,íl* 'tt^- ^i-^ VAULTED BRICK GON9WUCTJ0N (U GUADALAJARA ARCHITECTURE AT RICE UNIVERSITY NUMBER 18 ARCHITECTURE AT RICE UNIVERSITY is a series of reports on thoughts and investigations from the School of Architecture of the University. The series is published in the belief that archi- tectural education can best be advanced when teachers, practi- tioners, students, and laymen share what they are thinking and doing. HOUSTON, TEXAS AUGUST, 1966 VAULTED BRICK CONSTRUCTION IN GUADALAJARA I. FOREWORD 3 BY CAREY CRONEIS II. STRUCTURAL EVALUATION 11 BY NAT W. KRAHL III. DESIGN EVOLUTION 47 BY HARRY S. RANSOM Digitized by the Internet Archive in 2011 with funding from LYRASIS IVIembers and Sloan Foundation http://www.archive.org/details/vaultedbrickcons18krah I. FOREWORD BY CAREY CRONEIS CHANCELLOR, RICE UNIVERSITY Introducción por Carey Croneis * I I Desde mediados de 1963 la Universidad de Rice ha tenido el placer y honor de formar parte de un grupo de universidades Norteamericanas quienes, a través de sus representantes, se han asociado con el objeto de cooperar con el desa- rrollo académico y físico de la Universidad Autónoma de Guadalajara. Profesores y funcionarios administrativos de varias univer- sidades importantes — Arizona, California, Colorado, Colorado State, Dallas, Denver, Kansas, New Mexico, Rice, Southern California, Texas, y Tulane — se reunieron desde el 28 de junio al 1°. de julio de 1963 en Guadalajara, México. A ellos se les unió un grupo especial de invitados que incluían al Lie. Luis M. Farias, el Director General de la Oficina de Información del Gobierno Mexicano, el Doctor Saxton Bradford, quien representaba a Su Excelencia, Thomas C. Mann, Embajador de los Estados Unidos en esa época en México, el Sr. John Nagel, representante de la Fundación Ford para México y América Central, el Sr. Thomas Linthicum, Cónsul General de los Estados Unidos en Guadalajara, el Ing. Salvador Ochoa Montes de Oca, Presidente de la Junta de Directores de la Universidad de Guadalajara, representantes de las compañías Bacardi y Nestle de México, junto a un grupo de distinguidos profesores y administradores de la Universidad Autónoma de Guadalajara, incluyendo al Rector, Dr. Luis Garibay G., el Vice-Rector, Lie. Antonio Leano A, de Castillo, el Secretario General, Lie. Carlos Pérez Vizcaíno, y el Proboste, Dr. Ángel Morales Castro, Since mid-1963 Rice University has had the honor and pleasure of being one of a group of American universities which, through their representatives, have banded together as a consortium whose purpose it is to assist in the academic and physical development of the Universidad Autónoma de Guadalajara. Professors and administrative officers of a number of major universities — Arizona, California, Colorado, Colorado State, Dallas, Denver, Kansas, New Mexico, Rice, Southern California, Texas, and Tulane — met in Guadalajara, Mexico, on 28 June to July 1, 1963. They were joined by a group of special guests including Lie. Luis M. Farias, the Director General of the Mexi- can Government Office of Information, Dr. Saxton Bradford, who represented His Excellency, Thomas C. Mann, then Am- bassador of the United States to Mexico, Mr. John Nagel, the representative of the Ford Foundation for Mexico and Central America, Mr. Thomas Linthicum, the Consul General of the United States for Guadalajara, Ing. Salvador Ochoa Montes de Oca, president of the Patrons of the University of Guadalajara, representatives of the Bacardi and Nestle companies of Mexico, together with a group of distinguished professors and adminis- trators from the Universidad Autónoma de Guadalajara, includ- ing the Rector, Dr. Luis Garibay G., the Vice-Rector, Lie. Antonio Leano A. de Castillo, the Secretary-General, Lie. Carlos Perez Vizcaino, and the Provost, Dr. Angel Morales Castro, as como asimismo el Dr. Osear F. Wiegand de la Universidad de Texas, coordinador de los planes de desarrollo para la UAG. Estos señores han sido muy cordiales y una gran ayuda a los miembros del cuerpo docente de la Universidad de Rice, quienes, de varias maneras, han participado en este programa cooperativo. Durante el curso de la conferencia inicial del año 1963, se plantearon los objetivos del programa cooperativo, fueron consideradas las posibilidades de ayuda y de iniciativa privada vis-a-vis a un desarrollo tradicional de las universidades en México, y se llegaron a acuerdos con respecto a las distintas áreas de colaboración por establecerse con cada una de las universidades Americanas representadas. Ai término de las sesiones se planteó a la asamblea una resolución dirigida a la Fundación Ford y a la UAG, cuyo borrador original fué preparado por el autor. Esta resolución fué adoptada, y ha formado la base principal para los diversos propósitos cooperativos que han sido puestos en marcha con excelentes resultados. Como consecuencia de ello, fondos para funcionamiento y construcción han sido recibidos de varias instituciones, incluyendo a la Fundación Ford, al programa US-AID y al Departamento de Estado de los Estados Unidos. well as Dr. Osear F. Wiegand of the University of Texas, the coordinator of the development plans for the UAG. All of these gentlemen have been most cordial and helpful to mem- bers of the Rice University faculty who, in various ways, have taken part in the cooperative program. During the course of the 1963 initial conference, the objec- tives of the envisaged cooperative program were spelled out, private initiative and support possibilities vis-a-vis orthodox university development in Mexico were considered, and agree- ments were reached with reference to the various areas of collaboration to be established with each of the American universities represented. At the close of the sessions, a resolu- tion directed to the Ford Foundation and to the Universidad Autónoma de Guadalajara was put before the assembly, the original draft of the proposal having been prepared by the present writer. This resolution was adopted, and it has formed a principal basis for the various cooperative arrangements which subsequently have been put into effect with a gratify- ¡ngly high degree of success. As a consequence, development, operating and construction funds have been forthcoming from various agencies, including the Ford Foundation, the US-AID program, and the U. S. Department of State. En la adjudicación de las áreas académicas de la UAC a las distintas universidades Norteamericanas, Ingeniería y Arquitec- tura fueron las disciplinas asignadas a la Universidad de Rice, y además, a Rice le fué entregada la responsabilidad de desarrollar el plano regulador del nuevo campus de la UAC. Durante la conferencia inicial, William Caudill, Director de la Escuela de Arquitectura de la Universidad de Rice, y el autor consideraron junto a los funcionarios de la UAG posibilidades potenciales de sitios para el nuevo campus de la Universidad, comparado esto con las posibilidades prácticas de desarrollar una institución de tal envergadura en el terreno actual, en el cual en esa época ya habían varios edificios nuevos. En última Instancia, se llegó a la decisión de seguir adelante con el desarrollo del sitio actual de la UAC. Cuando los deseados fondos necesarios para situar el proyecto de la UAC en su primera etapa de planeamiento llegaron, le fué posible a la Universidad de Rice enviar al profesor Harry Ransom de la Escuela de Arquitectura a Cuadalajara a trabajar bajo la égida del Departamento de Estado de los Estados Unidos. El proyecto comenzó en octubre de 1964 con tres arquitectos y cuatro estudiantes, sin embargo, el 1°. de enero de 1966 participaban 30 estudiantes, 3 arquitectos y 4 ingenieros. A través de todo el programa, a la Universidad de Rice y al profesor Ransom les ha sido útil no sólo la eficaz cooperación In the apportioning of areas of academic concern to the various U. S. universities of the UAC consortium, engineering and architecture were disciplines assigned to Rice University, and, in addition. Rice was given the responsibility for develop- ing the master plan of the new Autónoma campus. At the time of the original conference, William Caudill, Director of the Rice School of Architecture, and the writer spent some time with the officials of the UAC in considering potential sites for a new campus for the University, as compared with the practicality of developing a major institution on the present University campus which at that time was already occupied by several new buildings. Ultimately, a decision was reached to go ahead with the development of the present site of the UAC. When the hoped-for funds necessary to put the entire UAC project into its detailed planning stage became available, it was possible for Rice University to send Professor Harry Ransom, of the School of Architecture, to Cuadalajara to work under the aegis of the United States Department of State. The project began in October, 1964, with three architects and four students. On January 1, 1966, however, there were 30 students, three architects and four engineers involved. Throughout the entire program. Rice University and Professor Ransom have profited not only from the effective cooperation de la administración de la UAC sino que además, la ayuda directa que han recibido de las facultades de Arquitectura e Ingeniería de dicha Universidad. Rice está especialmente agradecida al arquitecto Francisco Camarena, Director de la Escuela de Arquitectura y al Ingeniero José Luis Amezcua S., Director del Instituto de Ciencias Exactas y Terrestres. El dibujo de los cientos de planos fue hecho por un enorme grupo de alumnos avanzados, sin cuya valiosa ayuda este proyecto no podría haberse completado. Muy pronto en el trabajo de coordinación del diseño del campus, el profesor Ransom quedó particularmente intrigado con la vieja costumbre local de construir sistemas de techum- bre con bóvedas de ladrillo. Como en ello había ciertos problemas de orden estructural invitó al profesor Nat Krahl de la Universidad de Rice, a visitar la UAG con el propósito de estudiar esta técnica en profundidad y especialmente para iniciar una serie de pruebas y experimentos a fin de deter- minar la gama de posibilidades estructurales inherentes a este sistema de construcción local. El informe que sigue, documenta el procedimiento arquitec- tónico-estructural e ilustra como, en este caso en particular, la unidad ladrillo se transformó en la determinante de diseño de un campus universitario completo. Pruebas y análisis estructurales tuvieron éxito al demostrar la gran resistencia propia al sistema de bóvedas de ladrillo convencionales. Una serie de pruebas hechas en construcciones de tamaño natural determinaron con éxito las dimensiones de la trama estructural of the UAG administration, but from the direct assistance of the Autonoma's Architectural and Engineering faculties. Rice is particularly indebted to Arq. Francisco Camarena, Director, UAG School of Architecture, and to Ing. Jose Luis Amezcua S., Director, institute of Exact and Terrestrial Sciences. The draft- ing of the hundreds of drawings was done by a large force of advanced architectural students, without whose valued assistance the project could not have been completed. Early in Professor Ransom's work in coordinating the campus design of the University, he became particularly intrigued with an ancient local practice of constructing roof systems of vaulted brick. Because there were certain engineering prob- lems involved, he invited Rice Professor Nat Krahl to visit the UAG for the purpose of examining the technique in depth, and, in particular, to initiate a series of engineering tests and experiments to determine the actual range of engineering possibilities possessed by this local construction system. The following report documents the architectural-engineering procedure and illustrates how, in this particular circumstance, a unit of brick became the design determinant for the architec- ture of an entire university campus. The structural tests and structural analysis were successful in demonstrating the great inherent strength of the conventional brick vaults. A range of full-size construction tests was successful in determining the que puediese permitir la construcción de bóvedas de ladrillo de doble curvatura sin el uso de moldajes, y que como resultado final se consiguió un método más sofisticado de construcción de bóvedas, una vez más, sin moldajes. Resulta que el método no sólo es práctico, sino que permite una rápida construcción de bóvedas soportadas por una trama de vigas y pilares de hormigón armado. Además, no es únicamente un proceso relativamente barato sino que su resultado es estéticamente agradable. Esto se debe en parte a que el cielo de la bóveda puede ser dejado a la vista dado el extra- ordinario color y la interesantísima textura de los ladrillos utilizados. Al hablar en la conferencia original en Guadalajara, el autor hizo notar lo relativamente mucho que la UAG había realizado con el relativo poco apoyo financiero, a tal punto de situar en una posición incómoda a varias de las universidades Ameri- canas allí representadas que comunmente se quejan de la suficiencia de sus mayores recursos. Con respecto a esto, se hizo notar que la edad de una persona u organización se puede determinar por la cantidad de dolor que el individuo u organización experimenta al enfrentar una nueva idea. UAG es joven y vigorosa, y no ha experimentado dolor, por lo menos de esa naturaleza. Ha experimentado audazmente y con éxito desarrollar una institución privada e independiente, apoyada por particulares y corporaciones y especialmente por dimensions of a supporting structural network that would permit doubly curved brick vaults to be constructed without the use of forms and eventually resulted in a more sophisti- cated method of building such vaults, again without forms. It turns out that the method is not only practical but that it permits rapid construction of vaults which can be supported on a network of reinforced concrete beams and columns. Moreover, not only is the process relatively cheap but its results are also aesthetically pleasing. This is true in part because the underside of the vaults can be exposed ad- vantageously owing to the unusually attractive color and interesting surface texture of the bricks employed. Speaking at the original conference at Guadalajara, the writer observed that the UAG had accomplished relatively so much with relatively little financial backing that it put many of the U. S. universities represented in an embarrassing position because they commonly complain about the adequacy of their much greater resources. In this connection, it was also pointed out that the age of a person or an organization can be determined by the amount of pain the individual or the institution experiences when encountering a new idea. UAG is young and vigorous, and experiencing no pain, at least of this sort. It has boldly and successfully experimented by developing an independent private institution, supported by 10 "recibos de educación," procedimientos que son "nuevos" y heterodoxos en la educación superior Latinoamericana. Es por lo tanto propio y adecuado entonces, que en su moderno y futuro desarrollo UAC tome ventaja de un "nuevo" concepto arquitectónico-estructural, que sin embargo está firmemente enraizado en un antiguo y atractivo procedimiento de con- strucción característico del estado de Jalisco. La Universidad de Rice está orgullosa de haber tomado parte en el desarrollo de UAC, particularmente a través de los proyectos de arquitectura e ingeniería dirigidos por los pro- fesores Ransom y Krahl, delineados someramente en esta publicación. Su agradecimiento a las otras universidades Norteamericanas que forman parte del consorcio es grande, pero a la de los líderes de la Autónoma es aún mayor. Entre ellos los Drs. Luis Caribay C. y Ángel Morales Castro merecen especial mención por su dedicación eficaz e incansable a una causa extraordinaria. El esfuerzo cooperativo entre Rice y UAC se refleja en la mejor forma quizás en el símbolo integrado de diseño que aparece en todos los planos. Representa el apropiado, mas biológicamente imposible, matrimonio de las águilas Mexi- canas de la Universidad Autónoma de Guadalajara y los buhos Atenienses de la Universidad de Rice — colofón que orgullosa- mente presentamos en esta publicación. individuals and corporations, and particularly by tuition receipts, all of which procedures are "new" and unorthodox in Latin American higher education. It is altogether proper and fitting, therefore, that in its modern and future develop- ment the UAC should take advantage of a "new" structural and architectural concept which is nonetheless firmly rooted in an ancient and most attractive building procedure characteristic of the state of Jalisco. Rice University is proud to have been connected in several material ways with the development of the UAC, particularly through the architectural and engineering projects directed by Professors Ransom and Krahl, which are briefly outlined in this publication. Its indebtedness to the other U. S. universities of the consortium is great, but its obligation to the leaders of the Autónoma is even greater. Among them Drs. Luis Caribay C. and Angel Morales Castro deserve special mention for their tireless and effective dedication to an unusually worthy cause. The cooperative effort between Rice and the UAC is perhaps best characterized by the integrated design symbol which appears on all of the production drawings. It represents the appropriate, if biologically improbable, wedding of the Mexi- can eagles of the Universidad Autónoma de Cuadalajara and the Athenian owls of Rice University — a colophon we are pleased to display in the present publication. II. STRUCTURAL EVALUATION BY NAT W. KRAHL ASSOCIATE PROFESSOR OF STRUCTURAL ENGINEERING TABLE OF CONTENTS 1. Purpose 17 2. Origin and Description 17 3. Test Program 22 4. Analysis of Vaults 27 5. Evaluation 28 6. Conclusions 31 Appendix A. Test Results 33 Appendix B. Conversion Factors Between Metric And English Units 45 Part II is the text of an engineering report submitted to the campus design team at the Universidad Autónoma de Gua- dalajara for their use in designing the new campus buildings. LIST OF FIGURES FIGURE 1. FIGURE 2. FIGURE 3. FIGURE 4. FIGURE 5. FIGURE 6. FIGURE 7. FIGURE 8. FIGURE 9. FIGURE 10. FIGURE 11. FIGURE 12. FIGURE 13. FIGURE 14. FIGURE 15. FIGURE 16. FIGURE 17. FIGURE 18. Vaulted brick floor and roof systems under construction on steel framework 17 Vaulted brick roof system under construction on precast concrete framework 17 Terrado construction 18 Primitive Catalan vaulting 18 Techo de bóveda 19 Hand-made mud brick drying in the open air at Tateposco 22 Hand-made mud brick stacked for firing at Tateposco 22 Test vaults 23 Brick arches constructed for original tests 24 River sand bags being placed for uniform loading tests 24 Arch end view during early loading 24 Deflection meter placed at center of beam span 24 Instrumentation reading of vault deflection 25 Partially loaded test 25 At this point of loading the vault is supporting 388 pounds per square foot 25 With no appreciable damage to any structural element, vaults supporting 388 pounds per square foot. This is 5.4 times design live load 25 Line of thrust of test vaults 26 Calculations for: Horizontal Thrust Max. Compressive Stress Tensile Stress in Tie Rods 26 FIGURE 1. Vaulted brick floor and roof systems under construction on steel framework. FIGURE 2. Vaulted brick roof system under construction on precast concrete framework. 1. PURPOSE For some time now the School of Architecture at Rice University and the architectural firm of Caudill, Rowlett, and Scott have been assisting representatives of the Autonomous University of Guadalajara in the preparation of a Master Campus Plan for the development of their university. One concept v^hich has found favor with the planners is that of using an indigenous construction system — namely, roofs and floors of very flat barrel vaults of brick, supported on a struc- ture of contemporary design — as an architectural and structural feature of the campus buildings. This report is an outgrowth of a trip by the writer to Guadalajara for the purpose of observing this unique construction system and is based in part on a continuing series of tests which have been performed at the Autonomous University since that time. The purpose of this report is to present an evaluation of this construction system to the campus planning team for their use in the preparation of their plans. 17 2. ORIGIN AND DESCRIPTION OF VAULTED BRICK CONSTRUCTION IN GUADALAJARA 2.7 ORIGIN This construction system, known in Guadalajara as "techo de bóveda," is essentially a structural system whereby a roof or floor is constructed of brick multiple barrel vaults, almost flat, which are supported on a framework of steel (Fig. 1), or sometimes concrete (Fig. 2). The system is apparently unique to a portion of central Mexico and reaches its fullest expression in the state of Jalisco and its principal city, Guadalajara. This method of construction is an outgrowth of centuries of experience and experimentation with this and similar methods. 18 -Earth Fill Wood Beam • FIGURE 3. Terrado construction. ■ X' Fill arase Brick Barrel Vault ■ Wood Beam FIGURE 4. Primitive Catala'n vaulting. — Wearing Su rface Pill 1 Steel Beam 1 y-t ^^í-íS;;:^-;--jt-:-^-:\'^^-^-H' ~^;j^^g^^;i^áik^p¿mt¿jj^^^^^^ — Plaster 1 Brick Barrel Vault FIGURE 5, Techo de bóveda. It apparently stems originally from the Terrado system of building (Fig. 3) in which the Indians of Mexico used wood beams to support tree branches which were covered with earth. This method was modified by the Spanish to the Catalan system (Fig. 4), in which wood beams supported short-span multiple brick vaults which, in turn, supported an earth or concrete fill and wearing surface. With the introduction of structural steel, some builders used rolled steel beams in lieu of wood beams to support the brick vaults, and examples of this are to be seen in Guadalajara. Then someone observed that the shape of the steel beam would allow it to be placed between two adjacent vaults, thereby containing the depth of the arch rib and being virtually concealed itself, and the final step in the evolution of the method had been taken (Fig. 5). 2.2 DESCRIPTION Any description, analysis, or evaluation of this construction system must first recognize that this system is a combination of two separate structures: first, the brick vaults, which support the floor and the loads imposed on the floor and carry these loads to the supporting structure; and second, the supporting framework of steel or concrete beams, girders, and columns which directly support the vaults. 19 20 a. Vaults ' A typical cross-section through this type of construction is shown in Fig. 5. Typically, the supporting beams are placed 80-130 cm. (32-51 in.) center-to-center, and this distance becomes the span of the barrel vault. The bricks comprising the barrel vaults are laid without formwork, each successive row of bricks in itself becoming an arch rib spanning between steel beams. The vaults themselves are almost flat, the rise usually being only 3-8 cm. (1.5-3 in.). The bricks are handmade mud bricks of local manufacture, are very light in weight and are laid with the large side flat against the adjacent arch rib. The bricks are laid dry, and the suction of the mortar against the side of each brick helps keep it in place until its arch rib has been completed. Also, the vertical axis of each brick is tilted slightly so that each brick is partly supported by the newly-completed adjacent arch rib until its own rib is completed. The mortar in which the brick is laid is made from lime and sand. A level surface above the vaults and beams is achieved by filling with a lightweight concrete made from hydrated lime and "Jal," which is a local pumice sand and gravel and is the volcanic material from which the State of Jalisco takes its name. The thickness of the fill varies, particularly on roofs, which are usually sloped for drainage. For example, a variation of fill thickness from 5 to 20 cm. (2 to 8 in.) over a roof would not be unusual. A wearing surface of burnt clay brick or tile is usually placed on top of the concrete fill. On roofs a waterproofing layer of asphalt or cement is placed between the concrete fill and wearing surface. Finally, the underside of the vaults is almost always covered with a plaster made from lime and sand to give a perfectly flat ceiling which conceals the vaults and beams. In some cases a wire mesh is attached to the beam flange to support the plaster below the flange. In a few cases the brick vaults and steel beams have been left exposed to view from below and achieve a dramatic architectural effect because of the varying colors of the brick and the apparent daring of the flatness of the vaults. The brick vaults described above are widely used in Guadala- jara to support the typical floor loads and roof loads encoun- tered in houses and office buildings. The floor live load used locally for design of dwellings is 150 kg./m,= (30.8 Ib./ft.'), and maximum office floor live load for which this type of construc- tion would be considered suitable by local engineers is about 350 kg./m." (71.8 Ib./ft.-'). In addition, the brick vaults them- selves are not considered suitable for resisting heavy, con- centrated loads. If such loads are anticipated in buildings of this type of construction, it is customary to support these loads on small beams of steel or reinforced concrete, which replace the arch ribs in that immediate location and carry their loads directly to the supporting framework of beams, girders, and columns. The allowable loads and the practices stated above are based primarily on accumulated experience. Prior to the series of tests described in this report, there were apparently no test data available to aid in an evaluation of this structural system. Apparently the only criticism of this system by local architects and engineers is that a few buildings employing this system have floors which will vibrate noticeably when excited by such factors as nearby truck traffic, a person jumping on the floor, etc. Some floors which can be excited to vibrate notice- ably during construction show no objectionable vibration after completion of the building. The writer believes this behavior to be primarily dependent upon the supporting structure and discusses the point further in Chapter 5. b. Supporting structure As mentioned above, the beams which directly support the vaults are usually steel beams placed 80-130 cm. (32-51 in.) apart. Typical beam sizes range from 4-inch to 8-inch I-beams, depending on their span and their supported load. Usual spans are in the range of 4-6 meters (13-20 ft.). Most often the steel beams are supported by steel girders which, in turn, are supported by steel columns. Occasionally, small precast con- crete beams of 1-section are used in lieu of steel beams to support directly the brick vaults (Fig. 2), and the remainder of the supporting structure may then become reinforced concrete. At an interior beam under uniformly distributed load, the lateral thrust from the vault on one side balances that from the other side; but at edge beams under any loading and at interior beams under unsymmetric loadings there are unbalanced thrusts to be resisted. For this purpose, it is the common practice to use tie rods between adjacent beams at intervals along the length and to extend the lines of rods all the way across the building. The rods are eventually concealed within the floor construction. Individual members of the supporting structure are, of course, designed according to the live and dead loads, the conditions of support, and the span or length of members. 21 3. TEST PROGRAM In order to aid in the evaluation of this structural system, a series of tests was performed at the Autonomous University of Guadalajara. The original reports of these tests, in Spanish, and an English translation are presented in Appendix A of this report. The tests themselves can be classified into two groups: first, tests of individual materials used in this structural system; and, second, a test of full-scale vaults. 3.7 Tf5TS OF MATERIALS The brick used in Guadalajara for vault construction is made by hand in the countryside nearby, and is trucked into the city for use. The primary ingredient is a mud made from water and Jal, which is readily available throughout the entire area. In addition, certain amounts of manure and maguey fiber are used in the mixture. The bricks are formed in wood molds, dried in the open air (Fig. 6), then stacked in large piles and fired with mesquite logs, which are placed in slots left in the piles. (Fig. 7). The resulting brick is light in weight, very porous, and has beautiful variations in color, which depend largely upon the degree of firing. Predominant colors range through various shades of yellow, ocher, orange, red and brown. Appendix A contains the quantitative measurements of certain mechanical properties of samples of brick coming from the brickyards of Tateposco and Las Pintas. In addition to tests of the brick used for the barrel vaults, certain mechanical tests were also performed on representative samples of the mortar used in the vault construction, the Jal concrete fill placed above the vaults, and the burnt clay brick used for the wearing surface. Specifically, the samples of each of these materials corresponded to those particular materials used in the construction of the full-scale vaults that were load tested. Results are given in Appendix A. ^ - J-'S^'SL'^-jLa.'^a.'^ ■!?■ •■•¿^■P" *^>1^ "^-"^ ' ^ ■''^ íjÜ^ m^jHUB FIGURE 6. Handmade mud brick drying in the open air at Tateposco. FIGURE 7. Handmade mud brick stacked for firing at Tateposco. •Mif ¡Jr.: ■ -p:--^, ','.1i.'^u^fr^r'-"'ii''A- ■■,<*.■* fe it; ►_••■' --'x^j^r'-tJ^^:-:. e r-\ ■ 2V."* SECTION A-A Note: Dimensions shown in Meters, ^^nnr 3.2 TEST OF VAULTS Appendix A contains the report of the load test of full-scale brick vaults performed on the campus of the Autonomous University of Guadalajara. A sketch of the structure which was tested is shown in Figure 8 and photographs of the test are shown in Figures 9-16. The basic aim of the test was to observe the behavior and the strength of the brick vaults under uni- formly distributed load. As mentioned above, the structural system under consideration has two separate components, the brick vaults and the supporting structure. It was felt that a supporting structure of steel or reinforced concrete could be readily analyzed with confidence because of the wealth of research information which is available concerning such struc- tures. However, no test data were available for the brick vaults, since they seem to be unique in several respects. What was desired, therefore, was a measure of the strength of the vaults themselves, without a premature failure on the part of the supporting structure. To achieve this end, the steel beams supporting the test vaults were made arbitrarily oversize, 10- inch I-beams on a 3 meter (9 ft. 10 in.) span, while the vaults were constructed of conventional proportions and materials. Sacks of river sand were used to simulate a uniformly dis- tributed load on the structure. The sacks failed before the structure, but the test was successful in demonstrating that the structure was capable of sustaining a superimposed uniformly distributed load of 1893 kg./m.= (388 lb./ft.=) without apprecia- ble damage and without excessive deflection. Based on the maximum live load in current use for this type of construction, which is 350 kg./m.' (71.8 Ib./ft.-), we find a factor of safety which must be greater than 5.4. Additional details of the vault load test are found in Appendix A. 23 FIGURE 8. Test vaults. FIGURE 9. Brick arches constructed for original tests. FIGURE 10. River sand bags being placed for uniform loading tests. 24 FIGURE 11. Arch end view during early loading. FIGURE 12. Deflection meter placed at center of beam span. í f ■# # FIGURE 13. Instrumentation reading of vault deflection. FIGURE 14. Partially loaded test. FIGURE 15. At this point of loading the vault is supporting 388 pounds per square foot. FIGURE 16. With no appreciable damage to any structural element, vaults supporting 388 pounds per square foot. This is 5.4 times design live load. 25 26 FIGURE 17. Line of thrust of test vaults. FIGURE 18. CALCULATIONS FOR: HORIZONTAL THRUST, MAX. COMPRESSIVE STRESS, TENSILE STRESS IN TIE RODS Span: 1.30 m. Dead Load: Paving Brick 15 kg./m^ 3 Ib./ft.' Jal Concrete Fill 70 kg./m^ . . .14 Ib./ft.^ Brick Vault 145 kg./m^ . . . 30 Ib./ft.^ 230 kg./m^ 47 Ib./ft.^ For Design Live Load of 350 kg./m.^ (72 Ib./ft,^): Total Load = 230 + 350 = 580 kg./m.^ (119 Ib./ft.') L, . ^ I Tu ^ wL^ 580X1.30' Hor,zontal Thrust= — = ^^^OO^ Horizontal Thru5t = 1,630 kg./m. (1,097 Ib./ft.) ». r- • c* 2H 2X1,630X1 Max. Compressive Stress= ^ = ^^^^^^ ^ Max. Compressive Stress = 2.91 kg./cm.' (41 lb/in.') -r ., c. ■ T- D J H 1,630X1.50 Tensile Stress m Tie Rods= — = ., As 2 X U.71 Tensile Stress in Tie Rods=1,720 kg./cm.' (24,500 Ib./in.') For Max. Test Live Load of 1,893 kg./m.' (388 Ib./ft.'): Total Load = 230 + 1,893 = 2,1 23 kg./m.' (435 Ib./ft.') Horizontal Thrust = 5,960 kg./m. (4,010 Ib./ft.) Max. Compressive Stress = 10.64 kg./cm.' (151 Ib./in.') Tensile Stress in Tie Rods = 6,290 kg./cm.' (89,500 Ib./in.') 4. ANALYSIS OF VAULTS Figure 17 shows a scale drawing of a cross-section of the barrel vault used in the load test. If we consider a unit thickness of vault perpendicular to the plane of the paper, we can analyze the vault as an arch rib. Since these vaults in practice are con- structed without formwork, it must be realized that the actual curve of the underside, or intradós, of the rib will vary some- what from one cross-section to another. But, since the rise of the arch is so small, only about 3% of the span, all smooth curves of this rise and span will lie very close to one another. FHence the small variations in construction are likely to be un- important. For simplicity, the placement of bricks in Fig. 17 is shown approximating the curve of a second-degree parabola. Because this arch is of relatively short span and, in practice, supports a relatively light, uniformly-distributed load, the so- called "line of thrust" analysis is considered to be the most suitable basis for analysis.* According to this analysis, the line of thrust under uniformly distributed load becomes parabolic, the crown thrust is horizontal, and its resultant lies at the upper extremity of the middle third of the arch rib, while the resultant thrust at the skewback lies at the lower extremity of the middle third of the arch rib. •Harry C. Plummer, Brick and Tile Engineering, Structural Clay Products Institute, Washington, D.C., 2nd ed., 1962, pp. 199-214. The line of thrust following these restrictions is shown in Fig. 17. Since this line of thrust lies entirely within the middle third of the bricks making up the arch rib, no tensile stress will be developed in any part of the rib regardless of the magnitude of the applied load, and hence we can conclude that the arch is stable against failure by rotation of one section of the arch about the edge of a joint. Other possible modes of failure to be investigated include the sliding of one section of the arch on another, crushing of the masonry, and failure of an abutment to provide adequate thrust resistance. As a measure of sliding stability we examine Fig, 17 and find that the maximum angle between the line of thrust and the normal to the joint between arch sections is about 6 degrees. The significance of this number is evaluated in the next chapter. The rise of the arch itself is only .3.9 cm. (1.5 in.), but we see from Fig. 17 that the rise of the thrust line is 7.5 cm. (3.0 in.). Fig. 18 shows the calculations for horizontal thrust, maximum compressive stress in the arch, and tensile stress in the tie rods for two separate conditions: first, a design live load of 350 kg./m.- (72 lb. /ft.-); and, second, the maximum test live load of 1893 kg./m." (388 Ib./ft.'). The significance of these stresses is evaluated in the next chapter. 27 28 5. EVALUATION 5.1 VAULTS AND COMPONENT MATERIALS Two aspects of vault construction and behavior are worthy of comment here: first, the unique construction method whereby the brick vaults are laid without formwork; and, second, the ability of the vaults to carry uniformly distributed loads of considerable magnitude without damage or excessive deflection, despite their flatness. The handmade bricks used in constructing the vaults are ex- tremely porous because of the materials and methods used in their manufacture. This high porosity is reflected in the absorp- tion percentages reported in the test results of Appendix A. In addition, the bricks are very light in weight, being about one- third less in unit weight than the usual machine-made brick used in the United States. Thus, when a brick is laid dry in mortar against an adjacent, newly-completed arch rib, the suction generated on the contacting large face is sufficient to hold the light brick in place until its own arch rib is completed. The naturally low humidity of the Jalisco climate contributes to this process, of course. Also contributing to the support of the new rib is the fact that the vertical axis of each brick is slightly tilted toward the existing construction so that the newly-completed arch rib actually provides some positive mechanical support for the rib under construction. In order to evaluate the load-carrying capacity of the vaults, we shall combine the results of the load test described in Chapter 3 and the mathematical analysis reported in Chapter 4. Modes of failure to be considered can be listed as follows: rotation of one section of the arch about the edge of a joint, the sliding of one section of the arch on another, crushing of the masonry, and failure of an abutment to provide adequate thrust resistance. We shall examine each of these possibilities in turn. First, we see from Fig. 17 that the line of thrust under uniform load lies entirely within the middle third of the bricks making up the arch rib, regardless of the magnitude of the applied load; and this precludes the possibility of a tensile stress developing in any part of the masonry. Hence, we can state that the arch is stable against failure by the rotation of one section of the arch about the edge of a joint. Next, from Fig. 17 we see that the maximum angle between the line of thrust and the normal to the joint between arch sections is about 6 degrees. The coefficient of friction between the units is at least 0.50, without counting the additional resistance to sliding provided by bond between the mortar and the masonry units. This coefficient of friction corresponds to an angle of friction of about 27 degrees; and, since 6 degrees is much less than 27 degrees, the arch is quite stable against sliding. Concerning the possibility of masonry crushing, we should be able to conclude that since the test vaults sustained no appreci- able damage under a superimposed live load of 1893 kg/m." (388 lb. /ft.-), the stresses produced by this load must be lower than the failing stresses for the component materials. Maximum compressive stress in the vault is calculated in Fig. 18 to be 10.64 kg. /cm.-, well below the compressive strength of Tateposco brick, 65.5 kg. /cm.-, and even well below the com- pressive stress at first crack, 20.8 kg. /cm.' The calculated stress of 10.64 kg. /cm.- is slightly greater than the measured compres- sive strength of the mortar, 8.8 kg. /cm.-; but we should recognize that the mortar strength was measured in an uncon- fined compression test, whereas the mortar in the vault is stressed in a confined condition in which its strength is likely to be much greater than in the unconfined condition. Thus the arch rib is seen to be safe against crushing with a large factor of safety for a design load of 350 kg./m.", and this is seen to be true by both analytical and experimental considerations. Thrust resistance is provided by the steel beams and tie rods. The only question concerning the adequacy of the thrust resist- ance arises in the case of an unbalanced thrust, as on an edge beam. The edge beam transfers this thrust to the tie rods, which tie the two edge beams together across the width of all the vaults. Fig. 18 indicates a calculated tensile stress of 6290 kg./cm.= (89,500 Ib./in.") in the tie rods of the test vaults at maximum load. The test results give no data concerning the yield strength of the particular grade of steel from which the tie rods were manufactured; but for most grades of structural steel rods, the yield strength would be well below 6290 kg. /cm.- If the tie rods had yielded, this would likely have been reflected in a sudden increase of vault deflection; and, since no such increase was observed, it is likely that the tie rods did not yield and that the calculated stress in the tie rods is too high. Several factors help to explain this situation. First, the entire test structure was only 3 meters long between centers of end supports. Through friction at the bearings of the steel beams on the end supports, some force may have been transferred into the supports so that they participated with the tie rods in providing thrust resistance. Also, the Jal concrete and paving brick are considered to be only dead weight, and the only load-resisting masonry element in the floor construc- tion is assumed to be the brick arch. Actually, the jal concrete and the paving brick have structural strength in themselves and may well participate with the brickwork in supporting the loads. If so, the line of thrust would have a much greater depth available for its equilibrium position, the magnitude of the thrust would be reduced, and the resulting stresses would thereby be reduced. This is to say that the line of thrust analysis illustrated in Fig. 17 is conservative for design, but may be overconservative for the evaluation of test results when applied to this particular structure. It would seem prudent, however. 29 30 to provide tie rods in accordance with the requirements of the line of thrust analysis for any construction contemplated on the UAC campus. It would seem unnecessary to check the ability of the edge beam to withstand the bending produced by the horizontal thrust between tie rods, since the great stiffness of the vaults in the horizontal plane would make possible a horizontal arching between tie rods even if the beam were relatively flexible in this direction. A word of caution should be mentioned concerning the proper elevation of the tie rods. In order to avoid eccentricities which could produce secondary moments and stresses in the struc- ture, it will be desirable to keep the elevation of the tie rods as close as possible to the intersection of the line of thrust with the web of the steel beam. According to Fig. 17, it looks as if this requirement can be satisfied and still keep the tie rods concealed in the depth of the arch rib. A final word of caution concerns the handling of concentrated loads. The test program did not investigate concentrated loads, nor is the line of thrust analysis recommended for treating concentrated loads. Unless further investigations are made into this type of loading, it is recommended that the limitations on the use of this structural system which have been accumu- lated through experience be strictly followed where concen- trated loads are involved. In general, this would mean leaving out arch ribs in the immediate location of the concentrated loads and replacing them with beams of steel or reinforced concrete. 5.2 SUPPORTING STRUCTURE As mentioned above, a great deal of research information is available concerning the behavior of a structural framework of steel or reinforced concrete when subjected to a known loading condition. Therefore, the design of such a framework to resist certain specified loads is, to some extent, a straight- forward matter. However, two special conditions pertaining to the structures supporting brick vaults in Guadalajara deserve special consideration. The latest specifications of the American Institute of Steel Con- struction recommend* that the depth of fully stressed beams and girders in floors be not less than Fy/800,000 times the span; and, where subject to shock or vibration, not less than Fy/650,- 000 times the span. In their notation Fy is the specified mini- mum yield point of the type of steel being used, in pounds per square inch. If members of less depth are used, the unit stress in bending should be decreased in the same ratio as the depth •Section 1.13, "Commentary on the Specification for the Design, Fabrication anci Erection of Structural Steel for Buildings," April 17, 1963, American Institute of Steel Construction. is decreased from that recommended above. These specifica- tions also requiret that the depth of beams and girders sup- porting flat roofs be not less than fi,/600,000 times their span length whether designed as simple or continuous spans, where fb is the computed bending stress in pounds per square inch. These specifications are mentioned because apparently some of the beams used in Guadalajara to support the vaulted brick construction are too shallow to meet these standards. The excessive flexibility of some of these beams is the most likely source of the noticeable vibration which is occasionally present in a structure of this sort, and which was mentioned in Chapter 2. It is recommended that all of the AISC Specifications be strictly followed in the detailed design of structural steel for the buildings of the UAC campus, and that the latest specifica- tions of the American Concrete Institute be followed in the detailed design of any concrete framework. The second special condition worth mentioning is the fact that Guadalajara lies in a region where earthquake activity is frequent and, occasionally, severe. It has been some time since a major earthquake was felt in this area, but such earthquakes are a matter of historical record here. Earthquake engineering has advanced to the point where it is perfectly feasible for a tSection 1.13, "Specification for the Design, Fabrication and Erection of Structural Steel for Buildings," adopted April 17, 1963, American Institute of Steel Construction. building to be designed to pass through a minor seismic dis- turbance without appreciable damage and to survive a major seismic disturbance. Proper design for these conditions requires a consideration of the overall configuration of the building in addition to individual member sizes, connections, etc. It is recommended that due attention be given to seismic considerations in the design of the UAG buildings. 6. CONCLUSIONS Based on the considerations outlined in the first five chapters of this report, the following conclusions are offered: 1. it has been demonstrated by load test and by mathematical analysis that multiple barrel vaults of handmade brick, when constructed of the proportions and the materials customary in Guadalajara, can be used with a high factor of safety to support their own dead weight plus a uniformly distributed live load up to 350 kg./m.= (72 lb./ft.=). For a live load of 350 kg./m-. the factor of safety is greater than 5.4. 2. In any use of these brick vaults in the new buildings on the UAG campus, it is recommended that tie rods between sup- porting beams be designed to resist the thrust obtained by making a line of thrust analysis (or similar analysis) on the vaults themselves. 31 3. Since no tests have been made to determine the behavior of the brick vaults under concentrated loads, it is recom- mended that any use of the vaults strictly observe the limita- tions which have been accumulated through experience regarding the support of concentrated loads. In general, this would mean leaving out brick arch ribs in the immediate location of the concentrated loads and replacing them with beams of steel or reinforced concrete. If any modification of this recommendation is desired, additional tests should be performed to justify such modification. 4. It is recommended that any steel members, in the structural framework which supports the vaults, be designed in accord- ance with the latest specifications and recommendations of the 32 American Institute of Steel Construction. Special attention is called to Section 1.13 of the Specifications and the- Commen- tary on the Specifications concerning limitations on depth of flexural members, since apparently some steel beams used in Guadalajara to support brick vaults are shallower than these provisions recommend. 5. It is recommended that any concrete members, in the struc- tural framework which supports the vaults, be designed in accordance with the latest specifications of the American Concrete Institute. 6. It is recommended that due attention be given to seismic considerations in the design of the UAG buildings. APPENDIX A TEST RESULTS ^acultab he ingeniería Chnl REPORTE DE AHALIliia DE t.UTERIALeS EFECTUADOS EN EL LABORATüfilO DE LA FACULTAD DE INGEMJRIA Ü.A.G. MATERIAL ENSAYADO ; Ladrill* de lama hecli* a man* procadenta .......^ A,.,™ .. .„^,.i.,. ^g ^^g ladrilleras de "Tatepe«8«". Jiuarponiba ■ (■ enftimUlab lÜiobMb «nítitnitarta ^ulóaz Qltl. S-95-a6 34 Muestra 1 . 2 3,4,5 Pr»m. Fes* V*lum<- tric* Sec* K*/M3 1260 "1255 1200 '1210 '1370 1259 IlÜdul* de Bup- tura. (K«/o«2) 15.4 |9.41 16.7 '15.6 '25 16.42 Beslstencla a la o«inpreBi. . I. enit.n.ib«ft Reporte de análisis de materiales efectuados en el Labora Mnhicul bi ^cilcD aitob-b BBto.rtitari- A»i¿™«- torio de la Facultad de Ingeniería de la U.A.O. tnil. 9-95-38 Material ensayado í Ladrillo de lama hecho a Baño proceden- te de las ladrilleras de "Las Pintas". 36 1 f » t 1 t 1 1 , Kuestra ,1,2,3 ,4 , 5 , Prom. , ' Peso volume t t i f » t i • l'^h'^S ' 1180 ' 1070' 1180 • 1180 ' 1180 • 1158 ' Kg/n » seco lililí f ! Kídulo de - ' • i n i i i , ruptura lililí 1 (Kg/om'^) 11.4 11.9 13-55 16-50 11.35 12-94 • Reei Btencia " ' 'i • * » • , a la compre , 33-15, 43-2 , 72-2 ,6l-2 , 14.6 , 44.87 , Bien » Primera - - • » » » ' • • , grieta , 12.3 , 14-7 , 15-7 , 30-3 , 9-9 i 16.58 , Ka/cB'^ i "nfj" 1 ^^•«'^ ', 51-'*^ /36-35íí|^30-8lí 1 38-8Í | 35-03% ' • (Shr'e^ll.' -T-'^^- • ^"-^^I "5-5''; tlfí : 46?6 ,' 42.82# 1 Coeficiente ' de satura— • 1 • 0-71 ' 0.79 ' 0.75 ' 0.64 ' 0.8l5 • I oiín lililí I encargado del Laheratsrlo Jefe del Eepto. de Física Tng. Carlcs-^rujillo del Río % Fi^arjel ROO Niíñez Farias REPORT OF ANALYSIS OF MATERIALS PERFORMED IN THE LABORATORY OF THE SCHOOL OF ENGINEERING U.A.G. MATERIALS TESTED: Handmade mud brick coming from the brickyards of "Las Pintas.' Test Units Sample Average 1 2 3 4 5 Volumetric Weight, Dry kg/m3 Ib/ft3 1180 73.7 1070 66.8 1180 73.6 1180 73.6 1180 73.6 1158 72.3 Modulus of pupture lb/in 2 11.4 162 11.9 169 13.55 193 16.50 235 11.35 161 12.94 184 Compressive Strength ka^cm2 lb/in 2 33.15 472 43.2 615 72.2 1,026 61.2 870 14.6 208 44.87 638 First Crack kg/cm 2 lb/in 2 12.3 175 14.7 209 15.7 223 30.3 431 9.9 141 16.58 236 Absorption {24 hours) % 37.8 31.4 36.35 30.8 38.8 35.03 Absorption (5 hrs. boiling) % 37.4 44.2 45.5 41 46 42.82 Coefficient of Saturation - ' 0.71 0.79 0.75 0.84 0.818 37 Director of the Laboratory Head of the Department of Physics Engineer Carlos Tru¡illo del Rio Engineer Francisco Nunez Farias Jfaotltali be ^ngetrierta fflitti! BiiÜm»Hiift Antóiu Mnboi* bi ^«to aUnbab »ittti raita ri« ABt¿ «•t. s-a. -3B 38 LABORATORIO DE RESISTENCIA DE MATERIALES FACULTAD DE IBIJEHIERIA D.A.G. Informe de las pruetas de Resistencia de tiveda de ladrillo de lama apoyada en viga de fierr». Datos de Csnstruocijn.- La tiéveda se construya con ladrillo de lama proaedente de "Tateposoo" con un claro de 1.30 mtB. y una flecha de 3.9 eme. (3^ olaro) sóbre- la tóveda ee colocó un bormigiSn de Cal y Jal hidratada- de un espesor promedio de 7 eme. y por último se coloca en la parte superior un enladrillado consistente de la- drille de harro cocido con dimensiones 20 Z 20 X 1 eme* Las dimensiones totales de las bóvedas fueron: 3*15 X - 2.69 mts. dando una superficie de 8,4735 M2. Se usa mor tero de Cal y arena amarilla en proposiciones comunes. Prueba.— Se carg6 la bóveda con sacos de arena de- río que se fueron pesando independientemente, y colocan do en la bóveda para dar una carga uniformemente repara* tida en toda la superficie; ee midieron las deformacio- nes de los arcos de bóveda así como las de las viguetas de apoyOf llegando a tener una carga total de 16.033 ^g y una deformación de 6.9& milimetros sin tener fallas — considerables en todo el elemento estructural. Resultados: Resistencia de la bóveda sin llegar a la ruptura. 1892.92 Kg8/B2. Deformación de toda estructura. 6.98 milimetrod. No hubo fallas de consideración. Resistencia del mortero 8.8 Eg8./cm2. a la compre- sión. Guadalajara, Jal. 10 da Diciembre de 1964. lio del Río. STRENGTH OF MATERIALS LABORATORY SCHOOL OF ENGINEERING U.A.G. Report of the strength tests of a vault of mud brick supported on steel beams. Construction data.- The vault was constructed of mud brick from "Toteposco" with a span of 1 ,30 meters (4 feet 3,2 inches) and a rise of 3.9 centimeters (1 .5 inches), 3% of the span. Over the vault was placed a concrete of Jal and hydrated lime of an overage thickness of 7 centi- meters (2.8 inches), and lost, on the upper part was placed o firm paving of burnt clay brick with dimensions 20 x 20 x 1 centimeters (7,9 x 7.9 x 0.4 inches) . The total dimensions of the vaults were: 3.15 x 2.69 meters (10 feet 4.0 inches x 8 feet 9.9 inches) giving on area of 8.4735 square meters (91 .2 square feet) . A mortar of lime and yellow sand in the usual proportions was used. Test.- The vault was loaded with sacks of river sand which were weighed independently and were placed on the vault in order to give a load uniformly distributed over the whole surface; the deformations of the arches of the vault were measured as well as those of the supporting beams, reaching a total load of 16,033 kilograms (35,350 pounds) and a deformation of 6.98 millimeters (0.27 inch) without appreciable damage in any structural element. Results: Strength of the vault without reoching failure 1,892.92 kilogroms/square meter (388 pounds/square foot) Deformation of the whole structure 6,98 millimeters (0.28 inch) There was no damage of importance. Compressive strength of the mortar 8.8 kilogroms/square centimeter (125 pounds/square inch) Guada a ara. Ja isco Decembe Director Engineer r 10, ofth Carl 1964 s Laboratory DS Trujillo de 1 Rio 39 ^aadtab be ^ngenieibt (Shiil Bnlmmtiafi Astáaana bi 9iutula|ai Jmui^mrab» ■ 1< Bafimvlteb Mnfaiual til ^áltc Cfaibob SniStnitarÍB ^ftiitóminu QUI. S-S5-3B Oficina da Planeacidn de Ciudad Universitaria Autánoma. Atte. Arn. iiansoni. Informe de laa pruebas de ííeaigtencia de TOateri^lea aolicitaria por esa H, úficina. 1 .- A.- Peso volumétrico del hormigán de Jal 'isado en bóvedas de la prueba reali- zada en Oicierabre prdxlnio pasado. IluálFICACIüK EN VCLtaiSN. JAL 2 40 AH.íFí \':AaiLiA 1 CAL 0.75 AG a A 0.50 Peso volumétrico a la edad de 10 días 938.89 Kgs/M^ 6 días 1047.60 Kgs/M^ B.- I.'ODULO DE HUPTURA. 2.535 Kgs/cm.^ C- RESIáTirCIA A LA COIirPRESION la. grieta; 3.52 Kgs/om 2 Ruptura: 4.35 Kgs/om. 2 2o.- l'eso volumétrico del ladrillo de ba rro recüoldo usado cono impermeaoilisante de bóvedas. P£30 VOLUt.-iíRICO. 1464.29 Kga/cm2 Office of Planning of the Autonomous University City Attention: Architect Ransom Report of the strength of materials tests requested by that Honorable Office. A.- Volumetric weight of the Jal concrete used in the vaults in the test performed last December. PROPORTION BY VOLUME lAI 2 YELLOW SAND 1 LIMF 0.7.S WATFR n.-io Volumetric weight at lOdays 938.85>kg./m3 {58.5 lb./ft.= ) 6days 1047.ó0kg./mJ (65.5 Ib./ft.M B. MODULUS OF RUPTURE 2.535 kg./cm.' (36.1 lb./!n? ) C. COMPRESSIVE STRENGTH Cracking: 3.52kg./cm; (50.0 Ib./in.' ) Rupture: 4.35 kg ./cm? (61 .9 Ib./in.M 2. - Volumetric weight of the burnt clay brick used to moke the voults impermeable. VOLUMETRIC WEIGHT 1 464. 29 kg ./m .3 (91.4 Ib./ft? ) 41 J)[acultab be ^tigenirrta flihril JnmHiorab» ■ !■ Buftiinlftaa flintisil VnOitnttaiÍM ^linamm Sfl. 5-95-36 3.- 3EFLEXI0N JO^AL DE LA BOV"£DA 6.98 mm. DEFLEXION DS LA VIGA DE ACERO 2,446 mm, DEFLEXION ti -ILATIVA 'JE LA "BÓVEDA. 6,98-2.446= 4.534 mm. PEHALTE DE LA VIGA - 10" CLARO DE LA VIGA - 2.67 Kts. 42 Atentamente " CIENCIA Y LIBEaTAD " Guadalajara, Jal., Abril 9 de 1965. Encargado del Laboratorio de Resistencia de Materiales. ING. CARLOS J«ÍJILL0 DEL IilO 3. - TOTAL DEFLECTION OF THE VAULT 6.98 mm. (0.28 Inch) DEFLECTION OF THE STEEL BEAM 2.446 mm. (0.10 inch) RELATIVE DEFLECTION OF THE VAULT 6.98 - 2.446 = 4.534 mm. (0. 18 Inch) DEPTH OF THE BEAM - 10 inches SPAN OF THE BEAM - 2.67 meters (8 feet 9.2 inches) Sincerely, "SCIENCE AND LIBERTY" Guadalajara, Jal., April 9, 1965 Director of the Strength of Materials Laboratory Engineer Carlos Trujillo del Rio 43 APPENDIX B CONVERSION FACTORS BETWEEN METRIC AND ENGLISH UNITS METRIC UNITS TO ENGLISH UNITS 1 meter = 39.37 inches 1 square meter = 10.76 square feet 1 kilogram = 2.205 pounds 1 kilogram/meter = 0.6721 pound/foot ^^ 1 killogram/square centimeter = 14.22 pounds/sq. inch 1 kilogram/square meter = 0.2049 pound/sq. foot 1 kilogram/cubic meter = 0.06243 pound/cu. foot ENGLISH UNITS TO METRIC UNITS 1 inch = 0.02540 meter 1 square foot = 0.09290 square meters 1 pound = 0.4536 kilogram 1 pound/foot = 1.488 kilogram/meter 1 pound/square inch = 0.07031 kilogram/square centimeter 1 pound/square foot = 4.883 kilograms/square meter 1 pound/cubic foot = 16.02 kilograms/cubic meter DESIGN EVOLUTION BY HARRY S. RANSOM ASSOCIATE PROFESSOR OF ARCHITECTURE LIST OF FIGURES FIGURE 1. One way growth system 52 FIGURE 2. View of ceiling of "La Casa de las Artesanias" in Guadalajara, showing doubly curved, two way vaults 53 FIGURE 3. Two way growth system 52 FIGURE 4. Forming of test structure, December 1965. Note small wood pieces used for forms. Plywood is quite expensive.. 54 FIGURE 5. Reinforcing being placed in the test structure, 3 modules by 5 modules, two end modules cantilevered 54 FIGURE 6. Concrete being placed 54 FIGURE 7. Diagonal and circular doubly curved vaults constructed for study purposes 54 FIGURE 8. Diagonal vault completed. Labor time; 3 hours, 45 minutes. Labor time for circular vault; 7 hours 55 FIGURE 9. Forms removed; the effect of the concrete coffered ceiling 55 FIGURE 10. The diagonal vault system. Electrical outlet box fully recessed in brick depth 55 FIGURE 11. Rectangular brick vault. Labor time; 1 hour, 10 minutes. This is the vaulting system that will be used in the new university buildings 55 FIGURE 12. Aerial view of model of the proposed campus 56 The preceding sections of this report technically document the structural capacity of a traditional regional building technique: the building of brick barrel vaults. In the architectural programming for all of the proposed new buildings for the Universidad Autónoma de Guadalajara it became evident that a system of architectural flexibility must be an integral part of the structural skeleton. It is imperative that, with relative ease, walls may be relocated, modules added, spaces rearranged. Versatility is essential. The barrel brick vault system satisfied these requirements to a workable degree. But it was recognized as basically a one-way growth system with the opportunity to expand or move in only one direction. For example, partitions can be only reasonably relocated at the supporting beams, framing in the same direction (Fig. 1). A two-way directional system was therefore desired so that expansibility could occur in either of two directions. Such a sys- tem can be seen today (Fig. 2) in one building in Guadalajara, "La Casa de las Artesanías." In this handsome structure two- way doubly curved brick vaults are employed, spanning approximately 16 feet and supported upon square steel-framed bays. But elaborate formwork was necessary to achieve this striking result. Consequently, a subsequent series of tests was undertaken by the architectural design group in Guadalajara to discover that dimension of supporting structural network that would permit the doubly curved brick vaults to be constructed with- out the use of forms; and to retain a low rise of the vault capable of inclusion within a normal ceiling-floor thickness (Fig. 3). Several trial-and-error, full-size tests were conducted; calcu- lations reviewed and refined; and visual details studied. To transcribe all of the testing procedures would make this investi- gation a purely text oriented writing. Suffice to say that the original tests acted as a springboard for the development of a more sophisticated method of building vaults of brick; in this instance, doubly curved brick vaults — built without the use of forms — spanning 5'-2^Ir" (1.60 meters) rising 3.9" (10 centimeters). These vaults are in turn supported upon an aggregate network of reinforced concrete beams and columns. The underside of the brick vaults will be exposed to take advantage of their rich color and textura! surface (Fig. 4-11). This evolved system then becomes the structural design determinant for all of the architecture of the new campus of the Universidad Autónoma de Guadalajara — a system rich in both the past and the present (Fig. 12). 51 FIGURE 1. One way growth system. FIGURE 2. View of ceiling of "La Casa de las Artesanías" in Guadalajara, í> showing doubly curved, two way vaults. FIGURE 3. Two way growth system. w\ mt,:» 'M- * mmmm0mm0mt0mmmmBÍmgmí0^m 'if > 4 54 FIGURE 4. Forming of test structure, December 1965. Note small wood pieces used for forms. Plywood is quite expensive. FIGURE 6. Concrete being placed. FIGURE 5. Reinforcing being placed in the test structure, 3 modules by 5 modules, two end modules cantilevered. FIGURE 7. Diagonal and circular doubly curved vaults constructed for study purposes. - ^-^. FIGURE 8. Diagonal vault completed. Labor time; 3 hours, 45 minutes. Labor time for circular vault; 7 hours. FIGURE 9. Forms removed; the effect of the concrete coffered ceiling FIGURE 10. The diagonal vault system. Electrical outlet box fully recessed in brick depth. FIGURE 11. Rectangular brick vault. Labor time; ^ hour, 10 minutes. This is the vaulting system that will be used in the new university buildings. 55 ^ >yy t » ;.vv.i FIGURE 12. Aerial view of model of the proposed campus. 56 ARCHITECTURE AT RICE SERIES , 1 ON PEOPLE AND THINGS, William W. Caudill, September 1961 2 UNITED NATIONS' CONFERENCE ON NEW SOURCES OF ENERGY, Paul Jacques Grillo, October 1961 3 RICE PRECEPTORSHIP PROGRAM, William W. Caudill, December 1961 4 ALVAR AALTO AND THE ARCHITECTURE OF FINLAND, Scott D. Hamilton, Jr., March 1962 5 THE ALUMNUS CRITIC PROGRAM, Morton L, Levy, Jr., May 1962 6 ARCHITECTURE FOR OUR TIMES, Howard E. Eilenberger, Author; L. James McCuIlar, Illustrator, June 1962 7 THE PEOPLE'S ARCHITECTS, William W. Caudill, March 1963 8 SKETCHES, Charles Schorre, Special Editor, April 1963 9 WILLIAM WARD WATKIN TRAVELING FELLOWSHIP WINNERS, Coryl LaRue Jones, May 1963 10 THREE CITIES, Paul Jacques Grillo, September 1963 11 THE AESTHETICS OF FOLDED PLATES, Clovis B. Heimsath, January 1964 12 AN EVALUATION— THE RICE PRECEPTORSHIP PROGRAM, Coryl LaRue Jones, April 1964 13 THE RICE DESIGN FETE, AN EXPERIMENT IN EXPERIENCE, Coryl LaRue Jones, Author, Maurice Miller, Photographer, June 1964 14 FOUR PLANNING CONCEPTS FOR BAY CITY, TEXAS, William T. Cannady & Architecture 300 Students, September 1964 15 THE CONCEPT OF PLASTIC FORM, Bill N. Lacy and Frank S. Kelly, April 1965 16 LAKE HOUSTON DEVELOPMENT STUDIES, William T. Cannady & Architecture 300 Students, August 1965 17 POBLACIÓN ALMIRANTE GOMEZ CARRENO, Andrew Belschner, February 1966 18 VAULTED BRICK CONSTRUCTION IN GUADALAJARA, Nat W. Krahl and Harry S. Ransom, June 1966 with a foreword by Carey Croneis, August 1966 Direct requests to Publications, School of Architecture, Rice University, Houston, Texas 77001 ©ARCHITECTURE AT RICE, 1966. All contents are the sole possession of the contributors; partial or total reproduction of the material herein contained is prohibited by law.