we shall see, the 

Note. — See Appendix for farther details of Fine Si^inuiug Mules. 


reasons generally advanced for the use of some of the addi- 
tional movements are as api)licable to the spinning of very 
good coarse nnmhers as they are to fine numbers. To give 
an illustration of this we may point to the fact that several 
motions that were formerly only found on fine spinning 
mules are now to be seen on almost any mule from which 
good work is produced. INIoreover, high numbers are 
produced in a far less degree than formerly, and the skill 
that used to be displayed on counts such as ISO's to 300's 
is now turned to account in producing lower numbers of a 
superior qi;ality ; an<l where lOO's was necessary to give 
double 50's, we now find 50's by itself equalling the 
pre\'ious practice. It is no uncommon thing to see mules 
equipi^ed for spinning high counts used for much lower 
numbers. The following may be taken as suggesting the 
difference of treatment between fine and coarse numbers : — 

Fine numbers are spun from longer and better cotton 
than coarse niimbers. Long cottons are weaker than short 
cottons. More draft can be used when spinning fine 
numbers than in coarse numbers, because of the length of 
fibres. Fine numbers are twisted more than low numbers ; 
and fine numbers, owing to the delicate fibres, are strained 
through this extra twist, so that some means must be found 
to relieve them ; while for a similar reason the operation 
of spinning must be performed very slowly compared with 
the speed for low numbers. 

Double-Speed Driving'. — Some of the actions already 
described operate so prom])tly that the suddenness of action 
so produced tends to stiain the yai-n. To overcome this 
difficulty, a more gradual stopping and starting is ado[)ted, 
and, moreover, friction is reduced to the smallest possible 
degree. Some of the arrangements of mechanism for 
dealing with the points mentioned above Avill now be given. 


and tlie first example will illustrate what is generally 
termed " double-speed " driving. 

We have seen that the counter shaft controls the whole 
mechanism of the mule. It is at this point that a change 
is usually made if it is desired to alter the relative speeds 
of the various actions that are performed. Now in fine 
spinning it is absolutely necessary to perform the spinning 
process very slowly, but there is no necessity to work 
slowly while the other actions are in operation ; a form of 
driving is therefore adopted which is alternately slow and 
fast. Fig. 115 shows the usual method adopted. On the 
counter shaft, instead of a pair of pulleys, fast and loose, 
driven from the line shaft, there are arranged two sets of 
pulleys as at A and B, each set consisting of three pulleys, 
two loose and one fast. 

When spinning is taking place and the carriage is 
travelling outwards the counter shaft is driven from the 
line shaft through the fast middle pidley at B ; at tlie same 
time the other strap from the line shaft is running on the 
middle loose pulley at A. This driving continues until 
the carriage gets out, and, as backing-off can be performed 
quickly without danger to tlie yarn, a quicker speed is 
obtained by moving the stra[) forks P so that the straps 
are moved to the right from the fast to the loose joulley at 
B, and from the loose to the fast pulley at A. The set of 
pulleys at A being smaller than at B, we get by this means 
a quicker speed for the backing-ofF, and this extra speed is 
maintained during the run-in of the carriage. 

Fig. 115 fully illustrates the arrangement for moving 
the strap. A bar N, upon Avhich the strap-forks P are 
movmted, abuts against a lever V. It is connected 
directly by levers M, A and B to the upright rod L and 
through the lever at K to the setting-on rod J. The bar 



N is also connected indirectly by an arrangement of lever 
and wheels to the back shaft, from which some of its 
movement is controlled. A lever E, centred on the back 
shaft, rests upon a cam Q, which is driven by a train of 
wheels at a certain fixed speed. The revolution of this 



U--4i^ ---^,r;: 

R 1 («lJ ; 

cam lifts the lever li, and with it an upright rod S to 
which it is attached ; the rod S carries at its upper end a 
bowl T, which comes against one arm of the bell-cranked 
lever, whose other arm U bears against the lever Y. The 
lifting of the lever li takes U out of contact Avith V, ^.nd 
puts tension in the spring "W, which tends to pull the strap- 
forks from the douljlc-specd fast pulley at A and put the 


single-speed strap on the fast pulley at B. This cannot be 
done, however, until the carriage arrives fully in, when the 
setting-on rod J is unlocked, which releases the rod N and 
permits the spring W to pull the strap-forks P forward 
and allows the single-speed fast pulley to be driven. As 
the carriage moves out, the cam Q allows the rods S to fall 
and leaves the weight X pressing U against V and tending 
to force N back again. This j^ressure is exerted during the 
run-out, but the strap-forks are not moved until the car- 
riage, coming against Z, frees the setting-on rod and permits 
the strap-forks to be pushed back by the weight X. 

When it is necessary to stop the mule completely the 
straps from the line shaft can readily be moved on the end 
loose pulleys of each set at A and B. The drawing-up 
pulley E is driven from the jjulley on the counter shaft, 
and through the mule receives the change of sjjeed. 
The pulley at H, driven from G on the counter shaft, is a 
special Avinding motion, another example of which will now 
be given. When dealing with another maker's type of 
mule further on in the book, a second example is illustrated 
of double -speed driving obtained directly from the rim 
shaft. See Fig. 135. 

Winding Motion. — In spinning very fine counts, the 
change of the fallers when the carriage gets in, and wind- 
ing, as performed by the cpiadrant, is completed, results in 
a momentary freeing of a certain length of j'arn while the 
faller wires move into their new position. The fineness of 
the yarn and the twists it contains at once tend to form 
snarls and even cut yarn. Therefore a method is adopted 
to take i;p this length of yarn by giving the spindles a few 
extra turns, independently of the quadrant, just as the 
carriage is finishing the run-in and the fallers are about 
to chanire. 



Fig. IIG re2)rcsents an arrangement for performing 
this operation. The carriage is coming in, winding by the 
quadrant is in progress, and the strap is on the loose 
pulley C. On the rim shaft are placed two narrow pulleys, 
fast and loose, as at A and B. A strap from the counter 
shaft is on the loose piilley B, so that the rim shaft is 

Fig. 116 

stationary. On the carriage is fixed a stud and bowl J, 
which, as the carriage nears the finish of the inward run, 
comes into contact with an incline K carried by a lever L 
fulcrumed at M. The stud J lifts K upwards, and in doing 
BO sets free a projection N on the upright lever T, which L 
has previously held locked in the position shown in the 
drawing. Immediately T is free, a strong spring S attached 


to it pulls it over, and by means of a Ijar link P, connected 
to the upper part of T, the movement takes the strap-fork 
Q, which is attached to P, from the loose pulley B to the 
fast pulley A. Directly this happens the rim shaft begins 
to revolve, and consequently the spindles — which has the 
effect of taking up the yarn so that no snarls can be formed. 
At the same time a pin R on the link P is set so that it 
just comes into contact with the strap-fork H. The change 
of the main driving strap from C to D now takes place for 
the outward run and spinning, and as the strap-fork H 
changes, it moves back the link P by means of the stud R 
This changes the Avinding strap from A to the loose pulley 
B. For regulating purposes K can be adjusted so that the 
extra winding can be made to commence up to 8 inches 
before the carriage gets in, and in addition the adjusting 
screw enables the amount of strap that is considered 
necessary for driving A to be very delicately regulated. 
See also Fig. 117. 

Drawing-up by Belt. — For fine spinning, as already 
explained, the drawing-up cone clutch is dispensed Avith, 
and, in its place, drawing-up is performed by a strap, the 
" change " taking place by moving the strap from a fast to 
a loose pulley, as shown in the drawing, Fig. 33 ; see also 
Fig. 117. The arrangement is frequently employed on 
mules spinning counts 120's to 300's, and its object is to 
avoid the sudden change resulting Avhen the cone clutch is 
put suddenly into gear ; l)y the method shown a gradual 
movement is obtained, and all shock or suddenness of 
action is avoided. 

Gain and Ratch. — In spinning fine numbers, it is a 
frequent practice, in fact almost an unavoidable one, to 
cause the carriage to run at a slightly quicker rate than 
the surface speed of the front roller, which results in what 



is termed "gain." Further, this gain is augmented Ly 
sometimes stopping the rollers before the carriage has 

completed its run-out, so that the yarn already delivered 
is stretched still further, and, as it is popularly termed, 
"ratched." The terms " <i;ain " and "ratch" have thus 


become almost standard expressions for these two opera- 
tions, though the latter is frequently described as an " after- 
stretch motion." The effect of the " after-stretch " is 
naturall}^ to draw out the thick and thin places in the yarn 
and make it more uniform in thickness. 

Gain in the carriage is not confined to high numbers, 
though for ordinary medium numbers of twist it is seldom 
that gain is necessary. It is chiefly used for such numbers 
when weft yarn is made, and then only to a slight extent. 

For yarn containing an unusual number of t^\^sts the 
opposite effect is often produced, and the carriage travels 
slower than the surface speed of the front roller; the 
extra yarn thus delivered is taken up by the extra twist 
put into it, and in tliis way the yarn is relieved of the 
strain to which it would otherwise be subject. Fig. 117 
illustrates the gearing through Avhich the relative speeds of 
the carriage and front roller can be altered. A change of 
the wheel L, or if necessary the wheels L and K, will 
regulate the speed of carriage and front roller in relation 
to the speed of spindle, but it will not alter the relative 
speeds of the carriage and roller ; this is brought about 
by changing the pinion P through which the back shaft 
is driven from the front roller. This wheel is often called 
the "gain pinion," because of its function ; see also Fig. 118. 

Jacking" Motion. — In Fig. 16 a sketch was given 
showing how the front roller is driven from the rim shaft, 
and the back shaft from the front roller. Fig. 57 also 
showed a similar arrangement. In the drawing. Fig. 117, 
another full gearing plan of the mule is giveii, which 
exhibits the gearing employed on a fine spinning mule ; it 
therefore differs in a few details from the one illustrated 
in Fig. 57. The pulleys H are the extra winding pulleys, 
whose function was described in connection with Fig. 116. 


On reference to Fig. 117 it will be noticed that the 
wheel work, between the rim sliiift, the front roller, and the 
back shaft, is difl'erent from that given in Fig. 16, inasmuch 
as we find an extra pair of bevels M N driving the carrier 
wheel 0. To explain the function of this arrangement, and 
also to describe other features of the gearing, an illustration 

Fig. 1:9. 

is given in Fig. 118. As the rim shaft drives the front 
roller through J, K, L, C, R and S, we have the yarn 
delivered consistently^ with the requirements of twist, gain, 
and ratch ; and tlie back shaft, driven through T, 0, E, P 
and Q, drives the carriage out, in harmony M'ith these 
factors. At the same time, by introducing the bevel wheels 
]\I X, a connection is made between the rim shaft and the 
back shaft which is quite independent of the front roller, 


This independence is obtained by attaching to the boss on 
N a ratchet wheel A (see Fig. 119), which revolves within 
the carrier wheel 0. When the carriage is going out, and 
is being driven -from the front roller, all the wheels are 
revolving in the direction shown in Fig. 119; and since 
is receiving a greater speed from T than the ratchet wheel 
A is receiving fi'om the bevels M N, a number of catches 
or pawls B carried by simply slip over the teeth of A, 
and so the two wheels A and revolve independently of 
each other; and the bevels M N, so far as this part of 
their work is concerned, are useless. When, however, the 
front roller is stopped, by separating the clutch catch box 
between T and S (Fig. 121 is almost self-explanatory of 
how this is performed) the wheel will receive no motion 
from T ; but since M N continue to be driven from the 
rim shaft, the teeth of A will engage with the clutches 
carried by 0, and cause it to revolve and so drive the back 
shaft. In this Avay we continue the movement of the 
carriage when the rollers are stopped, and thus obtain 
what has been previously described as the " after-stretch " 
or "ratch." The wheels M, N and O are frequently 
spoken of as being the "jacking motion." Before leaving 
this feature it is as Avell to point out that this motion is 
not a necessary adjunct to the gearing through which we 
can drive the carriage at a quicker or slower rate than the 
surface speed of the front roller, and thereb}' obtain a drag 
or a gain. 

Roller-turning Motion whilst Twisting- at the 

Head. — Previous allusions have been made to what is 
termed "twisting at the head." By this we understand 
that, after the mule has completed its outward run, the 
front rollers are stopped, but the spindles continue to 
revolve and so put an extra number of twists into the 


yarn. These extra twists naturally put tension in the 
yarn because their tendency is to shorten it; the strain 
so occasioned would prove damaging by causing a good 
many breakages ; to relieve the yarn, the rollers are there- 
fore caused to deliver a very small amount of cotton at a 
much reduced rate as compared with that at which they re- 
volve when the carriage is moving. The effect is obtained in 
the following manner : — When the carriage stops, the catch 
box, Fig. 118, between T and S is naturally thrown out of 
gear, so that although S is driven, it simply rides loose on 
the shaft. On the side shaft, which carries M and E, is a 
pinion U, which drives through V another side shaft, on 
the other end of which is a worm W, from which the front 
roller can be driven through the Avorm wheel X and the 
pinions Y and Z. On the back of Y is a catch box, which 
is inoperative Avhen the front roller runs at its ordinary 
speed in the same way as A is in Fig. 119. But when the 
front roller is stopped the catches in Y permit the wheel to 
drive Z, and so Ave obtain from U a very sIoav movement 
of the front roller to compensate for the small amount of 
yarn taken up through the tAvisting action when the 
carriage is out and extra tAvist is being put in. " Jacking- 
delivery motion " is the name sometimes given to the 
arrangement, but it is much better to call it a "roller- 
turning motion Avhilst tAvisting at the head." 

Roller-delivery Motion whilst Winding. ^ — Another 
motion very often used, but upon the merits of Avhich 
there is an amount of reasonable scepticism, is the one 
called the "roller-delivery motion Avliilst Avinding." As 
its name implies, its object is to turn the rollers Avhile the 
carriage is coming in and Avinding is taking place. The 
reason for this action is generally sought for in the fact 
that an increased production is thereby obtained. This 


can readily be confirmed, for if tlie stretch is 64 inches 
and three more inches are delivered when the carriage 
comes in, tlie total length delivered each draw amounts 
to 67 inches. A better reason, however, than this of 
increased production can be deduced, namely, a strain- 
relieving effect on the yarn. "We know that the yarn is 
made to assume a line something like the letter Z Avhen 
the winding is taking place; this naturally puts some 
considei'able strain on the yarn, and, indeed, everything is 
done to balance this strain as much as possible. Now it 
will clearly be recognised that this bending of the yarn can 
be safely done in a long length ; but as the length gets 
shorter the strain will become greater, and to relieve it 
the rollers are made to deliver a little extra, and, of 
course, it comes in additionally as an advantage in the 

In this connection there remains an important point 
which is the cause of a difference of opinion among spinners. 
Twisting has been completed, and winding commences; 
untwisted roving is now delivered, and a question arises 
as to whether the extra three inches delivered is as well 
twisted as the remaining 64 inches. There can be no 
doubt that the twists already in the yarn Avill run up to a 
considerable extent into the extra yarn, but it by no means 
follows that the three inches will receive an amount equal 
to any other three inches in the stretch. The probability 
is that it does not, except in well-twisted yarns and fine 
numbers — in both cases because of the combination of 
natural and artificial elasticity of the fibres. This doubt 
leaves room for the difference of opinion mentioned. 

The gearing for giving the extra delivery is shown in 
Fig. 118, Avherein i is a wheel on the back shaft, and from 
it the front roller is driven through/. x\ ratchet wheel by 


/ is keyed on the front roller, and when the carriage is 
going OTit the wheel y runs in the opposite direction to the 
front roller, and so the ratchet wheel is not afi'ected by the 
pawl catches. When the front roller is stoj)i)ed, and the 
carriage runs in, the back shaft drives/ and the catch or 
catches which ) carries dip into the teeth of the ratchet 
wheel and turn it, and, consequently, the front roller. 
Another cause for the dissatisfaction as expressed by some 
for this motion will be understood from the fact that the 
extra matei'ial is delivered in a uniform manner from the 
beginning to the end of the ]un-in. Tliis is not in accord- 
ance with reason : there ought to be (on condition that 
such a motion is practically necessary) an increasing de- 
livery as the carriage approaches the beam ; or, in other 
w^ords, the front roller should deliver a little more in a 
given time towards the end of the run-in than Avhat it 
delivers in the same time at the commencement. It is 
motions of this kind that now and again make the governor 
and nosing motions more difficult to work than they would 
other^Wse be. 

Backing -off Motion. — A further illustration of a 
" backing-off " motion is given in Fig. 122. It represents 
a well-known arrangement, and one that has been ex- 
tensively applied to mules, especially to those of the " long- 
lever " system. Its action is as folloAvs : — As the carriage 
moves out, and is on the point of comjjleting the stretch, the 
end of the slide bar or gun lever F (this feature has already 
been fully described and illustrated, see Fig. 113) comes into 
contact with an adjusting screw A, carried by a hanging 
lever centred at B. To the lever at C is attached a link 
E, which carries one end of a long rod D ; the other end 
of D abuts against a stop G on the lower portion of the 
backing-oflF lever whose fulcrum is at H. When the 


carriage moves the hanging lever forward, the rod D is 
moved out of contact with the stop G, and at the same 
time a spring M, connecting the rod and the backing-ofF 
lever, is put into tension, and consequently pulls the lever 
H forward ; this action has the effect of moving the end J 
backwards, and so putting the backing-ofF cone wheel into 
gear with the cone clutch on the fast pulley. " Backing- 
off " now takes place, and when it is completed the faller 
leg locks ; as this occurs the slide rod F shoots back and 
releases the hanging lever B. A spring S, which has been 
compressed by the previous forward movement of the rod, 
is also now relieved from constraint, and at once forces the 
rod D backwards, and, abutting as it does against the stop 
G, it moves back the backing-ofF lever H, and takes the 
backing-ofF clutch out of gear. 

Fig. 123 shows a modification of the above arrangement. 
Instead of a hanging lever, a bell-crank lever is used 
fulcrumed at B ; a bowl A is carried \>j one arm, while the 
other arm is connected to the rod D through the link E. 
The slide bar F is extended, and is formed with an incline, 
so that, as the carriage moves forward, it comes into contact 
with the boAvl A, and depresses it, thus mo^ang forward 
the rod D. "When "backing-ofF" is finished, the shooting 
back of the slide bar F releases the bowl A, and, as before, 
the backing-off" clutch is taken out of gear. 

Roller Stand and Weighting. — The roller stand of 
the mule, Fig. 124, is very similar in most respects to the 
stand used on the fly frames. It consists of a principal 
bearing Q, bolted to the roller beam and carrying the 
front roller ; a projecting arm R supports a slide S, which 
acts as the bearing for the middle and back rollers. These 
two rollers being generally set a fixed distance from one 
another, the slide S is made in one piece ; but of course it 

VOL. Ill 



IS necessary in many cases to make 8 in two parts, each 
carrying one of the rollers B and C, in a way similar to 
that shown in the fly-frame roller stand. The cap bar, for 
keeping the top rollers in position, is pivoted at J so that 
it can readily be moved over out of the way when the 

Fig. 124. 


Fig. 125 

rollers require attention. The traverse rod carrying the 
thread guides is shown at H, and is generally connected at 
the outer end of the roller beam to some cam arrangement 
that gives it a to-and-fro movement, and whose object is to 
cause an equal wear of the leather of the top rollers. The 
necessity for this traverse exists wherever leather-covered 
rollers are used, and a large number of special motions 


have been introduced during the past few years for obtain- 
ing the maximum amount of use of the leather covering. 
The best motions are undoubtedly those depending upon a 
uniform cam motion, arranged with a slightly accelerated 
movement at the change in the traverse. Motions that 
depend i;pon eccentrics or cranks, in Avhatsoever form, for 
the traverse, are as a rule wrong in principle, and are 
generally complicated and unnecessarily expensive. 

The weighting of the rollers is an important matter. 
Two methods — dead weights and lever Aveighting — or their 
combination, may be adopted for obtaining the necessary 
pressure on the rollers. In Fig. 124 is shown a method 
frequently used in the mule. On the middle and back 
rollers B and C rests a lever D ; a raised point on the 
upper part of D supports one end of a lever E whose other 
end rests upon the front roller A. To E is attached a wire 
link K, which in turn is connected to another wire link L, 
and this, passing through a hole in the roller beam, is 
supported by means of a nut P by a lever M whose fulcrum 
is at F ; the lever M carries at its other end a weight W, 
the position of which can be varied for the purpose of 
obtaining a range of different pressures on the rollers. 
Fig. 125 will enable the effect of W to be thoroughly 
understood, and an example will be given showing how to 
calculate the pressure on each : — • 

The weif^lit of W = 4 lb. The distance of CE = * in. 

The distance of WF = 7i in. The „ (y& = \h ,, 

The ,, PF= 3 ^^ The ,, EB = r „ 

The ,, AD= 3 „ The „ ED = li „ 

The „ AE = 2 ,, 

The pull of the weight W at D will equal 
Weight x"\VF 4 x Ti ,„ „ 

PF -- r-=^Q"^- 

244 COTTON SPINNING chap, h 

This 40 lb. will l)e distributed, part of it on A and the 
remainder on the point E. 

The pressure on A will equal 

EDx40_ l§x40 „■, 
AE ~ 2 "~ '- 

The pressure at E = -iO - 27J = 12^7 lb., or the pressure 
at E will equal 

AD x40 _ § X 40 11) . _ 

AE ~ -2 '- -^-2. • 

The pressure at B will equal 

CExl2ilb. 1x1-24 lb. 

= 4-166 lb. 


The pressure at C will equal 12i- - 4-16 = 8-33 lb., or 
the pressure at C will equal 

BE X 124 lb. 1x124 lb. 

CB ~ U 

= 8-33 lb. 

Direct Aveighting of the rollers is performed 1)}^ placing 
a hook upon the roller and hanging a weight upon a link 
attached thereto. 

The driving of the rollers is illustrated in Fig. 126. 
The front roller through A drives a large crown wheel D ; 
on the axis of D is a wheel E, which drives the back roller. 
The back roller through C and the carrier E drives the 
middle roller wheel B. The necessary change (for draft) 
in the speed of C is obtained by changing the wheel E. 

Figs. 127, 128, 129, 130, 131 and 132 represent the 
complete sets of rollers for Avorking Japanese, Chinese, 
Indian, American, and Egyptian cottons. The particulars 
attached to them indicate the usual practice in the diameters 
and setting.^ 

Another Example of Long-Lever Mule." — The 

machine now illustrated, where the changes are produced 

^ Setting of rollers is further treated in the Appendix^ 
2 This type of mule is fully illustrated in the Appeadi?. 



^ (•tf--'j i'^iri 



i -x-1^6--r 


Fig 12S. 



through the medium of a long iever, will be familiar to most 
of our readers, and its position in the production of the finer 
qualities of counts entitles it to some consideration in these 
notes. We therefore give a few details of its principal 
actions and the mechanism emploj'ed in producing them. 


Figs. 133 and 134 illustrate the chief points of interest. 
In the former diagram the backing-ofF cone wheel and 
clutch are shown at Z. The bar or slide X is coupled up 
to the grooved boss of the backing-ofF cone wheel through 
a lever Y, so that any movement made by X will put the 
wheel in or out of gear with the cone clutcli, which is 



fast on the rim shaft. The method of doing this is as 
follows : — A projection on the bar X carries a stud AV, 
which locks itself into a notch cut in the under side of the 
connecting rod J ; another projection on X carries one end 
of a sj)ring 0, whose other end is fixed to a portion of the 


*■ 1;^ 

<!,-|/5->s |/5"^-"J> 

Fig. 130. 

framing of the machine. A cam G, driven in the direction 
of the arrow from the rim shaft through the wheels A, B, C, 
D, E and F, comes into contact with an inclined swing or 
finger H, which hangs pendant from the stud on which the 
compound carrier B and C revolves. The revolution of G 
has the effect of pushing H forward, and in so doing the 


^ ) = E6YPTIAII COnOH fe 


- r/s 




12- > * 




connecting rod S, AA-hich is fastened to the swing H, is also 
moved forward, and consequently pulls the backing-ofF 
slide X in the same direction, thereby putting the backing- 
ofF cone wheel Z into gear with the cone clutch. Backing-off 
at once takes place, and of course this is arranged, through 
the gearing from the worm A on the rim shaft, to happen 
just as the carriage has arrived at the termination of the 
outward run, as shown in the diagram, Fig. 13.3. Immedi- 
ately the backing-ofF is completed, the bar X is released in 


5 - * . -^1^ -, 



f.- IV -'r - lis- -1 


Fig. 132. 

the following manner :• — A long lever centred at ]\I is 
connected to the rod J by a link K ; its other end N carries 
an arm Q, whose loAver end passes through the holding-out 
catch V, which is fulcrumed at U. The position that can 
be taken up by the holding-out catch is carefully adjusted 
through the nuts at (^, so that, as the carriage comes out, 
the snug at T, carried by the scpiare, passes over the end of 
Vand becomes locked by the catch. Backing-off is finished 
by the faller leg being locked ; and, as this happens, the 
stud at S is oscillated and moves down the inclined finger 
at 1\, which presses against a projecting arm P of the down 



rod Q. This action at once forces the end X of the long 
lever in a downward direction, and correspondingly raises 
the end L, Avhich at once releases the rod J from the stud 
W ; K being set free is now pulled backwards by the 
spring 0, and the cone wheel is taken out of gear with the 

clutch, thus permitting the mule to run in. The same 
movement that lowers the long lever at N also presses 
down the holding-out catch V, and thereby unlocks the 

Drawing"- up. — The arrangement for drawing-up is 
illustrated in Fig. 134, The run-in of the carriage is 
effected through the pulleys E, and T, fast and loose 


respectively, on the shaft U. Dining the run-out the strap 
is on the loose pulley, as shown in the diagram. The same 
movement of locking the faller leg acting through the stud 
S, as in the first sketch, also moves the arm J upward, and 
J comes into contact with a bracket H carried by a lever 
F centred at G. The upward pressure of J lifts the lever 
F, and through the link E releases a catch finger D, and 
takes it out of contact with a stop-washer C on the rod 
; the weight W acting through L immediately pulls 
the rod backwards, and transfers the strap to the fast 
pulley T. This action, it Avill be seen, takes place precisely 
as the backing- off is finished, so that no sooner is the 
rim shaft stopped than the strap on T commences to drive 
the scroll shaft W, through the bevels P and Q, and so 
draws up the carriage. Just as the carriage is arriving 
against the stops, a bowl N on the square comes against 
the lower end M of a lever fulcrumed at K, and presses it 
backwards. As a consequence, the upper end of the lever 
at L moves forward the rod B, and changes the strap again 
to the loose pulley, thereby stopping the mule. At the 
same time the strap is moved from the loose pulley on the 
rim shaft to the fast pulley, and spinning immediately 
commences. The rod B is locked in position during the 
outward run, by the finger at D. 

Double-Speed Driving. — A recent improvement added 
to the mule is shown in Fig. 135, by which means a novel 
and satisfactory method of obtaining a double-speed effect is 
obtained. Briefly, it consists in the rim shaft being made 
in two parts, C and C^. One rim shaft C carries a large 
rim pulley B, whilst the other rim shaft Cj has a smaller rim 
pulley A keyed to it. The driving takes place through two 
fast pulleys D and F, the loose pulley being placed between 
them. When the strap is on the fast pulley F, tlic usual or 


slower speed of spindle is obtained through the rim pulley 

A ; but a change is effected, as the carriage gets out, by 


moving the strap on to the fast pulley D, whereupon the 
rim pulley B begins to drive the spindles at a greater 
speed than that obtained from A. The latter of course 
continues to revolve, but merely through its connection 
by band with B, and its movement does not afliect the 
spindles in the least. 

An additional and highly important improvement is 
effected in the arrangement by using two brake cone 
clutches at the points J and for backing-off. By their 
means a double amount of friction is obtained for stopping 
the rim shaft ready for backing-ofF. Naturally this opera- 
tion is performed very rapidly and effectively, and some 
time is saved in stopping and then reversing the spindles. 

The illustration will serve the pur2)ose of the relative 
positions of the details given in Figs. 133 and 134, the 
character in these two sketches being simplj^ diagrammatic. 

Snarls and Anti-snarling Motions. — AVe now touch 
on a sul)ject Avhich is always more or less a A'ery trouble- 
some feature in mules, and one that has been the occasion 
of innumerable devices being put on the market as remedies 
for '• snarls." During the complete cj'cle of operations on 
the mule the yarn is sup})0sed to be always slightly in 
tension ; slack yarn must be avoided, and this is one great 
reason why governor, nosing, backing-off" chain-tightening, 
etc., motions are employed, all having one object — that of 
keeping the yarn at a regular tension. If the yarn is 
permitted to become slack, it instantly doubles itself and 
forms into small curls or twisted loops, technically called 
'• snails " ; and motions to prevent snarls forming are 
generally termed "anti-snarling motions." Two illustra- 
tions will be given of such motions ; but first a few words 
as to why they are specially necessary Avill not be out of 


Directly the carriage is on the point of coming against 
the stops, a change occurs, which moves the copping faller 
Avire from the nose of the cop to a position just over the 
spindle point ; at the same time the under-faller wire is 
lowered to a position just under the spindle point ; and in 
these positions of the faller wires the yarn passes between 
them. The change of the faller wires to their new positions 
takes place very quickly, and, as it happens, a certain 
length of yarn is set free. The spindles are revolving 
during the change, so that the slack yarn resulting from 
the action just described is immediately wound on to the 
bare part of the spindle above the nose of the cop. 

The action of Avinding the yarn on the bare part of the 
spindle blade is a very delicate one ; moreover, it is ever 
varying ; for as the cop gets longer the amount of yarn to 
be wound on becomes less, and very exact adjustments 
have to be made to enable the result to be attained at all 
satisfactorily. In spite of all that can be done in this 
direction there remains an amount of slack yarn, which 
runs into snarls and thus becomes deteriorated. Extra 
motions are therefore applied, and to bring about the 
desired result two methods are generally employed : either 
the carriage starts out slightly in adA'ance of the rollers 
turning, or the carriage and rollers start simultaneously ; 
but a little extra speed is given to the carriage for a feAV 
inches of the initial part of its outAA^ard run. 

The first method is illustrated in Figs. 1.36, 137 and 
138. The front roller A is driA-en in the usual manner 
through the bevels D and C ; C rides loose on the shaft, 
and so does the other half of the catch box B. When the 
cam puts the catch box B and C into gear at the commence- 
ment of the outward run, no movement of the front roller 
Avill take place until the snugs L, cast on the catch box B, 


come against a disc K, which is fastened on the front 

Fig. 137. Fio 13S. 

Fig. 139. 

roller. The carriage in the meantime is travelling ont, 
and since the front roller is delayed in its starting, the slack 


yarn is made tight, and any snarls that may be in are 
quickly taken out. 

By referring to Fig. 138 it will easily be understood 
how the movement of the front roller, relative to the 
movement of the carriage, is regulated. When the cam 
takes the catch box B C out of gear preparatory to the 
running-in, B is free on the roller A; a leather band F 
passes over a groove on B, and each end of the band carries 
a weight ; H is the heavier Aveight, and conse(|uently it 
pulls over the part B and takes the snugs L with it out of 
contact with the disc K. The distance by which the snugs 
L can be moved away from the fingers of the disc is 
regulated by controlling the distance that H can fall, which 
is done by adjusting the vertical movement of the small 
weight at G ; a stop on J easily effects this, so that by 
means of a wing nut the motion is entirely under the 
control of the minder. This is almost a necessary provision 
to make, for, from Avhat has previously been said, it is clear 
that snarls will be larger and more frequent at the com- 
mencement of the building of the cop than at the finish. 
This arrangement permits the niinderto regulate the motion 
to suit the varying conditions. 

The second method is shown in Fig. 139. In this case 
the carriage is given a slightly additional speed over that 
of the front roller, and it is done in such an ingenious 
manner that we shall devote a few words to it. 

Instead of driving the back shaft J from the front roller 
through the usual wheels B, E, F, G and H, the two 
wheels B and E are put out of gear and the back shaft is 
driven through the wheels B, C, D, E, F, G and H. The 
wheels C and D are on movable centres, and connected 
by links L, the last one L^ being bell-cranked, with its 
centre on the stud-carrying wheel E and one end carrying 


a bowl M. Avhich fits in a cam-shaped groove on the back 
of a wlieel K driven from a -wheel J on the back shaft. 
The action of the motion is as follows : — When the front 
roller commences to revolve, the back is driven by the 
gearing already mentioned ; the Avheel J then drives the 
wheel K, and turns the cam disc. This movement of the 
cam lowers the bowl M, and naturally pulls over the two 
upper wheels C and D. The direct effect of Avheels moving 
over each other is to increase or decrease speed according 
to the direction in which they move. By observing the 
direction of the wheels C and D, it will be observed, first, 
that they -will not effect any change in the speed of B, 
because B is driven from the rim shaft direct. A slightly 
increased movement is therefore given to the Avheel E, 
which is transferred to the back shaft, and so we have the 
speed of the carriage accelerated. When the bowl M falls on 
to the circular portion of the groove no further movement 
of D and C takes place, and the carriage continues the 
remainder of its outward run at the usual speed. The 
amount of the excess speed given to the carriage is easily 
regulated b}' adjusting the cam so that a longer or shorter 
portion of the cam surface can be used. On the inward 
run of the carriage the revolution of the back shaft simply 
revolves the Avheels, and the cam takes the wheels C and D 
back to their original position, ready for the next run-out. 
One advantage of this motion is that there is no loss in 
production, because the roller is not stopped, as in the first 

A feature of some interest to many people is illustrated 
in Fig. 140. The question occurs — Do the spindles wind 
on yarn equal to the length of the stretch ? The sketch 
will settle the matter as far as actual measurements are 
concerned. When the spindle is close to the beam, the 

VOL. Ill S 


distances of the point A, horizontally and vertically, are 
shown ; from these dimensions we can readily prove that 
the length of the yarn A D is I'd inches. As the carriage 
moves out, the angle of the yarn varies ; and on the assump- 
tion that the spindle point travels 64 inches we shall get a 
length of yarn, between the spindle point at B and the 
front roller D, equal to 67 "52 inches for productive purposes : 
therefore (and without taking into account the stretching 
of the 3'arn that may occur during the rvm-in) we clearly 
see that there is 62 "91 inches to be wound on the spindle 
at each draw — which means that a stretch of 64 inches 
gives us a length of yarn 1 ^-^^ inch less, equal to a loss 
of about 1 "7 per cent. The investigation opens up several 
interesting questions, but for the present purpose it is not 
necessary to go any deeper into the subject. 

A variety of conditions arise to cause snarls, but these 
are usually remedied by attention to the following points : 
(1) Too great a movement of the nut up the quadrant for 
any given layer; when this occurs enough winding does 
not take place and slack yarn is the result. (2) Bad roAangs, 
whether through poor piecings or iri'egular sliver. (3) 
Slack scroll bands. (4) Faulty nosing motions. (5) Insuffi- 
cient weighting of the fallers. (6) Slipping of the winding 
catch. (7) Irregularities in the "changes." And (8) mis- 
calculation in the amount of the drag. 

Tubes and Starch for Cop Bottoms.— In commen- 
cing to build a cop bottom, we may either do it entirely on 
the bare spindle ; or build it upon a short or long paper 
tube ; or brush over the first two or three layers with starch. 
All these methods are adopted according to the class of 
work being done by the machine. The first one, however, 
is not often met with, so we will confine our attention to a 
few words on the use of tubes and starch. The object in 


using either of these methods is primarily to obtain a good 
foundation for the cop bottom so that in future use the 
cops can be passed on to a skewer without stabbing and 
spoiling the cop. The avoidance of waste in other direc- 
tions is an important factor, for it is desirable to use if 
possible every inch of yarn Avound on the sj^indle. From 
this point of view the use of a short tube pressed on the 
spindle where the cop bottom is formed ensures that a good 
opening is always left for a skewer to pass through, and 
another advantage is apparent, for in such a case all the 
yarn can be unwound without leaving waste. In some 
districts spinning finer counts, tubes are almost exclusively 
"used, but they have their disadvantages, among which 
might be mentioned the following : extra labour is involved 
in putting them on the spindles, and this means a slightly 
increased cost for such labour ; the tubes are pressed to 
their places, and sometimes in doffing they stick so fast 
that 'the cops are pulled out and of course waste is made; 
damaged tubes are a source of breakages whilst winding 
and unwinding, on account of rough edges ; when ends 
have been allowed to remain down for a few draws it is 
not so convenient to push the cop up a little, so that the 
cop is nicked and spoiled yarn made ; the few turns of 
3'arn round the spindle, previous to putting on the tubes, 
accumulates so much that it becomes a little troublesome to 
occasionally clear the spindles. There are several appliances 
that dispense Avith a good deal of the labour involved in 
putting on the tubes ; these are filled Avith the tubes during 
the Avorking of the mule, so that Avhen doffing is complete 
they are ready to be at once turned over on to the spindles 
without having to put on each tube separately. 

Starching is performed by ap{)lying Avith a brush a little 
starch to each spindle before starting to build the cop 


bottom ; when dry, it effectively prevents the hole closing 
under ordinary working conditions. If done properly and 
good starch is used very little waste is caused and very 
little labour is entailed, as a rapid movement of the brush 
(which is attached to a special box holding the starch which 
runs on to the brush) along the spindles enables the whole 
of the spindles to be starched in a minute or so. If a 
minder,' in his desire to have a better cop bottom, starches 
twice and also puts on a layer or two before doing it, he of 
course sacrifices a little time and in addition causes more 
waste to be made at the loom, but this gives a much better 
cop and the fact induces the practice to continue. Bad 
starch, carelessness in starching, and the starch running 
down the spindle to the bolster bearing, are its great dis- 
advantages, and a very frequent complaint results from the 
soft cop bottoms made. Longer tubes are generally used 
when a hard cop is desired and the yarn is spun rather soft. 

Horse -Power required to drive the Self- Acting 
Mule. — -It is now proposed to present, as briefly as 
possible in description and diagram, a digest of present 
knowledge as to the power required to drive the mule. 
Nothing will be said about the methods adopted to obtain 
the indications, beyond remarking that dynamometers of 
various kinds have been used, and careful observations 
taken of their results. 

If one were to ask the question — What horse-power 
will a mule take to drive it % he would probably be 
answered, in a general way, that 110 to 120 spindles per 
I.H.P. for low cottons, and 130 spindles for finer cottons, 
are good averages. An ansAver of this kind is quite 
sufficient for ordinary purposes ; and, as a rule, a result in 
such general terms can readily be obtained through the 
indications of the steam-engine. Like all general state- 


ments of facts, however, there is a tendenc}^ to overlook 
the circumstances and details which give to the statement 
its importance, and in consequence false ideas interfere 
with the true knowledge of the conditions that go to make 
up the average. Owing to the complicated and varying 
actions of the mule, it is by no means an easy matter 
to obtain exact results. When a dynamometer is used 
■without an automatic recording apparatus, a large number 
of careful observations must be made so as to include as 
man}' complete draws as practicable ; from the numbers 
thus obtained, as well as the intervals of time of their 
indications, a good average from each set of readings Avill 
be procured, from which it is possible to arrive at a com- 
paratively accurate result. This result can be represented 
in a diagrammatic form similar to the indicator diagram 
of a steam-engine, and therefrom, in a similar manner, 
much of the inner working of the mule can be rendered 

In watching the actions of the mule for the purpose of 
indicating its power, three distinct actions stand out from 
the others, namely : the outward run of the carriage, 
during which the spindles run at their greatest speed Avhen 
spinning ; the pause or rest at the finish of the stretch, 
during which backing-ofF takes place ; and the run-in of 
the carriage, during Avhich the spun yarn is wound on the 
spindles. It Avill be apparent to any one who has watched 
the mule working that all these actions require different 
degrees of power to perform them. To a close observer 
another marked feature connected with the power absorbed 
will not be overlooked. When the carriage is at the roller 
beam the machine is practically stopped, so that the 
commencement of the run-out requires a very high power 
to overcome the inertia of such a large, heavy, and 


stationary mass as the carriage, especially as it is driven 
from the front roller, and to drive the spindles at the high 
rate of sjDeed at which they run Avhen spinning. This 
feature is properly included in the power absorbed during 
the outward run ; but it is such a distinctive element in 
the power-diagram of the mule that it might almost be 
considered as quite separate from the power that really 
effects the drawing-out and twisting processes. 

Apart from the extreme care required in making the 
observations of the readings of the dynamometer, equally 
careful attention should be paid to the speed of the counter 
shaft. It is upon this speed that the accuracy of the 
results dejDcnd, and therefore means must be adopted to 
denote the slightest variation of speed that takes place 
during the time the indications are made. One of the 
best ways of doing this is by means of the tachometer ; 
when this speed indicator is applied to the counter shaft, 
variations are instantly shown. A noticeable feature in 
this connection will be observed when the carriage com- 
mences its outward run. The counter shaft is running at 
its full speed, with the driving belt on the loose pulley of 
the machine ; immediately the belt is moved on to the fast 
pulley the whole carriage has to move and the spindles are 
driven at their full speed. It is almost impossible for this 
to occur instantaneously — the shock would be too great ; 
so, in consequence of all the actions concerned being driven 
through belt or bands, we find, when this full power is 
thrown on them, a large percentage of slipping occurs, 
which allows the carriage and spindles to attain their 
maximum speed gradually. By the use of the tachometer it 
is easily seen that spindles do not attain their full speed 
until from 12 to 36 inches away from the roller beam, and 
in a few cases machines may be found in which the spindles 


only attain their maximum speed, just as the outward run is 
finishing. This of course means that the power is consider- 
ably reduced from what it would be if the speed remained 
normal throughout. During the course of a large number 
of power tests on the mule, the writer has found invariably 
a large percentage of reduction of speed in the counter 
shaft at the moment when the carriage begins to nioA'e 
outward. In some mules this reduction is much greater 
than in others, ranging from 10 per cent up to as high as 
35 per cent. The reader will therefore see the importance 
of observing very closely the variations in this important 
factor of the indications. 

In the accompanying drawing (Fig. 141) three power 
diagrams of the mule are given. The one marked A is a 
machine with a normal spindle speed of 9100 revolutions 
per minute. The speed of counter shaft started at 410 
and an interval of '2h seconds elapsed before it attained 
its normal speed of 460 revolutions per minute. The 
percentage of slippage is almost 11, which, in the opinion 
of the writer, may be considered a very low one. The 
fact that it is so low accounts for the very high power 
indicated as the carriage started out ; and, although only a 
short mule of G36 spindles, the initial power required to 
move the machine was over 24"5 horse-power. All the 
belts were newly spliced and the bands renewed for the 
test, so that slipping was reduced to a minimum. Directly 
motion was imparted to the carriage and spindles the 
power rapid!}' fell, and the speed of the counter shaft rose 
until, after an interval of 2i- seconds, normal conditions 
svere attained. From this point onward, the power 
required to drive the spindles and carriage remained 
stationary until the outward run was completed and 
backinsr-off commenced. 

264 COTTON SPINNING chap, ii 

The stoppage of the carriage and spindles naturally 
results in an almost instantaneous fall in the power 
absorbed, but there is sufficient movement going on in the 
mule to prevent its falling to zero. One of these move- 
ments is the action of backing-ofF, and as it occupies in 
Diagram A about 2| seconds, the power required in this 
interval is shown to be \\ H.P. This, it may be observed, 
is a comparatively high power to be absorbed during 
backing-ofF; it ought never to be above 1 I.H.P., even with 
the largest machine. It was noticed that the backing-off 
cone Avas extremely hot, and the cause of this may have 
given rise to the extra power. 

When backing-ofF is finished, the drawing-up commences. 
At this point the moving of the carriage from a state of rest 
seems to give no occasion for miich extra power — probably 
on account of the band working on the small diameter of 
the scroll, and so moving the carriage gradually in. The 
power during this action has risen to almost 3 H.P., and, 
naturally, after the carriage has passed its quickest speed 
halfway in, the force exerted naturally falls until the back 
stops are reached, when it is zero. The whole draw took 
12 seconds to complete it. 

The explanation of the other two diagrams, B and C, is 
similar to the above, but it is interesting to observe the 
difFerence in the power absorbed during the several actions. 
In B we have a mule of 1000 spindles of 8000 revolutions 
per minute. Its initial power is very low, in consequence 
of a large percentage of slippage (and this, it may be 
remarked, considerably reduces the average power of the 
machine, and gives to it an apparent advantage which it 
would not have if all its belts and bands were in perfect 
order). Its normal power in driving the spindles for 
twisting, bears comparison with diagram A, considering its 

i ^ 5 6 7 S £ 

< COM'P'LE.TE. TDK AW ->; 

I I 

Fig. 141. 



lower speed of spindle and the greater numlDer of spindles. 
As one might expect, B absorbs more power than A 
during the drawing-up, and it is quite clear the backing-oflf 
process is much better arranged and requires less power to 
perform it. In Diagram C we have a machine a little over 
one-half of B, and whose motions almost correspond in 
respect to time. Its speed of spindle, however, is much 
greater, namely : 11,100 revolutions per minute, and this 
gives it a greater advantage when we compare their 
average results : — 

In A the average number of spindles per horse-power = 7l "5 
111 B „ ,, .,, „ =847 

In C „ ,, ,, „ =87*3 

The above averages, it must be remembered, include the 
driving of the counter shaft. 

To show the difference between diagrams which are the 
result of a large number of individual readings, and one 
that is automatically recorded by the instrument itself, we 
give in Fig. 142 a drawing adapted from one issued by J. 
J, Rieter and Co., AVinterthur. Its chief characteristics 
correspond very closely with those given in Fig. 141. The 
erratic motion indicated from C to D is probably owing as 
much to the dynamometer as to the mule ; and the same 
may be said of the irregularity of the curve which indicates 
the backing-off and the drawing-up actions. A peculiar 
feature of the diagram is the line representing the backing- 
ofF from E to F. The writer has on several occasions 
observed a slight increase of power at the moment of 
putting the backing-ofF cone into gear, but it was so 
momentary and variable (in many cases it was entirely 
absent) that in the previous diagrams it is ignored. It 
represents the reversal of the spindles, and from this point 
the curve will be a gradual one to F, when drawing-up 



commences. Although the curve from E to F is automatic- 
ally recorded, it does not follow that it is correct ; the 

Fig. 1-12. 

sudden changes which are brought about in the nuile and 
the short intervals of time allowed for the actions do not 

268 - COTTON SPINNING chap. 

allow tlie indicator to fall suddenly to the real pressure 
before another action comes into play and causes it to rise 

In all dynamometrical indicators some means must be 
adopted to prevent the pointer from vibrating, on the same 
or a similar principle as the dashpot of a steam-engine 
governor. When the finger or pointer has been forced 
suddenly upwards, only a slow descent can be made, which 
depends on the character of the regulator used ; and if, 
during the descent, another action comes into play, then we 
lose the real diagram that ought to be produced. An 
example of this is seen at E to F. The pressure falls 
suddenly at D, when the carriage stops ; but before the 
figure has time to fall to the pressure that this stoppage 
represents, the backing-off takes place, and so the pointer 
must fall the remainder during the time backing-off takes 
place. The writer has experimented in this direction, and 
can speak from experience to the extent that such a curve 
as shown from E to F does not represent truly the actual 
conditions of power at that point. 

On reference to the diagram again, it will be noticed 
that three seconds elapse before the normal speed of spindle 
is obtained. This is accounted for either by the slowness 
of the pointer in falling, or by the great power required to 
bring the spindles up to their maximum speed. In either 
case the average power is increased as a coftsequence, 
though, even so, the number of spindles per horse-power, 
namely 90"9, compares very favourably with any of those 
given in the first diagram. On the whole this diagram 
may be taken as representing very fairly good average 
results of a mule. 

As showing to what extent good results can be obtained 
from a mule working under ordinary conditions, we 



reproduce in Fig. 143 an adapted diagram taken from 
one prepared by Sir Benjamin A. Dobson of Bolton ; its 


ZBi. _-! ' I. 



- r---r--, — , ^— .--p- , - .- 




: : i 1 

--{-■:— f—i—v-:--|— !—-;-- 



\ \ '' \ 

L ' ■ ' 

: 1 1 : ! 1 ; i" ! 

' 1 ' 1 

2!l|.- .i-.4..|._ 

ZO r- -\--J.--\- 

K" rr ii 


; ; : i 

vf) 1 ; ■ 1 

1 1 '' r 

9 L..,_.|...., 

1 1 j 1600 

— .-- r- -| — ( 1- ' -\ — (--j-j-- 

lK£v&ioF|>P|"N-raLEi&."J" '"'!'""!" 
Av'^^HfTc^ EfHio-Pv^'E povv'et^T y 




. 1 I 1 1 r--! i--r--r- 

: 1 1 1 : 1 ; ■ , 






i \ : : 

1 i _' ' i L 1 -l-.l 

! ; ; \ ; 

! 1 '^"1 - 

1 M H 

1 : 1 ! 1 ! i ;=!=^-.;.-4--L..; 

6r--; --|-'"'. 
S '..- -1. --;-.- 4 

4 !- . -j- -I.-.; 


2\.:. ■ .: 
1 i-AXjA. ; 

-^. .^^'j 

o I a. 3 4- A' £ 7 8 9 10 .11 la. ;ii h is i 

: StOONTDS. — ;■ : ; 

ii XWI&TINC-^ x*«CKIN^W|tTOIN£if 

!< COryTPUE-T-E. TiT=(.AW- >, 

Fig 143. 

average horse-power comes out very little lower than the 
normal power required to drive the sjjindles wliilc twistin^' : 


and the average for the number of spindles per horse-power, 
namely 105, is considerably higher than those obtained 
from the other diagrams illustrated. 

Before just comparisons can be made on the power 
required to drive various types of mules, several important 
factors must be known : for instance, tlie diameter and 
shape of wharve, length and diameter of spindle, speed 
of .spindle, and also the gauge. All these factors help to 
increase or decrease the power absorbed, according as they 
are greater or less. One-eighth of an inch increase in the 
diameter of wharve makes a considerable difference in the 
power; and when the length of spindle is increased even 
by only half an inch, the extra weight, revolved at 10,000 
revolutions per minute, has some effect on the force required 
to drive it. 

It may be pointed out that in none of the diagrams 
given does the number of spindles per horse-power ap- 
proach those given in the earlier portion of these notes. 
It is fortunate that large margins are usually allowed for 
in the steam-engine, and that is why in large mills it is 
seldom that more than two or three mules at a time are 
working synchronically. In smaller mills, however, and 
especially old ones, it sometimes happens that a number of 
the mules get working in unison — to the considerable detri- 
ment of an already overloaded engine. It is to be hoped 
that as our knowledge extends of what power the mule 
really requires to drive it, a reduction in the extreme 
estimates at present in vogue will be made, so as to con- 
form to the practical results of ordinary working conditions. 

Calculations. — lu connection with the calculations of 
the mule a remark made earlier in the book may be repeated, 

Note. — See the author's book on Cotton Spinning Calculations for the 
gearing plans of other makers' mules. 


namely — always get the speeds of quick running shafts, 
such as counter shafts, rim shafts, and spindles, by means 
of a speed indicator, which denotes the number of revolu- 
tions "without necessitating the use of a watch. In this 
way there is no possibility of mistakes happening in 
expressing the actual revolutions per minute. Xo mill 
ought to be without one or two of such indicators, and no 
calculated speeds ought to be used when a speed indicator, 
such as the tachometer, is applicable. In the absence of an 
indicator, the following rules will be found serviceable : — 
To Find the Speed of Bim Shaft per Mimite. 

Revohitions of line shaft x drum on line shaft x drum on counter shaft 

Pulley on counter shaft x pulley on rim shaft 

250 revolutions X 30 in. X 24 in. ,.,„ -, ,. -. . , ^ , 

; : — = /50 revolutions or rim shaft. 

15 in. X 16 in. 

From this point Fig. 144 wiU enable us to follow out the 
necessary calculations ; reference may also be made to other 
sketches of gearing and driving which have appeared in 
these pages on the self-acting mule. 

c, 1 r • 11 Revolutions of rim shaft x D x ?i 
Speed ot spindles = 

=- — — = 10,800revolutionsof spindles. 

T, , ^. /■ • 11 ^ c ■ Revolutions of spindle per min. 

lievolutious 01 spindle to one 01 nni^.,:^ = — : -—J — r — — ' .— 

Revolutions of rim shait per mm. 

— ^ = 14 "4 calculated. 


It will be observed that the speeds so far have been 
"calculated," but it is almost unnecessary to point out that 
the use of belts and bands for dri^nng occasion considerable 
slip. This must be taken notice of in all calculations, and 
for practical purposes 5 per cent is generally allowed. The 
above calculated speeds must therefore be reduced by this 




amount in order to obtain adncd speeds. The speed of 
spindle, then, becomes 10,260 revohitions per minute. 

Tlie Ratio of the Speeds of the rim and spindle is a very- 
useful nixmber to find, as it serves for a constant in finding 
the twist and twist wheel. The number so found as above. 


namely 14'4, simply means tliat the spindle revolves 14-4 
times faster than tlie rim shaft. 
Turns or Twist per Inch. 


Mule Twist. Multiply the si^uare root of counts by 3 '75 
Mule Weft. „ ,, ,, 3 '25 


Mule Twist. Multiply the sijuare root of counts by 3'606. 
Mule Weft. ,, „ „ 3-183. 

Tioist per Inch. 

_, . . , Length of yarn delivered or put up per draw 

'■ Revolutions of the spindle per draw 

This is a well-nigh impossible rule to apply in actual 

practice, so in its place it is sometimes modified, by assuming 

the rollers to run for one minute, and finding how much 

yarn would be delivered in that time. If the amount be 

then divided into the spindle speed per minute the result 

gives the twist per inch. 

Sometimes the revolutions of the rim shaft per draw are 

first found by dri'vdng the mule very slowly ; if this is then 

multiplied by the turns of spindle for one of rim, and the 

product divided by the length of stretch, Ave get the turns 

per inch. For instance — • 

Revols. of rim per draw X turns of spindle for one of rim rn • j^ ■ ^ 

S-rp -. -. — \ = i wist iier inch. 

total travel 01 carnage 

This cannot be relied upon for exact purposes, for there 

will clearly be far less slipping in di'iving the mule slowly 

than under ordinary speed conditions. If the twist wheel 

is used on the mule it is comparatively easy to adopt it 

as a basis for finding the twist per inch. For instance — 

Turns of si)indle for one of rim X twist wheel B _, . . , 

-:si \ h-- — T e z i = Twists per inch. 

IN umber 01 inches of yarn put up per draw ^ 

This is on the assumption that the tAvist wheel B moves 

VOL. Ill T 


the strap on to the loose jDulley after one revolution. If it 

revolves twice before changing the strap we should put the 

rule thus — 

Turns of spindle for one of rim X twice the twist wheel B m -^ • i, 

s- — — = Twists per inch. 

Number of inches or yarn put up per draw 



Twut Wheel. — The foregoing rule also enables us to find 
the twist wheel for any given counts. For — ■ 

Turns of spindle for one of rim _p 
Number of inches put up per draw 
Constant x twice the twist wheel = Twists per inch. 
Constant x twists per inch = Twice the number of teeth in the twist wheel. 

Example : — 


^^= -221 Constant 

■221 X 100 = 22-1 twists per inch, 
or -221 X 22-1 = 100 (Twice the twist wheel). 

The twist wheel would therefore have 50 teeth. 

Bevolutions of Front Roller per draw — 

Length of the stretch — the gain „ i i- e e ^ ^■< 

T^. " iT? 1 — v^ ;rv7r^ = R«'^'olntions of front roller. 

Diameter ot front roller x 3-1416 

If there is nogain in the mule, then this factor mustbe left out. 


Revolutions of rim shaft per draw X J X L X R ^ ,< e ^ ^^ 

TV — 5 p rrr-n .-, , ,.,,. =lvevs. of front roller. 

Kx S X diameter ot l.K. x 3-1416 

Revolutions of Front Roller per minute — 
Revolutions of rim shaft x J x L x R 

K X U X S 
Back Change Wheel 

= Revolutions of front roller. 

Twice the twist wheel x the rim spur_ . . „ 
~ Revolutions of front roller per draw 

2xBx J _ . p 

" Revolutions of front roller per draw 


The Drafts in the rollers of the mule are Avorked out 
practically in the same way as in the previous machines, 
the arrangement of the gearing Ijeing the same. A is the 
draft wheel. 

Draft = — — r (The front and back rollers are ec^ual in diameter.) 
mxlx diameter of front roller 

Constant for draft = 

Draft = 

Draft Avlieel = 

Draft wheel = 

Total draft in mule = 

k X diameter of back roller 

Draft wheel 

Hank roving x diameter of front roller xlxvi 
Counts wanted x diameter of back roller x k 

in xlx the stretch + roller motion 
A X ^- x length delivered by the rollers 

If the correct draft is known when spinning certain counts 
and it is Avashed to change to another count, using the same 
hank roving, it becomes a case of simple proportion of 
inverse order, for if count 40's has a change wheel of 26 
teeth, then 50's will require not a larger Avheel but a smaller 
one, so that the Rule is : — 

Diaft wheel x present counts -tn ^ , , . , 

; — T : = Draft wheel required. 

required counts '■ 

Under conditions of changing both the counts spun and 
also the hank roving the 

y. . r. .1 , Required haukrovingxpresentcountsxpresentdraft wheel 
Present hank roving x required counts 

When changing the twist Avheel for a change in the 
counts, it is as well to use a foundation rule occasionally, 
such as the one already given ; but for convenience, when 
once a correct twist wheel has been used, the practice is 


frequentl}^ adopted of using this as a standard from which 
to obtain the one required ■wlien the counts are changed. 
It is based on the fact that the twists per inch vary as the 
square root of the counts. From this we say that, if the 
square root of 60's counts requires an 80 twist wheel, the 
square root of 40's counts will require a jiroportionately 
less wheel. 

Example : — 60's cou.nts are being spun, and the twist wheel 
has 80 teeth. "What wheel is required for 40's counts? 

The square root of 60 = 7 "745. 

The „ ,, 40 = 6-324. 
If 7 "745 requires a wheel of 80 teeth, 
Then 1 ,, ,, SO 

And 6-324 


t „ 





80 X 


= 65 




Readers will perhaps be unfamiliar with this rule, but it is 
based strictly upon reason. The following Huh is the one 
generally used : — 

rp • , .. T _ /Twist wheel "^ x required counts 
■> Present counts 

The slightest acquaintance with ecpiations will enable any 
one to prove that this rule is derived entirely from the first 
one. For the sake of a few Avho desire to know whj' such 
a form of rule is adopted, the explanation is given, as 
follows : — 

If N 60 requires a wheel of 80 teeth, 
Then siT „ ,, ^ 


And \^40 „ ,, 80 X s/ iO 

.81^=65 teeth. 



Froof. — First square the expression, so- 

/SOx \'40' 

This eijuals 

Then take the square root of this result, which maj' be 
expressed so — 

/ 80^ X 40 
\^ 60 

This gives us the familiar form of the rule ; for 80 is the 
present twist wheel, 40 is the required number of the 
counts, and 60 is the number of the present count. 

It is sometimes necessarj' to change the front and back 
roller wheels as well as the change wheels. If the reader 
is acquainted with simple equations there is no necessity to 
learn off a number of rules applicable to each case. From 
one rule he would obtain any change required. 

Example : — Using the drawing for the letters on the 

m X I 

Draft = -7 r • 

A X A; 

,„ ., 7 

Front roller wheel ^• = 

Back roller Avheel m - 

Top carrier wheel l- 

A X draft 
A X Z; X draft 

A X i- X draft 

Chancfe wheel A = , = — '^r- • 

° kx draft 

It is thus seen that, by using a simple form of rule, expressed 
as a formula, any one of the factors can be deduced, pro- 
vided we know all the others. 



General Description, — The ring frame, so far as its 
general mechanism is concerned, is probably one of the 
simplest and most easily understood machines in a cotton 
mill ; and 3^et around the problem associated with its central 
action we find a peculiarly divided state of opinion, founded 
— as all opinions on ring spinning must be founded^ — on a 
mixture of theory and practice. The subject has a special 
interest of its own ; partly because of the great and in- 
creasing rivalry of the ring frame with the mule, and partly 
because of the unsolved or only partly solved proljlems 
connected with its twisting action and its effect ixpon the 

Before inquiring into the cause of this imusual interest, 
we shall first give a general description of the machine 
itself. The ring frame in its charactei'istics is practically 
the same machine as the flyer throstle, and in describing 
one we practically describe the mechanical arrangements of 
the other : the difference exists in the spindles and in the 
method of putting the twist into the yarn. In the flyer 
throstle the spindle is a plain rod of steel, surmounted by 
a flyer ; this is driven from a tin roller, and its I'cvolutions 
determine the amount of twist put into the yarn, exactly 



as in the fly frame. The differential speed between the 
bobbin and the flyer, for winding purposes, is obtained by 
letting the bobbin run loose on the spindle, and allowing 
the twisted yarn to drag it round. If the spindle runs 
say 7000 revolutions per minute, and yarn connects the 
flyer leg with the bobbin, the bobbin is dragged round at 
the same speed as the spindle ; but the rollers deliver 
roving, and this when twisted decreases the tension between 
the flyer and bobbin, and the bobl)in hangs back a little in 
its speed, and consequentlj^ has the deliA^ered yarn Avound on 
to it. Over-running is prevented by carefully grading the 
drag, which is obtained by resting the bobbins on some 
rough surface, such as flannel washers. Yarn made in this 
way is considered to be of a very superior qualit}-, but the 
system has now been practically discarded for the ring 
frame, so that it is unnecessary to enter into any detail as 
to its working. 

Fig. 145 illustrates our general remarks on the ring 
frame ; half the machine is shown in section and the other 
half in elevation. It is a double-sided machine, i.e. each 
side contains a long line of spindles, suitably spaced, and 
carried by strong rails, as at H. The spindles are driven 
by bands from tin rollers T and T — (in some cases only 
one tin roller is used) — the band passing from the wharve 
G over the top of the nearest tin roller T , and on round 
the farther tin roller T. The entire driving of the machine 
takes place through one of the tin roller shafts, and in the 
illustration given, X is the shaft chosen for the purpose ; 
the other tin roller T^ is frequently driven entirely by the 
spindle bands, which, passing from T over its upper surface 
and on to the wharves G, drive it simply by the friction 
of their contact in going forward. On the driving shaft X 
is fixed a wheel J, which by a system of gearing ultimately 



drives the front roller through the wheel P. A compound 
carrier wheel L M is introduced into the gearing, arid at 
this point any change of speed required in the front roller 

can readily be effected by replacing M with a larger or 
smaller wheel. The bobbins A from the last fly frames are 
placed in the creel, and the roving is passed over wire or 
wooden rods B, on to the rollers. Three lines of rollers 


are generally employed, and, as in all the previous machines, 
a draft is introduced for drawing out the slivers. From 
the front rollers the roving is passed through a guide Avire, 
which is placed directly over the centre of the spindle, and 
on through a small piece of steel, bent in the form of the 
letter C, called a "traveller," which clips loosely a specially 
formed steel ring, fixed on a movable plate called the 
"ring-plate." The spindle and bobbin pass through the 
centre of this ring, and thus after the yarn is threaded 
through the above-mentioned steel traveller it is wound 
round the bobbin. The revolution of the spindle, which 
is run at a very high speed, in its attempt to wind on the 
yarn, pulls the traveller round with it, and relieves what 
would otherwise be a tension in the yarn ; at the same time 
each revolution made by the traveller jmts a twist in the 
yarn, and as the bobbin can only wind on the amount of 
yarn delivered by the rollers, it follows that the traveller 
is made to revolve almost as quickly as the spindle, so 
that we get a most effective twisting operation performed. 
This is merely a general statement of the action ; it will 
be treated fully in subsequent pages. In order to build 
up a cop on the bobbin or spindle, the ring-plate is made 
movable, so that by a special lifting motion it is raised 
and lowered in a manner suitable for the formation of the 

Driving'. — Treating in detail the various features of the 
machine, we shall first briefly mention the driving. Sup- 
posing A in Fig. 146 to be the driving pulley of the ring 
frame, it is possible to drive A in three different ways, 
namely — direct driving ; gallows guide-pulley driving ; or 
driving by half-twisted belt. The last two systems are the 
ones most generally adopted, and the line shaft is in each 
case at riijht anules to the driving shaft of the machine. 


With gallows or guide pulley the line shaft may be some 
distance away, so that the gallows pulleys simply serve to 
guide the strap on to the machine below. 

As already shown, the tin drum on the driving shaft A 
drives the spindles on the side of the machine marked E. 
The bands, in passing from the top of the tin drum on 
A to the spindle, clear the top of the tin drum on B, 
but the same band from the lower side of A must pass over 
the top of the tin drum on B, so that it will be seen that 
the tin drum acts as guide pulleys to the spindle bands. 
Friction is set up by the large number of bands in a frame 
to such a degree that whenever two tin drums are used 
one of them receives no other motion than that obtained 
through this friction of the bands. Very effective driving 
is obtained in this way, but a little thought on the matter 
will lead to several conclusions to the disadvantage of the 
system. In the first place, the spindle bands, whose object 
is to drive the spindles, are called upon to also drive the 
tin drum at a speed of, say, 700 revolutions per minute ; 
the extra strain thrown on them of course soon destroys 
them, and rej^lacing is a frequent necessity. This is a 
tangible fault, and requires consideration from an economical 
point of view. Secondly, we may readily assume that, 
since the tin drum on B is driven merely by the friction 
of a number of bands, the spindles on the F side of the 
machine will exhibit a larger percentage of loss by slippage 
than the spindles at E. From a practical point of view 
this objection may be dismissed, for even when, through 
some local cause, a diflPerence is found, it is so slight that 
it may without disadvantage be ignored. 

The disadvantages mentioned weigh with some people ; 
consequently machine makers are called upon to adopt 
means to overcome them. In the accompanying sketch, 



Tig. l-iG, a general idea of one Avay of doing so may be 
obtained. On each tin roller shaft is placed a band pulley 
A and B. In some accessible jxxrt of the framing is fixed 

Fig. 140. 

Fig. 14S. 

Fic. 147. 

Fig. 149. 

a guide pulley C, carried by a bracket G, arranged so that 
adjustment can easily be made through the screw H Avhen- 
ever the band becomes slack. A guide pulley is also 
provided at D, with the object of ensuring a good grip of 
the band on the j)ulleys A and 1>. The band is threaded 


over tlie pulleys as shown in the sketch, and we can readily 
understand that B, by such means, can be driven from A 
with less probability of slippage than by the spindle bands 
alone. Its practical advantage is apparent in the greater 
lasting power of the bands, and an economy is at once 
effected in this direction ; but in regard to the speed of the 
spindles at F a merely fractional improvement is recorded. 
In one way the application of the band pulleys is a most 
decided disadvantage, and this in a direction that is very 
palpable if the trouble be taken to test it. A number of 
dynamometrical tests in ring frames shows an unmistak- 
able increase in the power required to drive the machine 
when fitted with the apparatus, from 10 to 20 per cent 
being no uncommon addition to the usual power. It need 
scarcely be pointed out that this loss outweighs the economy 
of spindle bands, and as a consequence many spinners 
refuse to use such a doubtful improvement. 

Roller Stands and Weighting. — Passing from the 
driving, we shall give some little attention to the roller 
stands and weighting. Figs. 1-47 and 148 illustrate the 
general arrangement of both features. It will be noticed 
that the arrangement of the rollers follows on the same lines 
as that of the fly frames and of the mules. One important 
difference exists, however, as seen in the tilting or inclination 
of the rollers as a body. The reason for this is a simple one, 
which can readily be understood ; the yarn as it comes from 
the front roller passes to the thread guide at such an angle 
that it must pass over a portion of the surface of the front 
roller before it is clear ; also the yarn is in contact with the 
thread guide (as at A, Fig. 148) all the time it is passing 
forward to the bobbin. Two points of contact are therefore 
tending to stop the twists put into the yarn by the traveller 
from getting up to the nip of the front rollers. Since in the 


ring frame "sve have not tlie agitated movement of jarn 
during the spinning operation, as in the mule, this inter- 
ference with the twist would cause a very weak spot to 
develop at the nip of the rollers, and a great souice of 
breakage would result. The diflficulty is overcome by 
inclining the rollers at such an angle that the yarn is in 
contact as little as possible with the bottom front roller, so 
that the twists get right up to the nip of the rollers. The 
four sketches in Fig. 149 will explain the action very 
clearly. In each case A B re2:)resent the front rollers, C is 
the grip, and C D a length of yarn delivered from the 
rollers. If C D be passed out as a flat ribbon, and twisted, 
say one turn, it would (as in 1 and 2, Fig. ] 49) become 
twisted right up to the nip of the rollers, as at C. In the 
ring frame the yarn passes forward in an inclined direction 
(Fig. 149), so that it is in contact with the roller B until 
the point C^^ is passed. Let us notice the eflfect of this by 
taking an extreme case (as in 4, Fig. 149), where the yarn 
goes forward at right angles. In such a case it is in contact 
with the front roller from C to E, and the twists would 
tend to stop short at the point E in the manner shown. 
To obviate this objection the three lines of rollers are 
bodily inclined (as in Figs. 147 and 148), and the effect of 
this is to move the top front roller from A to A (as in 3, 
Fig. 149), so that the nip of the rollers moves from C to C , 
and thus eliminates the objectionable contact surface of the 
roller B. The angle of the rollers varies from 15^^ to 35°, 
according to the cotton being used, but about 25'^ will be 
found most general and serviceable. 

Referring again to Figs. 147 and 148, two systems of 
weighting Avill be found employed— one on the lever system 
and the other by means of dead weights. It will l)e noticed, 
however, that in the first case only the front and middle 


rollers are weighted, the back being self- weighted by the 
large iron top roller. A saddle is put across the first rows, 
and a bridle or link is hooked on to it at a point much 
nearer to the front roller than to the middle one. This 
gives a preponderance of weight on the front roller ; the 
lever and weight arrangement is very similar to that shown 
on the mule x'oller stand. The dead-weighting of the rollers 
is illustrated in Fig. 148, and in this case only the front 
line is Aveighted, the middle and back rows being self- 
weighted. The dead weight hangs from a hook placed 
over the front roller, and it will be noticed that, instead of 
hanging a weight from each hook, the weight is made twice 
the necessary size, and long enough to go across the frame, 
so that its other end hangs from the front roller on the 
opposite side of the machine. A saddle and bridle lever- 
weighting, exactly similar to the one illustrated in the mule 
roller stand, is often adopted on ring frames, so that it is 
unnecessary to repeat the drawing here. There is one 
feature that may be of interest to mention, and that is to 
warn the reader against overlooking the inclination of the 
weight-hook in the systems mentioned. 

It will be sufficient to refer briefly to the matter, without 
entering upon actual calculations. If a saddle A B (1, Fig. 
150) rests horizontally upon two rollers A, and a weight W is 
hung at C, it is an easy problem to find how much pressure 
is put upon the roller at A and B. In the same way, if the 
saddle is inchned as in 2, Fig. 150, the hanging weight W 
can readily be found to give similar pressures upon A and B 
as in No. 1, because in such a case the relative horizontal 
distances of A and B from the direction of the pull of the 
weight W remain unaltered. 

Again, if the saddle is still inclined as at A B (No. 3) 
and the weight is made to pull in the direction of W, which 




is at right angles to A B, M'e should work out the pressures 
on A aud B exactly as in the No. 1 example ; but in the 

Fig. 150 

FiG 151. 

ring frame we generall}^ find a combination of the arrange- 
ments 2 and 3, in which a pull is exercised in the direction 
of W, this pull being produced by a weight hanging 
vertically. No. 4, Fig. 150, illustrates the meaning. AB 

288 COTTOy SPIN KING chap. 

is the saddle, C D a link hooked to the saddle and to the 
weight lever E W. W hangs vertically and produces a 
pressure at D in a vertical direction ; this pull, however, is 
exercised along the link C D, so that in consequence of the 
inclination of C D a portion of the pressure produced by W 
is inoperative on the saildle A B. It is a comparatively 
easy matter to find the effective pressure produced b}^ the 
weight W upon the saddle. Suppose a 2 lb. weight at W 
gives 10 lb. pressure at D : measure off on a vertical line at 
D ten given distances, such as quarter-inches ; now draw a 
line, from the upper end F of the divisioned line, at right 
angles to the line C D, cutting the line C D at G ; if we 
measure off D G and note how many quarter-inches there 
are in it we obtain the number of lbs. pressure along C D. 
Among several stands tested it was found that the pressure 
upon the saddle at C was about ten per cent less than that 
produced by the weight W at D. If this be duly noted 
Avhen calculating the pressure on the rollers, the rest of the 
calculation becomes an easy matter. 

Twisting". — After passing through the front rollers A 
(Fig. 151) the yarn goes forward through the thread guide B 
and is threaded through a bent piece of steel C, in this form, 
O, called a " traveller " ; from here it passes on to a wooden 
tube or bobbin D, fitted upon a spindle E, which is driven 
from the tin roller through the wharve H. The revolution 
of the spindle begins to wind on the yarn ; but since the 
rollers A only deliver a certain length, and the spindle 
revolves at a high rate of speed, the tension produced in 
the yarn acts on the traveller and pulls it round at almost 
the same speed as the spindle itself. Every revolution of 
C puts a twist in the j^arn, and at the same time the bobbin 
Avinds on the amount of yarn given out from the rollers. 
As winding commences, the rail G, which carries the ring F 




and traveller C, is caused to rise and full by lifting 
mechanism, so that the yarn is wound on in layers, and of 
a form similar to the cop of the mule. 

Thread Guide. — The features mentioned in the fore- 
going paragraph can now be dealt with in detail. Commen- 
cing with the thread guide : this is seen to be a curled piece 
of wire A, Fig. 152, screwed into V-shaped pieces of wood B, 
hinged to the thread board proper C. The board C is hinged 
to the roller beam D. As A is directly over the centre of 
the spindle, it is necessary to be able to move it out of the 

Fig, 152. 

Fig. 153. 

way whenever a bobbin is taken off. For this purpose B is 
hinged so that it can be turned over. For doffing purpose 
it is found convenient to be able to turn over the whole of 
the wires and thread board, and arrangements are frequently 
adopted for doing this. An illustration of one method is 
given in Fig. 153. To the under side of C is fixed a bracket 
carrying a pin E, to which are connected links F. The 
other ends of these links are centred on studs G, carried by 
a short lever H, pivoted on the shaft J. This shaft J also 
carries an arm K, to A\hich is attached a handle L, which 
can be locked in position by the slots M fitting over a 
projection N. If now the handle L be drawn forward in 
the direction of the arrow, the thread boards on each side 
VOL. Ill U 


of the frame will be raised bodily out of the path of the 
bobbins as they are being doffed. To serve the same 
purpose, an arrangement is sometimes adopted whereby 
the thread board is moved sideways to the extent of half 
the space of the spindles. 

The Ring. — The ring A in Fig. 154 is made of forged 
steel, carefully turned and afterwards case-hardened ; its 
general form is similar to that indicated in the diagram, 
and shown enlarged in Fig. 155. They are carried by and 
secured to a cast- or wrought-iron plate P, which in the 
modern machine is now flanged singly or doubly to prevent 
deflection. The diameter of the ring is its inside measure- 
ment, the usual dimensions for the diff"erent spaces of 
spindles and counts spun being as follow : — 

For 4's to 20's counts — 2| in. space, If in. dia. of rings 
For20's,, 40's ,, — 2g in. „ Ig in. ,, „ 

For 40's and up-«-ards — 2i in. ,, 1^ in. ,, ,, 

If balloon plates are used the space can be reduced a little. 
The ring is secured to the plate by a set-screw C. Other 
forms of rings are used, such as the double ring shown in 
Fig. 156 ; it is made in this form so that when one flange 
becomes worn the ring can be reversed. The method of 
fastening it to the ring-plate is to spring it into the grip £ 
of a special piece of sheet metal C, which is in its turn set- 
screwed to the plate B. The perfection to which rings are 
now brought renders it extremely doubtful whether there 
is any economy in the adoption of this system ; but some 
people still prefer it. An important American firm have 
introduced slight variations in both the ring and the plate, 
the plate being made out of sheet steel, while the ring is 
modified at the point marked A, with the idea of giving 
better hold to the traveller. Fig. 157 shows the comparison 
between the old and new form. 




From Fig. 154 wc obtain an idea of how the ring-plate 
is carried. At intervals along the frame the plates rest 
upon the upper part D of a shaft E; these shafts are 
termed pokers ; they slide vertically in bushes F fixed in 
the spindle rails G, and their lower ends are arranged to 
be actuated from the building motions in order to give an 
up-and-down motion to the ring-plate. 

Fig. 154. 

Fig. 155 

Pig. 157. 

Building* Motion. —The method of operating the pokers 
of the ring-plate, so as to give a reciprocating motion for 
building the cop, is very simple, and in most modern makes 
of frame there is such a similarity of consti'uction and 
principle that a single example will be sufficient to explain it. 

The principal mechanism employed consists of a cam, 
actuating a lever, on the end of which is attached a chain, 


leading to levers that act on the lower ends of the pokers. 
The speed of the cam is carefully regulated to give a 
motion to the ring-plate suitable to the counts being 
spun, and it will be noted that any alteration in the 
speed of the front roller (consequent on a change in the 
twist or counts) also results in a similar change in the 
speed of the cam. This will be pointed out more fully 
when dealing with the calculations of the machine. 

A view of the motion is presented in Fig. 158, which 
illustrates the essential features. A long lever A is centred 
on a stud at B, fixed in the frame end ; a bowl C is carried 
by a small dish bracket D, bolted to the long levxr, so that 
by the revolution of the cam E the lever A will be given 
a reciprocating motion. At the opposite end to B, the 
lever A carries a bowl G, and round it is wound a chain S, 
the other end of which is attached to a bowl T. As the 
lever A is acted on by the cam E, the pull of the chain S 
Avill turn T through a portion of a circle ; and if to a bowl 
U by the side of T is connected a chain which leads on to 
the levers actuating the pokers, we haA^e the motion of E 
transferred to the ring-plates. From the arrangements so 
far described we obtain the lift of the first layer of yarn, 
as at A in Fig. 159. It now remains to consider the 
further layers B and C. In the first place, the starting- 
point of each new layer is raised up the bobbin by a 
taking-up motion, as follows : the bowl G is carried on a 
short shaft, on which is keyed a worm wheel H, into which 
is geared a worm V. On the end of the shaft carrying V 
is keyed a ratchet wheel M, into the teeth of which is 
engaged a catch carried by a tumbler N. As the lever 
A is depressed, one end of the tumbler Q is brought 
against a stop W, and as its further movement down- 
wards is thus arrested it naturally commences to force 



round the ratchet wheel, and so turn the bowl G, which 
consequently winds on a small length of the chains S 
This action lifts the ring-plates a little higher, so that 
the next laj-er must begin higher up the bobbin than 
the previous one. The same action of the tuml)ler 
continues throughout the building of the bobbin, giving 

Fig. 158. 

whatever lift may be considered suitable. The amount 
of movement given to the tumbler, and consequently to 
the ratchet Avheel, is regulated by adjusting the tumbler 
so that its downward movement is not arrested luitil the 
stop itself is touched, or (as in the illustration) until the 
set-screw at P comes into contact with the projection R 
on the lever A. By careful adjustment of the set-screw 
we can regulate the number of teeth taken by the catch 


on the tumbler ; or a similar effect may be obtained by 
changing the ratchet wheel itself for one of a greater or 
less number of teeth. 

Fig. 160 enables us to follow the building motion to its 
connection with the pokers. The chain N from the bowl 
U passes to a bowl A, carried by a swing lever 'B centred 
on the lever E, whose fulcrum is at F. To the lever E is 
attached a lever Gr H, and as the chain S turns the bowl 
Tj the larger bowl U (acting through the chain N) moves 
the lever E, so that the ends G and F are raised and 
lowered. (We may add at this point that the ring-plates 
are not "lowered" by the direct effect of the lever A in 
Fig. 158; only the "lifting" of the plates is brought 
about by this means : the lowering is brought about 
purely by the weight of the plates and their connections, 
a series of balance weights being so arranged as to permit 
of this occurring.) 

It will be observed that the end G, Fig. 160, has a 
direct lifting effect on the poker ; but since the lifting of 
G means the lowering of the end H, a chain is used to 
transfer this movement to a lifting one ; a chain is attached 
to H, and, passing over the pulley J, is brought down and 
connected to the poker at K, thus producing a lifting action 
on each poker. The movement given to the lever E is 
transferred to each poker throughout the length of the 
frame by means of a rod D, AA'hich is coupled-up to similar 
levers as E at suitable intervals. 

The Traveller and its Action. — It will be found con- 
venient at this point to enter upon a discussion regarding the 
traveller and its action. From a superficial point of view 
the work of the traveller is comparatively easy to understand, 
and its effects in spinning and winding offer no difficulty to 
the observant mind. One reason for this is because the chief 




actions can be closely watched ; consequently experience 
can be obtained readily and quickly under a variety of 
conditions. From this statement we are led to remark that 
the best results are almost invariably dependent upon ex- 
periment, and very little reason is called into play. Never- 
theless there must be a decided advantage in knowing the 
reason for a certain line of action, and Avith this object in 
view a few remarks will be made explanatory of the functions 
performed by the traveller in the ring frame. 

Fig. ICO. 

We shall first point out how the traveller puts the 
twist into the yarn. Fig. IGl is a simple diagram illus- 
trating the point. Here let it be supposed that K 
represents the nip of the front rollers, and that a 
flattened portion of yarn A K is delivered from them ; 
the end A is carried round in a circle ABC, while 
the end E, is held fast by the rollers. As A is carried 
round with tlie small arrows always uppermost, it will be 
found that by the time B is reached, the tape will have 
been twisted half a turn ; and l)y carrying the end to C 
under the same conditions, a full twist will be found to 
exist in the tape — in other words, one revolution of the 

296 COTTON SPINNING . chap. 

end A puts one twist in the length A R. Now, as the 
traveller performs the duty of carrying the end A of the 
yarn A E, round the circle of the ring, it follows that 
the traveller is responsible entirely for the twists put into 
the yarn delivered from the rollers, and that the speed of 
the traveller regulates this factor in spinning. 

Why the traveller revolves? is the next question. In 
the first place, it must be understood that a traveller is a 
kind of guide for the yarn ; if it were a fixed guide, as at 
A, Fig. 163, the revolution of the spindle B would simply 
wind on the yarn as it was delivered by the rollers, and 
such yarn would be untwisted. On the other hand, if the 
traveller guide were attached to the spindle, as at A, Fig. 
162, the revolution of the spindle B would carry A round 
with it, and as a result no winding would take place, but 
every revolution of B would put a twist in the yarn. In 
the two cases given we have examples of all winding and 
all twisting; in spinning, these two operations must be 
performed ; so by making the guide A movable and yet 
unattached to the spindle directly, we get conditions that 
supply us with the requisite characteristics of the ring 
frame. The analysis of this action may prove interesting, 
but we defer it until another feature has been explained. 
A previous paragraph told us that the spindles are revolved 
at a constant speed throughout the building of a bobbin ; 
and we now know that the traveller, in addition to putting 
the twists into the yarn, also winds it on the bobbin. The 
question now arises — How is the conical part of the cop 
built up ? It is unnecessary to remind the reader that an 
enlarged diameter of bobbin or cop necessitates, so far as 
examples in other machines have shown us, a differential 
speed of spindle, in order to Avind the yarn on a varying 
diameter ; but in the case of a ring frame Ave fail to find 




any meciianism that performs this apparently required 
condition of building the bobbin. A few words will make 
this clear. Let us suppose that the front rollers deliver 
528 inches of yarn per minute, and the spindles rotate at 
the rate of 9500 revolutions per minute, also that the 
largest diameter of the cone is 1^ inch, and the smallest 
diameter \ inch. From these conditions we can readily 
find the necessary rate of speed of the traveller to wind 
on the yarn at the two extreme diameters. 

Fig. 161. 

To wind 528 inches on to \\ inch diameter the traveller 
must make 

528 ^>1% X 4 X 7 

1^x3-1416 5x22 

= 134 "4 revolutions 

less than the spindle, so that the speed of the traveller will 
be 9500- 134-4 = 9365-6 revs, per minute Avlien winding 
on the \\ inch diameter. 

To wind 528 inches on to \ iiicli diameter the traveller 
must make 

528 528 x 2 X 7 

4x3-1416 1x22 

= 336 revoliitious 


less than the spindle, so that the speed of the traveller Avill be 
9500 - 336 = 9164 revolutions per minute when winding on 
the I inch diameter. Now, comparing this variation of the 
speed of the traveller as it winds on the extreme diameters 
of the hohbin, Ave find that there is a difference of 9365'6 
- 9164 = 201-6 revolutions only, equal to a little over 2 jier 
cent. It is therefore easily understood why we find no 
apparent arrangement for obtaining differential speed dur- 
ing the formation of the bobbin. This difference, although 
slight, must be allowed for ; and since there is no method 
adopted for varying the speed of the traveller to the extent 
noted, resource is had to the lifting cam in causing the lift 
to vary ever so slightly by a small variation in the shape 
of the cam that actuates the lifting lever. 

In examining the action of the traveller it should first be 
stated that the yarn is wound on by the bolibin, and that 
the traveller simply regulates the amount wound on. A 
bobbin |- inch diameter revolving 9500 revolutions per 
minute would wind on 14,928 inches of yarn per minute, 
Avhile the roller only delivers 528 inches. The consec[uence 
is that no sooner does the boljbin cominence its revolution 
than a tension is set up in the yarn, and as this tension is 
exerted on the traveller, this small piece of bent wire natur- 
ally yields, and is pulled round by the bobbin before the 
tension becomes great enough to break the yarn. Fig. 
164 shows this clearly: the bobbin A pulls the yarn in 
the direction of the arrow; if B was a fixture and the 
rollers did not deliver yarn fast enough, the yarn Avould 
break, but the traveller B, being loose on the ring, gives 
Avay, and the yarn di'ags it round. This is an elementary 
statement of what occurs, but several important factors 
enter into the question, and we shall now consider the 
elements of these. 


A thorough investigation of the thread and traveller 
involves an advanced knowledge of mathematics and some 
knowledge of theoretical mechanics, and since definite 
statements of conclusions cannot be thoroughly relied 
upon without the proof that these sciences enable one to 
bring forward, a certain amount of what follows must be 
taken upon trust, as it would be outside the object of this 
book to use and repeat mathematical formulae which are 
not familiar to the average reader. The elements that 
enter into the question Ave are about to discuss are as 
follows : — 

(1) The counts of the yarn being spun. This has an 
important influence, because yarn has Aveight, and in 
different numbers the weight varies. For instance, one 
hank of 840 yards of No. 8's weighs 2 oz., Avhile 840 yards 
of No. 16's weigh 1 oz. This fact may be expressed by 
saying that the weight of yarn varies inversely as the counts. 

(2) Since the yarn between the traveller and the 
thread guide has weight, and as it revolves at a very high 
rate of speed (being almost equal to that of the spindle), 
it has a tendency to fly outwards from the axis around 
which it revolves. Centrifugal force is the name given to 
this tendency of a revolving body to fly away from a 
centre, and a common example suggests itself in a stone 
attached to a string, which, on being swung round, will, 
when the string is set free, fly off for some distance. The 
result of the centrifugal action of the yarn is such that, 
instead of the yarn passing in a straight line from the 
thread guide to the traveller, it flies outward and forms a 
curve. This always occurs in ring spinning, and the name 
" balloonins; " has been civcn to the buknng thread. 

(3) The amount of ballooning will depend to a certain 
extent on the counts of the yarn being spun. 


(4) The reasoning used in No. 2 is equally applicable 
to the traveller. Its weight and speed cause it to fly 
outwards, but being prevented from doing so by the ring, 
it naturally presses against the ring with a certain degree 
of force, which is dependent upon the centrifugal force 
and the tension in the thread. The pressure thus set up 
produces friction, and this has a retarding influence on 
the traveller's motion round the ring. From this cause 
winding is the result. 

(5) The pull of the thread between the bobbin and 
the traveller depends upon the diameter of the bobbin, 
which varies ; upon the weight of the traveller ; upon the 
diameter of the ring ; and upon the speed of the spindle. 

(6) The pull or tension in the thread will modify the 
friction between the traveller and the ring, and also modify 
the ballooning. 

These are some of the points that will next be in- 
vestigated, and from the analysis we hope, to deduce 
conclusions upon which are based the modern practice 
of ring spinning. 

Ballooning". — Ballooning, as we have already noted, 
is the thread flying away from the centre around which it 
revolves. The degree to which it takes place, of course, 
depends upon the counts of the yarn, in other words its 
weight ; the speed and the weight of the traveller ; atmo- 
spheric resistance to the thread has also some influence. 
Under normal conditions the curve of the yarn in Fig. 165 
represents the shape of the ballooning, and in plan view 
it will be noticed that another curve is produced showing 
that the atmospheric resistance has caused the thread to 
vary from the straight line between the thread guide (over 
the centre of the spindle) and the traveller. If the tension 
in the thread diminishes to any great extent, the ballooning 


will collapse and the yarn become entangled round the 
spindle, simply through the resistance of the atmosphere 
forcing it on one side. Under even ordinary conditions 
slight variations of tension occur, and the result is invariably 
shown in the effect on the shape of the balloon curve where 
it changes from the full line curve to the double one. A 
light traveller is usually the cause of a big balloon curve. 
It is no difficult matter to prove the conditions of balloon- 
ing ; but it will be sufficient to indicate that, granted we 
wish always to have the ballooning the same, Avhatever 
counts are being spun, the mass of the ballooned yarn 
multijDlied by its velocity squared and divided by the 
tension of the thread must equal a constant. . This may 
be represented as 

= constant. 

From this we can say that for the curve of the balloon 
to remain the same, the tension of the yarn must be in- 
versely proportional to the counts being spun, and directly 
proportional to the speed of the traveller ; or, since the 
traveller's speed A^aries so slightly from the sjiindle sjjced, 
the speed of the latter might be taken as the s})eed to 
work from. Explanatory of this statement, Ave might say 
that if the counts are changed and the speed remains 
the same, the tension must be altered by changing the 
traveller ; or if the traveller is not altered, the speed 
must be changed. 

To obtain the direction of the yarn as it enters the 
traveller is not an easy matter, but for all practical pur- 
poses it will be safe to assiune that it is at right angles to 
the portion of thread Avhich leads on to the bobbin. In 
dealing with the tension of the thread Ave come to the 
crux of the Avhole question, and the traveller plays the 


most important part in the matter. We have seen that 
the traveller is a loose piece of metal capable of gliding 
over the surface of the ring. Directly it begins to move 
it is affected by two forces, Fig. 1C6 — a tangential force, 
x, which tends to make it move in a direction tangent to 
the ring ; a centripetal force, y, which tends to draw it 
towards the centre of the ring. With these two forces 
acting upon it, the traveller is compelled to move in a 
circle. Suppose it moves at a rate of 70 feet per minute, 
and the mass of traveller and yarn equals -0000125 lb. ; 
the tangential force x^itix v, and the centripetal force 

m = mass, i7= velocity, of the traveller, and r = the radius 
of the ring = If inch. By working these formulae out we 
shall find that the centripetal force is about 612 times the 
tangential force. 

If we will now understand that the centripetal force is 
really exercised by the ring forcing the traveller back, we 
can easily see that what the ring is doing is being equally 
done by the traveller in trying to get away ; in other words, 
the centrifugal force of the traveller is equal to the centri- 
petal force of the ring, and to this extent the traveller in 
moving round the ring is pressing against it with a pres- 
sure 612 times greater than the force which tends to cause 
the traveller to tiy off at a tangent. Now the tangential 
force is due to the weight of the traveller and its speed, 
and we have seen that this force is a mere fraction 
compared to the centrifugal force, so from this demon- 
stration we can conclude that the momentum of the 
traveller, due to its being carried round the ring by the 
pull of the yarn from the bobbin, is so little that we can 


afford to completely ignore its Aveight proper, except so far 
as it influences its centrifugal force. This must be clearly 
grasped, as on it depends a right conception of the 
traveller's action. The chief lesson to be derived from it 
is that the tangential and centrifugal forces have nothing 
in common; one is a unit, the other is 612. In ring 
spinning there is never any attempt of one equalling the 
other ; they are so widely separated in their effect that the 
tangential force fails to have more than '016 per cent of 
influence in ring spinning, and it would require the 
tangential force of G12 travellers to equal the centrifugal 
force of one traveller ; consequently, outside the mere 
curiosity of knowing and comparing the two foi'ces there 
is absolutely no necessity for knoAving or mentioning the 
tangential force or pull of the yarn in dragging the weight 
of the traveller round the ring. 

If a frame Avas made in which the rings, travellers, and 
speeds Avere proportioned so that the tangential force 

inv— the centi'ifugal force, 

then the radius of the ring Avould have a dimension in feet 
equal to the A'elocity of the traA^eller in feet per second. 
Nothing can be more absolutely absurd than this result. 
The centrifugal force must be and ahvays is, to the extent 
of OA'er 600 per cent, in the ascendant. Moreover, travellers 
are so graded in their Aveight for different counts of 
yarn, that if the pull of the yarn reduces the centrifugal 
force beloAv a certain proportion the yarn AviU break im- 
mediately. There is, hoAvever, another very important 
point to consider, namely, the effect this centrifugal force 
of the traveller has in interfering Avith its movement round 
the ring Av^hen it is acted upon by the thread from the 
bobbin — or in other Avords, the effect the centrifugal force 


of the traveller has on the tension of the thread. This is 
a very important point, so we will consider it carefully. 

It must be fully realised that the centrifugal force of 
the traveller, or its pressure against the ring, due to its 
moinentum, is the chief factor to guide us ; it is equal, in 
the example previously given, to a weight of about 2i oz. 
resting on the ring. (In passing, it may be observed again 
that the weight proper of the traveller is so small, compared 
with this weight due to the centrifugal force, that it may 
with safety be ignored.) Now this 2| oz. is pulled round 
by the thread, and it is the act of pulling it round that 
causes the thread to be in tension. To find the tension, 
we must know the coefficient of friction between the 
traveller and the ring ; the ordinary coefficient of friction 
of polished steel and steel is not applicable to this case. 
Experimenters have found that it has a wide variation, 
and depends on such factors as the speed of spindle, 
diameter of bobbin, diameter of ring, and the dryness or 
otherwise of the surface of the ring. Professor Escher, of 
Zurich, found that if 

Oiled ring. Dry ring, 
the bobbin was g inch diameter the coefficient of 

friction was 0-27 0-465 

if the bobbin was If inch diameter the cocllicient 

of friction was O'lS 0-272 

From his experiments he concluded that the tension 

will vary the least, the greater the coefficient is between 

the ring and the traveller ; in other words, we might say 

that the less difference there is between the extreme 

diameters of the bobbin, the more uniform will the tension 

be in the yarn. Professor Liidicke, of Brunswick, found 

similar variations, and as an example of his I'esearches we 

give the results of experiments. 

Coefficient of frictions for 5000, 6000, 7000, 8000 revolutions 
= 0-4093, 0-3506, 0-2<)99, 0-252. 


He deduces a convenient rule from his inA'estigations, as 
follows : — 

Coefficient of friction = 0*65 - •00005 x revolutions of traveller. 

It will be noticed that the variable coefficient due to 
changes in the diameter of the bobbin is ignored in this 
empirical rule, and it is due to Professor Escher that Ave 
can now with certainty rely upon the fact that such varia- 
tion does exist. 

Mr. Bourcart, in a small pamphlet issued some years 
ago, used for convenience the fraction -1 as the proportion 
of a weight required to move it round a ring. This of 
course is much too little, the average being nearer \ than 
-1-. By taking \ for our basis as the coefficient, we find 
that \ of 2-5 oz. = 'SS oz. will be the tension in the thread 
required to pull the traveller round. It must be remem- 
bered, however, that this tension must be exerted at a 
tangent to the ring as at T, Fig. 166. If the direction 
of the pull T varies as at B its force must be increased, 
because, in addition to overcoming the friction of the 
traveller, we are now trying to pull the traveller away 
from the ring, and therefore some of the centrifugal force 
exists in the thread as tension. This would increase as 
the pull changes, until, when the direction becomes as at 
A, the traveller ceases to press against the ring, the Avhole 
of its centrifugal force is exerted on the thread, and so 
the thread would have a tension equal to this force. At 
the same time that this ol)lique pull of the thread is taking 
place another set of conditions exist also, due to the 
inclination of the pull. AVe know that if the traveller A, 
Fig. 1 66, is pressing against the ring, no amount of pulling 
in the direction of AC will cause it to move ; we also 
know that the least effort to move A Avill be along the 
VOL. Ill X 


tangent AT ; between these two lines a direction can be 
found along which, if a pull is exercised, as at AB, the 
traveller AvilL begin to move. Any pull exercised within 
the angle BAT will move the traveller, and the amount 
of the force ^vill become less as the direction of it approaches 
the line AT. On the other hand, no movement of the 
traveller can possibly take place if the pull is exercised in 
a direction that falls within the angle BAG. From this 
fact we can fix a limit to the diameter of the bare bobbin 
used on the ring frame, provided we know the diameter 
of the ring and the coefficient of friction between the ring 
and the traveller. Assuming the coefficient of friction to 
be \, Fig. 167 Avill give us the size of bare bobbin, while 
for a weft frame Fig. 168 will represent the conditions. 
From these diagrams we learn that the smaller the frac- 
tion representing the coefficient of friction, the smaller the 
bare bobbin can be, while if we wish to reduce the size of 
the bare bobbin, as in the weft frame, the diameter of ring 
must be reduced. 

It has already been shown that there are two forces 
affecting the traveller that have their origin in the mere 
fact of its revolution. It has also been shown that the 
centrifugal force is modified or reduced by the pull of the 
thread from the bol)ljin, such thread taking up the force 
that the traveller loses. A tension therefore exists in the 
thread, and it is the difference between the tension in the 
yarn and the remaining centrifi;gal force of tlie traveller 
that regulates the winding. For winding to take place 
at all, the yarn to the bobbin must always pull against a 
stronger force than that represented hy the tension of 
the yarn. For instance, if the tension equalled the centri- 
fugal force, the traveller would be in a balanced condition, 
and it would cease to press against the ring. In such a 




case it ■would lie carried round as if it were rigidly connected 
to tlie spindle, and while twists would be put in, no winding 
would be taking place. Ordinary observation confirms 
this remark ; for if too light a traveller be used, the balloon 
flies out directly, because of the slackness resulting from 

y r-J Pio. 166. 

i Fig. 167. 

Fig. 16S. 

insufficient 'winding, due to insufficiency of the centrifugal 
force of the traveller. In many cases a traveller is used 
that has only the slightest excess of centrifugal force over 
the tension ; when such conditions exist and circumstances 
happen that momentarily make the tension equal to the 
centrifugal force, unsteady ballooning occurs. On the other 


hand, when too heavy a traveller is used, the centrifugal 
force is so high that the tension in the thread, in its efforts 
to move the traveller, becomes so great that all signs of 
ballooning disappear, and if the tension necessary to do 
this be equal to or greater than the strength of the yarn, 
the end breaks. 

On examining a traveller that has been working on any 
ring frame under normal conditions, it .will be found that 
it is worn on that point which touches the inside of the 
ring ; and travellers will last as long as this point resists 
being worn away by friction. Any traveller that will 
develop sufficient centrifugal force to cause winding will 
show signs of wear only on the point Avhich is in contact 
with the inside of the ring. An interesting experiment 
will make this clear. Conditions illustrated in Fig. 165 
existed on an ordinary frame. Counts 28's were being 
spun from single roving, and a No. 5's traveller Avas being 
used. It was found that all numbers of travellers from 1 
to 10 would cause winding, and, moreover, that each neAv 
traveller after it had run a short time became worn. An 
hour's running in all the tests was sufficient to prove the 
point, but in the heavier travellers a quarter of an hour's 
spinning showed a comparatively large amount of wear due 
to the friction on the ring. If rings are somewhat soft, or 
are irregularly case-hardened, they will also be easilj^ Avorn 
out of shape, and for that reason l)Oth rings and travellers 
are made of the best material to resist frictional wear. 

From the fact of the centrifugal force regulating the 
tension in the yarn, several statements might be formulated; 
for instance, the tension is as the square of the speed of 
the traveller and in direct }>roportion to the Aveight of the 
traveller and the diameter of the ring. The rule for 
centrifuiral force, nameh' — 


mass oF traveller x velocity of trav eller^ 

radius of the ring 

will be an obvious proof of the statement. Briefly, the 
rule means that for a given weight of traveller, if the 
velocity of the spindle be doubled, the tension in the yarn 
will be " increased " four-fold ; or xke versa, if the tension 
is to remain the same, after doubling the speed, the weight 
of the traveller must be " reduced " four-fold. Again, if it 
be wished to double the tension in the yarn without alter- 
ing the speed of spindle, the weight of the traveller must 
also be doubled. Or, if the diameter of the ring lie increased 
and the speed of the spindle and Aveight of traveller be 
kept the same, the tension in the yarn will be increased in 
the same proportion. By similar reasoning \\q may con- 
clude that the weight of the traveller will vary in direct 
proportion to the size of the ring. Supposing the weight 
of the traveller Ije 1, the velocity of the traveller 2, and 
the diameter of the ring 2, then 

•my? mx2^_. 
r ~ 1 

If, now, the ring be doubled in diameter, the formula would 
work out 

vivP- m X 4- «i X 16 o 
r ~ 2 ~ 2 ~ "^ ' 

so that for double the ring we must have double the Aveight 
of traveller. 

Weight of Travellers. — Some interest attaches to 
the method adopted in grading the travellers as to their 
weight and their suitability for spinning certain counts of 
yarn. Generally speaking, a mill keeps the speed of spindle 
and diameter of ring the same for a range of numbers spun ; 
the weight of the traveller is therefore altered to suit the 
changed condition of the counts. The question now is — 



How must the weight of the traveller vary as the counts 
vary ? If the " weight " of yarn is taken as a basis, we 
shall find that 20's yarn is half the weight of lO's, 30's 
yarn is one-third the weight of lO's, 40's is quarter the 
weight of lO's, and so on. The following table will present 
a short series of numbers : — 





in weight. 










lighter than 28's 


„ uV 




)> TIT 

















„ A 



I', ^ 



,, irV 






>) T3- 



'', 5^ 




„ I1V 



„ -V 




„ ix 






„ -h 



I', Jv 






From this table Ave should conclude that the weights of 
the travellers must vary in the same proportion as the 
weight of the yarn varies. On the other hand, we might 
assume that the tension in any yarn is alwaj's a fixed pro- 
portion of the breaking weight of that yarn. For instance, 
suppose the tension in the yarn when spinning 40's is one- 
fifth of the breaking weight of 40's, then there would be 
reason in assuming that the tension in 20's ought to be one- 
fifth the breaking weight of 20's. Taking the breaking 
weight of yarn as published by jNIessrs. G. Draper and Co., of 
Hopedale, Mass., U.S.A., as a basis, we get the following : — 












































From this table it appears that the breaking weight of the 


yarn varies in an " increasing " proportion as the counts 
get lower. The proportionate increase does not vary so 
regularly as the weight of the yarn varies ; but taking into 
account the fact that the above breaking weights are from 
actual tests, the approximate result is near enough to give 
the assumption strength that the variation in the weight 
of the travellers might reasonably follow in a gradually 
increasing proportion as the counts vary. As a matter of 
fact, the United States and the Scotch standards do vary 
in the proportion the above reasoning suggests. Although 
makers of travellers are reluctant to impart information as 
to the weight of travellers, it is an easy matter to weigh 
a number of each, say 100, and form a table of the result. 
Very exact weighings are necessary, but a general idea may 
be obtained from the few following weights : — 








per 100. 


per ioo. 


per 100. 


200 grs. 


120 grs. 


60 grs. 


180 „ 


110 „ 


55 , 


160 „ 


90 „ 


50 ,, 


140 „ 


80 „ 


45 „ 


130 ,, 


70 „ 

The sj'stem adopted in this talkie may be seen at a glance, 
and one can readily understand that as the counts go lower 
the diflerence between the weights of each grade can be made 
greater, just as it can be made less in the higher counts. 

In dealing with the amount of twist put into the yarn 
\sY the traveller, a little repetition maj^ be necessary. 
Twist is the result of the traveller lagging behind the 
spindle. This lagging is due to the friction set up between 
the traveller and the ring as a consequence of tlie centri- 
fugal force of the former. While the centrifugal force is 
practically the same throughout the l;)uilding of the bobbin, 
the friction may vary because the coefficient of friction 
varies, and (as already shown) this has some influence on 


the lagging, because from this alone it is more difficult to 
move the traveller when winding is taking place at the 
nose of the cop than when the yarn is wound on the base. 
In addition to this, tlie tension in the yarn required to 
overcome sufficient of the friction to cause movement is 
greater at the nose than at the base, because on the nose 
the yarn is pulling very obliquely to the movement of the 
traveller. From these two causes, therefore, we conclude 
that the greatest tension in the yarn exists when winding 
on the smallest diameter is taking place, and the least 
tension when winding on the base. Now it must be 
remembered that whatever the tension may be, and no 
matter how it varies under normal conditions, it is always 
simply equal to a part of the pressure of the traveller 
against the ring ; the centrifugal force is always there, 
though reduced by the pull of the yarn. The direct 
consequence of this is, that if no delivery of yarn was made, 
the traveller Avould be carried round at the same speed as 
the spindle, whether the yarn was attached to the smallest 
or the largest diameter ; but the tension is greater in the 
former than in the latter case. On the other hand, if yarn 
be delivered, the tension will be reduced ; consequently the 
pressure of the traveller against the ring will be increased, 
and naturally a lagging liehind the traveller Avill be the 
result until the tension is restored. Continued delivery 
prevents its restoration, so there is always a lagging behind, 
as far as the smallest diameter is concerned. As the largest 
diameter is approached, a uniform continuance of the 
delivery also relieves the tension at this point \ but the 
addition thus made to the centrifugal force by reducing 
the tension is a less proportion to the total force than it 
was at the nose of the cop, and therefore the lagging 
behind is less at the base than at the nose ; in other words, 


the traveller revolves more quickly when binding on a 
large diameter than on a small one, and from this Ave 
deduce the fact that more twists are put in the yarn as the 
winding takes place from the nose to the base of the cop, 
A previous statement, which was used to show that there is 
no necessity for more than the slightest variation in the lift 
in order to compensate for the building of a conical cop, might 
have prepared the ground for this ; but another similar 
example will readily prove it. Let us suppose 530 inches 
of yarn are delivered per minute, and that the spindles 
make 9500 reA'olutions per minute ; the traveller must lag 
behind the bare bol)bin of |-inch diameter 225 revolutions, 
and behind the full bobbin of l|-inch diameter 122 revolu- 
tions. The speed of the traveller at these two points 
would therefore be 9275 and 9378 revolutions — a difference 
of only a fraction above 1 per cent. A difference such as 
this — indeed if it were much higher — would be impossible 
to discover ; so that from a practical point of view the ring 
frame 'is free from any tendency to produce irregularly- 
twisted yarn as far as its own twisting action is concernedc 
Irregularly-twisted yarn will naturally exist in ring yarm 
just as it does in the mule yarn, o^\•ing to the character 
of the cotton and its previous preparation. Differently 
coloured rovings spun on the mule and the ring frame Avill 
show a remarkable similarity in the irregularity of the 
twists, which are thrown out \(tx\ clearly by the contrast of 

In regard to the question as to Avhat number of a traveller 
must be used for any given counts, speed of spindle, size 
of ring, etc., no definite answer can be given beyond one 
that depends upon an accumulated mass of practical ex- 
perience ; and even then, local circumstances introduce an 
element of judgment that compels the use of a traveller 



■\vliich varies from the standard of general experience. 
The following tables Avill convey some idea of the general 
practice, as determined mainly from experience. Thej^ are 
given as a guide only ; each user must judge for himself as 
to how far other conditions necessitate variations from this 


H in. 


1^ in. 






12's 1 



lO's 1 



S's j 



6's . 

( s 


5's 1 






























If in. 



















1 2/0's 



1 4/0's 

1 4/0's 









1 9/0's 

Note. — The above table is given as a guide to select travellers re- 
quii'ed, but will, of course, vary according to circumstances. 

Travellers of from four to six numbers heavier than stated above are 
generally required for spiuuing Egyptian or Sea Islands cotton. 

It -will be noticed from the above table that as the ring in- 
creases in diameter, the Aveight of the traveller decreases. 

The space of spindle and diameter of rings for various 
counts may be gathered from the following lists : — 

With Ballooning Plates. 

4's to 20's 
20's to 40's 
40's upwards 

The traveller tal)lcs of Avell-known makers follow very 
closely on the tables just given. 

Space of 
2f in. 

If in. 4's to 20's 

Space of 
2§ in. 

Dia. of 

21 in. 

15 ill. 20's to 40's 

1h in. 

1* in. 

2| in. 

\\ in. 40's upwards 
l^V to \\ in. ■ 

2} in. 

li in. 


The Spindle. — Another important subject to which 
some space -will now be devoted is that of the spindle. A 
remarkable series of developments have taken place in this 
feature since the traveller system of spinning was introduced. 
The root idea of the chief improvements has been to make 
the spindle work satisfactorily at a high rate of speed with 
a minimum of power to dri\e it. It will be interesting to 
trace out the conditions of Avork to which the ring frame 
spindle has had to be adapted before it reached the 
present type of which Fig. 169 represents an example. 

The older form of spindle used on a throstle or flyer 
spinning-frame Avas essentially an upright spindle, supported 
in tAvo bearings, one at the bottom called a " footstep " 
and the other higher up the spindle, and as near as 
convenient to the bobbin, such support being called the 
"bolster bearing." The position of the AvharA'e or driving 
point on the spindle Avas generally betAveen the two bearings, 
but placed much nearer the top support than the bottom one. 
The mule spindle affords a good example of an arrangement 
of this kind, and in that machine AA'e see the perfection to 
Avhich such a system of driving has been brought, and 
the spindle made suitalile for revolving at very high speeds. 
The conditions of Avorking arc, hoAvever, different in the 
mule and the ring frame. In the former a plain spindle 
is used, upon Avhich the yarn is Avound directly ; it is not 
subject to the same tension on the yarn combined Avith a 
high speed as in the ring frame, nor is it surmounted by 
a heavy bobbin Avhose tendency is to become untrue and 
out of balance. This latter factor becomes of great im- 
portance Avhen the above conditions exist on a spindle 
running at a very high speed, and it AA'as soon recognised 
that some improvement of the well-known type Avas 
absolutel}'^ necessary Avhen greater production Avas required 

3i6 COTTON SPINNING chap, in 

from the machine. The chief objection to be overcome 
was, of course, the excessive vibration set up in the spindle 
when running at a high rate of revolution, which was 
caused by the spindle or bobbin being out of balance. A 
low speed does not disclose this vibration to the same 
degree as a high speed ; a bobbin and spindle slightly out 
of truth might not, at 4500 revolutions per minute, prove 
very inconvenient, or even show itself clearly ; but if the 
speed were douhled to 9000 revolutions per minute, the 
irregularity of balance would have a four-fold tendency to 
make itself felt, and it shows itself by setting up vibrations 
in the spindle. 

The first attempts at a remedy were made by Rabbeth, 
in a spindle which dispensed with the lower footstep bear- 
ing as such. He extended the bolster bearing in the form 
of a long tube firmly fixed to the rail, and at the top and 
bottom of this were bearings for the spindle. Above all 
was placed the bobbin. This arrangement, it will be seen, 
was only a slight move in the right direction, but it 
contained two features that formed the basis of the spindle 
of to-day, namely, a self-contained spindle, and greatly im- 
proved means of lubrication. The next move was made by 
Sawyer, Fig. 171, who recognised that a greater steadiness 
of running would be assured if the upper bearing could be 
raised. He effected this by extending the bush of the 
bolster bearing, and over this he placed the bobbin. By 
this means the bolster bearing was placed Avithin the 
bobbin, and to this extent a decided advantage accrued. 
He was compelled, however, to still use the lower separate 
footstep bearing, and to place his wharve between it and 
the upper support. The early Rabbeth and the Sawj^er 
spindles both contained the elements of a successful 
spindle to fulfil the requirements of that time, so that 


eventual!}^ both Avere combined — and in the result the well- 
known "Rabbeth" spindle was evolved. The improvements 
all contributed to greatly increased speeds and steadiness 
in running, and gave the ring frame the opportunity to 
compete to some extent with other spinning machines. 

On reference to Fig. 170 the characteristic features of 
the Rabbeth spindle will be observed. The steel spindle 
A is carried by a base or bolster B, which is firmly fastened 
to the spindle rail C by means of nuts D. The upper 
portion of the bolster extends to F, where it is bored out 
to ht the spindle ; the lower portion at M is also bored out 
to fit A. Between these two points the bolster is barrelled 
or recessed out, so that the open space thus formed serves 
as a receptacle for oil. The upper bearing at F is usually 
fitted with a thin bush of some anti-friction metal. A 
portion of a coarse spiral is left l^etween the ends of the 
sheet of metal which forms the bush, and by this the oil, if 
it reaches this part of the spindle, is distributed over the 
surface of F. 

Immediately above the bearing at F, a sleeve G is tightly 
fitted over the spindle, and is continued in a downward 
direction to form the wharve H. The position of the 
wharve is designed so that the top and bottom bearings 
each bears its share of the strain. It will be noticed subse- 
quently, that in this respect the pull of the spindle band in 
recent spindles is exercised almost entirely upon the upper 
bearing. As a rule a brass cup is fitted over the outside of 
the wharve sleeve at J, which serves for the reception of 
the lower end of the bobbin K. A loose fit of the bobbin 
is generally allowed at this point, for a purpose to be 
explained shortly ; but the up})er end of the bobbin is 
made to fit the spindle tightly. 

The Booth-Sawyer spindle had a very extensive vogue, 


aud {oiukI great favour with the users of ring frames; 


HA B BETH BOOTft-SAjWYEF^. D0650^l-^\^f\£,^^. 

Fio. 170. Fig. 171. Fn;. 172. 

but its introduction inaugurated a series of improve- 


ments that culminated in the spindle known as the 
Rabbeth, just described. Its chief characteristics may 
be summarised as follows: it is self-contained; it has a 
reservoir or bath of oil in which the spindle works ; its 
upper bearing is within the bobbin ; and the pull of the 
band takes place somewhere between the upper and lower 
bearings. In practically all modern spindles the two last- 
named features of the Rabbeth are entirely absent ; but 
the self-contained character and the oiling arrangement is 
such a basic feature that some authorities classify most 
recent spindles as being Rabbeth in principle. 

The Rabbeth spindle underwent a variety of alterations 
and improvements, chiefly with the idea of improving the 
lubrication. The reservoir of oil Avas very effective in 
lubricating the footstep bearing, but the upper bearing had 
to trust to capillary attraction for its oil. The metal bush 
had no power to take the oil in an upward direction, because 
the surface of the oil Avas kept too low for that purpose ; 
and, moreover, if through carelessness too much oil Avas 
placed in the spindle, it quickly rose to the top, ran over 
the bolster, and Avas dissipated by the Avharve, or it ran 
down and spread over the rail. 

A decided improA^ement Avas effected when an attempt 
was made to cause some kind of circulation of the oil 
Avithin the spindle, Avhereby the upper bearing might be 
kept constantly oiled. Another fault shoAved itself in the 
fact that Avhen the spindles required re-oiling the old dirty 
oil had to be pumped out, and indifference in doing this 
shoAved itself in accumulations of dirt and gummed oil, 
which largely increased the poAver necessary to driA^e 
the machine. This defect Avas also remedied ; and in 
Fig. 172 a Dobson-Marsh spindle is slioAvn in section, 
Avhich presents an extensiA'ely used method of OA'ercoming 


the objections mentioned. The lower end of the bolster is 
pierced at a point near to the bottom end of the spindle. 
Over the bolster is placed, by a detachable bayonet or 
other means, a cup, carrying a large quantity of oil ; on 
the spindle is placed a spiral of wire, which revolves and 
forces the oil upwards to the top bearing, and so keeps it 
constantly lubricated ; any oil carried over the top runs 
down, and, by means of the passage shown, flows back 
into the cup. When new oil was required after working 
a month or two, the cup was simply detached without 
stopping the spindles ; and the old oil was poured out, fresh 
oil supplied, and the cup hooked on or screwed into place 
again. Another improvement, copied from a still earlier 
spindle, was incorporated, namely, the set screw P on which 
the end of the spindle blade rested ; as the spindle wore at 
this point, the screw could be moved upwards to compensate 
for the wear ; the lock-nut Q effectively kept it in position. 
We have shown how a spindle revolving at a high speed 
is subject to strains through being out of balance, and how 
these strains are augmented through the uncertainty of the 
bobbin and cop maintaining themselves true. The demand 
for increased speeds brought about, as indicated, better 
spindles, in the Sawyer, the Rabbeth, the Dobson-Marsh, 
and other improved forms. These spindles for a long time 
served their purpose, but new conditions of speed, etc., began 
to show weaknesses in their construction, and inventors 
were thus led on to make further improvements. Eventu- 
ally a spindle was devised which solved the problem so far 
as principle was concerned, and the " gravity " or " top " 
spindle was introduced. Such spindles now assume in- 
numerable forms of construction in details ; but, the pur- 
pose being the same, a few words of explanation will not 
be out of place. 

VOL. Ill Y 


If a spindle is perfectly balanced and revolves at a high 
speed, well supported in bearings, its axis Avill permanently 
occupy one position, and any definite strain put upon it 
will always act in one direction ; and there Avould be little 
if any vibration set up in such a spindle. If, on the other 
hand, a spindle is out of balance, i.e. heavier on one side 
of its centre than on the other, there would be two sets of 
forces at work, and the strain would not be acting equally 
around the axis of the spindle. Such a condition as this, 
in which two opjDosing forces are at work, interferes with 
the uniform motion of the spindle round its axis, and there 
is a constant struggle going on to revolve round an axis 
which would be common to the two unequal sides. Vibra- 
tion is set up as a consequence of rigid bearings, and, 
together with considerable wear and tear, the spinning 
operation is performed under disadvantageous circumstances. 
The object of the improvement Avas to arrange the spindle 
so that it could revolve round its own axis and also round 
the real centre of its movement. A spinning-top is some- 
times used to illustrate this j)rinciple, and from the example 
the new spindle was formerly referred to as a "top" spindle. 
"Gravity" spindle Avas a name also applied. Either name, 
however, is only partially correct, and while a top may 
enable some idea to be obtained of the principle involved, 
the word "gravity" is entirely a misnomer. If a perfectly 
balanced top be set spinning at a high speed, it will revolve 
with its axis vertical ; but if it be moved out of that position, 
it will continue to revolve round its own axis and at the 
same time revolve in a circular path forming the outline of 
a cone whose apex is the point where it touches the ground. 
Now this is not what occurs with a spindle : an unbalanced 
top would represent the action much better. In such a 
case the top could not revolve with its axis vertical ; it 


would certainly revolve round its axis, but at the same 
time its free position Avould permit its axis to become 
inclined and a bodily movement to take place round the 
axis of a cone whose apex Avould be some distance in the 
ground. It will be readily seen that although this example 
is a better illustration than a balanced top, still it does not 
approach the actual conditions of a spindle; in the top, 
the axis is not supported in any way, while in a spindle we 
are compelled to have such support. 

If a bar of iron be taken, and a pound weight placed on 
one end and an ounce on the other end, the middle point 
of the bar is clearly not the centre round Avhich the bar 
could be set revolving ; neither would the bar revolve at a 
high speed if the point were taken, on which the bar and 
weights would be balanced. We require to know such a 
point in the bar that the energy developed by each weight 
would equal one another. This point is known as the 
"centre of gyration," and in the case of an unbalanced 
spindle it is this centre or axis round Avhich the spindle 
must be capable of revolving at the same time as it revolves 
round its own axis. For most practical purposes, pulleys, 
etc., are balanced on the principle of making their "centre 
of gravity " correspond with the centre of their rotation, 
but in important organs great care is taken to balance them 
so that the " centre of gyration " corresponds to the axis 
of the shaft on which they revolve. In a spindle this cannot 
be done, so the balancing effect is obtained by leaving suffi- 
cient room in their bearings to permit them to occupy and 
revolve round their natural centres. Spindles constructed 
on the above-mentioned principle are termed " elastic " or 
" flexible " spindles. 

Fig. 173 represents five of the prevailing types of flexible 
spindles used by machine-makers in this country. They 

324 COTTON SPINNING chap, hi 

are all self-contained, on the principle of the Rabbeth. 
Referring to A, it will be observed that the spindle D is 
fitted with a bolster or pillar E ; this bolster is itself 
fitted within the pillar F, which is firmly bolted to the 
spindle rail. The inside bolster E is only permitted to fit 
the fixed pillar F at its npper end, and even then the fit is 
very easy ; the length of the bearing is shown as from B 
to C. The lower end, it will be noticed, is quite free from 
contact with the lower part of F, so that if the upper part 
of the spindle and the bobbin fitted over it become un- 
balanced, it may be deflected from an exact vertical line, 
and revolve, as already pointed out, round its centre of 
gyration. Every possible precaution is taken to prevent 
or eliminate the tendency in the spindle and bobbin to 
become unbalanced, and consequently never more than the 
slightest tendency makes itself evident, even with the 
highest speeds. Allowance need only be made, therefore, 
to a limited extent, and this accounts for the small clear- 
ances shown in the drawings. 

It is quite evident that the pull of the band must be 
exercised on some part of the upper bearing, and this is 
marked clearly in the sketches, B and C representing the 
top and bottom of the bearing, while A is the centre of the 
Avharve. In this connection it may also be remarked that 
it is advisable to arrange the pull so that it may be some- 
where near the centre of the bearing. 

A variety of means are adopted for keeping the inside 
bolster in position and preventing it from revolving. Pins 
and slots, screw caps, and spring catches are the usual 
methods ; in the example B, a square end is provided on 
the bottom of the bolster, which rests within a correspond- 
ing but slightly larger hole in the outside pillar. 

There is one great inconvenience associated with the 


oW j,. 


e,< --*. 




spindles just illustrated : that is the method of renewing 
the oil. The machine must be stopped, all bands be taken 
off, the spindles taken out, and a pumj) used to extract the 
dirty oil ; then the whole operation must be performed 
again in the reverse order. All this means a waste of time, 
as well as offering an opportunity for carelessness to show 
itself. This objection was overcome by making the lower 
end of the outside pillar open, and attaching thereto a cup 
containing oil. 

The Dobson-Marsh spindle illustrated this method, but 
as modified, in its present construction, the oil cup is 
attached to the pillar by means of a spring ring fitting 
within a recess. Tlie oil is ciixulated the full length of the 
spindle blade by means of a spiral cut on the lower end of 
the spindle, and it returns to the cup by grooves cut in the 
inside of the outside pillar ; these can easily be traced in the 
drawing. Re-oiling can be performed without stopping the 
machine or touching the bands, all that is necessary being 
to take the cups off, empty the old oil out, put in the new, 
and replace the cui)s. The work is done quickly, and saves 
a deal of time. An arrangement of this kind has such 
decided advantages that the method has been applied, with 
slight variations in detail, to several makes of spindles, one 
of which is represented in Fig. 174. The cup C in this 
case is fastened on the inside of the pillar by means of a 
quick-threaded screw, as marked at B. The lower end C is 
made square, so that a key can be used to fix it firmly in 
position and make it perfectly tight. Similar inside cups 
have been used for some time by attaching them to the 
pillar in various ways, such as by means of clip rings, hook- 
and-slot, and bayonet joints. 

On reference to Figs. 175, 176, another improvement will 
be noticed. In order to prevent the spindle from lifting up 




from its position, a catch is so arranged over the wharve 
that such an action is impossible ; hefore the spindle can be 
moved, the catch nuist be moved on one side, and it is more- 
over necessary that the catch be so constructed that on the 
replacement of the spindle it will permit the spindle to fall 
with certainty to its j^lace. Fig. 169 and also Figs. 177 and 
178 illustrate similar catches. The lid shown in the oil cup 

Fig. 1V5 

Fig. 1V6. 

in Fig. 176 is to permit oil to be supplied to the spindle to 
compensate for any evaporation that may take place. 

It will readily be understood that the best-made spindles 
will wear very little indeed during the course of years, and 
if the oil 1)6 of the best quality (as it ought to be) there 
will be no gumming nor will it become very dirty, and 
evaporation therefore will reduce it in quantity only. The 
passage to the oil cup is thus a decided convenience to those 
who use the best lubricating oil obtainable. 

328 COTTON SPINNING chap, hi 

Fig. 175 illustrates a method of obtaining the same effect 
on a self-contained spindle. An extension is made to the 
outside pillai' at A, and it is bored out for the passage of 
the oil. A cap B prevents the entrance of fly, dirt, etc. 
An improvement on this is shown in Fig. 176, wherein the 
cap is replaced by a hinged lid D, so arranged that it serves 
also the purpose of a catch to prevent the lifting of the 
spindle. An indispensable adjunct to the ring rail is to be 
noticed in what is called the " traveller clearer." While the 
machine is working, a good deal of dirt and fine fibre is 
always flying about, which settles upon the frame, and 
some of it naturally rests upon the ring itself. In course of 
time accumulations would occur, which would interfere 
with the action of the traveller by clogging its action. A 
small projection is therefore placed on the ring plate by 
screwing or other convenient means, in such a position that 
the traveller in its revolution just misses it. In consequence 
of this any fibres adhering to the traveller are caught up 
by the projection, and the traveller passes on cleared of its 
encumbrances. A catch of this kind will be noticed in 
Fig. 1 .56 at D. 

The Ballooning Effect has already been explained ; it 
only remains to point out that under some conditions it has 
a tendency to caiise the space between the spindles to be 
greater than is desirable in order to avoid the adjacent 
threads coming into contact with one another. To prevent 
this, balloon plates, or, as some species of them are called, 
" separators," are adopted. Such appliances are only really 
necessary during the formation of the lower part of the 
bobbin, so that when this stage is passed they are gener- 
ally arranged to be automatically moved out of the way. 
Different ways for doing this have been introduced, but in 
essentials they consist of the introduction of projecting 




Fig. 17 


Fio. 178. 



pieces of metal between the spindles, so that the bulging 
thread is kept from coming into contact with neighbouring 
threads. A vertical plate of sheet metal is a favourite 
method ; its only disadvantage is that the open back and 
front causes the yarn to bulge out at these points, so that as 
it passes the sides it strikes against the plates, and of course 
such an action is a disadvantage. This can be neutralised 
to some extent by using plates that are closed in at the 
back, so that as far as practicable the yarn is always kept 
moving in a circle. Complete rings have even been adopted 
for anti-ballooning purposes, but they introduce difficulties 
in doffing and piecing, so have therefore not been very 

It has already been intimated that the yarn spun on a 
ring frame must be wound upon a bobbin whose diameter is 
relatively large. This has always been a great drawback 
and has prevented the machine competing with the weft 
yarn made on the mule. Weft yarn is of course made on 
the ring frame, and in large quantities ; but it is not done 
under the best conditions, and the size of the bobbins made 
is a great disadvantage. Many attempts, therefore, have 
been made to spin on the bare spindle for both twist and 
weft purposes, by getting rid of the bobbin, so as to obtain 
the greatest amount of yarn in the smallest space, as in the 
mule cop. A surprising amount of ingenuity and exertion 
has been put forth to solve the problem of spinning on the 
bare spindle, and, so far as making a cop is concerned, it 
may be added that the problem has been successfully solved 
in several ways. Commercial success, however, is another 
matter, and in this direction nothing but failure has 
rewarded the effi)rts that have been made. To be successful, 
a machine for spinning on the bare spindle must have a 
production equal to the present ring frame ; it must make a 


compact cop equal to the mule, which must possess the 
quality of " readying " to an equal degree ; the stopping and 
starting of the machine must present no difficulties, and the 
travellers or guides must be as permanent as possible ; the 
strains in the yarn must l)e uniform, especially in soft- 
twisted yarn (as in weft) ; and elasticity must be a quality 
possessed by the yarn produced. 

The chief difficulty, that of causing the traveller to 
ajjproach the spindle as the smaller diameters are being 
wound, has not proved insurmountable ; but most of the 
other points mentioned above have hitherto not been 
attained, and until these have been overcome, spinning on 
the bare spindle can only be said to be in its experimental 
stage. Bearing this in mind, it would be inadvisable to 
present the reader with the numerous methods that have 
been tried unless some claim to success could be made out 
for them. Every machine-maker is, more or less, devoting 
consideralile time and money to bring it to a successful 
issue, and no doubt something will be done soon to make 
the ring frame a satisfactory cop spinner. 

These notes upon the ring frame would be incomplete 
Avithout some reference to a comparison between the mule 
and the ring frame. There is such a divergence of opinion 
upon the matter that the subject can only be briefly touched 
upon, and it is done without the slightest idea of treating it 
controversially. Thus fa.r practical experience points to a 
limit beyond which ring 3\arn cannot excel yarn made on 
the mule. Between 60's and 70's might be taken as this 
limit — though the writer can point to a firm where as high 
as lOO's is spun equal to anything in strength and quality 
that the mule produces. AVeft yarns are not so easily 
produced on the ring system as on the mule, but improve- 
ments in the machine and conditions of workin<r enable 


weft up to 40's to be very successfully spun ; beyond this, 
practical difficulties arise, M'hich prevent commercial success 
being attained. It is frequently stated that the ring frame 
requires more power to drive than the mule ; but a 
considerable practical experience, extending over both 
machines, suggests no great disadvantage in this respect in 
the ring frame (especially with our high-class modern 
flexible spindles). The ring frame is much the cheaper 
yarn spinner, in some cases exceeding the mule by as much 
as 40 per cent. In medium and finer counts no advantage 
in this respect can be claimed, but below, say, 40's there is 
a decided gain. 

A comparison of the strength of yarn produced on the 
two systems gives to the ring yarn the claim to superiority, 
in some cases rising as high as 40 per cent. In regard to 
regularity and elasticity, there is room for doulit as to 
which claims the advantage, especially the latter cjuality ; 
but the mule appears to attain a higher degree than the 
ring frame in the elasticity of the yarn made, and is 
more uniform in that respect. The ring frame has the 
advantage over the nude in the space occupied, an economy 
of 50 per cent being claimed for it. Tlie cheapness of the 
labour and the ease with Avhich the ring frame can be 
learned and attended to are economical advantages to be 

The horse-power required to dri\ e a ring frame is a very 
variable quantity, depending upon a number of conditions 
that can scarcely be found alike in two machines. The 
spindles, of course, absorl) the greatest proportion of the 
power, and differences in spindles account for much of the 
variation in power betAveen one machine and another. 
Anything affecting the spindle, such as its speed, the pull 
of the band, the size of the bobbin, the length of the 


traverse, the lubrication and condition of the oil used — all 
are factors in the problem of the power. The construction 
of a machine, its erection, gearing, rollers, and weighting, 
are conditions which more or less must be considered in 
relation to the power. Therefore it is no easy matter to 
set up a standard by which the power of a ring frame can 
be gauged. From practical dynamometrical tests made 
by the Avriter, extending over scores of machines of all 
the best makers, the number of spindles per indicated 
horse-power has ranged from 60 to as high as 110. 
The following table, taken from one of Draper's pub- 
lications, presents in a convenient form the results of 
tests which were conducted to indicate the power absorbed 
by diflPerent parts of the machine. Speed of spindles, 
9300. Diameter of ring, \\ inch. Xos. spun, 43's. 

Power taken hy rollers, traverse motion, and gearing . 11 per cent. 

Power taken by weight of bobbin and yarn . . .11 ,, 

Power taken by the pull of the traveller . . .17 ,, 

Power taken by cylinder and bare spindles . . . (31 ,, 

100 „ 

The pull of the band has a very deciding effect in the power, 
and 20 per cent may easily be added by banding too 
tightly. To those who use a band tension scale a piUl of 
2 lb. is strongly recommended by the best authorities, and 
in no case ought it to exceed 3 lb. 

Another feature which is not sufficiently observed by 
many users of ring frames is the lubrication of the spindles. 
It has become an axiom among those who have devoted 
attention to the subject that only the very best oil it is 
possible to get ought to be used on a ring frame. The 
price of such an oil is an apparent objection, but when it is 
considered that large percentages of power are saved — 
which means a great saving in the coal bill, a longer life to 


the machine, and far better work — it is not difficult to see 

that lubrication is too important a matter to be ignored or 
lightly dealt with. 


Calculations.-speed of spindles =^^'^-°^^'^'^i^^^. 

^ Dia. of g. 

r> 1 ■• fi' i. 11 Revs. ofCxCxA 

J\evolutioiis of front roller = - 


E X D X P 

Turns of spiiuUe for one of front roller = -. 7= — k- 

'■ A X Cx Q 

T, . , . , ExDxP 

1 wist per inch = t — j:^ — pr — ^r^^ — _ . .-^ • 
'- AxCx QxNx3-1416 


Twist wheel = 

Twist per inch x C x Q x N x 3-1416 
E x D X P 

Constant nuinber for twist = 

CxQxNx 3-1416 

~ . , , , Constant number 

1 wist wheel = —rir-—, ■ — , — • 

iwist per inch 

m . ^ . , Constant number 
i wist per inch = — ^p; — ^- — ; — - — • 
^ iwist wheel 

rp . , , , Present twist wheel x ^Present counts 
1 wist wheel = — . 

vEequired counts 

rp . , , , /Present twist wheePx Present counts 

iwist wheel = A/ ^s ^ — s 

> Kequired counts 

^ .^ HxGxN 
Draft = 1 

Draft wheel = 


Draft X F x L 


Constant number for draft = „ _ 
If X L 

» -n ff _ Constant number 

Draft wheel 

T% p^ 1 1 Constant nuinber 

Draft wheel = =; — 


■D , 1 , , , _ Present shaper wheel x \/Kequirod counts 

\ Present counts 

Ratchet wheel = / Present sliaper wheel'^ x Required counts 
▼ Present counts 

336 COTTON SPINNING chap, hi 

Fig. 179 will enable all the above calculations to be 
easily followed. 

It may be observed that the above rule for twist 
is only approximate ; but it differs from exactness by 
such a small fractional amount that it may be used in 
all circumstances. 



It will only be necessary to briefly describe the uses to 
which cotton yarns are put after coming from the spinning 
machines, and these may be summed up in two chief 
purposes, namely, weaving and doubling. The former 
term includes all forms of cloth manufacture into which 
cotton enters, whether as the only material used or simply 
as the warp of the cloth, some other substance, such as 
silk, wool, linen, etc., being used as the weft. The latter 
term, doubling, signifies the twisting together of two or 
more strands of single yarn in a simple or compound form, 
for the purpose of making sewing thread, lace, embroidery, 
knitting, crochet, hosiery, netting and other fancy yarns. 
For the purpose of weaving, the Aveft yarn is generally 
used in the form in which it comes from the spinning 
machine, while, on the other hand, the warp threads 
require to be put through several important operations to 
fit them for their purpose. The only one of these opera- 
tions which concerns our subject is also employed in the 
doubling series of operations, so that by confining our 
attention to this branch a repetition will be unnecessary. 
The cops from the mule or bobbins from the ring frame 
VOL. Ill Z 

338 COTTON SPINNIXG ' chap. 

are brought to this machine to have their yarn wound on 
to large double-flanged bobbins, similar to those shown in 
the sketch Fig. 180 at A; they are generally termed 
warpers' bobljins, because yarn is practically always wound 
into this form before being transferred to the warping 
beam. The reason for this form of bobbin is that when 
the yarn has to be again unwound and placed on a beam 
or otherwise, it will come from the bobbin at a fairly 
uniform tension, because of the parallel layers ; from a 
cop or bobbin this would be impossible on account of the 
constantly changing diameter of the conical ends. In 
addition to being used for Aveaving purposes, the bobbins 
are largely used for the doubling frame for the same reason 
which prompts their use in warping. This point will be 
treated more fully in a subsequent paragraph. 

A general idea of the machine can be obtained on 
reference to Fig 180, which represents the upper portion 
of one side of the frame. Two rows of spindles (B) are 
carried in bearings D and E on each side, and driven from 
a tin drum in the centre of the machine through the wharves 
C. The upper ends of the spindles carry the bobbins A. 
The cops or ring frame bobbins are supported by small 
brackets at N, and from here the yarn is led forward over a 
flannel-covered board L, where it is likely to be cleared of 
any loose fibres or dirt adhering to it. On its way to the 
bobbin it passes through the bristles of a brush K, where a 
further cleaning Avill naturally occur, and then on through 
a guide H and on to the bobbin. The guide H is generally 
of a special form for the purpose of clearing the yarn of 
any persistent motes, slubs, etc., which stick to it, or to 
prevent the passage of badly-formed piecings or knots due 
to carelessness in the previous processes. This clearer is, 
therefore, an absolutely essential feature, and it has afforded 



innumerable opportunities for the displa}- of ingenuity in 
so arranging its parts as to obtain the greatest usefulness 

Fig. ISO. 

out of it ; it also adapts itself well to different counts and 
classes of yarn. This remark applies to other machines 


through which yarn passes and where such guides are 
employed ; in some cases a kind of winding machine is 
used simply for the purpose of clearing the yarn, so that 
the clearer is the chief feature. The guide H is carried on 
the top of a lifting rod F, which may be operated in its 
upward and downward movement by a cam or the well- 
known mangle-wheel arrangement. The resulting bobbin 
can be made barrelled, as in the sketch, or perfectly parallel. 
Bobbin boxes are placed under the row of cops M, and in 
the middle of the machine is a receptacle for the full bobbins 
after doffing. In most machines an arrangement can be 
applied, in the form of an endless band or apron running 
down the middle of the frame, which carries the full bobbins 
to the end of the machine and deposits them in a large box 
or skip. Instead of winding from cops, arrangements can 
be substituted in order to Avind from hanks. In Fig. 187 
will be found a fuller and better idea of the machine just 

Quick - traverse Winding Frame. — When yarn 
is to be used for doubling purposes, that is, a combina- 
tion of two or more ends into one, the yarns from two 
of these bobbins are passed together through the rollers 
of a doubling frame and then twisted together as one 
strand. This system, however, is not now so general as 
formerly, though it is still practised, and in the case of 
high numbers winding is dispensed with and the cops placed 
directly in the creel of the doubler. The usual method of 
doubling the ends together for the doubling machine is to 
take the cops or bobbins to Avhat is called a quick-traverse 
winding frame, where a bobbin is made upon an ordinary 
paper tube without flanges. A section of such a machine 
is given in Fig. 181. It is a double-sided frame, and, like 
the one just described, it will wind bobbins of any diameter 



at the same time. Two shafts run the full length of the 
machine, and on them are threaded and keyed drums M, 
whose lengths are suitable for the lift of a bobbin required. 
Eesting 011 the drums are wooden rollers N suj^ported in 

the slotted bracket carried by the beams, and on the rollers 
N rest the steel spindles upon Avhich the bobbin is to be 
wound. The rollers are driven by friction, so that the 
bobbin is built up through frictional driving, and as each 
bobbin is diiven from a distinct Avooden roller the diameter 
of the bobbin as it is formed does not interfere with its 


correct shape. The bobbin boxes or creels contain a scries 
of supports C for tlie cops or bobbins, and the yarn is led 
through a guide D and over a covered clearer E ; from here 
it passes tlirough tlie drop needles F and over a stationary 
guide Y and on to the bobbin through a spoon Q. This 
spoon can be placed very near to the nip of the bobbin and 
wooden rollers X, and it rec^ves a very quick to-and-fro 
movement from a cam or other suitable mechanical motion. 
The C|uick traverse of Q causes the yarn to be wound on 
the bobbin P in a series of very coarse spii'als, and these 
are such that the quick return motion enables a bobbin to 
be formed without the usual wooden ends, thus saving both 
weight and space. 

In this machine any number of ends from one to six can 
be doubled together and wound on one bobbin, and each 
end will have exactly the same tension, the arrangement 
of the parts being such that the tension can be readily 
adjusted. The importance of maintaining the exact numbei 
of ends continuously is obvious, especially if for doubling 
purposes, when only two ends are to be twisted ; it is 
therefore natural that an automatic stop motion is neces- 
sary, and for this purpose the needles at F are employed. 
Each end passes through a separate needle ; when the end 
breaks, the needle falls and comes into contact with a 
revolvins: grooved roller G. Since the needles are carried 
by a swivelled catch-lever H, the consequence of F falling 
into the path of G is to move the lever on one side, and 
in doing so it releases the catch end of H from a projection 
J which it has held in position. "When J is held bv H a 
slide K is drawn back so that its upper surface at L is out 
of contact with the wooden rollers N, and at the same time 
the spring is put into tension. "When the catch H is 
released, the spring forces K forward and the part L 


impinges against the Avooden roller X and lifts it clear of 
the drum ^I, thus stopping the hobbiii. This stoppage 
enables the broken end to be at once pieced. The im- 
probability of more than one end breaking at once or more 
than one cop becoming enqjt}' at the same time, enables the 
pieciug-up to be done Avithout bunch knots occurring ; it 
prevents waste and overrunning, and in keeping the yarn 
always at the same tension obviates the great fault of 
corkscrewing when the bobbins are taken to the doubler. 
The frames carrying the bobbins M are centred on the 
rod E, and wire hooks supporting weights AY are added to 
give grip and steadiness. A further improvement in this 
respect is obtained by the use of the spring T, especially 
when the bobbin is small. The hooks U and V are often 
formed with bent portions, so that the bobbin itself can be 
lifted out of contact with N and kept so by resting the bent 
portion of the wire upon a convenient projection, as at X. 

Another well-known and successful quick - traverse 
winding frame is illustrated in Fig. 182, where half the 
machine is shown in section. A shaft L drives a series of 
drums A, whose outer surface is in the form of a thin shell 
having a fine double helical slit piercing it all round ; this 
slit corresponds to the usual cam which gives the quick 
traverse to the other makes of winding frames. The spool 
or bobbin D carried by a lever B which is centred at F 
presses against the under side of A, and the yarn is led 
from the cop or bobbin through the usual detector wire 
G, over K, the roller H, and through the slit in the drum 
A on to the spool. The revolution of A will* naturally 
cause the yarn which passes through it to travel backAvards 
and forAvards the full length of the cam slit in its surface, 
and as the spool is driven at the same time through friction 
by being in contact Avith A, the yarn is Avound on D in a 


series of coarse spirals in such a manner that the ends of 
the spool are built up solidly and squarely, and are capable 
of being handled and transported safely and economically. 
A steel blade at E serves to keep the yarn always at the 
bite of the spool and drum, and the lever B is so fulcrumed 
that, as the spool enlarges, the point of contact with the 
drum remains always the same. The stop motion is suffi- 
ciently interesting to merit description. On the breakage 
of an end, the needle G falls into the path of the revolving 
wiper L, and is with its carrier at once moved backwards ; 
this pulls M with it and lifts up the catch at "m"; "m," 
it will be observed, has resting against it a finger " a," which 
is centred on the supporting lever A^, which carries the 
drum A. Directly "m" is moved out of the way of "a" 
the drum A falls back against a fixed brake N and is at 
once stopped in its revolution ; the broken end can at once 
be pieced, and if it is necessary to draw the spool away 
from the drum a catch fulcrumed on B^ enables this to be 
done by allowing the other end of B^ to come against a 
stop on the beam ; on pressing down at B^ the spool will 
fall immediately against the drum ready for work. The 
pressure of the spool against A is carefully regulated by 
the weight W, and is practically the same from the empty 
to the full bobbin. It may be observed that this machine 
is essentially a quick-traverse winding frame, and it could 
not be used for slow winding without great loss of produc- 
tion ; alteration in the speed of the traverse cannot be made 
irrespective of the speed of winding, and as a consequence 
the character of the winding practically remains the same. 
Where the traverse is independent of the revolution of the 
spool, half the pitch of the spiral can be obtained without 
altering the production, and a slow-traverse bobbin can 
even be made with the same efiect. 



Fig. 183 gives another full sectional view of a frame 
made by a Avell-knoAvn maker of this kind of machine, and 
the following; remarks Avill enable the working to be under- 

Fio. 182. 

stood. The cop or bobbin boxes A are carried from the 
spring pieces and run the full length of the machine ; in 
the boxes are mounted, as shown, the cops or bobljins B. 
The yarn from these is passed through wire guides C and 

346 COTTON SPINNING chap, iv 

on over an adjusting dr2,g board D, the regulation of which 
is effected through the screws E. After leaving the drag 
board, the yarn is passed through detector needles F carried 
by a short swivel cradle, which rests upon one end of the 
swivel frame G. The yarn is now taken in an upward 
direction over the wooden guide rollers R, and from here 
it passes direct to the flanged wooden winding bobbin M. 

The bobbin M is supported upon the upper end of a 
lever J fulcrumed on the swivel frame G ; the lower end of 
J is connected by a cord or chain to a weight K, which 
keeps M pressed against a central revolving drum X, so 
that the bobbin has a constant pressure and an unvarying 
surface speed. The threads Q are fixed to the traverse rod 
0, which is actuated by a slow-motioned traverse, either of 
the cam or mangle-wheel type. The automatic stoppage 
of the machine when an end breaks is brought about through 
the medium of the needles F ; these are kept out of contact 
with the revolving ratchet shaft I when the yarn is passing 
forward, but on an end breaking, the needle falls and is at 
once moved aside by one of the wings of I. This at once 
frees the end of G wliich carries the needle box, and it 
rises, thus lowering the other end, which carries the lever 
J ; in this way the bobbin is lowered from its normal posi- 
tion, and in doing so it is kept out of contact with the drum 
N by a brake L, whose knife edge projects almost to the 
nip of the drum and bobbin. This naturally stops the 
revolution of M; but to make this stoppage absolutely 
certain, a projection on J comes against an extension of 
the brake L, and the weight of J forces the upper end of 
L against the bobbin J\I, and so stops its further motion 
immediatel}'. For the purpose of piecing-up, the lever J 
and bobbin can be pulled forward and automatically hooked 
in a convenient position ; when all is in order, the setting-on 


handle H is depressed, and this action at once puts the 
whole arrangement in correct position for continuing the 
winding. One side of the machine is shown as when 
winding is being performed, while the right-hand side 
represents the altered positions taken up by the various 
levers when an end breaks. Xo difficulty whatever is 
experienced in finding and piecing a broken end, and any 
number of ends from one to eight can be wound together. 

The machine just illustrated is used, with very little 
alteration of construction, for quick -traverse winding ; a 
change in the method of giving the traverse being all 
that is necessary. 

The motion in this machine for making the " cheeses," 
as the quick-traverse bobbins are termed, is an interest- 
ing example of mechanism ; instead of the usual cam, 
an attempt has been made to use a crank, Readers 
will know that although a crank gives a to-and-fro move- 
ment, such a motion is not a uniform one, the middle of 
the throw giving a C|uick movement Avhile the ends pro- 
duce a slow one. In the example, this irregular action 
of the crank is overcome, or rather modified, by the 
introduction of a special cam groove, so formed that the 
crank pin travelling in this groove is made to give to a 
traverse rod an absolutely uniform motion. 

Fig. 184 gives a section through the traverse motion. 
Upon the drum shaft A are keyed a series of drums B ; 
the traverse rod J is connected to crossheads I, Avhich slide 
to and fro in guides. A crank pin H on the top side of 
the crank-plate slide F fits in a groove of the crosshead, the 
slide F sliding within a groove on the crank plate E. It 
Avill be seen that if F and E are fastened together and 
revolved, the pin H would impart to the traverse rod J a 
simple crank movement which is incapable of building a 



straight bobbin. However, F is made so that it can slide 
along guides carried by E, and by means of a bowl G which 
it carries, and which fits in a cam groove cut in a revolving 
plate D, the pin H is made to move in a special manner 
towards and away from the centre of the crank plate E, 
so that the irregularities of the crank motion are neutralised 
and a uniform traverse is the result. From the sketch an 
idea of how the arrangement is driven may be obtained. 
The shaft K is driven from the gearing C ; on K is keyed 

Fig. 184. 

a bevel L, Avhich gears into two bevels M and N. The 
lower bevel wheel M is keyed to the crank shaft P, and 
so drives the crank plate E ; the upper bevel wheel N is 
keyed to the boss of the cam D, and from it the cam 
receives its motion. 

Two partial views of a quick-traverse frame are given in 
Figs. 185 and 186. This frame is fitted up, as usual, with 
automatic stop motion, but the drawings will serve their 
chief purpose by showing the most common form of cam. 
The revolution of the cam A moves the pin B to and fro, 



and with it the traverse rod C, which runs the full length 
of the frame and carries the gnides D. 

A drawing has heen prepared, Fig. 187, to illustrate a 
more modern example of the ordinary Avinding frame for 
making warpers' bobbins than that given in Fig. 180. This 
machine is sometimes called a clearing^ frame. The ring 

Fig. 1S5. 

Fig. 186. 

frame bobbins G are mounted, as shown, upon a box 
arrangement F, which serves also the purpose of a receptacle 
for bobbins. The yarn is led upwards throTigh giddes over 
a drag board E and through another guide H ; from H the 
yarn is taken over a rod J, and when used as a clearing 
frame it passes through an adjustable yarn clearer C. This 
clearer is really a series of narrow slits, so arranged that 


the slits can readily be made wider or narrower according 

Fig. 1S7. 

to the counts or condition of the yarn, the object being to 
prevent the passage of knots or other imperfections in the 

352 COTTON SPINNING chap, iv 

way of " slubs," etc. The j^arn now passes on to the bobbins 
N, which are mounted upon Rabbeth spindles P carried 
from the beam R; the spindles are driven from the tin 
roller Q, the diameter of which is usually about 5 inches. 

To economise time when doffing, the full bobbins are 
taken from the spindles and put upon a travelling apron 
B, which carries them to the end of the machine and 
deposits them into a basket or box. The building motion 
takes effect through the Avheel M gearing into a vertical 
rack L, on the upper end of which is mounted the clearer 
arrangement. An ingenious contrivance is introduced at 
D : it is very desirable not to have the j^arn always passing 
through the clearer C at the same spot, so the rod J is 
carried by a lever D, whose centre is at S ; a projection on 
D rests upon an incline K, so that as L is raised and 
lowered the lever D will receive an oscillating movement, 
and the rod J will guide the yarn through the clearer C in 
a constantly varying position. 



The bobbins from the winding frame are now taken and 
placed in the creel of a doubling frame, or, as it is 
sometimes and more correctly called, a twisting frame or 
twister, Avhere the ends are twisted together into one 

The doubler bears a general resemblance to the ring 
spinning frame, and its twisting operation is exactly the 
same, but an observer would notice in most machinery three 
points of difference, namely — the bobbins in the creel are 
different, as already explained ; the feed rollers are not 
as in the ring frame — we find no drawing rollers at all in 
the doubler, for the reason that no drawing effect can be 
obtained from threads already so well tAvisted ; instead of 
three lines of rollers we only find a single line. The other 
difference noted would be the character of the bobbins built 
on the machine ; as a rule these are built up as parallel 
layers on double-ended bobbins and not in conical layers, 
as in the ring frame, though it may be observed that ring 
spinning frames are sometimes made Avith parallel lifts and 
doubling frames are made A\'ith conical lifts. 

Fig. 188 will convey some general idea of the doubler. 
As in the ring frame, there are tAvo roAvs of spindles HH, 

VOL. Ill 2 A 

354 COTTON SPINNING chap, v 

one on each side of the machine ; these are driven from the 

tin drums G. The driving of the single hne of rollers J 
starts at C on the tin roller shaft and through the gearing 


shown, to F. Ou account of tlie wide range of twists j^iit 
into the various doublings, two change places are introduced 
at A and B to enable this to be readily obtained. The 
lifting cam, it will be noticed, is an equal heart-shaped one, 
giving the up and down motion of the rail a uniform 

Fig. 19a 

traverse ; it is driven generally from the feed roller J in 
the manner illustrated. 

Creels. — Illustrations are given in Figs. 189, 190, and 
191 to show the method of doubling from flanged bobbins, 
from quick-traverse winding drum bobbins, and from cops j 
in all cases the system of doubling is called wet doubling, 



from the fact that the 3'arn before reaching the rollers 
•passes through a trough of water. Dry douljling is practi- 
cally the same system with the exception of the water 
trough, and iron rollers are used instead of bi'ass covered 
ones ; dry doubled yarn is used chiefly for warp threads in 

Fig. 191. 

weaving, and also in many cases simply for the selvedges in 
cloth where the general warp is single yarn. 

English and Scotch Systems. — The two illustrations 

in Figs. 192 and 193 will serve to illustrate the details of 
the trough in wet doubling. Two systems are employed, 
namely, the Scotch and English. Fig. 192 shoAvs the 
English system ; here the yarn coming from the bobbins 
passes down into the trough of water and under a glass rod 

Fig. 192. 




carried by a series of short aims 
centred as shown. On emerging 
from the water the yarn passes 
through a guide wire and on to the 
rollers ; these rollers are covered 
with brass so that the w^eft yarn has 
no corroding eflect on them. Apart 
from the fact that the rollers deliver 
the yarn at a fixed rate, they serve 
the purpose of pressing the surplus 
water from the yarn, the top roller 
being heavy and driven entirely by 
friction from the bottom roller. For 
cleaning purposes, etc., the glass rod 
can be lifted out of the trough by 
the handle shown in the illustra- 
tion, and the trough itself can be 
emptied by a tap placed at one end 
of the frame. The effect of water 
on the yarn is to give it a solidity, 
and strength is added from the fact 
that all loose fibres are smoothed 
down and readily incorporated in 
the double thread when twisted. 
The Scotch system is given in Fig. 
193 ; here the rollers themselves are 
located within the trough, so that 
the yarn passes direct from the creel 
to the underside of the bottom roller. 
The top roller is out of contact with 
the water in the trough, but it is 
constantly wet through its contact 
with the bottom roller, so that in 


Fig. 194. 



this case the yarn passes to the spindles in a fur vetter 
condition than in the English system. • 

Spindle. — The spindles of a doubling frame are practi- 
cally the same as those used on a ring spinning frame. 
Fig. 194: will enable a comparison to be made, and it "n^ll 
be noted that stronger and heavier spindles are required 
for doubling than for spinning. The bobbin is double- 

ended, and the yarn is "wound on in layers the full length 
of the lift, no crossing effect being given. 

Knee-brakes. — An appliance called a knee -brake is 
frequentl}" applied to a doubling spindle. This is done so 
that the attendant can, by pressing the brake with the knee, 
stop the spindle while the end is being pieced. Fig. 194 
shows the brake, and attached to it is a projecting wire 
used as a catch to prevent the spindle lifting out of its 


bearing. The brake ilhistrated encircles the base part of 
the spindle carrier, and on the nnderside of the front is cut 
an inclined surface resting upon the edge of the pillar base. 
A piece of leather is shown in the hatched part of the 
drawing, and it is so arranged that if the knee presses 
against the front of the l)rake, the inclined surface permits 
the brake to slide up until the leather comes into contact 
with the lower part of the spindle wharve ; the friction 
resulting from this pressure stops the spindle, and jiiecing 
can be performed quickly and conveniently. In withdraw- 
ing the knee the brake by its own gravity falls out of con- 
tact with the wharve and takes up its normal position. 

There are innumerable types of these knee-brakes, but 
one more example Avill be sufficient ; this is illustrated in 
Fig. 195, where a single casting, as in the first case, rests on 
the spindle rail and is prevented from any side play by small 
projections fitting round the pillar base. Pressuie applied 
by the knee to the front part at A causes the upper 
leather-padded end C to press against the spindle and so 
stop it. 

Stop Motions. — Two examples are given of stop 
motions ; the first is a very cheap and simple, but at the 
same time a very useful, type. Its object is to prevent the 
deliveiy of yarn when an end breaks ; it will readily be 
understood that twisted yarn, if it continues to be delivered, 
will either be wound on the roller in the foim of a lap and 
will require some trouble to cut away, with the possibility 
of damaging the roller, or it will be delivered and fly 
around in the path of other ends and cause several break- 
downs in addition to its own. Apart from the trouble 
involved, jauch waste is caused by it, and the necessity of 
a stop motion will be obvious. In the motion illustrated 
(Fig. 196) a metal holder D is loosely centred on the pivot 



of the top roller A; attached to D is a wire F curved 
around so that its lower end G, which is curled, allows the 
yarn to pass through on its way to the spindle. At E is 
fixed a strip of leather, so that when the end breaks the 
wire F instantly falls, and the leather E passes into the nip 
of the two rollers A and B, and is carried a little forward ; 
while the leather strip is in this position, it is impossible 


Fig. 19G. 

for yarn to be further delivered, so both waste, laps, and 
additional breakages are prevented. The second example 
is a Avell-known and well-tried arrangement, and, as will be 
seen in Fig. 197, it stops both the rollers and the spindles. 
The stoppage of these two organs was formerly considered 
an altogether unnecessary act for twofold yarn and even for 
threefold ; this opinion is still held by many, but experience 
is proving that substantial advantages accrue even when 



twofold yarn is being Jonbled, and the success of the 
method shown in the diagram is a strong proof of the 
efficacy of such an arrangement. The drawing is ahnost 
self-explanatory, and can be easily followed in its action. 

Pig. lOr. 

the connection between the stopj^ing of the rollers and the 
spindles being clearly depicted. 

Twisting". — The twisting action on the doubler requires 
no special description, because it depends upon the same 
principles as in the ring frame. In the dry doubler, ring 
and ti'aveller are precisely as in the ring frame. In wet 


douliling, however, a slight variation is introduced in the 
form of the traveller, with the object of obtaining a larger 
frictional surface between it and the ring. This will be 
observed in Fig. 198, where A is the ring and B the 
traveller. To prevent wear, doubler rings are oiled or 
greased ; several very ingenious methods have been tried 
to do this without hand labour, but so far the greasing in 
most places is purely a manual task. A point to notice in 
connection with a doubler ring a,n'd traveller is that any 
wear that takes place will be at the part A on the under 
side of the ring. It is a large surface, and wear must be 
considerable to become inconvenient. From this fact we 
find little effort made to use double rings in doublers, 
though they are by no means unknown. A list of suitable 
travellers for 2, 3, and 4-fold yarns for wet and dry 
doubling would be too long to insert here, but any machine 
firm of repute would, no doubt, willingly supply the reader 
with the information. 

An interesting subject is presented to us when we come 
to consider the twisting together of two or more yarns to 
form a cord. As this is the chief purpose of the machine, 
a brief mention of the operation will lie made. If a 
twisted thread of single yarn be taken from a cop that has 
been spun "twist way," the spirals will have the same 
direction as the threads of a left-handed screw (see Fig. 
199. If this thread is noAv allowed to sag until it becomes 
doubled, it will be observed that the parts of the doubled 
end immediately begin to twist themselves together, as 
shown in the sketch, Fig. 200. The peculiarity in this 
action lies in the fact that the twist of this doubled thread 
is opposite to the twist of the single thread which composes 
it. Moreover, the single thread, if left alone, would begin 
to untwist itself, while the double part has no such tendency. 



on the contrary, its tendency is to become more tightly 
twisted and to remain so. The action just described is a 
perfectly natural one, and has been performed entirely 
by the forces within the yarn itself. Its explanation is 
simple enough Avhen Ave remember that a thread twisted 
"twist way" will untwist itself by turning "weft way." 
If two threads twisted "twist way" are put together, each 

Fig. 198. 

Fig. 199. Fig. COO. 

Fio. 201. 

tries to untwist " weft way," and consequently they wind 
round each other and form a combined thread which is 
twisted "weft way," as the sketch illustrates. 

This example supplies us Avith the foundation upon 
■which to base our doubling operations. In doubling two 
or three-fold, the tAvist must be opposite to that of the 
single yarn. In four and six-fold tAvo operations are 
necessary. First, tAvo ends are made into one; these are 
then re- wound on a Avinding machine and tAVO or three of 

366 COTTON SPINNING *chap. v 

them are twisted together again on the second doubler into 
a single cord : such a cord would be denoted as a four or 
six-fold thread. Fig. 201 Avill illustrate how these ends 
must be twisted in order to obtain a thread that will be 
well twisted and will remain so without a tendency to 
untwist. Note, that since a six-fold is to be the object of 
our doubling, the three two-folds of which it is composed 
are not to be twisted as if they were to stand as simple 
two-fold yarn, but rather they must be so twisted as to 
make a permanent six-fold yarn. In the first place, single 
yarns A and B, both with the same twist, are taken and 
twisted together into one thread, as at C. The two are 
twisted in the same direction as the twist in the single 
yarn ; this twist is not a permanent one, for, as already 
mentioned, this two-fold thread would at once untwist 
itself if allowed to be free. We have, therefore, in the 
doubled yarn at C two forces at work — the twist in the 
single yarns A and B tending to untwist, and the same 
twist in the double yarn tending to untwist in the same 
direction. If three of these threads C are put together, 
they would among themselves twist into a cord in the 
opposite direction to the twist of the component threads ; 
they must, therefore, be twisted together in this direction 
if we desire a permanently twisted six-fold yarn. This is 
shown at D in the sketch, which, it will be noticed, has its 
twist opposite to that of the threads C and A and B. A 
thread made by this method is said to be " cable laid," to 
distinguish it from some threads, or rather cords, which 
are made by simply twisting six or more ends at one 
operation into a single cord. Commercially, doubled yarns 
are denoted by first stating the number of folds and then 
following with the counts of the single yarn of which it 
is composed — for instance, six-fold 120's means that the 

Fig. 202 

Fig. 203. 


combined thread has six strands of 120's in it, and is 
represented frequently as 6/120's. 

Rope Driving". — A feature now frequently seen on 
doubling frames and sometimes on ring spinning frames 
is an arrangement for enabling the speed of the tin roller 
to be altered. In each case, however, the necessity 
for its use only arises in case of a wide range of twists 
being Avorked in the same mill on the same machines. 
Several systems are adopted, an example of which is given 
in Fig. 202. The driving pulleys, instead of being placed 
on one of the tin drum shafts, are carried by suitable 
supports above the frame end, as at A and B. These two 
pulleys, it will be observed, are placed between two strong 
supports, but the short driving shaft is extended, and on 
it is fixed a rim band pulley C. An endless band passes 
around this pulley, and is threaded round pulleys E and F 
keyed to each of the tin drum shafts, and on over a guide 
pulley D. The passage of the band over the pulleys can 
be easily understood from the drawings. Figs. 202 and 
203 ; a point to observe is that no crossing of bands is 
necessary. The pulley D is carried by a slide bracket 
containing a screw, through which the band can be 
tightened and kept at a suitable tension. The change 
pulley C is exactly similar to those used on the mule, and 
the method of fastening it on the shaft is also the same, so 
that it is quite a simple matter to change it. From some 
points of view there is also an advantage in this arrange- 
ment of driving, inasmuch as both tin drums are driven 

Calculations. — Fig. 188 will enable the following 
calculations to be readil}- understood :— 

_ T - . T, Revs, of G X dia. of G 

Siieed ot siiinules = :pr^ rr^^ 

' ^ Dia. of H 

VOL. Ill 2 B 


Revs. ofCxCxBxA 

Eevs. of front roller 


Turns of spindle for one of front roller = ■: — tt — -yz — =i" 
^ Ax B xCxH 

„ . , . , FxExDxG 

Twist per ,^^\,^ -^--^-^^-,^-^-^^^^. 

rr • f 1 1 A FxExDxG 

Iwist wheel A= ^ . , — ^ — p^ — 5 — ^ — s-tttt:* 
Twist xBxCxHxJx 3-1416 

iwist wheel B^ 

A X Twist X (J x H x J X 3-1416 

In regard to the twist required in doubled yarns, the 

common practice is to find the counts of the combined 

thread and put in the twists that would be required for a 

single yarn of the same counts ; a slight allowance is to 

be made for the difference produced as a consequence of 

contraction due to twisting. 

Two threads of 40's doubled = 20's. 
Three ,, 60's ,, =20's. 
Four ,, SO's ,, =20's. 

The following rules, relating to doubled yarns when the 
numbers of the single yarns are not the same, are only 
occasionally required : — 

For two yarns of different counts, say A and 

-7 — ^ = doubled thread. 
A + B 

When we know the counts of the doubled thread, and 
it is desired to know the counts of another thread to use 
with a known yarn to produce it, all that is necessary is to 
proceed as follows : — 

-. — ,, = doubled thread. 
A + B 

We know the counts of A, and we want to knoAv the 

counts of B, therefore — 

A X doubled thread _ 
A - doubled thread 


If A = 40's and the doubled thread = 20's, Avhat must B 
equal 1 

40's - 20's 20 ~ ®' 

For three yarns of vmequal counts, doubled together, 
the usual rule is as follows : — 

Weigh a lea of each of the counts ; 
Divide the weight of each lea by 1000 ; 
Add the quotients together, and 
Divide the sum of them into 1000. 

Or, Divide the highest count by itself ; then 

Divide each of the other counts into the highest ; 
Add the three quotients together, and 
Divide their sum into the highest count. 

Or, Let three mixed numbers be «, i, and c, then the 
mixed counts equal 

ax Jx c 

-7— — ,~ = counts. 

ab +ac+oc 

Ex. — When ISTos. 20, 40, and 60 are doubled, what is the 
resultant numbers ? 


20 X 40 + 20 X 60 + 40 X 60 

= 10 "9 counts. 

Note. — 1 per cent is generally allowed when doubling different 



Reeling. — When yarn has to be dyed, bleached, and 
probably shipped abroad, it is usually made into as loose 
a condition as possible. Within recent years means 
have been adopted — by using perforated skewers and 
forcing dye or some bleaching liquid through the cop from 
its interior as it is immersed in vats or kiers — to bleach 
and dye the yarn while in the cop and bobbin form. In 
spite of the progress that has been made in this direction, 
however, the process of reeling is still largely used for 
unwinding the cops and bobbins, and, while the yarn is in 
the unwound state, bleaching and dyeing it by the usual 

In some processes the cop and bobbin are not most 
convenient for the yarn. This is the case especially in the 
hosiery trade : as the yarn used for this purpose is generally 
bought, it is first reeled and then made up into bundles for 
safety, convenience, and economy in transit. On its arrival 
at the mill it is re-wound into the special form of bobbin or 
cop that is most convenient for the purpose. 

Reeling is performed on a machine called a reel, the 
chief element of which consists of a "swift." This swift is 
built up of six longitudinal staves of wood, arranged in the 




form of a hexagon, each stave being suj^ported throughout 
its length by anus at an equal distance from the centre of 
a shaft. Fig. 204 illustrates this feature. Upon the shaft 

Fig. 204. 

A are mounteil arms B, C, and D ; one of the arms B is 
fixed to the shaft (or, instead of a shaft, a tin cylinder is 
used, as being lighter and larger in diameter). The staves 
E, Ej, and Eg are in pairs, and are carried by the arms B, 


C, and D. Two of the arms (C and D) are loose on the 
shaft, and can be moved round so that the staves can be 
brought together, as shown by the dotted lines at Eg ; a 
strap F is generally used to keep the staves in their correct 
positions when working. All that is necessarj-, when doff- 
ing, is to unhook one end of the strap at G, and the whole 
swift can be closed up as at Eg. 

The circumference of the swift is H yards = 54 inches ; 
the diameter, therefore, across two opposite staves will be 
approximately — 

f = 18 inches. 


Reels assume a variety of forms, according to the work 
they have to do, but, speaking generally, they may be 
divided into cop and bobbin reels — or, in other words, the 
differences between most reels consist of variations in the 
creel and method of driving. There are single reels for 
cops or bobbins arranged to be driven by hand or power, 
these having only one length of swift ; a double form for 
the same purjxse with a swift on each side of the machine ; 
a special form for reeling from the cheeses made on a quick- 
traverse gassing frame, and other types or variations. It 
will be sufficient to illustrate the essential features of one 
type only, this being the bobbin reel shown in Fig. 204. 
The bobbins G are placed upon spindles H, and the yarn 
led from them through guides or clearing plates at J on to 
the swift ; the swift is driven from the end of the machine 
(see Fig. 205) by the pulley K, L being the loose pulley. 
There are several methods of winding the yarn on to the 
swift so that it will l)e in a loose condition when taken off 
again. It must be remembered, hoAvever, that while the 
loose condition is essential, the winding must be so per- 
formed that no entanglement will occur in the subsequent 



processes ; therefore some definite method must be adopted 
tliat will facilitate re-winding. This, in conjunction with 
the fact that it is frequently necessary to have very exact 
lengths wound on the swift, introduces traverse and 
measuring mechanism, which form a chief feature of most 

Fig. 205. 

"When the yarn is wound on the swift in a definite 
length, the basis of the English system of numbering yarns 
— namely, 8-40 yards — is taken as the standard, and as 840 
j^ards of any yarn is termed a " hank " Ave have a very 
simple method of obtaining the necessary revolutirins for 
the swift to wind any given weight of yarn. For instance. 


the swift in one revolution winds on IJ- yards; it will 
therefore require 

-_ = 560 revs. 

to wind on one hank. Knowing the counts of yarn being 
wound — say 20's — there would he twenty of these hanks to 
1 lb. The usual practice is to make the lianks into bundles 
of 5 or 10 lb., so that it is an easy matter to say how many of 
the hanks from a reel are required for a bundle. The hank 
(or 840 3'ards) Avound on a swift may be put on by laj'ing 
one-seventh of a hank — that is, 120 yards — on one part of 
the swift, and then moving the guide wire a little to one 
side and laying another 120 yards — or "lea," as it is termed 
— by the side of the first ones. If this is done seven times 
we get the hank divided into seven parts (or leas), as shown 
at 1 in Fig. 205. Another method is to arrange the 
traverse motion to guide the j-arn over a sjiace on the swift 
equal to that occupied by the seven-lea motion. By doing 
this cpiickly the 3'arn is wound on in a " crossed " condition, 
as shown at 2 in Fig. 205 ; when definite lengths are 
wound on in this way the yarn is said to be "skeined," 
though it is as well to bear in mind that the system can be 
and is adopted for winding on any length other than a 
hank. The " Grant " system of winding is a modification 
of the crossed form, the yarn being crossed in a special 
manner, which enables a thread to be j)assed through the 
openings between the crossings, so as to tie the whole 
together and j^revent entanglements. The seven-lea ari'ange- 
ment also permits the seven diA^isions to be easily tied 
to<iether. Fig. 205 will enable the traverse mechanism to 
be understood. On the shaft A is keyed a worm M, which 
gears into a worm Avheel X. This wheel carries a finger 
which in the course of its revolution comes against a tooth 


in a vertical rack P, and lifts the rack bodil}'. The upper 
end of the rack P carries a stepped bracket Q, against 
which a j^in li, carried by the finger T, is kept pressed by 
a spring. The finger T being fixed to the guide rail S, the 
guide wires J Avill cause the yarn to be Avound on the swift 
at the points they happen to be opposite. Now suppose 
the rack is lowered and the jDin E. is on the step at 3, then 
when the finger on N lifts the rack P the pin will shoot 
into the next step at 4, and the guide plate S will move 
the depth of the notch away on one side of its previous 
position. This continues until the seven leas have been 
laid on the swift ; at the termination of the last length an 
automatic arrangement moves the strap on to the loose 
pulley, and a brake comes into contact with the pulley V 
and stops the frame instantly. A catch which drops into 
teeth on the side of the rack P prevents the rack falling 
back after it is lifted by the finger on N. 

The same illustration shows the method of obtaining the 
crossing motion. In this case the traA-erse is continuously 
moving to and fro, so that the parts M, N, P, and E, are 
not used; instead, Me have a wheel driving another 
wheel U, on the upper end of which is a crank W carrying 
a pin X, which also engages with a finger Y fixed to the 
traverse rod S. The revolution of AY will move X back- 
wards and forwards, and according to the " throw " of the 
crank will give a traA'erse of the yarn on the swift. The 
pin at X is the same as that used at R, so that if a machine 
is fitted up with both motions, the change from the "lea" 
motion to the " crossing " motion can be effected by simply 
changing the pin li to X. 

The doffing of the yarn from the swift is done as 
follows : — First, the staves are drawn together, as shown at 
Eg, Fig 204 As there are generally 40 hanks wound on 


one swift, all this requires to be moved to one end of the 
machine and then taken off. This is not a simple matter, 
for the swift and the yarn it contains are heavy, and the 
support of the swift if fixed prevents taking off the yarn ; 
the end support is therefore modified, and in its 
place the old style of bearing was that shoAvn 
at Fig. 206. Better systems are now adopted, 
one of which is shown in Fig. 204 ; this is 
called the "bridge" doffing motion. The shaft 

I'M. 206. . . ° . ^ 

A is carried by the framing, but surrounding it 
is a loose guide 6, which is loosely hinged on a pin 7 ; when 
the staves are brought together the yarn is drawn over the 
end and dropped into the opening 8. This done, the guide 6 
is turned over from the pin 7 as a centre, and then occupies 
the position shown in the dotted lines, leaving a clear 
opening for the doffBd yarn to be lifted out. 

Coleby's Reel. — A machine sometimes used instead of 
the long 40-hank machines illustrated in Fig. 204 is given in 
Fig. 207. It consists of four independent " swifts " (B, C, 
D, and E), all of which are driven from the middle of 
the machine by one belt. The swifts are of the ordinary 
construction, and each one is generally made to reel ten 
hanks, so that a complete machine will work forty hanks 
at a time. Every thread has an automatic stop motion, 
but only that section or swift is stopped which is at the 
time reeling the thread ; this is a decided advantage, as it 
saves considerable time and material ; only ten hanks are 
stopped for piecing instead of forty hanks, as in the 
previously described machines. 

The off end of each swift, as at J and H, is supported 
by a bracket K, which is provided with a stud J and a 
sliding bush H. AVhen doffing, the bush H is moved along 
the shaft of the swift and a clear opening is left for the 



reeled hanks to he slipped off the closed staves of the swift. 
The drawing shows the machine ready to he doffed, and 
one swift is represented as at the point ready for the hanks 
to he taken off. Any system of reeling can he applied to 
the machine and any required length wound on the swifts 
with automatic stop motion to each section when the 
correct quantity is put on. Cops, hohhins, cheeses, etc., 

Fio. 207. 

can he reeled, and one, two, or more ends reeled together. 
As in the previous machine, a change from "lea" to "cross" 
or "Grant" reeling can he effected hy simply changing a 
stud. The gain in production, through saving in time alone, 
over the ordinary reel is as high as 25 per cent, and one 
child with any smartness can take care of one side of a 
machine — that is, twenty cops — and produce as much as an 
average reeler from an ordinary 40-hank reel. 

Fig. 208 shows a sectional view of the machine, and it 



will be ohserved that the strap A drives both lines of 
swifts. This is not done directly, but through the arrange- 

FiG. 20S. 

ment shown in Fig. 209, which represents a j)lan A'iew of 
the gearing. The strap drives the pulley B which runs 
loose on the shaft C ; each side of the pulley has one part 
of a clutch wheel Avhich gears with the other half connected 
to the shaft C by a float key. The boss of this half of the 

Fig. 200. 

clutch wheel has a grooA^e for the fork D, so that by 
moving aside the handle E any one or all the swifts F 
can be stopped through disconnecting the clutch wheels. 
The shaft C drives the swift through the wheels G and H. 
The regulating and measuring mechanism is dri^-en from 
the worm J ; the finger on the wheel K lifts the vertical 


rack L, and so permits the traverse rod to escape a tooth 
after one-seventh of a lea has been Avound on at one spot. 
The stoppage of the swift after it has revolved a sufficient 
number of times is brought about by a stop M on the 
vertical rack coming against a projection N on the setting- 
on handle E and releasing it from a catch which holds it in 
position. The automatic stop motion when an end breaks 
is worked by the band P, which drives the shaft Q, on 
which are the spiders ; a needle falls on the spiders when 
an end breaks, but their continued movement forces the 
needle box on one side and causes the lever E. to lift up 
and release the handle E. 

The right-hand sketch in Fig. 209 will illustrate how the 
crossing motion is obtained ; a lever S centred at T is 
actuated b}' a cam at U ; this gives a to-and-fro motion to 
the end of the lever at AV ; its connection to the traverse 
bar produces a quick reciprocation motion, and so crosses 
the yarn on the swift. 

Gassing.^ — Gassing is a process in which yarn is passed 
rapidly over a light for the purpose of burning off the 
numerous ends of fibres which stand out from the body ; it 
is a very necessary operation for all purposes where yarn is 
required to be as round, smooth, and solid as possible. In 
general, it consists in taking the j^arn from the cops or 
bobbins A (Fig. 210), and after passing it through tension 
guides B and C, threading it backwards and forwards over 
small grooved pulleys D and E, and from here on to a 
quick-traverse drum-winding arrangement at F. Between 
the two pulleys D and E, just under the point where the 
yarn is crossed, is placed a gas-burner G, having a number 
of very small jets of flame ; in many cases the burner is of 
the atmospheric kind known as the Bunsen burner. The 
pale-blue flame, devoid of carbon, is intensely hot, and 

^ The subject of Gassinp is treated more fully in the Appendix. 



performs the function of singeing most effectively. The 
yarn passes through the light at from 200 to 250 feet per 
minute, and it does this from 7 to 11 times before being 

Pig. 2ia 

wound upon the " cheese." The coarser numbers — say, 30's 
twofold — go through tlic light at the slower rate and the 
higher number of times, while the higher counts — such as 
200's tAvofold — pass over the light at the quickest rate and 
the least number of times. Extra folds of yarn go slower 


stili, and the higher the counts the quicker the passage of 
the yarn must be. 

There is, of course, a h^ss in gassing yarn, the amount 
depending upon the extent of the gassing and the (juality 
of the yarn — 7 to 8 per cent representing the average, this 
meaning that if lOO's yarn is gassed the resulting yarn will 
be about 108's. The gas is supplied to the burners by a 
pipe Avhich runs the full length of the frame on each side ; 
the burners are connected to it by a swivel joint, and 
arrangements are made so that when an end breaks and 
piecing is being effected, or the machine is stopped, the 
burners move aside from underneath the y-ATn. On setting 
on the machine, the winding commences before the burners 
are moved, so that scorching or burning the threads is 
entirely avoided. The traverse motion is now almost 
universally an adaptation of the quick traverse motion, 
practically the same mechanism being used. 

Bundling Press. — As its name implies, this is a machine 
for pressing a number of hanks of cotton into a smaller 
compass, and while imder pressure the compressed yarn is 
tied up into bundles by hand. The size of the bundles 
varies, but generally 5 or 10 lb. bundles are formed in the 
machine. The illustration (Fig. 211) will convey a very 
good idea of the press ; its upper part is formed by a series 
of strong bars projecting above the table in two sets ; a 
narrow space separates each bar, through which string is 
passed ; packing paper is placed over the string, and on 
this are placed the hanks of the number to make up the 
weight to 10 lb. ; another paper is placed over the top, and 
the machine set in motion. Tlie action is as follows : — The 
driving pulleys, through strong gearing, turn a pair of 
eccentrics or cams; these lift up the sliding base upon 
which the hanks are i^laced, and as this base rises, the top 



of the yarn box is formed by a series of strong bars hinged 
to one of the sets of projecting bars above mentioned, 
being automatically caused to swivel down and become 
locked in the opposite set of bars ; a strong top is thus 
formed, and against it the bundle is pressed to the required 


Fio. 211. 

dimensions. An interval is allowed for completing the 
packing and tying by hand, after Avhich the base lowers and 
the top bars move upwards, leaving a clear space again for 
the withdraAval of the now complete bundle. The machine 
is generally made to be Avorked either by hand or power, 
and 180 ten lb. bundles can be made per day of 10 hours. 



Mill planning is essentially a branch of a student's work 
in the subject of cotton spinning, and for that reason it will 
be noticed in these pages. It is, however, so emphaticallj'- 
a distinct branch, from a practical point of view, that 
outside of giving an intelligent idea of how mill planning 
is done, it is not our purpose to do more than give suflScient 
information to enable one to use it as a basis for further 
practice, and as an aid in understanding the work of others. 
To plan a mill with all due regard to room, lengths of 
machines, arrangement of machinery, passages, driving, 
economy in carriage from process to process, speeds, hanks, 
productions, drafts, etc., etc., requires years of practical 
experience, and only those whose sole occupation it is can 
do it thoroughly. 

The total weight of yarn required to be produced and 
the counts spun are the two chief items it is necessary to 
know before commencing to plan the machinery ; with 
these as a basis, we Anil proceed to give examples of various 
mills spinning dift'erent classes of cotton. 

Example. — A 10,000 spindle mill, spinning average 20's 
on mules. 

A suitable space of spindle will be 1 J- inch. 
VOL. Ill 2 C 




A suitable length of mule will contain 1000 spindles, so 
that there will be ten mules. 

A mule spinning 20's will produce about thirty-two 
hanks per spindle per week. 

The ten mules Avill produce — 

1000 X 32 hanks 

16,000 lb. of yarn. 

20 counts 

Proceed now to make a table as follows, filling in the 
necessary data according to experience ; the data given in 
the tables are good average results. 




and lbs. 



Card . 





Draw Frame . 





Slubber . 






Intermediate . 





Rover . 










'Note. — Allow five per cent waste between the card and 
the mule. Of this, alloAv two per cent in the mvxle, the 
rest being divided among the other machines, this being 
sufficient for practical purposes. 

On referring to the table we note that the card produces 

700 lb. per week, so that 

.^ „^ , 16800 
No. of Cards = - ^,,., =24 

No. of Draw Frame deliveries = .„^^=16'68 
No. of Slubber Sinndles = - — ^^ — '^ = 165 


r, . ,, 16440x11 ,,, 
No. of Intermediate Spindles = jj = 514 

16320 x3J 
No. of Roving Spindles = —^ = 1 326 

No. of Mule Spindles = ^^ = 10000 




C3Cgi rn:n 

clnzzir c:]0 mr^ fU, 
a[z|[inrr] - '' 



< o 

tr £ 



H,-i -i> r-i rH 
.°° II II II II 

m pi -^ A r-l 

iJ m 

1-^ ^ 

S c :; - s 

fe & 

O o a> '■=' o 


1^ S2SS 


o ^ 




CS k" 

t'^ 5 . 


P, O. r^ C, ,^ 

o - - ~ 
r-i C^ w "^ 








9 p p p 




^-1 f 4i .|jh , 

-Tf i ^o'-y - 1^0 X;!t°''^ "'° T t' 110 ^ ivojr-110 T li oH 


I 1 


■o o 

— 0— OH}— 




For 20,000 Mule Spindles, Spinning No. SO's Combed Yam. 
20 Cardins Engines. 

2 Silver Lap Machines. 

2 Draw and Lap Machines. 
14 Combing Machines of S heads each. 

2 Draw Frames, each 3 heads of 7 deliveries. 

2 Slubbers, 64 spindles each, 7 in. spaca. 

4 Intermediates, 130 ,, ,, (>='?,, ,, 

iO Jacks, 200 „ ,, 4^,, ,, 

FiQ. 215. 




O (M O 



















1— ( 








CO -* »0 lO 

















































Containing 931(5 Rincc Spinning Spindlps. Spinning 16 s to 30 s Twist and 

20's to 36's Weft, and 2S0 Looms. 

Fig. 217. 



From these results we decide upon suitable lengths of 
frames, etc., and draw up a table of the machinery necessary 
for the mill, first noting that a vertical opener will produce 
30,000 to 40,000 lb. per week, and a single scutcher 15,000 
to 20,000 lb. per week. 

1 Double Vertical Ojoener. 

3 Single Scutchers. 
24 Cards. 

4 Draw Frames, 3 heads of 5 deliveries. 

2 Slubbing Frames, 86 spindles, 8 in. space, 10 in. lift. 
4 Intermediate ,, 132 ,, 6^,, ,, 9 ,, ,, 

8 Roving Frames, 170 ,, 5 ,, ,, 7 ,; ,, 
10 Mules, 1000 ,, \\„ „ 

From this example the general method adopted to obtain 
the number and dimensions of the machines will be easily 
understood. They are then planned out to scale to the 
best advantage, and to illustrate this planning several 
examples are given in Figs. 212 to 217. 

The following particulars may prove useful as a guide to 
the planning of a mill, together Avith other information 
given in various articles that have already appeared : — 



Kind of Cotton. 

Weight of 
Lap per yd. 



Lbs. per 

Card ill 10 




Indian or American . 

13i oz. 






13i „ 










American . 






) 9 

































,, Combed 











>» !J 








For a week of 56| hours multiply the above productions 
by 5-65 for a week's production. 

Production in lOliours: 

min. in 10 hours x revs, of calender roller x 
dia. of calender roller x 3 'Hie x weight of 
sliver in grains per yard 
" 36x7000 


Dia. of 

F. Roller. 

Revs, of 
F. Roller. 

of Sliver 
per yard. 


Lbs. per 
in 10 lirs. 























f Indian 
1^ or China 



















100 j 

32's r 

to \ 

40's \ 






or Low- 



























































Production in 10 hours = 

min. in 10 hours x revs, of F. E. x dia. of 
F.R. X 3-1416 X grains per yard of sliver 
36 in. X 7000 grains ~ 








o o o 
o o o 












■ ■ - cPcPcP 

o o o 

(»0 o 

■ ■ ■ o"o"o" 

I-l 1— 1 I-l 


:::::: o 7^ 

CO C2 u-5 

r^ O T^ "?* 

02 OS O O O r-l 

I-H 1-1 1— 1 T-H 


j-^ ^, in I.-; o Ci . . 
00 oo 00 i) C: ■— 1 : '■ 


. ■ 'P T' 

: : : : A- o 

CO (M o in t^ 

ip !p o CD in CD 'p 

-* -^ m in CO in ^ 


!■- in 

^ t^ 00 I^ CO 'I' OO fM 

CD '-D 

•"f ^ ^ '^ 00 

^ in m 

(N <J» C^ CO r- •;?< CD 

CD OO 4ll O O ^ C5 
I-l rH ,-1 (M (N I-l r-l 

■to O O OO C3> CC 'O Ol 

00 CO OO 00 in CO 


I-l 1— 1 

Oi OO 

CO oj in i^ "p "^ 


■— 1 I— 1 

u-5 in 

■511 cq tN 02 

02 r-^ «D 'O ^ I-l 

.-^ -M m in OS T-H 

T-H ,-1 ,-1 rH tM CO 

I— 1 OO in in 02 in i~~ 

02 in 00 <M --o >-i t^ 
<rq CO CO in CD CO c^ 

t^ 1^ rH O r-l in CO i-H 

^'^<inint^C200i— 1 


in in in 

CO <N <N (M '-0 O 
CO ^ "^^ '^i CD O 
.-1 ,-1 I-l .-1 rH (N 

-rf 00 

oi CO o in r-i CO t^ 

02 1— 1 O OO CO o l^ 
(M CO ^ ^ CO CO (M 


00 CO O 

■ : : ^ CO o 

■-I .-1 O) 

in tp 

in -* ci o 'i' c^i m CO 02 in — 1 o '~o CO 

«0 02 ^^ O rH m '-tl ;D -O fM r-H O 1^ CO 

^jt^inm i^oi«Dj>.cooi-<Or-ico 


in CO -H m I— 1 02 1^ 

CO r-l OO (M CO "-^ 1^ 



Indian or American 

Indian .... 

Cliiiia .... 
Japan .... 



Indian or Amc 







O V. o o o o o o 
r-l i-i !M (M CO 'a* ^ in 

so T 
90 T 
100 T 

H H rH H H H H 
CD o o o O' -r< ^l» 

I-l (M !M CO ^ I-H r-l 
. . 


U Eh 






~ " ~ "1 " " 










co T*^ 

■-I lO 

?D to 

00 00 oo 00 00 00 

CO CO CD CO g g g g g g^<» 


»0 m CO u-5 

i-H CO 1^ O lO lO Oi 

^ i-H CO >o 

o o 


CD CDt^t^lr^COQOO 
OlO(MCO-*<r-OCOCOr-l,-l05 00 

oocooi— lOjOr-Hi— icoc-i>ocoOi— ii-Hcooooco'M^O':D<r)-t<oo?DO 

t— CDin^i— iCOl^^iOiniOlOiOOOtDlO-^i— lOOa500?D«C>!0«OC0 1^ 
?-li-lr-li-l.-l i-li-Hr-li-li— li-Hr-l 

5 ■> 



00 CO ^ CO 05 CO 00 

CO CO lO t^ OD o o 

■-H r-l i-H ,-H .-H (M C-l 

r-l I-l .-H 1^ 

CO CO CO 1-^ 
0\ IM (M fM 

t^ CO 

C-1 T-H 





^COOSOOCOt-Hi— ll-HCOi— !■— lOO 
i-H .-1 i-H (M (>) IM CI C-l (M (>) IM C-l 





. =^ 


" " " " 
















" " 




n to 

<1 H 

= = " - 

g .2 

s "U 



I-l I-l 



Ring and Mule 

" " 

1) !S 






_2 bo ® 

00 Oi 'p 

iCi CC CO 

CO ■* 



4j( '^ 

Ip CO ip OT t-^ -tM -^ --Jl 00 ip CO Cq rtl lO 


CO . : . ; ::::::::::::::::::: 


. lO »c in ^ 

. 00 CO OO '^ 

(N CO CO 00 Oi Oi 05 
CO CTi l^ CO Oi C^ Oi 



t^ OS 

00*^^*0505 1^ CO t-.i-i?D!ncoi-icoco 

00<MOt^«DO i-ll^Ciai-*tOOiO 



Oi Oi 

OO00000S0000C0i--.l>.O05 (--.«>. I--. t^CDl/^ 

CO I-l 1-1 l-l 1-1 7-1 

I-l rH 


Comber 8 lids 
Slubber . 


z * 



" (3 








" s 


American . 

Egyptian . 

,, Combed 

Indian or American 

American . 
Egyptian Combed . 

Indian or 

1 1 



"-1 1-1 I-l 


Ci l^ -^ ~^ '~D Op rH «; C<1 «p O 

p. » 2 


lO OS o .-^ CO "^ ^ 


•;2 P^ 




Humidity of the air in cotton mills is a subject upon which 
much has been lately written, and so important as well as 
interesting is the subject that several writers and able 
observers have enabled the industry to benefit considerably 
by the results of their observations, experience and advice. 
Two names stand out very clearly in this connection — 
namely, Sir B. A. Dobson and Mr. W. W. Midgeley— and 
the fruits of their combined labours in books published by 
Messrs. Dobson and Barlow will also be looked upon for 
some time as standard works on this subject. Under these 
circumstances it is not intended to do more than give a 
mere outline of this feature of mill management. 

The essential meaning of humidity is dampness or 
moisture, and its association with spinning relates to the 
condition of the atmosphere of the rooms in which spinning 
operations are in progress. Now this moist condition of 
the air involves t\vo factors : First, the actual amount of 
moisture ; and, secondly, the relative amount. Strange to 
say, the actual amount of moisture in a spinning room is not 
the deciding factor in the case. For instance, yarn may be 
spun well when the temperature of the room is, say, 70° F., 



but if that temperature is raised to 90° F., everything else 
remaining the same, there will be a vast difference in the 
humidity of the room in spite of the fact that the actual 
amount of moisture in the air is the same in each case. 
When one enters a room that is well heated it will be 
noticed that it is very dry or has a parching effect, but as 
a matter of fact a cubic foot of air in such a room will have 
quite as much moisture in it as a cubic foot of outside air. 
The question may then be asked — Why do we say air is 
dry or moist ? The explanation lies in the fact that these 
terras dry and moist are not actual but simply relative 
terms, and that the human body is not capable of deciding 
from its sensations what the actual humidity of the atmo- 
sphere may be. We have an analogous example in the case 
of temperature. A spinning room may be 90° F. and in- 
conveniently hot, but any one placing a hand on the framing 
of a machine would feel it very cold, while in reality the 
iron is at practically the same temperature as the room. 

We all know that the air in summer is much drier than 
in winter, though it is equally well known that there is 
more moisture in the air diiring summer than winter. 
These considerations lead us to the conclusion that the 
actual amount of moisture in the air is not a deciding factor 
in our estimate of humidity, so that we must seek for some 
other element to solve the problem. If water is left to 
itself in contact with air it will slowly pass into a state of a 
gas or vapour, the phenomenon being known as evaporation. 
The water which has thus been transformed into vapour is 
in an extremely subdivided state, and diffuses very rapidly 
in the air without increasing the volume of the air with 
which it has become mixed. It is advisable to point out 
here that when moisture enters the atmosphere, whether 
by evaporation or by spraying, no chemical combination 


takes place : it is purely a mechanical mixture. The vapour 
of water, therefore, by virtue of its elastic force, Avhich it 
possesses in common with all other gases, takes up its 
position between the molecules of the air wherever it is 
free to do so, and moreover it always remains moist and 
acts just as it would if it were confined within a vacuum. 
Now, under a given set of conditions, vapour would continue 
to rise from the surface of the exposed Avater until the 
vapour tension exactly equals the tension which keeps the 
Avater in a state of water ; after this state has been leached 
no further evaporation can take place, for the air has now 
mixed up within itself as much vapour as it can hold, or a 
much Ijetter way to put it is to say that the particles of 
vapour in the air are putting forth all the pressure they are 
capable of exerting in keeping each other from changing 
back into the state of water from which they have arisen. 
The air is now said to bo " saturated," and the particles of 
vapour are exerting their " maximum pressure " and are 
also at their " maximum density." If the temperature of 
this saturated air is now lowered, if it is compressed into a 
smaller volume, or if an attempt is made to add more 
moisture to it, the moisture already in it will begin to be 
deposited in the form of dew, the temperature at which it 
does this being called the " dew point." 

A further characteristic to note is, that water will not 
evaporate into cold air to the same extent as in Avarm air. 
For instance, 2*13 grains of water will evaporate and 
saturate a cubic foot of air at 32° F., while 19-84 grains 
will evaporate before it saturates a cubic foot of air that is 
at a temperature of 100 ' F. This, of course, leads to the 
conclusion that the dew point varies according to the 
temperature of the moisture, or, in other words, the elastic 
pressure of the particles of vapour increases as their temper- 
VOL. Ill 2 D 


ature increases. In this connection we may point out that 
the air itself has nothing whatever to do with humidity, for 
all the phenomena of saturation, deAv point, etc., can be 
observed in a vacuum, and, as a matter of fact, it is from 
experiments performed in the absence of air upon which our 
knowledge of the dew point depends. Those, therefore, 
who speak of the property of air to retain moisture are 
wrong in principle ; the air happens to be a convenient 
vehicle for heating the vapour as it arises from the surface 
of the water, and in so doing increasing its elastic force and 
enabling still further evaporation to take place ; the applica- 
tion of heat to the water itself will cause vapour to be 
given off, and this, rising in the atmosphere, will heat the 
air, and so the same result naturally follows. 

We can now deal with the relative humidity. If air 
contains a certain amount of moisture, and this amount is 
only half of what would cause saturation, the humidity, of 
the air is said to be 50 per cent ; so that when we say that 
air is " dry " we simply imply that the proportion of 
moisture in the air is small compared Avith what the air 
would contain if it was completely saturated ; cold air with 
little moisture in it may be very moist, while warm air 
with much moisture in it may be very dry. 

It is now seen that the point of saturation or dew point 
is the foundation of our estimate of " humidity," and there- 
fore we must know this before the percentage of humidity 
in a room can be known. To do this would require skill, 
but fortunately Mr. Glaisher took advantage of a long 
scries of experiments made in England, America, and India, 
and from them constructed a series of tables by the use of 
which the humidity can readily be found. His first set of 
tables differed considerably from his later ones, published 
in 1856, but these last ones are now used as a standard by 



British observers, though other countries still retain tables 
based on their own observations, this accounting for the 
fact that in America, for example, the tables used are 
different from Glaisher's, and an American would give the 
humidity of a room slightly differently than we should in 
this country. It is simply a question of observation and 
experiment, and the tables are purely empirical. 

The instrument used to indicate humidity consists of 
two thermometers, one of which has its bulb covered with 
a thin piece of muslin cloth connected by an absorbent 
strand of material to a small well of water placed at a 
short distance from the thermometers. One thermometer 
Avill register the actual temperature of the air, the other, 
owing to the moistened covering, indicating a less tempera- 
ture ; this comes about, because water in changing into 
vapour expends heat and the remaining water becomes so 
much colder. The water in the muslin evaporates and the 
heat expended in this action leaves the water slightly 
colder. As this colder water is in contact with the bulb 
of the thermometer it causes the instrument to indicate a 
less temperature, and so we have two readings, one from 
the wet bulb and the other from the dry bulb thermometer. 
The difference between the two supplies us with the basis 
upon which to estimate the humidity, Glaisher's tables 
giving the amount at a glance. 

Since Glaisher's tables are intended to cover extreme 
conditions of temperature we find that the makers of 
the instrument just described — Avhich is knoA\Ti as a 
" Hygrophant " — issue a leaflet containing only the range 
of temperature likely to exist in the mill. A portion of 
such a table is given below, and it has some value to the 
cotton spinner, because it represents what Sir B. A. Dobson 
and Mr. Midgeley found to be the best relative humidity 



in the spinning rooms ; the complete table will be found in 
Sir B. A. Dobson's book on Humidity. 













per cent 

per cent 
















1 ''- 






^ 71 










72 -6 






























































An instrument is being extensively used now, that 
avoids the trouble of referring to separate tables. It is 
an American patent taken out by Huddleston in 1874. It 
consists of the wet and dry bulb thermometers, but between 
them is placed a cylinder on which is printed in upright 
columns a series of figures ; each column is headed by a 
number, Avhich represents the difference in temperature 
between the two thermometers. Close to the cylinder is 
a scale similar to the dry bulb thermometer. By turning 
the cylinder until the column of figures having the number 
on the top equal to the difference between the thermometers 
is close to the scale, we read the temperature of the dry 
bulb on the scale, and opposite to this numlier is the 
percentage of humidity in the room. This instrument, 
not being based uj)on Glaisher's tables, is not correct for 
use in England, but a Manchester firm (Casartelli) are now 

vrii HUMIDITY 405 

making a copy of this hygrometer having correct readings 
and specially constructed for mill use ; an illustration of it 
is given in Fig. 218. 

When the cotton industry was passing through its initial 
stages it was soon discovered that two essentials were 
necessary to obtain good results — namely, a Avarm atmo- 
sphere and a moist atmosphere, and both were obtained in 
the usual way by heating appliances and spraying the 
floors of the mill, the moisture arising in the process 
of evaporation. Improvements were effected and many 
methods have been adopted, chiefly on sanitary grounds, 
for obtaining the best and most permanent effects in both 
directions, for it was found that moisture played a very 
important part in the production of level and strong yarn. 
Cotton fibres are hygroscopic in character — that is, they 
have the property of absorbing moisture, and in doing so 
they become for the time being less brittle, more pliable, 
and capable of being incorporated more thoroughly among 
themselves in the 3'arn. Electricity in the mill produced 
by the friction of moving parts — chiefly in the belts — 
is a disturbing agencj- among loose fibres, and causes an 
additional fuzziness in yarn, which is naturally made fuzzy 
by the spinning operation ; in a warm, dry atmosphere this 
electricity is capable of exerting its full influence on the 
yarn, but the presence of moisture neutralises its effects 
considerably, and it is also for this purpose that a reasonable 
degree of humidity is desirable. Suitable climatic condi- 
tions such as exist in Lancashire supply a natural source of 
moisture, but taking into account the heat of a spinning 
room, it is found that artificial forms of moistening the air 
are requisite if the full benefit is to be obtained in the 
yarn. Mr. Midgeley, by micro-photographs of yarn spun 
under varying conditions, has been able to demonstrate this 


fiict to a certaint}^ and his experiments have led to the 
conclusion that the best results are obtained with the 
humidity as given in his talkie just quoted. 

Artificial methods of introducing moisture into a room 
are l)ased upon two properties of water : first, it is capable 
of being, as it were, pulverised into very fine particles ; and, 
secondly, its evaporation. So far the first method is the 
one chiefly adopted : the water is forced at a very high 
pressure in the form of a thin stream through a fine nozzle 
and made to impinge against a fixed surface ; the water is 
broken up into myriads of fine particles, and in this con- 
dition is sent into the room and caused to diff'use either 
artificially or naturally. 

The second method is to place open troughs in suitable 
positions about the room, fill them with water, and assist 
evaporation by running small steam pipes through them. 
Now, although moisture quickly diffuses in the atmosphere, 
it does not do so to a sufficient extent to give uniform 
results throughout the room. A recent improvement has 
been introduced, by means of which currents of air pass 
over the surface of the water in the trough and disperse 
the evaporated moisture uniformly in the atmosphere ; this 
is a very important matter, for, in addition to equalising 
the humidity, there is a constant supply of fresh air 
admitted to the room. 


Some of the following useful information may be found in 
other parts of the books, but it is sufficiently important to 
be gathered together and augmented so as to form a concise 
and useful reference. 


Single Acting Macarthy Gin 
Donble Acting Macarthy Gin 
Bale Breaker 

Willow .... 
Small Porcupine Opener 
Automatic Hopper Feeder . 
Vertical Beater Opener, Single Crigliton 
,, ,, ,, Double Ciigliton 

Exhaust Opener .... 

Single Opener (without Hopper Feeder) 
Donble Opener ( ,, ,, ,i ) 

Single Scutcher 
Double Scutcher 
Card, Revolving Flat . 
Sliver Lap Machine 
Ribbon Lap Machine (Draw and Lap IMaehine co 
Comber, single nip, 6 heads 

J) II IJ 8 5, 

,, double nip, G ,, 
„ „ 8 „ 
Draw Frame 


per 12 deliveries 1 









Slubbing Frame ...... 

90 spindles per 

Intermediate Frame ..... 

130 „ 

Roving Frame ...... 


Jack Frame ...... 


]\Iule, Indian and American cotton 


Mule, Egyptian and Sea Island cotton 

130 ,, 

Ring Spinning Frame 

Ring Doubling Frame .... 

100 „ . ,, 

Twiner, Yorkshire principle 

200 ,, 

Twiner, French principle .... 


Quick-Traverse Winding Frame . 

80 drums ,, 

Ordinary Winding Frame .... 

300 spindles ,, 

Gassing Frame ...... 

80 drums ,, 

Reel (Coleby's) 

Improved Reel (for gassed yarn) , 

Single Ordinary Reel . . , . • 

Double Ordinary Reel .... 

6 reels ,, 
8 „ 
. 16 „ 
8 „ 

Copping Frame ...... 

Bundling Press ...... 

300 spindles ,, 

Banding Machine ..... 


Tubular Banding Machine, 3 heads 


Balling Machine . per head \ 

The foregoing particulars represent average results, and 
on testing them on a number of mills through the steam- 
engine indicator, they were found in some cases to be 
below, while in others they appeared to be somewhat 
excessive. They may be taken as fairly accurate, a little 
judgment being necessary in fixing the spindles per horse- 
power for the mule and ring frame. 



Cotton. Front. 




Indian and American Cotton 
Egyj)tian cotton .... 
Sea Island cotton 






Some people prefer for American cotton — 

1st. 2iid. 3rd. 4th. 

20 1b. 18 1b 16 1b. 14 1b. 


Kind of Machine. 

Kind of Cotton. 






Indian and ) 
American j 

18 lb. 

24 lb. Saddle and Bridle. 


Egyptian, etc. 

16 lb. 

20 lb. Saddle and Bridle. 

Slubber . 


14 1b. 

12 lb. 1 Self- Weighted. 



Indian and ~\ 
American j 

16 1b. 

20 lb. Saddle and Bridle. 


Egyptian, etc. 

14 1b. 

18 lb. Saddle and Bridle. 



12 1b. 

10 lb. 1 Self-Weighted. 


Indian and "j 

IS lb. 

24 lb. Saddle and Bridle. 

American j 

Roving and Jack 

Egyiitiau, etc. 

16 lb. 

20 lb. Saddle and Bridle. 



10 1b. 

Self- Weighted. Self-Weighted. 

Jack . 


8 1b. 

Self-Weighted. Self-Weighted. 

Another firm adopts tlie followim 




Slubbing Frame 
Intermediate Frame 
Roving Frame (double . 
Roving Frame (single boss) 

18 1b. 
14 1b. 
18 1b. 
10 1b. 

14 1b. 
10 1b. 
14 1b. 

8 1b. 

10 1b. 

8 1b. 
12 1b. 

6 1b. 

diameteks of rings and spaces suitable for spinning 
Various Counts of Yarn 

For 4's to 20's counts, space 2j in., dia. of Ring 1| in. 
„ 20's „ 40's ,, ,, 21 „ „ „ li „ 

,, 40's counts & upwards ,, 2i ,, ,, ,, \\ ,, 

If an anti-ballooning motion is used, then 

For 4's to 20's counts, space 2f in., dia. of Eiug If in. 
„ 20^s „ 40's „ „ 21 „ „ „ If „ 

,, 40's counts & upwards ,, 2J ,, ,, ,, H ,, 

„ ^Veft . . • ., 2i ., ,, ,, l^^Vto li 



Fly Feames 

indian and low american cotton 

Slul)ber, sq. root of hank roving multiplied by 1'3 

Intermediate, ,, ,, ,, „ 1"2 

Roving, ,, ,, ,, ,, 1-5 


Slubber, stj. root of Lank roving multiplied by I'lS 

Intermediate, ,, ,, ,, ,, 1'25 

Rover, ,, ,, ,, ,, 11 

Jack, American, ,, ,, ., ,, I'l 

Jack, Egyjjtian, ,, ,, ,, ,, 0'9 


Slubbers, sq. root of hank roving multiplied by 0*7 
Intermediate, ,, ,, .. ,, 0'78 

Rovers, ,, ,, ,, ., 1"1 

Jack, Egj'ptian, ,, ,, „ ,, 0'9 

Jack, Sea Island, ,, ,, ,, ,, 0'95 

In regard to these tables, it may be remarked that some 
spinners use the multiplier 1-2 throughout the frames. 


Twist, Indian and American cotton, multiply square root of 

counts by ......... 3 "75 

Weft, Indian and American cotton, multiply square root of 

counts by , . . . . , . . . 3"25 

Twist, Egyptian cotton, multiply square root of counts by . 3 '606 

Weft, Egyptian cotton, multiply square root of counts by . 3'1S3 

Ring Frame 

Twist, Indian and American cotton, multiply square root of 

counts by . . . . . . . . . 4*00 

Twist; Egyptian cotton, multiply square root of counts by . 3"606 

Doubler Frame 
Multiply the square root of counts by . , . , . 4*00 




No. 1 Mill, No. of spls. {^™jo;i5'Sf"''^}53,000-48 spindJes i.h 

(all mules) 

69,300 = 72 

101,900 = 66 

82,000 = 69 

80,000 = 66 

Preiiaiing niacliinery is included in all the above mills. 


Kind of Cotton. 

Cylinder. i Doffer. 
Revs. 1 Revs. 

Feed Roller. 


Indian Cotton 
American „ 
Egyptian „ 
Sea Islands „ 

165 to 170 
170 to 180 
160 to 166 
150 to 160 

15 to 18 

14 to 20 

9 to 12 

5 to 9 

2 to 2-3 

2-3 to 2-5 
2-5 to 2-7 

400 revs. 

The flats travel about 2>\ inches pei" minute. 


Kind of Cotton. 





Indian — 

There are 




70 to 90 

firms who are 



100 to 110 

80 to 110 

noted for 

American — 

good work 





who use the 






Egyptian — 

counts of wire 


110 to 120 



given in 


120 to 130 


this table. 

Sea Islands 


130 to 140 

130 to 140 

Position of the Wharve on Mule Spindle. — In order 

to o1)tain the best results in driving the spindle, the spindle 
ought to set so tliat, if a straight edge be placed on the 



under side of the wharve, it will occupy the following 

positions : — 

T, , . • ■ ,1 1- • 1 i J -Hi 1 ii render side of tlie 

l"or 14 111. siundle straifirht edge will touch the- ,• ,, , r, 
^ ^ o 1^ ^jj^ i-oUei- shaft 


•will be 1-5 in. below 

Ends : Piecing-up. — The following table represents 
the numlier of ends pieced up j^er day (caused by breakages 
only) on the various machines. Three mills are taken, and 
they are the result of extensive observation for this specific 
purpose made by the secretary of Mr. Geo. Draper. The 
table is given by permission of Messrs. Geo. Draper and 
Sons, U.S.A. : — 





No. 1 Mill. 

No. 2 Mill. 

No. 3 Mill. 

Card . 




Drawing No. 1 




„ 2 


„ 3 




Slubber . 




Intermediate . 

13 -30 







Ring Frame, T. 


630 '00 


» w. 






The piecing-up on the preparing machine is estimated 
on the total number of spindles in the mill, while that of 
the spinning machinery is based on 1000 spindles : for 
instance, according to the table, a mule of 1000 spindles 
would have all its ends broken 1-G7 times during a day. 


24 grains = 1 pennyweight (dwt. troy). 

18 dwts. 5i grains = 439 -5 grains =1 ounce (oz. avoirdupois). 

16 ounces -=7000 grains = 1 pound (lb. avoirdupois). 


54 inches = 1 thread or circumference of wrap reel. 
4,320 ,, =80 threads or 1 lea or skein. 
30, 240 , , =560 threads = 7 leas = 1 hank = 840 yards. 
The number of hanks in 1 lb. is the count of the yarn. 
A bundle of cotton yarn is as many hanks as make 10 lbs. 


Circles, Areas, and Figures 

Diameter of a circle x 3'1416 or y =the circumference. 

Circumference of a circle x 0"31831 or -n^=the diameter. 

Square of diameter x "7854 = the area of the circle. 

Square of diameter x |-^ = the area of the circle. 

Square root of area x 1-12837 = the diameter of a circle. 

Radius of circle x 6 '28318 = the circumference. 

Circumference = 3 '5449 x ^^/area of circle. 

Diameter of a circle x •8862 = the side of an equal square. 

Side of a square x 1 "128 = the diameter of an equal circle. 

Area of triangle = the base x \ the perpendicular height. 

Square of the diameter of a sphere x 3"1416 = the convex surface. 

Cube of the diameter of a sphere x "5236 = the solidity. 

Diameter of a sphere x '806 = the edge of an equal cube. 

Diameter of a sphere x '6667 = the length of an equal cylinder. 

Surface of a cylinder = area of both ends + length x circumference. 

Solidity of a cylinder = area of one end x the length. 

Solidity of a cone = area of the base x \ the perpendicular heighte 

Area of an ellipse = long axis x short a.xis x 0'7854. 

Conversion of one Denomination to Another 

Feet X 0-0001 9 = miles. 

Yards x 0-0006 = miles. 

Square inches x 0-00694 = square feet. 

Square feet x 144 = square inches. 

Cubic feet x 0-037 = cubic yards. 

Cubic inches x 0-000579 = cubic feet. 

Cubic feet x 6 -2355 = gallons. 

Gallons x 0-16059 = cubic feet. 

Gallons x 10 = ]bs. of distilled water. 

Cubic feet of water x 62 '425 = lbs. avoirdupois. 

Cubic inches of water x 0-03612 = lbs. avoinlupois. 

Lbs. avoirdupois x 1-2153 = lbs. troy or apothecary. 


Lbs. troy or ajrothecary xO •8228 = lbs. avoirdupois. 
Lbs. avoirdupois x '00893 = cwts. 
Lbs. avoirdupois X 0*000447 = tons. 
Tons of water x 224 = gallons. 


Tables of the Horse-Power of Transmission Rope, by C. W. Hunt. 
The working strain is 800 lbs. for a 2-inch diameter rope, and is the 
same at all speeds, due allowance having been made for loss by centri- 
fugal force. 

Speed of the Rope in Feet per Minute. 


1500 ! 2000 


3000 ■ 3500 


4000 '4500 5000' 6000 









7-2 7-7 7-7 










9-8 10-8 10-8 








12-8 13-6 
















\\ 13-1 
















39-2 ' 41-5 








42-8 1 47-6 

51-2 1 54-4 ! 54-8 

1 i 

50-0 35-2 i 84 


Thickness. — Belts are of various thicknesses, but in a 
mill they are seldom below ^^ in., or above ^ in. The 
average may be taken as -r^^ in. 

Speed. — It is advisable to keep within the limits of 
3500 ft. per minute. 


1100 X Horse-Power of machine 

Width of belt = - 

Vel. of belt in ft. per miu. 


Power. — 

H. P. = Horse-power, 

W = Widtliof belt. 

r = driving force in lbs. 

T = Tension in belt. 

L = Circumference iu inches of pulley covered by belt. 

V = Velocity of belt in ft. ])er min. 

r= ,, ,, ,, second. 

A = Covered area of driven pulley in inches. 

Z= Circumference in inches of driven pulley covered by belt. 

l — x-^kx. 

y^ WxT rp ^,jjj,jgg fj.o„^ 70 to 150 lbs, 

oonnn ,, XT i"> 


3-iUUU X H. 1- 

:. AV - -02 T. 

Fx V 


38000 H.P. 




If a little less than half the pulley, viz. '4 of it, is covered by the 
belt, ^■ = l•l. 

H a little more than half the pulley, viz. '6 of it, is covered by the 

belt, Z;=-62. 

66000 X H.P. . , ,1 T 1^- 

lor double belting. 


= ^xV f°^ 




A = 

66000 X H.P. 


H = 

Wx V 

= 33-000 -«^°^«- 


36000 H.P. 

Diameters. — Pulleys ai-e not working nndei- good con- 
ditions if one of the pulleys is more than six times the 
diameter of the other. 

Width. — The pulley ought to l)e almost 1 j times the 
width of the belt. 

Preservative. — Castor oil applied to the back of the 



belt every few weeks, especially if the atmosphere becomes 

Splicing. — 

Width of belts, 1 in., 2 in., 3 in., 3 to 6 in., 6 to 8 in., over 8 in. 
Lap in inches, 2 in., 43 in., 5^ in., 6 in., 8 in., 10 in. 

Double Belts. — Double belts transmit \\ times more 
power than single belts. 


For Ascertaining the Weight of Hank or Decimal Part 
OP a Hank 

KuLE. — Divide 7000 grains (1 lb. of yarn) by 840 yards = 
dividend for 1 yard. 




































833 -333 





If 2 yards of card sliver weigh 80 grains, what hank is it ? Divide 
the dividend for 2 yards by 80 = 0-208 hank. 

If 30 yards of roving frame roving weigh 62^ grains, what hank is 
it ? Divide the dividend for 30 yards by 62i = 4 hank roving. 

What ought 60 yards of a 4i hank roving to weigh ? Divide the 
dividend for 60 yards by 4^ = 111 grains. 

VOL. Ill 2 E 











































































































































































































Note, : — For the square roots of higher numbers refer to the 
Yarn Table opposite. 






AND American | 

Egyptian Cotton. 






Root of 















5-30 ; 





6-49 ! 





















8-59 1 





9-19 1 














10 -10 











































4 000 











































































































































27 -12 
























' 22-94 





! 23-38 
















Indian and American 



Root of 


Egyptian Cotton. 1 







Mule Mule 
Twist. Weft. 

Ring I 



27-93 24-54 




28-39 25-05 




28-85 i 25-45 



























































33-83 29-84 




34-21 30-18 




34-59 : 30-52 




34-96 30-85 




35-33 ' 31-17 




35-70 31-50 






























37-81 33-32 





38-16 1 33-68 





38-50 1 33-98 


116 1 



38-83 34-28 




39-17 34-57 





39-50 34-86 




It has been thought necessary to give a few words of 
explanation of further improvements that have been 
eflfected upon the arrangement illustrated in Figs. 108, 
109, 110, and 114. These improvements are quite recent, 
but they have proved so valuable in enabling changes to be 
rapidly and certainly made that the machine-makers who 
make this type of mule are adopting them on all their 
newest mules. The young reader is advised to read up 
thoroughly all that has been said on the specific actions of 
the mule in the previous pages ; if this is done, the follow- 
ing brief summary of the actions now to be described Avill 
become comprehensive to him. 

Referring to Fig. 219 it will be noted that the long 
lever is retained, being fulcrumed on the side of the framing 
on the stud 2 ; it carries several studs or stops, as at 3, 4, 
5, 6, and 7, the purpose of which will be subsequently ex- 
plained. The drawing shows the positions of the various 
parts, as when the carriage is running out and spinning is 
in progress, under these circumstances : — • 

The strap is on the fast pulley on the rim shaft. 
The backing-off lever D is kept from permitting the 
backing-off cone wheel to go into gear with the fast 


pulley on the rim sliaft by the stud C on the backing- 
otf rod A. 
The dra"\ving-up cone is kept out of gear with the scroll 
shaft by stud 7 on the long lever, and by the catch 
G which is carried by the drawing-up lever centred 
at F. The catch G, it will be noticed, rests upon the 
end of the backing-ofF rod, and in this position it pre- 
vents the drawing-up lever, which is fulcrumed at F, 
from falling into gear with the scroll shaft. 
The long lever is kept from changing by the stud 3 
being hooked under the recess in the lever which is 
fulcrumed at X, also by weight 14 resting directly 
upon its end. 
It will also be observed that the spring K is in tension, 
and is tending to pull forward the backing-off lever 
D, but is prevented from doing so by the stud C on 
the backing-ofF rod. The spring " g " is also in 
tension, and is tending to pull the drawing-up lever, 
centred at F, into gear w^ith the scroll shaft, but is 
prevented from doing so by the catch G and the pin 
7 on the long lever. 
Since spinning is in progress, the faller leg is not con- 
nected to the shaper bowl Q'. 
As the carriage nears the completion of the run-out, a 
bowl 8 on the carriage square comes into contact with the 
inclined end 9 of a lever centred at 12 ; this has the effect 
of depressing it, and lifting up the end 1 3, upon which the 
weight 14 rests, and Avhich is, as a consequence, raised out 
of contact with the end of the long lever. 

The end M of a lever fulcrumed at N now comes against 
a projection L on the backing-off rod and moves it forward, 
thus freeing the backing-off leA^er D and causing the spring 
K to pull it forward and so jjutting the backing-off cone 


wheel into gear with the rim shaft. Backing-ofF now takes 
place ; the scroll on the tin roller shaft T winds on the 
backing-ofF chain and causes the faller leg to rise until the 
recess U is pulled over the upper part Q of the slide which 
carries the shaper bowl Q'. This action of course causes 
the lever M to be drawn backwards as well, and the 
backing-off rod, being now free from M, instantly shoots 
backwards under the influence of the spring K", and the 
backing-off cone is taken out of gear. When the backing- 
ofF rod is moved forward by M a stud B is brought under 
a part of the drawing-up lever at E, so that during " back- 
ing-ofF" the drawing-up lever is locked; the catch G is 
also disconnected by the same movement from the end of 
the backing-ofF rod. 

On the release of the backing-ofF rod and a simultaneous 
release of the lever centred at X the long lever is free to 
change, so that the drawing-up lever at once puts the 
drawing-up cone into gear with the scroll shaft and the 
carriage is drawn in. When the long lever has changed, a 
stud 5 carried by it falls under a recess 1 9 on a pendent 
lever carried on a stud at 18, so that the long lever there- 
fore becomes locked in this position. At the same time as 
the carriage runs in, the lever centred at 12 is free from 
contact with the stud 8, and consequently the weight is 
now only supported by hanging from the end of the long 

The carriage now approaches the roller beam, and as it 
does so an incline H' comes against a stud bowl H on the 
drawing-up lever and lifts the drawing-up cone out of 
gear with the scroll shaft, thus stopping the carriage ; at 
the same time the fallers come against the projection 16 
on the lever centred at 18 and release the recess 19 from 
the stud 5 on the long lever. There is now no resistance 


to the movement of the long lever, so that the weight 14 
hanging on one end of it at once falls, and in doing so 
causes the catch box on the back shaft to be put into gear, 
thus connecting the front roller with the back shaft ready 
for the run-out, which immediately commences. 

It will be noticed that it has not been considered 
necessary to go into detail as to the precise action of the 
various changes, these already having been thoroughly 
described and explained in the previous pages ; to those 
who understand the actions of the mule the drawing given 
will be practically self-explanatory. 


From page 135 to page 165 will be found a veiy com- 
plete description of the mule shaper, together with a full 
explanation of its principle. The short sliaper, however, has 
not been mentioned, though it ma}' be remarked that the 
greater part of the explanation is equally as applicable to 
the short as to the long shaper. An illustration is here 
given of the short shaper, and the following remarks will 
be sufficient to enable its working to be clearly xmderstood. 

A section of the carriage square is shown in Fig. 220 ; to 
the under side of it is bolted a strong framing in the form 
of a slide cover E. Into the grooves of E there is fitted a 
slide Q, as shown in the section. To the slide Q is con- 
nected a short rack >S, and into this rack the small pinion T 
gears ; on the boss of the wheel T is a larger wheel U, the 
two wheels T and U thus forming a compound carrier which 
runs on a stud carried by a bracket from the carriage squaie. 
The large wheel U now gears with the teeth of a long rack 
which is fastened to the floor, so that when the carriage 


moves backwards and forwards the wheel U will revolve 
by virtue of its being in gear with the long rack V. As 
U revolves so will the pinion T ; but T being smaller than 
U, it will only cause the rack S to move a proportionate 
distance to the carriage that the jjinion T is smaller than U. 
If T has 13 teeth and U has 43 teeth and the carriage 
travels 60 inches, then the rack S, and consequently the 
slide Q to which it is attached, will move -^-^ — =-18-1- 

4.3 ' 

inches, and, moreover, this movement of the slide will be 
in the opposite direction to the carriage. On a projection 
to the lower pai't of the slide Q rests the shaper, com- 
posed of the back, middle, and front plates ; these plates 
are connected to the slide Q by the nut M working on 
the screw which is carried from the slide by the brackets 
N and 0. 

Upon the shaper plates rest the shaper rail and shell 
through the pins A and B, the shell CGH being loose from 
the rail K for the purpose of adjustment. The pin B, in 
addition to resting upon the back incline, also projects into 
a vertical cut into the slide Q. From this description we 
can now see that any movement of the carriage will cause 
the slide Q, the shaper plates, and the shaper rails to be 
moved in the ojjposite direction, but to a less degree ; in 
other words, the shaper rail moves forward under the 
shaper bowl as the carriage runs in, and whilst the carriage 
travels 60 inches inwards the shaper travels 18 inches 
outwards, both movements occupying exactly the same time. 

As the carriage moves in, the end X of the slide Q 
comes against the lever Y carrying the catch Z, and this, 
gearing with the ratchet wheel P on the end of the shaper 
screw, moves the shaper plate so that the shaper rail is 
lowered and the coji is lengthened. 

For various purposes there arc several points of adjust- 





ment : for instance the ratchet wheel P can be changed ; 
the number of teeth taken can be regulated ; the position 
of the pins A and B can be adjusted ; and l)y means of the 
wheels T and U the distance moved by the shaper rail can 
be made to suit any diameter of cop required. 

When the pin A is at 1 and is set over 3 on tlie front 
plate, and the pin B is at 7 and set over 8 on the liack 
plate, twist cops can be made. When the pin A is at 2 
and is set over 4 on the front plate, and the pin B is at 6 
and is set over 9 on the back-plate pin, weft cops can be 
made. The starting-point in all cases for the shaper IdowI 
is at 5F, the finishing point for weft cops being at 6 and 
for twist cops at 7 ; the wheels T and U of course requir- 
ing to be altered to suit the stretch. The usual wheels 
used for changing are, for T 11 to 16 and for U 42 and 
43. Short shapers are used now only for very fine spin- 
ning mules, and their advantage lies in the fact that 
the stretch can be altered without changing the shaper : 
in the long shaper only one stretch can be made ; any 
variation from this would mean a new shaper. 

A further example of the Short Shaper is given in 
Fig. 221. Its connection to the copping-faller is clearly 
shown as well as other details connected with the locking 
and unlocking of the faller leg. 

Fine Spinning" Mule. — The following descriptions and 
drawings are given to amplif}^ the notes given on pages 
244 to 253. 

Backing-off Motion, etc. — A general view of the 
Single-speed Mule, Low Headstock, is given in Fig. 222. 
The backing- off motion is the chief feature illustrated. 
As the cai'riage completes the outward run the regulating 
bracket X is moved and the long backing-ofF lever, centred 
at Q, is unlocked. The end P of this lever falls and a 


ta 'I . r/l[lM l ,i!gilL ' f ' 

•famni vtiu XNOud 



recess in the rod L falls over the square stud M on the 
long backing-off rod L. The backing-ofF cam H, driven 
from the rim shaft through the worm, comes into contact 
with the swing J centred at K, and moves it forward. In 
so doing the slide L is dragged forward, and, through the 
lever at N, puts the backing-ofF wheel into gear with the 
cone clutch. In the meantime the carriage has been locked 
by means of the lever U locking on the square stud V on 
the square, and the down lever stud T is in contact with 
the strike finger W. During the locking of the faller leg, 
the strike finger W swivels downwards and depresses the 
down lever stud T. This action pulls down the long 
backing-ofF lever and locks it at S ; at the same time the 
slide L is released from the stud M, and, a spring pulling 
L backward, takes the backing-ofF wheel out of action. The 
depression of T also lowers U and unlocks the carriage, so 
the carriage is free to commence its inward run. A 
general idea of the other changes for moving the straps, 
etc., can be obtained by reference to the other detail draw- 
ings which follow. 

Setting-on and Drawing-up Motions. — The drawing. 
Fig. 223, gives a general view of the mechanism for these 
motions. Some portions are shown displaced from their 
correct position in order to bring them into view. 

A single rim shaft with one rim pulley is used, the 
double speed being obtained through gearing. C is the 


single-speed driving pulley, whilst A through — — — gives 

the increased speed. A bent lever centred at U carries a 
stud that fits into the three notches S, R, T, on the setting- 
on rod. The other arm of this bell-crank lever has an 
incline a which is, at the correct moment, moved aside 
by the cam driven from the worm t. The end of the 

H3iini <jn-ONiMvaa 




short arm has centred on it at c a rod d carrying the 
tumbler Y and safety catch X. The drawing-up rod is 
released by the swivel arm 11 on the carriage, coming into 
contact with Y when backing off, and so releasing the 
finger Z from the stop E, thus permitting the weight H on 
the gun lever to pull over the strap fork on to the fast 

Fig. 224. 

drawing-up pulley D. The carriage, as the inward run is 
being completed, moves the gun lever at F and puts the 
drawing-up belt on the loose pulley. 

For opposite side of tall headstock see Fig. 224. 

Backing-off Motion. — As the carriage runs out, the 
notch D, Fig. 224, is occupied by the projection C on the 
rod B. The backing-off cam K forces G forward round the 
VOL. Ill 2 F 



centre F, and so forces the rod B outwards and puts the 
backing-ofF wheel A into contact with the backing-ofF cone. 
When the faller leg is changed, a lever moves into contact 
with the incline M and lifts the lever L, thus unlocking D 
from C. The spring now pulls the backing-ofF wheel out 
of gear. At the same time, the carriage is released from 
the catch Q by virtue of tlic link rod Avhich connects the 
two levers L and P. 

Part of the roller gear rod is shown, but the back 
mechanism of the headstock is so similar to that shown in 


Fig. 225. 

Fig. 222 that it has not been considered necessary to repeat 
it here. 

Twist Motion. — This twist motion, Fig. 225, is a 
detail of Fig. 223 ; it is very simply arranged, and is 
actuated from the worm A on the rim shaft which drives 
the worm wheel B on cross shaft. On the end of this shaft 
is the change twist wheel C ; this gears into D part of 
compound carrier, whilst E part of same drives a 72's 
wheel on twist shaft, and thereby gives a motion to this 
shaft of one revolution per draAV. 

The twist shaft carries three cams — W, S, and T. The 
cam W actuates the backing-ofF motion (see Fig. 221). 

Amtu nm iiov9 




The cam S is for setting on (see Fig. 223), whilst the cam 
T is for twist (see Fig. 223). 

Roller-delivery Motion. — This motion, Fig. 225, is 
driven from the twist shaft by means of wlieel G and 
ratchet wheel R, which are in one piece, and can only 
revolve when the catch or pawl is allowed to fall into gear 
by the action of the two cams on Avhich one end of the 
catch rests. These cams can be adjusted to give motion 
to the rollers, so as to deliver the necessary amount of 
yarn which is required. 

This motion is mostly used when spinning twist. 

Special Mule. Section of Rim Shaft (see Fig. 226). 
— The drawing is practically self-explanatory. The single 
and double speeds are obtained by making the rim shaft in 
two parts and using two rim pulleys as G- and X. Each 
length of rim shaft is driven by a separate pulley A and C. 
The pulley C is the one through which the single-speed rim 
G is driven, as well as the one through which the general 
gearing receives its motion. Pulley A drives the double- 
speed rim pulley X and also drives, through the worm N, 
the roller-delivery motion whilst twisting at the head. 
The backing off is actuated separately from the pulley H 
on the boss of which is a Avheel J. This wheel drives K 
almost continuously, but the backing-ofF cone wheel D is 
only effective in driving the rim shaft when D is moved 
into contact Avith the cone clutch E. 

The wheel M is the point through Avhich the general 
gearing receives its motion (see plan of gearing. Fig. 234). 
Brake Motion (see Fig. 227). — When backing off 
is about to take place, the belt is moved from pulley A 
to the loose pulley B. The backing-off cone Avheel D is 
now forced into contact Avith the cone clutch E, and at the 
same moment the front part of the rim shaft is stopped 


de;id through the cone clutch M being forced into contact 
with the fast pulley A. 

Setting-on and Drawing-up Motions, Figs. 230 and 

231. — A general view is given of the above motions in 
Fig. 230, whilst an enlarged view of the out end of the 
headstock is shown in Fig. 231. The sketch illustrates the 
disposition of the mechanism as drawing up takes place, 
and the main driving belt in on the loose pulley B. The 
belt on the drawing-up pulley D drives the scroll shaft and 
through it the carriage. The setting-on rod is locked in 
position by a square stud at S. This stud will be raised 
Avhen the carriage comes into contact with the end F of 
the gun lever centred at G, so that the setting-on rod will 
be at liberty to move the belt on to the fast pulley C ; the 
movement of F will also move the drawing-up belt on to 
the loose pulley. After this happens the lifting lever at 
A^ is depressed by lever and bowl on carriage square, and 
the balance lever M taken away from the pin P. 

As the carriage completes the run out, the carriage 
moves V forward and releases the stud S from the notch T, 
and puts in tension the spring between the two balance 
levers M and N. The tension of the spring moves the 
setting-on rod forward, and transfers the belt from C to A, 
and so puts the double speed in action. 

The twist latch is now released, and the setting-on 
rod moves backward and puts the belt on pulley C. The 
stud S is now in the notch T where it remains during the 
drawing out. 

In the event of the carriage overrunning the catch, the 
fuller would come in contact Avith finger Y, thus moving 
forward the safety rod ; this would then move the L lever 
u and cause lever r to raise the twist latch o, thereby 
moving the strap on to the loose pulley B. 

'tamnd dn-ONiMvug- 




The stop Z is used for lifting square stud at S through 
lever a when running single speed. 

During the drawing out, the drawing-up rod has been 
locked in position through the finger K resting on the 



FlO. 231. 

recessed boss E. At backing off the lever c is raised 
(see Fig. 231) and the finger K released, thus allowing the 
drawing-up rod to place the belt on the fast pulley I). 

The stud 6 comes in contact with faller, and can be 
regulated so as to allow a small portion of the strap to go 


on fast drawing -up pulley D, thereby preventing the 
carriage from starting up too quickly. 

The drawing-u]) rod can be released if necessary by the 
knee lever i. Its release can be prevented when necessary 
by turning over the lever Q, Fig. 231. 

Fig. 231 also shows at N and P a small lever which 
locks X to the drawing-up rod. If P is turned over, the 
tumbler X is free to move aside without moving the incline 
W, and consequently the carriage will come to a standstill 
because the stud S is not raised, and so keeps the setting-on 
rod locked in position. 

Drawingf-out Motion, Fig. 232. — The driving strap 
is on the fast pulley C and the back rim jiulley is driving 
the spindles. 

The rim shaft pinion (see A in the plan of gearing, 
Fig. 234) is driving the front roller, the catch box Y being 

The rim shaft pinion is also driving the back shaft 
through the closed catch box y. This is also clearly shown 
in the gearing plan, Fjlg. 234. 

We thus have the spindles, carriage, and front roller 
driven from the rim shaft through the pulley C during the 
outward run. 

Soon after the carriage has started on the outward run 
the bowl a on the back of the square depresses the end 
V of the lifting lever centred at S, and raises the other end 
T ; a link connects the end T to a lever I, centred at H, 
so that the lever I is raised together Avith the weight carried 
by I. The upper end of this weight carries one end of a 
spring, which end is attached to an arm U of a T lever 
centred at H. The T lever is locked in position at L, so the 
spring (by the lifting of the weight) is put in tension ready 
to pull the end U upwards when the T lever is unlocked. 



The lifting of the lever I also raises the two pendent 
links or gearing legs Q and E, the longer one Q locking 
itself on a square stud carried \>j a bracket fixed on the 
floor. This stud is shown in the drawing betAveen Q and 
E, but the floor bracket carrying it is not shown. 

As shown in the drawing, the end AV of the T lever is 
coupled to the roller gear rod Z, on a reduced portion of 
which rests a bowl X carried by the lever Mhich actuates 
the catch box Y. The catch box Y is therefore locked so 
long as the T lever is locked in its present position. 

The end J of the T lever carries a weight K Avhich rests 
on a bracket L fixed on the headstock back. 

Ratching, Jacking- or After-stretch Motion.— This 

motion is one that stops the front rollers before the stretch 
is completed, but enables the carriage to complete its 
outward run. As the carriage nears the termination of 
the outward run, a projection or finger on the front of 
the square comes in contact with a finger c on the long 
gearing rod and moves the rod forward. A projection M 
near the back end of the rod bears against the weight K 
and moves it off the stud L. This action causes the weight 
K to fall, as Avell as allowing the spring to pull up the 
end U of the T lever. The etfect of the change is to cause 
the end "\V to move the roller gear rod forward and so put 
the catch box Y out of gear, thus stopping the rollers, and 
the same change takes the catch box y out of gear and 
puts the catch box x into gear. At the same time the 
strap moves from pulley C to pulley A, thus putting the 
spindles on to increased sj^eed, the rollers are stopped, and 
the carriage is now driven through the jacking wheels and 
the catch box x for the remainder of the outward run. 

Assistant Winding Motion, Fig. 233. — This motion 
is arranged to prevent snarls forming in the loose yarn at 



the moment the fallers change on the completion of the 
inward run. The amount of yarn set free varies through- 
out the set, and as it is beyond the control of the quadrant, 
a pulley A keyed to the rim shaft is utilised so that at the 
precise moment required the belt on B is moved on to the 




















Fig. 233. 

pulley A and so drives the rim shaft and, consequently, the 
spindles for a fraction of a minute, thus winding on the 
yarn that would otherwise be free to form snarls. The 
motion, it will be observed, is not one that allows snarls to 
form and then stretches such snarls out : its advantage lies 
in the fact that it prevents snarls forming on spindle. 

The motion is extremely simple. The biacket J is 
fixed to the carriage, and near the end of the run-in J 

fff 'AHTindlAlia 

'fSntnd dfi-ONiMvao 

a AamndiMW.. 



comes into contact Avitli a pendent lever centred on a rod 
at H. The lever is raised (as shown in dotted lines) and a 
projection on it comes into contact with the rod and raises 
it, thus allowing a stop finger G to pass through a slot in 
a fixed bracket M. A weight N on a lever centred at E 
pulls over the end F and forces forward a slide D on the 
other end. The slide D carries the strap fork so the strap 
is moved from pulley B to pulley A. The amount of strap 
allowed to move on to A can be carefully adjusted by 
the stop rod or adjusting screAV passing through a fixed 
bracket K, the adjustment being effected by a handle or 
nut L conveniently reached by the minder. The pendent 
lever at H can be set so that the carriage can actuate it 
at any required distance before finishing the inward run. 

Jacking Motion, Fig. 235. — This motion is for the 
purpose of driving the carriage during the last few inches 
of the outward run, after the rollers are stopped to stretch 
the yarn, as is customary in spinning fine counts. The 
speed wheel carrier, bevel A, drives the front roller through 
the bevel C and the catch box B and an internal disc which 
is keyed on the front roller shaft. "When the catch box is 
put out of gear the front roller is stopped. The bevel C 
is keyed on the boss of the jacking box D and E, and runs 
loosely on the front roller shaft. Inside this jacking box 
two pinions F and Gr are mounted, which are keyed 
together, but run loosely inside the box. These pinions 
gear with two Avheels H and J, H being keyed on front 
roller shaft and J on the long boss K. By reason of the 
wheels F and G being of different sizes, and being carried 
round the outside of the wheels H and J by the jack box, 
motion is transmitted to the Avheel J, which, being keyed 
on the boss K, on the other end of Y»'hich is secured the 



wheel L, drives the back shaft in the ordinary way hnt at 
a reduced speed. 

This motion can be operated to cause any desired 
amount of stretching of the yarn. 



Strap Relieving and Locking Motions, Fig. 236. — 
The object of this motion is to move the strap from the 
fast to the loose pulley when the carriage is within 2 inches 
to 12 inches of the completion of the outward run, the 
momentum of the carriage at this stage being sufficient to 
complete the outward run. 

Fixed on the carriage end is a frame B carrying a stud 
A. As the carriage moves outwards this stud comes in 
contact with the inclined surface of lever D, which it 
depresses, and, being centred at E, the swivel F is taken 
with it. To this swivel F the end of the rod G is secured, 
the other end of which is connected to the lever H, which 
in turn is secured to the lever J. This lever is centred at 
K and the strap lever is secured to it. 

It will be seen that if lever D is now pressed downwards 
the rod G will be pulled outwards, which, through the 
various levers, will operate on the strap fork lever and so 
bring about the desired change. 

The moment at which this motion comes into operation 
may be regulated by sliding the stud A in its frame B, the 
stud being held in any desired position by turning down 
the catch C into one of the holes drilled in the upper 
surface of the frame B. 

The locking device is for stopping the carriage at any 
part of the outward run, and operates as follows : — When 
the lever D is pressed down, the lug L, which is cast upon 
it, is caught by the catch M, thus preventing the levers 
resuming their original position, and keeping the carriage 
stationary until released. When it is not required to stop 
the carriage the catch is turned back to tlie position shown 
at N. 

Twist Motion. Driven from Tin Roller, Fig. 237. 

— This motion is designed to put the required twist in 



every (iraw alike. Tlie motion, being driven by a worm 
on the tin roller shaft, ensures this shaft, and therefore the 

spindles, making the same number of revolutions each 

The twist catch J is hinged at the back of the headstock 
as usual, and is connected by the rod I to the bell-crank 
VOL. Ill 2 a 

4 so cor TON SPINNING 

lever G, which is pivoted on a Ijrackct secured to the head- 
stock side. A bracket is fixed on the square which carries 
the twist motion wheels C and B, and to this bracket a 
slide is fitted carrying the twist wheel D. This slide can 
be adjusted to take any size of wheel from 50 to 100 teeth. 
It will thus be seen that the worm A on the tin roller shaft 
transmits motion to the wheels B and C, which in turn 
drive the change wheel D, whilst on the same stud, and 
secured to wheel D, is a finger E revolving with wheel D. 

When the carriage has completed the outward run, the 
tin roller shaft continues to revolve until the finger E comes 
in contact with the bell-crank lever G, which, being turned 
on its centre, exerts a pull on the rod I, and thus lifts the 
catch J, allowing the strap lever K to move the strap on 
to the loose pulley on the rim shaft previous to the mule 
backing off. 

Backing-off Motion, Fig. 238.— The object of the 
above motion is to unwind the coils of yarn formed on the 
spindle blade during spinning, in order that the spun yarn 
may be wound on the cop, and this is done by turning the 
S2)indles in the opposite direction, the slack yarn thus 
formed being taken up temporarily by the counter faller. 

The backing-off wheel A is mounted loosely on the rim 
shaft and is driven constantly in an opposite direction to 
the rim shaft during spinning. When the carriage com- 
pletes its outward run the lever K, which is mounted in 
the square, depresses the lever J, which moves the rod F 
and allows the spring G to turn lever I) on its centre and 
so force the backing-off wheel, the inside of which is turned 
conical, on to the friction cone connected with the fast 
pulley and thus driving through to the spindles in the 
usual manner but in the opposite direction. 

The winding faller is pulled down by the backing-off 



chains in the usual manner — through a click wheel on the 
tin roller shaft — until the bottom end of the boot-leg N 
rests on the locking bowl connected with the slide on the 
shaper rail ; in which position the fallers are said to be 
"locked." During this operation the lever K is raised, 
through the levers L and M, and the lever J is released, 
allowing the spring H to draw the backing-off Avheel out 
of gear. 

To prevent this motion coming into operation too soon, 
a finger C rests on a bowl E connected to the strap lever, 
and prevents it dropping until the cam is changed and the 
strap is moved from the fast to the loose pulley, thus 
preventing the backing-off friction going into gear before 
the strap is moved entirely on to the loose pulley. 

Whilst backing off, the button-head on rod F must be 
\ inch clear of the long lever D. 

Gearing Plan of S.-A. Mule.— In Fig. 239 is given 
a plan view of the gearing of a S.-A. Mule that is becom- 
ing more widely known, and so will be interesting to 
students. It may be remarked that the copping faller 
ought to have been shown thicker than the counter faller. 

CO o<l 

O w 



Gassing. — All yarns are made up of fibres of varying 
lengths, within the length of the staple being used, no 
matter what care has been taken to eliminate the short 
fibres. Further, a number of unstraightened fibres of all 
lengths are to be found in all yarns, even in the best 
combed cottons. In the spinning process, the vibratory 
action causes the ends of the fibres to stand out from the 
surface of the yarn. This roughened state of the yarn 
reduces its lustre owing to the diffusion of light. Also, 
the roughness destroys the impression of a smooth, round, 
and compact yarn. By burning oft' these projecting fibres, 
the lustre is restored and the yarn has a smoother and 
more compact appearance. As the projecting fibres do not 
add to the strength of the yarn, but rather increase its 
bulk, uselessly for many purposes, it naturally happens 
that yarn, after being gassed, is of a finer counts than before 
being gassed. 

Fisrs. 240 and 241 are rough sketches of single and 
double yarns respectively, showing the hairy condition of 
yarns. In the case of the single yarn. Fig. 240, it will 
be seen that practicatly the whole surface of the yarn -will 
come under the influence of the flame, and all the out- 
standing fibres will be burnt off. In doubled yarns the 
whole surface of each individual yarn is noi exposed to 




the flames, and so the amount burnt oft' is not as much as 
in the single yarn ; this can readily be understood from 
the sketch. 

The amount of projecting fibres will vary considerably 
according to the kind of cotton, its preparation for spinning, 

Fig. 240. 

Fic. 241. 

and the amount of twist put into the yarn at the spinning 
process. Soft twisted yarns, say for mercerising, will be 
more hairy than hard twisted voile yarns. The amount 
burnt off will therefore be a variable one on these grounds. 
In addition to this variation, however, a further increase 
or decrease will occur according to the heat of the flume or 
the length of time the yarn is under the influence of the 


flame. No hard and fast lines can be laid down on this 
percentage of loss, the range usually being from 5 to 9 per 
cent in weight and a corresponding increase in the counts 
of the gassed yarns. 

A few examples are given of the counts of yarns to be 
used in order to obtain given counts of gassed yarn : — 

Ordinary 56/2 becomes 60/2 gassed = 7 "1 per cent loss. 
74/2 ,, 80/2 „ =8-1 

94/2 ,, 100/2 ,, =6-3 
65 „ 70 „ =7-6 

Hard twisted single and doubled yarns, of course, will 
not result in large losses of this kind owing to the smaller 
amount of projecting fibre. 

Formerly only doubled yarns were gassed, and these 
were usually doubled on the wet doubler so that projecting 
fibres were fewer owing to such fibres, in their wet con- 
dition, lying in close contact with the body of the single 
and being twisted up with the rest of the fibres when 
doubled. The gassing of single now forms an important 
element of the trade, no doubt due to lietter cotton and 
more careful methods of preparing and spinning. "Whilst 
recognising the usual custom of the trade and methods of 
arriving at the loss due to gassing, it is as well to point 
out that the loss, in most cases, is not simply due to the 
amount of fibre burnt off. Testing for counts before 
gassing is done on yarn containing moisture, to an extent, 
in most cases, up and even above the regain moisture. 
Testing for counts after gassing is frequently done long 
before the gassed yarn has recovered and reabsorbed its 
previous amount of moisture which it has lost in passing 
through the hot flames. From the purely manipulative 
point of view this loss is not of importance and is ignored 
by custom, but economicallv its importance ought to be 


recognised and the carelessness associated Avith it elimi- 

Since the object of gassing is to free the yarn from its 
outstanding fibres without damaging the body of the fibres, 
the strength of the yarn will be maintained. This means 
that if a 70's yarn is gassed and becomes 75's, this 75's 
gassed yarn will be as strong as the original 70's yarn. 
From this fact it is frequently asserted that by gassing 
yarns we obtain stronger yarns ; this, of course, is purely 
relative, and even to obtain this result requires great care. 
Most firms are content if they can maintain the strength 
of the original yarn ; any reduction in strength naturally 
means that the body of the fibres has been injured. 

Gassed yarns are used for a variety of purposes, among 
which may be enumerated the following : — Sewing cotton ; 
lace ; embroidery ; poplins ; Venetians ; voiles ; crepes ; in 
borders of fabrics for India and special effects in a variety 
of woven materials ; mixing with silk ; mercerising for 
hosiery, fancy cottons, crochet cottons, etc. Var-ious 
defects arise during the process of gassing. These may 
be general or local. Ungassed yarn may be produced by 
some or all of the lights going out ; strong drafts or even 
the banging of doors may cause this. Too much air 
admitted to the burners may result in lights going out. 
In piecing an end or putting in fresh bobbins there will be 
a short length of ungassed yarn put through. 

Over-gassed yarn, of course, will weaken the yarn and 
darken it in colour. It may be caused by too slow a speed 
through the flame, too strong a flame, or too man}'' passages 
of the yarn through the flame. Sooty yarn is caused by 
yarn passing through a flame that contains a white portion 
due to careless adjustment of the mixture of air and gas, 
or to a change in the character or even pressure of tlie 


gas, causing incomplete combustion of the carbon in the 

Dirty yarns may be caused by soot from the flame, Itut 
more frequently it is caused by carelessness in handling 
the yarn, as the process is itself dirty, and there is always 
more or less of burnt fibre lying about. Dirt also accumu- 
lates in grooves of bowls, etc., and this comes off and stains 
the yarn in patches or even long lengths. 

Streaky and striped yarns. These faults are the most 
common ones in gassed yarns. They consist chiefly in 
variations of shade indicating that speed or heat has not 
been uniform. A general difference of shade or colour 
throughout the frame Avould suggest an alteration in the 
gas pressure, a condition that frequently occurs during the 
day, in almost all gas undertakings. Another cause is to 
be found in the partial choking of a burner by burnt dust 
particles and the consequent loss of heat. Burners of the 
ordinary bunsen type are more likely to cause this than 
those fitted with a pressure supply of mixed air and ga^. 
Irregular passage of the yarn through the burners or 
variations in the adjustments of the bowls will produce 
streakiness, and sometimes the tenter may have carelessly 
threaded the yarn, one more or less traverses over the 
bowls, and so caused an increased or decreased amount of 

Gassing Frame. — Fig. 242 shows the section of a 
horizontal gassing frame with a quick traverse and 
winding from bottle-shaped bobbins. One side of the 
machine shows a flannel drag P, whilst the other at E 
has the wire drag. Special provision is made for carrying 
away the vitiated air and burnt products due to the gas 
and the burnt fibres. This consists of a cased-in receptacle 
^y, running the full length of the frame, and containing 



openings indicated by the arrows. A fan connected to an 
extension on the outlet L, or a fan placed in a suitable 



position in the v/all of the room, carries away the foul 
air. A supply of fresh air is provided at J of sufficient 


capacity to prevent any currents that might interfere with 
the lights or health of the operatives. The guides G are 
unusually light and noiseless in action, and consist of two 
wires, to w^hich the guides are fastened. The wires are 
supported at intervals along the frame in brackets, which 
act as guides to them. Only sufficient of the machine is 
shown to illustrate the features already mentioned, but it 
will be understood that it is provided with mechanism for 
moving the burners aside instantly, when the cheese is 
drawn away from the drum. In regard to the gas used, it 
is now usual to force air into the gas pipe by a fan 
arrangement, in preference to the common bunsen flame 
method. Several systems are in oj)eration for mixing air 
and gas in correct proportions, and most firms who 
specialise in this class of machinery have their patented 
system applicable to the various kinds of gas that can be 

The production of the machine described will naturally 
vary according to the speed of drum. As an average it 
may be considered that 93 hanks per drum in ten hours 
can be obtained when the drum runs at 240 revolutions 
per minute, and 38 hanks per drum in 10 hours with a 
drum speed of 100 revolutions per minute, the inter- 
mediate productions being in simple proportion to the 
drum speed. About one H.P. is required for a frame of 
160 lights. One man can attend to 160 lights Avhen 
gassing from bottle-shaped bobbins, or 80 lights when 
gassing from cops. 

Fig. 243 gives a sketch of a vertical burner and split drum 
gassing frame. This type of machine has been growing 
in popularity, and several designs are on the market. 

The main feature consists of a gas tube F, perforated 
with a series of small holes, thus forming a vertical line of 



flame. The yarn is led from the bobbins B upwards, and 
guided through the centre of the flame, and is then passed 



■ i 



Fia. 243. 

over wire guides and led downwai'ds over guides Z on 
through the split drum D to the cheese C. 

The main gas pipe is shown at M, and to this is con- 


nected the flame tube F. This tube is enclosed in a casing 
whose upper end opens into a casing W, extending the 
full length of the frame, so that all the products of com- 
bustion can be carried away. Fresh air is introduced in 
a similar way to that already illustrated in connection with 
another machine. On examining the sketch it will be 
noted that the yarn is guided in its passage through the 
flame. These guides are carried by a light framework 
P, having a rack extension, into which a quadrant tooth 
segment Q gears. This quadrant has an arm connected to 
a rod A, which in its turn is connected by lever to the 
arm which carries the cheese C. 

When an end breaks, or the arm N is moved away 
from the drum, the rod A is raised, and this has the effect 
of at once carrying the yarn bodily away from the flame. 
On restoring the cheese to its running position, the yarn 
is drawn back into the flame. 

Hitherto the use of split drums for winding have had 
the disadvantage of causing a constantly varying tension 
in the yarn. This can readily be seen on reference to the 
two diagrams in Fig. 243. As the yarn enters the drum D 
at 1, it must pass diagonally across the drum, and emerge 
on the cheese at 2. When the drum has made half a 
revolution it will be as shown in the right-hand diagram, 
and the yarn entering at 3 will pass through the drum 
and emerge on to the cheese at 4. It will be seen that in 
passing from the angular position, 1 to 2, to the straight 
position, 3 to 4, there will be some slack yarn, and of 
course some tight yarn for the other half of the revolution. 

An ingenious device has been applied to overcome this 
difliculty, which consists in fitting within the dinim a 
specially shaped case E, which revolves Avith the drum. 
This is so formed that it takes up the slack exactly as it is 



formed, and of course releases it as the yarn becomes 
tight. With such a device as this, there is no longer any 
need to ignore split drums as a winding factor, especially 


Fig. 244. 

for gassing. The wire G is used to automatically place 
the yarn in the split of the drum. 

Upright Spindle Winding Frame.— An illustration 
was given in Fig. 180 of this type of machine. Another 
example is now given in Fig. 244 that embodies recent 


improvements. The drawing has been purposely made 
composite in order to show that double-flanged bobbins or 
bottle-shaped bobbin may be wound from cops, ring and 
doubler bobbins, or from hanks. 

The upright spindles A are provided with wharves, and 
are driven from the tin roller D. Single or double row 
of bobbins may be built. The clearer motion and guides 
are carried at the top of the rack R, operated from the 
building motion through the wheel W. 

For gassing and reeling the bottle-shaped bobbin is 
now the recognised form, but the building motion is so 
designed that the ordinary parallel shaped bobbin can be 
built by simply unhooking a chain. A creeper motion to 
carry the empty bobbins to the end of the frame can 
readily be applied in cases where the machine is run con- 

The drag varies according to the class of winding being 
done. Flannel is usual when winding endwise as at F, 
but it may be pointed out that this is not a satisfactory 
method owing to the very roughening effect it has on the 
yarn. Drag bands are used on self-contained spindles, 
and weights are hung from the barrel in the case of swifts. 
Makers of these machines will apply brushes, flannels, or 
ball clearers if required, but it is advisable to avoid both 
flannel and brushes as drag or clearing factors if good 
work is desired. 

A typical form of bottle-shaped bobbin is shown in 
dotted lines at X. Its chief advantage lies in the fact that 
it can be wound off" endwise without ravelling. The 
production varies according to the degree of clearing 
required, and also how the yarn is drawn — off cops, 
bobbins, and hanks. For cops and bobbins endwise an 
average of 360 hanks per spindle per 56 hours ; sideways 



240 hanks per spindle per 56 hours, and for hanks 220 
hanks per spindle per 56 hours. 

Quick Traverse Winding Frame. — In Fig. 245 we 
have another example of a qiiick traverse winding frame. 
Any number of ends up to 24 can be wound either on 
paper or wooden tubes. As shown, a stop motion is 

Fig. 245. 

applied, but of course it can be used without this device. 
The passage of the yarn from the cops C through tlie drag 
and the hook T now turns upwards to the top of the 
creel, where it passes over bowls or through guide wire, 
and returns in a downward direction to the guide G, 
which leads it to the drum D. The left-hand side shows 
the position of the stop motion mechanism when Avinding 
is taking ])lace, whilst the right-hand side shows the 
VOL. Ill 2 H 



changed positions taken up Avlien an end breaks. It will 
be noted that the guide Avire T is carried by an extension 
of the catch L, so that on this wire dropping when an end 
breaks, it comes into contact with the revolving spider 
shaft S, and the catch L is immediately forced away from 
beneath the lever H, Avhich it had supported. The lever 
H falls at once and in doing so raises the end K against 

an inclined lever B, which in its turn forces the lever A 
away from the drum D, and at the same time moves up 
the brake lever into contact with the cheese. 

The weight W keeps the cheese pressed against the 
drum D, so that on piecing an end, all that is necessary to 
restart the winding is to lift the end of H, and the catch 
L slips under it. Adjustments are provided for obtaining 
instantaneous action, and also for regulating the drag, 
especially on the flannel F. 

A sketch of a ball drag is given in Fig. 246, as applied 



to above frame, and associated with it is given a clearer 
view of the guide wire that operates the stop motion 
thi'ough the catch L. 

Fig. 247 shows the driving of the quick traverse winding 
frame. By changing the wheel A, a wide range of ratios 


E *3i'*! 





Fio. 247. 

can be obtained between the drum speed and the cam 
speed from 1'25 to 1 up to 56 to 1. 

In Fig. 248 a section of a reel is shown to illustrate the 
creel used when bottle-shaped bobbins or cheeses are 
beinsr used for hank winding. The small sketch in the 
upper right-hand corner of the illustration is an alternate 
arrangement of guide to that given in the general sketch. 

Roller Settings. — The usual mill practice in setting 



the front and middle rollei's in the mule and ring frame is 
to set them within the length of the stajjle. This means 
that the distance between the centres of the front and 
middle rollers must be somewhat less than the presumed 

■ r^nr 

length of the fibres of the cotton being worked. This is 
generally one-sixteenth to three-sixteenths of an inch less 
than the staple. In the preparing machinery of the card 
room, such as draw frames, flyer frames, etc., where 
drafting between rollers is performed, the front and middle 


rollers are set outside the length of the fibre, i.e. the distance 
between the front and middle roller centres is grc(ttcr than 
the length of the fibre. In the card room the cotton is 
principally drawn or attenuated in the space between the 
grip of the rollers, and in the spinning room the drawing 
action is presumed to take place by the front roller drawing 
the cotton from the grip of the middle roller. 

A number of peculiarities and difficulties arise during 
the progress of the cotton from the card to the spinning 
spindles, not the least of which is the apparent introduction 
of irregularities. Most of these difficulties would appear 
to be traceable to roller settings and the draft associated 
with these settings. 

A brief review of the subject is given in order to show 
the connection between draft and roller settings. On 
examining a card web, the fibres composing it are in a 
very unstraightened condition, and are curved and crossed 
and bent around each other in every imaginable direction. 
Very few fibres appear straight. The first problem to 
solve is to find the length of the filjres. This, of course, 
was done by the manager when the cotton was bought, 
and he adopted the usual course of testing the cotton by 
hand-pulling between finger and thumb. This action 
naturally straightens the cotton and practically pulls away 
most of the short fibres, thus leaving a tuft of the straight 
full length fibres. The same test is applied to note the 
length of the fibres in the card web as well as in other 
subsequent processes. In other words, Ave always obtain 
our idea of the length of the fibres by straightening the 
fibres, judging them in this condition, and setting lollers 
to the lengths based on this judgment. 

Since the setting of rollers and the drafting is so 
important, it is as well to observe the actual condition of 


the fibres in the web of a card and see whether length of 
fibre, as usually considered and estimated, is a good basis 
to work upon in setting rollers. 

A sketch of a number of fibres is shown in Fig. 249, 
which represents a portion of a card web. The fibres are 
made purposely of about equal lengths, in order to show 
almost ideal conditions of opening and cleaning at the card. 
Since this piece of web is gathered up into a sliver at the 
trumpet guide at the calender rollers of the card, it is clear 
that the arrangement of the fibres will be, at least, no 

Fia. 249. 

better in the sliA^er than in the web. (As a matter of fact, 
any bent fibres become more bent as they form into the 
sliver.) The sliver has now to be drawn out by passing 
between successive lines of rollers whose surface velocities 
increase between each pair of rollers. Tlie length of the 
fibres, since they are all about equal, can be judged by the 
fibres lettered E, F, and G, and this length may be taken as 
equal to the distance between the lines A and B. 

It is clear, however, that the length of the fibres, even 
in such an ideal set of fibres as shown in Fig. 249, cannot 
be considered equal to their straightened lengths, so far as 
setting rollers is concerned. The fibre F, for instance, is a 


47 > 

full length and straight fibre, but its position for roller 
drafting almost reduces its length to nil. The fibre E is 
the other extreme of position. All the other fibres occupy 
intermediate positions, and it is a judgment of all these 
positions that must be formed, in order to set rollers to the 
length of the staple. If it is now recognised that in the 
actual web we have all the peculiarities of shape and 
position of the fibres as shown in Fig. 249, and also that 
we have fibres of all lengths from, say, \ inch to the full 

length fibre F, it will not be difficult to understand that 
length of fibre, in the setting of rollers, is of very little 
importance. Our mills, however, do work on the length 
of staple, so let us examine Avhat happens. 

As a very simple illustration consider the cotton as it 
passes between the cages and calender rollers of a scutcher. 
Fig. 250 will illustrate the position. Eelative to the 
length of the fibre, the distance betw^een the calender 
rollers A* and the cage rollers at B is considerable. In 
order to maintain the continuity of the cotton between 


these two points the calender rollers have a surface speed 
greater than the surface sjjeed of the cage rollers, in other 
words, there is a draft between the two sets of rollers. 
This draft, however, is extremely small, and is not in- 
tended as a draft in the strict sense of the word, it is 
merely a carrying draft. Small as this draft is, its effect 
can be noted, in some machines and on some cottons, if 
carefully observed. The point, however, to emphasise is, 
that if this carrying draft is increased, the inevitable 
consequences would be that a quantity of fibres would be 
dragged forward from among their fellows, and gaps and 
ultimately breakages would occur in the layer of cotton 
between A and B. The maintaining of the surface speed 
of A at the lowest possible excess over the surface speed of 
B is an absolute necessity in order to prevent the intro- 
duction of great irregularities in the lap, and thus destroy 
a large part of the effectiveness of the regulating mechanism 
of the scutcher. The next step to note will be rollers of 
the card room. All drafting rollers in the card room, 
almost without exception, are set apart a little beyond the 
length of the normal straightened fibres, so that they are, 
in reality, set a considerable amount apai't in excess of the 
actual length of the fibres as they exist in the sliver, etc. 
From this condition it will be seen that between all drafting 
rollers in the card room there are spaces where considerable 
amounts of loose fibres exist. Rollers in these machines 
are heavily weighted, so that tlie fibres are drawn apart 
from each other in the spaces between the rollers and not 
from the grip between the rollers. Some of the fibres are 
straightened by this action, through the resistance to 
movement offered by the bulk of the fibres, but the very 
large numbers of fibres are drawn forward in an Ittenuated 
form in the same uustraightened condition as in the card 


sliver. At the same time, it must be noted that the draft 
between the most widely separated rollers, viz. the back 
roller and the next to it, is always small compared with 
the draft between the more closely set rollers, viz. front 
and preceding roller. Small draft and wide setting appear 
to be closely associated in our cotton spinning systems. 
It is frequently asserted that the draft between the back 
roller and the next following it is a carrying draft only. 
It is certainly small, relatively, but it is certainly far above 
a carrying di'aft considered in relation to the condition 
of the fibres and the distance apart of the rollers, and it 
requires but a glance at the movement of the fibres to see 
that the draft is a very effective one, small as it is, in dis- 
turbing the arrangement of the fibres and carrying through 
all manner of unstraightened fibres to the next pair of 

Now just as we saw in the case of the scutcher (Fig. 
250) that the draft must be kept very small indeed, so in 
the back draft of the card room rollers the draft must also 
be kept small if irregularities are to be eliminated or even 
to be kept from increasing. ExjDerience, however, supports 
reason in proving that the draft between the back and the 
following rollers is excessive, and introduces irregularities 
by tearing and dragging groups of fibres apart and carrying 
them bodily forward. 

When we come to the front and preceding rollers, these 
are set closer together, but still in the card room they are 
further apart than the straightened length of the fil)res, 
and, of course, there must, of necessity, be a lot of loose, 
free, and unstraightened fibres lying between the grips of 
these two pairs of rollers. It is between these two rollers 
that the bulk of the draft occurs, but a point for the student 
to observe is the fact that this draft is never very much. 


If the draft is made excessive, it would immediately show 
itself by spewing out the unstraightened fibres from the 
nip of the front rollers and even curling up many of the 
fibres that had previously been straightened. All this 
leads to the conclusion that drafting is very limited in the 
machinery of the card room, so far as drafting rollers are 
concerned, because of the existence of large quantities of 
curled and unstraightened fibres that lie between the grips 
of the rollers in all stages of the drafting processes. This 
limitation is compensated for by increasing the number of 
machines in order to bring about the required attenuation 
of the fibre. In spite of this, and even as a consequence 
of it, irregularities are increased, due to the small but yet 
excessive drafts. 

When cotton is combed two main objects are attained. 
The fibres are subjected to the straightening action of the 
needles, and the needles remove a quantity of the un- 
straightened fibres in the action that straightens many of 
the fibres that are left. This removal of unstraightened 
fibres is the main cause of improved appearance and feel 
of combed cotton. Comber waste contains all lengths of 
fibres, but they are the curved fibres, and almost always 
suggest simply short fibres, which is quite contrary to the 
actual conditions of the waste. The combed sliver also 
still contains quantities of unstraightened fibres and also 
fibres of various lengths, so that these are a bar to excessive 
drafts in card room machinery, and even the drafts that 
are used, with j^resent settings, produce irregularities. 

On arriving at the mule, the rovings are still subjected 
to the preliminary small draft of the card room method 
between the back and middle rollers with their wide 
settings, but a great change is to be noted between the 
front and middle rollers. Here the rollers are set within 


the presumed length of the fibres, so that there is also a 
presumed state of the fibres being held in the grip of both 
pairs of rollers. Under this new condition of setting, 
combined with the thin condition of the roving, the draft 
between the two rollers (front and middle) can be almost 
any amount. If the middle roller is weighted the draft 
cannot be taken beyond a certain amount, otherwise the 
presence of the short and unstraightened fibres will simply 
ooze out at the nip or in any case break up the roving into 
irregular patches and be incorporated in the yarn as such ; 
they are found in the very best yarn. With self-weighted 
middle rollers, greatly improved results are obtained, and 
higher drafts can be used as the fibres can be drawn from 
the nip of the rollers, but even then there is a certain 
amount of free space for the crumpled-up fibres that are 
in the roving, and these are mostly dragged bodily forward 
by the high draft and show as irregularities in the yarn. 
Extremely light middle rollers, made by reducing the 
diameter of the top roller or using a lighter metal than 
iron, will facilitate the use of higher drafts or improve the 
yarn. An improvement in our spinning mills is fore- 
shadowed in this attempt at an explanation of the drawing 
action of rollers, viz. : the use of drawing rollers of small 
diameter top and bottom in order to set as close as i)ossible 
well within the length of the staple; the use naturally 
of small top rollers, self-weighted when possible, and if 
weighting is necessary it must be of the smallest kind con- 
sistent with the thickness of the sliver, roving, etc. For 
mules and ring frames, the middle roller, in addition to 
being small in diameter, can be made lighter by using a 
lighter metal than iron ; even aluminium can be used. A 
greatly improved drawing effect can be obtained that will 
straighten the fibres, an increased draft can be used and 



distributed more equally among the rollers, a reduction 
in machinery will be possible, and a better yarn made 
from poorer cotton than is possible under our present 
system of drafting and roller setting. 

Costing. — The following notes are merely intended to 
give to the student a brief resume of how the price of 
yarn is obtained after cotton has been bought at a certain 

The term " margin " is a word frequently used in the 
cotton trade to represent the difference between the price 
of raw cotton and the price at which the yarn made from 
it is sold. 

The following table, representing a period of twelve 
months, gives these particulars, the prices being those of 
the date in each month : — 


Date. Egyptian 
per lb. 

60's Twist per lb. 
in pence. 

Margin per 
lb. in pence. 

Jan. 25 . . . 

Ql 3 

15 to 17 


Feb. 29 


151 ,, 17^ 


Mar. 28 


15i „ 17i 


April 30 


15f „ I7f 


May 29 


151 „ 17-2 


June 26 


151 „ IH 


July 24 


161 „ 181 


Aui;. 28 

10 1 

16 ,, 18 


Sept. 25 


16 ,, 18 


Oct. 30 


15i „ I7f 


Nov. 27 


16,V ,, 18 


Dec. 18 


16f ,, 18i 


It will be understood that all these figures undergo a 
variety of changes during the year and frequently during 
a single day, mainly due to the fluctuating price of raw 
cotton. Supply and demand have a strong influence in 
fixing the price of the yarn and cotton, and margins may 


be low or high, being more or less indicative of bad or good 
trade at the time. It may be stated, as a general rule, 
that this margin figure is used b}^ all classes of the trade 
as an indication of its prosperity or otherwise. Since the 
" margin " represents the difference between the price of 
raw cotton and the price of yarn into which the cotton is 
made, it follows that this margin includes the whole of the 
cost of running a spinning mill and also the profit on the 
business, if a profit is made. 

Very few businesses dealing with A^ery large quantities 
of material and having such a large turnover present so 
simple a problem as that of a cotton mill so far as getting 
out the costs is concerned. Most of the operations are 
performed on automatic machinery and paid for on piece- 
work rates based on mechanically operated indicators. 
Stocktaking is of the most simple character, so that it is 
possible, almost at any moment, to produce an analysis of 
the position of a firm. 

Assume a mill of 100,000 spindles, spinning 
60's counts, twist, carded, from Egyptian cotton. 
Capital £80,000. 

The production of this mill will be 23 hanks per spindle, 
or |-J = 0*3833 lb. per spindle per week. The output of 
yarn will therefore be 38,330 lbs. per Aveek at the spindle. 
To produce this yarn there are employed a variety of wage 
and .salary earning people ; also trade expenses, rents, rates, 
etc., and numerous items associated with the structure, 
machinery, accessories, transport, power jjlant, etc. Day 
wages and salaries are paid to manager, salesman, carder, 
overlookers in card and spinning rooms, engineer, clerical 
staflf, card tenters, grinders, a number of girl setters on, 
boys in warehouse, men in bale and waste room, etc. This 
item will amount to £90 per week. Tlie piecework wages 



on draw frames, fly frames, and mules will depend on the 
respective productions of these machines, and can be 
ascertained at once from the wage books. Since practi- 
cally most of the waste, visible and invisible, has been 
taken out of the cotton before reaching the draw frame, 
the productions of the total machines in the card room 
may be considered equal to the production of the mules 
during any given period. Such being the case we can take 
any single slubber spindle or frame for draw frame and 
slubber price, and single spindle or single frame for the 
wages of each passage of fly frames respectively. These 
items work out as follows : — 

Draw frame 

•0476 pence per lb 







Mule . 


Cleaning, day 

wages, and salaries 


per week. 

The trade expenses have now to be considered. These 
generally form a large item in the cost of making yarn, 
but they cover a very wide ground and vary somewhat 
in amounts from time to time. The following list will 
convey an idea of the items usually placed under the 
heading of trade expenses : — 




Ropes and bands. 

Paper and twine. 

Cleaning waste. 

Cop tubes. 

Repairs to macliinery. 

Repair to structure. 

Painting, etc. 


Bank conmiission. 

Transport of cotton and yarn. 



Rates and taxes. 



Interest on loans. 

Chief rent. 

Directors' fees. 



Stationery and stamps. 

Telephone, etc. etc. 



The total of these can only be estimated from past 
experience and by reference to previous years' accounts. 
Some of them are of a fixed character. Some items are 
bought, used, or paid at short regular intervals, others at 
irregular intervals, so that an average must be obtained 
from, say, a three years' experience associated with a fair 
judgment of the tendency of prices to vary. 

On the whole the estimate would be somewhere about 
2 "5 pence per lb. per week for the trading expenses. Trade 
discounts and commissions are important items, and may be 
put down as "5 pence per lb. 

The cost of producing a pound of 60's twist may now be 
set out as follows : — 

Cleaning, day wages, and salaries 


pence per 




Wages, draw frame . 


,, Slubber 


,, Intermediate. 


,, Roving .... 


„ Mule .... 


Trade expenses 


Discounts and commissions 




From the books it is found that there is a difference of 
18| per cent in weight between the cotton used and the 
yarn produced. Part of this is accounted for by 15 per 
cent of visible waste which is sold, and for which '71 pence 
per lb. is obtained on the basis of the total cotton used. 
The other part of the loss, viz. 3| per cent, is invisible, 
i.e. it consists of fine particles and evaporations of moisture 
during the passage of the cotton through the mill. This 
moisture is restored to the cotton, and generally a little 
extra, say 5 per cent; this is called the regain, but more 
often the cellar gain. 

The costing will now stand thus : — 



Cost of cotton used .... 

. 7-5d. 

Loss on cotton, 18 '5 per cent 


per lb. 

Wages ........ 


} J 

Trade expenses ..... 


) 5 

Discount and commission 



14 -Sd. 

Worth of waste . . . , . 


Cellar gain ...... 





Nett cost 

. 13-35d. 

Cost of cotton ..... 

per lb. 

Cost to clean and produce 60's twist 



10 per cent on capital .... 




The method just given is one based on general lines 
and reduces the costing to an unusual degree of simplicity, 
but a host of questions must arise in the student's mind 
on a variety of points. These can only be briefl}^ touched 
upon here. 

Cost of producing any given amount of sliver or roving 
in the card room and of cleaning the cotton in the scutching 
room should be carefully worked out both as regards value 
of the capital of the machinery and the value of the space 
occupied by the machinery. A series of different values 
will be found of the different hanks. Every effort should 
be made to get the full production for each machine, and 
by a few careful tests the twist to be put in a roving can 
be readily found so that it is not excessive on the one hand, 
and is not a cause of complaint on the other hand when 
put up in the mule creel. 

In mentioning 60's twist it will be recognised that this 
means the counts of the yarn after conditioning, so that 
the added moi.sture or cellar gain necessitates spinning 


higher counts in order to sell them as lower counts due to 
the added Aveight of moisture. For the internal economy 
of the mill it is therefore inaccurate to say 60's counts are 
being spun when it means that 60's counts are being sold. 
To sell 60's counts with 5 per cent of a regain we must 
spin 63's counts. 

Every overlooker in the mill should keep a notebook of 
productions, etc., and the wage cost of each, together with, 
and this is important, the time worked and the number of 
spindles working. Any percentage of spindles stopped 
simply means a corresponding increase in the cost in wages ; 
the rollers are running all the time and measuring wages. 

Interesting exercises for a student can be found in 
calculating the amount of cotton required to spin certain 
numbers, and base all costs on the price of cotton in order 
to find the price of the yarn. On the other hand, an 
assumed quantity of a certain count can be taken, which 
has to be sold at a certain price. Work back from this 
price and find the price of the cotton that will spin the 
counts and the amount of cotton required. 

VOL. in 2 I 


Action, principle of spinning, 17 

ot mule quadrant, 109 

of traveller in ring frame, 294 
Actual and calculated speeds, 271 
Adjustment of bands, 42 
After-stretch in the mule, 236 

motion, 442 
American cotton, diameter and 

setting of rollers for, 274 
Analysis of a mule cop, 99 
Anti-ballooning motions, 328 
Anti-suarling motions, 253, 442 
Application of twist wheel motion, 

Arrangement of the fibres in the 
yarn, 8 

of machinery in mills, 385 
Arrangements for "locking," 82 
Assistant winding motion, 442 
Automatic stop motion on Avinding 
frames, 342 

Backing-off by rope driving, 33 
chain, 86 

chain and faller motion, 220 
tightening motion, 220 
tightening the, 96 
and drawing-up, 55 
in the long lever mule, 214 
motion, 86, 222, 239, 428, 433, 

object of, 93 
Back shaft driven from the front 
roller, 37 
and its scrolls, 39 
Bad cops and their remedies, 154 
Balancing the faller wires, 165 
Balloon plates. 328 
Ballooning, 300 

VOL. Ill 4 

Ballooning effect, 328 
Band, governor motion, 192 
Bands, scroll, 41 

spindle, 49 

squaring, 44 

stretching of rim, 52 
Bare spindle, spinning on, 330 
Belt, drawing-up by, 232 
Belt driving, jiower of, 415 
Bleaching and dyeing the cop, 372 
Bobbin winding frame, 337 
Booth-Sawyer spindle, 316 
Brake for doubling spindle, 360 

motion, 436 
Breaking weight of yarn. 2, 8, 14 
Building or shaj^er motion on the 

mule, 135 
Building motions, 291 
Bundling press, Coleby's patent. 

Calculated and actual speeds, com- 
parison of, 270 
Calculations for finding the weight 
on the rollers in mule, 243 
ring frame, 285 
Calculations for the mule, 270 
mill planning, 385 
ring douliler, 369 
ring frame, 335 
Cam shaft, driving of, 77 
mule, changes in, 68 
drawing-up, 82 
Cap bars on mule, 242 
Card, counts of wire used for various 
cottons, 412 
productions of, 393 
speeds of the various organs in 
different cottons, 412 




Carriage of mule, movement of, 3o 

outward ruu of, 80 
Cause of twists flying to the smallest 
diameter in yarn, 6 
snarls in mule yarn, 253 
twist in the ring spinning frame, 
Chain, backing-off, 86 
tightening motion, 222 
winding, 112 
Change of speed in the mule carriage, 

Changes in the mule, 54 

on the cam-shaft mule, 68 
Changing the rim pulley, oO 

driving strap, 217 
Character of the miile's sjiinning 

action, 20 
Characteristics of a miile cop, 99 
Chase of a mule cop, 138 
Cheeses on quick traverse winding 

frame, 348 
Chinese cotton, diameter and setting 

of rollers for, 245 
Clearer frame, 337 
Click, winding, 132 
Coils on the spindle blade, 93 
Coleby's reel, 378 
Combed yarns, superiority of, 3 
Comparison of duplex ■ and single 
driving in mule, 60 
mule and ring yarn, 330 
Compensation for slippage in belts 
and bands, 50 
the taper of the mule spindle, 
Cone clutches, friction in, 90 
Coning parts of the mule shaper, 

Constants for twists per inch, 411 
Construction of rim shaft, 56 
Convenient multipliers, 414 
Convexity of long incline of the 

mule shaper, 149 
Cop, analysis of the mule, 99 
Cops, defective, and their ivnieclies, 
bleaching and dyeing of. 372 
Correction ofshaper for bad cops, 160 
Costing, 476 
Cotton, ideal state of, 1 

Cotton mills, power required to 
drive, 412 
yarn measure, 413 
Counter faller, weighting of, 167 
Creels and their arrangement, 29 
Creels of doubler frames, 356 
"Crossing," 138 
"Crossing" on the cop, 118 
Cross winding on the reel, 376 
Cycle of actions in the mule, 59 
Cylindrical form of yaru, 12 

Data for mill planning, 393 

Dead weighting, 243 

Defective cops and their remedies, 

Definition of twist and weft, 6 
Delivery motion whilst winding.. 

Details of fine spinning mule, 226 
Diagrams of mule power, 263 
Diameter of yaru, regularity of the 

rinss for different counts of yaru 
Difference in diameters of yarn, 2 

of yarn in weight and length, 5 
Dividends, table of, 417 
Dobson-Marsh spindle, 320 
Double-speed driving, 227, 251 

boss rollers, 244 et seq. 

rings, 290 
Doubled yarn showing variations in 

twist, 3 
Doubler, ring, 353 

calculations for, 369 

creels, 356 

English and Scotch s}"stems, 357 

knee brakes, 360 

rope driving, 369 

spindles, 360 

stop motions, 361 

troughs, 357 

twisting, theory of, 363 
Drafts for various cottons and 

counts of yaru, 395 
Drag, 236, 466 
Draw Iranie, productions of, 394 

rollers for various cottons, 245 
et seq. 

weights required for, 409 


Diawing-out motion, 440 
Dr.iwing-up liy rope driviug, 33 

and backing-ofl", 54 

by belt or strap, 64 

by strap, t)4 

friction cone, 62 

iu cani-sliaft mule, 81 

in long-lever mule, 215 

motion, 250, 431, 437 
Driviug of the mule, 32 

at the side, 33 

carriage, 35 

cam shaft, 77 

front roller from the tin roller, 52 

mule, duplex, 60 

quadrant, 130 

ring frame, 281 
Drum, winding, 132 
Duplex driving, 60 
Duration of backing-ofl', 85, 260 
Dynamometer, 260 

Easing motion, 169 

Eccentric traverses, 242 

Efl'ect of twist on the diameter of 

yarn, 6 
of an inclined spindle, 21 
of the varying inclination of the 

yarn during winding, 256 
Egyptian cotton, diameter and 

setting of the rollers for, 248 
Elasticity of yarn, 14 
Elastic spindles, 323 
Electricity in the mill, 405 
English system of doubling, 357 
Examination of the reason why 

twists fly to the smallest 

diameter of yarn, 7 
of the mule coji, 99 

Faller leg, 136 

rods and wires, 94 

sector, 136 

sector and backing-ofl" chain, 220 
Fallers, weighting of. 165 
Features of a cop, 106 
FiVires, arrangement of, in yarn, 8 
Fine spinning, drawiug-up by straj), 

mule, 226, 428 
Flexible spindles, 323 

Fly frame rollers, suital)]e weights 

for, 410 
Footstep bearing of spindle, 100 
Friction cones, 63 
Front roller driving the back shaft, 


Gain and ratch, 232 
Gallows pulley driving, 2S1 
Gassing, 381, 454 

loss in gassing, 383, 456 
Gearing for taking the mule carriage 
out, 37 
of mule, 272 
of ring frame, 334 
General slippage of bands, 52 
Governor or strapping motion, 188 
Grant system of reeling, 376 
Grajihic method of showing sj^eeds 
of spindle, 105 
method of showing speed of 
spindles produced by (quad- 
rant, 115 
explaining the action of the 
shaper, 146 
Gravity spindle, 321 

Half-twisted belt driving, 281 
Hank rovings suitable for various 

counts and cottons, 395 
Hastening motion, 217 
Hollow rim shaft, 53 
Horse-power required to drive the 
mule, 260 

complete cotton mills, 412 

cotton machinery, 408 

the ring frame, 332 
Humidity in cotton mills, 399 
Hygrometers, 403 

Ideal state of cotton, 1 
Imperfections of cops, 154 
Inclination of the mule spindle, 21 
of roller stands in ring frame, 

Inclines on the mule shaper, 136 
Indian cotton, diameter and setting 

of rollers for, 246 
Indicating the mule, 260 
Initial slippage of bands in the 

mule, 52 


Initial rate of speed of miile spindle, 

105, 126 
Intermediate rollers for various 

cottons, 245 et scq. 
Irregularities in yarn, 3 
due to bands, 43 
compensation for, 52 

Jack frame rollers for various 

cottons, 245 et seq. 
Jacking motion, 234, 442. 445 
Japanese cotton, diameter and 

setting of rollers for, 245 

Knee brakes for doublers, 360 

Lea winding on the reel, 376 
Leather-covered rollers, 242 
Length and weight of yarn, regular- 
ity of, 4 
Lever weighting of rollers, 243 
Locking arrangements, 83 

motion, 448 
Long-lever mule, 206, 244, 421 

changes in, 68 
Long shaper in the mule, 136 
Loss in gassing j'arns, 383 
Lubrication of spindles, 333 

Machinery, power required to drive 

cotton, 408, 412 
Measuring the diameter of yarn. 16 
Methods of judging j'arns not per- 
fect, 3 
of showing imperfections in yarn, 
Microscope, testing yarns under 

the, 2 
Mill planning, 385 

data for, 395 
Moisture in a cotton mill, 399 
Mule, analysis of cop. 99 
anti-snarling motions, 253 
arrangement of the creels, 29 
assistant winding motion, 442 
backing-otf, 86 
backing-off l>y band, 34 
backing- off motion, 239, 428, 

433, 450 
brake motion, 436 
calculations, 270 

Mule, cam-shaft jiriuciple, 68 
changes in the, 54 
changes in the cam shaft and 

long lever, 68 
crossing, 138 
cycle of actions, 59 
defective cops and their remedies. 

double-speed driving, 227 
drawing-out motion, 440 
drawing-up, 82 

and backing-off, 56 

by strap, 232 

by rope, 34 

motion, 250, 431, 437 
driving, 32 

the spindles, 49 

the cam shaft, 77 
duplex driving, 60 
easy motion, 169 
effect of a tapered spindle on 

winding, 170 
extra winding motion, 230 
fine spinning, 226, 428 
friction cones, 62 
gain and ratch, 232 
general description of, 24 
governor or strapping motions, 

hastening motions, 217 
horse -power required to drive 

the, 260 
imperfections of cops, 154 
improvements in, 421 
inclination of the spindle, 21 
inclines on the shaper, 136 
initial slippage of bands in the, 

initial speed of spindles, 105, 

jacking motion, 234, 442, 445 
lever weighting of rollers, 243 
locking arrangements, 83 

motion, 448 
long lever, 206, 244, 421 

backing-off, 214 

backing-off chain, 220 

backing-off motion, 223 

chain-tightening motion, 222 

changing the strap, 217 

drawing-up, 215 



Mule, long-lever, hasteniiis; motions, 

spinning' action, 209 
strap-relieving motion, 217 
long sliaper, 136 
modifying the results of twist, 8 
movement of carriage, 35 
movement of the nut up the 

quadrant screw, 125 
nosing motions, 174 
object of backing-off, 93 
outward run of the carriage, 80 
position of the spindle, 19 
principle of the scroll, 45 
principle of the spinning action 

in the, 17 
quadrant, 109 
ratching motion, 442 
rim shaft, 436 
rollers for various cottons, 245 

et seq. 
roller-delivery motion, 436 
roller stands and weighting, 240 
roller turning motion whilst 
winding, 237 

twisting at the head, 236 
scrolls, their shape and action, 38 
setting of the rollers, 245 et seq. 
setting-on motion, 431, 437 
shaper or building motion, 135 
shaper, long, 137 

short, 425, 428 
side-driven, 35 
snarls and anti-snarling motions, 

special, 436 
speed of carriage during spinning 

and winding, 35 
spindle, position of, 19 

inclination of, 21 

taper of, 22 
starching, 258 

strap-fork, movement of, 73 
strap-relieving motion, 90, 217, 

tightening the backing-off chain, 

tubes and starching, 258 
twist motion, 434, 448 
weighting of rollers, 240 

of fallers, 1 65 

:MuIe, winding, 109 

winding drum and tin roller, 132 
Multipliers for twist per inch, 411 

convenient, 414 

Nosing motion, 173 

Number of spindles per horse- 
power in the mule, 266 
of mule spindles to various 
machines, 395 

Nut, movement of, up the quadrant, 

Object of backing-off, 93 
Operations in the cam-shaft mule, 68 
Outward run of the mule carriage, 80 

Peg, nosing, 176 

Percentage of slippage in bands, 54 

of humidity in cotton mills, 404 
Plan of a pair of mules, 24 
Planning of mills, 385 

data for, 394 
Plates, front and back shaper, 142 
Position of mule spindle relative to 
the rollers, 19 

mule wharve, 412 
Power required to drive the mule, 

cotton machinerj% 408 

cotton mills, 412 

ring frame, 332 

transmitted by rope and belt, 415 
Prevention of waste in the ring 

doubler, 361 
Principle underlying the inclination 
and taper of a mule spindle, 21 

of the action of the traveller, 294 

cam-shaft mule, 68 

mule scrolls, 45 

nosing motion, 170 

quadrant, 109 

shaper, 141 

twisting effect in the doubler, 364 
Problems connected with the shaper, 

Productions of cards, 393 

draw frames, 394 
Proportions of machinery in a mill, 

Pulley, three-grooved rim, 50 


Quadrant, principle and action, 109 

and its connections, 130 

screw, 203 
Quick-traverse winding frame, 340, 

Rack governor motions, 196 

Rail, shaper, 135 

Ratch and gain, 232 

Ratching motion, 442 

Rates at which the mule sf)indle 

works, 106 
Reel, %\Tap, 5 
Reeling, 372, 468 

Coleby's reel, 378 

cross winding, 376 

doffing motions, 378 

Grant system of winding, 379 

lea system of winding, 376 
Regular and variable quadrant 

screw, 204 
Regularity of the diameter of yarn, 2 
Relationship between quadrant an I 

shaper, 150 
Remedies for defective cops, 154 
Rim band, stretching of, 50 

pulley, 50 

three-groo%'ed, 51 

shaft, 54, 436 

hollow, 53 
Ring frame, general description of, 

ballooning, 300 
effect, 328 

Booth-Sawyer spindle, 316 

building motions, 291 

calculations, 335 

calculations for weight on rollers, 

comparison of ring and mule 
yarn, 331 

diameter of rings for various 
counts, 290 

Dobson-Marsh spindle, 320 

driving of, 281 

flexible spindles, 323 

gearing of, 334 

gravity spindles, 322 

lubrication of spindles, 333 

power to drive, 332 

"Rabbeth" spindle, 320 

Ring-frame, rings, 290 

rollers for various cottons, 245 

et seq. 
rope driving, 369 
Sawyer spindle, 320 
space of spindles and suitable 

rings, 290 
spindles, 315 

theory of the traveller, 294 
tliread guide, 289 
traveller, 288 
twisting, 288 
weight of travellers, 309 
Roller turning motion whilst turn- 
ing at the head, 236 
delivery motion 'whilst winding, 

237,' 436 
diameters and settings for various 
cottons, 244 et seq. 
Roller stands, inclination of, 284 

and weighting, 2^0, 284 
Rollers, weighting of, 409 
Rope driving, 282, 369 

jiower transmitted by, 415 
Roving frame rollers for various 

cottons, 245, et seq. 
Rules for mule calculations, 270 
ring-frame calculations, 335 
the diameter of yarn, 17 

Saddle and bridle weighting, 285 
Scotch doubler, 357 
Screw of quadrant, 203 
Scroll bands, 41 

the principle of a, 45 
Scrolls, position, action, and con- 
struction of, 39 
Scutcher laps, variations in, 5 
Section of the diameter of yarn, 

Sector and backing-off chain, 220 
Self-weighted rollers, 286 
Setting of rollers for different 

cottons, 245 etseq.y 467 
Setting-on motion, 431, 437 
Shaft, rim, 54 
Shaper or building motion, 135 

short, 425, 428 
Side-driven mules, 35 
Slippage of rim band, 50 

general, 53 



Slippage of bands, percentage, of, 54 

initial, 53 

of spindle bands, 282 

of straps and bands, 51 
Slubber rollers for various cottons, 

245 et seq. 
Snarling motions, 253 
Spaces of spindles in ring frame, 

Special mule, 436 
Speed, double, 227 

of carriage during spinning and 
winding, 35 

of spindle necessary to form a 
cop, 100 
Speeds in the card, 412 
Spindle in tlie mule, j^osition of, 19 

inclination of, 21 

tai)er of, 21 
Spindles, driving of tlie, 49 

bands, 49 

eflect of the taj^er on winding, 

per horse-power, 266 

of ring doubler, 360 

of ring frame, 315 

taper of, 101 
Spinning, fine, 226 

on the bare spindle, 330 

theory of, 1 

various forms of, 17 
Spinning action of the mule, 17 

in the long-lever nmle, 209 
Square roots, talde of, 418 
Squaring band, 44 
Starching, 258 
State of cotton, ideal, 1 
Steady bracket in mule, 153 
Stop motions on doubler, 361 
Strap, changing from fast to loose 
pulley, 217 

drawing-up by, 66, 232 

fork, movement of, 73 
Strap-relieving motion, 60, 90, 217, 

Strapping or governor motion, 188 
Strength of yarn, 8 
Stretching of rim bands, 50 
Strongest yarn not maile from tlie 

strongest cotton, 8 
Structure of a mule cop, 99 

Suitable hank rovings and drafts 
for various cottons and counts, 
counts of wire for cards, 412 
percentages of humidity for cotton 

mills. 404 
speeds in the card, 412 
Suj)eriority of combed yarns, 3 
Swift for i-eel, 372 

Table of dividends, 417 

of square roots, 418 

of twists per inch and square 
roots, 419 
Tachometer for indicating speeds, 

Taper of spindle, 22, 101 
Tapered spindle, etl'ect of, in wind- 
ing, 170 
Tension of tlie yarn, 166 
Testing yarn for diameter, 2 

under the microscope, 2 
Theory of spinning, 1 

of the traveller, 294 
Thick and thin places in yarn, 6 
Thread guides, 289 
Three-grooved rim pulley, 50 
Tightening the backing-ott' chain, 96 
Tin roller, 132 
" Top " spindle, 321 
Traveller, 288 
Travellers, weight of, 309 
. for different counts, 314 
Traverse motions, 242 
Troughs used in ring doublers, 357 
Twist, effect on the diameter of the 
yarn, 6 

and weft, 6 

change jilaces for, in doublers, 

how produced iu the mule, 17 

motion, 434 

from tin roller, 448 

wheel motion, 90 

why it flies to the smallest 
diameter, 7 
Twister, 353 
Twisting at the head, 76, 90 

roller motion during, 236 
Twists for doubled yarn, 370 

per inch, multipliers for, 411 



Tubes and starching, 258 
Types of spindles, 323 
Typical defects in cops, 15-4 

Uncertainty of bands in mule, 43 
Unilbrniity in yarn, 2 

of the twist in ring yain, 313 
Useful information, 408 
Use of cotton yarns, 337 
Usual diameters and settings of 
rollers, 244 

Variable screw in quadrant, 204 
Variations of diameter in yarn, 2 
illustrations of, 3 
in doubled yarn, 3 
scutcher laps, 5 

speed of spindle required for 
building a cop, 103 
Various forms of reels, 374 

of spinning, 17 
Varying movement of mule carriage, 

Warpers' bobbins, 338 

Waste, percentage of, in doubler, 

Weakest yarn not made from the 

weakest cotton, 8 
Weight and length of yarn, regu- 
larity of the, 4 
Weighting the fallers, 165 
and roller stands, 240 
in ring frames, 284 
Weights of travellers, 309 

and measures of cotton yarns, 

for draw-frame rollers, 409 
for tiy-frame rollers, 410 
Wharve, position of, in mule 
spindle, 412 

Winding drum, 111 
driving of, 132 
click, 133 

effect of tapered spindle on, 170 
frame bobbin, 337, 463 

quick traverse, 340, 465 
motion, 442 

for tine spinning mule, 230 
variations of reel, 375 
Wire, suitable counts of, for cards, 

Wrap reel, 5 

Yarn, arrangement of the fibres in, 8 
cause of thiclc and thin places, 7 
comparison of ring and mule 

yarn, 331 
diameter of, 17 
elasticity of, 14 
irregularities in, 4 
judging, method of, not perfect, 3 
measures and weights of, 413 
preparing machine, 372 
regularity of diameter, 2 

in length and weight of, 4 
rotundity of, 12 
section of, showing fibres, 15 
strength of, 8 
strongest and weakest, 8 
supei-iority of combed, 3 
table of twists per inch in, 411 
tension in, 166 
testing under the microscope, 2 

the diameter of, 2 
uniformity of, 2 
use of, 337 
variations in, 3 

doubled, 3 

Zig-zag creels, 32 

Piinicd in Great Hritain hy K. & 1\. Ci.akk, T,ii\iiin:t), l''ifinl'urt;h. 



/// Three Volumes. Crown Zvo. Illustrated. 

Vol. I. — Including all Processes up to the end of 
Carding. Eighth Edition. los. net. 

Vol. II. — Including the Processes up to the end of 
Fly Frames. Sixth Edition, with Appendix. 
8s. 6d. net. 

Vol. III. — -Including Theory of Spinning — Mechanism 
and Working of the Mule — The Ring Spinning 
Frame — Winding Frames — Doubling — Yarn 
Preparing — Mill Planning — Humidity — Use- 
ful Information — Appendix — Index. Fifth 
Edition. los. net. 

Cotton Gazette. — "Most excellent and exhaustive work. 
. . . The first edition was published in 1896, and the 
fact that this highly useful technical work has been steadily 
corrected and kept up-to-date, with improved methods 
and machinery, is the best tribute to its great value to 
practical cotton spinners." 

Cotton Factory Times. — " ' Cotton Spinning ' is a helpful 
treatise, and is deservedly popular." 

Nature. — "This book should certainly be in the hands, 
not only of every cotton spinner, but of every spinner 
of materials of a like nature." 



A Book of Sketches of all Types of Machinery of 
a modern Cotton Mill, with full details and 

Third Edition. Sniper Royal ^vo. ys. 6d. net. 

Cotton. — " A very interesting sketch-book of illustrations 
of all types of cotton machinery. The drawings are taken 
from the three volumes of the same author's book on 
Cotton Spinning, and should be found most useful for those 
wishing to become thoroughly acquainted with the subject." 


A Book of Cotton Mill Calculations illustrating all the 
Gearing of the Chief Types of Machinery, and 
Worked Examples of all Rules necessary to a 
Spinning Mill. 

Third Impression, with Corrections. Illustrated. 
Crown %vo. ys. 6d. net. 

Nature. — " This admirable work should be in the hands 
of every one, at home and abroad, who is interested in the 
practical working of textile machinery." 


A Practical Guide for Managers, Carders, and Over- 

Illustrated. Crown Svo. los. net. 

Textile Manufacturer. — " The book contains a great deal 
of useful information, and at the modest price charged can 
be warmly recommended to the notice of all engaged in 
cotton spinning." 


Tele^yanis : ^' Globe, Castleton-Lancashire." 
Telephone: yi Rochdale. 

Royal Exchange, AfanchesiCf . 
Pillar H.-2. 


(1920), LTD. 


Hopper Bale Openers. 
Hopper Feeders. 
Porcupine Feed Tables. 
Crighton Openers. 
Exhaust Openers. 
Large Cylinder Openers. 

Flat Carding Engines. 

Drawing Frames. 

Stubbing Frames. 

Intermediate Frames. 

Roving Frames. 

Twist and Weft Ring Frames. 

Wet and Dry Ring Doubling Frames. 

(4-Coilcr and Condenser Systems) for WASTE. 


"ASA, OLDHAM" 4th and 5th Editions No. 777 OLDHAM 




























Iron Castings, 



Illustrated Technical Dictionaries 

In Six Languages : 


Indispensable to all Engineers and Business Men 
who have Dealings with Foreign Countries. 




English Editor : WM. SCOTT TAGGAKT, MJ.Mech.E. 

"... the Jiiost complete and well-arranged work of this nature whicli 
we have had before us for review. . . . 

" It is difficult to speak too highly of this publication, which will be 
sure of great appreciation in this country. . . ." — The Textile Recorder. 

Agents for the British Section : 


25 Victoria Street, LONDON, S.W.I. 









68 Gold Medals and Diplomas 
8 Grands Prix 



And at Todmorden. 


Telegrams: NUGGET. LIVERPOOL. Codes: ABC, A I , and MARCONL 


Castle Iron Works, STALYBRIDGE. 

Telegraphic Address: " Lang, Stalybridge." Telephone No. 107. 

Manchester Exchange. "E8" Pillar. Tuesdays and Fridays. 






For all classes of Condenser Yarns from Cotton Waste, 
Vigogne, and Angola Yarns, Hosiery Yarns. 

Hopper Bale Opener and Cotton 

Conveying and Distributing 

Lattice Work. 
Exhaust Fans for Transit of 

Hopper Feeder with Patent Feed 

Supply Apparatus. 
Buckley's Patent Combined 

Opener with One Cylinder 

and One Beater. 
Buckley's Patent Combined 

Opener with Two Cylinders 

and one Beater. 

Buckley's Patent Opener, specially 

arranged for Sea Island Cotton 

(without Beater). 
Crighton Opener. 
Roving Waste Opener. 
Scutcher or Lap Machine (Single 

or Double). 
Self-acting Mules for Coarse, 

Medium, or Fine Counts. 

Patent Self-acting Mules, with One, 
Two, or Three Spindle Speeds, 
for Spinning Waste, Barchant, 
or Woollen Yarns. 


& CO. 

Specialists in machinery 


For winding yarns 
for the Creel of 
Twisting Frames, 
the machine illus- 
trated is the most 
efficient on the 
market. High 
Speed. Maximum 
Production. Mini- 
mum Waste. In- 
stantaneous Stop 
Motion. No Cob- 
webbed Ends. 
Winds any class of 
\arn, with or with- 
out Stop Motion. 
Damping or con- 
ditioning ar- 
range m e n t, 
specially ar- 
ranged for 
worsted and 
similar yarns, 
fitted if re- 

Patent Quick Traverse Winding Frame. 




for the doubling trade. 

Arundel — Coulthard Products 

Patent Quick Traverse Winding Frames 
Patent Slow Traverse Winding Frames 
Patent Ring Doubling Frames 

Patent Flyer Doubling Frames 

for cotton, silk, worsted, ramie, and 

linen yarns 
Patent Clearing Frames 
Patent Vertical Burner Split Drum High 

Speed Gassing Frames 
Patent Quick Traverse Gassing Frames 
Patent High Speed Reels 
Patent Yarn Preparing Machines 
Patent Bundling Presses 
Gravity High Speed Ring Spindles for 

Spinning and Doubling Frames 

Rings for Spinning and Doubling Frames 

Top and Bottom Rollers for Spinning and 
Doubling Frames 

Patent Roller Traverse Motion for Spin- 
ning Frames 

Bobbin and other Wheels for Slubbing, 
Intermediate and Roving Frames 

Long and Short Collars for Intermediate 
and Roving Frames 

Inner Tubes for Gravity Spindles 

Spindle Wheels 

Flyer Frames and Ring Frames — The 
Coulthard Patent Variable Traverse 

Appleby Patent Simplex Lappet 

Murphy-Simpson Patent Shuttle Threader 

CATALOGUES will be sent to all responsible buyers on request. They are fully 
illustrated and should be in the hands of all buyers of textile machinery. Please 
mention the machines or accessories in which you are specially interested. 

Agents and Representatives 

For HOLLAND and SCANDINAVIA: Vlies and Benson, Ltd., Manchester 

For FRANCE and BELGIUM: Benson and Vlies, Paris. Rouen, Lille, and Mulhouse 

For JAPAN : Takata and Co., London, Tokio, New York, Osaka, Shanghai, eind Dairen 

For SPAIN : Senor Carlos Salles Gerona, 50, I', 2' , Barcelona 

For INDIA: Duncan, Stratton and Co., 9 Bank Street, Bombay 

MANCHESTER EXCHANGE, Tuesday and Friday, Nos. H3 and F2 Pillars 

Telephone : 
Stockport 606, Preston 140. 

Telegrams : 
'Arundel, Stockport." "Coulthards, Preston' 



Send for a 
copy of the 
Belting Book. 

For over 1 00 years 
we have been pro- 
ducing leather 
products of quality. 

From the raw hide to the 

finished product every stage 

manufacture is carried out 

in our tanneries and factories. 

Every leather belt bearing the 
mark Walker, Bolton, is a firm, 
sound belt of uniform strength 
and substance — points which 
have made Walker's Belting 
famous throughout the world. 

Wm. walker & SONS, Ltd., BOLTON, ENGLAND. 



For durability and resili- 
ency use Walker's Buffalo 
and Oak Tanned Pickers, 
Chrome Picking Bands. 

We make every descrip- 
tion of leather fittings for 
all makes of looms. 
Standard patterns can 
be supplied from stock — 
other shapes to suit every 
make of loom can readily 
be made. 

Send for our Illustrated Lists. 

Wm. walker & SONS, Ltd. 


Spindle and Flyer Manufacturers and Mach inists, 

Hargreaves Works, OLDHAM. 


Of every description and with any kind of Presser 
for Cotton, Silk, Woollen, or Flax Spinning. 

Improved MULE SPINDLES for High Speeds. 

speciality :— BODDEN'S IMPROVED 



Which entirely dispenses with necessitj' of stopping frame, taking out 

spindle, or use of pumps when re-oiling, thus saving a great amount of 

oil, bands, time, and labour. 

4,000,000 Spindles isith the above Appliance at Work. 


Ordinary and Elastic Rabbeth and other Ring Spindles. 

New Long and Short Collars. 

Wheels, Steps, Brackets, Plain or Loose Boss Top Rollers. 

Patterns of every description for Lawson's Frames. 




And put into thorough working order on our own 
Special Improved Methods. 



By F. H. Bowman, D.Sc. 

APPLICATIONS. Illustrated. Crown 8vo. 
I OS. 6d. net. 

POSES. Illustrated. Crown 8vo. los. 66. 

By Thomas Woodhouse and Thomas Milne. 

Illustrated. Crown 8vo. I2s. 6d. net. 

Crown 8vo. 15s. net. 

By Thomas Woodhouse. 

MECHANISM. Illustrated. 8vo. 25s. net. 

By W. Lawrence Balls, ScD. 

in Physiology and Genetics. Illustrated. 8vo. 
6s. 66. net. 

COTTON GROWERS. 8vo. Sewed. 3s. 66. 



By W. Lawrence Balls, Sc.D. {continued) 

trated. 8vo. Sewed. 3s. 6d. net. 

By T VV. Fox. 

Edition. Illustrated. Crown 8vo. 12s. 6d. 

By J. Can NELL Cain, D.Sc. 

PRODUCTS FOR DYES. Illustrated. 8vo. 
I OS. net. 

I 2s. 6d. net. 

By Prof. Arthur G. Green, F.I.C 

COLOURING MATTERS. Founded on the 
German of Drs. G. Schultz and P. Julius. 
Third Impression. Imperial Svo. 25s. net. 

By G. S. FRAPS, Ph.D. 


By M. S. WOOLMAN, B.S., and E. B. McGowan., B.S. 

TEXTILES. A Handbook for the Student and 
the Consumer. Crown Svo. ' 12s net. 



X S 


^ z 

^ S 

2 0^ 



n t^nmtmm mmumm wmmmmimmKmmtfmimttmtfiifBm