Skip to main content

Full text of "Cotton spinning"

See other formats


Slip i. B. ItU IGibrarg 

Nortli (Earolina ^tnU Hmtteraitg 


















'quadrant and SHAPER of the S.A. mule'; CYLINDER OIL 






] 930 


First Edition 1S97 
Second Edition 1901 

Third Edition 1905 
Fourth Edition igo8 

Fifth Edition 1913 
Reprinted 1917, igig 
Sixth Edition 1921 
Refrin'ed i^'H, 1930 



This, the second volume, reprinted from The Textile Mercury, 
brings the subject of cotton spinning down to the end 
of what is generally termed the preparing processes. It 
includes all the machinery and manipulations between the 
card and the self-actor or ring-frame. As the operations 
included in this section have a most important influence 
upon the future character of the yarn, it has been con- 
sidered necessary to go a little more deeply into the subject 
than is usually thought necessary. Attention has not only 
been confined to the principles underlying the actual pro- 
cesses themselves, Avhich is of itself a most interesting and 
important feature, but in conformity with the objects that 
prompted the writing of the first volume, an effort has been 
made to give to the mechanical details an interpretation 
which some of them have hitherto not possessed. 

It Avas with this purpose in view that the writer remarked 
in the preface to the first volume that the book would not 
be one of " mere description." The remark was meant to 


be applied to the whole subject of cotton spinning ; in this 
volume a reasonable claim may be made that it is justified, 
and the third volume, dealing with the spinning processes, 
will still further prove the writer's objects and method. 

The fact that at present the education of the average 
reader is not sufficiently thorough to enable him to under- 
stand fully some aspects of the question discussed is no 
reason why an attempt at completeness should not be made. 
Such a state cannot last very long. The systems now 
being adopted in our technical schools will raise the general 
level of intelligence, and, moreover, a feeling will grow that 
we " must " so raise ourselves if we are to maintain the 
practical superiority over our competitors which we at 
present claim. W. S. -T. 

Bolton, 1897. 


Improvements and corrections have been made and the 
drawings renumbered, so that the book is entirely self- 
contained. This system of dividing the subject of cotton 
spinning into three volumes has been justified : it fits in 
with the arrangements for study and the examinations, and 
meets the requirements of those employed in the various 
departments of the mill. W. S. - T. 



The book has been brought up to date and considerable 
additions made both to the matter and the illustrations. 
The author and publisher hope that these improvements 
will cause it to maintain its career as a useful handbook 
to the student and the practical man. 


Bolton, 1913. 




Drawing 1 

Combing 47 

Fly-Fkames . 118 

Supplementary Notes 241 

INDEX 289 


1. Section of Draw-Frame 

2. Tandem System of Draw-Frames 

3. Alternate System of Draw-Frames . 

4. Zigzag System of Draw-Frames 

5. "i 

' pVeigliting of Rollers in Draw-Frame 

7. Solid and Loose Boss Rollers 


Diameters and Spaces of Draw-Frame Rollers for various 
classes of Cotton 

Diagram showing effect of Doubling and Drawing 

Diagram illustrating Draft in Draw- Frame 

Front and Back Stop Motion 

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 

Ceiling'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 





Dobson and Barlow 

27. Diagram of Roller Gearing in Draw-Frame 

28. ,, „ „ . . 
28a. Draw and Lap Machine. Dobson and Barlow 

28c. „ „ ,, ,, 

28e. Gearing of Ribbon Lap Machine 

29. Section through Comber (Duplex). 

30. Star Feed Wheel 

31. Section through Comber and Nipper Cam 

32 "1 

' j- Two Arrangements of the Nipperd .... 

OOt J 

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 Mechanism . 

40. Section of Single Nip Comber . 

„ Double Nip Comber 

'Diagrams explaining the Combing Action 


46. Section through Nasmith's Comber . 

47. Gearing Plan of Nasmith's Comber . 

48. Detail of Nasmith's Comber .... 

49. ,, „ „ . . . , 

50. ,, ,, ,,.... 

51. Diagrams explaining Action of Nasmith's Comber 

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

53. 1 

> Details of Nasmith's Comber . . , , 
54. } 

55. >! 

56. VGauges for Nasmith's Comber . . . • 












58. Stop Motions on Comber. Hetherington . 


60. "Whitin Comber. Howard and Bullougli . 


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

61. Section of Comber . , . . 

62. Gearing Plan of Comber . 

63. Section through Fly-Frame 

64. Plan of the Spindle Rail . 

65. Section through the Rollers and Stands 

66. Cap Bar 

67. RoUers and Stand in Fly- Frame 

68. Diameters and Spaces of Rollers in Fly-Frame 

69 "k 

■ 5-Fly and Bobbin with Drivin 

71. Spindle Footstep Bearing 

72. ~» Diagrams explaining the Action of the Flyer 

73. / Presser .... 

74. Flyer Legs with Straight and Curved Slot 

75. Driving the Bobbins and Spindles 

76. Diagrams explaining Winding in the Fly Frame 

77. Diagram explaining "Flyer Leading" 

78. ,, ,, " Bobbin Leading " 

79. "v Diagrams explaining Variations of Speed of the Bobbin 

80. i during Winding 151,153 

81. Gearing of Fly-Frame ....... 155 

82. \ 
!■ Diagrams explaining the Curves of the Cone Drums 

83. J 

84. 1 Diagrams explaining the Construction of the Cone 

85./ Drums 

86. Epicyclic Train of Wheels 


88. ,, ,, ,, 











136, 137 















Differential Motion (Sun and Planet) 
Section through 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 Skewers 

Draw-Frame, Plan and Elevat 

,, Section 

Ermen's Clearer 
Diagram of Hand-pulled Cotton 

Lensrth of Cotton Fibres 

ill Xatural Condition 

Draw-Frame Rollers 
Flyer-Frame Rollers 
Diagram of "Web of Card 

• Diagram of Drafting in Draw-Frames 




127. \ . . ''•^°E 

> Diagram of Draftin" in Draw-Frames .... 261 
128./ ^ 

■ I Aspirator for Comber "Waste / 

130./ 1 270 

131. Diagram of the Cotton between the Nipper of the Comber 273 


144. Comber, improved Nasmith Type. Dobson and Barlow , 286 

145. Roving Frame. Section through Driving Shaft. Piatt 

Bros, and Co. Ltd 287 


Diagram illustrating the "Weighting iTffect of Rollers .' to 

on the Cotton passing through ..... I 284 


The Web from the Card. — A close examination of the 
web or film of cotton as it comes from the doffer of the 
carding engine will disclose an arrangement of fibres 
quite contrary to what might at first sight be expected 
when the action of the card is understood. Instead of 
order, a very irregular result will be noticed. The specific 
action of the condensing process between the cylinder and 
doffer, however, will account for much of the crossed 
condition of fibres, and the stripping operation of the 
comb will still further increase the irregularity of their 
disposition in the web. It is this irregularity of arrange- 
ment that enables the cotton to be stripped from the 
doffer and carried forward to the calender rollers. Any 
parallelisation in such a thin web would render it practically 
impossible to free the fibres from the doffer — much less to 
carry them away, as at present. A point, however, that 
must not be overlooked, is seen very clearly in the web as 
it approaches the calender rollers of the card. Between 
the doffer and the calender rollers there is always a draft ; 
and, moreover, the mere fact of causing a web of cotton, 
which is the full width of a card, to converge very quickly 

Note. — A very complete set of practical notes on these macliiues will 
be found in the author's book, Cotton Mill Management. 

VOL. II |£ B 


towards a point, will make every fibre in the web partake 
of the convergence, in whatever crossed condition it may 
be. This constitutes, of course, a distinct tendency to 
laying the fibres side by side, which is further augmented 
by the action of the draft in pulling the fibres apart. 
Such an action as this cannot be performed without 
straightening them considerably, the friction existing 
between them being sufficient to cause the individual 
fibres to straighten in sliding over one another. Although 
this is apparently an insignificant action, it is in reality 
the very keynote of the process of drawing ; for while in 
this process great reliance is placed upon obtaining a 
regular sliver, it must not be overlooked that success in 
the operation almost depends entirely upon the fibres 
being laid parallel, which state is brought about because of 
the friction between them during the drawing action being 
sufficient to straighten them out. 

Description of the Drawing-Frame. — Before enter- 
ing upon a detailed investigation of the drawing-frame and 
its action, a general description of the machine will be 
given. To do this a drawing is represented in Fig. 1, 
which represents a section, through the chief features, of 
a well-known type of frame as made by Dobson and 
Barlow. The full cans of sliver are taken from the card 
and put behind the draw -frame, so that the sliver can 
be passed up in the direction of the arrows through holes 
in the guide-plate A. On going forward, each sliver passes 
over a spoon-shaped guide B, and on between two rollers 
C and D, whence it is guided to the back-roller F by 
means of the traverse guide E. Four successive lines of 
rollers are now passed, during which the sliver is consider- 
ably drawn out or attenuated in consequence of each roller 
having a greater surface speed relative to the one next 


to it. This is an essential feature of the machine, and it 
is from this fact that the name " drawing-frame " is derived. 
On emerging from the roller J the drawn sliver is taken 
over a polished plate K, through the funnel L, and on 
between the calender rollers to the coiler. The above is 
a general description ; Ave now enter into details. The 
guide -plate A is arranged so that it can be carefully 
adjusted according to the position of the cans from Avhich 
the slivers are taken. Large quantities of waste are easily 
made by carelessness in setting it, due to the breakages 
that occur through too great a drag being put on the sliver 
in passing through the holes. The draw-frame is arranged 
so that each sliver that enters the rollers is drawn out four 
to eight times its original length, and four to eight slivers 
are passed through together, so that, although each one is 
so lengthened out, their combination at the funnel L, 
through which they pass, produces a sliver which differs 
very little in weight or length from any one of the entering 
slivers ; it is more regular in substance, and its fibres are 
drawn out, comparatively speaking, parallel to each other. 
(The principles underlying these actions are explained on 
page 14.) Otherwise there is no noticeable difference 
between a sliver fed at the back and that delivered at the 
front of the machine. Six ends or slivers are generally 
passed through, and in such a case the total draft between 
the rollers will be six. Each drawing-frame is split up 
into what are termed heads, a head being that portion of 
the rollers through which a group of slivers pass. For 
instance, if six slivers, gide by side, pass through the four 
lines of rollers and are combined into one at the funnel, 
that section of the machine is called a delivery, and a set 
of deliveries is called a head. As a rule, there are from 
two to four heads in a complete draw-frame (see Figs. 2, 3, 


and 4). This diagram illustrates three methods of arrang- 
ing draw-frames. In Fig. 2 we have the tandem system 
in which the deliveries are all one way. It will be 
observed that there are 24 cans behind the machine ; six 
ends pass through each of the four deliveries, so that only 
four ends emerge at the front. These ends would be 
taken to another passage of draw-frames, and possibly to 
a third. Fig. 3 illustrates the alternate system, in which 
the slivers follow the arrow, and an extension of this 

Fig. 2. Fir,. 3. 


Fig. 4. 

arrangement results in a modification known as the zig- 
zag system. These different systems are the outcome of 
convenience to the work hands, economy of labour, and 
suitability of driving, as well as a saving in room and 
power. It must be understood that a wide variation may 
exist in the size of draw-frames. Each head may have 
eight deliveries, instead of four, as shown ; there may be 
three heads of seven or eight deliveries ; in fact, there is 
scarcely a limit to the utilisation of the draw-frame for 
its purpose of drawing and doubling slivers. The length 


of rollers in each delivery varies from 15 in. to 18 in. ; this 
length refers to the top rollers. Each of the bottom rollers 
is generally made in one length, or in sections pieced 
together so that they revolve as one length. These bottom 
rollers are fluted — that is, grooves are cut therein length- 
wise — in order that a grip may be obtained of the cotton 
as it passes through, and also that this grip shall j^revent 
the cotton from slipping or being drawn from between the 
rollers in consequence of the draft. The number of the 
flutes, of course, A^aries in different frames, but they are 
usually as follows: — 1 in. diameter, 36 flutes; \\ in. 
diameter, 45 flutes; 1| in. diameter, 50 flutes; 1| in. 
diameter, 54 flutes. The arrangement of the flutes, 
especially Avhere the top rollers are covered with leather, 
is such that they have a slightly decreasing pitch on the cir- 
cumference. It will be seen that by this means the chance 
of the flutes making corresponding grooves in the leather, 
or actually cutting it, through continual working, is 
considerably reduced when compared with the effect pro- 
duced if the flutes be all pitched alike. The general term 
used to designate this kind of fluting is hunting, and its 
effect is similar to that of the 'hunter cog' adopted by 
clockmakers, and formerly by engineers, in having the 
number of the teeth of wheels in gear so arranged that the 
same teeth come into touch Avith each other only after a 
number of revolutions. 

Rollers. — It will readily be seen that since the drawing 
action of the machine is such an important factor in its 
usefulness, every means must be adopted to enable it to 
perform this part of its work perfectly. The grip of the 
various pairs of rollers, or their power to draw the fi])res 
over each other without breaking or straining them, must 
be carefully attended to, for carelessness in this respect will 


simply mean -waste, and considerable irregularity in the 
resulting sliver. 

The top rollers, althotigh heavy, are not sufficiently so 
to dispense "with additional weighting, and, generally speak- 
ing, draw -frame rollers are weighted by some system of 
supplementary weights. As a rule, dead weights are 
adopted, i.e. weights sufficient for the purpose are lumg by 
means of wire or cast-iron hooks from each end of the 
roller (see Fig. 5). In such a case the weights would 
probably vary from 1 4 to 25 lbs. if each end were weighted 
separately, but very frequently one weight is used for both 
ends ; Avhen this method is adopted the weights must be 
double what they Avere in the first case. The actual weight 
required depends upon the special circumstances in each 
mill, and is largely a question of experience. Many people 
would adopt Aveights of something like the following : — 
Front roller, 22 lbs. ; second roller, 17 lbs. ; third roller, 17 
lbs. ; and fourth roller, 17 lbs. Here the front roller is 
naturally more heavily Aveighted than the others, because of 
the greater bulk of cotton going through at this point ; but 
the others are all alike in weight, the draft between them 
not being so great, and consequently there is no great 
necessity for any variation. There are, howeA'er, many 
authorities who advocate different weights on each line of 
rollers, and they would most probably arrange them as 
follows : — Front roller, 20 lbs. ; second roller, 18 lbs. ; third 
roller, 16 lbs. ; and fourth roller, 14 lbs. It may also be 
noted that some arrangements for the lower class cottons 
and heavy slivers have slightly heavier weights than for the 
finer classes of cottons. It is, however, very much a 
question of experience, and in deciding upon the question 
a full consideration must be given to all the factors of 
the case. 


Leather RoUers.^ — The leather covering of the top 
rollers has already been mentioned. The adoption of such 
a covering on the top rollers of cotton machinery is 
necessitated by the fact that two iron rollers revolving in 
contact, with pressure upon them, would crush such delicate 
fibres as those of cotton when it passed between them. 
The top roller is therefore almost invariably covered Avith 
some elastic or yielding material. Usually it is first 
covered with a specially woven woollen cloth, Avhich is 
firmly cemented to the iron surface of the roller : this gives 
a good elastic foundation. Over this is tightly drawn a 
thin leather covering, which thus forms a smooth, regular, 
and firm surface, which is capable of gripping the slivers 
and yet at the same time yielding sufficiently to prevent 
damage to the fibres. The maintenance of a perfectly 
round leather-covered roller is of great importance in the 
draw frame, and under all circumstances it ought to be 
maintained. One way of doing this has already been 
indicated in the use of the variable pitch in the flutes of the 
bottom roller, but this is only applicable whilst the machine 
is working. When a stoppage of the machine takes place 
for any length of time, such as a week-end or holidays, the 
top roller can be damaged considerably by the effect of the 
hanging Aveights producing a depression in the yielding 
material of the top roller. This, when the machine works 
again, is easily seen in the slightly eccentric running of the 
roller, and its effect is to produce irregularity in the sliver. 
In most mills, when a stoppage occurs, the weights are 
disconnected from the rollers, and this can be done either 
by going round to each of the weights and uncoupling them 
separately, which means a great waste of time and energy, 
or by the jise of some system of raising all the weights 
in one head at once. Such a system is shown in the 

Fia. 5. 


accompanying draAving (Fig. 5). The four lines of rollers 
are shown with the hooks A hanging from the top rollers. 
To these hooks other hooks B are attached, and to these are 
connected the weight hooks C, to which the weights D are 
hung. In each weight is a specially formed hole, through 
which passes an eccentric G, so as not to touch the weights 
in any way whilst they hang from the roller. The eccentric 
is carried at each end by brackets H bolted to the beam. On 
one end is connected a handle J, conveniently placed for 
being used when required. A quarter turn of this handle 
turns the eccentric sufficiently to bring it into contact with 
the weights on the upper part of the hole, and a slight 
continuation of the movement will naturally raise them 
bodily, and so relieve the pressure on the rollers. The 
elongated cast-iron washer F, together with the small 
washers E, are used to prevent the hooks B falling through 
the holes in the beam whenever they are uncoupled from 
the weight hooks. 

A further refinement in weighting is adopted in some 
mills, and consists in making the hook C in two parts, and 
placing a spring between them. This has the effect of 
neutralising any slight shock that might come upon the 
rollers, its purpose being simply to act as a cushion. Fig. 6 
shows a modified form of this system, as made by Brooks 
and Doxey, and the effect of the spring A is to reduce 
vibration due to the high speed of the roller. 

Loose Boss Rollers. — We have treated of the top roller 
so far simply as a roller covered with leather and working 
in slides or bearings at each end. Such a roller is called a 
solid roller, and is represented in the drawing (Fig. 7). 
They are, as a rule, used only for the last three lines of 
rollers. The front line of rollers is now almost invariably 
made with loose bosses. This type, introduced by Evan 

1 DRA WING 1 1 

Leigh, is shown in Fig. 7 in section, and from it wc see that 
there is a centre spindle having a barrelled body, over which 
is fitted an outer shell which runs loose on the centre 
roller ; and since it is only in contact with it at the points A, 
friction will be considerably reduced. Such a loller can 
easily be lubricated, the tendency of the oil being to run 
down to the points A and stop there ; the oil is consequently 
prevented from getting on to the cotton as it passes 
through. The centre roller does not revolve, so that no 
lubrication is required on its pivots. It is much easier in 


a roller of this class for the outer shell to revolve over the 
stationary centre roller, especially when such a thorough 
lubrication can be obtained ; consequently it has become 
very generally used. In spite of these advantages, however, 
there are many who still prefer the solid form of roller. It 
is contended that the very ease with which it revolves is a 
disadvantage, inasmuch as the grip will l)e reduced, and to 
restore it heavier weights must be used. It is also claimed 
that a better draft can be obtained with a solid roller ; but 
when a practical examination is made of the matter the 
contentions are found to be of no value whatever. In the 
facility with Avhich the loose boss roller works there is an 
advantage that has not yet been mentioned — that is, the 


prevention of the possibility of the top roller remaining 
stationary, if only for a moment, whilst the bottom roller 
is still running. When such a thing happens the slivers 
that are passing at the time are practically crvished or 
very much weakened by the rubbing they receive. This 
cannot occur with loose boss rollers, and from this 
fact alone there exists a strong recommendation for 
their use. 

Loose Bush Rollers. — Another form of top roller, 
which has even greater advantages than the loose boss 
type, is one in which the roller itself is solid, but each end 
works in a loose bush ; a good surface for lubrication is 
thus provided, as well as greater facility for the oiling, and 
friction is almost eliminated. It was formerly the rule to 
have double boss rollers in the draw-frame, but this practice 
has fallen almost out of use. Their advantages over single 
boss rollers was apparent — for instance, there were fewer 
hooks, wires, and weights, and consequently the cleaning 
of the machine was a much easier matter and quickly per- 
formed ; the weighting Avas also much more simple, the 
machine was less costly, and in the lubrication less oil was 
required, this, of course, reducing the probability of staining 
the slivers. The single boss roller, however, in spite of the 
above, has one great advantage which the double boss does 
not possess, viz. each sliver, or group of slivers, is treated 
by itself, independently of others, so that more regular 
yarn is produced. In connection with this question of 
loose bush rollers, one well-known firm (Brooks and Doxeys) 
writes as follows:- — "Where manufacturers prefer double 
boss rollers we strongly advise the front top roller being 
made with loose bosses, as these obviate cutting the roller 
leather arising from one of the bosses being larger than the 
other to defective covering. For single boss rollers, how- 


ever, Avhich have the advantage of only two selvages instead 
of four, Ave advise the use of loose bush rollers." A single 
boss roller would be weighted from each end, and if 16 in. 
long on the boss, the leather-covered portion would be 
about 8| in. long. The double boss roller is weighted from 
its middle, and the leather-covered portion on each side of 
the hook would be about 5 in. long. 

It is the practice in most mills to have the bottom 
rollers case-hardened in the necks, whilst in some the whole 
of the bottom rollers are case-hardened throughout. As a 
result they are rendered stiffer and stronger, and more 
capable of resisting torsion, while the flutes are less likely 
to be damaged either through accident or carelessness. 

Diameters and Setting of Rollers. — Some attention 
will now be paid to the sizes and the conditions of setting 
the rollers, together Avith the circumstances that must be 
taken into account in their arrangement. The importance 
of the relative position of the rollers to each other, accord- 
ing to the cotton being worked, cannot be overestimated. 
If the operation is carelessly performed, nothing can after- 
wards remedy the bad Avork that is sure to result. There 
is one broad principle that must always be used as a guide 
in setting the rollers. The last pair of rollers must be so 
set that the distance apart of their centres just exceeds the 
aA'erage length of the staple of the cotton passing through. 
The previous pair of rollers are then set |^ in. farther apart 
than this, and the back pair -J- in. farther still, so that if the 
staple used is 1 in., then the distances would be 1^ in. 
between front and second, 1;^ in. between second and third, 
and If in. betAveen third and back. The above is a good 
plan to folloAA^, but variation may be introduced and the 
distances made slightly less, especially Avhere the cotton is 
soft and not heavy or Aviry. The following sketches, AA'ith 

See Appendix for Drawing :iuJ Drafting of Cotton Fibres. 



dimensions, will conA'ey a good idea of the arrangement of 
the rollers and their sizes for different classes of cotton. 
The dimensions of the top rollers in each case represent the 

Fig. 8. 

diameter before it is covered. Fig. 8 is for Indian cotton, 
Fig. 9 American cotton, and Fig. 10 Egyptian and Sea 
Island cotton.^ 

The following summary will represent the main points 
that are deducible from what has already been said : — The 

Fig. p. 

short-stapled cottons require small rollers and short distances 
between them ; as the staple increases in length the rollers 
must be enlarged and the distances increased. If a heavy 
sliver is being used, the distances between the centres must 

^ Vol. III. gives more complete details on this subject to which refer- 
ence mav be made. 

DR. 4 ]VING 


be greater than when a finer sliver is passing througli. If 
the draft is rather small, or, as it is sometimes called, 
" easy," and the sliver is fine, the distances between the 
rollers can be a little less than Avhen the draft is a high one 
and the sliver heavy. If the staple of the cotton is 
irregular, the best thing to do is to bring out the roving as 
fine as possible and use easy drafts. When a big draft is 
used, the rollers should be run slowly. As the draft is 
lessened a quicker speed may be run ; but it should always 

Fio. 10. 

be remembered that a large percentage of waste Avill result 
if a big draft and a big speed are run together. 

Principles of Draft. — -Now that we understand the 
general arrangement of the machine and the disposition of 
the rollers, it will be beneficial at this point to make an 
examination of the principles upon which the action of 
the machine is based. In the draw-frame there are two 
absolutely distinct operations, each one serving its own 
purpose, and producing a result totally different from the 
other. By combining the two operations into one process, 
we obtain the desired result of parallelisation of the fibres 
and regularity of the delivered sliver. The two actions 
will be considered in their order. 

Note. — See Cb. IV., Supplementary Notes, for draftiug of cotton fibres. 


It has alread}'^ been mentioned that the sliver passes 
through four pairs of rollers, each pair of which, after the 
first, is accelerated in speed, the result being an attenuation 
or drawing-out of the fibres. It is advisable to understand 
this clearly. In the first place, the sliver is fed to the first 
pair of rollers, which carry it forward ; the second pair of 
rollers now grip the cotton, and since they are revolving at 
a greater surface speed, they will take the cotton forward 
quicker than it is being given to them. It will be readily 
understood that if both pairs of rollers held the same fibres 
of cotton at the same time the fibres would naturally be 
broken ; it is to prevent this happening that the rollers are 
set apart a little wider than the length of the fibre. The 
fibres under these conditions, therefore, yield among them- 
selves under the action of the quicker roller, and in doing 
so their contact with each other sets up sufficient friction to 
cause each fibre, as it is pulled along, to straighten itself 
out into a comparatively level condition. The surrounding 
fibres are all being acted upon in the same manner, so that 
although the sliver as a whole is gradually being made 
thinner, its fibres are also being made to lie in the same 
direction as the length of the sliver. There is another 
point that it is as well to understand. By referring to the 
remarks on the setting of the rollers, it will be noticed that 
the fibres, when passing from one pair of rollers to another 
pair, are not immediately gripped, but lie free, as it were, 
between the two pairs. Such fibres are in reality being 
drawn in almost the same way as if they Avere gripped, 
because their ends are being acted upon by the fibres that 
are passing through the rollers ; in addition, the friction 
thereby set up carries them forward, and in doing so drags 
the other ends over the fibres that are being delivered, and 
so straightens them. The reason why the different pairs of 


rollers are not uniform in their distance apart is not far to 
seek. As the sliver from the card or comber is passed 
through, the first two pairs of rollers must be wide apart, 
because the sliver is thickest, and also, for the same reason, 
the least drawing action must be performed, so as not to 
damage the fibres, but to act on them gradually ; the time 
of leaving the grip of one pair and being gripped by the 
next pair must therefore be sufficient to permit of this 
being done. The sliver is considerably reduced by the 
time it reaches the last, or, as it is frequently called, the 
front pair ; so here we have the least distance and the 
greatest draft consistent with good work. 

Nothinsr has been said so far about this attenuatinsr 
process having the effect of obtaining any species of regu- 
larity in the thickness of the sliver, and to a thoughtful 
reader it is quite obvious that no kind of regularity can 
possibly be obtained by a purely drawing-out process as 
it exists in the draw -frame. If a thick and thin sliver 
were passed throi;gh the machine, their relative condition 
would be practically the same when delivered, no matter 
what amount of draft be given. As an illustration we will 
take an exaggerated example of an irregular length of 
sliver. Suppose Ave had a length of sliver 2 ft. long, each 
6 in. of Avhich had a diameter of 1 in., | in., \ in., and \ 
in. respectively, and this were passed through a machine 
having four of a draft, the result would clearly be that 
each 6 in. woiild be lengthened to 2 ft., and the cross 
section of each sliver would be reduced to one quarter of 
what it Avas on entering; we cannot reasonably expect any 
other result. The fil^res would certainly be in a more 
parallel condition, but the irregularity, so far as thickness 
is concerned, would still exist, because nothing has been 
done to lessen it. On the other hand, great irregulaiities 



can be introduced by carelessness in the drafting and 
■weighting of the rollers. 

The laying of the fibres in parallel order is a ver}' 
important duty in the draw-frame ; but the equalisation of 
the sliver is equally important. The process of doubling 
the slivers enables this condition to be obtained, and it 
must be repeated as often as is found necessary to obtain 
the required degree of regularity with the least strain on 
the fibres. The principle underlying the action was toiached 
upon Avhen dealing with the doubling of the laps in the 
scutcher (see vol. i.) ; but here we will go into the matter 
a little more fully, because it is the very foundation upon 
which regular yarn is made. If a number of slivers, say 
six, that may be very irregular in diameter, were placed 
side by side or made into one thick sliver, it is generally 
assumed that there would be great probability of the thick 
and thin places coming together; and the assumption 
would be a correct one even if it depended only upon 
experience for its sanction. One may make thousands of 
tests from the carded sliver, knowing beforehand its vari- 
ation from regularity, and if six are simply placed together 
the regularity will be improved. It may and does happen 
that two, three, or even six thick or thin places come 
together, but the chances of their doing so are very remote, 
and do not neutralise the argument that there is an 
enormously increased probability of greater regularity 
occurring. "We will try to demonstrate this by an example, 
and whilst taking numbers that are large and give a wide 
variation, it will readily be understood that, whatever 
numbers are taken, similar results will be given. Suppose 
a sliver was very irregular, and its different thicknesses 
were represented by the numbers 1, 2, 3, 4, 5, and 6. 
Such a sliver would be six times as thick in one place as in 


another. Now take, say, three slivers like this one, and 
place them side by side, and let us consider the question 
whether the irregularities will be reduced by so doing. 
Granted that the irregularities are equally disposed in each 
sliver, we can see at a glance that there is only one position 
they can occupy side by side so as to obtain their greatest 
irregularity, and that is when the thickest and thinnest 
places all come together; but when we know that the 
possible combinations of the different thicknesses in the 
three slivers are innumerable, it is easy to see that the 
probabilities are all in favour of regularity. A very simple 
illustration will make this point perfectly clear. A diagram 
is given in Fig. 1 1 and portions of three irregular slivers are 
shown at A, B, and C. Measv;rements of these slivers were 
made at seven points in their length, and the figures over 
each point represent the diameters to scale. It will be noted 
that greatest difference between the thickest and thinnest 
place is 12. Similarly the difference in B and C is 8 
and 19 respectively. Now if these three slivers are placed 
together and then drawn out to three times their length 
we shall obtain a sliver D which is equal to A, B, and C 
divided by 3. If the diameter numl^ers on A, B, and C 
are totalled at each respective line and divided by 3 we 
obtain the diameter number of the new combined sliver D ; 
and it will be noted that the new diameter numbers only 
represent a difference between the thickest and thinnest of 
6^, so that the new sliver is greatly improved in regularity 
upon each of the original three slivers composing it. If 
more slivers are combined the probabilities increase, and 
when we consider that irregularities in a sliver are also 
very irregular in their position along the sliver, we increase 
still more the chances in favour of regularity. Upon all 
this, the fact should be considered, that when six slivers 











are put through the draw-frame, six of the resulting slivers 
are passed through another head, and six slivers that are 
the result of the second drawing are passed through a 
third time. Each time they have been submitted to a 
draft of six, so that the total doubling the sliver has 
received is 6xGxG = 21C). The diagram in Fig. 12 will 

Fio. 12. 

perhaps make this clear. A represents six card slivers, and 
they are doubled, drawn out, and form one sliver at G. 
A similar thing happens at B, C, D, E, and F, and results 
in single slivers at H, J, K, L, and M. The six slivers 
G, H, J, K, L, and M are doiibled and drawn out to one at 
N, so that N is the result of 36 original card slivei's being 
doubled together. Each of the slivers P, Q, R, S, and T 
also represents 36 original card slivers, so that if these six 
slivers N, P, Q, li, S, and T are doubled together, the 
resulting sliver W is the result of doubling 216 original card 


slivers. "We can thus easily realise that in the operation 
of doubling we have a process upon which every reliance 
can he placed in obtaining regularity of the sliver, and 
that the principles upon which it depends are perfectly 

Stop Motions. — It is quite manifest that the value of 
doubling will depend very much for its success upon the con 
tinuous feeding to the rollers of the same number of slivers. 
If six slivers are fed this number must be maintained, for 

Fig. 13. 

if one of them broke, and it was not immediately pieced up, 
there would be a weak spot in the deliA^ered sliver of over 
16 per cent less than the adjacent portions ; so it is imper- 
ative that this should be avoided, and consequently in all 
draw-frames means are taken both at the front and back of 
the machine to prevent the possibility of such a thing 
happening. A variety of mechanical methods are employed 
to serve this purpose, one or two of which will now be 
given. Fig. 13 shows the first example. After the slivers 
have passed through the sliver plate from the cans, each 
one is conveyed over a lever B, centred on a pivot at A. 



The upper portion is formed as a kind of spoon, so shaped 
as to keep the sliver in position, and the lower part is 
made with a liook-shaped projection "a." The spoon lever 
B is so pivoted as to be almost balanced. Its heavier 
portion is, however, at the bottom. When a sliver is 
passed over the spoon there is sufficient weight, together 
wnth the tension in it, to counteract this weight and cause 
the upper portion B to be depressed ; and so long as a 
sliver is going forward or its tension is maintained, the lever 

Fig. 14. 

will always occupy the position shown in the drawing (Fig. 
13). Directly an end breaks, through a knotted portion 
not being able to pass through the sliver plate, and also in 
consequence of Aveak spots existing in the sliver, or when a 
can runs empty, the upper part is relieved of its weight 
and the low^er portion " a " immediately falls, and in doing 
so occupies a position which prevents a \ibrating arm " b " 
from working. This causes almost instantly the stoppage 
of the machine. Its precise action is worthy of a detailed 
examination, so, although the sketch (Fig. 14) is similar to 
the drawing (Fig. 13), it is reproduced here stripped of the 



accessories, and the stop motion only shown distinctly. An 
eccentric X, acting through the arm Y and the bell lever 
on the shaft Q, gives to the vibrating knife " b " its recipro- 
cating motion. Under normal conditions of working " b " 
passes through a short arc of a circle, and in doing so just 
misses the hook portion " a " of the spoon lever. It will 
be noticed that the eccentric arm Y is in two parts, Y and 
Z, and that they are connected by a pin at the centre V. 
The end "d" is in connection with the vibrating lever 
working on the shaft Q. The whole arrangement is so 
balanced as to offer very little resistance to the free 
movement of the two portions of the eccentric arm, and it 
therefore works as if it were one piece. If anything, how- 
ever, interferes with this freedom of movement — for 
instance, when an end breaks and the hook part "a" of the 
spoon lever B drops and prevents the motion of "b" — 
something must yield, for the eccentric continues to work. 
This yielding takes place between the two parts of the 
eccentric arm, and results in the lifting up of Y bodily. 
In this action it is brought into contact with the upper 
portion "e" of a knocking-ofF slide "f," which keeps the 
strap on the fast pulley. As " f " is lifted up, the shaft J, 
which is kept in position by the slide, is released, and a 
strong compression spring placed upon it, acts through 
suitable stops, on the strap-fork rod " h," and so changes 
the strap from the fast to the loose pulley, which stops the 
frame. The machine cannot be started again until the 
cause of stoppage is put right. Fig. 1-4 is given to represent 
the position occupied by the eccentric and levers when an 
end breaks. The eccentric is in its highest position, 
whilst the centre V has also been raised from its normal 
position. This clearly lifts up the arm Y, and when this 
is effected the stoppage of the machine is a comparatively 


simple matter. This motion is generally called the BdiCk- 
Stop motion. 

There is also in front of tlie machine another stop 
motion, its ol)ject being to prevent a decreased number of 
slivers passing through to the coiler. This reduction in 
the sliver may occur either through breakage or roller laps, 
i.e. when a sliver sticks to the leather of the roller and 
begins to be wound round it. Crossed and knotted 
portions may occur, and these also stop the machine. The 
slivers coming from the front roller pass over the plate K 
(Figs. 1 and 13) and down through the funnel L into the 
coiler can. The funnel L is carried by a lever centred at 
M, and is so balanced that any diminution of tension of 
the slivers will cause the funnel to lift up, which causes 
the other end N to fall, and in doing so it comes in the 
way of a vibrating feeder bar W. This bar is connected 
with the eccentric arm, and is actuated from it. Directly, 
therefore, that its motion is stopped, the two portions 
X and Z separate, as in the back motion, and the frame 
stops. If from any cause too much sliver is going through 
the funnel L, it will do so with difficulty and naturally 
depress it. This depression raises one end R of a lever, 
which is centred at P, and depresses the end S. A stop 
pin T on the vibrating feeder bar is by this means pre- 
vented from passing forward, and as a consequence its 
movement is stopped, and, through its connection, the 
machine also. The position of the weight E, on the lever 
is used to regulate the amount of sliver that can be passed 
through the funnel without stopping the frame. 

Another form of stop motion, made by John Hetherington 
and Sons, is shown in Fig. 15 ; enlarged views are also 
given of some of the details. The action is as follows ; — 
The sliver passes forward from the cans in the direction of 



the arrow, and is guided to the " single preventer " rollers G 
and H by the guide J. From here it moves towards the 
rollers over the spoon F ; through this spoon, means are 
adopted for stopping the machine when an end breaks or 
a can runs empty ; the enlarged view will enable the action 
to l)e understood. As the sliver goes forward, the spoon F, 
fiilcrumed on the knife edge 0', is pressed down upon the 

Fig. 15. 

rod 0, and in this position its other end is kept out of the 
way of a revolving spider Q. If an end breaks, however, 
the pressure is taken from the spoon, and its lower end, 
being the heavier, falls and comes into contact with the 
revolving spider Q ; through inclines on the face of the 
spider a sliding motion is produced which, actuating a stop 
rod, moves the strap on to the loose pulley, and so stops the 


The front stop motion acts as follows : — The sliver 
passes through a funnel E which is pivoted as shown in 
the drawing. It is balanced h}' means of the Aveight P, 
and this can be so carefully done that if too thick or too 
thin a sliver is passing through, the funnel will fall or rise. 
For instance, if a knot or twisted sliver tries to go through 
it will depress the funnel, this will raise the link K and so 
act on the lever L (see enlarged drawing) as to bring one 
end of it into contact with the 1st spider N and stop the 
machine. On the other hand, if too thin a sliver passes 
through, the link K will fall and come into contact with 
the 2nd spider and also stop the machine. 

It is very desirable, whenever an end breaks, that the 
piecing should be a direct one. It is not sufficient that the 
two ends are placed just in contact; there must be no 
interval whatever between the two. The fibres of one end 
must interlock with the fibres of the end to which it is 
pieced. To do this effectually we must arrange the 
machine so that the breakages occur at points where the 
piecings can be performed successfully, and Avith a minimum 
of trouble. If a breakage of sliver happened just as it Avas 
entering the rollers there Avould be considerable difficulty 
and loss of time, as Avell as Avaste, in making a successful 
piecing ; so, to prevent this, all draAV-frames are noAV made 
Avith Avhat is generally termed a " single preventer " motion. 
This usually consists of tAA'o rollers (see Figs. 1, 1.3, 14, and 
15) placed between the sliver plate and the spoon levers. 
The bottom roller is driven at a slightly slower speed 
than the back roller, so that the sliver betAveen them has 
a small draft and is in tension. Any breakage that occurs 
Avill almost invariabl}' be betAveen these rollers and the 
cans, so that immediately the broken end pa.sses betAveen 
the two rollers the spoon is relieved and the machine stoj)s. 



The piecing at this point is easily effected. The stop 
motion also acts much quicker with the single preventer. 
The tension of the sliver between it and the back roller is 
always uniform, and is sufficient to keep the spoons in their 
correct position, as they act instantly when the tension is 
removed. Without this motion the tension would be very 
irregular, since the sliver coming from the can would one 
moment be tight and the next very slack, and formerly it 
was no uncommon thing for the machine to stop simply 
because of slack sliver. 

Fir.. 16. 

Owing to the multiplicity of parts and the separate 
motions necessary to obtain thorough control of the 
machine by means of the mechanical arrangements just 
described, a large firm of machine-makers, Howard and 
Bullough, introduced several years ago a method of attain- 
ing the same, or even better results, by means of electricity. 
In order to understand the question fully, a drawing of 
a section of the draw-frame is given in Fig. 16. Here the 
sliver is shown in its passage through the machine, and it 
is quite apparent that the whole arrangement is of the 


most simple character. The principle upon which the 
action of the machine relies for its effective workina; is 
based upon the important fact that cotton is a very good 
insulator, or, in other words, a non-conductor of electricity. 
Electricity, generated by means of cells placed near the 
machine or wherever desirable, is conducted by wires to 
certain parts of the frame. This electricity is rendered 
inoperative so long as the opposite poles of the current are 
not allowed to be brought into contact Avith each other. 
Directly one part of the machine having positive electricity 
is connected to another part having a negative current, the 
current begins to circulate, and if means are taken to 
introduce in the path of the current some appliance like 
the electro-magnet, the machine can be readily stopped by 
its action upon catches or otherwise. In the drawing the 
single preventer motion is shown at AH. The bottom 
roller is continuous, and is supplied by electricity from one 
pole of the battery. The top rollers are in short lengths, 
each length having two slivers passing under it. They are 
connected to the opposite pole of the battery, and insulated 
from the rest of the machine. So long as the top and 
bottom rollers are kept apart by the slivers which pass 
between them nothing will happen, because the current is 
disconnected, and therefore powerless ; but immediately a 
sliver breaks, the rollers come into contact, and the current 
begins to flow through the electro-magnet P. This, as a 
consequence, is made sufficiently powerful to draw on one 
side a hanging catch X, which is thus brought into the path 
of a revolving cam 8. The motion of the cam is stopped 
by this action, and, in stopping, it actuates the strap shifter 
in such a manner as to stop the machine. When the 
sliver breaks in front of the frame, after passing through 
the rollers, the stop motion is arranged to be actuated from 


the calender rollers L and D. These are insulated in a 
similar manner to the back roller motion A and H, and the 
slivers in passing between them prevent the flow of the 
current ; but when a breakage does occur the rollers come 
together, and the machine is instantly stopped by the same 
electro-magnet and catch as in the first case. When the 
sliver ^vraps round the roller, either top or bottom, its 
immediate effect is to lift up the top roller. This completes 
the circuit by bringing roller K into electrical contact with 
the top clearer at the adjusting screw C, and so puts into 
operation the electro-magnet P. A similar result happens 
when the can is full. In this case the machine is stopped 
through the excessive pressure of the sliver in the can 
lifting the tube wheel slightly, and thus connecting the two 
poles of the battery. 

Whatever may be said against the introduction of 
electricity into a room which, in the majority of cases, is 
already greatly charged with the fluid in a most incon- 
venient and unmanageable form, its results in the attain- 
ment of the desired ends that necessitated its introduction 
have caused it to be held in high repute by all who have 
had occasion to use it. 

Top Clearers. — We now come to another detail of the 
draw-frame, viz. the top clearers, and they are an essentially 
important feature of it. Owing to the large amount of 
friction that is set up in a cotton mill, and the practically 
complete insulation of the whole ironwork in the building, 
the electricity that is generated is gradually accumulated 
until its effect on the A^arious machines and the fibres of 
cotton is to tend to draw them away, with the result that 
these fibres cause a wiriness to appear which, unless suitable 


humidifying influences are brought to bear on it, will con- 
tinue in the subsequent stages. In the draw-frame the 
slivers are untwisted and relatively parallel to eacli other, 
so that this effect of electricity is seen in the ease with 
which the fibres attach themselves to the rollers and are 
carried round. If this continues for some time the loose 
fibres will accumulate, and eventually, by gravity or other 
disturbing force, fall back on the slivers and be incorporated 
in them, to their considerable detriment. It is, therefore, 
necessary to clear the rollers by some continuous process, so 
that as little labour as possible is introduced into the 
operation. One method of doing this is by means of 
what is termed a stationary flat. A sketch of this form 
is shown in the section of the draw-frame in Fig. 1. It 
consists of a piece of flannel attached at the front and back 
to a portion of wood in such a manner that the Aveight of 
both wood and flannel rests on the top rollers. The rough- 
ness of the flannel naturally clears the rollers of the loose 
fibres and particles of dirt that may adhere to them. In 
the arrangement there are no means taken to carry away 
what is generally called the flat waste, so it gathers into 
lumps, and at last Avill drop into the slivers unless great 
care is taken in cleaning them at regular intervals of about 
two hours. To avoid this unnecessary labour, and the 
probable spoiling of the cotton, several good types of top 
clearers have recently been introduced, which minimise to 
a very large extent the inconveniences associated with the 
stationary flat. A typical form, made by Dobson and 
Barlow, is given in the accompanying drawing (Fig. 17). 
It consists essentially of two Avooden rollers covered with 
flannel, one of which rests between the front and second 
rollers, and is driven i)Ositively at such a speed as to prevent 
any damage to the leather rollers upon Avhich it rests, and 




yet so that it can clear them of their adhering fibres. The 
other roller simply rests between the third and back 
rollers. This roller will receive a turning motion, due to 
the fact that it is in contact with two rollers that are run- 
ning at different speeds. The friction set up between them 
clears the leather. It is quite obvious that the revolution 
of the two clearer rollers will, in addition to their cleansing 
action, also form around themselves in a kind of lap all the 
dirty and loose cotton they collect. There can, therefore, 


patent revouvihg clearer motion 

Fig. 17. 

be no accumulations with their attendant evils. The 
stripping of the roller is an easy matter, and only requires 
to be performed about once a week. This is a decided 
improvement upon the frequency of stripping Avhich was 
necessar}^ in the earlier form noted above. It will be 
noticed that the weight of the flannel-covered rollers must 
be carefully adjusted, for it is bad policy to allow the full 
weight of any arrangement, unless it is very light, to rest 
indiscriminately upon the top rollers. The requisite ar- 
rangement in the apparatus is attained by the employment 
of a balanced lever, pivoted as shown, one end of Avhich 



rests under the frame carrying the rollers, "whilst the other 
end is weighted with a movable weight. By this means 
the pressure required on the rollers is readilj^ obtained. 

Two other examples of clearers are given in Figs. 18 
and 19. Fig. 18 consists of a flannel A carried over rollers 
and driven positively; the friction of the flannel on the top 
rollers gives a cleansing action. The flannel is automati- 
cally cleaned by means of a reciprocating knife B resting on 
the flannel and being carried by an arm C centred at the 
end D of a lever cariied on a stud at E. An eccentric 
gives to the end F a to-and-fro movement which is trans- 
ferred to the knife B, and which enables the knife to scrape 

Fio. IS. FiQ. 20. Fig. 10. 

off the fibres, etc., collected by the flannel. This clearer is 
generally known as Ermen's Clearer. An improvement on 
it, known as Colling's Clearer, is shown at Fig. 19 ; the arm 
C has a double knife B, the fly and dirt scraped off the 
flannel A is collected in the receptacle shown, and thus 
prevented from escaping and falling down on to the 
emerging slivers. 

Full Can Stop Motion. — Fig. 20 represents a full can 
stop motion : it is very simple in jjrinciple and quite 
effective. As the can fills, the sliver presses upwards 
against the plate D, and this moves a plate F also upwards. 
In contact with the i)late F is a pin G, which is forced up 
until it comes into the path of a reciprocating rod 11 which 



it stops, and this stoppage, through tlie usual means, leads 
to the stoppage of the machine. 

A full can stop motion is also made Avhich acts when a 
certain length of sliver has been put into the can ; a worm 
and worm wheel, actuated from the rollers, cause a stop- 
piece to move forward on a screw until eventually it comes 
into the path of the reciprocating rod through which the 
stop motion acts. 

Metallic Rollers. — As their name implies, these rollers 
are of metal and intended to displace the top leather- 
covered rollers of Draw and Fly frames. They are fluted 
and spaced in such a manner that they practically gear 
with the bottom rollers, so that as the sliver passes between 
the two it receives a crimping effect. Claims are made 
that this is an advantage and adds to the elasticity of the 
yarn, but it is an extremely doubtful advantage even where 
it exists. A further claim made is that an increased pro- 
duction is obtained from the same speed of roller. This 
may be considered. In the first place the increased length 
delivered, due to the crimping, implies an increased speed 
of spindle, in order to put in the desired twists per inch. 
Moreover, because of the increased length, the draft must 
be regulated by running the back roller at a higher speed. 
These are two palpable results to be taken into account if 
a comparison is to be made between a machine working 
with and without metallic rollers. The cpiestion of their 
advantage in higher production may therefore be considered 
as a very small matter and of no practical value in a mill. 
It has also been found that the passage of the sliver 
between the flutes has a crushing effect on the fibres, 
though this has been modified in the most recent form of 
roller. The older form of metallic rollers simply consisted 
in the flutes, of the top roller resting in the spaces of the 


bottom roller. The next iiiipruvenicnt was to prevent this 
by turning collars on each end of top and bottom rollers ; 
these were of such a size that when the collars were in con- 
tact, the tops of the flutes of one roller did not come into 
contact with the bottom of the spaces of the other roller. 
This, however, did not prevent the crushing effect, so a 
further improvement was to cut at each end of each roller 
flutes and spaces that rested on and in each other ; but in 
the middle part of the roller, through which the sliver passed, 
a less diameter with smaller flutes and larger spaces was 
made ; this naturally is the best form of the arrangement. 

A section of Asa Lees' draw-frame is given in Fig. 21, 
and whilst it follows the usual design the stop motions 
are sufficiently interesting to be illustrated and described. 
The motion shown in the drawing is driven from the coiler 
shaft to a wheel on which is cast an eccentric A, on 
which fits the strap B, with its rod C weighted at D, 
with an extension on which is fixed a stud E. This stud 
E enters a V-shaped slot in the lever F, whose fulcrum is 
at C ; the other end of lever F is extended to I. The 
weight D maintains the stud E at the bottom of the V slot, 
and the lever F will simply oscillate to and fro freely. If 
the end S of the spoon or the end of the trumpet lever H 
comes into the path of the vibrating ends I or J, then the 
V-shaped end of F will be locked, and the stud E, being free 
to move, will slide up one of the grooves in the V, and on 
coming into contact with the arm K move its other arm. 
out of the locking position on the stop rod L, and so stop 
the machine. 

In Fig. 22 we have Dobson and Barlow's improved 
design of the machine illustrated in Fig. 1. It will be 
noticed that the rollers D, C take up the slivers from the 
cans and that the single preventer is placed between these 


rollers and the back roller F. Every sliver is thvis treated 
alike and all have equal tension, so that a more sensitive 
action is introduced on the spoon B, and the trouble caused 
by mere slackness of the sliver in the cans is no longer a 
source of annoyance through stoppage of the frame. 

Calculations/ — We now come to deal with the question 
of gearing, and the changes to be made in order to obtain 
the necessary productions and conditions for the purpose in 
view. A special drawing (Fig. 23) has been prepared, 
showing at a glance the arrangement of wheels and parts 
that constitute the chief features of the frame. Whilst 
mentioning that the machine represented is the method 
adopted by a leading firm of machine makers, Dobson and 
Barlow, it is almost unnecessary to point out that other 
makers' designs vary so little from it that readers can very 
readily adapt the following calculations to suit the gearing 
of the machines with Avhich they are directly in touch. 

The driving pulley shown on the outside of the frame 
end receives its motion from a line shaft above. The size 
of this driving pulley, of course, varies according to the 
speed required and pulley on the line shaft, but under 
ordinary conditions an 18 in. x 13 in. will be found a good 
standard to adopt, although 21 in. is frequently used. 
On the driving shaft, just inside the framing, is keyed the 
inside driving pulley B, or, as it is sometimes called, the 
bottom shaft pulley. In different makes of machines this 
pulley varies, but for our present purpose 16 in. will be 
^aken as its diameter ; it drives, by means of a belt, a pulley 
C on the front roller, whose diameter as shown is 12 in. 
From this point the driving of the whole frame takes place, 
and it is here where the machine is stopped and startetl, 

1 Full calculations and drawings of the chief machine-makers' draw- 
frames are given in the author's book Cotton Spinainij Calculations. 


.1 fast and loose pulley being used for that puiposc. 


front roller receives its motion direct from here, the other 
portion of the draw frame receivimr motion throuiih the 


gearing as represented in the sketch The front roller 
drives the back roller, through D, a compound carrier A E, 
and the wheel F; the second roller is driven from the 
back roller through the wheels G and H, a single carrier 
coming between them to preserve the direction of revolu- 
tion ; the back roller also drives the third roller by the 
wheels J and K, a carrier again being necessary here to 
turn the roller in the right direction. The system shown 
is the one generally adopted now, but in some makes of 
machines the two intermediate rollers will be found driven 
at the other end of the frame, ostensibly with the object of 
reducing any tendency to twisting, or at least of neutralis- 
ing it somewhat; but this is too palpable a fallacy to need 
explanation, and in all the best or newest designed machines 
the arrangement shown in the drawing is adopted. The 
calender rollers are driven from the front roller by a train 
of wheels whose continuation also drives the top coder 
shaft. This shaft, through pairs of bevel wheels, drives 
each coiler top, whilst it is also the means of conveying 
motion through bevels to the bottom coiler shaft, from 
which the coiler can itself is turned. Elevations of the 
roller gearing are given in Figs. 24, 25, and 26, by which 
their disposition is clearly shown. To facilitate making the 
calculation, and putting to a practical test the rules that 
are given, the following table of particulars is presented : — 

A Draft wheel . . . Yaiious, 40 to 90 teeth. 

B Inside driving pulley . . Various dias., say 14 in. 

C Front roller pulley . . . .12 in. din. 

D Front roller wheel . . . .20 teeth. 

E Crown wheel on top carrier . . . 115 ,, 

F Back roller wheel . . . Various, say 80 ,, 

G Back roller wheel driving second roller . ,, 45 ,, 

II Second roller wlieel . . . ,, 20 ,, 

J Back roller wheel driving third roller . ,, 26 ,, 

K Third roller wheel . . . ,, 21 ,, 

Diameter of front roller .... li in. 

,, second roller . , . . l| ,, 

Diameter of third roller .... H in. 

„ fourth roller . . . ■ l| ,, 

Ko. of slivers per delivery . . . . 8 ,, 

S])eed of front roller per minute . . . 264 revs. 

Obtaining the correct draft is the most important calcu- 
lation of the draw-frame, and this is done in the usual way 
hy finding the surface speed of the rollers and dividing the 
slowest into the quickest, which gives a ratio or number that 
represents the draft. The front roller has the quickest speed, 
whilst the back roller runs the slowest, so if we find their 
respective surface speeds the question becomes an easy one. 

(1) Draft between front and "1 _ExF x dia. of fiont roller 

back rollers J ~l)x Axdia of back 1 oiler 

_l]5x80x U_^ „ 
~ 20x5Sxl| ~^ 

It will be seen that this formula is made by simply con- 
sidering the back roller as the driver and working back 
to the front roller. In most cases this is the best plan to 

When the draft is already given or supposed, and it is 
required to find the necessary wheel, the same rule is 
applicable by substituting the required draft in place of the 
Avheel A, as this is the change place. 

,„, „, 1 r^ 1 n . F X E X dia. of front roller 

(2) Change or draft wheel A = D,„t , p ^ aia. of back 7olk-r 

8 X 20 x IJ in. 

For this result we take the next highest wheel, viz. 58 
teeth. In order to save repetition of the above calculation, 
finding the "constant number" is advisable. This is done 
b}' using the above rule, but leaving out the draft, or, in 
No. 1, by leaving out the draft wheel. 

F X E X dia. of front roller 

(3) Constant number = 

D X dia. of back roller 
^ SOx 115 x11 in. 


From this constant number we can obtain either the 
draft or tlie draft wheel as follows : — 

Constant nnniber 

(4) Draft ==- 

(5) Draft wheel = 

Draft wheel 
Constant number 


(G) Draft between first and second ~| _ H x F x E x dia. of front roller 

rollers J "GxAxDx dia. of second roller 

20x80x115x1-5 in. _ 

— 4 '23. 

',7) Draft between the second and back 

45 X 58 x 20 x 1-25 in. 

G X dia. of second roller 

H X dia. of back roller 

_45 X 1:|: in. 

20 xU in, 

= 1 

, ^ ^ , , ,, • n 111 J X dia. of third roller 

(8) Draft between the third and back = ,, — ^. ^. — . n — 

^ ' K X dia. of back roller 

21 X 1^ in. 

(9) Draft between the second and) _K x G x dia. of second roller 

third rollers / " J x H x dia. of third roller 

21 x 45 X \\ in. , ,. 
26 X 20 X li m. 

(10) Production in 10 hours is found as follows : — 

ilin. in 10 hrs. x revs, of F. R. x circumference of F.R. 
X gi-ains per yard of sliver 
36 in. X 7000 grains 

(11) The "constant" number for production may be obtained for any 

diameter of front roller by using rule No. 10, but leaving out 
the grains per yard of sliver. When this is done the produc- 
tion is found as follows : — Constant number x grains jier yard 
of sliver = production. 

,,^ . , ^ , . number of ends X weight of carding 

(12) AVeight of drawings -^^^^^ ° • 

(13) The draft can be found by the following rule : — 

number of ends put up x weight of carding 
weight of drawing 

(14) The draft wheel can also be found by the proportionate method : — 

^ ,, , , required weight x draft wlieel on 

Draft wheel = — ^-pr • 

present weight 

, , T^> ,., 1 1 present hank X draft wheel on 

(15) Draft wheel = ^' ^ — r^, , • 

^ ' required hank 



(16) The draft between the lirst ami .second x dralt butwecu the spcdikI 
and third x the draft between the third aud fourth — tutal (h-aft 
of the machiue. 

The horse-power required for driving the draAV-frame is 
generally put down as 1 i.h.p. for 12 deliveries. 

Draft. — On referring to rule (1) on page 41, it will be 
noticed that, although we are asked to consider the back 


3^." ROLLER. 
2~-° ROLLER. 

1 Ybdia 



Fio. 27. 

roller as the driver, we place the back roller among the driven 
wheels and the front roller among the drivers. This is puzz- 
ling to the student, so an attempt will be made to explain it. 
Consider the gearing of Brooks and Doxey's draw- 
frame in Fig. 27, having the following particulars : — 

Wheel A 20 T. drives B TOO T. "j 
,, C 40/70 T. ,, D 70 T. 
,, E 43 T. „ F 16 T. 
,, r, 22 T. ,, H 18 T. 
, K 22 T. „ M 48 T. 

Wheels N, J, and L are 
carrier.? and are not 
used in the calculations. 


The diameters of the rollers are stated on the drawing. 
Since draft simply means dividing the surface speed of one 
roller into the surface speed of another roller, we must find 
the surface speed of the front roller and divide it by the 
surface speed of the back roller. 

Now suppose the front roller has 200 revs, per min., 
its sui'face speed will be 

350 X n X 22 

350 X 1| >c --=^ = — ;: — ^ = 1512 inches per miu. 

* ' 8 X / ^ 

The surface speed of the back roller will be 

350 X A xC X circ. of B.R. _ 350 x 20 x 58 x If x y 
BxD ~ 100x70 

350x20x58x11x22 „-„ . , 

= — — - — ~ = 2o0 inches iier mm. 

100x70 ^ 

Surface speed of F.R. _ _^ 

Surface speed of B.R. 

I'll 2 

= £^^=6-04 total draft. 

We see from this that if we start our draft calculation 
from the back roller the front roller diameter must be 
on the top line and the back roller below. If we simply 
combine the two calculations into one we do so as follows : — 

^50 X 1 => X .. ^?£0iL?2.x 58 X If X -V 
SSOxl^x-,.. jQ-^— ^ 

350x11x22 350x20x58x11x22 

8x7 ■ 100x70x8x7 
_ 350xll x22xl00x 70x8x7 
~ 8 X 7 X 350 x 20 X 58 x 11 X 22 

= ^^°Jii. = 6-03 total draft. 

If, in thislast calculation, letters are used instead of figures, 

and using only the diameters of the rollers. Me obtain— 

. A X C X dia. of B.R. 

Dia. ofF.R. -r is — f=;^ = Draft 


Dia. of F.R.xBxD 

~Dia. of B.R. X AxC~ '^ 



It will thus be seen that a simple form of reasonin"' 
easily expluins why the fi'ont and l)ack roller diameters 
occupy the positions they do in rules for draft. The 
driving of the rollers of the draw-frame is not always the 
same. Several systems are adopted by machine-makers. 

fx. 18. 



B<t< •to.^e. 




Fin. 28. 



For instance, HoAvard and Bnllough's have the three systems 
shown in Fig. 28. The calculations for these systems all 
follow on the same lines as the examples already given, so 
it is unnecessary to work them out here. 










1 i ^ 



i s rt rt i § 

- -. _o _o _^ .3 


c3 rt's-'^*^^ "^Z^^-^^^I^ 

s ^ <u o o r"^ 


:5 'H S S £ SB 


Q >-i<! <1 <i; j: tq 




o 'S'oio Ci'iS ovr;o 


I^q (NCO-* -<}'"^-*^u-5 


o ooo o>^oco 


+j -l^+j-tJ +^fcC-t^-<--w 


33 tn m m oj H m x. 'r. v. 'r. 'j-. x k 

O 0-*(M (M OOU-.OOOOO 

,-1 (N(Nco CO co-^^«oi^ooaio 

/^"^"N /-^""^^^—N 


t- "■ K 

in (>] i^ 00 05 (>) (M -^ c^5 .^ r-i oi t^ »o in 



00 to m -t< Ol Ol .— 1 O 0) .— 1 O Oi CO l^ CD CD 




•V o 

fc- ^ . 

o CO i^ 1^ i^ C5 CO >--: Ti -»< 1^ >n CO T-H o oo 

(M ?1 OV C ■— 1 r^ C-l CD Ci CO t^ "-1 CD .— 1- o 

OCi00C01^«0CD>ncDCD>OO-3''*00 00 

S 1* 


o ^ 

CO XI CO -* CO ct C-. cci CO cr. 30 CO ^ CO 1— 1 t~- 

^ ID 

Ol CO CO >rn^ 1^ CO O t- CO O O CO O CO 1- 

ci '^ 

,— •.— lrtr-l.-ir^rH-M^.^7^1(M(NCI(M(N 

K " 

t^"?' . 

O '-^ y. 

— EV 





O a; 1^* 

• '-**'t^ 


?^ • ^ 

o o >n m >o o o o oo CO oo lo >n o c o 

r" 2i ■— 


^ ^ " 

— ^ 

§ ^ 

-g o 

O <E 7*1 




o >> 



Hx r^x -H- ^+f H-j- cc*r ?*r ?ax ^H?i H?i -*?i -^|?» -471 -♦?i -♦M -w 










■^ CO " 



■"^ CO • 






'^ CO • 


• h 


■ O) 


. y^ 



' * > 

* cr 



" C c; 


O - o 


t. •— t^ 

t^ c o 

o^ .^ 

't: '^ 

cs ci :: 





Object of Combing. — It is advisable, now that the 
subject of combing has been reached, to explain clearly the 
position this process occupies in the cotton - spinning 
industry, and also to give the reasons why the treatment of 
the operation has been placed after drawing instead of after 
carding, which is the plan usually adopted by textile 
writers. The exj)lanation necessitates a consideration of 
several important matters, a knowledge of which is almost 
essential in order to gain a clear understanding of much 
that will follow. 

In the first place, it must be thoroughly realised that, 
although cotton spinning has the one great aim of bringing 
an irregular mass of cotton fibres into comparative order, 
and making them into a strong I'ound thread, there is such 
a wide variation in the condition of the raw material, and 
also such a great range in its ultimate product, that the 
industry is, as a consequence, split up into several branches ; 
and these — principally for economic reasons, as well as 
structural details of machinery conducing to variation of 
result — might also be termed distinct processes. Four 
branches or departments may be enumerated, viz. waste 
spinning, spinning low numbers, medium numbers, and high 

NoTK. — A very complete set of jiractical notes on these machines will 
be founil in the author's book, Cotlon Mill Management. 




numbers. Waste spinning Ave are not concerned about, 
but the other three are part of our subject. The variation 
of the raw material used for either of these three purposes 
compels an almost corresponding variation in details of 
structure, in type of machine, r.nd in the arrangement of 
the order or extent of the various processes. To make this 
clear, a list of machines in their order is given for spinning 
from low numbers up to high numbers, and it will be seen 
how the conditions vary as the better classes of cotton are 

Nos. 3 to 10 

1. Double vertical opener. 

2. Single scutcher. 

3. Single scutcher. 

4. Single scutcher. 

5. Carding engine. 

Nos. 10 to 20 

1. Double vertical opener. 

2. Single scutcher. 

3. Single scutcher. 

4. Single scutcher. 

5. Carding engine. 

Nos. 20 to 50 : 

1. Single vertical opener. 

2. Single scutcher. 

3. Single scutcher. 

4. Single scutcher. 

5. Carding engine. 

Nos. 40 to 100 : 

1. Double opener, with lap p^art. 

2. Single scutcher. 

3. Carding engine. 

4. Draw-frame. 

Drawing frame. 

7. Slubbing frame. 

8. Roving frame. 

9. JIule or ring frame. 

Indian Cotton 

6. Draw-frame. 

7. Slubbing frame. 

8. Intermediate frame. 

9. Roving frame. 

10. Mule or ring frame. 

American Cotton 

6. Draw- frame. 

7. Slubbing frame. 

8. Intermediate fiame. 

9. Roving frame. 
10. Mule or ling frame. 

Egyptian Cotton 

5. Slubbing frame. 

6. Intermediate frame. 
Roving frame. 

8. Mule 

Nos. SO to 100 : Egyptian (double-carded) 

Double ojiener, with lap part. 6. Draw-frame. 

Single scutcher. 7. Slubbing frame. 

Breaker card. 8. Intermediate frame. 

Lap machine or Derby doubler. 9. Jack-frame. 

Finisher card. 10. Mule. 

^ Very full details of these systems are given in Vol. III. 

ir COMBING 49 

No. 100 u]iwan]s: Ska Island Cotton 

1. Single ojiencr, with laj>j>;iit. 

2. Single sctittluT. 

3. Carding engine. 

4. Drawing before combing. 

5. Sliver lap niaeliine. 

6. Ribliiin lap machine. 

7. Comber. 

8. Draw-frame. 

9. Slubl)ing frame. 

]0. luternieiliate I'ramc. 

] 1. Roving frame. 

12. Jack-frame. 

13. Mule. 

It will be noticed in the above lists that a more delicate 
treatment is afforded to the better-class cottons, whilst at 
the same time the operation is lengthened in order to obtain 
a greater regularity of yarn. "When Ave come to the highest 
numbers a new process is introduced called combing, closely 
following a jjreliminary course of draAving. It Avill be seen 
therefore that, throughout, the operation of drawing always 
follows that of the card. 

The above lists show that combing is a process which is 
only introduced into the spinning of high numbers or counts 
of very good quality. As its name imjjlies, it is primarilj' 
a combing operation — that is, every fibre of cotton is practi- 
cally isolated and straightened out, in which condition it is 
maintained afterwards by the contiguity of the surrounding 
fibres. The mechanical method of performing this action 
introduces of necessity several of the most important effects 
that make combed yarns possible, and gives to them the 
qualities that make them so valuable for the special purposes 
tliey serve. 

In the notice Avitli which the chapter on carding was 
prefaced,^ attention was called to the condition of the cotton, 
Avhich rendered that process an indispensable one in the 
manufacture of yarns, and those remarks are doubly appli- 
cable to the subject we Avere noAv considering; but, in 
addition, there are several important properties that combed 
yarns must possess, Avhich are practically unattainable by 

1 See Vol. T. 


the use of the card alone. These properties may be sum- 
marised as follows : — First, the fibres composing the yarn 
must be uniform in length. From the description of the 
card, we are prepared to acknowledge that this result is im- 
possible of attainment with the most perfect card of the 
present day, although it approaches much nearer to such a 
state than machines of a former period. An examination 
of a sliver will show beyond doubt that large quantities of 
irregularly stapled cotton exist, and it is to eliminate the 
shorter fibres that we have recourse to combing. Secondly, 
each fibre must be combed out as straight as possible, and 
must be maintained in this position. We saw that this 
combing action was performed very effectually by the card, 
but in that machine the perfect freedom of the fibres gave 
them full liberty to return to their naturally curly state 
because of the elasticity they possess. In combing, the 
large numbers of adjacent fibres mutually prevent this, and 
consequently Ave get the fibres laid side by side in a very 
level condition. Thirdly, we must have the fibres incor- 
porated together in some kind of order. They must overlap 
each other in such a way as to exclude any possibility of 
chance in their arrangement, and so that thorough reliance 
can be placed upon the strength that such an arrangement 
gives. In the card it was seen that the fibres were massed 
together in a haphazard fashion, without order or regularity 
so far as local conditions were concerned, and Ave can easily 
see that the binding of the fibres into a strand of sliver 
depended on no fixed method in their distribution, and to 
that extent there is a palpable Aveakness. In the comber 
Ave have a machine that renders these objects easy of 
accomplishment, but the combination of mechanism that is 
employed in attaining this successful result has given us a 
machine that may be considered one of the most complicated 


and ingenious of the many that are used in the cotton- 
spinning industiy. 

Before combing can be resorted to, the sliver from the 
card must be thoroughly prepared for what is necessarily 
a very delicate operation. In the first place, the arrange- 
ment of the fibres in the card sliver is (us already pointed 
out) of such a nature that the needles of the comber would 
be considerably damaged if an attempt were made to comb 
them in this state ; and, secondly, the great irregularity of 
the sliver would be re})roduced in the comber, and at the 
same time cause very unequal work to be thrown on the 
various organs of that machine, with a corresjionding waste 
of good fibres. These difficulties are avoided by first passing 
the card slivers through a draw-frame, with the result that 
the fil)res are partially parallelised and made more regular 
length for length. 

In doing this, another consideration must be taken into 
account. When the cotton is passed through the comber, 
it is combed out by a series of very fine steel needles, each 
row having a Avidth of from 7 to 11 in. This introduces 
the necessity of forming the slivers into a lap, so that a 
number of them lie side by side, making up the required 
width. The slivers are therefore taken from the draw-frame, 
and passed through a machine called a sliver-lap machine. 
Tills lap is very compactly made on a wooden core, and is 
about 9 to 12 in. in diameter. The cans are placed behind 
tlie machine, in number from 14 to 20, according to the 
size of lap desired, and tlie slivers guided through two or 
three pairs of small rollers having a slight draft. From 
these it passes forward between one or two pairs of calender 
rollers, whose object is to consolidate the lap and form a 
kind of fleece of the combined sliver. On emerging from 
these rollers it is wound on the wooden core, whose revolu- 


tion is obtained by the friction produced by its resting upon 
large revolving bowls. The selvedge is kept perfectly free 
from damage by circular end-plates, which revolve with the 
lap, and in this way friction of the lap ends with the framing 
of the machine is avoided. A stop motion is an absolute 
necessity in a machine of this description, and in all makes 
of it the slivers are first passed over spoons or some other 
system of stopping the machine, so that directly an end 
breaks or a can nms empty it instantly stops. 

The small draft alluded to above varies slightly, but 
it ought never to exceed two. 

The jiower required for driving is usually put down as 
\ i.h.p., and its production under ordinary conditions is 
450 to 500 lbs. per day. 

The speed of the driving pulley on the machine varies 
greatly in different makers' machines, but this depends 
very much on the arrangement of the gearing. ■ One large 
firm, who have a speciality in this machine, advocates 200 
revolutions, whilst another gives a speed of 120 revolutions 
jier minute. 

The remark ought to be made here that the principle of 
the machine just described is also the same for the Derby 
doubler, with which it is often confounded. 

The Derby doubler varies from the lap machine 
principally in the method of feeding the slivers. A long 
V table is used, the point being farthest away from the 
machine, and the cans are arranged on each side. This 
allows a much larger number of slivers to be used (22 to 
60), and also gives a greater width of laj) (10 in. to 37 in.). 
It was formerly extensively emj)loyed in making the laps 
for double carding, and is now chiefly used for the card in 
spinning waste and low numbers. The sliver-lap machine has, 
at present, almost entirely taken its place for other purposes. 


111 a large numlicr of cases the lap is noAv taken to the 
comber, but before considering its treatment there, it Avill 
be interesting to notice another system which is becoming 
more generally adopted in the preparing of fine yarns. 

The fleece taken from the sliver-lap machine, when held 
up to the light, is seen to consist of thick and thin places 
running in the direction of the length. This, of course, is 
caused through the slivers not being incorporated with 
each other, simply lying side by side, the outer fibres of 
each attaching themselves to those of the one next to it. 
This condition is unavoidable in the previous machine, and 
as it is not a desirable state of things, a remedy is sought 
in the following manner : — The slivers are taken direct 
from the card, and passed through the sliver-lap machine. 
The lap thus formed is taken to a ribbon-lap machine 
and draw-frame combined. The laps put up vary in 
number, but are generally six. The fleece from each is 
passed through four lines of rollers arranged on a similar 
plan to those of the draw-frame. Here they are sub- 
mitted to a draft — in this case six — and they emerge 
attenuated to this extent. They are now carefully 
guided along specially curved plates, Avhich bring them 
down on to a smooth table, upon which they are drawn 
lengthwise of the machine to its end. Each fleece in 
passing forward is brought under the other fleeces, as they 
are guided in a similar Avay, and by the time they reach 
the end of the machine they are all superimposed, and a 
combined lap is the result, with the irregularities reduced 
to an extent that shows little, if any, variation of light and 
shade when held up to the light. The combined lap is 
now passed through a series of calender rollers, which 
effectually consolidates them, and neutralises any tendency 
to licking. 

54 COTTON SPINNING chap, ii 

A good idea of this machine, as made by Dobson and 
Barlow, can be obtained by an examination of the foiir 
illustrations. Figs. 28a, 28b, 28c, and28D. It will be noted that 
Figs. 28a and 28b give the full gearing plan of the machine, 
the driving end when standing in front of the machine is 
shown in Fig. 28b, whilst the lap end is given in Fig. 28a. 
Elevations of the machine are represented in Figs. 28c and 
28d. Absolute smoothness of the curved plates and table is 
necessary, and frequently these are nickel-plated and highly 
polished. The drawings are sufficiently clear and detailed 
as to render a further descrijDtion unnecessary. 

Draw and Lap Machine. 

(Reference to Illustration). 

A Draft wheel. 

B Driving pulley. 

C Front roller driving wheel, 65 to 78 teeth, 

D Front roller wheel, 65 to 78 teeth. 

E Front roller wheel driving F. 

F Cliased boss carrier. 

G Back roller wheel. 

H Bai'k roller wheel driving second roller through 70^ carrier. 

J Second roller wheel. 

K Back roller wheel driving third roller through 52*' carrier. 

L Tliird roller wheel. 

M 3" dia. calender roller driving wheel. 

N 3" dia. calender shaft wheel. 

Lap end driving bevel. 

P Cross shaft bevel. 

Q 6" calender roller driving wheel. 

R 6" calender roller wheels. 

S Lap drum driving wheels. 

T Back lap drum wheel. 

U Back lap drum wheel driving front drum wheel. 

V Front lap drum wheel. 

W Chanire wheel for draft between wood lap roller and back 
roller, 37 to 62 teeth. 

Another example of the ribbon lap machine as made 
by John Hetherington and Sons is j^resented in Fig. 28e ; 
it will be found a simple matter to work out the calcula- 
tions from the gearing shown. 

V50 " u '■ 

Flo. 2.8a. 

Fio. 2SB. 


52 cabbies; ;,■■,-•;. -\ 
\ vuboo' L*p cbilLeijs 

FlO.. 2Sd. 



Less waste is made in the comber than would otherwise 
be the case, and the combing process is rendered easier 
by the uniform thickness and equally distributed arrange- 
ment of the fibres in the lap Avhich is fed to it. 

One horse-power is required to drive the machine, and 
its production is equal to about 450 to 500 lbs. per da\-, 
this varying with the class of cotton used. Its driving 
pulley speed is usually 260 revolutions per minute. 

For the very best yarn it is sometimes considered 
advisable to comb the cotton twice over. When this is 
done, the method generally adopted is as follows : — 

When a ribbon-lap machine is used, the cotton is passed 
through the following machines in their order : — 

1. Sliver-lap iiiacliiiie. 

2. Ribbon-lap macliiiie. 

3. Comber. 

4. Sliver-lap raacliiiie. 

5. Ribbon-lap Tnaeliine. 

6. Comber. 

If double combing is resorted to without the ribbon 
lapper, the order of the machines is as follows : — 

1. Draw-frame. 

2. Sliver-lap macliiue. 

3. Comber. 

4. Draw-frame. 

5. Sliver-lap machine. 
C. Comber. 

In double combing, a S3'steni is sometimes adopted as 
follows : — 

1. Sliver-lap machine. 

2. Comber. 

3. vSliver-lap machine. 
4 Comber. 

Description of the Comber. — We now^ come to deal 

with the combing machine itself. In the first place, a 
brief description as to its metliods of working will be 
advantageous, as it will then enable a better grasp to be 
obtained of the various operations, when describing its 
mechanical movements, which effect the result. 

The lap is taken from the ribbon-lap machine or the 


sliver -lap machine, .and placed on corrugated wooden 
rollers behind the comber. The machine is divided up 
into several sections called heads, six or eight being the 
usual number. Each head takes a lap, and is com})lete 
within itself, except that the driving arrangement of the 
complete machine is placed at one end. The lap is passed 
through the feed rollers intermittently, a short length at a 
time, this length depending upon the staple of the cotton 
being treated. In passing through it comes between two 
plates called nippers, and here it is held whilst a revolving 
cylinder partially covered with rows of needle comlis passes 
through it, its needles combing out all the short and nepp}' 
fibres during the revolution. Directly this is done, a 
fluted portion of the cylinder comes under the combed 
cotton, and at the same time a movable roller is allowed 
to drop upon it, and as the revolution continues the combed 
cotton is carried away and overlapped by a backward 
motion of a fixed detaching roller upon the lap that has 
previously gone forward. Just as this is being done, 
another comb is brought douTi, and lies in the path of the 
cotton as it passes onwards. The finished sliver is con- 
ducted to calender rollers, Avhich carry it forward. The 
various actions are repeated at a rapid rate, and a continu- 
ous fleece of combed cotton is delivered into coder cans at 
the end of the machine. 

In order that the above-mentioned actions may be better 
understood and illustrated, a section through the principal 
features of a machine is shown in Fig. 29. Although the 
drawing represents a double form of comber, it will be 
seen, as the description is given, that it diflers very slightly 
from the single machine, the difference being in the cylinder, 
Avhich contains oidy one set of combs and one flutted 
portion. (Compare Fig. 31.) 



The lap is jilaced in position upon the wooden rollers, 
whose revolution unfolds it and feeds the sheet of cotton 
down a highly polished plate in the direction of the arrow. 

The feed rollers F F carry it forward intermittently, the 
exact amount of the feed being regulated according to the 
staple of the cotton. The length fed goes forward between 
the nipper II and a cushion plate G. When sufficient has 


passed through the cushion plate, H descends and i)resses 
against the cushion G, thus holding the cotton very firmly. 
At the same time the needles B of the revolving cylinder 
come round into contact with the portion of the lap pro- 
jecting from the nipper, and, in passing tlirough it, 
thoroughly free it from its short fibres and impurities left 
in b}' the previous processes. Immediately the combs have 
finished their work, the rollers D D are actuated so as to 
turn back a short portion of the cotton that they took 
forward at the preceding operation, and just as this is clone 
the fluted portion C of the revolving cylinder has moved 
so that the leather detaching roller E can be brought into 
contact with it. This leather-covered roller, driven through 
friction by the fluted detaching roller D, with which it is 
pressed into contact by springs or weights, directly it 
touches the fluted segment C, takes the combed cotton 
forward. Simultaneously with this motion, the top comb 
comes down right in the path of the cotton, and so the fleece, 
in passing through it, is cleared of any short fibres that may 
have been held by the nipper. It will be seen that this 
action completely separates the combed portion of the 
cotton from the rest, and consequently a piecing must be 
cfFticted. This is done by causing the fluted detaching 
roller D to return a portion of the combed cotton it has 
taken forward, and, whilst the cotton is being drawn 
through the top comb, the roller E overlaps a portion of 
i:s delivered cotton upon that ])art that has been returned 
by roller D, and in this way a piecing is brought about. 

The forward and backward motions are obtained in the 
following manner : — A pinion P lides loose upon the end 
portion of the detaching roller D. The motion that P 
receives is transmitted to D through a clutch Avheel, which 
is actuated by means of a cam. Gearing into P is the 


quadrant shown in dotted lines, one end of which carries a 
bowl X, which receives its motion from the quadrant cam. 
This cam is formed so that it will cause the detaching roller 
D to turn backwards a given length and immediately turn 
forward a greater distance, a portion of which is used for 
piecing purposes. 

When this is done, the rollers D and E are at rest 
whilst the combing process is going on, and so the 
quadrant cam has a portion of its revolution idle so far 
as the quadrant is concerned, the clutch wheel at the same 
time being out of gear. When the combing process is 
finished, the clutch is put into gear, and the cam, acting 
on the quadrant, repeats the operations of backward and 
forward motions. The leather detaching roller E is put 
into and taken out of contact with the fluted segment C 
by means of the lever S centred at T. This is actuated 
through the lever E by the roller cam. 

The top comb is centred at 0, and its setting is effected 
through the set screw at V. 

The nipper H, centred at I, receives its motion through 
the lever L, and the rod K from the nipper cam, its 
movement being regulated by the cam on the one hand 
and by the setting screw Y on the other. The centre I is 
carried by a kind of cradle, upon which the cushion plate 
G is fixed. The cradle, being centred at W, allows the 
cushion plate to yield a little, and so be depressed quite 
close to the cylinder, when the nipper H presses against it. 
This enables the needles to effect a better combing, since 
the cotton is brought as near to them as possible. 

The foregoing is only a brief description of the comber's 
action. An examination of its details and full explana- 
tions of its operations will follow. 

When considering in detail the various movements of 


the comber and their results, the broad fact must be kept 
in mind that the machine depends entirely upon an inter- 
mittent action of the several parts, and each operation is 
so dependent upon the other that the slightest variation 
from the correct time of acting destroys the efficacy of the 
machine, or neutralises to a considerable extent the objects 
for which it is used. These several operations may be 
summarised as follows : — The feed motion, in which the 
lap is fed to the cylinder to be combed ; this is of necessity 
intermittent ; the length that is fed is also dependent upon 
the staple being used. The nipper motion, for holding the 
cotton during the process of combing ; it is intermittent 
in its action, and is arranged to allow the cotton to go 
forward after it has been acted upon by the cylinder. The 
actual combing operation follows next, by means of the 
rows of needles on the cylinder passing through the 
lap, after which the combed portion is taken on by the 
detaching roller. The backward and forward motion of 
the detaching roller; and, finally, the delivery of the 
finished material by calender rollers to the draw box and 
coiler. In addition to the above, there are several move- 
ments taking place either intermittently or continuously 
during the cycle of operations enumerated, but their 
dependence is so close on those given that it is not 
advisable to speak of them as distinct actions. Attention 
will first be given to the feed motion, and whilst the 
descriptions will be illustrated as fully as possible as the 
various details are dealt Avith, it is recommended that 
reference should be made to as many of the drawings as 
possible, in order to gain a clear idea of their relative 
positions and importance. 

A drawing is given in Fig. 30 showing the method 
generally adopted for obtaining the intermittent action of 


the feed rollers. A is the cylinder shaft, driven from the 
driving shaft by gearing through the large wheel B (see 
Fig. 29). On this wheel a disc plate C is fastened con- 
taining a pin D. A little distance from the axis of the 
cylinder is a stud carrying a star wheel E, into the teeth 
of which the pin D gears during a portion of a revolution 

'i^* ^ ^--G 

of the cylinder. The stud also carries a wheel F, whose 
pitch line is sliown, and which works into a wheel G on 
the feed roller H. This arrangement contains all the 
1 equisites for the desired action, and the following descrip- 
tion may be given as to its precise working. The cylinder 
is built up alternately of needles and a fluted surface — the 
needles for combing and the fluted surface for carrying 
the combed cotton away. Clearly the feed cotton must be 

11 COMBING (>^ 

delivereil to the cylinder just before the fluted segment 
comes forward ; this permits the delivered cotton to be 
drawn forward when the leather detaching roller comes on 
the fluted segment. This introduces the necessity of some 
setting arrangement that will enable this delicate adjust- 
ment to be made ; and so we find, as in the sketch (Fig. 
30), that the pin D is attached to a movable disc C, which 
in its turn is connected to the wheel B by the screw J. 
In this way the engagement of the pin D with the star 
wheel E can be very carefully and exactly made so as to 
fall in with the correct portion of the cylinder's revolution. 
The star wheel itself is made with five teeth, as shown. 
The pin D can only act on it during part of a revolution 
The remainder of the time it is stopped and made 
immovable through the concave formation of its outer 
circumference being in contact with a circular portion of 
the boss on C The connection of E with the feed roller 
by gearing is a necessity, because the machine must be 
capable of working various staples of cotton, and con- 
sequently there must be some facility for obtaining a 
variation of the length delivered by the feed roller H. 
This is readily obtained by the interchangeability of the 
wheel F with larger or smaller ones just as they are 
required. In this way the revolution of H can be regulated 
as to the amount it delivers to the cylinder. 

The above arrangement is the one generally adopted 
for a single nip comber, but when a "duplex" or double 
action comber is made, it is clear the feed roller must 
deliver the required length of cotton twice during one 
revolution of the cylinder, because in such a case the 
cylinder has a double set of combs and two fluted segments, 
and for each set the feed rollers must deliver material. 
The necessary effect is obtained by using two of the pins 


D in the disc C, one being placed diametrically opposite 
the other. In this Avay the star "wheel is acted upon twice 
during a single revolution of the cylinder. 

It will be understood that the intermittent action of 
the feed rollers must result in the same motion being given 
to the fluted wooden rollers upon which the lap rests when 
placed behind the machine. The requisite motion for 
doing this is transmitted by means of suitable gearing, 
w^hich ■will be illustrated at a subsequent part of the 

The next feature to demand attention is that part of 
the comber knoAvn as the "nipper." "We have already 
seen that this is a combination of levers which are brought 
into play for the purpose of holding the cotton firmly, 
without injuring it, whilst the cylinder combs out the 
portion of the lap that protrudes. In explanation of this 
action a drawing is given in Fig. 31, to which reference 
Avill be made in the following description. The principal 
features shown are similar to those in Fig. 29, but in the 
present case a single combing cylinder is exhibited ; to 
this the remarks will apply. A is the cylinder, whose 
fluted portion C has just made the piecing and carried 
forward the combed portion of the sliver. Directly this 
has happened, the feed rollers F F must deliver a sufficient 
length of the lap for the next combing operation, and at 
the same time the cushion plate G and the nipper knife H 
must open to allow this length to go forward. Immediately 
this is done, the nipper and cushion plate are brought 
together, and hold the cotton as the teeth of the cylinder 
pass through it. The details of this action will now be 

The method adopted for obtaining the opening and 
closing eflfcct of the nipper is generally by means of a cam, 


as shown at E ; this cam is grooved in such a manner that, 
acting through a series of levers, it produces tlie necessary 
movements of the nipper. A shaft M goes the full length 
of tlie machine, and on it are placed a series of levers L, 
one being used for each head. These levers have connected 
to them short connecting rods K, the upper part of which 
is connected by the pin J to one end of the nipper arm, 
whose fulcrum is at I. On the shaft M a lever N is fixed, 
which carries a bowl P, working in the groove on the face 
of the nipper cam. AVhen the bowl receives movement 
from the cam it gives a rocking motion to the shaft M, and 
this is transmitted by the levers L and connecting rods K 
to each nipper in the machine simultaneously : the cam is 
formed so that whilst revolving it keeps the nippers opened 
a sufficient length of time to allow the fed cotton to go 
forward. This action is shown taking place in the sketch, 
but as the cam continues its revolution, P will be moved 
a farther distance from the centre to P^ until the point Q is 
under the shaft. When this occurs, it will be seen, the 
nipper is closed, and in this position it will remain as long 
as the bowl is working in the concentric part of the groove 
of the cam QRQ. 

In connection with this intermittent action of the nipper, 
one or two considerations have compelled an arrangement 
of the levers to be made which is not so simple as the 
description just given would imply ; therefore it will be 
well for the reader to closely follow the description. The 
cam, as already described, causes the nipper arm to turn 
round I as the fulcrum, but this is only a part of its action: 
for directly the nipper H touches the cushion plate G (the 
cotton, of course, between them) the plate is forced down- 
wards. Now it will be noticed that G is carried by a frame 
or cradle which also carries the fulcrum I ; this frame is 



centred at W, and kept in position by a strong spring 
attached to a projection S and the framing of the machine. 
When, therefore, Gr is depressed by the pressure of H it 

Fig. 3-J 

Fig. 33. 

commences to move in a smaller arc of a circle, having W 
as its centre. This movement, it will be seen, brings Gr 
very close to the surface of the cylinder, and consequently 
enables the cotton to be combed very close to the nip. 
When the combing is finished the nipper begins to return 
to its original position ; but it must clearly be seen and 

11 COMBING 69 

understood that the cotton is not free from H and G until 
G occupies such a position that when the cotton is at 
liberty to be taken forward it is compelled to pass through 
the teeth of the top comb. This position is a very delicate 
one to adjust, and is regulated by means of the setting 
screw Y, which bears against a portion of the framing 
(shown by shaded lines in the drawing). 

If some arrangement of this kind was not made, the 
cotton, after being combed, Avould lie in the teeth of the 
cylinder or on the surface of the fluted segment. In this 
position it would be impossible to pass it through the top 
comb, and as a consequence the combing operation would 
give only a portion of the good results that are now 
obtained from it. By allowing the cushion plate, whilst 
still holding the cotton, to be raised away from the 
cylinder, we raise the combed fibres into a position where 
they are obliged to pass through the top comb as they are 
taken forward by the roller E. 

It was formerly the custom to place the nipper cam at 
one end of the machine, but the absolute exactness that 
was required by the simultaneous movement of each nipper 
was found to be slightly interfered with by the torsion 
that was produced in the shaft M. Strengthening the 
shaft was the first remedy tried, but ultimately the best 
arrangement was adopted of placing the cam in the middle 
of the comber, in which position it will be found in all the 
latest and best machines. 

We have already spoken of the great degree of accuracy 
that must be maintained between each action. Every 
means is adopted to obtain this precision. It will readily 
be understood that the closing of the nippers and their 
opening at the right moment is a very important matter, 
so we find that adjusting arrangements are provided to 


obtain the necessary regulation, as shown at U U and at the 
lower part of K, where it is connected by a swivel joint to 
the lever L. From these points the exact movement of H 
can be regulated. N in reality is a lever working loose on 
the shaft M, but is so arranged that by pressing against 
the screws U it moves the lever T, which is keyed on the 
shaft. This enables an adjustment to be made of a very 
delicate character when the cam is being set. 

An enlarged view of two kinds of nippers is showm by 
Figs. 32 and 33. In Fig. 32, as made by Dobson and 
Barlow, the cushion plate G is made with a dull, thin edge, 
and this is pressed against a strip of leather inserted in the 
nipper H, A firm hold of the full width of the lap is 
obtained by this means, and whenever the leather requires 
renewing it is a simple matter to turn it round or replace 
it. The other illustration (Fig. 33) shows a well-known 
method of getting the same results, but it is not of so simple 
a character as the previous one, and the replacing of the 
leather is more difficult, and requires greater care. 

The movement of the top comb is obtained in the 
following manner : — On cylinder shaft is fixed a cam X, 
shown in dotted lines, which, during a revolution, actuates 
a lever Z which rests upon it. This lever is fastened to the 
.shaft 0, to which is also fastened the top comb, so that the 
cam X, through the lever and the shaft, raises and lowers 
the comb as required. The comb's adjustment is obtained 
by means of the setting screw V, which rests upon a fixed 
portion of the machine. Several other adjustments can 
usually be made— for instance, the position of the top comb 
can be altered by the set screws at " a," a radial slot being 
provided to allow of this, and also the cushion plate G can 
receive a slight regulating through the set screw shown in 
the sketch. As a rule, small projections are fastened on 


the cushion plate in order to direct the lap on to the 
cylinder, as well as to prevent had selvedges forming. 

The next motion that recjuires consideration is that 
known as the "roller motion," or, more correctly speaking, 
the detaching roller motion. A general idea of its action 
has already been given, hnt a reiteration of its main features 
will enable its functions to be l^etter apprehended in the 
following explanation. After the combs of the cylinder 
have passed through it, the cotton is raised up out of contact 
with the teeth by the action of the cushion plate. Directly 
this happens, the comljed portion must be pieced up to the 
cotton that was acted upon during the previous combing 
action, and which has been carried forward hj the rollers 
D. To do this a part of the finished sliver must be returned 
so that the new portion will overlap it, after which the 
combined length will pass on to the coder. To obtain this 
backward motion, and then a forward motion, to immedi- 
ately follow it, has been the occasion for the disjilay of 
much ingenuity. Two of the principal means employed 
for effecting it will be given at this stage. The first is 
called the quadrant motion, and is illustrated in Fig. 34. 
Reference will also be made to Fig. 29, which represents 
this motion in another position, and is suitably depicted to 
serve as an illustration to this part of the subject. 

On the shaft F is fixed a cam called the "quadrant cam." 
Working in the groove on the face of this cam is a bowl X, 
carried by one end of a specially formed lever, whose centre 
is at B, and whose other end is formed as part of a wheel 
with teeth. From this feature it receives its name of 
quadrant. This toothed portion gears into a small Avheel 
P, which rides loose upon the detaching roller D, but 
which, when occasion demands it, can be put into gear with 
a clutch wheel fastened on the detaching roller, and in this 



Avay it gives motion to D. In the drawing (Fig. 34) the 
position of the cam is such that the wheel P is on the point 
of being reversed, or, in other words, receiving its backward 
motion. The amount turned back varies of course according 

Fig. 34. 

to the staple of the cotton being worked, but from | in. 
to 'I in. is usual. "When the bowl X has approached the 
centre as near as the groove will allow, it immediately 
begins to move outwards again, and it will be noticed that 
this outward movement Avill result in the roller D taking 
the cotton forward, and as the groove of the cam extends 
further from the centre of the shaft than tlie point where 


the backward motion begins, we can readily see that a 
longer length will be delivered forward than the amount 
returned. The difference between the two lengths will, of 
course, give us the total length of finished sliver the machine 
delivers per "nip," as it is termed. 

A clearer understanding of this complicated action may 
be obtained by referring to Figs. 35 and 36. Here we have 
the cam detached from the rest of the machine, and a plan 
of the mechanism is also given, so that the two actions can 
be easily followed. 

The black spots in the centre of the groove represent 
the position the centre of the bowl occupies as the cam 

When the bowl is on the line A the backward motion is 
on the point of beginning, as we saw in Fig. 34. At the 
same time the wheel P (Fig. 36) must be put into gear 
with the clutch F, which is keyed on the detaching roller 
D, otherwise the roller will remain stationary. So at point 
B (Fig. 35) the clutch H must be set so that the clutch 
wheel P is put into gear. As the cam revolves, the line C 
will represent the lowest position the bowl X can attain, 
and thus the backward motion is finished, and is immedi- 
ately followed by the forward motion. Usually, however, 
a very slight interval is allow^ed between the finish of the 
backward motion and the beginning of the forward part, so 
as to allow the leather detaching roller to be brouglit down 
on to the fluted portion of the cylinder, which has by this 
time been brought round, and occupies the position suitable 
for it. This interval is shown from C to D. Afterwards, 
onwards to F, the forward motion contiinies. "When this 
point is reached, as is shown in Fig. 29, the bowl is working 
on a concentric part of the cam groove, and of course no 
further motion of the detaching roller can take j)lace, and 



in addition the clutch is brought out of gear at the same 
time. It takes a slight interval to do this, and the drawing 
represents it from G to H. Just before the forward motion 
is completed, it is necessary to move the leather detaching 
roller E (Fig. 34) from the flutes of the cylinder, so as to 
be clear of the teeth as they come forward. The point E 
rejiresents that part of the cam occupied by the bowl when 

Fir.. 3.-). 

this is done. From H round to A again the cam is inopera- 
tive so far as moving the roller D is concerned, because the 
clutch is out of gear, and the movement of the quadrant is 
simply one of preparation for the next backward motion. 
The above description can be summarised, so far as regards 
the action of various parts of the cam, by the following 
table : — 




References to Fig. 35 

A Beginning of the backward motion. 

B Begiuuiug of the clutch going into gear. 

C Finish of the backward motion, and beginning of the forward 

D The leather detaching roller touches the Hated segment of the 

E Tlie leather detaching roller leaves the fluted segment of the cylinder. 
F Finish of the forward motion. 
G Beginning of the clutch coming out of gear. 
H Finish of the clutch coming out of gear. 

The above cycle of actions are practically the same 

Fig. 30. 

in a double form of comber, the only difference being 
that the cam revolves twice during one revolution of 


the cylinder, and also that the curves and intervals are 
slightly different, owing to the flutes and combs not 
being proportioned on quite the same lines as in the 
single form. 

Another method of actuating the detaching roller is by 
means of the notch wheel arrangement. This is the most 
general system of accomplishing the above purpose, and a 
drawing is given in Fig. 37 of its essential featm'es, so that 
its action can be clearly understood. A is the cam shaft, 
on -svhich is fastened the face cam B ; working in the cam 
groove of B is a bowl C, carried by one end of a lever 
centred at D, and which is shown in the drawing in dotted 
section lines ; the other end of the lever carries a catch E, 
a projection on whose end fits in the notches cut on. the 
periphery of a circular plate F, which also has D as its 
centre. It will be noticed that the projection on the catch 
E is square, and that the notches are also of the same 
shape, so that whichever w^ay the catch moves the plate 
must follow. The catch is kept in position by the spring 
G. An examination of the drawing will show that the 
movement of the cam B will cause a backward and forward 
motion to be given to the bowl C, exactly in the same 
manner as explained in the previous method. The position 
occupied by the bowl in the sketch shows it to be on the 
point of commencing the backward movement. This effect 
is transmitted by the catch E to the notched plate F. On 
the same centre D as the plate is an internal wheel H, into 
which gears a small pinion J fastened on the detaching 
roller, so that any motion given to the notched plate will 
be given in a less degree to the detaching roller, according 
to the relative number of teeth in each Avheel. Now, since 
the detaching roller is stationary except during the back- 
ward and forward motions, some means must be jjrovided 



of taking the catch out of gear with the notch Avheeh This 
is provided for in the drawing by connecting to the catch a 
short lever carrying a bowl K, which rests upon the outer 

Fig. 37. 

circumference of a cam plate L, fixed on the shaft A. So 
long as it is necessary for the catch E to engage with the 
notched wheel F, the form of the cam plate L is such as to 
have no effect on it, but when the forward motion is finished 
the catch must be taken out of gear, and so L is made with 
a curved projection M, which during its revolution comes 
under the bowl K and lifts the catch E out of the notch. The 



necessary position of the projection ]\I, so that the timing 
of the various actions is correct, can be readily made. 

The cylinder is shown in dotted lines at X, and also the 
direction of its rotation. 

In Fig. 38 Ave have an almost identical arrangement 
■\vitli the last one. A close inspection of the ilhistration 
will, however, disclose one or two detaik that are Avorth 


observiiif;;. In the first place, the cams are a little different 
at the points where the changes are made. This is a 
detail which does not affect the actual work the cam has to 
perform, but from a practical point of view it is advisable 
to avoid an extreme or sudden change in the cam's action, 
so that slightly more rounded corners are the best. In the 
next place, there is a different method of lifting the catch 
E out of gear with the notched {)late F. In this case the 
cam plate is dispensed with, and in its place a movable 
plate L is fastened to the back of the grooved cam B. Its 
position can be easily altered to suit the requirement by 
means of the screws ]\I, this plate during the revolution of 
the cam B being brought under the bowl K (which in the 
drawing happens to occupy the same position as the centre 
D of the notched wheel), and raises it so that the catch E 
is lifted out of gear with F. The cam B itself can be 
adjusted to the necessary conditions of timing, etc., by its 
connection with the plate P and bolt Q. 

The cylinder X is shown in its relative position to the 
other portions of the drawing. 

"We noAV come to the consideration of the means 
employed to raise the leather detaching roller E out of 
and into contact with the fluted segment of the combing 
cylinder. It will be understood from previous descriptions 
that the leather-covered roller is always in contact with 
the bottom fluted detaching roller D, and is kept so by 
means of weights or springs. During the backward motion 
its revolution is given to it purely by the friction of itself 
with D, but directly the forward motion commences, it 
falls into contact with the fluted segment of the cylinder, 
and this motion helps to cause E to rotate. Now, unless 
the surface speed of D is timed exactly like the surface 
speed of the cylinder, we can readily see that the roller E 



between the two surfaces will suffer considerable damage 
to its leather covering, as well as producing inferior results. 
To prevent this occurring, the cam grooves which actuate 
the quadrant or the notch wheel are very carefully designed, 

and in some cases every cam 
is milled out so as to ensure a 
perfect action. 

In Fig. 39 an illustration is 
given which will show very 
clearly the method of actuating 

Fig. 30. 

the leather detaching roller. A cam B fixed on the cam shaft 
has working in its groove a bowl X, which is carried by a 
lever L centred on the shaft N. The shape of the cam groove 
is such as to cause X to approach and recede from the cam 
shaft centre, and through this motion the shaft N receives 
a rocking action. On the same shaft N is fixed a series of 
levers K, which carry a stud P. This stud works in a slide 



formed in one end of the lever 8, -which has T as its fulcrum, 
and Avhose other end provides the surface upon which the 
ends of the leather detaching roller rest. AVe can now see, 
by following the movement of X as it is changed by the 

•vmm/m/imm»:» jn/^A 

Fig. 40. 

special form of the cam B, that its motion Avill affect, 
through the levers R and S, the jiosition of the roller P]. 
In the position shown in tlie drawing (Fig. 39), E is 
working in contact with the cylinder, and so assisting in 
the forward motion, l)ut the further revolution of B will 
lift it out of this position, and keep it so until the flutes of 
the cylinder again return to the necessary place for the 




repeated action. Adjustments can be made for the ex- 
treme exactness that is required, by the setting arrangement 
on the lever L, and also by the regulating of the stud P, 
Fig. 40 shows Dobson and Barlow's improvement on 

Fig. 41. 

their old method as depicted in Fig. 39. The levers have 
been rearranged and reduced in number, there are there- 
fore less moving parts, and an action free from vibration 
and of a more capable and delicate adjustment is ol^tained. 
Fig. 41 illustrates its application to the double form of 
comber, and this drawing may be found useful in connection 
with the description on page 58. 


A review may now be made of the {)revio\is descriptions 
of the various actions produced by tlie mechanisms just 
described, as far as they all affect that portion of the 
machine just above the cylinder. A series of drawings has 
been prepared illustrating that region of the machine, and 
from them the following explanations will be summarised. 

In Fig. 42 the feed rollers F have delivered a suitable 
length of staple, the nippers G H have closed up and hold it 
fast, and the cylinder teeth are upon the point of com- 
mencing to comb it. It will be noticed that the first rows 
of teeth are of a coarser pitch than the others that follow. 
The teeth themselves are also longer, stronger, and farther 
apart. Each row gradually becomes finer in this respect 
until the last row, which is composed of a large number of 
very fine needles close together. The object of this arrange- 
ment is to act progressivel}' on the cotton. The coarser 
needles prepare it for the finer ones which follow, and so 
the fibres are treated in a way that ensures the minimum 
of damage with the maximum of cleaning. The top comb 
is out of the way, and the sliver previously taken through 
is shown to be in a position ready for the backward 

Fig. 43 is practically the same drawing, but showing 
the conclusion of the combing action. It will be seen that 
the cotton already taken forward is quite separated from 
that just combed, so it is necessary to piece it up, and this 
is performed by an overlapping process. 

In Fig. 44 this process is shown. The roller D has 
been turned in a backward direction, and presents a length 
of cotton ready for piecing. The roller E has been placed 
on the fluted segment, and natixrally grips the end of the 
combed cotton, Avhich, during its revolution, it carries 
forward and overlaps on the returned portion, so that an 


effective joining is the result. Just before this action 
however, the nipper is opened and the top comb T comes 
down into the path of the lap, so that as E carries the 

Fig. 42. 

Fici. 43. 

Fin. 44. 

Fig. 45. 

cotton forward it is drawn through the needles of T and 
receives a good combing. Fig. 45 illustrates the finishing 
cycle of movements, and shows E to be on the point of 
being raised out of contact with the cylinder ; the rollers 

11 COMBING 85 

are delivering cotton for the approtaching combs, and 
directly this is done the nipper will close. 

It may be as well at this point to note that the amount 
of cotton taken forward by the detaching roller is not equal 
to the amount combed ; only the longest fibres are taken 
forward, so that the remaining fibres, augmented by the 
fresh fibres fed by the feed rollers, are again combed, and 
probably the majority of fibres, before being absolutely free 
to pass forward, are combed several times. 

In regard to the cylinder itself, it is now generally made 
so that the greatest facility is afforded for taking it to 
pieces and replacing very quickly damaged portions. As a 
rule, it is built up in segments around the centre, and is 
composed almost entirely of turned work, in which form 
duplication becomes an easy matter. Each row of needles 
can be removed for repairs without interfering with the 
rest, so that the former arrangement, which caused great 
difficulty in this respect, is completely altered and im- 

It used to be the common practice to make combers to 
take only laps up to 1\ in. wide, but at the present time 
improvements have been made in the way of guides on the 
cushion plate, M'hich prevent bad selvedges through spread- 
ing, and enable a wider lap to be used without increasing 
the length of the machine. But machines are not now 
confined to narrow laps. Up to lOi in. laps are frequently 
made both for the single and double form of the comber. 
Six to eight heads are the usual number forming one 

So much improvement has taken place in the comber 
during the last few j'cars that its ])roduction has been 
greatly increased, and the number of nips per minute, which 
used to be about GO, has gone up by degrees until now 


many machines at work giving good results are running 
at 96 and even 100 nips per minute. 

The comber takes to drive it about | i.h.p. to | i.h.p. 
according to the followincf table : — 

Ordinary . . 
"Dux)lex" . 

6 heads, f i.h.p. 8 heads, | i.h.p. 
6 „ 3 „ 8 „ I „ 

The pulley varies in size and speed according to the make 
of machine, one maker using a 12-in. pulley at 230 
revolutions, Avhilst another, with a 10-in. pulley, runs at 
325 revolutions. 

The following tables will be useful as conveying much 
valuable information on the comber, and presenting several 
points of interest to the student :— 

Ordinary CoiMber. 

o ^ 



Lb.s. per 


of combed 


Kind of Cotton. 








Sea Islands. 






>> )) 






>> )> 





7 '5 

Egyptian or American. 






>> ). 







The above j^roductions are based, as will be seen, on a 
speed of 80 nips per minute ; but if, as is now the case, 
machines run up to 100 nips, the production, of course, 
will be correspondingly increased. 



Duplex " or Dorisi.E Comber. 


5 s. 

P. . 



Kind of Cotton. 







Sea Islands. 






,, ,, 






>> !) 






Egyptian or American. 






>> )i 






)> )i 

Heavier productions than these can be obtained if more 
medium qualities are desired, but for the best work the 
above results represent a good average. 

The necessity for great care and experience in the 
management of the comber will be apparent to all who 
have followed carefully the description of this complicated 
machine. The setting of the various actions demands 
extreme exactness and a knowledge of the material being 
worked, or else considerable waste and probable damage to 
some part of the machine will result from such negligence. 
Waste may be increased more than is desirable by several 
causes — for instance, if the nijipers do not close at the 
right moment, that is, before the needles reach the cotton. 
This is generally spoken of as the nipper closing too late. 
Feeding too late produces excessive waste, and the angular 
position of the combs on the cylinder gives a similar result. 
The setting of the top comb in relation to the cylinder and 
nipper, unless carefully performed, yields more waste than 
is necessary, and consequently too close setting is to be 
avoided. The method of settinif and the gauges used for 


the operation vary of course with different makes of 
machines, so that little purpose would be served in giving 
the modm operandi. It will be sufficient to intimate that 
the following points are important items in the process : — 
The distance between the flute of the detaching roller and 
the front edge of the fluted segment of the cylinder ; the 
distance between the detaching roller and the cylinder; 
the distance between the edge of cushion plate and the 
fluted detaching roller ; the position of the cushion plate 
relative to the needles on the cylinder when the nipper is 
closed ; the distance apart of the feed roller and the fluted 
detaching rollers. In addition to these, facilities are pro- 
vided to enable each action to follow in its proper order. 
An index wheel is marked with numbers suitable for 
various settings, and by turning these numbers to the fixed 
pointer, the cams, cylinders, etc., can be fixed on their 
shafts, and adjusted correctly for any given set of con- 
ditions. In order to convey an idea of the numbers used, 
the following table is given, and represents the practice 
and arrangement of a well-known firm of machine- 
makers, whose work on this machine amounts almost to 
a specialty : — 



Feed rollers move at . . . 

4i to 5* 

6i* to 4i 

Detaching roller moves forward at . 

6* „ 6| 

6| „ 5| 

Detaching leather roller touches 

segment at 

6 „7i 

6| „ H 

Toji comb comes down at 

oi ,, 6i 

not lifting 4 

Nipper closes at . 

8f „ 9i 

lOi ,. 8i 

Clutch wheel in gear at . 

Of .. n 

Os „ Oi 

Those marked with an * in 

the Egyptia 

n column are 


considered very good times, Avhilst the ones marked with 
an * in the cohinin for American cotton are settings for 
the best quality. 

There still remain a numljer of details to be pointed out 
which have an important bearing upon the work of the 
comber, but they can, however, only be briefly mentioned 
here. In the first place, absolute cleanliness of the machine 
is a condition that cannot be too emphatically insisted upon. 
It must be clear to all that it is useless making an effort to 
obtain the very best fibres from any class of cotton, and 
arranged in a manner calculated to give the strongest 
result if, through carelessness or ignorance, some of the 
impurities taken out are allowed to be returned, or even 
when the effective capability of the machine is somewhat 
neutralised by the same cause. It is a comparatively easy 
matter to allow the comber to do its work indifferently, 
but the greatest care must be exercised if the very best 
results are desired from it. The continued or intermittent 
movements of the various parts set up vibrations and 
currents of air, having a strong tendency to dissipate the 
delicate fibres or the loose cotton taken out of them as 
waste. This must be j^revented, and we find that arrange- 
ments are made with this object in view, so that the 
collection of the waste and the cleansing of the rollers is 
performed in some systematic manner. 

On referring to Fig. 29, where a section of the comber 
is illustrated, it will be noticed that a circular brush is 
shown, the bristles of which are pressed against the needles 
of the cylinder to the depth of \ in. to \ in. Its revolution 
in the direction indicated freely clears the teeth of the 
fibres combed out of the cotton, and does it in a way that 
reduces to a minimum the possibility of fluff flying about. 
Its speed is greater than that of the cylinder, but the mere 


fact of its movement at the point of contact being the same, 
and also because the angular position of the needles is 
favourable, a thorough clearing is ensured. It must be 
understood that the cylinder can only be kept clean if the 
various rows of needles are perfectly joined up to each 
other. When even the slightest space exists between the 
rows of combs, an opportunity is given for accumulations of 
waste as well as interference Avith the work of the needles. 
Much damage is done to cotton through bad workmanship 
in this respect, and what is technically known as " flocking" 
is a frequent occurrence. The brush itself, through its 
continual movement and rubbing against the cylinder, 
naturally wears to a smaller diameter, and consequently 
requires readjustment both for position and speed, or 
otherwise the needles are not cleaned. The cotton on the 
brush, in its turn, must be taken away, or ultimately the 
cylinder would pick it up again, so a spiked doffer is used, 
whose teeth clear the waste from it. The dofler is stripped 
by means of a vibrating comb in a similar manner to the 
comb of the card, but, of course, running much more 
slowly. The waste comes off in a thin fleece or web, and 
falls into a suitable rece^Jtacle arranged to receive it. In 
some machines the web of waste falls from the doffer upon 
a slowly revolving shaft, which winds it in the form of a 
lap, and in this way it is collected in a very neat and 
compact manner. 

What is termed an aspirator is sometimes applied in 
collecting the comber waste. It consists of a revolving 
perforated drum or cage set close to the circular brushes 
in place of the doffer and comb. The cage is fitted with 
dampers, and a fan draws air through the portions left un- 
covered and so sucks the fibres from the brushes. These 
fibres, on the surface of the revolving cage, are carried round 

See Appendix for illustration and description of the ' ' A.'^pira'.or. " 


and are deposited on a travelling lattice which conducts them 
to a coiler at the end of the machine, where they are coiled 
into a can. At the same time all the dust and loose fibres 
flying around the comber are sucked into the aspirator, and 
the machine is kept much cleaner and the air purer by 
the device ; its cost probably keeps it from a general use. 

The feed rollers are kept clean by means of a brush 
pressing against them, whilst the detaching rollers are 
supplied with movable clearers covered with flannel. In 
order to prevent the escape of waste, and as a protection 
also, the cylinder, etc., is carefully covered in by tin 
casings. These are shown in Fig. 29, at Q and U, They 
are made so as to be readily removed in case of necessity. 

Nasmith's Comber. — A general view of the section of 
this machine is given in Fig. 46. Some of its features are 
similar to the Heilmann comber previously described, so 
a brief description only is necessary in connection with 
this illustration. 

The lap from the ribbon lap machine is placed on the 
two lap rollers and the sheet of cotton led downwards 
and underneath the roller B mounted on a plate D. The 
sheet passes over the edge of this plate, where it is 
nipped by the nipper E, whilst it is combed by the teeth 
of the cylinder A. From this point it is taken forward 
by the detaching roller X after this latter roller has made 
the overlap or piecing. The rollers Y now continue the 
forward movement of the sheet, and after it has passed 
through the usual funnel it is taken for^^•ard by the 
calender rollers "m " in the usual manner. 

The plan view Fig. 47 will now enable a few more 
essential facts to be noted. The driving shaft has a 23'* 
wheel driving a 90'' wheel on the shaft A or cylinder shaft. 
On this shaft is fixed the index disc, and on this disc is a 


pin to which is connected a rod which extends backwards 
to a lever on the shaft K ; the revolution of the index 
disc acts as a crank and gives a rocking motion to the shaft 
K. The sector or c^uadrant rides loosely on K, which thus 
acts as its centre. An eccentric on the cylinder shaft A 
gives a rocking motion to the shaft S. The cam on A 
operates the sector or quadrant through which the back- 
ward and forward motions are given to the detaching 
roller. The rest of the drawing partakes very much of the 
features with which the reader is already familiar in the 
Heilmann comber. 

By keeping these brief notes of the main features of the 
machine in mind, the following descriptions of the details 
will be easily followed. On reference to Fig. 48, a lever 
H is centred on J ; this lever carries a bracket to which 
a plate D is fixed and a bearing for the roller B. The 
same bracket is also designed to be coupled up at N by a 
link or adjusting rod M to a lever L on the rocking shaft 
K ; we can now see that the rocking shaft K will give a 
to-and-fro movement to the lever H and consequently to 
the plate D and its roller B. This plate D is the nipper 
plate and its roller is the feed roller. On the end of the 
feed roller is a wheel (see small sketch in Fig. 46) and a 
lever "n" carrying a pawl "p," the whole being cairied 
forward by the lever H. The lever "n" during this 
movement comes against an adjustable stop " q " and is 
arrested ; but lever H continues to go forward, and as a 
consequence the pawl " p " acts as a driver to roller B, 
and causes it to revolve and so feed the sheet of cotton 
forward in the direction the nipper plate D is moving. 
The stop " q " can be regulated to give the feed required. 
This brief description is sufficient to show us that the 
nipper plate is moved by a simple crank motion, no cam 

96 COTTON SPINNING chap, ii 

being required, and that the intermittent feed motion 
dispenses with the usual Geneva stop or star wheel feed. 
It will also be noted that there is always a sheet of cotton 
between the feed roller B and nipper plate D and ex- 
tending to the edge of the plate. The forward movement 
of the lever H results in the previously combed length 
being detached, and a new length is delivered ready for 
combing ; the backward motion of the lever H brings this 
newly-fed length of cotton into position for the needles 
of the cylinder to pass through it; but previous to this 
happening, the top nipper has come down and gripped the 
projecting cotton against the nipper plate, the grip being 
metal and metal, for no leather is used. It is interesting 
to note how the nippers are opened and closed. The top 
nipper E in Fig. 48 is carried by a lever centred at N and 
whose other end carries a bowl P. This end of the lever 
has a spring connected to a fixed part of the framing and 
its tendency is to close the nippers. As the forward 
movement of the lever H is finishing the bowl P comes 
into contact with an inclined foot " 1," and so opens the 
nippers against the pull of the spring. On the return 
movement the bowl leaves the inclined foot " I " and the 
spring at once closes the nippers, and the projecting cotton 
is gripped ready for the combing action. The inclined 
foot " 1 " is adjustable, so that the opening and closing of the 
nippers are easily regulated. The drawing Fig. 49 shows 
clearly the parts in their backward position, so that a 
comparison of the two drawings is advisable in following 
the description. In reference to the top comb a separate 
sketch is given in Fig. 50 that will convey some idea of 
its action. It will be seen that the top comb F is carried 
by a bracket bolted to a lever Gr that is centred on an 
extension of the lever H. The movement of H will 





therefore carry the top comb to and fro. On the top 
comb lever is a bowl or pin " 14," which, during the back- 
ward motion of the lever H, comes into contact with a 
fixed inclined arm " 15 " so adjusted that the top comb is 
raised and clear of the cotton, but on the forward move- 
ment nearing completion the bowl passes off the incline 
and falls, the extent of the lowering of the comb being 
regulated by the adjusting screws "x." The exact position 

Fig. 50. 

of the comb is obtained by the slots "b" in the lever G 
and the regulating screws "a." 

A brief reference may now be made to the detaching 
mechanism, for which purpose a glance at Fig. 51 may be 
made. The upper figure shows the cylinder needles 
combing the cotton, the nippers being closed and in their 
backward position. Before this combing action is com- 
pleted, and whilst the finer rows of needles are in action, 
the nipper plate commences to move forward, and as this 
movement is in the same direction as the movins; needles 



the combing action l)ecomes gentle and the fibres freed 
from strain. In the right hand side figure the needles 
have passed and the nipper j)late has advanced about half 

Fio. 51. 

way towards the detaching rollers, but just previous to 
this position, and as the last row of needles pass the 
detaching rollers, the latter are given their backward 
motion through the sector or quadrant ; this cotton that 


has been returned is projected into the space between the 
needles and the plain segment of the cylinder, and this is 
the more easily effected by reason of the forward position, 
relative to the cylinder, of the detaching roller X. The 
fleece is therefore bent under the bottoin roller, and a clear 
surface of cotton is presented for the newly combed 
portion to be pieced to it. As this is taking place, the 
nippers have opened and the released combed cotton rises 
and points in the direction of the detaching rollers are 
indicated in the sketch. The top comb in the meantime 
begins to fall. The detaching rollers now commence to 
turn forward and the nipper advances, so the combed 
cotton is laid on the previously returned portion and the 
whole taken forward until the nipper completes its move- 
ment ; the top detaching roller is moving away slowly 
during this period, whilst the nipper is advancing more 
quickly ; the detaching rollers continue to turn a moment 
longer after the nipper has stopped, and this commences 
the separation or detaching, an action that is completed 
when the nipper begins its return movement. It will be 
seen from this description that the nearness of the grip 
of the detaching rollers to the nippers is obtained by 
moving the top detaching roller towards the nippers and 
then away from them in the delivery ; this method entirely 
dispenses with any contact with the cylinder, so no fluted 
portion is necessary in this latter organ. A plain circular 
portion is therefore substituted. Both Figs. 48 and 49 
show how the movement of the top detaching roller is 
effected. The rocking shaft S, through the lever T and 
connecting rod U, operates the front levers and so brings 
about the to-and-fro movement of X, the roller being kept 
in contact with the bottom roller C by a weight hung by 
chains to the lever Q at Z. The eccentric, on the cylinder 


shaft A, and its connection to the rocking shaft S is sho^vn 
in the sketch Fig. 52. Adjustments are fully provided 
for in the setting of the roller and its precise moment 
of action. 

A further illustration in Fig. 53 is given to show the 
method of rocking the nij^per shaft K. It is a crank 
motion, but, owing to the sliding block moving along the 
lever " r " as the index disc 8 revolves, a variable leverasre 
results, giving a form of quick return motion. The small 
diagram in Fig. 54 will give some idea of the variable 

character of the movement produced by the crank action ; 
the commencement and ending of both the forward and 
return stroke are gentle, and all shock is avoided. It is 
characteristic of the whole machine that all the various 
operations are free from sudden actions. It remains to 
point out that this machine, whilst capable of combing 
a wide range of staple (it will equal the Heilmann 
in combing the longest Sea Islands cotton Avith an ad- 
vantage of doi;ble its production imder ordinary con- 
ditions), finds its greatest value in the combing of shorter 
stapled cottons, even cotton of ^ inch staple being easily 
manipulated. This arises chiefly from tlie fact that the 




piecing operation enables a very long piecing to be made, 
much longer than the length of the staple is itself, and the 
result gives a uniform and regular sliver not disfigured by 
the frequent uneven- 
ness seen in the 
Heilmann comber. 
The waste extracted 
is well under control 
and can be brought 
down to a Ioav per- 
centage on short 
stapled cotton. 

The setting of the 
various parts are readily carried out and the timing of 
the various operations clearly indicated, so that when the 
machine is once understood, it presents little difficulty to 
the practical man. 

Fig. 55. 

//" ^^^ 

Fk;. 56. 

Figs. 55, 56, and 57 illustrate the gauges and the chief 
dimensions requiring attention in setting. 

In Fig. 56 the gauge shown is -J^ inch and it is used to 
fi.x the distance of the top comb from the bottom detaching 


roller C, the simple purpose of course being to so set the 
comb that it is free from any contact with C. 

Fig. 57 represents the distance between parts that 
iiave an influence on the cleanliness and regularity of fibres 
of the cotton or the amount of waste taken out. The 
space between the bottom nipper plate and the surface of 
the bottom detaching roller is set by means of a stejDped 
gauge marked with numbers 8, 9, 10 up to 16, these 
numbers representing 32nds, so that we have spaces equal 
to 3^, -i^i \% up to i#, or a variation of settings from \ 
to \ inch. It may be noted that it is extremely difficult 
to set this distance to the \ inch gauge. At the same 
time a compensating setting that has some considerable 
influence is the distance of the feed roller from the bottom 
detaching roller, and by moving the feed roller nearer to 
the front of the bottom nipper less waste is made, and 
vice versa. Of course the greater the distance between 
the nipper D and the roller C the greater the percentage of 
waste. The usual settino-s are ijiven as 

\ inch for American. 



TO to f » Egyptian, 
to Vtt „ Sea Island. 

The gauge used (doctor gauge), that rests under tips of 
top comb and on the toj) of roller C, will usually leave a 
space between the gauge and the top of front detaching 
roller Y. The greater this distance the larger the percent- 
age of waste taken out, simply because the top comb will 
enter the web to a greater length of its needles, and so 
take out more fibres. In some cases, of what may be 
termed semi-combed yarns made from heavy laps and 
medium cottons, the top comb is set so that the gauge 
also rests on the front detaching roller Y. 

11 COMBING los 

Weight of Laps. — AVidth of lap lOi inches wide. 

For longest Sea Island cotton . . 12 to 18 dwts. ])er yard. 

,, other ,, „ . . 18 ,, 22 ,, ,, 

,, Egyptian Cotton . • 24 ,, 27 ,, ,, 

,, American ,, . . 26 ,, 32 ,, ,, 

Speeds. — The speeds are variable for different classes 
of cotton, but in general they may be taken as follow : — 

For best Sea Island cotton . . 335 revs, perniin., 86 nips ])er min. 

,, other ,,,,.. 350 ., ,, 90 

,, Egyptian cotton . . 370 ,, ,,95 ,, 

,, American ,, . . 390 ,, ,, 100 ,, 

Production. — This of course depends on several factors, 
such as speed, weight of lap, and amount of waste ex- 
tracted. If allowing 15 per cent waste when working a 
25 dwt. lap at 100 nips per min., the production of a 
six-head comber will be about 800 lbs. in a week of 50 

Power. — The Nasmith comber of 6 heads requires 
about I of a horse-power. 

An enlarged view of a portion of the Heilmann comber 
as made by Hetherington's is shown in Fig. 58. Incidentally 
the drawing illustrates a sliver stop motion which can be 
applied if necessary. The funnel C is made separate from 
the sliver tin and carried by a lever centred on a knife 
edge fulcrum A ; when an end breaks or no cotton passes 
forward, the other end of the lever at B interferes with a 
vibrating lever D, actuated by an eccentric on the cam 
shaft through the levers G and F, which releases the slide 
bar J, unlocks the stop rod K, and so stops the machine. 
Stop motions ought to be applied to all combers, their 
extra cost is quickly saved in the quality of Avork pro- 

A further sketch, from the same make of comber, is 



given in Fig. 59, which shows a full can measuring or 
stop motion. The bottom calender shaft actuates a lever 
which through the pawls turns the ratchet wheel, this 
revolves the worm gearing into the worm wheel, a pin on 

the latter is brought into contact with the slide bar and 
releases the stop rod, thus stopping the machine. Changing 
the ratchet enables a control of the length of sliver requii'ed 
to be easily obtained. 

Whitin Comber. — This comber, made liy Howard and 
Bullough, is an adaptation of tlie Heilmann comber. In 



essentials it is the same machine, but a simplification of 
various parts enables a higher speed to be run by the 
cylinder with a corresponding reduction in vibration. A 
section of the machine is given in Fig. 60, and the various 


features can be followed out by a reference to the names of 
the parts accompanying the sketch. 

It will be noticed that the section could practically be 
taken as representing the Heilmann comber, and the 
description of this latter machine is applicable. The 
improvement or rather the variations from the Heilmann 
consist chiefly in eliminating the movement of the leather 




detaching roller ; this roller is kept in one position, so no 
cam is required. As a consequence the cylinder speed can 
be greatly increased, so that up to 130 nips per min. are 
readily obtained. The speed of the cam shaft is kept 
down by the simple metliod of making a double cam for 

Fio. 60. 
A P]ain segment. 
B Brush. 
C Cylinder. 
D Dofter. 
E Nipper shaft. 
F Cam shaft. 

G Steel drawing-off or detaching roller. 
H Brass detaching roller. 
I Leather detaching roller. 
J Top feed roller. 
K Bottom feed roller. 
L Nipper arm fiilciuni. 
JI Top comb shaft. 
N Lap roller shaft. 

O Top comb arm. 
P Toi) comb. 
Q Cushion plate. 
R Dotler cover. 
S Nipper knife. 
T Waste chute. 
U Nipper cam. 

V Horse-tail holder shaft. 
W Calender roller shaft. 

X Nipper shaft lever. 

Y Half lap. 

Z Top feed roller saddle. 
A] Lap plate. 
Bj Waste packer. 

operating the nipper. The cycle of actions in combing the 
cotton can easily be understood on reference to Figs. 60a, 


60b, and 60c, which show the various organs at different 
stages of their movements. In general this comber is 
applicable for all purposes for which the Heilmann is used, 
whilst in addition it is also well adapted for reclaiming the 
large proportion of good fibres found in card strips, these 
latter amounting to sometimes 60 per cent of the waste. 

The setting of this comber follow-s somewhat similar 
lines to that of the Heilmann. For ordinary purposes it 
will be found that the nippers close at \\\ on index wheel. 
The detaching roller moves forward at 6, and the feed 
roller moves at from 3| to 6|^ according to the percentage 
of waste required to be extracted. The increase of waste 
may be obtained by setting the top comb closer, by 
feeding later, by putting more angle on the nipper knife 
and top comb. Weights of laps weigh from 400 to 550 
grains per yard. 

The production varies from 700 to 900 lbs. per week 
of 55 working hours. 

All the usual motions for reducing wear and tear, stop 
motions, taking up wear of brushes and running at correct 
speeds, collecting waste, etc., are applicable to this machine 
as to the other combers described. 

Nippers. — The nippers previously described are made 
by most firms who manufacture combers, but a special form 
has been adopted by Dobson and Barlow's which has 
advantages. An illustration is given in Fig. 61. It will 
be noticed that the nipper knife is covered with leather or 
other soft material This simple device enables the setting 
of the nipper to the needles of the cylinder to be extremely 
fine ; there is no danger of the needles being destroyed 
by contact with the metal as frequently happens in the 
pre\'ious types of nippers, so that, from this point of view- 
alone, the new method is a means of saving considerable 


time .171(1 money in the course of a year. In addition to 
its protective advantages, the finer setting enables a 
thicker lap to be nsed and a corresponding increase in 
production ranging from 30 to 60 per cent. 

Fio. 01. 

In Fig, 62 is presented a part plan of the comber, 
arranged with the object of showing the whole of the 
driving mechanism and gearing. From it can be traced 
the method of driving each separate action, and by using 

112 COTTON SPINNING chap, ii 

it also for reference, when reading tlie previous description 
of the machine, a more intelligent idea of its motions will 
probably be obtained. It will be noticed that one complete 
head is shown, and part of another. If the whole of the 
comber had been drawn in plan there would have simply 
been a repetition of the first set of rollers and cylinder to 
six or eight heads, each head of which works precisely 
alike, and delivers its own combed sliver. These slivers 
travel along a smooth, polished plate, parallel to the length 
of the machine, to the end, where they are passed through 
a draw-box, consisting usually of three lines of rollers, as 
shown in the drawing. At this point the combined slivers 
undergo a draft, which, of course, varies according to the 
purpose required. From the draw -box it continues its 
course, after passing through a pair of small calender 
rollers, to the coiler, a full section of which is shown on 
the right of the illustration. 

The driving of the machine takes place through the 
pulley shown on the left. This pulley is keyed on a short 
shaft firmly carried by the framing, so that no vibration 
can possibly exist. Its motion is balanced by a fly-wheel 
in order to prevent fluctuation of speed, owing to the 
intermittent movement of some of its actions. On the 
driving shaft is fixed a pinion, which, in the single comber, 
drives direct on to the cylinder through a large wheel of 
80 teeth. This wheel, in its turn, drives the cam shaft 
through a similar wheel, so that the cylinder and cam shaft 
revolve at the same speed. On the cam shaft is shown the 
disposition of the various cams, etc., an enlarged view of a 
portion of which has been given in a previous drawing. 
The calender roller is also driven from this shaft liy worm 
and worm-wheel arrangement. 

The feed roller is driven from the cvlinder shaft, as 




already explained, by the star wheel and gearing at A. 
A is a change place in order to alter the feed according to 
the length of the staple being worked. As pointed out, the 
feed roller drives the lap roller, the gearing of which is 
clearly shown in the sketch. 

The brush is driven from the driving shaft through 
carriers, which may be Avorked as a simple carrier, as 
represented in Fig. 62, or as a compound carrier, by 
simply changing the wheel into which the 34's on the 
driving shaft gear. 

On the opposite end of the cylinder shaft to the driving 
pulley is an arrangement of gearing by means of which the 
draw-box is driven, and also the coiler and doffer. Very 
little changing is done on the comber in regard to the 
gearing after it has left the maker's hand. The only 
places where this is effected are the feed wheel A and a 
change in the draw -box, but this latter is not often 
changed. Changes, of course, can be made to suit almost 
any special conditions, but these do not come under the 
general head of change places, and consequently they cannot 
be dealt with here. 

"Calculations, "^ — The following particulars of the gear- 
ing will enable the necessary calculations to be made : — ■ 

Driving shaft wheel .... 21 teeth 

Cyliniler index wljecl . . . . 80 ,, 

Cam shaft wheel . . . . . 80 ,, 

Cylinder wheel . . . . . 60 ,, 

Coiler wheel . . . . . 59 ,, 

Block wheel . , . . . 40 ,, 

Front roller wheel . . . . . 22 ,, 

„ M „ 34 „ 

rlO ,, 
Compound carrier in draw-box . , '4'" 

^ Several makes of combers are fully illustrated in gearing plans and 
the calculations given for them in the author's book on Cotton Sinnning 


Side shaft wheel . . . . .14 teeth 

Back roller wheel . 

. 50 „ 

Diameter of back roller 

. If in. 

Diameter of bottom block in draw 


2^ „ 

Diameter of calender in coiler 

2 „ 

Coiler driving bevel 

22 teetli 

Coiler bevel 

22 ,, 

Star wheel . 

5 ,, 

Cam shaft worm 


Cam shaft worm wheel 

14 teetli 

Calender mitre bevels 

20 ,, 

Diameter of feed roller 

5 in. 

Diameter of calender roller 

2| „ 

Feed wheel 

18 teeth 

Feed roller wheel . 

38 ,, 

I '5 

The above particulars are taken from a machine working 
fine Egyptian cotton. 

Draft of draw-box - 

22 X 40 X 50 X 50 X 2| in. 

= 5 -IS. 

40x34 X 45x14x1 
Draft from the calender in 1 _ 40 x 34 x 45 x 60 x 2 in. 

draw'-box to the coiler / ~22 x 40 x 50 x 59 x 2| in 
Draft from the feed roller ) _38 x5x 2x20 x 2| in. 

to the calender block j 18 x 1 x 14 x 20 x | in. 
Total draft of machine = 5 -13 x 1 -01 x 5 "52 = 28 "6. 

a -03. 


The total draft can also be obtained 1)}' finding the draft 
in one operation between the feed roller and the draw-box 
block, and multiplying it by the draft between the draw- 
box and the coiler. For instance — 

38 X 5 X 50 x 40 x 22 x 2g in. _ ^_ 
18 X 1 X 45 X .34 X 40 X % in. ~ "' 

27-826 X 1 -03 = 28 -6 total draft. 

38 X 5 X 60 X 2 
18 X 1 X 5y X ^ 

= 28-6 total diaft. 


Ill calculating draft it frequentl}'' happens tliat the 
actual draft differs from the residts arrived at by calcula- 
tion. This occurs chiefly in connection Avith machines 
dealing with slivers, such as the draw frame, sliver lap 
machine, ribbon lap machine, and comber. This diflerence 
must arise, because, for calculation purposes, the exact 
diameter of the bottom roller is taken as the basis. If the 
top roller were driven positively and exactly the same 
surface speed as the bottom roller there would be no differ- 
ence, but since the top roller is driv^en by the bottom roller 
through the layer of cotton between them, and the top 
roller is weighted, the thickness of the cotton passing 
through has some influence on the draft, and brings about 
a difference between the calculated draft and the actual 
draft. The two factors of thickness of slivers and weight- 
ing of the top rollers must be taken into consideration 
when dealing with the question of draft. An important 
factor that sometimes arises in this connection is the speed 
of the front roller. As a rule it is presumed that the top 
roller will be driven regularly and easily by the bottom 
roller through the friction of the fibres between them, but 
it will 1)6 clearly seen that this cannot always take place. 
If the weighting is not carefully adjusted to suit both the 
sliver and the speed, the fibres in contact with the bottom 
roller will travel forward quicker than the fibres in contact 
Avith the top roller, and there will naturally be a tearing 
away of fibres, thus giving rise to the condition termed 
"spewing"' as the sliver emerges from the nip of the rollers. 

Comber. — Difference in weighting the leather detach 
roller alters the waste. 32 lbs. on detach roller gave 9%, 
changed to 16 lbs. gave 12%, due to less grip and probably 
long fibres taken out. 

To find Percentage of Waste. — Have doffer comb at 

11 COMBING 117 

the bottom of its swing, now remove all A\aste at the hack 
up to the doffer comb. Break the sliAer at the draw-box 
calender rollers. Work the machine for, say, 40 Jiips, 
leaving the dofter comb at its lowest point. Now weigh 
respectively the sliver and the waste made. The two added 
will equal the original cotton, and the waste will represent 
a percentage of this total. For instance — 

Good sliver = 60 grains 
Waste = 15 ,, 

Total cotton = 75 grains. 
If 75 grains of lap liave 15 grains of waste, 

Then 100 grains of lap have ". , — grains of waste, 
° '■ I :> 

]5-<100 ^. , „ , , , , 

.•. — — - — =20 per cent 01 waste taken out. 
75 ^ 



Object of Fly-Frames. — Hitherto, the processes described 
here have all been directed towards obtaining a uniform 
strand of cotton, free from impurities, and whose com- 
ponent fibres approach within a reasonable degree of 
equality in their length. The actions that have been 
employed to produce this residt are Beating, Combing, 
Drawing, and Doubling, and the cotton in being subjected 
thereto has been reduced from an irregular mass of tangled 
fibres to, comparatively speaking, a condition of regularity 
and uniformity. 

The next process is one primarily intended simply as 
a continuation of the Drawing process, with or without 
the combination of Doubling. Owing, however, to the 
extreme delicacy of the sliver, any further reduction of its 
diameter would make it so weak as to practically j^revent 
its further treatment, unless such reduction Avere accom- 
panied by some action which strengthened it, so that it 
could withstand the strains to Avhich it would be sub- 
jected when undergoing the next steps in the process. 
In addition to this, we can easily understand, from what 
we have seen of the coder, that the use of this for coiling 
a much finer roving or sliver would l)c a very clumsy 

Note. — A very complete set of practical notes on these niacliiiics will 
be found in the author's book, Cotton Mill Management. 


ciiAr. Ill FLY- FRAMES iiO 

method, cand conse([Ucntly a slight twist is given to the 
reduced sliver, and it is then wound upon a bobbin. The 
small amount of twist given to the attenuated and ex- 
tremely loose sliver is sufficient to give it the necessary 
strength to enable it to be built up in the form of a 
bobbin, in which condition it is very convenient for further 
treatment. The drawing-out or reduction of the sliver, 
from the diameter as it exists Avhen passing through the 
drawing frame to a diameter suitable for the process of 
spinning, is so great, and results in such a weak sliver or 
roving, that a very delicate and gradual operation must 
be exercised. The steps, by means of which the reduction 
is made, depend upon the degree of fineness required in 
the resultant yarn, and consequently they vary in inimber. 

The machines used for effecting the attenuation of the 
sliver are called by the different names of Fly-Frames, 
Roving Frames, and Speeders : and according as one 
or more of these are used, we get specific names for the 
machines in each step. For instance, the first fly-frame 
used after the draw-frame is called a Slubbing Frame ; 
following this is the Intermediate Frame, and then the 
Roving Frame ; after whidi (for fine spinning) a liner 
roving frame is used, called a Jack Frame. The order of 
these machines for their several purposes will be seen on 
referring back to p. 48, vol. ii., where their sequence was 
shown for various number's ; and it will also be noticed 
that as the numbers or counts increase, so do the number 
of sets or passages of fly -frames. The objects of each 
machine are exactly the same, and so for all practical 
purposes their structure and mechanism are alike, the only 
difference being in the strength and dimensions of the 
various jiarts. 

The necessity that arises at this stage in the building 


of the roving or sliver in the form of a bobbin, also 
introduces in its wake complications of mechanism for 
automatically performing it, and these give to cotton 
machinery, from the mechanical point of view, one of its 
greatest sources of interest. 

The bobbin is made by winding the sliver round a 
wooden cylinder in layers until a suitable diameter is 
obtained. When finished it has the appearance of a cylinder 
with its ends tapered, in which condition it is the more 
readily, and with the least possibility of injury to itself, 
taken from one machine to another. The various problems 
connected with the building of the bobbin will be dealt 
with as thoroughly as possible as the description proceeds. 
It is sufficient at this stage to mention that the placing of 
layer upon layer of roving on the bobbin necessitates a 
varying speed being given to it in the flj^- frames, the 
delivery of sliver during the process being constant. . There 
is also a A'ariation in the length of ti'ansverse, to give the 
tapered character to the bobbin. These two motions 
present some very ingenious and mechanical problems, 
which it is advisable should be thoroughly understood, if 
more than a mere superficial knowledge of the subject is to 
be obtained. 

In regard to the twist put into the roving, it Avill be 
found that twisting is an inevitable consequence of this 
mode of forming a bobbin by means of a flyer ; but arrange- 
ments are made in the machine Avhereby the amount — ■ 
within certain limits — can be carefully regulated. For all 
practical purposes the twist or turns per inch is largely a 
question of experience, and depends upon several factors, 
which must always be taken into account when deciding 
ui)on this important j)oint. It is, however, never more 
than might be correctly described as a "slight twist" ; for 


it must be clearly understood that the merest excess of 
twist in the roving would prevent any further drawing in 
the following machines. From this we can readily compre- 
hend that, although the twisting is taken advantage of for 
giving cohesion to the fibres during the winding operation, 
the ultimate purpose of the yarn must be kept in view, 
and the twists must also be arranged so that the eflfective- 
ness of any future drawing action will not be destroyed. 
In following the passage of the cotton through the several 
fly-frames it will be remarked that the reduction of the 
sliver takes place graduall}'^, a little at each frame. The 
exact amount of the draft, of course, depends upon the 
cotton used, and also upon the numbers to be spun, but 
the accompan3-ing table will convey an idea of the usual 
course adopted in most mills : — 

Dr.AiTS FOR Indian and Amekican Cotton. 

Slubbing Frame . . . . . 4 to 5 

Intennediate Frame . . . . 5 ,, G 

Roviny Frame . . . . . Si ,, 65 

Drafts for Egyptian and Sea Island Cotton. 

Slubbing Frame ..... r> to 5^ 
Intermediate Frame . . . . iii ,, C^ 

Roving Kiame . . . . . Ci ,, 8 

Jack Frame . . . . . . 5i upwards 

There are many variations in drafts introduced in order 
to obtain identical results, and this is so true that probably 
no two men would use the same drafts in spinning similar 
classes of yarn. The reason for this is almost apparent : 
the number of machines through which the cotton must 
pass, readily permits a give-and-take policy in the arrange- 
ments of the drafts, and many men would alter the drafts 
of one or two machines only, instead of making a change 
on all the machines — provided, of course, there Avas nothing 


excessive in such a method. It is, however, always advis- 
able to let each machine do its own share in the work of 
making the roving finer : this will mean more time and 
labour in making changes, but better results are certain 
to be obtained by the extra trouble involved. It has been 
remarked that very little difference exists between the 
four passages of fly-frames, and what difference does exist 
is caused by alterations in the diameter of the full bobbins 
and in their lengths. In the slubber the bobbins are large 
and long; they get smaller as the roving is made finer, 
until in the Jack we have a bobbin Avith only one-quarter 
to one- fifth the amount of roving on it. The following 
table presents the usual practice in regard to this feature, 
l»ut it must be understood that slight variations exist on 
either side of the dimensions civen : — 















Indian and Low 









American . 
American and 







Low Egyptian . 

Good Egyptian 

and Sea Lslands 







American . 



Egyptian . . . 



Sea Island 



Description of Fly-Frame. — In the accompanying 
sketch (Fig. 63) a transverse section through a fly-frame is 
given. As will be seen, it is not complete, but a sufficient 
portion of the machine is shown to enable its essential 




features to be j)oiiitcd out and explained ; a full detailed 
examination of its actions will afterwards be made, and, as 

far as possible, illustrated by numerous drawings and 

On reference to the drawing (Fig. G3) it will be noticed 
that the sliv^er is fed to the rollers at the back, from bobbins, 
l)laced in a suitable structure called a creel. (This name is 



given to almost all arrangements in cotton machinery by 
Avhich bobbins are carried.) It must, however, be pointed 
out that in the slabbing frame this creel is not necessary, 
because in that case the cans from the drawing frames stand 
behind the machine, and their slivers are taken over a 
slowly revolving tin drum and passed on to the rollers. 
When bobbins are formed, a creel becomes requisite for 
carrying them, and its exact form varies according to the 
number of bobbins it has to accommodate, and also the 
number of heights in which it is convenient to make it. 
In the sketch a single row is used, and it is made in three 
heights. The full bobbins are taken from the previous 
machine, and placed on wooden skewers ; a shoulder on 
the skewer forms a resting-place for the bobbin, so that 
they are prevented from touching the long wooden rails, 
which go the full length of the machine, and which consti- 
tute the chief features of the creel. These skewers are 
pointed at each end. The upper portion enters the rail, 
and rests in a hole protected by a small iron ring to reduce 
friction, whilst the lower end rests upon a small recessed 
cup -shaped porcelain step, well glazed, so as to offer as 
little resistance as possible to the revolution of the bobbin 
as the sliver is drawn from it. As a rule, in modern mills, 
the rails of the creel are really made of long lengths of 
thin angle iron, with the wooden part screwed to them on 
their under side. This gives a much stronger creel, and 
one that is practically indestructible. The rails are carried 
by brackets fixed to upright rods J, which are in their turn 
securely fastened to the spring pieces of the machine. This 
arrangement of the creel enables the distance apart of the 
rails to be readily adjusted. The top of the creel is made 
so that a stock of full bobbins can be placed there ready 
for immediate use. 


The rovings arc taken from the bobbins E, F, and G, 
and passing over the guide rods X are led to the three 
lines of rollers A, B, and C In going through these they 
are subject to a drawing action the same as in the draw- 
frame, and to this extent it is simpl}' a continuation of 
that process. The amount of this draft for general purposes 
has already been given, see p. 120. 

"We are now in a position to see the need for the intro- 
duction of twist into the roving. The strain upon the 
sliver as the back roller takes it forward is considerable, 
when we take into account that it has to pull round the 
full bobbins in order to unwind itself, and so an additional 
cohesive power is given to it by twisting, in which the 
fibres are rather more firmly bound together, but not 
sufficiently so to interfere with the further drawing to 
which it must submit in the next operation. 

The drawn-out sliver or roving is taken from the rollers 
and threaded through the flj^ers L M, and wound upon 
bobbins w^hich loosely fit the long spindles that carry the 
flyers. The bobbins are all driven separately and in- 
dependently of the spindles, whilst the spindles also are 
individually driven. The spindle consists of a long steel 
rod, whose diameter varies from ^ in. in slubbing frames 
to -^-^ in. in jack or fine roving frames. Its length also 
varies according to the machine. In consequence of the 
speed at which it revolves, it requires to be well supported 
in suitable bearings, so as to prevent vibration and reduce 
friction as much as possible. It is, therefore, carried in a 
footstep bearing at Q, and in a bolster bearing at P. 
The bearings at these points are made as long as possible 
by means of the collars R R firmly fixed to the top rail 
R These collars are generally made in two lengths, and 
hence they are usually known by the terms of "long " and 


" short " collars. The toj) rail P has resting upon it all 
the bobbins of the machine. Since the spindles and flyers 
are stationary so far as vertical movement is concerned, 
the bobbins must be given this motion in order to have 
wound upon them the roving Avhich passes through the 
flyer. With this object in view, the rail P is given a per- 
pendicular movement, which constitutes the lift or traverse 
of the machine, and defines the length of the bobbin. It 
receives the motion through a rack T and wheel S, the 
rack being fastened to the rail and the wheel obtaining its 
movement through suitable gearing from the driving shaft. 

Fig. (U. 

Arrangement of Spindles. — All fly-frames are made 
with two rows of spindles, so disposed as to economise 
space and yet obtain a maximum number of spindles in a 
given length. Fig. 64 is given as an illustration of this, 
and from it we see that they are arranged in a zig-zag 
order — in many cases regularly so, but in other makes of 
machines the back row is not placed exactly midway 
between the centres in the front row, but a little to one 
side, the object being to facilitate doffing, etc. The " space " 
of the spindles is the distance from the centre of one to 
the centre of the next, as at A B or C D, whether we take 
the front or back row for the measurement. In many 


cases, however, the word " space " is replaced by the word 
"gauge," and instead of expressing the space of the 
spindles the machine is spoken of as having a certain 
"gauge." This to a certain extent is an advantage, because 
both rows of spindles are taken into account when the 
number or length of the machine is required, whilst in the 
first case only one row is expressed, and so it is necessary 
to double or half, as the case may be. In the sketch the 
" gauge " of the spindles would be denoted by saying that 
there are six spindles in the distance E F. This measure- 
ment, it Avill be seen, includes three spindles in each row. 

The following will show the two methods of denoting 
the space of spindle : — 

Distance of spindle from centre to in. in. in. in. in. in. 
centre . . . . . . 5^ 5f 6 6:^ 6i 7 

Or eqnal to 6 spindles in . . . 16^ 17| 18 18^ 19i 21 

Roller Stands. — A general view of the roller stand and 
the method of setting the rollers is given in the accompany- 
ing drawing (Fig. 65). On reference to it, it will be noted 
that the stand itself is a fixture on the roller beam X, and 
that it carries the fiont roller D. The other two lines of 
rollers are carried by separate bearings E and F, which are 
so arranged on the main stand that their distance from 
each other can be adjusted so as to suit various lengths of 
staple, thus giving every facility for setting ; after which 
they are readily fixed in position by means of the set 
screw. The recess G is occupied by the traverse rod, 
which moves the roving to and fro along the roller Mith 
the object of preventing an undue wear of the leather on 
the top roller. The top rollers — made either with a single 
or double boss, and also with or without loose bosses — are 
covered with leather in the usual way, and carried by an 
arrangement of cap bars, these latter not being used as 



bearings, but simpl}' as side supports for the ends of the 
rollers. They are made so as to enable the rollers to be 
readily taken out, and also so that any given set of bars 
may be bodily removed or turned over out of the way of 
the bottom rollers. It was formerly the practice to make 
the cap bars of cast-iron, but difficulties were experienced 
in several directions when so made, there being irregularities 
in the spaces, owing to moulding and casting, a slight 
damage necessitating an entirely new cap bar ; there was 

Fig. 65. 

also a lack of siinplicit}' in the adjustment of the slides for 
the different rollers, and, owing to the strain caused by 
screwing the loose parts togetlier, various other parts Avere 
twisted out of truth. Very great care in the making can 
obviate some of these faults, and irregularities are the 
more easily rectified by means of milling machines, so that 
some makers still adhere to this method of making the cap 

Ill FL y-FRAMES 129 

bars. Another system, now extensively followed, is that 
shown in Figs. 65 and 66. Here a support K is fixed to 
the roller stand, and carries a shaft or stud F, and on this 
shaft, at the necessary intervals, are fixed small brackets 
E (Fig. 66). These carry a long finger D of a pentagonal 
section (see H), and on it are threaded the cap nebs. A, B, 
and C, which form the supports for the top rollers. The 
hole through the nebs is similar to the section of the finger, 
and consequently they are prevented from turning, and, in 
addition, no side strain is introduced, because they are 
firmly screwed in position by set -screws bearing on a 
perfectly flat surface of the finger. The projections on A 
serve the purpose of a rest for the flat or clearers on the 

Fig. 66. 

upper part, and, on the lower one, adjust the centre 
vertically for the top rollers. As it may easily happen 
that carelessness or other causes might allow the finger or 
nebs to fall on the fluted rollers, precautions are necessary, 
and a slight recess is cut in F, and a tapered pin driven 
into the bracket E, this pin eff"ectively preventing any twist- 
ing of the bracket, and also holding it securely in position. 

Roller Weights. — A front view of the rollers and 
stands is given in Fig. 67. The sketch shows four spindles 
to a box, i.e. between two roller stands rovings are delivered, 
Avhich supply four spindles. Double-boss top rollers are 
used, and the cap bars would be placed to support the 
pivoted ends at L M, J K, and N P. The Aveights applied 
to give the required pressure between the rollers would in 




each case hang from G H. These weights, Hke many- 
other details of cotton machinery, vary in their amount, 
but the following may be taken as representing the usual 
practice : — 

Slabbing Frame . 
luterniediate Frame 
Roving Frame, single boss 
,, ,, double boss 



















In Roving and Jack Frames it is usual to have the back 
and middle rollers self -weighted, i.e. the weight of the 

Fig. 07. 

I'oUers themselves is sufficient for the purpose, in Avhich 
case the back roller is made much larger in diameter, as 
will be seen on reference to the next sketch. 

Diameter and setting of Rollers. — In this diagram 
(Fig. 68) a representation is given of the usual diameters, 
and the distance apart of fly-frame rollers for various classes 
of cotton • but it must be impressed upon the reader that no 
hard and fast line is to be drawn in respect to the 
dimensions given, and it will always be necessary to 
exercise judgment upon the special characteristics and 
staple of the cotton worked.^ 

1 See Vol. III. for fuller details of rollers, etc. 



Slabbing Frame. 


termediate Frame. 

• Roviiif; and Jack Frame. 

A Indian cotton 

B American cotton 

C Egyptian and Sea Island cotton 

D Indian cotton 

E American cotton 

F Egyptian and Sea Island cotton 

G American cotton 

H Egyptian and Sea Island cotton 

Note. — In the jack frame the distances apart of the rollers will be 
slightly greater than in the roving frame. It must also be observed 
that the top roller diameters are for the rollers when uncovered. 



Fio. 68. 

A general idea of the number of flutes in fly-frame 
rollers may be obtained from the following : — 

IJ in. dia. =5-1. 1^ in. dia. =60, !§ in. dia. =65. 

Twisting. — It has been remarked that directly the 
roving emerges from the front roller it undergoes a 
twisting operation — a somewhat necessary effect of winding 
it upon a l)obbin by means of a flyer. "We can now 
examine this action in detail. Wlien it is desired to put 
twist into any arrangement of filjres, etc., the essential 



condition is that one end must be held while the other is 
twisted. This statement is so expressed because in cotton 
spinning machinery the definition fits in with actual 
practice. A better method of defining how twist is 
produced may be by stating that one end of the substance 
must be revolved round its axis at a quicker rate than the 
other end, and in the same or the opposite direction. 
Even this definition might be simplified to some minds by 
saying that the angular velocities of each end must var}', 
when measured in the same direction, in order to produce 
twist or to cause an intertwining of the component parts 
of the substance. In the example of the flyer, this condi- 
tion is carried out in a very simple manner. 5igs. G9 
and 70 are presented to illustrate the description of its 
form and action. The spindle upon which the flyer is 
placed is a long steel rod carried by a footstep and a 
bolster. At the footstep end it is slightly reduced in 
diameter, as shown in Fig. 71, and its bearing is usually a 
recess fitted with a brass bottom ; or it can be made self- 
lubricating with a loose brass bottom part. In the sketch, 
however, an improvement is shown that enables the oiling 
of the spindle footsteps to be done very eff"ectively, and 
with a minimum of trouble, and at the same time a 
reservoir keeps the bearing well lubricated. The spindle 
is grooved for a short distance above the point where the 
small bcA'el is fixed, and a slight recess is cut in the upper 
part of the bevel wheel. By oiling at this point the oil 
descends by the groove to the reservoir, and in this way 
the necessity of going through the trouble of lifting the 
spindles is dispensed with. As shoAvn in the drawing, it is 
well-nigh impossible for dirt, etc., to enter the oil-cup. 

Collars. — The bolster bearing is also a very important 
matter, and consists of special bearings, called collars, 



securely fixed to the spindle rail, f^itlier long or short 
collars are used, and these are generally fastened as shown 

in Fig. 70. The lower portion of the collar fits a hole 
bored in the rail, and, by means of a shoulder, is bedded 
very truly to a milled facing provided for that puipose. 



It is then firmly fastened on its under side by a nut R, 
and frequently by a set-screw at the side of the snug. A 
long collar is shown in the drawing, but much difference of 
opinion prevails as to the merits or demerits of the two 
kinds. Each, how^ever, has its advantages, those of the 
short collar, being, of course, obvious. It is made of just 
sufficient length to serve the purpose for which it is in- 
tended, viz. to support the spindle ; but when we consider 
the great length of spindle above the bearing, and also the 
flyer, which is practically unbalanced throughout the whole 

Fig. 71. 

of the time it is building the bobbin, it will readily be 
seen that it is advisable to give more support than is 
obtained by the short collar, especially for high speed. 
Practical difficulties, however, used to stand in the way of 
their use : the correct boring of a long collar was no easy 
matter, and it was found that, owing to this irregularity, 
friction was developed, and more power was required to 
drive the frame. The bearing points were generally at the 
top and bottom of the collar, the intermediate part being 
barrelled or recessed out. Modern tools have now over- 
come these difficulties, and a perfect long collar can easily 


be made. The recess is also dispensed with by several 
makers, as dirt and fly accumulate on the inside and 
interfere with correct working. The spindle thus bears 
the whole length of the collar. Another disadvantage of 
the long collar is the fact that a large bobbin is necessary, 
owing to the larger hole required to fit over the collar, 
this, of course, causing extra weight. 

Flyer and Presser. — The flyer fits upon a reduced 
portion of the spindle by means of a socket, a recess being 
cut across the top, into which drops a pin J, inserted in 
the boss part of the flyer (Fig. 70). In this way the two 
parts are made one, so far as revolving together is con- 
cerned. The roving is inserted in a hole A in the flyer 
top, and passed through a small opening B in the side. 
The mere fact of passing the roving through this last 
opening gives to the flyer its ability to produce twist, for 
B is clearly out of the centre of the spindle, and describes 
a small circle as the flyer revolves. The other end of the 
roving is held by the roller; every revolution of the 
spindle naturally gives a twist, and, according to the 
relative speeds of the flyer and front roller, we get varying 
degrees of twist — generally expressed as twists per inch. 
The length of the bobbin to be built, and the weak nature 
of the roving, necessitate that it should be guided on the 
bobbin at a point much lower than where it emerges from 
the hole B, and we thus get a long arm, made hollow, 
down which the roving is passed. From the bottom of 
this arm it is wound round, and then threaded through 
the eye of a projecting arm loosely attached to the flyer 
leg, and from this it is drawn forward by the Ijobbin. 
The projecting piece D is termed a presser, its function 
being to give as light pressure to the roving on the boljljin 
in order to obtain a firmer result. It is specially arranged 


to do this, and reference will now be made to the method 
adopted ; but it ought to be first remarked that, in order to 
balance the arm C, another arm H is made on the opposite 
side, of such dimensions that its weight balances that of 
the hollow arm. 

The Presser and its Functions. — The "presser" 
is composed of two parts — one is the projection called the 
paddle, or presser, and the other a thick wire rod running 
up the side of the leg, and centered at E, the two being 

connected at F. The paddle D is made to fit loosely over 
the flyer leg, and as E is capable of swivelling from its 
centre, which is practically that of the arm C, it is clearly 
seen that any movement of the wire rod will be transferred 
to the paddle. To illustrate this further, a plan view of a 
flyer is given in Fig. 72. G is the rod running up the side 
of the leg, H is the paddle working round the leg F as a 
centre ; G also, for all practical purposes, works round the 
same centre. The weight of G is greater than that of the 
paddle, and, in addition, it revolves at a farther distance 


from the centre of the spindle, so tliat it has a greater 
tendency to fly away from that centre. It is free to do so, 
but is guided by its connection to F ; tlie centrifugal force 
thus tending to move G outwards being resisted by the 
bobbin, which prevents the other end of the paddle moving 
inwards. A pressure is consequently set up at H, which 
causes H to press against the roving on the bobbin, and we 
therefore get a more solid bobbin. 

As made in our machine shops, the flyer is perfectly 


c. ... 



IlG. 73. 



balanced when the presser occupies the position shown at 
H in Fig. 72; but this is the only place where such a 
condition exists. The presser, being a movable piece that 
is continually changing its position, disturbs the balance 
directly it assumes any other position than that shown. 

A diagram is given in Fig. 73 from which the foregoing 
description may be supplemented, and the unbalancing 
character of the presser explained. C is the bare bobbin, 
D its middle diameter, and E the full diameter. The flyer 
leg is marked at F, and the paddle, in its several positions, 


at J F, H F, and K F, tlie respective positions of tlie Avire 
rod being shown at G\ G, and G^^. "When the paddle is 
pressing on the bare lioljbin at J, the Aveight is at G^\ at 
its farthest distance from A. In this position it is exerting 
its greatest tendency to fly outwards, and, as a conse- 
quence, the paddle will press the more firmly at J. As 
the bobbin is building the increasing layers will move the 
paddle outwards, this action, of course, bringing G nearer 
to the centre A, in which position its surface velocity is 
not so high, and therefore it exerts less force to fly out- 
wards, so that the pressure on the bobbin at K is less than 
at H or J. It must not l)e forgotten, also, that the paddle 
itself has a tendency to fly away from the bobbin owing 
to centrifugal force, and is only prevented from doing so 
by the superior weight and the greater distance of G from 
the centre A. This tendency on the part of the paddle 
increases as the bobbin fills, so that as the centrifugal force 
of G decreases, the same force in the paddle increases, and 
both these result in a diminished pressure of the pressor as 
the bobbin fills. By altering the relative weights of the 
Avire rod G and the paddle, almost any degree of firmness 
or softness can be obtained on the bobbin. 

It will readily be comprehended from the above reason- 
ing that this altering of the centre of gravity of the presser 
brings a slightly additional weight nearer or farther away 
from the centre, and so disturbs the balance of the flyer. 
With a single pressure this is inevitable, and as nothing 
very serious results from it, it is almost ignored. Some 
time ago, however, an attempt was made to obtain a 
perfect balance throughout the building of the bobbin by 
the introduction of double pressers, one on each arm, but 
they are never made at present. 

The different surface velocities of the bobbin and tl:c 




flyer cause a nibbing action l)etween the pressor and the 
roving on the bobbin, and everything is done by careful 
workmanship to neutralise the evil eff"ects that may arise 
from this cause. In a double flyer this evil would be 

Flyer Leg". — The slot in the flyer leg is usually made 



straight, as shown at A in Fig. 74, but for 
high speeds and fine rovings there is an 
advantage in making it with what is called 
"winding" in it — that is, with a slightly 
curved form, as at B. This prevents the 
centrifugal force sending the roving through 
the slot. A flyer, it must be said, is one of 
the most highly finished appliances in cotton 
machinery, and a very large number of opera- 
tions have to be gone through before a 
finished article is obtained. It must be per- 
fectl}' smooth all over, and made of the finest 
material, to prevent accumulations of fly, etc., 
which would be fatal to good yarn. 

Previous sketches will convey some idea 
of how the driving of the spindles and bob- 
bins is performed, but a complete view is given ' 
in the following sketch (Fig. 75). A is the ^'-'- '^• 
bobbin wheel on the driving shaft (but not driven direct 
from it) ; the two rows of bobbins are connected by a \ya\x 
of wheels C and D, which necessitate the arrangement of 
the bevels as shown at E F and H G ; the spindles are driven 
direct from the driving shaft through the wheel N, which 
is fixed to it, the necessary gearing being shown in the 
latter case simpl}' by dotted lines. 

A description has been given of the flyer and the bo])bin, 
with their driving arrangements, sufficient to enable the 



following explanation to be made of the method adopted 
in placing the roving upon the "wooden cylinder that forms 
the foundation of the bobbin. There is always a real 
difficulty experienced by students in comprehending the 
proper meaning of the terms bobbin leading and flyer 

Fic. 75 

leading : and although a general statement of their 
purpose can easily be made, and, perhaps, as readily under- 
stood, yet the analysis of their operations is not so satis- 
factorily apprehended. An attempt will therefore now be 
made to present the subject in as clear a light as possible, 
and the diagrams accompanying the description will materi- 
ally help towards making it intelligible. 

in FLY-FRAMES 141 

Principle of Winding. — As already explained, the 
roving comes from the rollers at a continuous and regular 
rate, which is dependent upon their surface speed or 
revolution. The problem to be solved, therefore, is how 
to place this roving upon the bobbin at exactly the same 
rate as it is delivered from the rollers. Two methods have 
been adopted : one, in which the flyer "vvraps the roving 
upon the bobbin \ and the other, in which the bobbin 
winds it round itself. These give rise to the two terms 
mentioned above, but before indicating the present practice, 
an examination will be made of the principles upon which 
" winding," as it is called, depends. 

A simple illustration will be given in the first instance 
on the effect of the relative velocities of two points ; — it is 
upon this feature that the winding has its basis. In Fig. 
76 (top left-hand corner), A and B represent two points, 
connected together by some material that can be " paid out " 
if the points separate. If these points are caused to move 
in the direction of the arrows, at equal velocities, they will 
arrive at C and D without altering their relative positions, 
so that the material connecting them has not been influenced 
in any way. If, however, only one point moves, as at E, 
the material must be paid out in order to keep the points 
connected, the amount being denoted by the line F, G. 
This length, of course, Avould be doubled if the point F, 
instead of being fixed, were caused to move in the opposite 
direction. A modification of this case is shown in H and 
I, where H moves to J at the same time as I moves to K, 
only half the distance separating them, which denotes the 
amount paid out ; it would give the same result if the 
relative movements of H and I were reversed. 

Another illustration, approaching more nearly to the 
conditions of the special case of the fly-frame, is represented 



by the first five diagrams, and shows the eiTect when a 
circular movement is made by the points instead of a 
horizontal one, as in the preceding case : A and B are two 
points, connected as before with material capable of being 

Fig. 76. 

paid out by one or the other or both — but for simplicity 
say B. Each point moves round the centre, and we 
require to know the effect of the respective velocities of A 
and B upon the material as they move round the centre. 
In the first case they move at equal velocities in the same 
direction, so that on reaching the position A^ B^ no change 

Ill FL Y- FRAMES 143 

ill the line joining them has taken phice ; it remains exactly 
the same length as before. In the second diagram, A 
remains fixed while B moves, and it is readily seen that B 
no sooner begins to move than the material must be given 
up in order to maintain its tension, so that when B arrives 
at B^ the amount of this material is represented by the 
semicircle C D. By causing A to move in the direction 
opposite to B (as in case 3), half a revolution of each would 
cause a complete turn of the material to be made round E, 
half of which would be wrapped round by A, and the other 
half by B. 

In Diagrams 4 and 5 a medication of the first two 
diagrams is given, both points, A and B, being allowed to 
move in the same direction, but at diflferent velocities. By 
reference to Diagram 4, B is supposed to make a complete 
revolution at the same time as A makes half a revolution 
to A^ ; this has the result of winding on half a reA^olution 
of the material on E. It will be seen that B's movement 
Avould give a complete turn ; but since A moves in the 
same direction, half of it only will be laid on the centre E. 
Diagram 5 represents the reverse of this, though producing 
the same results : A is made to go through one revolution 
during the time that B makes half a revolution. The 
peculiarity to notice here is the direction of motion compared 
with that shown in Diagram 4 : this is a necessity if the 
tension of the material between the two points is to be 
maintained, for if the direction were reversed in No. 5, 
the superior speed of A over B would instantly cause the 
material to go slack, and the first condition of winding 
would be destroyed. There is one way of keeping the 
same direction of rotation in the two cases, and that is by 
reversing the positions of A and B. This would make 
No. 5 case almost similar then to No. 4, with the exception 


that the material is given up from the opposite side of the 
circle that forms the path of the movement of the points. 

From this illustration a step farther may be taken, of a 
more practical character, and directly connected with the 
actual effect of winding. AVe have seen from the examples 
given that a material can be wound round a centre bobbin 
when two points through which it passes are revolving at 
different speeds. By giving a diagrammatic view of the 
conditions existing in the fly -frame this may be fully 
exemplified and made obvious. For this purpose the 6th, 
7th, and Sth diagrams have been prepared, the reference 
letters corresponding in each case : E is the centre bobbin 
upon which the roving is wound ; B is the flyer through 
which the roving passes to the bobbin ; its path is shown 
by the dotted outer circle ; A is the point where the roving 
is laid on the bobbin. 

In Diagram 6 it is assumed that the bobbin is stationary 
and the flyer B revolving. "When B has gone through half 
a revolution it will naturally have wrapped upon the bobbin 
a length of ro\ang equal to half of the circumference of E, 
as shown by the line A C. Now this is a perfectly' natural 
way of placing the roving upon a bobbin as far as a single 
layer is concerned ; but practical considerations in respect 
of placing a number of layers upon it necessitate a modifi- 
cation of this case, in which the bobbin itself is given a 
motion, but in its degree A'arying from that of the flyer. 
Two cases are given : one (Xo. 7) in which the flyer moves 
more quickly than the bobbin, technically called "flyer 
leading " ; and the other case (No. 8) in which the speed of 
the bobbin is much the quicker — from which we get the 
term "bobbin leading." 

On reference to Diagram 7, if the flyer B revolves 
through half a circle to B^ while the point A on the bobbin 

Ill FL y-FRAMES 145 

only goes a quarter of a revolution to A\ it is obvious that 
the two points A and B will separate to the extent of the 
quarter of the circle A^ C ; in other words, since the tension 
remains constant this length of roving has been drawn 
through the flyer leg B and wound upon the bobbin. Now 
take the case of diagram 8 : here A and B occupy the same 
position as in No. 7, but the liobbin is assumed to go half 
a revolution in the same time as the flyer B goes a quarter 
of a turn to B\ The first thing to notice is the direction 
of movement : it is clearly impossible, as the}^ are at present 
arranged, for them to move in any other direction ; other- 
wise, since A moves quicker than B, the roving would go 
slack between the two points. It is therefore necessary in 
bobbin-leading, as compared with flyer-leading, either to 
change the direction of driving or reverse the position of 
the presser of the flyer which corresponds to the line A B. 

In respect of the winding, it will be seen that the point 
A moves half a turn to A^ ; at the same time the flyer B 
has gone a quarter of a turn to B^, which clearly causes a 
separation to take place between the two points A and B 
to the extent of the portion of the circle A^ C. This 
length has been drawn from the flyer owing to the superior 
speed of A, and for the same reason it Avinds it upon itself, 
as shown. At the present time all fly-frames are made 
with bobbin-leading", as this system is found to possess 
superior practical advantages over the flyer-leading'. The 
reason for this adoption will be given subsequently when 
dealing with the problem of building the bobbin. 

Flyer leading. — If the foregoing description has been 
closely followed it will have prepared the reader for the 
next step explanatory of the process of building the bobbin. 
The two diagrams. Figs. 77 and 78, have been prepared 
in order to elucidate it, and in connection therewith a little 




recapitulation Avill be necessary. Fig. 77 represents the 
case of the " fiyer leading " ; in it the flyer B is shown as 
having moved through half a revolution to B^ in the same 
time as the bobbin E has revolved one quarter of a 
revolution to A\ This results, as already shoAvn, in the 
flyer Avinding on the bobbin a length of roving equal to 
a quarter the circumference of the bare bobbin, represented 
by the thick line A^ C. The relative velocities of the 
fl3'er and bobbin will keep the same until the first layer 
is completed, but when we come to wind the next layer 

■■ >f:' 

upon the first one, it must be done on a larger diameter ; 
— and this fact introduces a new order of conditions, Avhich 
will now be dealt with, and which brings us face to face 
with the real problem of Avinding. As a preliminary, two 
conditions must be remembered, viz. the flyer revolves 
at a constant speed, and the roving is delivered regularly 
from the rollers whatever diameter the bobbin may be. 
The conclusion to be drawm from this latter fact is, that 
the bobbin must run at such a speed as to wind on exactly 
the same amount of roving as is delivered from the front 
roller, whether it be full or empty. The effect of Avinding 
on a larirer diameter will now be considered, and in order 

Ill FL Y-FRAMES 147 

to emphasise the matter the difference between the empty 
cand the full bobbin will be taken as an illustration. On 
the empty bobbin, C A\ Fig. 77, is wound on during the 
same time as the flyer B moves through half and the 
bobbin E through a quarter of a revolution. Now, 
on the full bobbin, this length C A^ must be wound on 
in exactly the same time as on E ; the position of the 
presser (along which the roving travels) has moved from 
B C to B C^ (or from B A to B D) ; CMs therefore the 
point of eontact where the roving enters on the full bobbin, 
and from here to A^ (shown by a thick line) is represented 
a length C^ A^, equal to C A\ Whilst the flyer has 
therefore moved half a revolution, the bobbin must have 
gone through a much larger angle than a quarter of a 
revolution, as it did when the bobbin was empty : in other 
words, it has had' to increase its speed from a quarter to 
almost half a revolution, the angle D E A^ representing 
the exact amount. From this it is seen that when the 
flyer leads, the bobbin, starting at a certain speed when 
empty, must gradually increase its rate of revolution as 
it gets larger in diameter. 

Bobbin leading. — The case of the " bobbin leading " 
will now be taken. Fig. 78 being used for reference. As 
before observed, the bobbin E has the quickest speed, and 
while it goes through half a revolution, from A to A^, the 
flyer moves through a quarter of a turn, from B to B^, 
with the result that the empty bobbin winds on itself a 
length of roving equal to C A^. As the bobbin fills, the 
presser will move outwards from A to D, and when the 
flyer makes its quarter of a revolution it will occupy the 
position B^ C\ From this point a length of roving C^ A' 
is shown on the full bobbin equal to the same length C A^ 
on the empty bobbin. On the empty bobbin it required 


half a revolution to wind this length on, but on the full 
bobbin it will be seen that only a little over quarter of a 
revolution is required, as shown by the angle D E A". 
This means that as the bobbin fills it must gradually 
decrease in speed from Avhat it started with as an empty 

To sum up the questions that have just been discussed : 
we may say that with the " flyer leading " the flyer 
revolves quicker than the bobbin, and, as the bobbin 
increases in diameter, its speed must increase in order to 
have wound on it the same length as on the smaller 
diameter. When the "bobbin leads," the bobbin revolves 
at a quicker speed than the flyer, and as it increases in 
diameter it must decrease in speed ; its direction of revolu- 
tion is opposite to that when the flyer leads, or else the 
flyer must be on the opposite hand. 

At the present time the " flyer leading " has fallen into 
disuse. Several reasons are assigned for this ; one 
objection is the increase in speed necessary for the bobbin 
as it enlarges and gets heavier; another is the fact that 
through the indirect driving of the bobbin by means of a 
strap on the cone drums the flyer is caused to start a little 
earlier than the bobbin, which produces a strain on the 
roving, and results in frequent breakages. This evil is 
traceable also to the general gearing, and is said to be the 
result of more backlash existing in the larger number of 
wheels used in the driving of the bobbin than in the 
driving of the flyer. Each may be accredited with its 
share of the condemnation of this principle, and although 
the same conditions exist, yet they do not appear as evils 
when the " bobbin leads," for, instead of the late start of 
the bobbin resulting in a strain and breakage, the roving 
is slackened a little ; this, however, is quickly taken up in 


the course of a revolution or so as the strap and wheels 
drop into their working jmsitions. A simple illustration 
will show the necessity for slowing the bobbin as it fills. 
It is as follows : — Su})pose a bobbin one inch in diameter 
turns once round ; in so doing it will wind on itself 3'1416 
in. of roving. If it be now enlarged to 3 in. diameter 
one revolution will wind on 9-4248 in., so that for the 
larger diameter to wind on the same amount as the smaller 
one it must make one-third of a revolution. The reason 
for not giving this example at an earlier stage is obvious : 
it would not have been consistent with the example of the 
"flyer leading," in which case we saw that the bobbin must 
increase in speed as it enlarges ; it was therefore con- 
sidered preferable to explain the matter in the first instance 
on the general principles applicable to both cases. The 
above illustration, however, is a very valuable one as 
enumerating the principle of the example when the bobbin 
leads, but a warning must be given that such a reduction 
in speed as is there mentioned never actually occui^s in a 
fly-frame, although a greater difference than 3 to 1 exists 
between empty and full bobbins. 

It must be firmly impressed upon the reader's mind 
that reduction in speed as the bobbin fills relates only to 
that portion of its speed which is in " excess " of the speed 
of the flyer. An example of this may be seen in Fig. 78, 
where the empty bobbin turns half a revolution while the 
flyer only turns a quarter. Now the full bobbin is clearl}' 
more than twice the size of the empty one, and yet its 
speed is obviously more than half what it was originally. 
Such a result is very puzzling to those who rely ui)on the 
conclusions drawn from the simple illustration given, 
without considering its real application. In Fig. 78 the 
point to notice carefully is that the excess speed of the 



empty bobbin over tbe flyer is represented by the angle 
C E A\ shown bordered by the black line, and indicating 
the amount of roving wound on, when the bobbin is full, 
and over twice the diameter of the empty bobbin. Its 
excess speed is shown by the angle C^ E A", which is, as 
it ought to be, less than half of C E A\ and is bordered 
by exactly the same length of an arc as on the smaller 
circle. It may be added that, no matter how large the 
bobbin, its speed would never be reduced to one-half, and 
the excess speed, although gradually reduced, would never 
be eliminated. 

Principle of Winding. — We can now make a closer 
inquiry into the reduction of speed necessary for winding 
on the same amount of roving on different diameters of 
bobbin. Suppose that Q (Fig. 79) represents the bobbin, 
and that, after a number of revolutions, its excess speed 
over the flyer has enabled it to wind the roving once 
round itself, as at A B. Further layers are added, as at 
C E Gr, etc., and the question is — What reduction must 
take place in the excess speed at these points in order to 
Avind on the same length of roving % It is quite unnecessary 
to give the actual calculation required to obtain it, it is 
sufficient to point out that since the excess speed is 
represented by the length wound on, its amount can easily 
be found for any given diameter. For instance, the outer 
circle at T was 7 in. diameter in the original drawing, and 
the smaller circle 1 in. diameter ; so that one-seventh of 
the large circle equals the full circumference of the inner 
one. This is what is shown in the diagram at T U ; at 4 
and 5 the circle is 4 in. in diameter, consequently one- 
fourth of its circumference equals A B, and for the same 
reason the circle 2, 3, which is 2 in. in diameter, has half 
its circumference marked off as equal to the circumference 



of Q. An}' of the otlier points can be found in the same 
manner, and this becomes a very simple matter indeed 
when we recognise the principle underlying the diagram, 
wliich may be expressed as follows : — As the diameter of 
the bobbin increases, its rate of revolution must be reduced 

\ '■> "' 

'•■/// / / / / 

--••' ..-•■ /-■' / / 

•••• y* /■' 


Fig, 79. 

in inverse ratio; for, as just shown, twice the diameter 
requires half the speed, four times the diameter, one- 
quarter the speed, etc. The diagram (Fig. 79) is con- 
structed in this manner, and the amount of each circle or 
diameter occupied by a similar length of roving is shown 
by the thickened line ; the ends are joined by a curve, 



which accurately defines the limits of the roving, equal to 
the circumference A B. This curve is a most important 
factor in designating the values of the varying speeds for 
the different diameters, and it is evident that its founda- 
tion depends on the intersection of the radial lines U Q, 
S Q, etc., with their respective circles U T, S E, etc. 
These radial lines form angles with the foundation line 
T Q, while the length of the arc is the same in all the 
circles ; the angles enclosed by it vary considerably, and it 
is their variation that produces the curve. By plotting out 
this curve, either by taking the angles for our values, or by 
taking the proportion of tlie circumference occupied by 
each arc as our basis, and calling the smaller circle one, Ave 
obtain the curve represented in Fig. 80. In this form the 
curve presents its true characteristics, and it is at once 
seen to be a hyperbola. In the first place, it was made 
from the angular values in the following manner : a line, 
T A, was taken and divided into parts equal to the layers 
on the bobbin in Fig. 79, perpendicular lines were erected 
at the points of division, and on them were marked off 
lengths equal in value to the various angles; A B, for 
instance, represents 3G0°, because a full circle at Q contains 
360° ; again, at T U, this line represents 51° 25' 43 " as 
equal to the angle T Q U, and so on with the other lines. 
In order, however, to adapt it to the other method, the 
length T A has been made equal to the full diameter, 7 in. 
At each inch, then, erect a perpendicular, and make the 
one at A B any convenient length. The length of the 
others can readily be obtained as follows : — Call A B one, 
as representing an inch in diameter ; each division will 
then be numbered as two, three, four, etc., up to seven, 
and will be the corresponding inches from the end of the 
line A. The increase in diameter will therefore be repre- 



sented on the liorizontul line, and the decrease in speed on 
the vertical lines, and these latter will Ije reciprocal to the 
former. For instance : — 

Diameters are 1 in. 2 in. 3 in. 4 in. r> in. 6 in. 7 in. 
Sjifeds will l)e 1 i ^ i i ^ f 

Represented as AB AAB ^AB ^AB lAB JAB iAB 

It will be noticed from the above that this relation mnst 
always hold good, A'iz. the diameter mnltiplied by its 
speed must always give the same result throughout the 
building of the bobbin. By carefully measuring the 

diagram (Fig. 80), or making a new one, the following 
relationship between the speed and diameter will be found 
to hold good : — 

At A = l in. dia., lias its speed A B represented by AB. 

At 2 = 2 in. dia. (twice the dia. of A), has its speed 2 -3 represented 
by * A B. 

At G = 3 in. dia. (three times the dia. of A), has its speed G 11 
represented by ^ A B. 

At 4 = 4 in. dia. (four times the dia. of A), has its sjieed 4 5 re- 
jn-esented by ^ A B. 

At i[ = 5 in. dia. (five times the dia. of A), has its speed M X 
represented by ^ A B. 

At 6= 6 in. dia. (six times the dia. of A), lias its speed 6 7 re- 
presented by I A B. 

At T = 7 in. dia. (seven times the dia. of A), has its speed T U 
represented by 3 A B. 


We may now summarise this explanation as follows : — - 

In order to wind on the roving, the bobbin must always 
have a greater speed than the flyer (bobbin leading). 

As the bobbin increases in diameter this excess speed 
must be decreased. 

The reduction in the excess speed must be the reciprocal 
of the increase in diameter : for instance, if the bobbin be 
made twice the diameter, its excess speed must be reduced 
to one-half ; if the increase in diameter be three times, the 
excess speed is reduced to one-third, etc. 

The curve representing the combination of an increase 
in diameter with a reciprocal decrease in speed is known 
as the " hyperbola." 

The variation in the speed of the bobbin as it increases 
in diameter must be consistent with the principles of the 
above curve. 

Driving the Bobbins. — In applying the arguments 
just concluded to the actual operation of winding, it will 
be unnecessary to refer to previous methods. At the 
present time one system is practically universally adopted 
of effecting the required change in the speed of the bobbin, 
viz. by the use of cone drums, on the same principle as in 
the scutcher and opener. On reference to Fig 81 (which 
exhibits the full gearing of the fly-frame) it will be noticed 
that the spindles are driven direct from the driving shaft 
through the wheels H, K, L, and M. It will also be seen 
that the bobbins are driven direct from the same shaft 
through the bevel 0, which is fixed to the shaft and forms 
one of an epicyclic train of wheels, called a " diff"erential 
motion." The wheel G- is connected with this, and from 
here the bobbins are driven through the wheels N, P, and 
Q. As near an approach as possible is given to the speed 
of the spindle and bobbin consistent with the function of 

'" FLY-FRAMES 155 

the differential motion, and it must be carefully noted 

that both, to this extent, are driven direct by positive 



Cone Drums. — Tlie excess speed of the bobbin must 
now be considered. Cone drums arc introduced, one of 
which is driven from the driving shaft through B and V, 
the bottom one being connected to the differential motion 
by the wheels F, R, and E. This has the effect of giving 
the necessary " additional " movement to this motion (the 
action of which will be subsec[uently made the subject for 
examination), and this excess speed is transferred to the 
bobbin through the wheels previously mentioned. 

The statement was made in some previous remarks that 
the amount of roving wound on the bobbin represented 
the excess speed of the bobbin over the flyer, so that if a 
bobbin starts with a certain excess speed, its reduction 
must take place inversely to its increase in diameter. If, 
therefore, the full bobbin be four times the diameter of the 
empty one (which is about the limit in fly-frames), the 
excess speed must be reduced to one-quarter. This gives 
a basis to work upon, and there is no occasion to know 
the actual nixmber of revolutions of this speed, as it is 
merely a question of gearing, and need not be taken into 
account in constructing the cone drums for their special 
purpose. Let us take for illustration the empty bobbin as 
being 1 in. in diameter and the full bobbin as 4 in. in 
diameter. This will mean that the diameter of the cone 
drums must be arranged to give a reduction of four from 
one extreme to the other. The diameters suitable for 
this, which several of our principal machine makers adopt, 
are : 3iin. for the small end, and 7 in. for the large end of 
each cone drum. When the strap is on the large end of 
the" top cone and driving the small end of the bottom one, 
the bare bobbin is winding. As layers are added the strap 
is moved gradually by a special appliance to the other end, 
where a small diameter of the top cone drives the large end 



of the bottom cone, in -wliich case the largest diameter is 
■winding on the roving. 

Formation of Cone Drums. — Two diagrams arc 
given in Figs. 82 and 83 to ilhistrate the further reasoning 
and to show its result graphically. In order to reduce 
the question, at first, to its simplest form, the speed of the 
bare, middle, and full bobbin vill be found, each of which 
is represented by A, B, and C respectively in Fig. 82 ; 
these diameters may be taken as 1 in., 1% in., and 4 in. 
The top drum is the driver, and, for simplicity, it is 


Fio. 82. 

Fig. 83. 

assumed to run at a constant speed of 100 revolutions per 
minute. The extreme diameters E and D Avill therefore 
drive the bottom drum diameters G and F at 200 and 50 
revolutions per minute respectively. The middle s[)eed is 
readily found, for, since the diameter is 1\ in., which is 2^ 
times larger than the empty boljbin, the speed must there- 
fore be the reciprocal of this, Avhich is 1-^2^. Or perhaps 
a more simple way of putting it would be to say that the 
speed must be the inverse of this increase, 2i = Y; so its 
inverse order would make it ?. Now, ? of 200 (the speed 
which drives the empty bobbin) = 80 revolutions; conse- 
quently the top cone must drive the 1)ottom cone at this 


speed when the strap occupies the central position. An 
important point to notice here is the great reduction in 
speed that has taken place from the speed of the bare 
bobbin, and how small this reduction becomes as the 
bobbin fills. It will serve to impress the reader with the 
obvious lesson to be learnt from the hyperbolic curve 
given in Fig. 80, where the reduction in speed is shown, 
by the quick descent of the curve, to be very rapid for the 
first layers, and then by the more gradual curvature to 
diminish at a slower rate as the bobbin gets fuller. It is 
now merely a question of proportion to get the diameters 
to suit the speeds, remembering, of course, that the sum of 
the opposite diameters must be the same, viz. 10| in. ; 
therefore — 



' = 5'833 in. dia. of bottom drum, 

and the corresponding diameter for the top drum Avill be 
10'5 - 5*833 = 4"66 in. If diameters be measured on the 
diagram equal to these dimensions, and a curve be drawn 
through their extremities, Ave obtain the hyperbolic curve 
that is so characteristic of the cone drum : the top cone 
becomes depressed or concave in its outline, and the 
bottom one correspondingly becomes convex. This simple 
case Avill enable a more complete example to be the more 
easily understood, and this we now proceed to give. 

The bobbin, as before, is 1 in. in diameter when empty, 
and 4 in. when full. In order to obtain a sufficient 
number of points through which to draw the correct form 
of the curves, the necessary diameters of the drums Avill 
be found for every quarter of an inch additional diameter 
of the bobbin. The lengths of the cone drums are at 
least thirty inches, and the strap is moved from one end to 
the other by the transverse motion, step by step, as each 




laj'cr is adilcd. Then in tlie present example the cone 
dnims are divided so as to show the position of the centre 
of the strap for each diameter of the bobbin. This gives 
us thirteen position lines for ■which diameters are required, 
but, as the two end ones are known, viz. ?>\ in. and 7 in., 
only eleven require calculating. As in the last illustration 
the top driving cone will be assumed to run at 100 revolu- 
tions, in which case the extreme speeds of the bottom cone 
will be 200 and 50 revolutions, corresponding to the empty 
and full bobbins respectively. 

The following table presents in a very concise form the 
elements and method of calculating the speeds and 
diameters : — 


















1 X 200 = 200 








% X 200 = 160 








t X 200 = 133 -33 







f X 200= 114-28 








\ X 200 = 100 








% X 200 = 88-88 








I X 200 = 80 








T^rx 200 = 72-72 








i X 200 = 66 -66 








^*jx 200 = 61-5 








f X 200 = 57-14 








T*jx200 = 53-33 








\ X 200 = 50 





A are the actual diameters of the bobbin in inches. 

B are the diameters of the bobbin expressed fractionally. 

C represent the reciprocals of the figures in column B, 
and are expressed by reversing the order of the fractions 
in that column. These fractions represent the speeds in 
the same Avay as column B represents the diameters. 

D column gives the speeds of the bottom drum as the 


various diameters of the bobbin are built up. They are 
found, as shown, by taking the fractional proportion of 200 
revolutions for each diameter, as represented in column C. 

E gives the corresponding speed of the top cone drum, 
which is constant. 

F gives the calculated diameters of the bottom cone 
drum. These are found by a simple proportion, as 
follows : — If the sum of two opposite speeds equal the 
sum of two opposite diameters corresponding to them, 
what diameter will equal either of the two speeds which 
make up the sum 1 Example (for 1 in. diameter) : — 

Top speed X sum of the dia.'s ,. r, ^^ j 

— =i — i = — ^— =— =dia. of bottom cone drum. 

lop speed -f bottom speed 

(1 in. dia. of 100 x 10i_ 1050 _ 

bobbin) 100 + 200- 300 -'^■'^ '"■ '^'''• 

(2 in. dia. of 100xlOi_1050_ . ,. 

bobbin) 100 + 100 - ^00 - "^ "^"^ '''• '^'^• 

(3 in. dia. of 100 x 10| _ 1050 _ ., . ,. 

bobbin) 100 + 66 "66 "166 -66 "^ ^"^^ ^^' 
(4 in. dia. of 100xl0i_1050_^ . ,. 

bobbin) 100 + 50 - 150" - "''• '^'^• 

G represents the diameters of the top cone drum, which 
are found by subtracting the diameters of the bottom cone 
drum from 10* in. 

H is the sum of the opposite diameters. It is scarcely 
necessary to point out that this is requisite in order to 
allow the strap to fit regularly throughout the length of 
the cone drums. 

Figs. 8i and 85 embody the above tabulated results. 
When the strap is at A the empty bobbin is being driven. 
By the time a quarter of an inch increase in diameter has 
been made the strap has been moved to B, and so on the 
full length of the drums, equal divisions representing equal 
increases in the diameter of the bobbin. 




The above method of constructing a pair of cone drums 
for Ijuilding the l:)obl)in of the fly-frame is obviously so 
simple that it is surprising so much difficulty is experienced 
in explaining it. The whole process may be summarised 
in a few words : the various diameters of the bobbin are 
expressed fractionally ; these fractions are reversed or 
inversed, and multiplied into any speed we care to take 
as representing the empty bobbin ; the results give all the 
speeds from empty to full bobbin. Knowing the speeds, 
it is a question of simple proportion to calculate the 
diameters required to produce these speeds ; not only 

J .! 


f -^ f * '7 ^ V, — o — 1-:5 — 

"■■ i^. i-i. M. 3-. 2x. d. d, t. li ,i. ,i ,- 

—^lft»ot BOBBIN ■ '• 

Fig. 84. Fio. 85. 

is the correct form of the drums given, but the actual 
hyperbolic character of the curve is denoted, and it is 
strange that so many mistakes have arisen as to its correct 

Whilst appreciating the above method, it is well for 
the reader to be acquainted with other methods of arriving 
at the same result, so one or two of them will now be 
given. In this case, as in the previous one, the extreme 
diameters of the two cone drums must be decided upon. 
We will assume the same diameters as before, viz. 3i in. 
and 7 in. ; and for purposes of reference Fig. 85 may be 
used. Now divide the length of the cone drums into a 


number of divisions in such a way that when they are 
numbered, the last number is four times the first. (The 
reason for this is, because four is the proportion between 
the extremes of the speeds.) In the diagram each division 
is numbered, commencing at 4, and on up to 16; we thus 
obtain 13 divisional lines upon which the diameters must 
be measured. Of two of these we know the dimensions, 
so 11 remain to be found. Assuming the top cone to be 
running at 100 revolutions, the following table will show 
in a convenient form the elements to be used in calculating 
the diameters and speeds : — ■ 

V Speed of driving cone =100 revs. 

D Required dia. of driven cone . . Tliis must be found 

I\[ Greatest speed of driven cone . . . . = 200 revs. 

P Required speed of driven cone . . This must be found 

R Ratio of sjseeds at the extreme of driven cone . =4 

E Smallest dia. of bobbin (bare) represented by . =4 

G Any dia. of bobbin represented by . . . =4 to 16 

S The sum of the dias. of the cone drum . . =10^ in. 

The speeds of the bottom drum are obtained by inverse 
simple proportion : for instance, if when the strap is at 
4, the driven drum has 200 revolutions, what number of 
revolutions will it have when the strap is on 5 ? Inverse 
proportion gives 160 revolutions. Again, when the strap 
is on 6, what speed Anil the bottom drum have % The 
result is 133 '33 revolutions — and so on through the Avhole 
numbers from 4 to 16. It will be observed that the 
inverse proportion is the chief point to be taken into 
account, and the reason for it is too obvious, after the 
explanation just given, to require further explanation. 





P speeds. 

D (lia. 


P speeds. 

D dia. 








































The above gives a tabulated result of the speeds and the 
diameters deducible from them. Convenient formulae for 
both speeds and diameters are : — 


-7^- = r (speeds). 

r + 100 

= D (diameters). 

Another Method : — The same diagram (Fig. 85) will 
also serve for illustrating this method. As in the last 
example, the cones are divided into 1 3 diameters or division 
lines, commencing at 4 and finishing at 16. Instead of 
finding the various speeds of the driven cone as the strap 
moves forwards, we obtain the "ratio" that must exist 
between any two opposite diameters. With the end 
diameters this is a very easy and direct calculation, for 
at the beginning end the ratio is 



and at the finishino; end it i.s- 



The intermediate ratios, however, cannot be obtained in 
this way because the diameters are not known ; therefore 
the numbers representing the division lines 4 to 16 must 
be made use of. It will be noticed that the ratio is a 
fixed proportion of the numbers which represent the two 
ends: for instance, \ of 4 = 0-5, and \ of 16 = 2; and in 
the same way \ of any of the numbers from 4 to 16 will 
give the ratio at these points. The formula is as 
follows ; — 

Ratio of diameters at any point =^-p,. 

The letters referred to are in the following table : — 

A Largest dia. of cone . . . . . 7 in. 

B Smallest dia. of cone . . . . . Z\ in. 

C Required dia, of driving cone (to be found) . S - D 

D Required dia. of driven cone (to be found) . S - C 

E Smallest dia. of bobbin, represented by . .4 

F Largest dia. of bobbin, represented by . .16 

G Any dia. of bobbin, represented by the numbers 4 to 16 


R Ratio of dia. of cones of any given point . = -r-=, 

S Sum of any two opposite dias. . . . =10i in. 

The formula for the ratio works out as follows : — 

R = |5 = 'l:^ = i|i^ = 0-125G. 
AE 7x4 28 

If this be worked out for each number, 4 to 16, the 
column R in the accompanying table will be obtained. 
It is now an easy matter to get the speeds from the ratio, 
and it is needless to do more than present the method by 
a formula : — 

The diameters of bottom drum = :p; — -. 


From this we obtain the column D in the table, and 



on compai'ison with the previous table it Avill he found 
that the same diameters are detained in each case. 



R = 4G. 

D = 


4 . 



5 . 



6 . 



7 . 



8 . 



9 . 



10 . 



11 . 



12 . 



13 . 



14 . 



15 . 



16 . 



It will be seen that the principle in the above method 
depends upon the important fact that the ratio of the 
corresponding diameters on each cone drum increases in 
exactly the same proportion as the diameter. To those 
readers who have carefully followed out the reasoning 
employed in the foregoing examples, there will be nothing 
difficult in understanding the statement just made ; indeed 
it is a natural deduction from the statement that the 
speed of the bobbins is inversely proportionate to their 
diameters. The following will make this clear ; the speed 
of the bobbin depends directly upon the speed of the cone 
drums, so that the speed of the cone drums must vary 
inversely as the diameter of the bobbin ; but the ratio of 
the diameters of the cone drums are inversely 2_)roportionate 
to the speeds which they give — for instance, a 3-in. pulley 
driven by a 6-in. pulley is one-half the size, but its speed 
is doubled — therefore the ratios of diameters of the two 
cone drums must increase in the same proportion as the 
diameters of the bobbin increase. 


Another Method. — A modification of the above method 
can now be given, which depends on finding the ratios of 
the cone drums from the diameters of the bobbin ; but by 
incorporating the sums of the diametei's in the calculation 
the obtaining of the ratio becomes unnecessary. Still it 
must not be overlooked that this ratio is the very principle 
upon which the method is founded. As before — 

A represents the largest dia. of driving cone drum . Say 7 in. 

B ,, the smallest dia. of driven cone drum . ,, 3^ ,, 

X ,,. any other dia. of driving cone. 

X ,, ,, ,, driven ,, 

D ,, dia. of empty bobbin . . . .,,!,, 

d , , any other dia. of bobbin . . . . 1 to 4 , , 

S ,, sums of opposite dias. of cone drums . 10^ ,, 

We can now reason as follows : — • 

As the initial ratio of the driving cone is to the empty 
bobbin, so will the ratio of any other opposite diameters 
be to the corresponding diameter of bobbin. Or, to express 
it in letters — 

(1) As|:D::^:rf. 

To those unacquainted with algebra, this could easily be 
worked out by proportion to get the ratio of the opposite 
diameters of cone drums, with the same results that appear 
when the above is formed into an equation : — 

(2) a"^=X^°^' A=X- 

... , BdK 

(4) and x= -r-^ . 


But X + x = S. 

.-. X =H-x. 

And.x- =S-X. 


By substituting in (3) the value of X Ave obtain — ■ 
„ AxD 

V>d (S-a;) = AxD. 


Bt;S = Aa;D + Bcte. 

„..,,,, ., , Bf^S AccD + Bc^a; 
Divide both sides by x : = -. 

- — =AD + B<^. 


Divide both sides Bc?S... - = — ^^-^jq — . 
X BctS 


•"•^~AD + B(«" 


In a similar manner, by substituting the value of x in 
(4), it can be shown that — 

(^^ ^-b^Tad- 

By substituting figures for the above formulae Ave can 
readily obtain diameters of the cone drums corresjjonding 
to any required diameter of bobbin. For example, Avhen 
the bobbin is 2 in. in diameter — 

,-. x=-^ — 4i = 5'25 in. 

7 X l+3ix 2 

and X=: 10^-5 -25 = 5 -25 ,, 

Again, Avhen the bobbin is 3 in. in diameter — 

3i X 3 X lOi ^ .^ . 
.' . x = — ^^- — —. — =- = 6 "3 in. 
7 X 1+34 x3 

and X = 10|-6-3 = 4-2 ,, 

These results Avill be foiuid to correspond Avith those 
given in Fig. 85. 

Many efforts have been made to improve on the present 
method of driving the bottom cone drum, but hitherto 
Avithout sufficient success to Avarrant their extensiAe use. 
The attempts have chiefly been directed toAvards dispens- 


ing with the strap and connecting the drum by some 
arrangement of friction driving, either directly or by 
friction bowls. Innumerable mechanical contrivances have 
been devised for the purpose, but they have proved of 
little value except from an experimental point, ^yhen the 
subject of the differential motion is being treated, it will 
be shown that there was a reasonable necessity for trying 
to effect some improvement in the driving of the cone 
drums, but it will also be seen that the real difficulty, 
caused through constant slippage and breakage of belts 
existed in the motion itself, and, consequently, whilst some 
people sought for better results in a new form of driving 
the cones, others looked for the remedy in a new differ- 
ential arrangement. In the latter case a success has been 
gained, and the faults previously complained of are now 
practically eliminated. 

The Differential Motion. — In Fig. 81 a gearing 
vieM' of the fly-frame was given, from which could be traced 
the whole of the driving. It was especially given to show 
the driving of the bobbin and the spindles, and in the 
accompanying remarks it w^as pointed out that the bobbins 
as well as the spindles are both driven positivel}^ from the 
driving shaft. This positive system of driving the bobbins 
must be connected in some way with the varying excess 
speed given to them through the cone drums, and the 
problem of how to obtain a combination of two speeds has 
been solved by introducing into an ordinary train of wheels 
other wheels on movable centres. An arrangement of this 
kind is called in mechanics an epicyclic train of wheels; 
but in the phraseology of cotton S2:)inning it is known as a 
differential motion. The principle underlying an epicyclic 
train of wheels is so important that it will ])e useful to 
make a few remarks on the elementary part of the subject 




before describing the differential motion itself. With this 
object in view a series of sketches have been prepared that 
Avill now be described and the principle of the actions 
which they illustrate will be explained. 

In Fig. 86 a diagram is given showing the simplest 
form of an epicyclic train of wheels. It consists of two 

D ; i 

Fig. 86. 

wheels, B and C, in gear, and, for convenience, they are 
made with an equal number of teeth in each. The wheel 
B is carried on a fixed centre A, Avhile C is carried by a 
link or arm J, which is capable of turning round the centre 
A, Under ordinary conditions the two wheels would 
revolve at equal speeds if they had fixed centres, but by 
carrying one of them on a centre that can be moved, a new 


set of conditions is introduced, and these we will proceed 
to analyse. 

In the first place, suppose the wheel B is fixed so that 
it cannot revolve. If the arm J is now moved through a 
complete circle round the centre A, it will carry the wheel 
C with it. The teeth of C will therefore be in gear with 
the teeth B during the whole revolution, and, as a con- 
sequence, C will be compelled to revolve on its own centre. 
On examination it will be found to have made two 
revolutions during the time the arm J has made one ; at 
first sight this result is puzzling and it is not very easily 
understood. The reader naturally reasons that since C 
cannot possibly gear with more teeth than are contained by 
B, and as C has exactly the same number of teeth, there- 
fore C ought only to turn once whilst travelling round B. 
This reasoning, however, leaves out the important consider- 
ation that the mere fact of carrying C round the centre A 
gives it a revolution of its own, quite independent of its 
connection with the wheel B ; and this revolution must be 
added to that obtained through the wheels B and C being 
in gear. This is the reason why we obtain two revolutions 
of C for one of the arm. 

The independent revolution of C is one that students 
find very difficult to understand. They readily follow the 
reasoning which shows that C must revolve once through 
its being in gear with B, but the additional revolution is 
not so easily understood, and therefore an attempt will be 
made to explain it as clearly as possible. 

In the diagram (Fig. 87) the wheel C is shown, carried 
by the link J, which is capable of turning round the centre 
A. The Avheel is marked with an arrow, so that any 
change in its direction can be noted. As the arm carries 
it bodily in a circle round A, its position in each quarter of 




the circle is shown ; and it is seen that, when at D, the 
arrow points in a direction at right angles to its original 
position, whilst at E it points in an opposite direction to 
the position at C. The wheel must therefore have made 
half a revolution in going from C to D, for it is impossible 
to conceive of its turning upside down without revolving 
round its axis to do so. In the same way, when passing 

from E to C, the arrow resumes its first jiosition, and has 
thus completed a full revolution. It is this revolution, 
which has been obtained independent of any gearing, that 
must be added to whatever revolutions are given to it 
through its connections with the centre or other wheels. 
To show by an example that this explanation is correct and 
can be confirmed, a diagram is given in Fig. 88. The 
centre wheel B is loose on the centre A, and as the arm J 
is revolved, the wheel C will turn on its axis and assume 



tlie positions at D, E, and F ; and it will be noticed that in 
doing this it will also cause the wheel B to revolve to the 
extent that one revolution of tlie arm will give one revolu- 
tion to B. A further confirmation that C in Fig. 86 turns 
twice in being carried round the centre A can be obtained 
by a consideration of Figs. 89 and 90. In Fig. 89 a side 

view of the wheels is shown : is fixed on A, so that it 
cannot revolve ; the wheels B and D are carried by the 
link J, centred at A ; B gears into the fixed wheel 0, 
whilst D gears with a wheel C working loosely on the 
shaft A. If the link J is now turned through a complete 
circle, it will carry the wheels B and D along with it. B, 
as we have said, will turn twice, and as it is on the same 




centre as D, this Avlieel will naturally revolve twice. Since 
D is in gear wath C, it is quite clear that C will also obtain 
two revolutions, so that for one revolution of the wheel B 
round the centre A, the w^heel C can be made to revolve 
at double the speed. This fact is taken advantage of in 
many cases, one of Avhich occurs in the well-known wrap 
reel of our cotton mills ; but instead of pinions being used 
an adaptation is made with bevel wheels, as shown in Fig. 
90. As in the previous case, is fixed on the shaft A, 
B is geared into it, and also into another bevel C, working 

Fig. 90. 

loosely on the fixed shaft A. If B, which is capable of 
being moved bodily round the centre A by means of J, is 
now revolved round A, the two revolutions it will receive 
(one because of its being in gear with the dead wheel 0, 
and the other because of its turning round A) are trans- 
ferred to C, and so w-e find that one revolution of the 
handle J produces two revolutions of the wheel C. 

So far we have confined ourselves to the consideration 
of all the wheels being equal ; but it is not difficult to see 
that they may be unequal to each other in any degree that 
may be found convenient, without altering anything that 


has just been explained. For instance, in Fig. 86 B may 
contain 65 teeth and C 24 teeth ; then, if B is fixed and 
the arm J revolved, C will revolve |-^ times by reason of 
its gearing with B, and once through its revolution round 
the centre A, so that its total turns will be 

If 4- 1 = Iff revolutions 

and so on with whatever wheels are used. It ought also 
to be pointed out that any number of wheels can be geared 
together and carried by the arm J, and these can be either 
mere carriers or compounded. If carriers they do not 
affect any part of the calculations any more than they 
would in an ordinary train of wheels, but they do affect 
the direction of revolution, and if the train consists of an 
odd number of Avheels, such as 3, 5, 7, etc., the direction of 
revolution will be opposite to that obtained when an even 
number of wheels is used, as 2, 4, 6, .etc. Another feature 
in connection with this motion that must be clearly under- 
stood is what is called " relative " motion ; that is, the 
speed of one part of the epicyclic train compared with any 
other portion. For instance, although the wheel B in Fig. 
86 may be fixed, and, therefore, can have no "absolute" 
motion, it may still have a "relative" motion compared 
with the arm J, and this relative motion will necessarily 
be opposite to that of the arm, just in the same way as the 
impression is given to a person in a moving train that 
stationary objects passed on the way are apparently moving 
in the opposite direction ; in reality they are fixed, but 
"relatively " to the train they are moving. The amount of 
this relative motion can be readily obtained by subtracting 
the speed of the train from the speed of the object. In 
the case of a fixed object the result is that it appears to 
move at the same speed in the opposite direction ; if the 


object itself moves, the result Avill vary according to its 
speed, but the means of obtaining it is always the same, 
viz. by subtracting the speed of the train from that of the 
object. This illustration is given because it is familiar to 
every one, and it applies equally well to the epicyclic train 
of wheels. This reasoning therefore enables it to be said 
that the " relative " speed of the centre wheel equals the 
speed of itself minus the speed of the arm ; and the 
" relative " speed of C (or the last wheel of the train, Avhat- 
ever it may be) equals the speed of itself minus the speed 
of the arm. This can be expressed much better if letters 
are used instead of Avords, and for the convenience of 
students who have Goodeve's Elements of Mechanism, the 
same reference letters will be used as are there employed 
for explaining the same subject. 

"Whilst the arm J makes a revolutions, 
let the first wheel B make m revolutions, 
and the last wheel C make 11 revolutions ; 
also let e be the value of the train wheels. 

The value of a train of wheels is found by dividing the 
speed of the last wheel by the speed of the first Avheel. 
The above reasoning gives the " relative " speed of — 

B = m- 


and the relative speed of- 



a ; 

therefore the value of the train 



- a 

This gives an equation from which can be obtained any 
of the elements that may be unknown, provided the others 
are given ; and it will suggest itself to the reader that 


three principal methods of arranging the system can be 
made. For instance : 

(1) The wheel B can be fixed, in which case it has no 
absolute motion ; and therefore m = o. 


e = 

11 — u, 

■111 - «' 

e = 

n -a 

- a ' 

le — 

()i-a)~ a 

— (\ 

-a ' 

- ac = 

■ n - a. 


e = 


and e- 

1 = 



■ l-e. 

Th ere fore 




a = 


Or (2) the wheel C can be fixed, in which case n = o. 

Therefore e = — , 

m - a 

— a 

e{m -a)= - a, 


m-a= — , 

Therefore m = al 1 • 

, m 
and a= -, 



Therefore a = . 



Or, again (3), both the arm and the wheels are free to 
move — 


in - a 
ei^in- a) — n- a, 
e m- e a = n - a, 
em- ea + a = n, 
e 1)1 -a{e- l) = ?i. 
Therefore n — e7)i + a{l~ e). 

This latter case is the one to which most interest will 
be attracted, as it bears directly upon the actual conditions 
of the differential motion, and the reader will see that it 
allows for the driving of the wheels or arm independently 
of each other. In working out numerical examples of the 
cases just given it is as well to bear in mind that the value 
of e will be positive when the train is composed of an odd 
number of wheels, and negative when an even number of 
wheels are used. To take an example from Fig. 86. 
Suppose B = 65 teeth and C = 24 teeth, and the arm 
revolves 12 times a minute, how many revolutions will C 
make 1 From our first case 

n = a{l -c) 

65\ The minus sign because there are 
~ 24 j two wheels in the train. 


?i = 44i revohitions. 

An example can also be taken from Fig. 90. Here 
and C are equal, and it is immaterial what the size of B is. 
The value of the train e is therefore 1, and as C revolves 
in the opposite direction to (which corresponds to the 
effect produced by an even number of spur wheels), its 
value becomes -1. 

n = a{\ -e), 
and c= - 1. 
Therefore ?(, = «(! - ( - 1), 
n = a{\ + \), 
n — 2a. 
VOL. n N 


C is thus shown to revolve twice as quickly as the arm 
J. This result can also he obtained in another way, which 
is perhaps more suitable for our present purpose. It has 
already been shown that 

Also in Fig. 162 f= - 1. 

Therefore n-a — a- m, 
and ?i+ m = 2a, 

which shows that the wheel C revolves twice whilst the 
arm turns once. 

We will now proceed to describe how advantage has 
been taken of the foregoing principles in producing the 
diflferential motion of fly-frames. It is part of the subject 
that must be thoroughly studied if the students desire to 
obtain a proper knowledge of it, and for this reason it will 
be presented in such a way that the reasoning can be 
readily followed. 

In Fig. 91 a drawing is given showing an enlarged 
view of the differential motion seen in Fig. 81. Only 
sufficient of the gearing is now represented to serve our 
present purpose. The first thing to notice on comparing 
it with Fig. 90 is the method of carrying the wheel B. 
Here it Avill be observed the arm J is dispensed with, in its 
place a spur Avheel J being used, and upon its face is fixed 
an arrangement for supporting the bevel wheel B, at the 
same time allowing of its free revolution. A is the driving 
shaft, upon which is fixed the bevel 0, and on the opposite 
end is fixed a wheel H, from which the spindles are driven. 
The spindles are thus driven direct from the driving shaft 
A. Referring to the bevel 0, we notice that it drives the 
bevel C through B, and on the elongated boss of C (which 



is loose upon the shaft A) is fixed a wheel CI, from Avhich 
the bobbins are driven. The bobbins are therefore driven 
direct from the driving shaft, and during the whole build- 
ing of the bobbins the motion received from the cone 
drums is only a slight proportion of tlic motion they 
receive direct. 

By mounting the bevel B upon J we are enabled to give 
an independent motion to what corresponds to the arm in 
Fig. 65, and it is through this wheel J that the cone 

Fig. 01, 

drums transmit to the bobbins the excess speed necessary 
for winding. The connection of gearing for this purpose 
is clearly shown in Fig. 81, only the shaft F and the 
wheel E being reproduced in Fig. 91. We are now in a 
position to analyse the motion. In the first place, the 
bevel C will revolve at exactly the same speed as 0, on 
condition that the wheel J is held stationary ; but its 
direction will be opposite to that of the shaft A. Now 
the only way to vary the speed of C (through which 
the bobbins are driven) is to cause B to be moved round 


the shaft A, and this is done by causing J to revolve, 
which carries B round with it. According to the direction 
and speed of J it is possible to increase or decrease the 
speed of C, and as this is the point round which the whole 
interest centres, we propose to examine the subject care- 
f\illy, and will commence by assuming that J is revolved in 
the same direction as 0. 

(1) Suppose J has the same number of revolutions as 0, 
and in the same direction, it is easy to see that B will be 
carried round with its teeth locked in 0, and therefore it 
cannot cause C to revolve by its own axial rotation, but 
simply carries it round at the same speed as J and ; 
consequently O, J, and C have the same rate of revolution, 
and all in the same direction. A reference to the formulae 
n-\-m=1a would show this to be the result; for, since 
a = m, we have 

7i.-t-m = 2»i. 
Therefore n=77i, 

which proves C to have the same speed as 0. 

(2) Suppose J to be fixed, will then drive C at a 
speed equal to itself, but in the opposite direction : or by 
using the formulae, a = (m = 1 in all these formulae) — 

n + m = 2a, 
n = 2a - m, 

71 = 0-1, 


C is thus shown to revolve at an " equal " speed to 0, but, 
as the minus sign denotes, it is in the " opposite " direction. 
This is therefore an extreme case when J is stationary. 
We shall now notice the change that takes place as J is 
revolved in the same direction as 0. 

(3) Suppose J revolves i the speed of 0, then — 


n = 1a- m, 
TC = 2xi-], 

Therefore C revolves at " half " the speed of and in 
the "opposite" direction. 

(4) Suppose J revolves h the speed of 0, then — 

n = 2a-m, 
?i = 2xi-l, 
?i = l-l, 
n = 0. 

The speed of C is therefore reduced to nothing when the 
speed of J equals half the speed of 0. 

This result represents another extreme, or rather, it 
might be termed a zero point, for, as we shall see, the 
speed of J when still further increased causes an increased 
velocity of C. For instance — 

(5) Suppose J revolves at | the speed of 0, then — 

n = 2a- 7)1, 
n = 2x|-l, 

« = i 

So that C now revolves at " half " the speed of 0, but in 
the " same " direction. 

(G) Suppose J revolves at the same speed as 0, then — 
n = 2a- m, 

71 = 2-1, 


This is simply the first case repeated, and shows C to 
revolve at the same speed as 0, and in the " same " 

It will be observed from these two last cases that any 
increase in the speed of J, above half the speed of 0, causes 


an increase in the velocity of C, and it also causes C to 
revolve in the same direction as the shaft A ; but it will 
be noticed that J requires a high speed, and although it is 
compensated for somewhat by C running in the same 
direction as the shaft, which of course minimises friction, 
its high speed, and the fact that with the flyer leading its 
speed must be still further increased as the bobbin is filled, 
result in the adoption of the present method of driving J 
in the " opposite " direction to 0. A similar course will 
now be followed to show the result of this action on the 
speed of C. Of course when J is stationary, we get the 
same effect as was given in case (2). 

(7) Suppose J revolves at \ the speed of 0, then — 

n= -2a- m, 

n= -H- 

C thus revolves at 1| times the speed of O, but in the 
" opposite " direction. It must be understood, as already 
explained, that the speed of J becomes a minus quantity 
when it revolves in the opposite direction to 0, and this is 
the reason for prefixing the minus sign to 2a. 

(8) Suppose J revolves at h the speed of 0, then — 

71= - 2a- m, 


n= -2. 

The speed of C is shown to be doubled, but the direction 
remains opposite to 0. 

(9) Suppose J revolves at the same speed as 0, then — 

71= - 2a - m, 
?i= -2-1, 
71= -3. 

This result shows a still further increase in the speed of C, 
but the direction is still opposite to 0. 


These three illustrations will exhibit the advantages of 
this method of driving J. It is manifest that, starting 
from zero, any increase in the speed of J will cause an 
increase in the speed of C ; and, moreover, a high speed of 
C can be obtained by a comparatively slow speed of J 
[compare (3) and (7)]. Given this slow speed, when the 
bobbin leads, it is reduced as the bobbin fills. Its great 
disadvantage, however, is that the revolution of C takes 
place in the "opposite" direction to the shaft A upon 
Avhich it revolves. Great friction is consequently set up 
between the two surfaces, and naturally there exists a 
greater strain upon the gearing and the strap of the cone 
pulley, which gives rise to considerable Avear and tear. 
The whole motion has to be constantly lubricated and 
the wheels made perfect in their contact with each other ; 
but in spite of these practical difficulties this form of 
differential motion remained, with only slight alteration 
as originally designed by Houldsworth, up to Avithin a 
feAV years ago, and is even now largely employed. But 
several improved and ingenious methods have been devised 
recently for overcoming chiefly the great fault of the "sun- 
and-planet" motion, or the "Jack-in-the-box" — names 
by which it is frequently called, viz. the revolution of 
C in the opposite direction to the shaft A, and also by 
using a small number of wheels. The reader will see 
from the explanation that the wheel D, although it forms 
part of the motion, and is driven exactl}' like B, is not 
really a necessar}- part of it. The only purpose it serves 
is to balance B on the opposite side of it ; it therefore 
simpl}' represents its weight, and keeps the arrangement 
running steadily. 

One of the most ingenious methods of overcoming the 
practical objections to the " sun-and-planet " motion is the 

i84 COTTON SPINNING chap, hi 

one now presented in Fig. 92 (Howarth and FalloVs patent, 
made by Dobson and Barlow). As the description proceeds 
it will be found that the epicyclic system is entirely dis- 
placed, and a complete novelty in the Avay of differential 
gearing is introduced; it is one of the most important 
mechanical arrangements that have been given to the 
world for a considerable time, and is superior in every 
respect to the motion Avhose place it has taken. The 
following is a description : — A is the dri^^ng shaft to 
which the driving pulleys are fixed, on it is fastened the 
wheel H, through which the spindles are driven ; a little 
farther on to the left is a bevel wheel B, also set screwed 
to A ; this bevel, it will be noticed, gears into another but 
larger bevel C, which is perfectly free from the shaft, and 
is simply mounted upon a portion of a ball or spherical 
bearing D, the long boss of which carries the wheel G, 
from which the bobbins are driven by suitable gearing. 
On the boss of D is also fixed a clutch-wheel S, which 
gears into the teeth P on the back of the bevel C ; if the 
shaft A is now driven, B will carry the bevel C round 
with it. It will be observed that for all practical purposes 
they may be considered as clutch-wheels, the same teeth 
for the time being remaining in contact with each other. 
The same thing happens between the teeth P and those of 
S, so that when A is revolved, the wheels B, C, S, the ball 
bearing D and the wheel G (which is keyed to D) all 
revolve at the same speed as the shaft A, and all in the 
same direction. This condition exists when the bobbin is 
commencing, and therefore no friction whatever is present 
to cause wear and tear on the motion. As a result the 
belt of the cone drums is perfectly free from the usual 
strain to which it is subjected in the older differential 


We now come to the characteristic feature of the motion. 
In order to maintain the bevels B and C in contact with 
each other at the same point, a cam E (which is simply a 
cylinder cut on the slant) bears against the extended outer 
edge of the bevel C, and is revolved at exactly the same 
speed as the shaft. Tliis revolution is produced by keying 
on the boss of the cam a wheel F, and driving it from the 
bottom cone ; the cam is therefore the only means of 
transferring the varying motion of the cone drums to the 
bobbin wheel G, and it is done in the following manner : — 
As the cam E is gradually slowed by the cone drums, it 
causes the bevel C to oscillate on its spherical bearing, and 
at the same time compels it to roll round the bevel B. 
This rolling action will clearly bring into action fresh teeth 
of each bevel, and thus their point of gearing will be 
altered. This, however, would serve no purpose if the two 
bevels B and C were equal in size, for after a single 
revolution the same teeth would come into gear again, but 
in consequence of C being slightly larger (^ larger) the 
effect of its rolling round B is to cause the point of contact 
between the wheels to move backwards. This backward 
motion, although a relative one, results in a direct loss in 
the speed of C. (The point can be readily tested by 
taking two bevels of unequal size and rolling the larger 
one round the smaller; it will be found that after the 
larger one has geared into all the teeth of the smaller one, 
it has a few teeth left that have not been brought into 
contact with those of the smaller wheel. It has therefore 
not made a full revolution, in fact we can say that it has 
lost in speed compared with the smaller bevel.) The same 
thing occurs in this motion, for, directly the speed of the 
cam E changes, fresh teeth on the two wheels come into 
gear, and this means a change in speed of the wheel G 


which is driven from C through the wheel S. It will, of 
course, be understood that it requires a number of revolu- 
tions of the cam E to cause a small difference to exist 
between the speeds of the two bevels B and C, and even 
when they have their greatest variation it is so little that 
they may be said to drive, because they are in gear as 

The function of the teeth P and the wheel S is easily 
comprehended. They save the purpose of transferring the 
motion of C to the wheel G, and are not wheels for driving 
purposes, the only two active wheels being B and C, and 
these, as we have already remarked, differ so little in speed 
that the whole motion is almost silent, and it is not difficult 
to see that very little friction can possibly exist, especially 
since at the commencement of the bobbin the Avhole 
arrangement revolves in the " same direction," and at the 
" same speed " as the shaft A, and it continues in the same 
direction whilst the bobbin fills. The excess speed is 
obtained for the bobbin by making the bobbin wheel G 
slightly larger than the spindle wheel H, The elimination 
of the epicyclic principle in this motion frees the strap of 
the cone drum from the strain, which in the older motion 
it has to sustain in consequence of having to retard the 
large sun wheel ; this is a distinct advantage. 

The calculations connected with this motion are ex- 
ceedingly simple. The wheel C has 36 and B has 32 teeth, 
as the cam E causes the wheel C to roll round the wheel B 
it follows that C will gain or lose speed to the extent of the 
difference of the number of teeth in each wheel ; wheel C 
will therefore gain or lose 36-32 = 4 teeth, i.e. it will gain 
or lose to the extent of ^^ of its motion. 

To find the speed of the bobbin wheel G Ave must 
subtract the speed of F from the speed of the shaft A, 

1 88 


divide the result by 9 and subtract the quotient from the 
driving shaft A, the answer gives the speed of G. 
As a formula it could be expressed thus— 

G = A-i (A-E). 

Speeds are implied in this formula. The speed of the cam 
E is calculated from the gearing through the cone drums 
to F. 

The oiling of the motion is very complete ; the drawing, 
Fig. 92, shows the method of doing it. An outer casing is 




used, which covers up the wheels and also serves the 
purpose of retaining the oil, the projections N distributing 
it on to the teeth and bearing. Its appearance when on the 
machine is that of a large-sized coupling, thus occupying 
very little space, and in addition it is practically noiseless. 

In Fig. 93 is represented Curtis and Rhodes' patent as 
made by John Hetherington and Sons and Piatt Brothers, 
which overcomes effectually the disadvantages of Houlds- 
worth's motion. It retains the epicyclic principle in the 
train of wheels used, but instead of bevel wheels small 
pinions are employed. The number of these wheels that are 


necessary will, perhaps, at first sight, cause the reader to 
think that the advantages of the motion have been obtained 
at the cost of greater noise and friction, and more compli- 
cations than existed in the older motion. That such a 
conclusion, however, would be erroneous, will be seen as 
the description proceeds. In Fig. 93, A is the driving 
shaft, and on it is fixed the wheel G for driving the 
spindles at a constant speed ; on A is also securely fastened 
a disc H, which therefore revolves with the shaft, the 
disc is specially prepared to carry a stud, on one end of 
which is fixed a wheel L, and on the other end a wheel D ; 
the latter is connected to the wheel K through a compound 
carrier N and C. The wheel K, and its long boss B, it 
will be noticed, are loose on the driving shaft, and therefore 
the wheel J, which is fastened on the boss B, and is driven 
from the cone drums, is able to transfer this motion to the 
wheel L through the train of wheels mentioned. The 
pinion L gears Avith an internal wheel E, which is loose on 
the shaft A ; the long boss of E has fixed to it the wheel 
F, from which the bobbins are driven. The connection of 
the bobbin wheel F with the cone drum can therefore be 
clearly traced through the motion. Now let it be supposed 
that the disc H is disconnected from the shaft A, and that 
A is driven, it Avill then be an easy matter to see that the 
wheel J Avill drive the Avheel F through the Avheels K, C, 
N, D, L, and E. The direction of rotation of each Avheel 
can also be followed, which shows F to revolve in the same 
direction as the shaft A ; its speed, however, would be 
rather slow, moreover the revolution of all this gearing 
would result in considerable wear and tear and noise, but 
such conditions do not exist in the motion ; they are 
practically neutralised by fixing the disc H to the shaft, 
thus cau.sinf? it to revolve with it. This revolution of H 


carries round the wheels D, X, and C, and as C revolves in 
the same direction as K, and almost at the same speed 
when the bobbin is empty, it follows that there is scarcely 
any movement of the gearing within H ; consequently wear, 
tear, and noise are reduced to a minimum. During the 
revolution of H the pinion L will be carried round, and the 
mere fact of its teeth engaging with E will cause E to be 
carried round at the same speed ; E will also receive a 
slight additional speed, owing to the small gain existing in 
the gearing of the wheels. As the bobbin fills, the Avheel 
J will be reduced in speed and the wheels within H will 
naturally increase in speed, but the amount is never 
suiJicient to result in any appreciable wear and tear, the 
whole design and the arrangement of the wheels being 
such as to reduce this probability to the smallest degree. 
The motion is balanced by placing a weight on the opposite 
side of the shaft to that occupied by the wheels L, D, etc., 
so that any tendency to vibration is entirely neutralised. 

The speed of the bobbin can readily be determined : it 
depends upon the excess speed given to the internal wheel 
E, and this excess speed can be found by subtracting the 
speed of C (due to its being carried round by H) from the 
speed of K, and following this through the train wheels to 
E. For instance — - 

Revs, of J - revs, of A x - — — — ^ = excess revs, of E. 
Cx D X E 

Xow E has the same number of revolutions as the shaft 

A, plus this excess speed, so that the speed of E or F 

equals — 

F = revs. of J -revs, of Ax^ — ^; — — + revs. of A. 
G X D X E 

Very complete arrangements are made for lubricating 



the motion, and as it is entirely covered in liy projecting 
flanges on the disc H it is thoroughly protected from dirt 
and prevented from causing injury to those attending to 
the machine. 

In Fig. 94 another excellent motion is shown (Tweedale's 
patent), which fulfils all that is claimed for it in overcoming 
the defects of the old sun-and-planet motion. It is an 
epicyclic system of gearing, and bevel wheels are retained 
to effect the desired transfer of motion to the bobbins. 
Eeferring to the drawing, A is the driving shaft, B the 
Avheel driven from the cone drums, C the bobbin wheel 

from which the bobbins are driven, and K the spindle 
wheel. The direction of motion of each part is clearly 
indicated in the sketch by the arrows. Attached to the 
boss of the wheel B is a bevel E, both of which revolve at 
the same speed, and run loose on the shaft A, and in the 
same direction ; E gears into a bevel F, carried by a stud 
which passes through the driving shaft, whose other end 
carries a bevel H ; this gears into a large bevel wheel D, 
to the enlarged boss of which is fixed the bobbin wheel C. 
The two latter wheels revolve as one, and run loose on 
the shaft. The action of the motion can now be made 


clear. As the shaft A revolves, it carries bodily round 
with it the wheels H and F, and in so doing wiU impart 
a similar motion to the bevel D ; but in addition to this, 
the driving through B from the cone drums will give a 
motion to H round the centre of the stud on which it is 
carried, which will considerably modify the motion given 
through the bodily motion of H and F. The combination 
of these two movements enables the bobbin wheel C to 
be driven at the required speed, and also in the same 
direction as the shaft. This brief statement is, however, 
not sufficient to explain the principle of the motion, and 
as it possesses one or two features of interest it will be 
advantageous to examine it a little more in detail, in order 
to see why such a result can be obtained, and more 
especially why the direction of motion should be as de- 

C is the wheel through which the bobbins are driven, 
and therefore a resistance is offered to the driving of D, 
and it remains stationary until some force acting upon the 
wheel D sets it in motion by overcoming the resistance. 
Now the shaft A carries round the cross shaft G at its 
own speed, say 400 revolutions, but this motion can have 
no effect direct!}' on D simply because the bevel H com- 
mences to yield instead of D, and H transfers its motion 
to F and so on to the wheel B. Owing, however, to the 
direction of the axial revolution of F and the direction of 
the revolution of the cross shaft, B would not revolve at 
a great speed because the two movements partially neutralise 
each other. 

What is now needed is to drive B at such a rate and 
with sufficient power to revolve H in the opposite direction 
to that given to it through the movement of the cross 
shaft, "When H is thus driven it will cease to move on 

Ill FLY- FRAMES 193 

its axis, and yet the cross shaft carries it round, conse- 
quently it takes D with it and the bobbins arc driven 
through C. The formula is very sim})le, and depends on 
the same formula as laid down on p. 141. 

Revs, of A - (revs, of A - revs, of B) x — ^ — =revs. of C. 

1'' X D 

. 1 ExH 18x30 7 

As a rule — = = - 

FxD 16x48 10" 

If this be condensed into symbols where the speed of A 
= m, the speed of B = a, the speed of G = », and the ratio 
between the wheels 

we get 


n = m — {m — a)r. 

By substituting actual speeds and wheels in the formula 
just given, it will be an easy matter to find the speed of 
the bobbin Avheel C (which is part of D) and from it the 
speed of the bobbins. 

- The dotted line shows how the motion is covered in to 
prevent accidents to workers. The lubrication of the 
various parts is well arranged for, and the Avhole is well 
balanced, so that friction is reduced to a minimum, and 
the strain on the cone drum strap diminished considerably. 
The next example is the one given in Fig. 95. Its 
essential features arc as follows : — A is the driving shaft, 
to which, as usual, the spindle wheel H is fixed. A little 
farther along the shaft is also fixed a wheel B, whicli gears 
into a wheel E, carried l)y a disc C ; E is one of a compound 
pair of wheels E and F, both of which are cast together 
and work on the same stud N ; F gears into a wheel D, 
which runs loose on the shaft A. Cast upon 1) is a long 
boss, upon which is fastened the bobbin Avheel G. We 



thus see that G is driven direct from the shaft, and by 
following the gearing we shall notice that it runs in the 
same direction as A, so the motion fulfils its principal 
object in reducing friction to the lowest possible point. 
The connection with the cone drums is made through the 
wheel J, which is set screwed to the boss of the disc C, the 
disc running loose on the shaft A, and in the same direc- 
tion. By this means the wheels E and F are carried bodily 
round the shaft A, and by this action influence the speed 

C K 

of D and consequently G — and it will be noticed that this 
action is one which does not interfere with D revolving in 
the same direction as the shaft. 

The whole motion is of such a simple character that it 
can easily be understood, and the following remarks be 
readily understood: B has 30 teeth, E and F 18 each, and 
D 33 teeth. Tlie irain between B and I) is therefore — 

18 '^33^33' 


Now if C be revolved at the same speed as B, E and F will 

Ill FLY- FRAMES 195 

haA'c a "relative" motion, which will result in D leAolving 
at exactly the same sj^eed as the shaft. If C be slowed, D 
will lose in speed by an amount equal to the proportion in 
which B is less than D, Avhich is ^^ ; so that if C revolves 
eleven revolutions less than B, the wheel D will have one 
revolution less than A. In the same way any increase in 
the speed of C over B will cause an increase in the speed 
of D in the same proportion as above, viz. y^p so that if 
C revolves eleven revolutions more than B, D will revolve 
one more than A ; C, however, is never required to run 
quicker than A. From this statement it is an easy matter 
to calculate the speed of the bobbin Avheel G, and from it 
that of the bobbin. 

The motion is thoroughly cased in by C and K, and 
special attention has been paid 10 the arrangements for 
lubrication through the studs N. Wear and tear due to 
friction is almost eliminated, and comparatively very little 
motion of the wheels E and F can take place, so that the 
gearing adds little, if anything, to the noise of the frame. 
In all the examples that have been given, one great advan- 
tage is the more regular winding that results from their 
working. The power hitherto transmitted thiough the 
strap of the cone drums has always had a tendency to 
cause slippage, but since the important factor of friction 
has been so considerably reduced this element of uncertainty 
in winding has almost disappeared, and the bobbins can 
now be relied upon to start at the same time with the 
spindle and maintain the true relationship of speed through- 
out the building of the jjobljin. 

Traverse Motion for Bobbin.— The next feature of 
the fly-frame to which attention will be directed is that of 
the traverse or reversing motion. Its functions are of a 
very important character, and therefore before describing 


it, it will be necessary to make a few preliminary remai'ks, 
in which the reason for its use will be defined and illus- 
trated. In the building of a bobbin, the operation must 
be performed in such a way that no change of shape can 
take place in it, and also so that its form shall permit of 
its being handled and carried from one machine to another 
Avithout causing any damage to the roving. There are 
three methods of fulfilling these conditions, viz. building a 
bobbin by a quick traverse motion ; using bobbins with 
flanged ends ; and building bobbins formed with conical 
ends on the cotton itself as 'the bobbin is filled. All these 
methods are practised in the processes of cotton sjDinning, 
but in the machine under discussion the latter method is 
the one usually adopted. 

The extreme weakness of the roving renders it impera- 
tive that very light bobbins be used, in order that they 
may permit the roving to be taken off from them Avhen in 
the creel. Plain barrels are therefore used, and the roving 
is wound on in a manner that gives the result shown in 
Fig. 96. An examination of this sketch will show the 
necessity of shortening the traverse from A, B to C, D, in 
order to give a conical form to the ends. The bobbin rail 
must therefore be reversed earlier after each successive 
layer has been added. (It is from this fact that we get 
the term reversing motion.) Incorporated with the revers- 
ing motion is the arrangement for moving the straps 
forward along the cone drums as the bobbin fills. Each 
additional layer requires a slower speed than the preceding 
one, and consequently each change of the traverse auto- 
matically removes the strap to the required position on 
the cone drums, the movement of course being equal for 
each layer. Now since the hank of the roving may vary 
through a long rani^e it will be necessarv, when a coarse 



hank is being wound on the bobbin, that each layer will 
require a greater lateral movement of the strap than when 
a finer hank is used. This is obvious : for instance, suppose 
the strap starts and finishes the same for a fine and a coarse 
hank bobbin ; the number of layers in the diameter of the 
bobbin containing the finer roving will be many more than 
the layers in that of the coarse roving, and consequently 
the larger number of movements 
of the strap must be correspond- 
ingly reduced in amount in order 
to equal the fewer movements 
which occur in Imilding the coarse 
roving bobbin. For example : — 
SujDpose a 2 -hank roving be 
wound on a bobbin 5 in. in dia- 
meter, there will be about 120 
layers, and each layer will require 
a movement of the strap. The 
amount of this movement on a 
drum 30 in. long will be j%i) = \ 
of an inch. 

If a 5 -hank roving be wound 
on a bobbin 5 in. in diameter, there Avill be about IGO 
layers ; each layer will require a movement of the strap 
of -^\^Ti = tV of an inch. This lessened movement is 

1 o O 10 

arranged for in the reversing motion. 

Another feature quite evident in connection with the 
number of layers or coils in the length of the traverse is, 
that for coarse roving the movement of the rail must be 
much quicker than for fine rovings, and an alteration will 
therefore be required when a change of roving is made. 
This point will be dealt with more fully at a later stage. 
It is sufficient to mention here that if a 2-hank roving has 



• !;; 
: li' 

; p5 

: i 
; '^> 


Pig. 9G. 


twelve coils per inch lift, a 5-liank roving will have 23 
coils per inch lift; this means that the bobbin rail will 
make its traverse at only a little over one-half the speed 
for a 5-hank roving as compared with the 2-hank roving. 

It will also be obvious, especially on reference to the 
gearing plans of the machine in Fig. 81 and Fig. 97, that 
the coils depend on the speed of the cone drums, the rack 
being driven from this point ; and, as this speed is a varying 
one, it would seem to follow that the pitch of the coils on 
each layer would vary. We certainly do get a slower 
traverse, but at the same time the bobbin is slowed to 
the same extent, and consequently the pitch of the layers 
remains constant throughout the bobbin. 

The actual building of the bobbin or traverse of the 
bobbin rail is performed by the bottom cone drum, through 
the wheels F, R, S, T, D, U, and the train of wheels leading 
up to the rack. The reversing of the traverse is performed 
by the reversing motion, whose position on the machine is 
shown at C in Fig. 97 ; its action is such that at the finish 
of the traverse the bevel U is thrown out of gear with D, 
and the other bevel is brought into gear with it. At the 
same time an effect is produced which causes a lateral move- 
ment of the cone drum rack, and through it the strap is 
moved along the drums. 

The motion itself will now be described. It is suffi- 
ciently complicated to require a careful following of the 
description and reference to the illustrations. 

In Fig. 98 a complete drawing is represented of a well- 
known form of building or reversing motion. Explanatory 
sketches also are given in Figs. 99 and 100, Avhich will 
enable the description to be more easily understood. 

It will be remembered that the motion has a threefold 
purpose, viz. it alters the position of the strap on the 

Ill FLY- FRA MRS 1 99 

cone drums so tliat the bohljin is slowed in speed as it 

enlarges in diameter : this action irivcs it the character of 


a building motion — a name sometimes applied to the whole 
arrangement. It regulates and reverses the traverse of the 
bobbin or lifting rail, and is frequently called the "traverse" 
or "reversing" motion because of this action. It makes 
conical ends on the bobbin, by causing the reversing 
arrangement to take place a little earlier as each layer is 
added to the diameter. 

Each of these effects will be noted separately. The 
first-named is effected through the upright shaft B and the 
wheel U, Avhich moves the long rack V forward in its 
slide. The connection of the rack with the strap forks 
gives the required movement to the strap on the cone 
drums. The second effect is obtained by an oscillating 
movement of the cradle J about the centre A ; this gives 
a backward and forward motion to the rod Z, which 
ultimately puts the bevel wheels in and out of gear with 

Ill FI.y-FKAMES 20I 

D. The third action is governed by a small pinion A (see 
R, Fig. 99) gearing into the rack S, and its revolution 
altering the position of the pin I in the slide T. 

The slide T is the governing element in all three 
motions. It is attached to a rail which is connected 
directly to the bobbin or lifting rail, and the vertical 
movement along in, n, which it thereby receives, controls 
each of the actions mentioned above. The whole consists 
of a frame, firmly fastened to a cross rail. The frame 
contains a centre boss A, on which are placed two cradles 

J and W, in a manner that leaves them perfectly free to 
oscillate independently of each other. On a stud A, Avhich 
passes through a hole in the boss, is fixed a ratchet wheel 
C, and also a bevel, which gears into the bevel H on the 
upright shaft B. A chain weight at ^l' passes over a guide 
pulley p, and is wound on a bowl Jc keyed to the upper 
part of the shaft B. The weights to will naturally tend 
to cause the shaft B to revolve, and consec^uently the 
ratchet wheel C, but this action is prevented by detents 
or catches D or E, which engage in the teeth of C, and it 
is only when these are freed from contact with it that the 
weight IV gives motion to the shaft. Such a motion, when 


produced, turns the wlieel U, and the long rack V is 
moved forward. 

The method of releasing the ratchet can now be ex- 
plained. The rack S passes through a slide, which forms 
part of the upper cradle W. Its other end is connected 
to the pin I, Avhich slides in the guide T. The up-and-down 
movement of the rail gives an oscillating motion to the 
cradle W round the centre A. On the cradle arc fixed 
two sets of screws a, h, and c, d. The first pair are for 
releasing the tumbler or j^igeon catches X and Y, while 
the second pair are connected with special hooks e and /, 

Fig. 100. 

to which latter are attached strong springs K and L. The 
springs are held by the ends of a cross piece M, which is 
centred on a knife-edge at carried by the rail. It will 
now be seen that the oscillations of the cradle W will have 
the direct effect of putting either of the springs in tension 
according to the direction of its movement ; for instance, 
if d moves upwards, the hook / will be free from the cradle 
J, the tension in the spring L will draw the opposite end 
of M downwards, and when the projection on the hook e 
prevents further movements by its coming into contact 
with J, the spring K Avill be stretched and put into tension, 
therebj'' exercising a strong pressure on J, tending to turn 
it round its centre A. A projection on g on the upper 

Ill FL Y- FRAMES 203 

part of J enables tlie tumbler catcli X to prevent any 
movement of J, and it is only when the lifting rail has 
reached the limit of its traverse that the oscillation of W 
brings the setting screw a into contact with X, the down- 
ward pressure of a causing the tumbler to be freed from 
the projection g. Directly this occurs the tension of the 
spring K produces a quick movement of J, which bodily 
moves round the centre A ; the lower part of J carries a 
double finger N, and as J suddenly changes position this 
finger, moving with it, gives a smart blow to the catch E, 
and releases it from the teeth of the ratchet wheel C. 
When this release takes place the weight w instantly 
commences to turn the ratchet wheel, but only a fraction 
of a revolution is permitted, for the spring which connects 
the two catches D and E instantly brings D into the path 
of the teeth, and prevents more than a regulated amount 
of movement taking place. The amount which is obtained 
is imparted through the bevels H to the upright shaft B 
and the wheel U. The long rack is correspondingly 
affected, and produces the change of the strap which alters 
the speed of the bottom cone drum. The same action is 
repeated during the descent of the rail ; h releases the 
catch Y, Avhich engages with the projection cj ; the cradle 
J under the pressure exerted by the spring L, oscillates 
backwards, and in doing so forces D out of contact with 
the teeth of the ratchet, and permits E to be brought into 
action. This alternate effect, due to the oscillation of the 
cradle "\V, continues throi;ghout the building of the bobbin. 
The to-and-fro motion of the lower cradle J is taken 
advantage of to obtain the reversing of the lifting rail. 
Attached to the low^er portion of J is a rod Z, which is 
connected with the bevels that gear into the bevel I). As 
the change is therefore made in the position of J the rod 


Z is moved nt the same time, and a change in the direction 
of the movement of the rail is obtained. 

Ill regard to the ratchet wheel, it will be seen that the 
amount of movement given to the long rack depends 
primarily upon the number of teeth contained in C. A 
smaller number of teeth will be required when a coarse 
hank is being wound than when a fine hank is used. The 
quickness of the movement of the catches prevents more 
than a single tooth of the ratchet "escaping," no matter 
(within reasonable limits) how small the pitch of the teeth 
may be. As a rule in all frames the catches are set in 
relation to each other, so that only half a tooth of the 
ratchet escapes, the catch E being made with a setting 
arrangement, which allows a regulation to be made for 
this purpose. 

We can now deal with the part of the motion which 
affects the shortening of the traverse, and produces the 
taper of the ends of the bobbin. On the stud A, Avhich 
carries the ratchet C, is fixed a small pinion E. (Fig. 99). 
Each escape of C causes a similar movement of E, and as 
it gears wdth the rack S it moves it forward. This alters 
the position of the pin I in the slot of the guide T. Now 
it will be seen that the position of this pin regulates the 
time when the setting screws A and B depress the tumblers 
and relieve the cradle J. When the bobbin commences, 
the lift from I to K is arranged to produce the change at 
the termination of the up-and-down lifts. Each successive 
change moves the pin forward, and thus shortens the 
leverage of the rack about the centre A. As a consequence 
the setting screws A and B will be brought into contact 
Avith the tumbler catches at an earlier period of the traverse, 
and the reversing will be performed before the completion 
of the original traverse. Extreme positions are shown in 



the Diagram 100, and it will readily be understood that 
the traverse from J to L will have exactlj^ the same effect 
in oscillating the cradle W to the same extent as Avhen the 
traverse was from I to K. The sketch shows, by means of 
full and dotted lines, the positions of the cradle, catches, 
and raclv, for each extreme position. 

Fig. 101. 

Another form of reversing motion is represented in 
Fig. 101. Its principle of action is exactly the same as 
in the example just described, but it differs in one or two 
features of a practical character from the arrangement in 
Fig. 98. 

The upper cradle W is centred on the stud A ; the rack 
S passes through slide brackets fastened to W, so that the 
cradle oscillates with the rack al)Out the centre A. On the 


stud 1\. is fastened the ratchet wheel T, the bevel wheel for 
giving motion to the upright shaft (whose centre only is 
shown), and also the wlicel li, gearing into the rack S. 
The oscillation of the cradle W causes the set screws M 
and N to relieve the tumbler catches X and Y alternately. 
This action sets free the lower cradle J, and it makes a 
short sudden movement about its centre at B. This 
movement causes a pin fixed in its lower part at J^ to 
move the lever C sideways, and in doing this the upper 
part H of C knocks the catch which holds the ratchet 
wheel out of position, and permits the ratchet to make the 
necessary amount of movement. The releasing lever is 
centred at C ; its other end D is attached to the rod Z, 
Avhich puts the reversing bevels in and out of gear. 

It will be noticed that the two cradles, W and J, are 
on different centres. This arrangement is one that shows 
a variation in the practical way of obtaining the same 
result, and as they both give equally good results com- 
parisons are unnecessary. Another difference of more 
importance consists in the substitution of the weights K 
and L for the springs in the })receding motion. Weights 
have the advantage of being of a definite character, and 
always remaining so. Springs are not so well defined in 
the pressure they exert, and they are liable to change 
through continual use, unless very well made. Weights, 
however, have the disadvantage of their Avhole effect being 
brought to bear suddenly on the upper cradle, and when 
the tiimbler-catcb is relieved the full force of the weight 
K or L produces the change of the cradle J. The shock 
of the sudden stoppage of the fallen weight is a distinct 
disadvantage, which in the case of a spring is almost 
absent. A spring exerts its full pressure at the moment 
ot the release of the cradle, l)ut the pressure of the spring 


is reduced as the cradle yields to it, and -when the cradle 
reaches its extreme position the spring is free from tension. 
As the cradle W oscillates, it will be observed that al- 
ternately the full pressure of the -weight or spring comes 
upon it. The pressure exerted is considerable, and as 
it acts either at P or Q it produces a frictional effect 
between the rack and the pinion, which prevents the pinion 
turning freely. The drag weight w in Fig. 98 is the only 
means that enables R to revolve ; consequently a very 
heavy weight is required here to overcome the friction, and 
even then a slight delay is occasionally caused, Avhich 
prevents the strap from being moved in order to wind on 
the new layer, and tight winding is the natural result. 
It is only fractional in its character, and scarcely observable, 
but it is one of the many small irregularities that may 
interfere with perfect winding. 

A well-known firm of machinists, Howard and Bullough, 
have successfully introduced a method of relieving the 
upper cradle of the pressure mentioned above. It is illus- 
trated in Fig. 102, in which a reversing motion is shown 
complete. Both cradles are on the same centre as the 
ratchet wheel. Now, instead of the weights H and H^ 
hanging from the upper cradle, an arrangement is intro- 
duced Avhich is entirely unconnected with them. A 
hanging lever A, fastened to the guide B, has its lower 
end fitted with stop screws A^ and A^, so that during its 
rise and fall they come into contact with the end of a lever 
C, whose fixed centre is on the framing at D. The weight- 
hooks E and E^ pass through holes in the projections F 
and F^ of the lower cradle. They are also prepared with 
stop-pieces, which fit special notches cut in the lever C. 
As the lever C oscillates, it alternately raises one of the 
weights and frees the other. For instance, suppose the 



lifting rail is mo\ing upwards, the stop A" lifts up the end 
of the lever C ; the notch coming under the stop on the 
hook E raises the weight H, ami thus takes its pressure off 

the lower cradle. At the same time the other end leaves the 
stop on E^, and the weight H^ is then compelled to hang from 
the cradle at F, and in this position it is ready to produce the 


change which results in the escape of the ratchet wheel. The 
opposite effect is obtained on the descent of the lifting rail, 
Avhereby H^ is relieved, and H is brought to bear on the 
cradle. It will be seen that the upper cradle is uninfluenced 
liy the weights, and therefore the rack and pinion work freely 
with each other. A normal drag-weight W only is required to 
turn the upright rod as a tooth escapes ; the change is instan- 
taneous, and all chance of delay from this cause is eliminated. 
In view of the importance of obtaining as great a degree 
of freedom from friction the arrangement shown in Fig. 103 
is used by Brooks and Doxey. Its main features will be 
readily understood if the previous descriptions have been 
closely followed. The diminishing or tapering rack is 
shown in its middle position, and the top cradle has moved 
so that the left-hand weight (not shown in drawing, but 
the w^eight hook indicates its position) is already hanging 
from the bottom cradle ready to pull it over immediately 
the adjusting screw on the top cradle comes down low 
enough to release the pawls that hold the bottom cradle in 
position. The movement of the bottom cradle actuates a 
vertical lever fixed on its shaft, and the lower end of this 
lever is connected to a bracket fastened to the reversing 
shaft or rod. This bracket carries an arm hollowed out 
at each end into Avhich the lower extremities of the levers 
carrying the ratchet catches or pawls are inserted. As the 
lower cradle rocks to and fro due to the rise and fall of 
tlie lifting rail of the frame, the dished arm will act on 
the ends of the levers alternately and push the catch out 
of gear with the ratchet wheel, and at that moment of 
freedom the weight hanging from the winding wheel boss 
will move the diminishing rack forward and so through the 
rack wheel give motion to the cone drum rack. Also at 
the same time, the displacement of one of the catch lever 



arms puts the spring between the two arms in tension, and 
the other catch lever arm is pulled into gear with the 

ratchet teeth and so prevents further movement. It will 
lie noted that the drag chain is attached to a wheel separate 
from the building motion, and also that the winding back 


operation is done through the gearing shown in shaded 
lines and in a position convenient to the worker. 

To eliminate probable errors and obtain as great a degree 
of accuracy as possible in the mechanical details of the 
machine, attention is directed to the method of driving the 
bobbin-shaft wheel at the time of its making the traverse. 

Fio. 104. 

In several makes of machines an arrangement is adopted 
consisting of four wheels, as shown in Fig. 104, Avherein A 
is the driving-shaft wheel and D the bobbin-shaft wheel. 
This latter wheel, through its being carried by tlie top rail, 
is compelled to move vertically, as shown by the arrow. In 
consequence of this movement the wheels B and C?, w^hich 
connect it to E on the driving shaft, must be carried on 
movable centres in order to compensate for the vertical 


motion of D. The usual method of doing this is to co'nnect 
B, C, and D by a bell-crank lever in one piece H J ; Avhile 
B is connected to E by the link G. By this arrangement 
D can readily be moved in the required direction. 

A careful examination of the action of this system will 
show that it presents an epicyclic train of wheels, carried 
by arms G, H, and J round the centre A. This being so, 
we shall find that while the centre of D moves vertically, 
the wheel itself turns slightly on its own axis. This axial 
rotation is transmitted to the bobbins, and introduces a 
perceptible variation in the roving; the amount of the 
turning is shown on the wheel at 1. If X was a point 
marked on the wheel, it would be found that it would 
occujiy the position Y after three-quarters of the lift had 
been performed, and the position Z after the remaining 
one-quarter. The effect of this would be to stretch the 
roving during the latter part of the up lift, and to produce 
the opposite result during a similar period of the down 
lift. (The exact amount cannot readily be calculated, but 
it is an easy matter to test the question practically ; it 
is then found to give an extreme variation of about 3 per 
cent, or H per cent on either side of the hank required.) 

To overcome the above defect and neutralise the rota- 
tion of D, a Avell-known firm have introduced the improve- 
ment shown in the drawing (Fig. 104). Instead of carrying 
B, C, and D by a bell-crank lever, H and J are made in 
separate links ; at the same time C is also carried by an 
extra lever, centred at K. The position of this centre is 
such that the movement thus jDermitted to C compensates 
for the axial movement of D, and it is found that with 
this improvement more regular roving is produced, which 
demonstrates its effectiveness. 

In several other makers' machines the disposition of 



the wheels is made to difler from the above. Fig. 105 
will show the arrangement : here only three wheels are 
used, the centres being connected by the links D and E ; 
A is the driving-shaft wheel, and C the bobbin-shaft wheel. 
As C is moved vertically, the wheel B will describe a 
circular path round the centre of A, the link D keei)ing the 

two wheels in gear. 

Now this arrangement is also of an 

Fm. 105. 

epicyclic character ; but since it consists of an odd number 
of wheels (three) instead of an even number, as in the 
previous illustration, the wheel C is by this very fact 
prevented from making any axial rotation. In spite of this 
advantage, however, the movement of wheel C in a vertical 
direction, instead of a circular one round A as centre, changes 
the point of contact between the Avheel C and B, and so 
interferes slightly with the speed of A ; otherwise it is one 


of the best mechanical arrangements known for preventing 
axial rotation. In both illustrations the extreme move- 
ments of each wheel are shown by the dotted lines. 

Several minor matters of detail still remain, which are 
worthy of a little consideration, but the remarks concern- 
ing them must of necessity be brief. The foregoing 
examination into the principles and construction of the 
fly-frame, however, will enable them to be readily under- 
stood Avithout going to undue length in explanation. 

A feature that is at first puzzling to the student is the 
varying speed at which tlie different lifts take place as 
the bobbin fills. It will be noticed that the first layer 
occupies a much shorter time in its performance than the 
last lift. This is explained when it is found that the 
gearing which operates the lift is directly connected to 
the bottom cone drums. The reason for it is also obvious : 
for it is quite clear that if the layers are wound on quickly, 
as on the bare bobbin, the lift must be made quick enough 
to put the layers or coils on at a suitable distance apart 
according to the hank being wound. AVhen the bobbin 
increases in diameter the winding is slower, and therefore 
the lift must be slowed to compensate for this : in fact the 
connection between the lift and the Avinding is a direct one, 
and a reduction in speed of the one must be followed by 
a similar reduction in that of the other. This explains 
why the pitches of the coils are constant throughout the 
building of the bobbin while the lift that lays them is 
itself altering in speed. 

The balancing of the bobbin rail is one of those smaller 
questions that prove troublesome occasionally, but the 
present-day machines are carefully attended to now in 
this respect. The usual method is to hang weights by 
means of chains passing over pulleys from the rail brackets. 

Ill FLY- FRAMES 215 

winch woi'k in slides on the face of the spring pieces; the 
amount of the weights is arranged to halance the rail Avhen 
its bobbins are half-full. In most cases this is a matter 
of guesswork, but it is sufficiently near for practical 
purjjoses. Another method is to attach the weights by 
means of chains to a lever, which rests under the centre 
of the rail, and of which the fulcrum is situated as far away 
as possible in a line with the middle of the lift. This 
latter method prevents any possibility of the rails sticking 
in its slides^as is quite possible in the former case, 
unless great care be taken in forming the slides. An 
ingenious method was tried some time ago, which provided 
for the balancing of the rail automatically under all cir- 
cumstances. The chain, connected to the slide brackets 
of the rail, passed over guide pulleys, and was fastened to 
a lever, which carried a weight heavy enough to balance 
the empty rail. This weight, by means of a screw and 
ratchet wheel, could be moved along the lever aAvay from 
the point of suspension of the chain, which was done 
through the ratchet wheel being acted upon at each up- 
and-down lift. Greater pressure was in this way exerted 
by the weight in its new position on the lever, and a 
well-balanced rail was the result. 

When a set of bobbins has been completed, the strap 
on the cone drums must be wound back to its starting- 
point. In order to do this it is necessary to raise the 
bottom cone drum a little, so that the strap is slackened ; 
when this is done by the tenter from the front of the 
machine, she is able to wind the strap back, ready for the 
next set. In most machines the two operations, viz. 
lifting the cone drum and winding back the strap, are 
performed from the front of the frame, thus saving time 
and trouble in iroine: round the machine to the back. 


Although the point has been mentioned before, it is 
important the student should thoroughly understand that 
the full length of the cone drum is not used for all sizes of 
bobbins : it is only when the ratio between the empty and 
full diameters corresponds to the ratio between the extreme 
diameters of the drums that this is the case. \Ye have 
seen that the form of the drums has been obtained irre- 
spective of the speed of the bobbins, and it wall be readily 
understood that the strap can always start at the same 
point (no matter Avhat the diameter of the empty bobbin 
may be), provided the initial speed is made correct for 
that diameter by arranging suitable gearing between the 
cone drum and the bobbin shaft. When the bobbin is 
large enough for the purpose in view, the strap will be 
on the drums in the position necessary for that diameter ; 
and it is clear that, for some bobbins, especially in the 
finer roving frames, very much less than the full length of 
the drums is utilised. 

Ratchet Wheel. — In all fly-frames there is an element 
of uncertainty in the star ratchet wheel in regard to the 
number of teeth it ought to contain for different hanks. 
Experience and trial in most cases settle the matter in the 
first ])lace, and afterwards the Avheels thus found are used 
for finding others in proportion. Attempts have been made 
to arrange a table from which the star "wheel can be 
calculated, which depends on the number of layers per 
inch in the diameter of the bobbin. Approximate results 
are in this way obtained, but there are so many degrees of 
tension put into the roving — ranging from soft bobbins to 
tightly wound bobbins — that very little reliance is placed 
on any one basis by practical men. These calculations are, 
however, useful as a guide, and therefore the following 
extract is given from Mr. W. H. Cook's Tlaiix Series of 

in FLY-FRAMES 217 

Cotton Spinning Calculations, in which such a table is 
published : — 

Layers per inch Lift 

1 liank 





1-1 to 2 lianks 





2-1 ,, 3 ,, 





3-1 ,, 4 ,, 





4-1 ,, ,, 





General Notes. — In bringing to a conclusion the 
description of the fly-frame, it will be convenient to make 
a few general remarks on the ultimate results of its work. 
AVe have seen how imperative it is that a high ideal of 
accuracy should be kept in view, and the descriptions will 
have shown to what extent mechanical action has attempted 
to attain to this result. It is only when an examination is 
made of the roving on the bobbin that a correct judgment 
can be formed of the real work of the machine. Bad work, 
when it exists, may be traced to the following different 
causes, viz. faults in the mechanism ; faulty setting of 
rollers, etc., and speeds; carelessness in attending to the 
machine ; faulty rovings, the results of previous operations ; 
and poor cotton. Each of these gives rise to inequalities in 
the roA-ing, which may be reduced to a minimum with 
proper care and supervision ; but, in spite of all that can 
be done, there remains the inherent property of the fibre 
itself to be taken into account, and the apparent impossi- 
bility of obtaining ideal mechanical conditions for dealing 
with it prevents absolute uniformity from being obtained. 

Faults in the mechanism are to be seen in broken 
teeth ; too much backlash in gearing ; loosened brackets ; 
slides sticking in grooves through friction ; slippage of 
wheels on their shaft, etc. Faulty setting of rollers, etc., 
and speeds, give rise to a lot of bad work, which at first is 

2 1 8 CO TTON SPINNING chap. 

not easily traceable to the real cause ; drafting, spacing, 
and twist are all-impoi'tant points, which require attention 
in getting the best results from the cotton and the machine, 
and, in a measure, the required exactness depends ulti- 
mately upon experience and knowledge of the special 
conditions governing the case. 

Carelessness, of course, is responsible for many faulty 
rovings. If the top rollers are not oiled at regular intervals, 
or a sticky oil that gums be used, it causes the rollers to 
momentarily stop, and a thin place is instantly made. Long 
piecings are a distinct evil. Roller laps increase or reduce 
the draft, and so alter the hank slightly. Slippage of the 
cone drum strap occurs through slackness or allowing oil to 
run on it. SlijDpage of the bobbins and also the momentary 
slippage of the flyer occur when it has not dropped into its 
slot before the machine has started; allowing "single" to 
go through ; and also when a case of "back double" occurs. 

Previous operations are also responsible for bad work. 
Unequal laps from the scutcher naturally transmit the 
evil to all future processes, and too much care cannot be 
insisted upon in attending to this matter. Bad carding 
and single in the draw-frame lead, to inequalities, which 
the fly-frame cannot get rid of, and therefore they exist of 
necessitj'^ after passing those machines which produced them. 

In regard to poor cotton, the inequalities from this 
cause are to a certain extent expected, but bad judgment 
in mixing will lead to poor results in the roving and yarn, 
and of course these can be detected in each process, but 
the cause is not always so readily found. 

Calculations.^ — An illustration of the gearing part of 
the fly-frame is given in Fig. 107 (reproduced from i)age 155), 

' Full calculations with geai'iiig jilaiis of tlie cliiet type of fly-fraiiios 
arc given in the author's book Volton SidnniiKj Calculations. 

Ill FLY- FRAMES 219 

and fov the purpose of the calculations reference may also 

be made to Fig. lOG (which is practically the same, witli the 


exception of the differential motion, and is reproduced from 

page 199). There are six distinct change places, at all of 


wliich alterations can be made to attain or suit various 
conditions. These are specially marked in the drawing by 
an asterisk, as at A, B, C, D, E, and F. 

AVe liave already referred to the driving of the spindles 
and bobbins, and shown that they both receive their 
motion direct from the driving shaft, except the " excess 
speed of the bobbin," which, of course, is transmitted from 
the cone drums. This ought to be distinctly noted, as it 
will prevent much misunderstanding in regard to the 
purpose of both the differential motion and the cone drums. 
The rollers are driven from the driving shaft through the 
twist wheel B, which, it will be seen, also drives the top 
coue drum. If, therefore, the front roller be altered in 
speed, the speed of the cone drums will be altered to the 
same extent ; the reason for this will become clear Avhen 
we make a practical application of the rules. 

The bottom cone drum drives the sun wheel of the 

differential motion as well as the lifting motion. The 

gearing for the latter can be traced through the following 

wheels, the top line representing drivers, and the bottom 

line the driven wheels : — 

FxSxDxaxcx e 
R X T X U X 6 X f^ X rack' 

The objects of the change j^laces and the usual extent of 

the change, in the machine illustrated, are given in the 

following list : 

A is the change place for Draft . . . Wlieels 35 to 60 teeth 

B „ „ Twist ... ,, 26 „ 48 ,, 

C „ „ Winding . . ,, 10 ,, .'iO ,, 

D ,, „ Traverse . . ,, 16 ,, 26 ,, 

■r, ia. change in the dia. "1 1 - o/^ 

\^ 01 enii)ty bobbin J " " " 

■p /botli traverse and dia.\ ,, „j, 

" " \ of empty bobbin J " " " 

The latter change is one that is very seldom resorted to, 


as the other change places do all that is usually necessary. 
A careful observation of the result of a change in the 
number of teeth of F will show us that it will affect two 
features of the bobbin, viz. the tension of the roving and 
the closeness or pitch of the coils in the traverse. When, 
therefore, a combination exists of slack ends and coils 
that are too close together, a remedy may be found in 
putting on a large wheel at F, which naturally increases 
the speed of the bobbin and also increases the speed of the 
lift, thus tightening the ends and making the pitch of the 
coils greater. 

The necessity for the changes may be best illustrated by 
an example. Suppose a certain hank is being produced 
on the machine and it is desired to change to a higher 
hank, the same roving being used in the creel. The 
necessary changes in gearing are as follows : — 

Draft. — This must be altered, in order to make the 
delivered roving thinner ; A will therefore be made smaller 
in order to drive the back roller more slowly. 

Traverse or Strike JJliecl. — This change must be made, 
because the thinner roving which is now delivered will 
require to be Avound in coils of less pitch than for the 
thicker roving; consequently D is made smaller, so that 
the lift can be reduced in speed. 

jrindinr/ or Star IJlieel. — The thinner roving also requires 
that C should be enlarged in order that a lessened move- 
ment may be given to the rack at each up-and-down lift. 
There will be moi'e la3'ers per inch of diameter for the finer 
than for the coarser roving : therefore, each action of the 
lift must give a shorter traverse to the rack which moves 
the cone drum strap ; a larger number of teeth in C will 
effect this 

Tivist. — A fine roving has more twist put into it than a 


coarse one ; Tj is changed for this purpose, and by making 
it smaller the speed of the front roller is reduced, but since 
the speeds of the spindles are not altered more twists will 
be put into the roving. From this latter change it Avill be 
noticed that the speed of the cone drums is also changed, 
which is a necessary consequence of changing the speed of 
the front roller : for if the front roller be slowed the 
bobbins must have the same reduction in speed in order 
to wind on the smaller amount of roving that is delivered 
to them. 

The following table presents the reader with all the 
essential particulars required in working out any of the 
calculations in connection with the fly-frame. They, of 
course, A^ary greatly in other machines, but the applications 
of the calculations are similar : — 






Speed of frout roller = 

revs, of B X B X "W 


T , ,, .,. 270x41x 40 ,«n^„^. 

In a slubber this = =160 revs. 

24 xll5 

Speed of spindles = 

revs, of H X H X L 

T 1 1,1 ii • 270 X 56 X 50 r^n ^„„„ 

In a slubber tlus = = 500 revs. 






A Draft wheel . . Change 


B Twist wheel . 

C Scar or ratchet wheel ,, 

D Lifter wheel . . ,, 

E Jack wheel . . ,, 

F Bottom cone end wheel ,, 

G Driving wheel for bobbins . 





H ,, ,, spindles . 





I Back roller wheel . 





J Sun wheel .... 





K Outside spindle wheel . 





L Skew gear wheel for spindles 





M Spindle bevel wheel 





X Outside bobbin wheel . 





Driving bevel for differential 

motion .... 





P Skew gear wheel for bobbins . 





Q Bobbin bevel wheel 





R Jack shaft wheel . 





S Lifter bevel wheel on Jack 

shaft ..... 





T Upright bevel on lifter shaft 





LT Strike or lifter bevel wheels . 





\ Top cone drum wheel . 





"W ,, end wheel 





X Large front roller wheel 





Y Small front roller wheel 





Z Top carrier wheel . 





a Strike shaft pinion for lifter . 





{■Compound carrier for lifter . 





d Lifter wheel .... 

: 100 




e Bobbin rail rack wheel . 




i " 


(3) Length delivered from the front roller eqtials- 


revs, of B X B X ^V X dia. of F.R. x 3-U1 6 

V X X 

T 1 ui 4.1,- 270x41x 40xli"x22 c^,.^« • , 

In a slubber tnis= 5 ^ = 565'/ inches. 

24x115 X 7 — 

Or (4) Revs, of front roller x dia. of F.R. x 3-1416 

= 160 X l|x3-1416 = .')65-7 inches. 

(5) Turns of spindle to one of front roller can be found 
by dividing the speed of spindles by the speed of front 
roller, which equals — ■ 

Or (6)— 

•1^ = 3-12. 

Xx VxHxL 

W x B X K x :S1' 


(7) Twists per inch = 

In slubber = 

40 X 41 X 58 x 26 

= 3-12. 

W X B x K X M x dia. of F.R. x 3-1416 

40 X 41 X 58 X 26 X 1| x 3-1416 
= -88 twists per inch. 

. . speed of spindles 

^ ■' '■ length delivered by front roller' 

In slubber= . „ = -88 twists per inch. 

(9) Twist wheel = 

X X Y X H x L 

W X twists per inch x K x M x dia.of F.R. x 3-1416" 

115 X 24x56 X 
'^''"= 40X-88X58X 

(10) Constant number/ _ X x V x H x L 

, , , , 115 X 24x56x50 

^° ^lubber= ^^ ^ .gg ^ -8 X 26 X 1| X 3-1416 = ^^' 

for twist t W X K X M x dia. of F.R.x 3-1416 


In slubber = 

40 x 58 X 26 xl|x 3-1416 

= 36-08 constant number. 

.,,. rr. • i 1 constant number 

(11) Iwist per inch = -. ; — ; — 

'^ twist wheel 

= — 7^ = 88 twists. 



constant number 

(12) Twist wheel = 

twist iier inch 

36-08 ,, . ,, 

^-•88 = ^' ^^^*^^- 

^ , , ,, dia. ofF.R. X IxZ 

(13) Total draft =T^ TW^ S — v 

^ ' dia. ot B.R. X Ax Y 

_ Hx60x90 ^^ 
1| X 57 X 18 

(11) Change wheel f _ dia. of F.R. x I x Z 

for draft \ dia. of B.R. x required draft x Y 
_ l|x 60x90 
~l|x 5-25x18 
= 57 teeth in wheel A. 

(15) Constant num- \ ^ cjia. of F.R. xl xZ 

ber for draft J dia. of B.R. x Y 
_ Hx 60x90 
~1^ xl8 

= 300 constant number. 

, . constant number 

(16) Draft wheel A = ~^—j7^ 

= ^ r = 57 teeth. 


,, ^, T-v c-L. constant number 

(!') Draft = — - — r-p 

dratt wheel A 

300 ,„, 
= vi^ = 2o. 


(18) Hank Roving". — To find the hank roving it is 
necessary to take a certain number of yards and weigh 
them very carefully. "When this is done reference can be 
made to specially tabulated results, on which the hank is 
denoted at once. If such a table is not available, the hank 
can be fonnd by the use of jDroportion. "When it is known 
that 1 lb. (7000 grains) of roving equals 840 yards, and 
that this is the standard for one hank, it becomes an easy 
matter to calculate the hank, granted the weight of a 
certain length is known. For instance, suppose ten yards 
of roving from a slubber weighs 166'5 grains, then since 

Ill FLY-FRAMES zi-] 

7000 grainsH-8-10 yards =1 hank, therefore 166"5 grains 
-7-10 yards = "5 hank. 

A convenient form of presenting this is as follows : — 

,,_, TT 1 ■ 7000 X number of yards taken 

(19) Hank rovinfc= r-j — . -. 

° 840 X weight in grains 

_ 8"33 X number of yards taken 
weight in grains 
8-33 X 10 ^ , , . 

It frequently happens that changes are made in the 
gearing under circumstances which enable us to dispense 
Avith a long calculation and resort to that of simple propor- 
tion. The draft wheel is often found in this way, and from 
the fact that the difference in the hank of rovings depends 
upon their weight, and consequently upon the area of their 
cross-section, we can by noting this obtain several other im- 
portant change wheels by the following simple methods : — 

(20) Draft wheel when changing the hank : 

Present hank x present change wheel 
Required hank 

^5;^= 38 teeth. 

The rack wheel, etc., are readily found— by remembering 
what has just been said about the area of the cross-section 
of the roving. As this area depends upon the square of 
the diameter, the rules which follow will be easily under- 

,„, , ,, , , 1 ./Present rack wheel^ X required hank 

(21) Rackwheel= V =- --;j -* 

Present hank. 

^ . , , , _ ^ /Present twist wheel'' x iiresent liank 

Twist wheel B = V -, — — — L 

Required hank 

T .^^ , , TA ^/Present lifter wheeP X present hank 

Lifter wheel D = V ^ — 5-,~- , 

Required hank 

,,,, r,^ 1 1 /> ./Present .stiir wiieel- X required hank 

(24) Star wheel C= V ~ , ^ 

Present uauk 



Production of Fly-frame 

The production of these machines depends primarily 
upon the speed of the front roller and the length of time 
they work per day or week. A knowledge of these two 
factors will enable us to obtain very exactly the production 
of any machine. It is, however, sometimes necessary to 
be able to draw up a table of productions, and it is also in- 
teresting to do so, as showing what the machine is capable of 
doing under normal conditions. To do this we must know 
several things : for instance, twist per inch ; hank roving ; 
weight of bobbin ; revolutions of spindle per minute ; and 
the time lost per set in doflSng and piecing. It will be 
noticed that the speed of the front roller is not required, 
and it will also be noticed that all the other factors are of 
a more or less variable character. Such factors are 
tabulated in the following list for the different machines, 
and also for various classes of cotton :— 




















































.2 ^- 











3 p. 










1 (S 






















a ~ 














o ^^ 


























O m 




















SC£ J 













^ n 


































S c 






T— 1 







»— t 


T— 1 





























■— ' 





































p , 











, — 1 




































^— ! 













































































• > 
















While recognising that in no two mills is to be found 
the same set of conditions, it is worth noting that those 
given in the table are the result of a thorough investiga- 
tion in a large number of mills spinning the classes of 

Fig. lOS. 

cotton named. A long time was spent in noting the time 
that was spent in doffing on the different machines, 
together with that lost through piecing, etc. In many 
cases a machine would be doffed very quickly through six 
or seven helpers assisting, while the same machine, at the 
next doffing, would only have two helpers ; this necessitated 

ni FLY-FRAiMES 231 

a large number of observations being taken, and an average 

Fifi. KKi. 

of them gave the results as given in the talilc. The 


multiplier for the twist per inch is also the result of a 
practical investigation in a large number of mills ; and, as 
will be seen, they follow very closely the necessary con- 
dition that less twist should be put in as the cotton gets 
finer in staple. 

We are now in a position to calculate the production. 
In the first place, the time occupied in building the bobbin 
must be found. This is obtained as follows : — Weight of 
bobbin in lbs. x hank roving X 840 X 36 = length in inches 
on bobbin. Length on bobbin X twist per inch = total 
twist on bobbin. 

total twists on bobbin . ,,-,,. , .,.,,. 

= minutes the bobbin takes in buildinfr. 

'revs, of spindle per min. 

These three rules can now be incorporated into one. 
For instance, time to build a set of bobbins equals 

840 X 36 X twists per inch x h ank x weight of bobbin in lbs. 
revs, of spindles. 

When we know the time for one set, we can then find 
the number of sets made in any given time (for instance 
in 10 hours, which is convenient as a base). 

-KT n , ■ -,,^ ^ 10 hrs. X 60 niins. 

jMo. or sets m 10 hrs. = — ■. — ■ , — m-; — t-t-. ■. :— 

minutes to build bobbin + tune tor 

doffing, etc. 

Lbs. per day of 10 hrs. =Xo. of sets in 10 hrs. x weight of laobbin. 

Hanks per day of 10 hrs. = lbs. per day x hank roving. 

Example. — Production of a slubbing frame working good 
Egyptian cotton, hank roving 1 -2. 

840x36x -766 X 1-2x26 ,,„ 

-rprr: — zrz =112'92 miiiutes per set. 

400 X 16 ■■ 

600 , ^ . , 

= 4'/2 sets 111 10 Lours. 


4-72x26 „ . 

-- _- — = / 6/ lbs. in 10 hour.s. 

7-67 x 1-2 = 9-2 hanks in 10 hours. 




The waste from fly-frames scarcely ever exceeds 2 per 
cent, the average heing represented as less than 1 per cent. 
The power required to drive the different machines will of 
course vary according to the size of the spindles ; the 
following may he taken as an approximation of the number 
of spindles per horse-power : — 

Slubbing frame of 8 in space 40 spindles per horse-power. 
Intermediate frame of G5 ,, 65 ,, ,, ,, 

Roving ,, {•'\ ., SO ,, ., ;, 

Jack „ 4i „ 100 

For the purpose of assisting those readers who desire 
to thoroughly understand the gearing of the fly-frame, 
three drawings are given in Figs. 108, 109, and 110. One 
(Fig. 108) represents the side view of the full gearing, and 
from it all necessary calculations can readily be made in 
the same manner as those given on p. 224. 

Fig. 109 represents another well-knoAvn maker's arrange- 
ment, and whilst giving an end view it is an eaiy matter 
to follow throuii'h the various trains of wheels. Each 


gearing point between the wheels has been marked with 
a dot, and names given to several features to assist in 
tracing the driving. 

Fig. 110 is given partly as a side-view of the gearing of 
the previous figure, but primarily as an example of a method 
of readily making sketches of gearing with the object of 
saving continual reference to the machines themselves. 
It might be termed a notebook sj'stem, and is far prefer* 
able for the purpose than the other sketches of gearing 
given in other parts of the book. 

Drawings that will be of interest to many readers are 
given in Figs. Ill and 112. They represent a species of 
standard bobbin for different lifts and the skewers adapted 
for them. Fig. Ill shows the dimensions suitable for 
short collars, whilst Fig. 1 12 is adapted for long collars. 

In order to convey some idea of what production can be 
obtained from the various fly-frames for dift'erent varieties 
of cotton, a table is appended. These productions are of 
course dependent upon the conditions given at the head 
of each set, and are based upon actual results gathered 
from a large number of mills working under the conditions 





SLUBBING FRAMES.— Indian and Low Amekican Cotton. 

Speed of Spindles = 550 revs. 

Dia. of Full Bobbin = 5J in. 
Weight of Full Bobbin = 30 o/. 


Dia. of Front Roller = lJ in. 
Lift = 10in. 

Time lost in Doffing, etc. = 14 min. 
per set. 

Turns of 

iKevs. of Spindle to 

one of 



^H*"'^ I Front 
^°^''ng- Roller. 









Hanks per 


per day of 

10 hours. 












Lbs. per 


per day of 

10 hours. 




week of 


age of 

Time lost 

in Doffing, 









SLUBBING FRAMES.— American and Low Egyptian Cotton. 

Speed of Spindles = 500 revs, per min. 
Dia. of Full Bobbin = 5i in. 
Weight of Full Bobbin = 2S oz. 

Dia. of Front Roller = li in. 

Lift=10 in. 

Time lost in DofRng,etc. = 14 nun. per set. 













9 054 




























SLUBBIKG FRAMES.— Good Egyptian and Sea Island Cotton. 

Speed of Spindles = 400 revs, per min. 
Dia. of Full Bobbin = 5i in. 
Weight of Full Bobbin = 2(i oz. 

Dia. of Front Roller = 12 in. 

Lift = 10 in. 

TimelostinDoffing,etc. =14min.perset. 











































































INTERMEDIATE FRAMES.— Ixdiax and Low American Cottox. 

Speed of Spindles = 700 revs, per 

Dia. of Full Bobbin = 4i in. 
Weight of Full Bobbin = 24 oz. 

Dia. of Front Roller = lJ in. 
Lift = 10 in. 

Time lost in Doffing, etc. = 14 min. 
per set. 




age of 

week of 

Time lost 


in Doffing, 













I Hank 
1 Roving, 


„ J o^"^"If °/ rr • I. Hanks per Lbs. per 

Revs, of Spindle to Twist ^^^^^^ gpindle 

per day of per day of 
10 hours. 10 hours. 



one of 













INTERMEDIATE FRAMES.— American and Low Egyptian 

Speed of Spindles = 080 revs, per 

Dia. of Full Bobbin =4* in. 
Weight of Full Bobbin = 22 oz. 

Dia. of Front Roller = IJ in. 
Lilt = 10in. 

Time lost in Doffing, etc. = 14 min. 
per set. 




1-27 ! 








1-376 ; 








1-467 I 








































1 -832 





INTERMEDIATE FRAMES.— Good Egyptian and Sea Island 

Speed of Spindles = 650 revs, per 

Dia. of Full Bobbin = 4J in. 
Weight of Full Bobbin = 20 oz. 

Dia. of Front Roller = 15 in. 
Lift = 9 in. 

Time lost in Doffing, etc. = 14 min. 
per set. 

































2 •.■30 



1 -232 








1 -293 















6 04 
















ROVING FRAMES.— Indiax and Low Ameuican Cotton. 

Speed of Spindles = 1100 revs, per 

Dia. of Full Bobbin = 3J in. 
Weight of Full Bobbin = ll oz. 

Dia. of Front Roller=l| in. 
Lift = r in. 

Time lost in Dofflng, etc. 
per set. 

13 min. 


Revs, of 

Turns of 

Sjiindle to 

one of 


Hanks per 


Lbs. per 


I)er day of 

10 hours. 


week of 

age of 
Time lost 





10 hours. 


in Doffing, 











7 "5 





























































5 '9% 










Speed of Spindles = 1050 revs 

Dia. of Full Bobbin = 3A iu. 
Weight of Full Bobbin^ lOJ oz. 

American and Low Egyptian Cotton. 

per Dia. of Front Roller = 1J in. 
Lifc = 7 in. 

Time lost in DoflBng, etc. = 13 min. 
per set. 

























































of Spindles = 900 


Dia. of Full Bobbin=3J in. 
Weight of Full Bobbin =10i oz 

Good Egyptian and Sea Island Cotton. 

revs, per I Dia. of Front Roller = ] J in. 

Lift = Sin. 
Time lost in DoflBng, etc. 
per set. 







































35 -93 




















CHAP. Ill 

Speed of Spindles = 1150 revs, per inin. 
Dia. of Full Bobbin = 2Jin. 
Weight of Full Bobbin = 9 oz. 

-American Cottox. 

I Dia. of Front Roller=l J in. 

I Lift = 7 in. 

I TimelostinDofRnf;,etc. = 13niin.perset. 


Revs, of 

Turns of 

Spindle to 
one of 


Hanks per 

Lbs. per 


week of 

age of 
Time lost 




10 bours. 

10 hours. 


in Doffing, 




























2-58 ■ 













































JACK FRAMES.— Egyptian Cot 


Speed of 


= 1120 revs 

. per min. | Dia. 

of Front Roller = lJ in 

Dia. of Full Bobbin = 2Jin. 

Lift = 

= 7 in. 

Weight of Full Bobbin = 8 o? 



min. per set 













































































6 -58 













—Sea Island Cotton. 

Speed of Spindles = 1050 revs 

. per min. | Dia. 

of Front Roller = 1^ in 

Dia. of Full Bobbin = 2i in. 

Lift = 

= 6 in. 

Weight of Full Bobbin = oz 

1 Time 


ng,etc. = 13 






5 -.^35 






















































































A GENERAL view of Dobson and Barlow's draw-frame is 
given in Fig. 113, which supjDlements those ah-eady given. 
An enlarged view of the upper part of the machine such as 
the rollers, stop motions, etc., will be found on p. 37, 
Fig. 22. 

The upper view of Piatt's draw-frame is sketched in 
Fig. 114, from which the stop motions can be readily 
understood. The arm which operates the front stop 
motion is carried in slides. The strap-fork is coupled to 
S. Full can stop motion for K consists of a plate in the 
space M Avhose upward movement puts a pin in the path 
of the arm and thus stops the machine. 

A more complete view of Ermen's Clearer is given in 
Fig. 115 as fitted to Dobson and Barlow's draw-frame. 

Drawing and Drafting of Cotton Fibres. — The 
drawing or drafting of the cotton fibres occupies an un- 
usually important position in the preparing and spinning 
processes. There are two distinct phases of this attenua- 
tion of the fibres. In the first place, a short, bulky length 
of fibres is divided up, and the fibres distributed so as to 
form a longer and thinner or less bulky mass. This kind 
of attenuation or draft is to be found in bale-breakers, 
hoppers, openers, scutchers, card, drawing rollei-s in \)vc- 

NoTE. — A very complete set of practical notes on these details will be 
founil in the author's book, Cotton Mill Manaijemoit. 

VOL, H 241 R 

/.A- ""^,,,5- 


I 52'^ /' 

Fir.. 113. 



paring and spinning machines, and in tlie excess movement 
of the carriage of the mule. All these examples are 
definite drawing methods, and are spoken of as draft ; the 
amount of this draft can be readily calculated. The 
dimensions of a bale of cotton compared with the dimensions 
of the laps of the opener produced from it, will give a 
ratio that represents the draft, just as the resulting sliver 
in the draw-frame compared with the number of ends put 
up will give us the draft. 

Fia. 114. 

It is as well to note that the drawing action in most of 
the examples above is quite independent of what are 
called drawing rollers, and, therefore, the particular length 
of the cotton fibre, so far as the draft is concei-ned, is not 
a factor of much importance. For instance, the 3000 to 
4000 of a draft between a bale and the lap (produced from 
it at the opener) has not necessitated any consideration 
being given to the length of the fibre so far as the draft 
itself is concerned. The same may almost be said of the 
draft of the card. When the length of the fibre is taken 
into account in either opening or carding, it is done for 
other purposes than drafting. Draft therefore in opening, 



cleaning, and carding is purely and essentially a lengthen- 
ing and a thinning process. 

When the fibres have been reduced to a somewhat small 

Fio. 113 

and fairly well-ordered condition, as in a sliver and a 
roving, another method must be adopted to attenuate the 
fibres, and this brings us to the use of drawing rollers. 
These consist of pairs of rollers adjacent to each other, one 
pair of which runs at a higher surface speed than the pre- 


ceding pair. The cotton is passed forward by one pair of 
rollers, and the next pair gripping it carries it forward at 
a faster rate, and consequently draws the fibres apart 
longitudinally, and so lengthens the sliver or roving at 
the same time that it thins it or reduces its bulk. 

Confining our attention to the purely attenuating process, 
it will be seen that a new set of conditions arises, for now 
the filjres are not carried along by air currents, needle 
points, or by rollers set Avide apart, and between which 
there is the smallest travelling or carrying draft to keep 
the cotton slightly in tension. Presumably, Ave must have 
the various pairs of rollers carefully adjusted in their 
distance apart if w^e are to obtain a large draft between 
any given pair of rollers. This may be considered a funda- 
mental practice in cotton spinning, and the setting of 
rollers to suit the average length of the fibres of the cotton 
is a basis throughout our spinning mills. Noav, this 
universal practice and the tolerably fair results, in the 
majority of cases, that are attained with draAving rollers, 
as at present arranged, has led to the conclusion that the 
adjustment or setting of the rollers and their weighting is 
the real basis of good drawing and its accompanying effect 
of parallelising the fibres. 

In spite of this, no student of cotton spinning ought to 
accept it as an axiom ; in fact, no process or action in 
cotton spinning can be considered as other than tentatiA'e 
steps in the making of yarns. " Cotton Spinning " is not 
merely accepting present-day facts of our cotton mills, and 
the resiilts of their mechanical operations, nor does it lie 
in remedying mechanical defects or delicate adjustments of 
parts ; good mechanics can do this, but it lies in the reason- 
ing faculties that can accurately deduce causes of failure 
or the correct procedure to adopt under given conditions. 


Cotton spinning 

Fif). 116. 

Before we can commence considering the drawing action of 
rollers, it is necessary to have a clear conception of the 
fibres that are to pass through them. Judgment of fibres 
is usually based on hand-pulling, which gives a nice set 
of fibres of fairly regular length and 
quite straight, as in Fig. 116. This 
suggests the length of the fibres, and 
also, incidentally, their strength, if 
one tries to break them. It is certain 
/ j M \\ I that no practical man ever uses this 
hand-pulling test as a real test of the 
average length, but it supplies a basis, 
and in the mill, adjustments are 
made that allow for a shorter length 
than hand-pulling would indicate. 

If a small pinch of any cotton is 
taken and the fibres measured and 
arranged, the result would be somewhat as in Fig. 117, 
wherein all the fibres are shown and not simply all the 
long fibres as in the hand-pulling test. 

Now, cotton spinning has to do with all the fibres in a 
bale of cotton, and not merely the long fibres or even the 
average length of fibres. Wastes and short fibres may be 
taken out, but they leave the bulk of the fibres very little 
better than before, and even the product of the comber has 
its wide range of lengths of fibres from the shortest to the 
longest. Going a step further, it must be remembered that 
the cotton, up to the time it is in the sliver form, has been 
practically in a free condition, and, as in the card web, the 
fibres take up a natural position and are twisted and bent 
in a variety of forms ; straight fibres are unusual in good 
cotton. In the drawing process we have to deal with 
these indiscriminately varied forms. Fig. 118 represents a 



few of the naturally -shaped fibres. All of them are of 
exactly the same length, but all of the fibres in cotton, both 
long and short filjrcs, are of these varied forms, and they 
exist in these forms in the card sliver throughout its 
length. It has already been pointed out how difficult it 
is to even approximate to an average length on the sup- 
position that all fibres lie straight in the bulk, but when 


their actual disposition is realised, as depicted in Fig. 118, 
or even better when examined in the web of the card, the 
difficidty of estimating the average length is practically 
impossible. Suppose the six fibres in Fig. 118 were passed 
through two pairs of drawing rollers. We know definitely 
the length, and that all six are equal in length, a condition 
unusually favourable to ready decision in setting the rollers 
according to the prevailing practice. AVe know, of course, 
that the setting would be made so that the distance apart 
of the centres of the pairs of rollers Avould be slightly 
over the length of, saj', No. 1 fibre. AVe also know that a 
mai'ketable product Avould 1)6 the result of such a setting of 
the rollers. The facts in the case, however, are no reason 



for not clearly understanding the process, and, as a pre- 
liminary, the student, as well as the practical man, must 
clearly realise the actual condition of the cotton that is 
submitted to the drawing action between rollers. 

Fio. lis. 

Present Arrangement of Drawing Rollers. — We 

will first consider the matter from the present-day arrange- 
ment of drawing rollers. These are of two kinds, viz. four 
pairs of rollers and three pairs of rollers, as shown in Figs, 



119 and 120, the usual arrangement for average American 
cotton. Modifications, of course, are to be found, but they 
will not affect what follows : — 

Fig. 110. 

fig. 119 arrangement is used for the drawing of cottons 
in the sliver form, whilst Fig. 120 represents the di-awing 
rollers for rovings whether on the flyer frames or the 

Fio. 120. 

spinning machines. A draft or drawing action takes place 
between each pair of rollers so that the cotton emerges 
much finer than when it was introduced. This draft is not 
distributed equally between the pairs of rollers ; for in- 
stance, in Fig. 119 the draft between C and D would be 


about 1 "4, between B and C and between A and B about 
2*83, so that the total draft would be about 8. The same 
occurs in the three lines of rollers, for between B and C in 
Fig. 120 there would be a small draft, and the bulk of the 
draft put between A and B ; in other words, the main part 
of the draft or drawing action occurs between the pairs of 
rollers that are nearest together, and whose centres apart 
differ very little from what is supposed to be the average 
length of the fibre. The rest of the drawing action occurs 
between pairs of rollers whose centres are set apart a 
greater distance than the length of the staple. 

Drawing" Short Staple. — The action that takes place 
between two rollers whose centres are wider apart than the 
staple of the fibre, is somewhat as follows : — 

Let C and D in Fig. 121 represent the centre lines of 
two rollers ; these will be spoken of as the gripping points 
(small circles separate so rapidly that the grip of two small 
circles may be considered to act only at the line joining 
their centres). Several full-length fibres are shown in 
dark lines, and these are surrounded by the usual mass of 
fibres of the sliver or roving. All the fibres between A 
and C will move forward at the speed of the rollers at C, 
and they will continue at this speed as individual fibres 
until they are free from the grip of the rollers at C. The 
mere fact of the rollers being able to draw the sliver 
forward from the cans indicates that the fibres are all 
entangled, and the accumulated pull of the fibres causes the 
sliver to move bodily forward without being pulled asunder 
or even drawn ever so slightly. (Interesting and very 
valuable tests may be made on the strength of slivers by 
allowing a sliver to hang vertically, and paying it out until 
breakage occurs ; the weight of the broken -off piece of 
sliver represents the breaking Aveight. Their value would 



lie in tlie direction of indicating variation in carding, 
mixing, and character of the cotton. In draw-frame and 
comber slivers much valuable infor- 
mation could be obtained as to the 
effectiveness of the operations and 
extraction of waste.) 

When the fibres emerge from the 
grip at C, and even while passing 
through the rollers, they come under 
the influence of the higher surface 
speed of the rollers at D, and those 
fibres already within the grip of D 
will be pulled forward and slide over 
those that are still under the influence 
of C. This sliding action will tend to 
straighten both fibres and so cause the 
fibres to become more parallel to each 
other. At the same time, a large 
number of the fibres that are free 
from the grip of both C and D are 
carried forward just as they are, and 
in many cases are not only not 
straightened out, but, on the con- p^^. ^^i. 

trary, are still further bent or dis- 
torted. According to the degree of entanglement, we find 
fibres that remain under the influence of the grip at C 
until they pass to the grip at D, and many others that 
come under the direct action of the grip at D immediately 
they leave the grip at C. These several states of varied 
movement of the fibres leads to irregiilarities in the 
reduced but longer sliver delivered by the rollers at D. 
These irregularities will not be too pronounced as long 
as the distance between C and D is not too much in 


excess of the length of the fibres. The expression " too 
pronounced " is used ad^dsedly, and applies only to the 
testing methods used in our mills to detect A'ariations. 
A series of successive yards of sliver from a card or 
first head of drawing will always exhibit a fairly wide 
range of variation, but if each yard were cut up into 
inch lengths, the variations in weight of each inch would 
be surprisingly great. 

Wide spaces between successive lines of rollers are 
responsible for the perpetuation and increase of these 
irregularities, and they are found in the draw-frame and 
the spinning machine rollers. 

Spacing" of Rollers. — In the openers and scutchers 
there is a long space between the cages or cage rollers and 
the calender rollers. A draft exists at this point, but it is 
merely " a carrying draft," yet even this small draft pro- 
duces a tearing apart effect which is noticeable, if carefully 
observed, on some machines. In any two pairs of rollers 
set at a distance apart in excess of the length of staples, 
there will always be this tearing action as distinguished 
from a true drawing effect. In the very early stages of 
cotton spinning, it Avas recognised that the further apart 
rollers were spaced the less draft could be used, and Ave 
have to-day a kind of compromise, based on that experience, 
that fixes the space between the back and middle roller at 
such a distance in excess of the length of the staple that 
the draft between them has to be the lowest possible con- 
sistent Avith this setting. These two factors are of such a 
fixed nature that they haA^e assumed the character of mere 
structui'al details outside the consideration of the spinner 
and concerning only the manufacturer of the machines. 
So much so is this the case that 90 per cent of spinners 
could not tell one what the draft is betAveen the back and 


middle rollers of his fly fnimes or spinning machines, nor 
the distance betAveen the centres of these two sets of rollers. 
Something that is bred in the bone, that has almost become 
instinct, tells our spinners that if this spacing is exceeded, 
even Avith the low draft, or if the distance is kept and the 
draft increased, the result will be chaos. 

Roller Settings. — The matter does not end here. 
The experience that gave us our back and middle roller 
settings also decided, in a way, the front and middle roller 
settings. These old spinners found that by reducing the 
space between the rollers one could increase the draft 
without the irregularities being too noticeable, so they 
closed up the front and second pairs until the distance of 
their centres apart was almost equal to the presumed 
length of the staple, and this has remained the practice 
until the present time in our card-room machinery. A 
limiting factor, however, entered into the question, for it 
was found that care had to be exercised as to the amount 
of draft between the two rollers, and as the draft between 
the back and middle rollers must of necessity be small, 
owing to the wide spacing between their centres, so the 
draft also must be limited between the front and middle 
rollers so long as their distance apart is in excess of the 
length of the staple. 

This limitation of draft accounts for the small drafts 
used in the card-room, and the necessity that follows of 
repeating operations and multiplying machinery to attain 
some desired result. Immediately the rovings are passed 
on to tlie spinning-room, a new aspect of the drawing 
action opens out. We still note that the back and middle 
rollers are set far apart, but experience forbids, even here, 
anything more than a low draft between them, but between 
the front and middle rollers there is scarcely a limit to the 


draft that is put in, and this becomes possible simply 
because the front and middle rollers are set Avithin the 
length of the fibres, i.e. the distance between the centres 
of the front and middle rollers is a little less than the 
presumed length of the fibres composing the roving. 

Before asking ourselves why this large increase is 
possible, we must recapitulate a little. In the first place, 
it must be noted that a judgment of the length of the 
fibi-es is based on the fibres drawn straight by the hand- 
pulling test, and on this test is based the settings of rollers. 
In experienced hands, this test for length allows for the 
shorter fibres, and so judges an average length, Avhich is 
somewhat shorter. It has, however, been pointed out that 
there are comj)aratively few straight fibres in raw cotton, 
and that the great bulk of the fibres are bent and curved 
in all directions. Since this is the condition as they lie in 
the sliver, and to a less extent in a roving, it naturally 
follows that the averaoje length of fibres is much shorter 
than is generally assumed as a basis for setting the rollers. 
After passing the first head of drawing, some of the fibres 
are a bit more straightened, and this effect is increased at 
each succeeding drafting process, but a proportion of these 
bent and curved fibres, of all lengths, remain in the roving 
and are spun into yarn. Excessive draft in the draw and 
fly frames would spew these fibres out in a very prominent 
manner from the front rollers, and any indication of this 
spewing is always a clear sign of too much draft in the 
rollers, or, what amounts to the same thing, that the rollers 
are set too far apart, and so allow too many of the bent 
and curved fibres to lie free between the grips of the 

If the fibres of a comber lap are carefully measured and 
compared with the fibres of the finished comber sliver, the 


difference between them is only comparatively small in 
spite of the enormous difference in appearance and feel. On 
measuring the Avaste that has been taken out, it will be 
found that this also presents no very great difference from 
the lap or sliver. What has really happened is that the 
comber has extracted a certain percentage of the curled, 
bent, and curved fibres which form the roughening element 
in the cotton, and consequently leaves the remaining fibres 
smooth and straight, hence the great change seen in the 
resulting sliver. Curved fibres still exist in the combed 
sliver, combed yarns will show any amount of them, short 
and long, and these, of course, are not straightened, rather 
the opposite, by the setting of the rollers in the comber 
draw-box, and in subsequent drafting processes, before 
reaching the mule. 

A further point to note is the use of some form of 
weighting ajiplied to the rollers to increase the grip. This 
weighting is so heavy, on account of the bulkiness of the 
material (slivers and rovings) in the card-room, that the 
draft and the parallelisation of the fibres must, of necessity, 
take place in the free space between the rollers. It is the 
sliding of the fibres over each other in the space between 
the grips that we rely upon for parallelisation or straighten- 
ing out of the fibres. Practically, no fibres are drawn out 
from between the grip of the bottom and top rollers in 
card-room machinery, and whilst this is effectual to some 
extent, as already explained, an opposite effect is always 
associated with it which brings in its train considerable 
irregularities whether of carded or combed material. 

It may be interesting to note now the action of the 
mule rollers. These are supposed to be set within the 
length of the fibres so that, presumably, the front and 
middle rollers both grip the same fibres. As a whole, 


however, tliis cannot be the case if the rollers are only set 
a short distance within the length, say -^^ or 1 of an inch. 
Owing to the great amount of drafting that has taken place 
in the card-room and multiplicity of machines the cotton 
has passed through, there are undoubtedly a certain pro- 
portion of fibres that have been straightened. As these 
are momentarily stretched between the two holding grips, 
any curved fibres in contact with them will be drawn 
somewhat straighter if they are in the grip of one of the 
rollers. This action, however, can only affect a compara- 
tively few fibres, so that most of the curved fibres are still 
free between the closely-set rollers, and consequently are 
taken forward in their curved condition and incorporated 
in the yarn. If the rovings are of combed cotton, the fact 
that a proportion of the curved and irregularly shaped 
fibres have been removed reduces the possibility of this 
hapjDening to the same extent, so that more regular yarn is 
made. Theoretically, there is no free space between the 
front and middle rollers of a mule, but, actually, the 
presence of unstraightened fibres always constitutes a free 
space. The more there are of these unstraightened fibres 
the less is the di'aft that can be put in any machine, and 
the less of curved fibres that exist in the cotton, so the 
draft can be correspondingly increased. The whole prob- 
lem of drafting between rollers depends, therefore, on the 
degree of straightness that we can produce among the 
fibres from the earliest stage onwards. 

Underlying this obvious statement is a mass of practical 
problems Avorthy of consideration and solution. At the 
moment, attention must be directed to the draft and 
setting of rollers, and accept the web of the card as it is 
given to us. Collect carefully a piece of this web and, 
realising that it will be passed through drawing rollers. 

SUrri.EMENTAR ] ' NOTES 257 

ask oneself wliat is the average length of the fibres. In 
Fig. 122 a small piece of such a web is given. A close 
examination will show that practically every fibre is equal 
in length, and this length is about equal to the distance 
between the lines A and B. A hand pull of this cotton 
would give us ideal conditions for judging length, but 
clearly this length cannot be used as a basis for setting our 
rollers. Take a fibre that lies parallel to A and B ; it is 
straight, of full length, but evidentl}^, for the purpose of 
setting rollers and drafting, it has no length at all, and in 

Fio. 122. 

the same connection all the other fibres vary consider- 
ably in length according to their position in the web or 
sliver. "With even the ideal conditions shown in Fig. 122, 
the length of the fibres for drafting purposes is something 
extremely difficult to estimate, but in any case such an 
estimate must, of necessity, be far shorter than the actual 
length of the fibres. If an actual case is taken, and we 
look at our web and see every gradation of length among 
the entangled fibres, a judgment of an average length for 
the purpose of drafting and straightening becomes a 
mathematical problem of some complexity. After some 
considerable amount of drafting, length becomes a more 


tangible factor, especially after combing, but this is due 
chiefly to the fact that we have reduced the quantity 
of fibres and kept the draft low at each stej) of the 

What has been said leads logically to the conclusion 
that the parallelisation or straightening of the fibres can be 
carried out more effectually by eliminating all possibility 
of any free space for any shape or length of fibre between 
the drafting rollers. This means, practically, that centres 
of rollers must be brought as close together as possible, so 
that one pair of rollers draws the fibres from between the 
nip or grip of the other pair of rollers, and straightens the 
fibres as they lie in the grip. The natural consequence of 
this condition implies that rollers must be small in diameter 
to enable them to be closed up, and also that little or no 
weight must be put on those rollers through which the 
cotton is drawn. Length of fibre becomes of small con- 
sequence under these conditions ; draft can be increased 
considerably, and greatly impi'oved straightening of the 
fibres effected. In the case especially of mule rollers, the 
middle and back top rollers must be as light and small as 

It will be an easy matter to deduce from what has been 
described in regard to present methods of drafting and 
setting of rollers, that they are not conducive to regularity 
of slivers or rovings. Granted that a perfect card sliver 
could be produced, i.e. every inch of it uniform in weight, 
if this sliver were put singly through the drawing frame, 
flyer frame, and spinning machine rollers it would emerge 
from each set of drawing rollers in a very irregular con- 
dition. The irregularity thus caused by the drawing rollers 
would not of necessity be cumulative, as some irregularities 
would be neutralised in passing through several drafting 


operations. The great factor which neutralises some of 
the irregularities, however, is the doubling process that 
occurs in all or most of the machines mentioned, and 
so produces a much less irregular roving from the front 
roller of the spinning machine than would be the case if a 
single sliver only were subject to the drafting in all the 
machines. Drawing rollers, under present conditions, are 
invariably a cause of great irregularities in every machine 
on which they are used. 

Draft in the Rollers of Draw- and Fly-Frames. — 
As a rule the drafts between the four pairs of rollers on a 
draw-frame are not based on any definite system. The 
totiil draft is known, and this usually takes the form of a 
whole number such as six or eight. This number is then 
divided up into a small draft between the 3rd and 2nd 
lines, and a much larger draft between the 2nd and 
front rollers. It is not often that the draft is altered 
between the back and 3rd or 3rd and 2nd line of rollers, 
they remain as set by the machine maker, and as small 
wheels are used on these lines of rollers, a change of a tooth 
would make a big change in the draft. Added to this, it 
happens that nothing can be seen as to what is occurring 
between these lines of rollers, so beyond calculating the 
total draft, and making any change by altering the speed 
of the back roller, very few people trouble themselves 
about the distribution of the drafts. The amount of draft, 
of course, ought to depend on the setting, and the setting 
naturally, in an orthodox mill, depends on the length of 
staple, so that it scarcely seems advisable to have a fixed 
rule for the drafts that is applicable to any cotton and to 
any head of drawing. This fact, however, does not prevent 
the following rule being followed in a number of well- 
managed mills : — - 


Draft between the 1st and 2nd rollers = square root of total draft. 
,j ,, 2nd „ 3rd ,, = cube root of total draft. 

,, ,, 3rd ,, 4th ,, = the rest or remainder. 

(1) Example — 

Total draft to be 6. 

Draft between the 1st and 2nd rollers = v^6 = 2*449 increase 36 % 

2nd „ 3rd ,, =V~=1"817 „ 35% 

» 3rd „ 4th „ =__A_^=i.36 „ 36% 
^^-'^^^ over sliver. 

2-449 X 1-817 X 1-36 = 6, total draft. 

(2) Example — 

Total draft to be 8. 

Draft between 1st and 2nd rollers = v 8 = 2 -828 increase 41 % 

„ 2nd ,, 3ra „ =V8 = 2 „ 43% 

„ 3rd „ 4th „ =^_^^ = l-il2 „ 40% 

^ 1,^0 X ^ over sliver. 

2-828 X 2 X 1-412 = 8, total draft. 

It will be noted in the first example that the draft between 
the 1st and 2nd rollers is 1-388 times more than between 
2nd and 3rd. The draft between 2nd and 3rd is 1-331 
times more than between 3rd and 4:th. In other words, 
the drafts are almost equally proportioned between the 
rollers. The case is even more clearly seen in the second 
example, for although there is a slightly larger percentage 
increase of draft between the 2nd and 3rd over that between 
3rd and 4th, than exists between the 1st and 2nd over the 
2nd and 3rd rollers, the proportion is fairly equal. 

The meaning of this will be seen at once if the drafts 
are plotted on squared paper as in Figs. 123 and 124. 
Commencing with one of a draft (which of course is no 
draft, but represents the condition of the cotton between 
the can and the back roller), the drafts are marked out on 
the 1st, 2nd, and 3rd vertical lines in each diagram, and 
the curve joining them is almost a straight line. In using 



numbers to express draft we obtain a fairly clear idea of 
what has happened to the cotton in its totality, but neither 
the figures nor the diagrams in Figs. 123 and 124 give us as 
clear an idea as is requisite, especially in the intermediate 
drafts ; we get a bulk idea instead of a detailed view of the 
drafting. Now drafting is simply the sliding of fibres over 

































• } 




' r 












1st. 2nd. 3rd. 
Fio. 123. 

1st. 2nd. 

Fio. 124. 


and among each other and their rearrangement in a longer 
and thinner condition. If we have " tv>^o " of a draft this 
means that, say, one foot of sliver has been lengthened out 
to two feet, and in this drafting some fibres have not moved 
whilst others have moved one foot from their original 
position. In moving this distance there has been a sliding 
effect among the fibres. It is this effect of drafting that 
must be clearly grasped, and perhaps one way of enabling 


this to be done is to plot the attenuation of the fibres 
resulting from a series of drafts. Let us first calculate the 
attenuation from the drafts obtained by rule in Examples 
1 and 2. For this purpose we will assume one inch fed 
into the back roller. 

(3) Total draft = 8. 

Length fed to back roller =1 inch. 

,, delivered by 3i-d roller=l X 1'4 . . . =1-4 inches. 

,, 2nd ,, =1-4x2 ^ . . =2-8 

,, ,, 1st ,, =2-8 X 2-82 . . =8 ,, 

(4) Total draft = 6. 

Length fed to back roller . . . . . . = 1 inch. 

,, delivered by 3rd roller=l X 1-36 . . . =r36 inches. 

2nd „ =1-36 X 1-817 . . =2-47 ,, 

„ 1st ,, =2-47x2-45 . . =6 

By plotting the lengths obtained in (3) and (4), which 
is done in Figs. 125 and 126, we obtain a very distinct 
impression of how the cotton has been drawn out by the 
drafting based on the rule given in (1) and (2). The 
original inch shown at in a thick line has been lengthened 
out by an amount A in the first draft, by a length B over 
the length of the 1st draft, and by a length C over the 
length of the 2nd draft. 

It scarcely needs to be pointed out that the lengths A, 
B, and C are the real factors to be decided upon in all 
drafting problems between successive lines of drafting 
rollers, and that these lengths ought to be decided upon 
before fixing the drafts. They "vWll naturally vary accord- 
ing to the condition of the fibres in the sliver or roving ; 
on the staple, and the setting of the rollers. It scarcely 
seems reasonable to suppose that there ought to be the 
same attenuation of the fibres in the intermediate stages at 
the last head of drawing as at the first head, and a mere 
glance at the diagrams in Figs. 125 and 126 will suggest 
that the first draft between the back and 3rd roller is 



too small, and tlie draft between the 2nd and front is too 

The rule of "square root, cube root, and the remainder" 
woidd ajipear to be one of those haphazard rules formed 
for the convenience of memory and to relieve the judgment; 





H 8 inch 

tion / 
es / 








tal / 
enuation / 
6 inches / 


















1st. 2nd. 3rd. 

Fio. 125. 

1st. 2nd. 3rd. 
Fig. 126. 

it has no practical basis, and is certainly not scientific in 
spite of square root and cube root being incorporated in it. 
A practical method of fixing the drafts would be to 
decide how much you will attenuate the fibres in each stage 
of the drawing rollers, and this must be a question of 
judgments based on experience and knowledge of the con- 
dition of the cotton and rollers. An example is given to 
illustrate this. 


Assume one inch of sliver from the can. "We ask our- 
selves how much can this be drawn out, and decide that 
the fibres can slide over each other to the extent of a 
maximum of \ of an inch. This Avill give us If to be 
drawn out at the next pair of rollers, the 3rd and 2nd pairs. 
Again use our judgment as to how much we can lengthen 
this 1|- inches. As it has been somewhat straightened, we 
may reasonably lengthen it to 3i inches. This 3| inches 
must now be lengthened finally, between the 2nd and front 
roller, to 8 inches, thus increasing its length by i\ inches. 
The above can be stated thus — 

(5) 1 inch in the first draft has been drawn out to 1| inches. 
1| inches ,, second ,, ,, ,, 3i ,, 

3i ,, „ third „ „ „ 8" „ 

or — 

In the first draft the fibres slide over each other | inch. 
,, second ,, ,, ,, ,, \\ inches. 

„ third ,, ,, ,, ,, 3i ,, 

With these lengths decided upon by judgment, it is an 
easy matter to calculate the drafts to obtain them. These 
drafts will be as follows — 

(6) Draft between back and 3rd roller = lg = l'625. 
3rd „ 2nd ,, ==2-153. 

2nd „ front ,, =: 2-285. 

It must be acknowledged that these drafts do not ajjpear 
satisfactory, but this is simply because our ideas of drafts 
are so hazy. If the figures in (5) are plotted on squared 
paper we obtain a picture of the lengtliening process between 
the rollers, and in Fig. 127 we note that a quite reasonable 
amount of drafting is indicated. Further, if the lengthen- 
ings at A, B, and C are represented as "drafts" as in (6), and 
these are plotted, we get the diagram as in Fig. 128, which 
is quite different from the curve in Fig. 124 which is based 



on the rule. To show this difference the curve in Fig. 127 
has been drawn on Fig. 1 24: in a dotted line at A. 

Almost all drafting between drawing rollers results in 
the production of irregularities. This is due to the setting, 
weighting, and drafts. A perfect card sliver, if such a thing 

1st. 2ncU 3rcL 
Fio. 12s. 

1st. 2nd. 3rd. 

Fia. 127. 

could be produced, would become imperfect in passing 
through a series of drawing rollers. This can readily be 
tested by passing a single card sliver through a head of 
drawing or successive heads. Doubling is the factor that 
disguises somewhat this action of the rollers, but even 
doubling fails to eliminate the irrei^ularities that are beins; 


constantly added as the cotton passes from one machine to 

A further factor that makes for irregularities is the 
driving of the top roller by frictional contact with the 
cotton passing between the pair of rollers. This has a 
disturbing effect on the fibres, and on some machines makes 
quite a difference to the draft, but even when this is not 
influenced, the disturbance and rearrangement of fibres 
passing under a weighted roller, and through which this 
weighted roller is driven, must be a serious cause of 
irregularities. Our methods of testing resvilts by long 
lengths have lowered the standard to which we ought to 
attain. So long as this lower standard exists most work 
done on cotton machines may be thouglit fairly good, but 
the student is asked to study carefully the inner actions 
of the machines, and to realise that much remains to be 
done and new ideas must be formulated that will inspire 
the responsible man to set up a high standard and adopt 
newer methods to attain it. 

Long Fibres in Comber Waste. — The presence of 
crossed and also curled fibres of good length in the comber 
lap almost of necessity leads to their elimination in comb- 
ing, and they go into the waste. Good straight fibres are 
by no means uncommon in the comber Avaste, and these are 
caused by faulty settings and timings of the various organs. 
Every action of the comber requires exactness in its 
mechanism, and the moment when the action begins and 
ends must be definitely timed. The student of the comber 
scarcely needs to be told that late nipping, opening nippers 
too soon, too early action of top comb, too premature or 
late detaching, etc., will result in long fibres being mixed 
up with the waste. The adjustments of rollers and nippers 
in regard to being parallel to each other, to the uniformity 


of the pressure of weights and springs, to the sufficiency of 
the pressure to ensure the right giip, are all necessary to 
prevent waste of long fibres. Faulty laps, as already ex- 
plained, Avill be a serious source of good fibres being carried 
away, and if the grip of the nippers is not uniform, or the 
holding surface or line of the nippers is irregular through 
wear or damage, it Avill be impossible to prevent long fibres 
getting into the waste. 

Combing Action and Number of Fibres in a 
Comber Lap. — If No. 60's is being spun from sakel cotton, 
there will be an average of sixty- five fibres in the cross- 
section of this yarn. Suppose a 410 grain per yard lap 
is used on the comber, the cross-section of this lap will 
contain about 192,000 fibres. The seventeen rows of 
needles on the cylinder of the comber will contain about 
11,000 needles, and these needles will pass through the 
cotton, say, four times, so that 44,000 needles will be used 
to comb 192,000 fibres. This gives, on an average, one 

needle to every ( 44 qqq ) four fibres. From this we de- 
duce that the comber does not comb every individual 
fibre but only small groups of fibres. Tliere may be a 
number of fibres individually combed, but if so there will 
be so many more in groups that are not combed. It is 
suggested to the advanced student, and to those occupying 
responsible positions in the mill, that it may prove bene- 
ficial to realise the number of fibres in slivers, rovings, and 
yams, and to make it a kind of basis on Avhich to reason 
out some of the problems that confront the practical man 
in his daily work. 

Comber Waste Collector. — A brief mention is made 
on p. 90 of a comber waste collector termed an aspirator. 
This system is coming into more general use, so that a 
drawing is now given showing its general features. Two 


factors are responsible for the use of the aspirator, viz. 
the large amount of dust resulting from the brush, and also 
the neppy character of the waste when a doffer comb is 
used, this neppy waste resulting in a reduced value when 
the waste is sold. In Fig. 129 a back view of the comber 
is shown with the apparatus in position, and a section of the 
machine giving an end view. A slowly revolving drum R, 
driven from the lap rollers at N, runs along the back of 
the comber. The drum is in communication with a pipe 
L at its middle point, and this pipe L is coupled to a 
fan F which exhausts the air from the revolving drum R 
The drum is fitted with perforated sections at each head, 
and these sections are cased in by sheet metal at the front, 
the casing extending so as to include the brush and back 
of the cylinder. Dampers are fitted within the drum or 
cage, so that as the waste is bruslied from the cylinder it 
is drawn at once on to the uncovered part of the perforated 
section. All dust and minute fibres are thus sucked 
through the cages and on to the fan; the longer fibres 
adhere to the drum and are carried round, being slightly 
consolidated by a roller E resting on the drum. The waste 
thus takes on the form of a sheet, and as it passes over the 
back of the drum and being free from air pressure, it falls 
naturally into the receptacle AV. 

A more detailed view of the section of the apparatus is 
given in Fig. 130, the reference letters being the same 
in each drawing. 

Preparation of Cotton for the Comber.— In con- 
sidering the question of preparing cotton for the combing 
process, our starting-point must be the card sliver. The 
condition of the card sliver is more or less an open book to 
us, for the card web presents a clear picture of the whole 
of its structure, and every individual feature of which this 

Fig. 130. 


whole is composed is, or ought to l^e, well known. Much 
of this knowledge is, of course, easily obtained by the 
unaided eye ; some require the aid of optical methods, and 
a not unimportant part is only definitely ascertained by a 
careful testing of weight for length. This latter test for 
irregularities is seldom carried out, except in a crude form, 
for the simple reason that general experience indicates the 
probability that all card slivers may be expected to show a 
considerable range of irregularity, and consequently that 
doubling is an absolute necessity in order to reduce the 
irregularities to a minimum. An important question now 
arises as to where this doubling process should take place. 
Before this can be answered, a further feature must be 

The needles of the comber pass through the sheet of 
cotton i)resented to them. These needles will straighten 
those fibres that are held by the nippers, and comb out or 
extract all the fibres that are not so held. These extracted 
fibres will naturally consist of short fibres, together with 
the longer fibres that happen to lie in the cotton in a bent 
or curved condition. Any needle coming into contact with 
small entangled groups of fibres, such as neps, will also 
carry them away. Irregularly disposed fibres therefore 
form a fair proportion of the Avaste taken out by a comber. 
From this it may be concluded that an important feature in 
the efficient working of a comber depends on the sheet of 
cotton being composed of a continuous series of parallel 
fibres. A second question now arises as to where these 
fibres can be made parallel. 

Irregularity of the thickness of the lap, or sheet of 
cotton across its width, is a further factor of importance. 
If the nippers press upon an irregular thickness of cotton, 
this pressure will be of a varying character, and the presence 


of thick and thin places across the lap Avill i-esult in the 
cotton not being held at all in some parts of the nipper, or 
only feebly held, This condition will naturally residt in 
not only excessive waste containing good long fibres, but 
also in the production of irregularities in the resulting 
comber sliver. A third question arises, therefore, as to 
how can the lap be made uniform in thickness across the 
width % 

The accompanying sketches (Fig. 131) will show in a 
somewhat emphasised form the varying character of comber 

No. 1 represents a uniform thickness of cotton lying 
between the nippers, and this may be assumed to be normal. 
If the lap is irregular lengthwise of the lap, but uniform 
across its width, the normal thickness will only occur 
occasionally, for the thickness between the nippers will 
vary, say, between the two extremes, as in Nos. 2 and 3. 

Any one acquainted with the working of the comber Avill 
recognise what this constantly varying thickness of the 
feed must mean in lowering the efficiency of the machine 
and the value of its products. 

The varj'ing irregularity of the disposition of the fibres 
projecting from the nippers may be roughly illustrated as 
in No. 4. 

All the loose fibres in the projecting sheet A are in 
a position to be taken out as the needles pass through. 
Many of these free fibres are taken out, but since there is 
an excess of fibres over the number of needles, there must, 
of necessity, be many free fibres that are untouched by 
the comber needles. 

It must be recognised that the projected sheet of fibres, 
held by the nippers and presented to the action of the 
cylinder needles, is in a very entangled and disordered 



condition, and thcat looped fibics, both ends of Avliich are 
held by the nippers, are not uncommon. Observation 
shows this to be the case, and it is confirmed by the broken 
and distorted needles which indicate that some unusual 







S N 



Fig. 131. 

effort has been necessary in passing through entangled 
fibres. Broken and bent needles are a sure indication of 
an inferior preparatory process, and it may be taken for 
granted that, where broken or bent needles are found, some 
needles have escaped damage by simply breaking the fibres. 


One very important effect has been produced by 
the passage of the needles, viz. the fibres have been 
straightened and laid in some kind cf parallel order, 
thus giving a uniform lustre and a smooth silky feel to 
the cotton. 

At the same time it must be emphasised that it is highly 
desirous the fibres should be presented to the comber in as 
parallel a condition as our present system is capable of 

Our next point to notice is the irregularity that may 
exist along the width of the lap. 

In No. 5 a rough sketch is given to illustrate an irregular 
thickness of lap between the nij^pers. 

Thick and thin places are shown at A and B respectively. 
The pressure between the nippers cannot be uniform along 
their full width if irregularities of this kind exist in the 
lap. The natural consequences would be the plucking out 
of fibres and groups of fibres from the parts B, a high 
probability of considerable broken fibres at the points A ; 
also bent and broken needles may be anticipated. 

Three important characteristics that should be possessed 
by the lap put up at the combers must be regularity in the 
length ; regularity in the width ; the fibres composing the 
lap to be in parallel order. At this point the reader may 
well ask — Why comb the cotton when it has attained the 
conditions just mentioned % The query may be dismissed 
by casually observing that a comber lap never possesses 
the characteristics enumerated ; they are purely ideal. In 
practice, however, the ideal ought to be kept in view, and 
the cotton from the card prepared in such a way as to 
approach the ideal as near as our present methods and 
machinery will permit. 

The machinery at our disposal in preparing comber 


laps consists of drawing frames, sliver lap machine (Derby 
doubler), and ribbon lap machines. 

The methods might be stated as follows : — 

(1) Card sliver to sliver lap machine and direct to 


(2) Card sliver to lap machine, then to ribbon lap 

machine, and on to comber. 

(3) Card sliver to draw-frame, then to sliver lap 

machine, then to ribbon lap machine, and on 
to comber. 

(4) Card sliver through two heads of draw-frames, 

then to sliver lap machine, then to ribbon lap 
machine, and on to comber. 

or put into a concise form : — 









Sliver lap 

Sliver lap 


2 heads Draw-frame 


Ribbon lap 

Sliver lap 

Sliver lap 


Eibbon lap 

Eibbon lap 



This list is sufficient for the mill man to recognise the 
amount of doubling and drawing a comber lap has under- 
gone, and his reason alone will enable him to work out the 
relative efficiencies of each of the four methods. Normal 
conditions, within a reasonable range of counts and a due 
sense of economy of' production, are important factors in 
deciding the system, but the making of the best yarn out 
of a given cotton enters largely into the question, and is 
sometimes the most important factor of all. 

It is not difficult, and possibly it has often been done, 
to test the merits of the four methods, but there is such a 



lack of definite information that one is led to give the 
figures of some tests that have been made. The cotton 
used was the same in all the tests, the same card sliver and 
the comber settings remained the same throughout. 

(1) Card 

Sliver lap machine 

(2) Card 

Sliver lap machine 
Ribbon lap machine 

(3) Card 
Sliver lap machine 
Eibbon lap machine 

(4) Card 

2 heads of Draw-frame 
Sliver lap machine 
Ribbon lap machine 

-18 per cent of waste at comber. 


per cent of Avaste at comber. 

10 per cent of waste at comber. 

9 per cent of waste at comber. 

The mere fact of taking out a certain percentage of 
waste is not a sufficient indication of good combing unless 
one has a clear knowledge of how the comber lap has been 
prepared. It would certainly be of interest to the practical 
man and the student to seriously reason out the problems 
that arise as to what has actually taken place among the 
fibres of cotton in the four experiments, and not dismiss 
the matter by merely suggesting or asserting that extra 
doubling and drawing has caused the difference between 
the four different methods. 



Factors governing- Weighting. — The weighting of 
rollers is extensively practised in textile machinery, and for 
various purposes. From this fact alone it might be con- 
sidered obvious that the pressure put upon rollers would 
be of unusual importance in effecting some given purpose 
The measure of the efhciency of the weighting would be 
indicated by the care exercised in adjusting the pressure 
to suit particular conditions of length of staple, drafting, 
spacing, thickness of cotton, hanks, speeds, etc., and even 
extends into the relative value of springs, dead weights, 
and lever weighting. All these factors are parts of the 
problems involved in a rational system of weighting roller^^ 
and their importance is fully recognised by the men in 
authority in our mills. 

No Standard. — Paradoxical as it may appear, efforts 
at utilising weights at their best advantage do not prove 
successful in the direction anticij^ated, with the consequence 
that for apparently identical purposes authorities will have 
heavy weights, others light weights, and others again any- 
thing between extremes, and the peculiar fact remains that 
each authority that has settled on these weights considers 
them the correct weights for his purpose. On the other 
hand, so many have experimented with the weighting and 
found such varying results, that they have simply left the 
matter in the hands of the machine-maker. It may almost 
be said that in the bulk of cases a mill accepts the machine- 
maker's decision as to weighting. If fairly normal results 
are obtained by this weighting, then the mill authority will 
go through life under the imjiression that those particular 
weights are the best, and they become a kind of standard 
for all future work. 



A Varying" Factor. — Interesting as the subject is, it 
is not our purpose here to work out the Aveighting of 
rollers for any particular set of conditions. Our object is 
rather to show how weighting is practically a constantly 
varying factor on any given roller, and that Avhen we are 











-X - - 


-y - 

; C 

— ->: 


Fig. 132. 

under the impression that, say, a 10-lb. pressure is on a 
roller, it is not 10 lbs., but a pressure that goes through a 
cycle of variations in pressure. These variations, it may 
be added, are responsible for a considerable amount of the 
irregularities in the products turned out by the machines. 




;c -!- 

^S.. ... -y 

i ^ 

Two Ways to exert Pressure.— There are two ways 
in which pressure may be applied to a roller, viz. by 
pressure being exercised on the centre of a roller, and by 
the pressure being applied at each end of the roller. 

Fig. 132 represents a diagram of pressure being applied 
to the centre of a roller, the top roller T being free in 



its end bearings E, and capable of rising and falling verti- 
cally to accommodate the thickness of the cotton C that 
may be going through. B is the bottom roller. 

In Fig. 133 the weighting is applied at the ends of the 
roller. The top roller T is free, as in Fig. 132. 





>t x-x^- 


Fig. 134, 

The pressure in each of these two cases may be applied 
by springs, by lever weighting, by dead weights, i.e. hang- 
ing weights at the points P, or by self-weighting, i.e. with 
no additional weight beyond the weight of the roller itself, 
in which case the weight or pressure will act at the middle 
point in the roller. 


;t-x-'^ y— >; 

Fio. 13"). 

Now suppose a pressure of, say, 20 lbs. is acting at 
P in Fig. 132, and the two strands of cotton CC are at 
equal distances on either side of the direction of the 
pressure P, then the pressure on each of the strands will 
Ije 10 lb?., and this will be the pressure whether the strands 
are as in Figs. 132, 135, or 136. 

If a traverse motion is used which does not maintain 



the strands at equal distances in respect to the direction of 
pressure, this equality of pressure vanishes. In Fig. 134 
the strands of cotton are shown at unequal distances from 
the line of pressure. The total pressure on the two strands 
C and D will, of course, be 20 lbs., but each strand will 



X y 


Fio. 136. 


be subject to a different pressure. 

be 13| lbs. and that on D will be 6| lbs. = 20 lbs 

The pressure on C will 
This, of 

course, is a wide difference in pressure, and it is naturally 
varying from moment to moment as the traverse guide 
moves the cotton to and fro. This variation of pressure is 
taking place on two strands of cotton that are presumably 





Fir.. 137. 

alike, and working under exactly the same conditions of 
draft and speed. In face of this it is surprising that such 
dogmatism still persists in cotton mill circles as to the 
efficiency of the certain weights or pressure to be applied 
to rollers. Here we have (and it is common throughout 
our mills) two rovings, one weighted twice that of the 
next ro^'ing, and yet no one would examine this particular 



feature in looking for a course of irregularity or other 
peculiarity of the resulting roving or yarn. 

A self-weighted roller, as in Fig. 137, works under the 
same varying conditions of varying pressures for the two 

strands of cotton, save that the weight is very small, and 
being small it can scarcely be considered a weight but 
merely a holder and guider, so that the pressure does not 
act or function in the same Avay as in rollers that are 
deliberately weighted. There are cases where the self- 



Fig. 130. 

weighted roller only has one end going through, as in 
Fig. 138. 

It will be seen that the only spot for the cotton to 
receive the full weight of the roller is when it passes 
directly under the middle. If the cotton passes to one side 
of the middle line, the top roller looses its balance and tips, 


so that the pressure is lessened the farther away the cotton 
moves from the centre. This can be seen in reference to 
Fig. 139. 

Referring to Fig. 136, a condition may arise similar to 



- - X -> 


that shown in Fig. 140. Here we have one end down and 
the other going through. This causes one end of the 
roller to rest on the bottom roller at M, whilst the other 
end rests on the cotton at N. It will readily be seen that 
the cotton will be subjected to a varying pressure as it 
traverses its length of the roller. 





So far, the thickness of the cotton is of little consequence 
on the pressure effect, but in rollers weighted at their ends 
the thickness becomes important. In Fig. 133 it will be 
seen that so long as the two strands of cotton are each at 
equal distances from the ends or from the middle, they will 
be subject to equal pressures, but if from any cause they 



are at unequal distances there may be a Mnde variation in 
pressure. Fig. 141 will illustrate this. 

Now if, instead of taking tAvo strands of cotton, we use 
a lap or a series of strands, we obtain a uniform pressure 
throughout the length if the cotton is uniform in thickness, 
but if the cotton is irregular in thickness, then the pressure 
will vary and very frequently there may be no pressure at 
all. This, of course, is easily seen, and the consequences 
in the production of irregularities are only too apparent in 
scutchers, cards, draw-frames, etc. Fig. 142 will indicate 
the condition. 

If the normal thickness were going through at A, and a 
thicker portion at B, the pressure would act at A and B, 
and leave the intervening space between A and B free from 
pressure, so that there would be no holding or gripping 
effect either to bring the cotton forward in the correct 
form, or to prevent the cotton already between the rollers 
from being drawn or plucked out bodily by such organs as 
beaters, takers-in, or other rollers. On the supposition 
that the two ends were weighted equally, a thick portion 
of cotton passing through the middle of the roller would 
balance the roller and act as a fulcrum to it, with the 
consequence that there would be no pressure on the cotton 
going through on either side of the centre line. This will 


be quite clear from Fig. 143, where an equal pressure is 
applied at P and P, and a thick piece of cotton or a thick 
portion of sliver is passing through at C, Avhich is the 
centre of pressure resultant of the two pressures of P 
and P. The cotton at C will naturally be subject to the 
full pressure or total pressure of the two levers or dead 
weights acting at P and P, and thus leave the cotton on 
either side of C practically free from pressure. 

The examples already given are based on the sujjposition 
that the axes of the top and bottom rollers are parallel to 



each other. It is possible that there are cases where the 
pivots or bearing of the top roller have become worn ; 
when this has occurred so that both ends have worn 
equally, the top roller will occupy a position forward of 
its correct position, but its axis Avill still be parallel to that 
of the bottom roller. This altered position will naturally 
result in the lengthening of the distance between the grip 
of two pairs of successive rollers. This can easily be 
allowed for in the setting, and is scarcely a matter of 
importance unless in any given machine the two roller ends 
of one roller have worn more or less than the two roller 
ends of another roller. This would, of course, cause a 
difference of setting between the two rollers. Adjustable 


cap bars would enable even a case of this kind to be 
readily set correctly. If, however, the two ends of a top 
roller are worn unevenly, or the axis of the top roller is 
not set parallel to the bottom roller, the top roller will not 
bed evenly on the bottom roller. In such a case the grip 
of the top roller, Avhether self-Aveighted or weighted, will 
be of a very varying character and unreliable as a drafting 

A Neglected Subject. — All the diagrams used are 
purposely somewhat exaggerated, but they point out clearly 
a phase of cotton spinning that is too often neglected. 
The principles involved belong to a very elementary phase 
of mechanics, so that it has not been considered necessary 
to work out numerically the varying pressures. If our 
technical schools would fit up a few models of rollers, and 
by means of spring balances actually test the pressures on 
rovings and yarns at diflferent positions along the rollers, 
it would be a matter of intense interest to students, and 
probably of considerable value in suggesting solutions to 
some problems associated with irregularities that ave con- 
stantly cropping up in the mill. 

Fig. 144. 

RoviNc Frame 



Fig. 145. 


Action of the rollers in draw-l'raine, 

Adjustments for comber, 70, 87, 103 
top comber, 70 
draw-frame rollers, 8 
Alternate systems of draw-frames, 5 
Ajnericaii cotton, draft in fly-frames 
for, 121 
machines for spinning, 48 
Arrangement of fibres in card 
web, 1 
machines for single and double 

combiug, 58 
machines for spinning various 

counts of yarn, 48 
S( indies in tiy-franie, 126 
Aspirator for comber waste, 90 

Back-stop motion in draw-frame, 

25, 27 
Backward and forward motion in 

comber, 61, 71 
Balancing bottom rail of fly-frame, 

Bobbin, building up of, 195 
leading, explanation of, 147 
number of layers on, 197 
shape of, 195 
standard sizes, 234 
traverse motion for, 195 
Bobbins, diameters and lifts of, 
driving of, in fly-frame, 154 
lifts and diameters of, 122 
Bottom rollers of draw-frame, 6 
Building of the fly-frame bobbins, 
VOL. II 289 

Calculations for comber, 114 

draw-frame, 38 

fly-frames, 218 
Card, condition of fibres in web 

of, 1 
Case-hardened rollers, 13 
Cause of waste, 1 5 
Clearers, Coding's top, 33 

Ermen's top, 33 

roller top, 31 

stationary top flat, 31 

top roller in draw- frame, 30 
Coils and layers of yarn on the 

bobbin, 217 
Collars, long, 125, 132 

short, 126, 132 
Comber, adjustments for, 70, 87 
of top comb, etc., 70 

aspirator for waste, 90, 267 

backward and forward motion, 
61, 71 

calculations for, 114 

cushion plate, 66, 70 

cycle of operations, 75, 83 

cylinder, construction of, 85 

descrijition of, 58, 91, 106 

detaching roller motion, 71 

dott'er brush in, 89 

double-action, 65 

draft in, 114 

driving of, 59, 111 

duidex form of, 65 

full can stop motion, 106 

gearing of, 114 

leather detaching roller, 79 

Nasmith's, 91 
gauges for, 103 




Comber, nipper, 66, 67, 70, 110 

nips per minute, 85 

notch wheel arrangement, 76 

percentage of waste in, 116, 276 

piecing of sliver, 71 

power to diive, i'o, 105 

preparation of cotton for, ol, 269 

production of, 86 
Whitin, 110 
Nasmith's, 105 

quadrant motion, 71 
cam, 71, 73 

speeds of, 86, 105 

star wheel for feeding rollers, 64 

top comb, 70 

waste, 86, 116, 266 

Whitin, 106 

width of laps for, 85 
Combing, 47 

objects of, 47 
Cone drums, 156 

formation of, 157 

curves of, 157 
Construction of comber cylinder, 85 
Covering of rollers, 8 

leather of rollers, 8 
Creels, 123 
Curves on cone drums, 157 

hyperbolic, 154 
Cushion plates, 66, 70 
Cycle of operations in comber, 

75, 83 
Cylinder of comber, 85 

Deliveries, 4 
Derby doubler, 52 
Description of comber, 58 

fly-frames, 122 
Detaching roller motion, 71 
Diameters and lifts of bobbins for 

various cottons, 122 
Diameters and setting of rollers in 

drawing-frame, 13, 130 
Differential motion, 168, 184. 188, 

191, 193 
Defter brush in comber, 89 
Double-action comber, 65 
Doubling, principles of, 18 
Draft, 43 

equalising effect of, 19 

in comber, 116 

Draft in fly - frame for Indian, 

American, Egyptian, and Sea 

Island cotton, 121 
in the web of card, 1 
principles of, 15 
Draw and lap machine, 53 
Drawing-frame alternate system, 5 
back-stop motion, 25 
bottom rollers, 6 
calculations, 38 
case-hardened rollers, 13 
Colling's top clearer, 33 
deliveries, 4 
description of, 2 
diameters and setting of rollers 

in, 13, 130 
draft, 43, 241. 259 
electric stop motion, 28 
Ermen's top clearer, 33 
tlaunel covering of top rollers 

of, 8 
front stop motion, 25 
full can stop motion, 33 
hank slivers suitable for, 46 
heads, 4 

hunting in flutes of rollers, 6 
leather rollers, 8 
loose boss rollers of, 10 

bush rollers of, 12 
metallic rollers, 34 
number of flutes in bottom rollers 

of, 6 
power to drive, 43 
principles of clraft, 15, 241 
production of, 46 
rollers, 6, 249 
roller top clearer, 31 
set of deliveries, 4 
" single preventer " motion, 27 
speed of front roller, 46 
stationary top flat clearers, 31 
stop motions, 22. 35 
tandem system, 5 
top clearers, 30, 2 '4 
waste made in, 4 
weight relieving arrangements of 

rollers, 8 
weighting of top rollers, 7 
zigzag system, 5 
Drawing process, kepiote of, 2 
Driving of bobbins. 154 



Driving of comber, 59, 111 
Dujjlex combei", 65 

Egyptian cotton, draft in fly-frames 
for, 1-21 

niacliines for spinning, 48 
Electric stop motion of drawing- 
frame, 28 
Epicyclic gearing, 169 
Equalising effect ot doubling, 18 

draft, 19 
Ermeu's top clearer, 33 
Essential features of good yarn, 50 

Flannel covering of rollers, 8 
Flat top clearers, 31 
Fluted rollers, 6 

"hunting" in the flutes of, 6 
Fiy-frames, 118 

arrangement of spindles, 126 
balancing bottom rail, 214 
bobbin driving, 154 

leading, explanation of, 147 
building of the bobbin, 195 
calculations, 218 
cone drums, 156 
creels, 123 

curves of cone drums, 157 
description of, 122 
diameters and lifts of bobbins for 

various cottons, 122 
diameters and setting of rollers, 

differential motion, 168, 184, 188, 

191, 193 
draft for various cottons,121, 259 
driving of bobbins, 1 54 
epicyclic gearing, 169 
flyer leading, explanation of, 145 
flyer leg, 139 
flyer and presser, 135 
formation of cone drums, 157 
gauge of spindles, 127 
gearing of, 218 
general notes, 217 
hank roving, 226 
hyperbola, 154 
Jack-in-the-box, 183 
layers per inch lilt, 217 
lifts and diameters of bobbins 

for various cottons, 122 
long collars, 125, 132 

Fly-frames, object of, 118 
passages of, 119 
power to drive, 233 
presser of, 135, 136 
principles of cone drums, 157 

winding in, 141, 150 
production of, 228, 235 
ratchet wheel, 216 
reversing motion, 198 
roller stands, 127 

weights, 129 
shape of the bobbin, 195 
short collars, 126, 132 
space of spindles, 126 
speeds of front roller, 235 
spindle footstep bearings, 132 
standard bobbins and skewers, 

star wheel, 222 
strike wheel, 222 
sun and planet motion, 178 
table of drafts in, 121 

hank roving, speeds, and multi- 
pliers for twist, 229 

production of, 235 
theory of cone drums, 157 

winding, 150 
traverse motion for bobbin, 195 
twist per inch, 135 

in rovings, 120, 125, 131 
winding cone drum strap back, 

Front stop motion in drawing-frame 

25, 27 
Full can stop motion, 33 

Gauge of spindles, 126 
Gearing of comber, 114 

epicyclic, 169 

fly-frames, 218 
General notes on fly-frames, 217 

Hank slivers, suitable for drawing- 
frame, 46 

Heads of comber, 59 
drawing-frame, 4 

Horse-power required to drive the 
comber, 86 
draw and lap machines, 58 
draw-frame, 43 
fly-frame, 233 



Horse-power required to drive tlie 
ribbon-lap macliiue, 58 
sliver-lap machine, b'l 
Nasraith's comber, 105 
"Hunting" in the flutes of rollers, 

Hyperbola, 154 

Indian cotton, draft in flv- frames, 
machines for spinning, 48 
Intermediate frame, 119 
Irregularity of web, 1 

Jack-in-the-box, 183 
Jack frame, 119 

Laps, width of, for comber, 85 
Layers per inch lift on bobbins, 217 
Leather detaching roller in comber, 

rollers of drawing-frame, 8 
Leg of flyer, 139 
Lifts and diameters of bobbins lor 

fly-frames, 122 
List of machines used in cotton 

spinning for variijus counts, 48 
Long and short collais, 125, 132 
Loose boss top rollers, 10 
busli rollers, 12 

Machines for spinning American 
cotton, 48 
counts Nos. 3s to 10^ 48 
Egyptian cotton, 48 

(double carded), 48 
Indian cotton, 48 
Sea Islands cotton, 48 
list of, used in cotton spinning, 

for various counts, 48 
sliver-lap, 51 

power to drive, 52 
production of, 52 
speed of, 52 
Metallic rollers, 34 
Multipliers for twist per inch, 229 

Xasmith's comber, 91 
gauges for, 103 
power to drive, 105 
production, 105 
speeds, 105 

Xasmith's comber, weight of laps, 

Nipper in comber, 66, 67, 70, 110 
Nijis per minute Lu comber, 85 
Notch wheel arrangement in comber, 

Notes, general, on fly-frames, 217 
Passages of fly-frames, 119 
Percentage of waste iu comber, 116 

Piecing of the sliver in comber, 71 
Power to drive comber, 86 

draw and lap machines, 58 

drawing-frame, 43 

Nasmith's comber, 105 

sliver-lap machines, 52 
Preparation of cotton for combing, 

51, 269 
Presser of fly-frame, 135, 136 
Preventer motion, single, 27 
Principles of cone drums, 157 

douTfling, 18 

draft, 15 

the presser, 136 

winding in fly-frame. 141 
Production of comber, 86 

draw and lap machines, 58 

drawing-frame, 4 6 

fly-frames, 228, 233 

Nasmith's comber, 105 

sliver-lap machines, 52 

Whitin comber, 110 

Quadrant cam for comber, 71 
motion in comber, 71 

Rail, balancing of, in fly-frames 

Ratchet wheel in fly-frames, 216 
Relative motion, 195 
Reversing motion, 198 
Ribbon -lap machine and draw- 
frame combined, 53 

power to drive, 58 

production of, 58 

speeds of, 58 
Roller stands of fly-frame, 127 

top clearer, 31 
Rollers, 6 

case-hardened, 13 

diameters and setting of, in draw- 
ing-frame, 13 



Rollers of (Irawing-fraiues, 6 
Haiiiiel covering of, 8 
hunting in the tlutes of, 6 
leather covering of drawing frame. ^ 

detaching, 79 
loose boss, 10 

bush, 12 
metallic, 34 
setting of, ia drawing-frame, 13, 

vs'eightiug of, 7 
weight - relieving arrangements 

of, 8 
weights for, in flj'-frame, 129 
Roviug frames, 119 {see Fly-frames) 
Roviugs, twist in fly-frame, 120 

Sea Islands cotton, machines for 
spinning, 48 
draft in fly- frames for, 121 
Set of deliveries in drawing-frame, 

Setting and adjiL-tmeuts in comber. 

87, 103 
Setting of drawing-frame rollers, 13 

fly-franie rollers, 130 
Shape of tly-frame bobbin, 19.1 
"Single preventer" motion, 27 
Sliver-lap machine, 51 
Slabbing frame, 119 
Space of spindles, 126 
Speeders, 119 {see Fly-frames) 
Speeds of comber, 88 

draw and lap machines, 58 
front roller of drawing-frame, 46 

fly-frames, 235 
Nasraitli's comber, 105 
sliver- lap machines, 52 
Standard bobbins, 234 
Star wlieel, 222 

in comber for feeding, 64 
Stop motions in drawing - frame, 
22, 35 
electric, 28 
motion, full can, 33, 103 

Strike wheel, 222 

Sun and planet motion, 178 

Table of diameters and lifts of 
bobbins, 122 
drafts in fly-frames, 121 
hank roving, speeds, and ninlti- 

pliers for twist, 2^9 
production of fly-frames, 235 
sj>eeds, hank slivers, etc., for 
drawing;- frame, 46 
Tandem system of drawing-frame, 5 
Theory of cone drums, 157 

winding, 150 
Top comb, 70 

clearers in drawing-frame, 30 
stationary flat, 31 
Traverse motion for bobbin, 195 
Twist per inch in roving, 135 
in rovings for fly-frame, 120, 
125, 131 

Waste, cause of, 15 

in comber, 86, 116, 263 

made in drawing-frame, 4 
Web, arrangement of fibres in, 1 

draft in the, 1 

from the card, 1 

irregularity of, 1 
Weight of laps for Nasmith's 
comber, 105 

relieving arrangements of rollers, 
Weighting of top rollers of drawing- 
frame, 7, 277 
Weights on rollers in fly-frame, 

129, 277 
Whitiu comber, 106 
Width of laps for comber, 85 
Winding cone drum strap back, 215 

principles of, 141, IT.O 

theory of, 150 

Yaun, essential features of good, LO 
Zig::ag system of drawing-frame, 5 

rTinlcd in Great Britain hy R. & R. Ci.ahk, I.iMn ed, Eiiinhi7^/t. 


(1920), LTD. 



Makers of 





Royal Exchange, Manchester, Pillar H.2. 
Telegrams : " Globe, Castleton - Lanes.' 
Telephones: 5871, Castleton-Rochdale. 



Wilson Brothers Bobbin Co. Ltd. 


Telegrams: "Nugget, Liverpool" /^ ar-c^-rkt-i 

Telephona : 900 Garston Vjrdl S lUll , 

Offices and Showrooms: 

MANCHESTER ROYAL EXCHANGE, 324 & 325 (Tower Entrance). Pillar G2 
Also Bradford E.x change, Monday and Thursday 




Manchester Office : King's House, King Street West. 

Royal Exchange : Tuesdays and Fridays, Pillar L6. 

London Office : 71 Lincoln's Inn Fields, Twyford Place, Kingsway, W.C.2. 


you will obtain SATISFACTION 

from the 









Codes : 
A1, A. B.C. 4th, 5th and 6th Editions, Western Union, Bentley's. 

Telephones : 

No. 1826 OLDHAM. No. 5344 CITY. No. 2242 HOLBORN. 

Telegrams : 



Consulting Engineer. 


Greenmount Lane, 
tele. : 2313 bolton. 







Codes used: 

"Al." "ABC," 
Telegraphic Address: 4th and 5th Editions, Telephone: 

"ASA, OLDHAM." ''^"^°*'°'',', No. 777 OLDHAM. 









Cotton. Wool. Worsted, etc. 


Telegrams : 
'Globe, Accrington" 

Telephone : 
No. 2121 (4 lines) 








Revolving Flat Carding Engine. 

Roving Frame. 



1- ^r 



1 )M) 








u * 



^^ +J 










• • 


e • 

















_l ^ 






































Driving of 



Frames a 





Wm. Kenyon & Sons, Ltd. 

DUKINFIELD, Cheshire, England. 

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

Cablerie du Nerd, Armentieres, France. 95-97 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. 



gives the world's Textile News in its most complete, 
comprehensive, and concise form. Acknowledged 
as the most progressive Textile Journal in the World, 
its contents include invaluable technical articles by 
the leading authorities on Cotton. Wool, Silk, Jute, 
Flax, Hemp, Ramie, Hosiery, and Lace Manufacture ; 
Bleaching, Dyeing, Printing, Proofing, and Finishing. 

Published the 1 5th of each month. 

Post Free 24/- per annum. 



gives each month technical and trade information 
unobtainable in any other publication. It circulates 
in every country in the world using or producing 
Silk or Artificial Silk, and is indispensable to all con- 
nected with these industries. Permanent features 
include special illustrated articles on Raw Materials, 
Yarns, etc. Exclusive Market Reports, Throwing, 
Spinning, Doubling, Winding, Weaving, Knitting, etc. 
Special articles on the manufacture of Artificial 
Silk and Knitted Goods, Dyeing, Printing, Finishing, 
Engineering, etc. 

Published 20th of each month. 

Post Free 9;'- per annum. 


is the only Journal published in the United Kingdom 
exclusively devoted to the Making-up Industry and 
to the buying of Fabrics for the wholesale and 
shipping. It covers the whole field of " READY- 
MADES." It is invaluable to Dyers, Printers, 
Finishers, Proofers, Embroiderers, Embossers, 
Pleaters, etc. ; Makers and Merchants of all classes 
of Machinery used by the Making-up Industry, 
Packing Warehouses, etc. ; Manufacturers of Woven 
Names, Tabs, Labels, etc. 

1/- monthly. 10/- per annum. 
Send for Specimen Copies 



Central 7400 (7 lines) 
Also at London and Blackburn 

Improved Ring 
Spinning Machine 
incorporating Special 
Motor Driving. 
4 lines of Rollers 
High Draft System 
and Tape Driving 
Arrangement for the 

Textile Machinery 

on the most modern 
Principles of Design 

Comprising — 



Complete Plant, Equipment, and 
Accessories for the Manufacture of 
RAYON by the Viscose Process. 


Established 1790 

Telephone: No. 60\. 


Telegrams : Dobyons, Bolton. 



In Three Volumes. Crown 8vo, With llki.stratious. 

Voh I. Including all Processes up to the end of Carding. 
Xinth Edition. 10s. net. 

Vol. II. Including the Processes up to the end of Fly 
Frames. Sixth Edition, with Appendix. 8s. 6d. 

Vol. III. Including Theory of Spinning — Mechanism and 
^\'orking of the Mule — The Ring Spinning Frame 
— Winding Frames — Doubling — Yarn Preparing 
— Mill Planning — Humidity — Useful Informa- 
tion — Appendix — Index. Fifth Edition. 
10s. net. 


Third Editiou. Super Royal 8\o. 7s. 6d. net. 

A Book of Sketches of all Types of Machinery of a modern 

Cotton Mill, with full details and diagrams. 


Illustrated. Crown 8vo. 7s. 6d. net. 


A Practical Guide for ^Managers, Carders, and Overlookers. 

Illustrated. Crown 8vo. IDs. net. 



Illustrated. Crown 8vo. 12s. 6d. net. 


Illustrated. Crown 8vo. 15s. net. 


Illustrated. 8vo. 25s. net. 


Second Edition. Illustrated. Svo. 18.s. net. 


JUTE AND .lUTE SPINNING. In 2 parts. Part I. 
Production of Fibi'e : Cultivation, Batching, Pre- 
paring and Carding. (Second Edition.) Part II. 
Drawing and Roving Frames. Illustrated. 8vo. 
20s. net each pait. 

By F. H. BOWMAN, D.Sc. 

TIONS. Illustrated. Crown 8vo. 10s. Hd. net. 

TECHNICAL PURPOSES. Illustrated. Crown 
8vo. 10s. 6d. net. 


Medium 8vo. 20s. net. 

GROWERS. 8vo. Sewed. 3s. 6d. net. 

COTTON HAIIJS. Illustrated. 8vo. Sewed. 
3s. 6d. net. 




Its. Di'l' ,,.,^,., 

REW Oil