This action might not be possible to undo. Are you sure you want to continue?

# Y

OF

ISSUED

WEEKLY

21, 1918

October

Her December

11,

No. 8

August

1912, at the post office at Urbana, Illinois under the Act of f or mailing at the special rate of postage provided for in section 1103, ^ct of October 3, 1917, authorized July 31, 1918)

**ANALYSIS AND TESTS OP RIGIDLY CONNECTED REINFORCED CONCRETE FRAMES
**

BY

MIKISHI ABE

BULLETIN

No. 107

**ENGINEERING EXPERIMENT STATION
**

PUBLISHED nr THE UNIVERSITY OF ILLINOIS, URB/

PRICE: FIFTY CENTS

EUROPKAN AGENT CHAPMAN & HALL, LTD., LONDON

EXCHANGE

.

UNIVERSITY OF ILLINOIS ENGINEERING EXPERIMENT STATION BULLETIN No. 107 OCTOBER. URBANA . 1918 ANALYSIS AND TESTS OF RIGIDLY CONNECTED REINFORCED CONCRETE FRAMES BY MIKISHI ABE FORMERLY STUDENT IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS ENGINEERING EXPERIMENT STATION PUBLISHED BY THE UNIVERSITY OF ILLINOIS.

- ./V.

. 5.. . 12. . . . .. and ..CONTENTS I. . . . . 9... Moment . . .. 4.. ..". |? . Square Frame under Horizontal Load The Nature of the Resulting Formulas. Number 6.. .. .. 39 44 13. Notation Statically >.. .. Height of zontal 16. . . . 48 Effect of Variation Moment Moment Member Inertia on Bending 52 399192 . Member in Vertical . Relation between Horizontal Reactions in Frame under Uniform Load and under Concentrated Load 15. INTRODUCTION PAGE 9 Preliminary 2. . 45 Effect of Variation in Moment of Inertia and Relative in -. . 15 18 Principle of Least Work of . 20 25 Simple Frames under Vertical Load .. Frame on in Bending of . . Scope of Investigation and Acknowledgment II. . . Single Story Construction with Three Spans . I . Frame with Three Spans . 30 34 35 10. 14. . Trestle Bent with Tie .. Hori. .. . Building Construction with Several Stories and Spans . 12 THE ANALYSIS OF RIGIDLY CONNECTED FRAMES .. Systems . 11.. .. . . The Use and the Advantages of Reinforced Concrete Frame the Rigidly Connected 10 3. . . . .. 7. . 14 Determinate and Indeterminate . . 1. . .. The Effect of Work Normal Force on the Magnitude of Statically Indeterminate Forces 8.. .. of Statically Indeterminates . . . .

98 98 99 Continuity of the Composing Stresses at Corners Members of a Frame .. 19. 25. . 29.4 III. . 18. .. Columns . 96 96 Distribution of Stress over the Cross-Section Position of Point of Inflection in 27. 23. tfc c \ * * 'CONTENTS :(CONTINUED) TESTS OF RIGIDLY CONNECTED REINFORCED CONCRETE FRAMES PAGE 17... Test Specimens Materials 55 59 of Test Making and Storage Testing Frames 60 61 20. Conclusions and General Comments 100 . . 26. 21. Comparison Effect of of Test Results with Analyses 92 End Condition of Column on Results .. 30. 28. 22. Explanation of Tables and Diagrams 70 71 Phenomena of Frame Tests Comparison of Conditions Considered in the Analysis Tests with 91 24. .

Bending Moment 51 5 . . 49 Coefficient of Bending Moment at Center of Middle Span for Single Story Three-Span Frame Having Extreme Ends of Beams Hinged and Column Ends Fixed 26.41 42 Single Story Three-Span Frame with Four Columns Having Middle Span 21. 16. 1. Lower Ends of Columns Fixed Rectangular Frame with Rigidly Connected Tie at Base of Columns . . 36 36 Two-Story Three-Span Frame Having Middle Span Uniformly Loaded Coefficient of Bending Moments for Top of Lower Column and Foot of . for Two-Story Three-Span Frame Having All 38 External Connections Fixed Formulas for Single Story Three-Span Frame with Four Columns Having Columns Hinged 19. L-frame with Column Hinged at Support 29 . at Lower End Three-Span Frame with Four Columns Having . 14. Single Concentrated Load Middle Span for Single Story 47 Coefficient of Bending Moment at Center of Three-Span Frame Having Extreme Ends of Columns and Beams Hinged 25. 15.. . . . 4. 31 13.. at Lower End . L-frame with Column Fixed at Support . . 30 31 Continuous Single Story Spans. 44 Locus of Intersection of Reaction Lines in Single Span Single Story Frame under 24. 28 28 . 9. .. . . PAGE Simple Hinged Frame with Concentrated Load Illustrations of Degree of Indeterminateness by Means of Joint Bars 15 . 23. Uniformly Loaded Formulas for Rectangular Frame under Horizontal Load and for Octagonal and Rectangular Reservoirs Horizontal Reactions for Distributed Loads and Concentrated Loads . Frame under Concentrated Load 34 35 A-frame Showing Form of Moment Curve Frame with Single Panel Cross-Section of Building Loaded . 16 Types Degrees of Indeterminateness Type of Structure in which Effect of Normal Force is Great Unsymmetrical Frame under Concentrated Load of Frames of Various 17 .'.. .. 50 Coefficient of Middle Span for Single Story All Frame External Connections Fixed .' . Upper Column 18. 18 20 24 6. Simple Frames under Vertical Load. . . Having Middle Span Uniformly Loaded Single Story Three-Span . One Panel Loaded Single Story Three-Span Frame. 8.LIST OF FIGURES NO. . 3. . . 7.. 11. 43 22. 40 Formulas for Single Story Columns Fixed 20. Lower Ends of Columns Hinged Simple Frames under Vertical Load. 10. . 5. . 17. 2. . ... . 12. Three-Span Having at Center of .

.. Observed and Computed Stresses Observed and Computed Stresses Frame Frame 4 41...6 NO...56 56 57 57 61 62 63 64 65 66 ... 49.... .. 45. LIST OF FIGURES (CONTINUED) PAGE Coefficient of Bending Moment at Center of Middle Span for Single Story Three-Span Frame Having Lower Ends of Columns Hinged .. Frames 4......... 27. 31.. ..... ..... 40.... . and Showing Typical Location of First 98 Fillet 99 ...... .. of Reinforcing of Reinforcing Frame Having Lower Ends Bars in Frames 1 Columns 54 Size Size Size Size and Distribution and 6 . 36.. ... View View View of of of of of Frame Frame 5 51.... 35..... 6... and Distribution and Distribution and Distribution Bars in Frames 2 and 4 Bars in .. .. ...... after Test 1 after 1 in Testing .... 48..... .. Stresses ... 7 83 83 84 84 Test 2 after Test after Test after Test Frame 3 Frame 4 Frame 7 85 85 86 6 after Test after Test after 86 89 Frame 8 Test '..55 ... 53 29. Frames 3 Load-Deformation Diagrams for Additional Gage Lines............ 55.... . .. ...... Connection of Cracks 59.. 4..... Frame ... 2. Frames Gage Machine ... Frame ...... 37..... Frames Location of Additional Gage Lines.. 52 28............... 42. 43.. .. ... Load-Deformation Diagrams for Additional Gage Lines.. Coefficient of Bending Moment at Lower Ends of Interior Columns of for Single Story Three-Span Fixed 30... ..... .. Rectangular Joint with and without . Size and Distribution of Reinforcing Bars in Stress-Deformation Diagrams for Cylinders Observed and Computed Stresses Observed and Computed Stresses in Frame Frame 1 in in in 2 3 39. View View 54. 47..... 56.. 44.... Beam Column ..... ....... .. ..... . 67 68 69 76 Computed and Observed 1.... 38. . 32. - of 58.. .. . 33.. ......... 6... 46. 50...... 52....... Coefficient of Bending Moment at Center of Middle Span for Single Story Three-Span Frame Having Lower Ends of Columns Fixed . . ... of Reinforcing of Reinforcing Frames 3 and 7 8 Bars in Frame 5 34..... Location of Additional Lines........... 5..... 2... Frame .... 53. . View of of of View View View 57.. Frame ...... 1.............. 5.... Observed and Computed Stresses in Frame 5 Observed and Computed Stresses in Frame 6 Observed and Computed Stresses in Frame 7 Observed and Computed Stresses in Frame 8 Points Used for Comparison of ......... 3 78 79 80 81 7 .

11. . 6. 13. Formulas for Two-Story Three-Span Frame Bending Coefficients of Moment for Two-Story Three-Span Frame with ... of Columns and Beams Fixed . 12. 74. and Points for Single Story of Inflection for Single Story Three-Span Frames 4. 9. PAGE an Indeterminate 21 Elements Used Structure in Conditional Equations for Solution of 2. 33 37 5. 3. and Back Members Frames 1. 15. Ends 7.' . Computed Stresses for Test Frames in Terms of Total .42 Having Four Columns Data of Test Frames 58 . 75 95 97 100 Computed and Observed Stresses in Bars at Front Stresses in Frames of . 94.39 Coefficients of 8. . Bending Moments.. . of Test Frames . Applied Load . 32 Coefficients of Bending Moment Three-Span Frame . 1. Elements Used in Conditional Equations for Solution of Inverted U-frame 26 Formulas for Reactions. Observed Stresses at Sharp Corners in 3 and 6 . Compression Tests of Concrete Cubes and Cylinders Values of Modulus of Elasticity of Concrete Used in Stress Computations General Data of Tests of Frames 72. 16. Tension Tests of Reinforcing Steel 59 60 71 73 10..LIST OF TABLES NO. . Bending Moment for Single Story Three-Span Frame . 14..

In these structures connections are used between members and in many or most of are statically indeterminate. a arch is ple. There is also a tendency in America to use reinforced concrete frames for buildings and bridges. the Advantages of the Rigidly Connected Reinforced Since about 1905 reinforced concrete frame construc- tion has been extensively used in continental Europe. The Use and Concrete Frame. girders. . The field of the application of rigid frames is almost unlimited. culverts and sewers. balconies.10 ILLINOIS ENGINEERING test EXPERIMENT STATION and the The general rigidity at the places where The work is presented are joined has been investigated. In England also frame constructions of reinforced concrete viaducts and other structures have been built in recent years. tures. subway construction. in the thought that it will be helpful in bringing into wider use the have been made. and most bridge structures can be designed In highway bridges. slabs are all rigidly connected with each other. and therefore the design should be made as a rigidly connected frame. members principles applicable to the design of rigidly connected reinforced concrete constructions. and reservoirs and water tanks. beam and bent construction. and the building as a whole. In the design the requirements and the advantages of the frame are taken into account. spandrel-braced frequently used. 2. for columns. In such a case columns are rigidly connected to the arch ribs and to the superstructure. trestles of elements of such constructions as buildings. Box culverts and the box type of construction for subways give sections which are examples of the rigidly connected frame and which may not be rationally designed without a sufficient knowledge of rigid frames. rigid and viaducts. bridge struc- retaining walls. Arches. beams and the effect of this condition continental European countries it is most common to use frames in building constructions. for examas frames on a rigid analytical basis. towers. such as roofs. Bridge structures are in the field of the rigid frame. even though In is not fully considered in the design. them the bending moments It is clear that may every building construction of reinforced concrete be considered as a rigidly connected frame. Many examples can be found in the German texts and magazines. will secure safety The designing of trestles and viaducts as a frame and at the same time obtain the best proportioning of parts. for most reinforced concrete structures are composed It covers rigid frames.

In building and structural design. tion the element of rigidity is usually not fully taken advantage of. It is not known that an experi- be reliable close . will reliability of rigidly connected rein- It is evident that reinforced concrete frames if there is perfect continuity or complete rigidity of joint. agreement between theory and experiment. the rigidity of the connection of and structural point the members may be used. the fundamental underlie the static considerations can seldom be coming to a conclusion as to the forced concrete frames. These moments will suggest are modified by the relative thickness of walls and the other dimen- sions of the structure. The rigid frame is capable of exact design. The question is also naturally raised if the formulas deduced from the elastic work of deformation of a non-homogeneous material like reinforced concrete will hold good for such composite members with fair agreement. and a small effect of stresses of a secondary character. These things must be considered before conditions. the unknown negative bending moment due to a rigid connection of wall to wall or base to wall will exist at each corner. can thus be seen that most monolithic construction falls within the field of the rigid frame. One reason why some engineers hesitate to use concrete frames extensively is that they hardly believe in the continuity of the parts of the structure and doubt the effect of the rigidity of the connection. The bending moment for a beam is frequently taken as an assumed fraction of PI (where P is the load and I is the span) while bending moments at the ends and in the columns are disregarded entirely. A study of the rigid frame will assist in developing judgment for use in the design of such construction. expense The reinforced concrete frame is can be saved and a much better result obtained advantageous in that material from the theoretical In ordinary concrete building construcof view.REINFORCED CONCRETE FRAMES 11 In water tanks and reservoirs of a rectangular or a polygonal form. however. as is well known. With the concrete frame construction. Under actual assumptions which more than partially fulfilled even under carefully prepared specifications and well executed designs. furthermore the secondary stresses may act to modify the results. insufficient attention is often given to the bending of columns caused by the rigidity of connections. A knowledge of the rigid frame the proper It method of solution. and therefore the economical distribution of materials can be realized. thus the structure inadequate or making one part stronger at the leaving of the other.

three spans. In making these tests the purpose was to obtain experimental information along the following lines which have a bearing on the design of rigidly connected reinforced concrete frames : The amount and the distribution of stresses in the rein(1) forcement and in the concrete The continuity of the composing members of a frame (2) (3) (4) (5) The The The location of sections of critical stress reliability of a reinforced concrete frame applicability of the theoretical formulas in the design of frames. The experimental work was done as a research problem of the Engineering Experiment Station of the University of Illinois. (3) trestle bent with tie. In the design of the frames requirements not touched upon by the analyses referred to were provided for in a practical way. to whom. single span. and (3) rectangular reservoir or tank. 1914. acknowledgment is due for valuable suggestions and of The limits of space set for this bulletin will not permit publication even a small part of the details of the derivation of the formulas . and January. 3. The analyses and the results of the tests have been subjected to critical study and discussion. February. 1913. In order to put to practical test the reliability of these formulas for reinforced concrete structures. TALBOT. (2) octagonal reservoir or tank. and (5) bridge trestle.12 ILLINOIS ENGINEERING EXPERIMENT STATION mental study of this subject has before been made. and the deformations produced in the various parts of the members by the series of test loads were measured. single span. eight test frames designed according to the formulas found by the analyses were made. The work was under the charge of PROFESSOR ARTHUR N. For horizontal load the following cases have been analyzed: (1) single story. and were tested in January. of the staff. (4) building frame with several stories and several spans. In this bulletin formulas for several types of statically indeterminate structures which have been deduced by the use of the principle of least work are given. It may be expected then that careful experiments and investigation will give information which will help to settle these questions. For vertical load the following cases have been analyzed: (1) single story. (2) single story. 1914. The specimens were made in November and December. single span. Scope of Investigation and Acknowledgment. and March. as well as to other members aid.

calculations. and other data of the tests. in the case of the experimental work.REINFORCED CONCRETE FRAMES 13 found nor of the observations. of showing graphically the stresses that were observed at the principal loads. and of not including details of the data. . The original and the reduced data and more detailed work of the analyses are on file at the Laboratory of Applied Mechanics of the University of Illinois. Instead the plan has been followed generally of giving the formulas found from the analyses without the details of the derivation main and.

frame to length of span. of a uniformly distributed load. = total vertical height of frame. w = ratio of moment of inertia of horizontal member of vertical to that member. = moment of inertia in general. the ends of In the diagrams representing the forms of the frames analyzed. modulus of elasticity (considered as constant) of the of application of a concentrated load material. = total length of horizontal span of frame. p = intensity V = vertical f8 =unit 8 f' reaction. THE ANALYSIS OF RIGIDLY CONNECTED FRAMES The following notation is Notation. n =-j- = ratio of height of M = bending moment in general. H = horizontal reaction I h I acting at the end of a column. Numerical is suffixes are used for individual members when a frame composed of members of different sizes. 4. = unit stress in steel in compression. N = normal P=a force or stress section). from the right corner or axis of a frame to the point of application of a concentrated load on a top beam. used generally throughout the bulletin: A = area of cross-section of a member. stress in steel in tension. on a section (total internal force normal to the concentrated load. s = length of an inclined member. to indicate the members and moments use these letters as subscripts to which they apply as well as the point of .14 ILLINOIS ENGINEERING EXPERIMENT STATION II. distance from the left corner or axis of a frame to the point = distance E on a top beam. members are indicated by lower case letters and the symbols for the properties. fe =unit stress in concrete in compression. forces. A a b is also used as a coefficient to represent certain algebraic expressions.

= M bc bending moment at any point in be. 6: Ha I ab hbc l bc = horizontal reaction at a. or the algebraic sum of all vertical forces acting on a body (2) is equal to zero. = length of horizontal projection of be. X = 0. are (1) three. that is to say. The by the supports of the structure.REINFORCED CONCRETE FRAMES 15 The method and by reference examples application. other external forces are the reactions due to the loads. the vertical FIG. reactions are exerted and in order be determined from the statical conditions the total number of unknowns must not exceed three. The conditions of static equilibrium for any number of forces in a plane. A force is said to be statically determinate when its direction and magnitude and its point of application are known from the conditions of static equilibrium. The The loads to which structures may be subjected are always given. = moment of inertia of member vertical height of be. 1. 1. two unknown forces exist at the joint. Statically Determinate and Indeterminate Systems and Number of Statically Indeterminates. ab. SIMPLE HINGED FRAME WITH CONCENTRATED LOAD . is equal to zero. 5. as is generally well known. or the algebraic sum of all horizontal forces acting (3) M= on a body 0. The ordinary trusses without redundant members are always statically determinate if a frictionless pin is used at each joint and if in the determination the effect of the longitudinal deformation of members on the stresses is If a case in which two members meet at a joint is considered neglected. that they may as shown in Fig. of use will be made clear by the following to Fig. 7 = 0. or the algebraic sum of the is moments of all forces equal to zero.

In studying the behavior of statically determinate and statically indeterminate systems. They are entirely free to move horizontally and 0] of 2. necessary to have three joint bars at rigid joint between connection each members. and they are prevented from moving vertically. In Fig. and therefore this is a statias stated three statical conditions before.16 ILLINOIS ENGINEERING EXPERIMENT STATION and horizontal forces. that is. In Fig. An . but This are free to move horizontally and to rotate about a point A. 2 (f). ILLUSTRATIONS OF DEGREE OF INDETERMINATENESS BY JOINT BARS MEANS OF vertically and also are free in rotation. two members are connected by three joint-bars. and therefore the total number of unknown But each member will give forces due to the external force P is six. example as of this arrangement is the frictionless roller end of a cantilever shown in Fig. is an example. the conception of the connection of members by means of joint bars is a convenience. it is is To make a . If the structure is rigidly connected to the ground. 2 (b) by a single bar. the rotation of a member about A. The crown hinge of an arch. An example of this arrangement is a touching joint between the free ends of cantilever In Fig. 2 (d). the intersecting point of two bars. 2 (c) two members are connected beams as shown in Fig. may be called the arrangement having two freedoms in motion. 2 (e) two members are connected by two connecting bars and have only one freedom in motion. 2 (a) the two members are not connected. an example of this arrangement. FIG. accordingly it may be called the arrangement having three freedoms in motion. Fig. and 3S conditions of equilibrium must be set up to determine 3$ unknowns when the structure is composed of S members. and may be called a The restrained rigid connection which allows no freedom of motion. 2 (h). cally determinate system. Fig. 2 (g). In Fig. end of a cantilever beam.

REINFORCED CONCRETE FRAMES more conditions than 3S are required. 3. a+6-3*S = 36+3-30 = 9. Casee. a+6-3S = 6-f-12-9 = 9. a+6-3S = 18+18-21 = 15. TYPES OF FRAMES OF VARIOUS DEGREES OF INDETERMINATENESS Case a. a+6-3 = 6+4-9 = l. Such a system is called ra-fold statically indeterminate. . therefore statically deter- minate. Casec. 9-fold statically indeter- minate. Fig. FIG. a+6 35 = 5+4 9 = 0. 15-fold statically inde- terminate. and m additional equations of condition are necessary to determine these unknowns. and b be the number member to the ground. is necessary to make When the members in the structure are all rigidly connected to each stati- other. 3 gives a few examples. Caseb. 9-fold statically indeter- minate. then from the connect of joint bars needed to the existing 3S conditions the following relation the structure statically determinate. 17 Let a be the number of jointto member one bars needed to connect another. Cased. a always exceeds 3>S and therefore the case becomes a cally indeterminate system in which +b where m represents the number of the statically indeterminate forces. 1-fold statically indeter- minate.

the criterion stated does not apply. 4 illustrates such a case. 1 . Consequently in any member in which the change in length due to the direct stress has an appreciable effect on the stress developed.18 ILLINOIS ENGINEERING EXPERIMENT STATION stresses in the In general the change in length of a member due to the direct member will have an effect on the magnitude of the stresses developed. as is very slight in all ordinary shown in a later paragraph. effect is However. this forms of construction and has been neglected in stating the method of determining the degree of indeterminateness. Fig.

. P n in an elastic system. According to the theorem of Castigliano dW dPi J CM_ dM El dPi 7 HV SN + J EA dPi f + JfKS GA dS dP. however. the expression to shearing stress is for the internal work due w= where 2GA G expresses the shearing modulus of elasticity of a material. . the shearing stress is not uniform over the cross-sections. and 'K&dx 2GA~ where K is a for known the Therefore total factor for a specified form of the cross-section. fKS dS *Normal to a cross-section member. Since. f N dN . work of resistance. . .. normal stress. and shearing stress.. that there are n statically indeterminate forces PI. . ' ' The total internal work may be subdivided into the parts due to bending moment. the expression for the internal work due to shear is modified. Suppose . dW ap 5 CM dM N dN fKS dS dW dP n [M dM of the . in a member will The total work due to the bending moment be ~M2dx M -r2EI / 2EA For the total internal work due to a total normal stress* N on a section If shearing stress S is uniform over the cross-section.KEINFORCED CONCRETE FRAMES of the 19 work of resistance with respect to statically indeterminate forces which are so chosen that the forces themselves perform no work are equal to zero.

and therefore the horizontal reaction is the only indeterminate term which enters into the expression of the normal force. is that having a sloped column under a The frame shown in Fig. the horizontal reaction. unknown The solution of these equations will give the exact formulas for statically indeterminate quantities. El dM . Generally the normal force in any member contains these reactions as factors. and in the horizontal members only are the normal forces affected by When a frame zontal reactions a statically indeterminate force. . *Normal to a cross-section of the . namely. the columns are vertical. 7. . the horizontal reactions do not affect the normal forces in the columns. If. the vertical and horiand the bending moment at the fixed column ends are statically indeterminate. fKS dS dP is the second term will disappear when -^p dN equal to zero or. 5. in other words. is fixed at its column ends. when the normal force* does not contain any of the statically indeterminate quantities. The Effect of Work of cally Indeterminate Forces. Normal Force on the Magnitude of In the general equation Stati- J M_ fM. f N EA dN .20 ILLINOIS ENGINEERING EXPERIMENT STATION of condition as there are These furnish as many equations quantities. 5 is used to illustrate the method of FIG. the vertical reactions become statically determinate. From these statements it will be seen that the form of frame which will be largely affected by the normal force vertical load. at the same time. UNSYMMETRICAL FRAME UNDER CONCENTRATED LOAD member. But if a frame having a single span is symmetrical in form and in the manner of loading.

REINFORCED CONCRETE FRAMES 21 MM . t g ttj ttj \ I - 1 ' 02 P 1 a d bfl fi 1 i H s| 1 I 53 o o I 5 " 22- N N~J N a s s a o 2 2222 ent rtia .

'Jo Ji co V >cos0t b + Hsin6t \sinB I 2 secdz'dy = /] I i J .=P-KPoTV . In general the internal work due to shearing stress and the general equation becomes may be neg- ( J M EI dH LQ dx + ^ J E~A ~dH f N dN d dx - All necessary elements in forming this equation are arranged in Table 1. in which it is assumed that E i rh I i jj- (Vb r Jn l > tanev-Hy)(-y)8eceidy+jjjj-l ' (V^ Jo EI ^ 'A* y) secd^dy .' / n k f V I Jo cosd HsinO sine seed d L -i EA ! fsldx / T / CC . In this case is the only statically indeterminate force. Inserting these values in the general equation gives the following and / are constant: expression. .=(l-K)P b b K being used as a general coefficient. lected.22 ILLINOIS ENGINEERING EXPERIMENT STATION analysis and to bring out the effect of the work of the direct normal force on the magnitude of the statically indeterminate forces. H Taking the moment of all forces about b' V l-Pfa+lw) = b or V b and V .

in which the work of the normal force is fully counted: 7 "bc^bc C t 9/ _|_ ^ Jl'bc l'cc'i 7 72 l/ 7 cc' ^b'c'^b'c' Cf Qt>i*Ul Q7'^j /} .REINFORCED CONCRETE FRAMES 23 Integrating and simplifying this equation gives the following general expression for the statically indeterminate force H.

.24 ILLINOIS ENGINEERING EXPERIMENT STATION In the foregoing example it is seen that the final formula is very complicated by taking the direct force into consideration and that the effect of the internal work of all the direct stresses on the final much value for statically indeterminate stresses is inconsiderable when compared with that of the bending moment. (**&*- ^ . force making the analyses of the frames A& // A W> V . The work of the normal is therefore disregarded in treated in this bulletin. 6. SIMPLE FRAMES UNDER VERTICAL LOAD. ^/f//^ S FIG. LOWER ENDS OF COLUMNS HINGED .

uniform load. 6 are given forms The form at the left of the figure is the of a type of simple frame. As a specimen application of the method given. and a frame the statically indeterminate force is the horizontal reaction H. Formulas for H. are given in the figure for the three forms of frame and for the three loadings. 8. In Fig. For this forces at Knowing the horizontal reaction H. of the top beam for the For example. the of the vertical reaction V about the section considered. derived by analysis using the principle of least work. of least work. 6 is The minate force is H. of the inverted form U-frame. The direct stresses may be added algebraically after the statically indeterminate stresses are found. the bending moment and the any section of the frame may be determined by the ordinary analytical method. Three forms of vertical loading are presented a single concentrated load. all The column connections at the base are hinged and other joints are rigid. The effect of is normal forces being neglected. The effect of the so insignificant work of the deformation due to shear is generally when compared with that due to the bending that it may be entirely neglected without sensible error in the calculation of the internal work. two concentrated loads. and of a load with a moment arm moment p about the section. . the general equation of condition J IM_ El d# v All quantities necessary in forming the conditional equation for this case are arranged in Table 2. and the others are special cases general of this frame.REINFORCED CONCRETE FRAMES Attention is 25 internal even though the effect of the may be neglected in determining the reactions the direct stresses themselves can not be neglected when called to the fact that work of the direct stresses calculating the total stress hi any member. Simple Frames under Vertical Load. the general solution of the frame and loading shown of using the principle in statically indeter- the upper left-hand corner of Fig. 6 is equal to the algebraic sum of the moment of the horizontal reaction H moment the equal to the height of the frame. the bending moment at the middle frames with two concentrated loads shown in Fig.

26 ILLINOIS ENGINEERING EXPERIMENT STATION 2 TABLE ELEMENTS USED IN CONDITIONAL EQUATIONS FOR SOLUTION OF INVERTED U-FRAME Member .

the considered. moment of the vertical reaction of V about the section Fig..7 ends of the columns fixed. (5) J^ Substituting l bc sec 6 = Sbc and h bc = Ibc ' - tan d .. 7 is equal to the algebraic sum of the a the moment of bending moment at the lower end of the column. H the frame. as determined from analyses. (6) 2^bc ... h Fig. 7 and the moment one load p 2 about the section. the horizontal reaction with a moment arm equal to the height of M . .~2^~z\ ' -P a . \ i &bc 6c I '^ab . . and the bending moment at the lower end of the column.. . and the forces at any section of the frame may be determined by ordinary analysis. f^bc c - Sl c r 2hl b Mice' r~T^\.REINFORCED CONCRETE FRAMES Collecting the terms involving 27 H and those involving P f ' i-L H 2 ab h" ^ ^ 9T " Ll 2sece Pnih ab II 2hP l cc '+- 1 CC . . the horizontal reaction. gives sketches of the inverted U-frame having the lower Equations of the statically indeterminates. indicates the manner in which the moment varies along the members composing the frame. Knowing these indeterminates. Thus the moment at the middle of the top beam for the frame in the upper left-hand corner of Fig. . . Solving for H. the moments. . are given in the figure.

RECTANGULAR FRAME WITH RIGIDLY CONNECTED TIE AT BASE OP COLUMNS Fig. 8 the horizontal cross shows a rectangular frame in which all the joints are rigid. The equations of the indeterminates M b and H b .28 ILLINOIS ENGINEERING EXPERIMENT STATION . 7. SIMPLE FRAMES UNDER VERTICAL LOAD. 8. LOWER ENDS OF COLUMNS FIXED hlcc - - + 2 hlee - +9$* FIG. tie at the bottom of the frame having rigid connec- tion to the columns.

the wall. and The values of b or M. which is also given in the figure. Fig. For under uniform load and with the column hinged at the support. 9. FlG. The maximum positive bending moment in this frame occurs at the distance x from c where 3+ be figure. On account of the fixity of M . Vb. here termed an L-frame. this frame the horizontal and vertical reactions Hb and Vb at the column end are given in the figure. Hb.-. The figure indicates the manner which the moment varies along the members. The value of the maximum positive bending moment is given in the The maximum bending moment is the negative moment at is fixed at the supthe column there are three statically indeterminates for this frame. 10 shows an L-frame in which the column port. L-FRAME WITH COLUMN HlNGED AT SUPPORT Fig.REINFORCED CONCRETE FRAMES for this in 29 frame are given in the figure. 9 shows an unsymmetrical frame. these are given in the figure.

9. 9 and 10. a beam-and-girder or a flat-slab construction of a single story. Authors tried to analyze the stress distribution without taking this bending into account. In the design of Single Story Construction with Three Spans. L-FRAME WITH COLUMN FlXED AT SUPPORT The formulas for the the distance of the section of maximum positive moment in the beam and maximum positive moment from c are Mmax = pos _IQ_ 1 2 1+ ( ll 24 III) + T 1 + mem- The manner in bers composing the frames which the bending moment varies along the is indicated in Fig.30 ILLINOIS ENGINEERING EXPERIMENT STATION FlG. many engineers do not take the effect of the bending of columns on the moments also in other portions of the structure into consideration. and stresses in the slab will be modified by variations in the ratio of the moment of inertia of the girder or slab to that of the column and of the column . Obviously a bending in the columns will allow an increased bending moment at the center of the span of a girder or have slab loaded unsymmetrically with respect to the column. 10.

and consequently an exact analysis is hardly possible with any assumption.REINFORCED CONCRETE FRAMES 31 that columns may be height to the span length. with different span lengths and different cross-sections of members. subjected for others. there may be twenty or more spans in succession. . Tests have shown this will be more that seen be it will and severe to bending. than structure a for single story important In actual cases.

32 ILLINOIS ENGINEERING EXPERIMENT STATION panel. AND POINTS OP INFLECTION FOR SINGLE STORY THREE-SPAN FRAMES . conditions of beams and columns outside the loaded The formulas which have been derived for these four cases are given in Table 3. TABLE 3 FORMULAS FOR REACTIONS. BENDING MOMENTS.

at top of column.REINFORCED CONCRETE FRAMES TABLE 4 COEFFICIENTS OF BENDING 33 MOMENT FOR SINGLE STORY THREE-SPAN FRAME moment moment at end of middle span. a = coefficient /3 = of pi2 for bending coefficient of pi 2 for bending .

this frame becomes the same as 0. SINGLE STORY THREE-SPAN FRAME UNDER CONCENTRATED LOAD For a three-span frame with concentrated load at the center of the middle panel. c . /bc = 0. c are given in the figure. and l bc approaches = 1).34 ILLINOIS ENGINEERING EXPERIMENT STATION 3+4 lite // I =f CC ' these formu/as become: Vf- ftlce' FIG. 8 and the formulas for cc c reduce to the M > H same form as those given in Fig. When = 0. and values of t M H cc >. 13. making h cd =h (that is when & and that shown in Fig. . 13. Assuming symmetry about the vertical center line of this frame the formulas for the horizontal and vertical reactions are given in Fig. 13 shows the manner in which the moment varies along the several members composing the frame. 8. fre- quently termed the A-frame. ' 10. and also formulas for bending moments at top of column and at end of middle span. Fig. and Hb. Formulas of the for M ce Knowing the Hb. H may trestle construction. Trestle The frame shown be used in in Fig. Bent with Tie. 14. the stresses at any section frame may be computed.

qKhJ FIG. cd. 14. such as is shown in Fig. A-FRAME SHOWING FORM OF MOMENT CURVE 11. Building Construction with Several Stories and Spans. In the actual construction of buildings of reinforced concrete. of course. ?2cc' Ice' * . the bending. stories. 3Icd\ * . may be approximately true for the lower . loading which produces serious bending in columns is. frequently little attention is and columns are assumed to be This assumption rigid enough to resist paid to this point. and c'd'. it is common to use a continuous slab for floors supported by a number of columns. is greatly modified In present practice. The moments of inertia of columns are sometimes smaller than those of slabs. an eccentric arrangement. b'c'. Ac- A by the cordingly. the bending moment in the floor slab fcc'f flexure of columns be.REINFORCED CONCRETE FRAMES 35 e -K) . 15.

it is easily understood that the (see Fig. and c'd' are practically fixed at 6. cd. bending moment in floor slabs. and d'. c'b'. 15) and eWe'. but it is not true for the upper stories. 16. TWO-STORY THREE-SPAN FRAME HAVING MIDDLE SPAN UNIFORMLY LOADED . and serious bending stress may exist in the column due to eccentric loading. &'. If the floor slabs are not thick enough to keep the FIG. where the cross-section of columns is usually small. d.36 ILLINOIS ENGINEERING EXPERIMENT ST ATION b' e' FIG. An exact analysis is hardly possible because there are many unknown From a practical standpoint. gdd'g' due to the load on the floor cc' is so small as That is. to be inconsiderable if the floors are of moderate thickness. respectively. 15. conditions entering into the solution. the columns c6. CROSS-SECTION OF BUILDING FRAME WITH SINGLE PANEL LOADED where the columns have large diameters.

CASE fO) SPECIAL CA5(c) Horizontal >" 'reaction at b Pic H* Horizontal reaction atd P* PJ./K ' 'omen<t at ef co/ ends dtd' eL* 84 plc at Mf .ENDS HINGCD GENERAL. M.Vertical reaction offer PjUc- 4 IB Vertical reaction afbeb' M. Moment at " 1 Mt M3 t ZM. ~ Vertical reaction at f .Moment beam endsftf * Horizontal y pi 2f ^ f^clfc f^ff lcc\ Ifcltc f reaction atbtti 2 hcjlttc %sL* f tec-Ike' tyclcc- } * f rforizonta/ reaction atdid' Y * ' .Bending moment at 3 Sending M4 moment at 4 . 44 Pl 44 .Vertical & 'reaction at b pi' M ^Bending moment at I ~ ' e ^Bending ~ moment at 2 pi PJ' 44 M3 . M4 ' -3 -4 .REINFORCED CONCRETE FRAMES TABLE 5 37 FORMULAS FOR TWO-STORY THREE-SPAN FRAMES _#/*! *s v f . M3 FIXED (a) GENERAL CASC SPECIAL CA6E(b) Ice" I** IIcc-I/c SPECIAL CAS(C) **co/ ends b^l cctlbc PJ' 64 tec' I be .

For such a frame. 17. a. . Fig. will 2 Fig. the end condition of the columns be between the hinged and the fixed state. 17 gives numerical values of the coefficients of pi for the case in which ends of beams and columns are fixed. and moments of inertia of beams and columns. being the coefficient of the bending moment at the top of the lower column and & the coeffiIn Table 6 are given value? cient at the foot of the upper column. the end spans being equal in both cases. Using these assumpThe resulting tions an analysis which is almost exact is possible.38 ILLINOIS ENGINEERING EXPERIMENT STATION column ends in a fixed condition. formulas may be used in the design of buildings. Formulas are also given for equal spans and equal moments of inertia of the beams and for equal spans. story heights. COEFFICIENT OF BENDING MOMENTS FOR TOP OF LOWER COLUMN AND FOOT OF UPPER COLUMN FOR TWO-STORY THREE-SPAN FRAME HAVING ALL EXTERNAL CONNECTIONS FIXED .0/0 JO FIG. The formulas for the horizontal and vertical reactions and the bending moments are given in Table 5 for ends of columns and beams hinged and for ends of columns and beams fixed.035 . 16 shows the manner in which the moment varies along the members composing the frame for two combinations of end conditions of beams and columns.

.REINFOKCED CONCRETE FRAMES TABLE 6 COEFFICIENTS OF BENDING 39 MOMENT FOR TWO-STORY THREE-SPAN FRAME WITH ENDS OF COLUMNS AND BEAMS FIXED a = coefficient of pP for bending moment at top of lower column. = coefficient of pi2 for bending moment at foot of upper column.

.40 ILLINOIS ENGINEEHING EXPERIMENT STATION ^ 14* X ^ s? \ ^ e ^ K B > x * n H s ZA FORMULAS U >.

REINFORCED CONCRETE FRAMHS 41 .

at top of intermediate column.42 ILLINOIS ENGINEERING EXPERIMENT STATION FIG. SINGLE STORY THREE-SPAN FRAME WITH FOUR COLUMNS HAVING MIDDLE SPAN UNIFORMLY LOADED TABLE COEFFICIENTS OF BENDING 7 MOMENT FOR moment moment SINGLE STORY THREE-SPAN FRAME HAVING FOUR COLUMNS a = coefficient /3 of of = coefficient pP pP for for bending bending at end of middle span. 20. .

A* M A .REINFORCED CONCKETE FRAMES 43 M A.

44 ILLINOIS ENGINEERING EXPERIMENT STATION at the top of the middle columns and at the end of the middle span for the case in which the three spans are equal. stresses frequently necessary to solve for the statically indeterminate due to a horizontal force. the moments of inertia of the beams are equal. 13. a uniform load being applied over the middle span. Hitherto only the cases in which the load It is was applied vertically have been discussed. Square Frame under Horizontal Load. such as a wind pressure or the braking Horizontal Thrust. and the moments of inertia of the columns are equal.ri riorizonto/ Thrust -H A (b) h n(3+2mn) 2m n) 6 Pi n(?+mn) b' <fer rr: .

**REINFORCED CONCRETE FRAMES
**

force of a locomotive.

45

The method

of determination of the statically

imknbwns

is

the same as that used for frames with vertical loads.

A

**few cases have been taken as illustrations and the resulting equations
**

are given in Fig. 21.

Another application

is

the water tank or reservoir subject to the

static pressure of water, such as may be found in filter plants. Fig. 21 this character. It is of framed constructions of gives two examples the the walls on the four sides tank same a seen that for having square

negative

moment from

**the formula for rectangular tank becomes
**

pP, as

is

2

-rjr-

L&

pi

and the positive moment

known from

other sources.

14.

The Nature of

It is

the Resulting

Formulas; Relation between Hori-

zontal Reactions in

Load.

Frame under Uniform Load and under Concentrated interesting and important to note from the results of the

foregoing analysis that there is a fixed relation between the horizontal reactions in the symmetrical frame under distributed loads and those

in the

same frame under concentrated

loads.

To show

this relation

**a few cases have been selected as illustrative.
**

Fig. 22.

These are shown in

It has been stated previously that there is also a fixed relation between the horizontal thrust and the bending moment at the fixed column or beam ends, and the bending moment can be expressed in terms of the horizontal thrust. The bending moment at any section of a frame is a function of the horizontal thrust. Therefore, it may be

stated that the statically indeterminate stresses in the symmetrical frame have a fixed relation under distributed and concentrated loads.

From

loads

the foregoing illustrations

it will

be seen that the horizontal

thrusts at the

column ends due to uniform and centrally concentrated

may

be expressed in the following forms:

Uniform

load,

H=K

PI H = K-$o

Centrally concentrated load,

The

coefficient

K

is

of frame.

The formula

the same in both cases, but varies with the form for the horizontal thrust in the frame under

concentrated load

may

be written directly

if

the formula for thrust

46

ILLINOIS ENGINEERING

EXPERIMENT STATION

An analysis of the statically in the frame under uniform load is known. indeterminate forces for a given case should first be made to find the form of the function. The bending moment at the end of the span in these frames is:

For Case

a,

M=

ForCaseb,

M=9

g

PI PJ

"

For Case

b',

2+mn

8

8

It is known that when a beam is perfectly fixed at its ends the negative bending moments due to a distributed load and a centrally concentrated 2 PI vl

load are

*JL.

12

and-, 8

respectively.

It

is

seen, therefore, that for these

cases the bending

value of the end bending

moment at the end of the beam is moment of a fixed beam by

**obtained from the
**

multiplying

by K,

which depends upon the form of the frame, but is independent of whether the load is applied uniformly or is concentrated

a

coefficient

at the center of the span. Returning to the nature of the formulas for the horizontal thrust

column end of a frame, it is further seen from Fig. 22 that the given constant relation between the values of the horizontal thrusts for a frame under a distributed load and under a concentrated load

at the lowr er

still

holds for the case in which a frame

is

metrical load.

**These simple frames are
**

remains constant and

sufficient

subjected to a non-symto illustrate the

general relation.

coefficient

It appears, therefore, that for the

same frame the

independent of the method of This statement can easily be extended to the case of multiple loading. concentrated loads, for then the horizontal thrust is the sum of the It will be horizontal thrusts due to the individual concentrated loads.

K

found that this statement applies also to the non-symmetrical frames of Cases d and d'.

REINFOKCED CONCRETE FRAMES

47

FIG. 23.

**Locus OF INTERSECTION OF REACTION LINES IN SINGLE SPAN SINGLE STORY FRAME UNDER SINGLE CONCENTRATED LOAD
**

it will,

**For a concentrated load,
**

,

be

of interest to find the locus of

y the point of intersection of the lines of action of the reactions with In a complicated form of the line of action of the load (see Fig. 23).

frame, there are, of course, is an important one. but

many

**statically indeterminate quantities,
**

statically indeterminate s

it is

H

The remaining

have always the same factor in the denominator as H. Therefore, It is very interesting to know the form of the expression for H.

evident that

H H

(Fig. 23)

is

a function of

I,

h, I,

and

P

hi a given case.

Since the

moments

at b

polygon for the load P and Vb about the point o moments of

and &' (Fig. 23) are zero, the equilibrium must pass through these points. Taking the

V

y

b lco>

H

and

l co '

H

and Vb are known

in this case

when

P

are given,

and

J_Y

h)

V,=

fP

15.0 and Ibc = 1.48 Therefore ILLINOIS ENGINEERING EXPERIMENT STATION _ Ico'lo'c' P or This equation is entirely free from lco'> kv. if it is remembered that The equation for loads which gives the ?/ permits the determination of the position of when a column is fixed at its end the point of application of the reaction deviates from the neutral line of the is column by where ~^A Vb M b the end moment and Vb is the vertical reaction at that point. Fig. 24.0. Accordingly the locus of the point o is a straight line parallel to W. and 28 give bending moment coefficients for the beam of the of central span for several cases of a three-span frame in which the spans moments of inertia of the three beams are equal. y is a constant quantity for a given frame and is not changed by the change of the point of application of a load P. the moment inertia of caused by variation in the relative values of the moments of members and in heights of frames is shown in these figures. For 3 Icc ' = Ibc and _i. maximum reaction and stress in any member. 25. The effect on the bending are equal. The same method may be extended to any case. and the moments of inertia of the columns are equal. 26. and P. Effect of Variation in Moment of Inertia and Relative Height Frame on Bending Moment in Horizontal Member. moment For a general comparison it is only necessary to consider the bending at the center of the span for load applied eccentrically with the columns. therefore. ^. y = ^L. since the effect on moments at other places to respect . 27. I In the case in which ^1 = 1.o. .

24. .040 /^ ZO 3. but beyond that range the increase in bending moment comparatively small. From the general nature of the curves shown the following conclusions are drawn: of ~ J-bc (1) The bending moment from is is increased rapidly as the value increases to 1. COEFFICIENT OF BENDING MOMENT AT CENTER OF MIDDLE SPAN FOR SINGLE STORY THREE-SPAN FRAME HAVING EXTREME ENDS OF COLUMNS AND BEAMS HINGED and on thrusts will be similar.0 fa/ues of &' FIG.5.REINFORCED CONCRETE FRAMES 49 .

O FIG.070 3. . AND COLUMN ENDS FIXED of An increase in the height of the frame has an effect (2) the same nature on the bending moment as an increase in the ratio (3) -Ibc in the variation in coefficient of bending moment is wider frame hinged at ends of columns and beams than in the case The of fixed ends. (4) By coefficient of positive the fixing of ends of columns and end beams the bending moment is slightly decreased from that for hinged ends.50 ILLINOIS ENGINEERING EXPERIMENT STATION . COEFFICIENT OF BENDING MOMENT AT CENTER OF MIDDLE SPAN FOR SINGLE STORY THREE-SPAN FRAME HAVING EXTREME ENDS OF BEAMS HINGED . 25.

COEFFICIENT OF BENDING MOMENT AT CENTER OF MIDDLE SPAN FOR SINGLE STORY THREE-SPAN FRAME HAVING ALL EXTERNAL CONNECTIONS FIXED (5) In most common cases of panels under uniform load where the ratio the bending r.5 to 3.is not far from 1.assumed as JLo .&*! 60 .KEINFORCED CONCRETE FRAMES 51 . and may be conveniently . 26. moment at the center of the loaded span (case of equal spans) varies from about TTr to about -rrpl2 .065 ..0 and -~ J-bc varies from 1.060 \ ?$ f __ 7 5l/_J< h %k&> Values o FIG.0.

27. 6 and 7. AT CENTER OF MIDDLE SPAN FO: SINGLE STORY THREE-SPAN FRAME HAVING LOWER ENDS OF COLUMNS COEFFICIENT OF BENDING MOMENT HINGED 16. . stor. 17 and 29 and in Tables In Table 6 the three spans are taken as equal and the height is taken equal to half the span. Effect of Variation in Moment of Inertia on Bending Momen in Vertical Member.52 ILLINOIS ENGINEERING EXPERIMENT STATION . The variation in bending moments in columi ends due to the variation in properties of the members for several case is shown in Fig.040 . of a three-span frame 4. FIG.

higher bending stress will exist at the top of the . 17. COEFFICIENT OF BENDING MOMENT AT CENTER OF MIDDLE SPAN FOR SINGLE STORY THREE-SPAN FRAME HAVING LOWER ENDS OF COLUMNS FIXED Values of coefficients of pi for various values of and are y^' It is seen from the diagram that for structures plotted in Fig. having the relations between span lengths and moments of inertia 2 assumed in Table 6. 28.0 FIG.REINFORCED CONCRETE FRAMES 53 3.

COEFFICIENT OF BENDING MOMENT AT LOWER ENDS OF INTERIOR COLUMNS FOR SINGLE STORY THREE-SPAN FRAME HAVING LOWER ENDS OF COLUMNS FIXED lower column than at the foot of the upper column and that the variations in moments of inertia assumed cause less variation in the moment in the upper columns than in the lower columns.030 .005 3.54 ILLINOIS ENGINEERING EXPERIMENT STATION . . 29.O FIG.

Several bars were welded and these welds were located at points where the bending moment was very small. TESTS ON RIGIDLY CONNECTED REINFORCED CONCRETE FRAMES 17. 30. Test Specimens. The length of span of the frames was 6 ft. 8. Bars continuous from one end to another were used for all frames. Care was taken in designing the test specimens to secure continuity of connected members and to obtain such proportions between moments of inertia and spans as would result in high bending stresses in the columns and beams at nearly the same time. on Frame No. The ends of the steel reinforcing bars were bent into hooks in the specimens having columns fixed at the ends. The size and disposition of the reinforcing bars and the dimensions of the frames Data of the frames are given in Table 8. The the frames varied from about 5 ft. in. The cross-section of the Five types of frames were selected for the composing members varied from 8 by 8 to 81/2 centers except height of by 17% in. are shown in Fig. 30 to 34. FIG. The radius of bends of the main rods was about 5 in. to about 10 ft. U-shaped or double U-shaped stirrups were used.REINFORCED CONCRETE FRAMES 55 III. The size and spacing of the stirrups are given in Table 8. 8 in. They passed under the longitudinal bars and extended to the top of the beam. tests. SIZE AND DISTRIBUTION OF REINFORCING BARS IN FRAMES 1 AND 6 . In the frames with stirrups. which had three spans of 4 ft.

31.CC !&m*t All steel i ^r IN FIG.56 ILLINOIS ENGINEERING EXPERIMENT STATION -n j_rnj^Tff_ s***\ TTTT7 - FIG. SIZE AND DISTRIBUTION OF REINFORCING BARS IN FRAMES 2 AND 4 ^ 5ect. 32.A-A SHI 5ect. SIZE AND DISTRIBUTION OP REINFORCING BARS FRAMES 3 AND 7 .

34. SIZE AND DISTRIBUTION OF REINFORCING BARS IN FRAME 8 . 33. SIZE AND DISTRIBUTION OF REINFORCING BARS IN FRAME 5 FIG.REINFORCED CONCRETE FRAMES 57 FIG.

s.a isa 1-fflrf I CQ I ? 05 n_ a 444. .s.4.4.4 8^ ! I fl g.s U -9 00 00 00 00 00 00 00 00 o 1-s oooooooooo 'x'x'x'x x'x'x x OOOOOOOOOOOOOOQO bC c o . 3 O 05 3s S o CO & 3 SJ I-H >> <N co ^ 10 <r> i> oo .58 ILLINOIS ENGINEERING EXPERIMENT STATION I*.s.g.4.2 |t|.s.4.

Table 9 gives the results of the tension tests of the steel. A good quality of crushed limestone ordered to pass over a J^-inch sieve was used. sharp. Materials. The sand was of good and generally clean. and through a 1-inch sieve quality. stone. TABLE 9 TENSION TESTS OF REINFORCING STEEL Nominal Diameter inches . hard. The sand. steel. were similar to those ordinarily used in reinforced concrete construction. and cement were taken from the stock of the Laboratory of Applied Mechanics. well-graded.KEINFORCED CONCRETE FRAMES 59 The materials used in making the test frames 18. The reinforcing bars were plain round rods of open hearth mild Test pieces were taken from the test frames after the test.

.60 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE 10 COMPRESSION TESTS OF CONCRETE CUBES AND CYLINDERS Frame No.

the load being applied through a number of spiral springs. per FIG.000-lb. In Frame 8 in order to see the effect of the eccentric load on the adjacent spans and at the same time to produce high bending stresses in the middle beam and in the central columns. 36 to 43. 35. 20. a uniform load on the middle span was selected. room ranged from 55 to 70 degrees F. versity of Illinois. STRESS-DEFORMATION DIAGRAMS FOR CYLINDERS The Fig. zsoo Unit Deformafion. Riehle of testing machine in the Laboratory Applied Mechanics of the Uni- The Deflections were read on some of the frames.REINFORCED CONCRETE FRAMES 61 of dampened occasionally after that time. in.in. . . positions of the loads for the different frames are shown in The specimens were tested in the 600. was used to develop as high a flexural stress in the columns as possible. The temperature the Testing. To develop high stresses in the beam and in the columns nearly at the same time one-third point loadings were used In Frame 5 the centrally concentrated load for many of the frames. deformations of the steel and of the concrete were measured at the various parts of the frames for each load applied.

Computed stress curve stee/ for or concrete. front. 36. o /ooo zooo 4ooo Stress /n 5ree/ -/b. //?.per 50. - \ * * * bacff. OBSERVED AND COMPUTED STRESSES IN FRAME 1 . O 2OOOO 4OOOO FIG. " concrete. Stress in Concrete -/b.62 ILLINOIS ENGINEERING EXPERIMENT STATION Observed stress in " -o steel.

REINFORCED CONCRETE FRAMES 63 Observed stress in steef. FIG. front * * * * * concrete. OBSERVED AND COMPUTED STRESSES IN FRAME 2 . I Computed stress carve for stee/ or concrete. 37.

64 ILLINOIS ENGINEERING EXPERIMENT STATION .

Stress in Concrete-Jb. 39. in. FIG. front. " " * back. $0000 40000 per sq. Computed stress curve for stee/ or concrete. / P=3000Olb./b. concrete. OBSERVED AND COMPUTED STRESSES IN FRAME 4 .REINFORCED CONCRETE FRAMES 65 Observed stress in steel.persq.L o jooo eooo 4000 Stress in Steel .

.

41. 40000 FIG. 1000 EO 00 4000 Stress /n SteeJ.per sq.fb. front. Computed stress curve for stee/ or concrete. concrete. \ 7 O o Concrete. OBSERVED AND COMPUTED STRESSES IN FRAME 6 ./b. &ooo in.REINFOECBD CONCRETE FRAMES 67 P*30QOO/b Observedstress in steel. " " " " " " back.persq.

68 ILLINOIS ENGINEERING EXPERIMENT STATION Observed stress in steel. front Stress in Concrete o 1000 ^ooo - /b. per SQ. 42. in. Stress in Steel -/b. in 4ooo so. per o ^oooo 40060 FIG. OBSERVED AND COMPUTED STRESSES IN FRAME 7 .

KEINFORCED CONCRETE FRAMES 69 .

The load at first crack is the load noted when the first fine crack was observed during the test. where the concrete Small gage holes. the greatest number.055 in. The The faces of the test frames were whitewashed to enable the appearance of cracks growth to be more easily observed. The smallest number of gage lines on any frame was 75 on Frame 1 . long. deformations were to be measured. temperature corrections were made. These were based on an assumed linear variation of length with time between successive readings on the increment of load. were drilled in the reinforcing bars and in the steel plugsTwo sets of initial readings were taken before the application of loadcomplete set of observations of the deformations was taken at each A In reducing the strain gage readings to stress. 4 in. 163. The average deformation over the gage length was used. and 8 in. v represents the vertical shearing unit-stress in the concrete. 21.70 ILLINOIS ENGINEERING EXPERIMENT STATION deformations." Variation in temperature is sufficient to cause an appre- and Hence observations change in the length of the instrument. b the breadth of the V the total vertical shear beam. on an unstressed standard bar of invar steel were taken for the purpose ciable of making temperature were corrections. The tensile and compressive stresses are the highest stresses observed at the points specified. The gage lengths used were 2 in. in diameter. and jd the distance . Explanation of Tables and Diagrams. ' Extensometers of the Berry type were used in measuring the The method of using these instruments is described in Bulletin 64 of the Engineering Experiment Station of the University 'Tests of Reinforced Concrete Buildings under Load. extent of the cracks at the several loads was marked on the speci- and their men during the test. 0.. The loads given in the various tables and figures are the loads applied by the testing machine and do not include the weight of the frame itself. about one inch in the concrete. at the end of the beam. set in plaster of paris Small steel plugs." of Illinois. in a paper in Proceedings of the American Society for Testing Materials for 1913. "The Use of the Strain Gage in the Testing of Materials. The ultimate load is the highest load applied to the maximum specimen just before the load carried began to decrease slowly.. The vertical shearing stress was calculated with the ordinary formula v = where r-r-j. on Frame 8. standard bar.

m Loads are given in pounds.REINFORCED CONCRETE FRAMES from the center of the in the concrete. and o surface of the reinforcing steel. TABLE 11 VALUES OF MODULUS OF ELASTICITY OF CONCRETE USED IN STRESS COMPUTATIONS Frame . of the steel to 71 the center of the compressive stresses The bond rv stress in the is beam was computed by means formula u = where u the bond stress per unit of area of the the number of reinforcing bars. of reinforcing steel and the amount to reference with selected were jd the modulus of elasticity of the concrete. unit-stresses in pounds per square inch. The values of the circumference or periphery of one reinforcing bar. and moments in inch-pounds.

^55 CM a* CO a 8 8 8 S-S o ^^ .9.3 PS .s 02.72 ILLINOIS ENGINEERING EXPERIMENT STATION % c i g> [3 2.3^ a H'3 3 a M I rt 4* <B & S So .

KEIN FORCED CONCRETE FRAMES 73 Ill HA s &e 8. 8 CO ' S over . fl ' 3. TJ O |M| C I .

74 ILLINOIS ENGINEERING EXPERIMENT STATION .

REINFORCED CONCRETE FRAMES 75 PL. Co 10 . PL.CO II II oo ooo PO P-iPn PH 00l>^ rH T^ ^ 000 I + w * o *r. -2 PnPn T^t^ CO CO do" coco co oo do PnPn "^ Oi (N C^ CO<M do" t^ CO T-H O CO 0(MO CO CO N.

The tensile steel at the fixed ends of the columns were rather low. in most cases. loads were increased the cracks developed and new cracks appeared on the tension side between the points of application of the load on the beam. The bent-up bars in the beam came into action as soon as tension cracks formed in their vicinity. *~ the tensile 3 \^~ Jr ''^-J '~^ a U U r ul LJ r . After the formation of these cracks. a formed on the beam underneath the load and then a crack formed at the side near the juncture of the column and the beam.76 crete ILLINOIS ENGINEERING EXPERIMENT STATION on the tension side of the vertical tension crack member was sufficiently stretched. the tension As the in these parts was taken mainly by the reinforcing bars. and in the bent-up portions tensile stresses as high as 22. at regular intervals in the due to the negative bending moment within the the juncture of the beam and the column was small. and horizontal cracks formed columns.000 pounds per square The tensile stresses in the inch were developed in several instances. space occupied by and in these places tension cracks did not form in many frames until The tensile stress high loads were applied. strength of the concrete in this part apparently reduced stress in the steel.

. on each side of the center of the top beam and extended upward 2 in. respectively. and the load was held for a few minutes and then dropped very slowly. load the first fine crack in the beam appeared directly under a load point and extended from the bottom of the member vertically 2 in. from the bottom surface of the beam. to the level of the reinforce- cracks shown in Fig. The frame carried 40. two cracks appeared 2 in. The location of the 12. which show the location of cracks are given of the tests of the individual frames are in Fig. At 12. and a new crack appeared just inside the right-hand corner 10 inches from the center of the beam and extended diagonally toward the load point. 52. from the top corner of the frame and extended vertically downward. At the same load four cracks appeared at the outer shoulders. The frame failed by tension in the reinforcement of the beam. and the maximum compression was observed along the sharp corner at the juncture of the beam and the column. as might be expected. Frame 2 Inverted U-frame with Columns Hinged at Lower End. The cracks were well distributed in the tension zone of the frame and no crack due to diagonal tension was formed. 49 shows the frame in the testing machine. 2 in. is .000 Ib. 5 in. 51. Nominal span length and total height were each 6 ft. edge of the column on a level with the surface of the beam bottom and one at 2 ft. At 14.500 Ib. At the ment. Frame 1 Square Frame with Columns Hinged at Lower End. Nominal span length was 6 feet. 49 to 57. . At 8.000 Ib.000 Ib. Total height was 5 ft.000 Ib. from the bottom of each column end. Frame was loaded at the one-third points of the span of the horizontal member. The location of cracks is shown in Fig. and 3 in. At that load cracks appeared in. first noticeable cracks appeared in the in one the outside columns. Load was applied at the center of the span of the horizontal member. The general phenomena given in the following brief notes. No crack appeared in the top side of the beam until At the same load the 8 the load was increased to 36. This is due to the curved beam action at the rigid joint. Fig. In each frame the maximum Views of the frames load was higher than the load expected. these cracks had extended vertically 6 in.000-lb.REINFORCED CONCRETE FRAMES 77 High compress! ve stresses were developed in the concrete in the upper portion of the columns below the intersection with the beam.

FRAMES 1. LOAD-DEFORMATION DIAGRAMS FOR ADDITIONAL GAGE LINES. 2. 3 . 45.78 ILLINOIS ENGINEERING EXPERIMENT STATION 40000 30000 IOOOO Frnme 3 Unit Deformation FIG.

REINFORCED CONCRETE FRAMES 79 Concrete no/es on of frame J .

6.80 ILLINOIS ENGINEERING EXPERIMENT STATION 30000 20000 7 i/n/'f Deformation FIG. 7 . 5. LOAD-DEFORMATION DIAGRAMS FOR ADDITIONAL GAGE LINES. 47. FRAMES 4.

REINFORCED CONCRETE FRAMES 81 .

. Accidentally the frame was built slightly out of form. paratively high in the horizontal beam and at the rigid joint between the columns and the sloped beam. The distribution The frame. the first noticeable crack appeared at the left-hand inside top corner. 11 in. At 10. Frame 5 Trestle Bent with Tie Span length center Total height from 1^ Load was applied the top at the center of the top beam. upward. to center at the supported column ends was 6 feet. when three cracks appeared at the bottom between At 30. Nomi- nal span length was 6 ft.). carried 60.000 Ib. factory information (32.900 Ib. the base to the top of the frame was 10 ft. and extended 2J/2 in. However. downward from the extended line of the bottom face of the beam. had been applied. the columns being out of plumb 1J4 in. No The frame crack was observed at the fixed ends of the columns. due to the lack of sufficient friction to resist the horizontal thrust at the support. before the slipping occurred. Failure was by tension in the steel load (50. The cross-section of beam . Loads were applied at one-third points. diagonal tension crack appeared in the beam. in the height of 4 ft. loads and several cracks appeared in both columns. Nominal span length was 6 ft. Loads were applied at one-third points of span of member. and there appeared to be no danger of sudden failure. satiswas obtained because very high tensile stress in.82 ILLINOIS ENGINEERING EXPERIMENT STATION the cracks were extending. from hinged end of columns. Total height of frame from fixed column end was 4 ft.000 Ib. 54. however. and one crack appeared at the top of the beam.000 Ib. per sq. The cracks in the upper part of the columns were located within 14 in. then dropped No very slowly.000 Ib. carried a comof the cracks shows this clearly. and the load was held for a few minutes.) had been developed at the center of the beam Frame 3 Square Frame with Columns Fixed at Lower End. and the frame failed by tension in the longitudinal steel of the beam. in. Total height of frame was 6 ft. The location of the cracks is shown in Fig. 53. and more stress was thrown to the left-hand column than to the other.000 Ib. 3 in. Cracks are shown in Fig. horizontal Frame 4 Inverted U-frame with Columns Hinged at Lower End. new cracks appeared in the beam and columns. (A-frame). Unfortunately at this load the foot of one of the columns slipped outward about J^ in. No noticeable cracks appeared until 21.

VIEW OF FRAME 1 IN TESTING MACHINE FIG.FIG. VIEW OF FRAME 5 AFTER TEST . 49. 50.

VIEW OF FRAME 2 AFTER TEST . 52. T f T \ FIG.NPI. 51. VIEW OF FKAME 1 AFTER TEST FIG. .

FIG. VIEW OF FRAME 3 AFTER TEST FIG. 54. VIEW OF FRAME 4 AFTER TEST . 53.

VIEW OF FRAME 6 AFTER TEST FIG. 56. 55. VIEW OF FRAME 7 AFTER TEST .N?6 * 30 A* 3I 3< JJ- it* FIG.

At this load sudden as failure and the cracks extended took place at both inside corners of the lower ends of the columns horizontally and vertically almost through the . new cracks appeared in the beam where the longitudinal bars were |bent up and these cracks ran diagonally almost to the load points.000 Ib.REINFORCED CONCRETE FRAMES was the 87 At 40. three in the beam and three in the columns.000 Ib.000 Ib. and the column section 8^ by 8% in. The frame carried 46. The location of the cracks is shown in Fig.000 Ib. a view of the frame after the shows the appearance of the cracks. one near the center of the beam. and one at the upper part of the right-hand column.000 and 36. The first noticeable cracks appeared at 21. 50. The outer points of the beam. first cracks appeared in the column at the outside with the beam. the cracks extended further and additional cracks appeared in the beam and columns at regular intervals. from the outside face of the columns. applied at one-third points of span of test. 16 in.000 its near with a breaking sound.000 Ib. At the same load three cracks formed in the bottom half of the beam. load was controlled and the load Frame 6 Square Frame Same as Frame 1.000 Ib. Load was horizontal with Columns Hinged at Lower Ends.. member. and the cracks at both ends of the beam extended The vertically downward nearly to the bottom side of the beam. Same Frame 3.000 juncture the occurred at right-hand rigid joint between the tie and the column Ib. At Ib.000 Ib. At 18. The maximum by the failure of the top beam which failed by tension in the reinforcement. the crack located on the outside of the left-hand load point extended diagonally almost to the load point. two noticeable cracks appeared. 56. ultimate load carried by the frame was 61. Fig. At 24.. At 30. 55. four Cracks appeared. At 100. As the load increased. load a crack suddenly 140. Failure was by tension in the longitudinal steel in the beam. carried 146. The frame gradually dropped.000 Ib. At 30. These were under the load For location of cracks see Fig. two of them under the load points. the cracks had extended further and two cracks due to the negative bending moment appeared at the ends 8 in. No crack appeared on the top side of the beam. and the load then dropped slowly. 8% by first cracks of the beam the finally extended diagonally nearly to the load point.000 Ib. Load was applied at one-third points of the span of the horizontal member. Frame 7 Square Frame with Columns Fixed at Lower Ends.

but the structure as a whole probably was not appreciably affected. At this load the reinforcement at the bottom of the middle beam was The frame. Consequently the ends of the frame projected beyond the base of the testing machine upon which the center to center. tion of the ends of the frame under test an Ames dial was attached at each end of the frame. At this load the crack at the center of the middle span had opened considerably and the steel at this place had scaled. 8 in. over 20 hours. The movements of both ends were observed The maximum movements (upward) were as the load increased. when three cracks appeared in the middle span. Total height of frame was 6 ft. 7J^ in. they extended deeper. The base of the frame was 15 ft. Frame 8 Frame with Three Spans. failed by tension in the longitudinal steel of the top beam. the location of the cracks. in length. while the length of the base of the testing machine was 10 ft. At 60. 57 shows the appearance of the frame after the test with span. At the same time the conindicating failure by tension in the steel. observed at 60. To observe the upward deflecframe was bedded in plaster of paris. carried an increasing load in good condition and the highest load was 134. 6 in.000 Ib. several new cracks appeared at both ends of the middle span and also in the upper part of the intermediate columns.000 Ib. but the fall in the applied load was only 300 Ib.000 Ib. .000 Ib. one in each outer span on the top side of the horizontal member near The former are due to the positive bending moment and the latter are due to the negative moment. Also the crete at the top of the intermediate columns had crushed. and span may the intermediate columns. The frame also considerable positive bending was developed there. It is noted that the stresses in the outside columns were very low. Therefore the steel stress in the beam have been slightly modified by the movement. stressed in tension beyond the elastic limit of the steel. 4 in. at the east end and 1/300 of the side in.000 Ib. however. at the west end. Uniform load was applied to the upper horizontal member of the middle Fig. At 75. concrete base was cracked at the bottom end of the right-hand intermediate column. and the amounts were 1/263 in.000 Ib.. No cracks were observed until the load had reached 45. Span lengths were 4 ft. The frame was subjected to the load of 60. even at the maximum load.88 ILLINOIS ENGINEERING EXPERIMENT STATION This fact shows that concrete base and almost through the columns. These cracks were located symmetrically and they extended vertically about 6 in.

--1-7-1 '""-- r - jr...<.- _. VIEW OF FRAME 8 AFTER TEST ._ FIG.- i i* ..- < : ..._. .*l*ii>A*^t if<. N5S - . 57.

.

and also of those in columns having fixed ends. It appears reasonable to take the average value of the observed stresses in the part in question for the purpose of comparison. The design of Frames 1 to 7 was such as to cause high stresses (4) in columns and beams at about the same time. There was more stress in the steel at the rear side of the member than in the steel at the front side. The concrete on the rear side was stiffer and stronger than that on the front side and the distribution of steel stresses seem to have been modified by this fact. In the comparison of test results with analysis. In most of the test pieces. two cases must be conone in which the concrete is considered to take tension and sidered. the following modifying conditions have been taken into consideration: (1) The quality of the concrete was not uniform over the cross- section of the member. stresses with the observed stresses. Under these conditions the bending moment at the center of the beam will be less than that calculated on the basis of the nominal span length the difference being greater than that occurring with the dimensions found in practice. and the concrete on the rear side of the frame (the bottom side for the position in which the frame was made) seems to have been richer than that on the front side. In addition to this the corner at the juncture of the beam and the column was provided with a fillet. the column width occupied nearly one-seventh of the nominal span (distance center to center of the columns). the numerical values of the bending moment in the beam. computed were designedly made than would commonly be used in practice. The steel stresses are greatly modified by the presence of (2) Therefore in comparing the tension in the concrete for the low loads. the other in which the concrete is considered to be broken in tension.EEINFOKCED CONCRETE FRAMES . In Frame 8 the presence . In the computations for the horizontal reactions the nominal span and In finding height of the frames (distances center to center) were used. 91 23 Conditions Considered in the Comparison of Tests with A nalysis. but when the equations involved further use of span lengths the nominal span length was replaced by (3) The cross-sections of the test frames larger in proportion to the span the clear length of the span. the horizontal reactions computed as described previously were used. The frames were made in a horizontal position on the floor of the laboratory.

in 43. i. tension side. intermediate columns and in the beams of the side spans are also small except at the extreme end. The observed Comparison of Test Results with Analyses. plotted process already described. a straight line stressdeformation relation was assumed. 24. The selection of a proper value of the of the moments and made Frame 8. The values of modulus of elasticity of the concrete used in the computations of stress are given in Table 11. A probable value for strength the moment of inertia of these members will be an average between that obtained by using the full cross-section and a section which negstresses for lects the part outside the tension rods. less than the Naturally the modulus of elasticity used was generally initial modulus.. It will be seen then that for Frame 8 there are two cases to be considered in comrestraint as to give the paring the experimental results with the analyses. modulus of elasticity was somewhat dependent upon a knowledge of the qualities of the concrete of the various frames and a comparison of the behavior of the frames with that of the corresponding control cylinders. by the e. Therefore in the calculation of the tion for the second case it is moment of inertia of the cross -sec- not correct to neglect entirely the tensile the in of concrete these two members. the dashed line the stress in the farther side of the specimen. the stresses have been plotted from the central longiplane. The dotted line represents stress in the concrete. This assumption was in the numerical computation In making the computations of stress in the concrete of the frames a constant modulus of elasticity was used. Consequently loads which would cause high stresses in the beams would not produce cracks even in the intermediate columns although the moment there would be greater than in the outer columns.92 of the ILLINOIS ENGINEERING EXPERIMENT STATION members to be much of unloaded panels caused the smaller than in the loaded beam. that is. The 36 to have been Fig. moments in the columns The base also offered so much column ends almost a fixed condition. one in which tensile strength in the concrete is considered in all members and the other in which the beams and the intermediate columns are cracked on the The bending moment in the outside column is very and there is no chance for tension cracks. Because of tudinal axis of the member in which they were observed. the possibility of confusion resulting from this method an exception . deformations the from observed those obtained stresses. The moments in the small. and only one crack appeared in this member. usually at the median In general. light full line represents stress in the steel at the nearer side of the specimen.

but the maximum difference between experimental and theoretical values is higher than in the beams. steel stresses outer surface. owing to the fact that the direct stress is not equally distributed over the cross-section of the column. tension in steel and compression in concrete when measured at or near the outer surface of a member have been plotted out. The point at which concrete stresses. (2) The experimental and the computed values of the steel stress columns are also in fair agreement. The location of the In Table 14 are given values of both observed and computed stresses at three points for several loads for all the frames. the stress line crosses the coordinate axis represents the position of the point of inflection and not a change in the sign of the stress. In Fig. consequently. The heavy full line represents the computed stresses for both steel and concrete which have been calculated for various points by means of the formulas given in preceding pages.REINFORCED CONCRETE FRAMES 93 was made in Frames 2. 45 and 47 are given load-deformation curves for gage lines gage not represented in the diagrams of Fig. toward the central part of the frame as a whole. where the stresses have been plotted from the line showing the location of the reinforcing bar. likewise. that is. A study of the tables and diagrams seems to justify the following statements: (1) The experimental and the computed values of steel stress at the center of the loaded top beam are in fair agreement for each kind of frame tested except in a few instances in which the load was com- paratively low or extremely high. In only a few cases have tension in concrete -and compression in steel been plotted. represent compression. Tension in steel and compression in concrete when measured at or near the inner surface of a member have been plotted in. the computed stress is based on the assumption of no tensile strength in the concrete. and 8. where II precedes it and where neither I nor II is given. 46 and 48. 36 to 43. Where I precedes the computed stress the calculation considered the tensile strength of the concrete. represent tension regardless of whether in the diagrams they appear Plotted as positive or as negative with reference to the coordinate axes. 5. in the . Both have been plotted out if measured on the inner surface of a member and in if measured on the With very few exceptions. lines is shown in Fig.

Frame .94 ILLINOIS ENGINEERING EXPERIMENT STATION 14 IN TABLE COMPUTED AND OBSERVED STRESSES Computed Values f s is FRAMES are given in Roman Type. fc is Unit Stress in Concrete. Observed Stresses in Italics. Unit Stress in Steel.

are given in Roman Type. Observed f c is Stresses in Italics. Unit Stress in Concrete. is Unit Stress in Steel.REINFORCED CONCRETE FRAMES TABLE 14 95 (CONTINUED) COMPUTED AND OBSERVED STRESSES IN FRAMES Computed Values f. Frame .

gate the distribution further. The results of these observations showed practically no bending 25. per sq. In Table 15 are given stresses in bars at front and back at two places on the frame for one load generally seen that generally the stress in a bar near (top of the member as poured) is less than that To investiin a bar near the back (bottom of the member as poured) near the the front of the maximum. 2. 7. In some instances the somewhat higher than the discrepancies ran up to 50 per cent. 26. which would be expected as a result of the friction in the bearings at the free ends of the columns in Frames 1. special measurements were made in the columns of Frames 6 and 7. agree reasonably well with the computed values though in most of these cases the observed concrete stresses were computed stresses. stresses stress for all loads except the ultimate load. the front of the member being the top side of the frame as poured and the back side being the bottom. in. and for higher stresses the discrepancies were frequently even greater. The front outer corner developed the lowest tensile stress and the front inner corner the highest compresThe back outer corner developed the highest tensile sive stress. Other matters difficult of explanation probably cause further discrepancies. The concrete bases used for Frames 3. . and 6 seem to have been very small and may be neglected without appreciable error. and the made at each load. and a slight bending in the base due to a load may be expected to have an influence on the bending in the other members. The gage lines observations were were located on the four faces of the column. It member is . Deformation readings at the middle point of the base were taken at each increase in the load. of course. The secondary Effect of End Condition of Column on Results.96 (3) ILLINOIS ENGINEERING EXPERIMENT STATION stresses in the concrete at the The observed compressive low loads which developed a unit-stress up to about 800 Ib. the gage lines being placed where bending was not sufficiently large to produce tension cracks in the concrete. Undoubtedly these differences are partly due to the fact that the modulus of elasticity used does not represent correctly the modulus of elasticity of the concrete in the frames. not entirely rigid. tions of a it was found that differed In the observaDistribution of Stress over the Cross-Section. and 8 to secure the fixity of the column ends were. stress in the steel on bars near a front corner member from that on bars near a back corner of the mem- ber. 4.

REINFORCED CONCRETE FRAMES 97 CO |||8S8 CO r-l w oooooooo >oe <M (N (N (N CO <N Ol Ti & <NO5iO"*TjHOOiOOl (MCOCOCOCO(NCOCO C^ C^l OS *^ CO O^ CO CO .

The deformation in the columns of Frames 3 and 7 were thus separated. of inflection in the 'as is indicated by the analysis. the point of inflection in a Position of Point of Inflection in Columns.98 stress ILLINOIS ENGINEERING EXPERIMENT STATION and the back inner corner the lowest compressive stress. and the The position of the point position of the point of inflection found. For these frames the point of inflection was found to be almost exactly at one-third the height of the column. 58. a straight line stress-deformation relation being assumed. The observations indicate a lateral bending and twisting of the frame. It seems probable that the main source of the difference in stresses from front to back was the lack of homogeneity of the concrete. The position of member of a structure which is subject to flexure is an important element for use in designing the frame. B ^ . 28. by direct it is stress necessary to separate the deformation and that due to flexure of the member. In the Continuity of the Composing Members of a Frame. columns of these frames changed very little during the progress of the loading. To determine from observed deformations the position of the point of inflection for a into that caused member. The was not much altered by the increase of load. CONNECTION OF BEAM AND COLUMN SHOWING TYPICAL LOCATION OF FIRST CRACKS 27. A difference in stiffness would at least partially distribution account for the phenomena. whatmembers of of was no frames there discontinuity sign the stresses the frames and from the of It is apparent from action . that at the bottom of the member as poured being stronger and stiffer than that at the top.-Beam Column FIG. tests of the ever.

rigid connection at the lower A (Fig. 59.REINFORCED CONCRETE FRAMES 99 mitted from it is felt observed that the stresses and therefore the moments were well transmember to member by the connection. Stresses at Corners. at gage lines in the neighborhood of the fillets and within the space occupied by the intersection of the two members are plotted in Fig. the crack at B appeared at nearly the same time as that at A. In the design of a frame a square corner such as that shown at A in Fig. Some of these values have been converted into concrete stresses by the use of the moduli of elasticity of the concrete already assumed and are given in Table 16. for it is well known that theoretically in resisting bending the material at the corner would develop excessively high stresses. These deformations are of interest. 59 (b). 58) at a lower In the frames with column ends. It is therefore common to design such corners with fillets as shown in Fig. RECTANGULAR JOINT WITH AND WITHOUT FILLET 29. 45 and 47. FIG. From the results that there is every reason to have confidence in the rigidity of connections in frames that are properly designed. No attempt will The observed deformations . 59 (a) should be avoided for all con- nections. be made to compute the stresses at the fillets. In the frames free to turn at the lower column ends there was a tendency for a crack to form near the juncture at load than that at which a crack appeared at B.

Load Ib. per sq.100 ILLINOIS ENGINEERING EXPERIMENT STATION TABLE 16 OBSERVED STRESSES AT SHARP CORNERS IN FRAMES 1. 3 and 6 Stress in Ib. . in.

brought about by inclining the inner surface and making the outer surface vertical will add materially to the rigidity of a frame without a proportional increase in the at the end. a stress in the member. is increased by fixing the lower column ends This is obviously due to the increase in horizontal of a frame. frames as for steel structures.REINFORCED CONCRETE FRAMES (5) 101 No sudden failure took place in the frames tested. tie Due (11) member to A to cause a sudden breaking of such a joint accompanied an increase of bending moment in a main member. amount (10) of material used. in increase the slight top width of a vertical member and a slight decrease in its bottom width. vertical line. indicating as great relia- bility for reinforced concrete juncture of at which the first fine crack appears near the members. marked tendency . high compressive stresses in several cases local failure occurred by the crushing of the con- crete at these corners under high loads. that (9) Owing o to the existence of to a horizontal thrust (which of simple frames) varies from P -^P lo in most common cases at the ends of a vertical or inclined incline the member slightly member. (8) A slight deviation of the axis of vertical is members from a ' " out-of-f orm' of the vertical to say. it is advisable to toward the direction of the reaction Such arrangement will greatly reduce the bending If this arrangement is not practicable. it may be form of frame in such a way that the column take no bending stress throughout its length. possible to select the will For a frame having an inclined column. produced an appreciable variation in the stress distribution in the frame. a slight columns. attention should be paid to the rigid joint of a insure the stiff connection with a main member. thrust at the lower column end over that developed when the lower (6) The load end is free to turn. were and action beam curved due to so-called developed in the concrete (7) At sharp inside corners. The increase in the deflection was uniform.

102

(12)

ILLINOIS ENGINEEKING EXPERIMENT STATION

The use

of a footing

is

rigidly connected to the lower

advisable, for it will reduce the bending moment at the juncture of the vertical and inclined members. frame having such a footing is solvable analytically, since it ap-

end of a vertical member

A

proaches the case halfway between that of the hinged end and that of the fixed end of the vertical member, provided the foundaA little consideration is needed to tion is sufficiently unyielding.

provide proper reinforcement at the juncture of the column and the footing.

The formulas derived by analysis may be applied to a (13) variety of forms of frames and are of wide applicability.

**LIST OF PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION
**

Bulletin No. 1.

Teats of Reinforced Concrete Beams,

**by Arthur N. Talbot.
**

1905.

1904.

None available.

and Henry

Circular No.

1.

High-Speed Tool

Steels,

by

L. P. Breckenridge.

None

available.

Bulletin No. 2. Teats of High-Speed Tool Steels on Cast Iron, None available. 1905. B. Dirks.

by

L. P. Breckenridge

Circular No. 2. Drainage of Earth Roads, Circular

by

Irs O. Baker.

1906.

None

available.

Fuel Tests with Illinois Coal (Compiled from tests made by the Technological Branch of the U.S. G. S., at the St. Louis, Mo., Fuel Testing Plant, 1904-1907), by L. P. Breckenridge and Paul Diserens. 1908. Thirty cents.

No 3

Breckenridge.

1906-

Bulletin No. 3. The Engineering None available. 1906.

Experiment Station of the University

of Illinois,

by

L. P.

Bulletin No. 4.

Bulletin No. 5.

Tests of Reinforced Concrete Beams, Series of 1905, Resistance of Tubes to Collapse,

**by Arthur N. Talbot.
**

L. Carr.

Forty-five cents.

by Albert

P.

Carman and M.

Webber.

1906.

1906.

None

available.

Bulletin No. 6.

Holding Power of Railroad Spikes, by

Roy

I.

None

available.

**Bulletin No. 7. Fuel Tests with Illinois Coals, 1906. None available. Dirkd.
**

Bulletin No. 8.

available.

by

L. P. Breckenridge, S.

W.

Parr,

and Henry B.

1906.

Tests of Concrete:

I,

Shear;

II,

Bond, by Arthur N. Talbot.

None

Bulletin No. 9. An Extension of the Dewey Decimal System of Classification Applied to the Engineering Industries, by L. P. Breckenridge and G. A. Goodenough. 1906. Revised Edition

1912.

Fifty cents.

Bulletin No. 10. Tests of Concrete N. Talbot. 1907. None available.

and Reinforced Concrete Columns,

Series of 1906,

by Arthur

Bulletin No. 11. The Effect of Scale on the Transmission of Heat through Locomotive Boiler Tubes, by Edward C. Schmidt and John M. Snodgrass. 1907. None available. Bulletin No. 12. Tests of Reinforced Concrete None available. 1907.

T-Beams,

Series of 1906,

by Arthur N. Talbot.

Bulletin No. IS. An Extension of the Dewey Decimal System of Classification Applied to ArchiNone available. 1907. tecture and Building, by N. Clifford Ricker. Bulletin No. 14. Tests of Reinforced Concrete None available. 1907.

Beams,

Series of 1906,

**by Arthur N. Talbot.
**

1908.

Bulletin No. 15.

How

to

Burn

of

Illinois

Coal without Smoke,

by

L. P. Breckenridge.

None

available.

Bulletin No. 16.

A

Study

Roof Trusses, by N.

of Coal,

Clifford Ricker.

1908.

None

available.

Bulletin No. 17. The None available. 1908.

Weathering

of

by

S.

W.

Parr, N. D. Hamilton,

and W. F. Wheeler.

1908.

Bulletin No. 18. Forty cents.

The Strength

Chain Links, by G. A. Goodenough and L. E. Moore.

Bulletin No. 19. Comparative Tests of Carbon, Metallized Carbon and Tantalum Filament Lamps, by T. H. Amrine. 1908. None available. Bulletin No. 20. Tests of Concrete and Reinforced Concrete Columns, Series of 1907, by Arthur

N. Talbot.

cents.

1908.

None

available.

**Bulletin No. 21. Tests of a Liquid Air Plant,
**

Bulletin No. 22. Tests of Cast-Iron 1908. None available.

by C.

S.

Hudson and C. M. Garland.

1908.

Fifteen

**and Reinforced Concrete Culvert Pipe, by Arthur N. Talbot
**

and Weight

of

Bulletin No. 23. Voids, Settlement Fifteen cents.

**Crushed Stone, by Ira O. Baker.
**

Distillation,

1908.

**^Bulletin No. 24. The Modification of Illinois Coal and C. K. Francis. 1908. Thirty cents.
**

Bulletin No. 25. Lighting

by Low Temperature

by

S.

W. Parr

1908

**Country Homes by Private Electric Plants, by T. H. Amrine.
**

is

Twenty

cents.

*A

limited

number

of copies of bulletins starred

available for free distribution.

103

104:

PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION

Bulletin No. 26. High Steam-Pressure in Locomotive Service. A Review of a Report to the Carnegie Institution of Washington, by W. F. M. Goss. 1908. Twenty-five cents.

**and Duff A. Abrams.
**

Bulletin No. 28. 1909. Fifteen cents.

Bulletin No. 27. Tests of Brick Columns and Terra Cotta Block 1909. Twenty-five cents.

Columns bv Arthur

N

Tilhnt Talbot

A

Test of Three Large Reinforced Concrete Beams, by Arthur

N

Bulletin No. 29. Tests of Reinforced Concrete Beams: Resistance to 1907 and 1908, by Arthur N. Talbot. 1909. Forty-five cents.

Web

Stresses, Series of

**^Bulletin No. SO. ment, L. H. Adams,
**

cents.

the Rate of Formation of Carbon Monoxide in Gas Producers, and C. N. Haskins. 1909. Twenty-five cents.

On

by

J.

K. Cle-

^Bulletin No. 81. Tests

Bulletin No. 32.

cents.

**with House-Heating Boilers, by
**

in Coal,

J.

M.

Snodgrass.

1909.

Fifty-five

The Occluded Gases

by

S.

W. Parr and

Perry Barker.

1909.

1909.

Fifteen

cents.

Bulletin No. S3. Tests of Tungsten

Lamps, by

T. H. Amrine and A. Guell.

Twenty

^Bulletin No. 34. Tests of Two 1909. Snodgrass. Fifteen cents.

Types

of Tile-Roof

Furnaces under a Water-Tube Boiler, by

J.

M.

Bulletin No. 35. A Study of Base and Bearing Plates for Ricker. 1909. Twenty cents.

**Columns and Beams, by N.
**

High Temperatures, by

S.

J.

Clifford

and W.

Bulletin No. 36. The Thermal Conductivity of Fire-Clay at L. Egy. 1909. Twenty cents.

K. Clement

F. Wheeler.

**Bulletin No. 37. Unit Coal 1909. Thirty-five cents.
**

Bulletin No. 38.

five cents.

**and the Composition
**

of Coal,

of

Coal Ash, by

Parr and

W.

Parr and

W.

The Weathering

by

S.

W.

W.

S.

F. Wheeler.

1909,

1909.

TwentySeventy-

^Bulletin No. 39. Tests of

five cents.

Washed Grades

in

of Illinois Coal,

by C.

McGovney.

Bulletin No. 40.

A

Study

Heat Transmission, by

J.

**K. Clement and C. M. Garland.
**

1910.

1910.

Ten

cents.

Bulletin No. 41. Tests of

^Bulletin No. 42.

Timber Beams, by Arthur N. Talbot.

Thirty-five cents.

The

Effect of

Key ways on

the Strength of Shafts,

by Herbert

F. Moore.

1910.

Ten

cents.

Bulletin No. 43. Freight Train Resistance,

Bulletin No. 44- An Investigation of Built-up Herbert F. Moore. 1911. Thirty-five cents.

**by Edward C. Schmidt. 1910. Seventy-five cents. Columns under Load, by Arthur N. Talbot and
**

1911.

*Bulletin No. 45. Thirty-five cents.

The Strength

of

Oxyacetylene Welds in Steel, by Herbert L. Whittemore.

***Bulletin No. 46. The Spontaneous 1911. Forty-five cents.
**

ty-five cents.

Combustion

of Coal,

by

S.

W.

Parr and F.

W. Kressman.

1911.

^Bulletin No. 47. Magnetic Properties of Heusler Alloys,

by Edward B. Stephenson.

Twen-

and Melvin

^Bulletin No. 48. Resistance to Flow through Locomotive L. Enger. 1911. Forty cents.

**Water Columns, by Arthur N. Talbot
**

F.

^Bulletin No. 49. Tests of Nickel-Steel Riveted Joints, 1911. Thirty cents.

by Arthur N. Talbot and Herbert

Moore.

1912.

^Bulletin Fifty cents.

No. 50. Tests of a Suction Gas Producer, by C. M. Garland and A. P. Kratz.

Bulletin No. 51. Street Lighting,

by

J.

M. Bryant and H.

G. Hake.

1912.

Thirty-five cents.

^Bulletin No. 52. Fifteen cents.

An

Investigation of the Strength of Rolled Zinc,

by Herbert

F. Moore.

1912.

1912.

*Bulletin No. 53. Inductance of Coils,

**by Morgan Brooks and H. M. Turner.
**

by A.

Guell.

Forty cents.

*Bulletin No. 64. Mechanical Stresses in Transmission Lines,

1912.

Twenty

cents.

^Bulletin No. 55. Starting Currents of Transformers, with Special Reference to Transformers with 1912. Twenty cents. Silicon Steel Cores, by Trygve D. Yensen.

^Bulletin No 56 for Structural Steel

Tests of Columns: An Investigation of the Value of Concrete as Reinforcement Columns, by Arthur N. Talbot and Arthur R. Lord. 1912. Twenty-five cents. A Review of Publication No. 127 ^Bulletin No. 57. Superheated Steam in Locomotive Service. 1912. Forty cents. of the Carnegie Institution of Washington, by W. F. M. Goss.

*A

limited

number

of copies of bulletins starred

is

available for free distribution.

*Bulletin No. Parr and H. 66. *Bulletin No. The Coking of Coal at Low Temperatures. by F. Anderson 1913. by Herbert F. Slater. Bulletin No. Yensen. Trygve D. Watson. The Analysis of Coal with 1914. F. 1915. Goodenough. L. the Strength of Concrete. The Coking Coal at G. and H. 78. M. Thirty-five cents. Becker. Laboratory Tests of a Consolidation Locomotive. Study of Boiler Losses. Slater. by A. Fifty cents. of Iron Silicon Alloys. 1915. The Mortar-Making Qualities of Illinois Sands. R. The Effect of Cold Weather upon Train Resistance and Tonnage Rating. Bulletin No. 77. Seventy cents. Williams. M. Fifty cents. Twenty-five cents. Talbot and The Steam Consumption 1913. Ten cents. 80. 76. R. W. 71. Dunn. A. Fifty cents. with a Preliminary Study of the By-Products. by A. Melted J. Thirty-five cents. Keller. M. 62. 1912. The Strength I-Beams and Girders. by Edward C. Talbot and W. Entropy-Temperature 1913. Talbot' 1913. Fifteen cents. Twenty Bulletin No. by C. Yensen. Willis A. Temperature on. 88. 64. 1916. 70. W. Hadley. Correction of Echoes in the Auditorium. Washing in Illinois. Abrams. of Locomotive Engines from the Indicator Diagrams. Thirty cents. Wilson. White. 73. 74. cents. by F. A. by F. 1915. and Transmission Diagrams by C. Snod- ^Bulletin No. 75. 67. Low Temperatures. 85. and R. R. Yensen. W. by G. Influence of 1915. Twenty-five cents. Tests of dollar. Fifty cents. Twenty-five cents. Tests of Reinforced Concrete Buildings 1913. 65. Ammonia Vapor. Wind Stresses in the Steel Frames of Office Buildings. Magnetic and Other Properties 1915. 86. 1916. Schmidt and F. A New Analysis of the Cylinder Performance of Reciprocating Engines. 83. Good- Bulletin No. Kratz. by J. *Bulletin No. ^Bulletin No. 84. Phenol as a Solvent. L. Parr and H. grass. Olm. Reinforced Concrete 1913. Olin. by Duff A. 1914. Moore. Bulletin No. 68. *Bulletin No. Sixty cents. The Strength of I-Beams in Flexure. B. Lincoln. T. Iron Melted in Vacuo. by Trygve cents. Dry Preparation of Bituminous Coal at Illinois Mines. 1915. 1913. Wiley. J. and *Bulletin No. Bond between Concrete and Steel. Paul Clayton. Maney. Woodrow. McDaniel. R. of Electrolytic *Bulletin No. J. 79. C. Watson 1916. by G. A. One D. by Arthur N. of A by A. 68. C. A. 87. by S. by Arthur N. by A. *Bulletin No. Twenty Bulletin No. *Bulletin No. Coal Bulletin No. Schmidt. Richards. A. ^Bulletin No. Thirty-five cents. Twenty-five cents. Marquis. *A limited number of copies of bulletins starred is available for free distribution. by Arthur N. M. C. 1916. 1913. P. Twenty *Bulletin No. Moore and W. 1915. by E. *Bulletin No. Melted in 1914. ^Bulletin No. Bulletin No.PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION 105 *Bulletin No. The Tractive Resistance Thermal Properties of of a 28-Ton Electric Car. for Air. in Vacuo. 1913. Bulletin No. . 63. University of Illinois. Holbrook. W. Wilson and *Bulletin No. by Herbert F. and of Stiffness of Steel under Biaxial Loading. *Bulletin No. Bulletin No. Forty cents. Parr and H. Tests of Reinforced Concrete Flat Slab Structure. Twenty-five cents. 78. 59. by Elmer H. 61. Characteristics and Limitation of the Series Transformer. Forty cents. Sixty-five cents. 60. 1914. Vacuo. Magnetic and Other Properties of Electrolytic Iron 1914. 82. The Effect of Boron upon the Magnetic and Other Properties by Trygve D. by S. B. Bulletin No. by Bulletin No. Sixty-five cents. under Load. Thirty-five The Electron Theory of Magnetism. Bulletin No. Fifteen cents. The Strength 1916. by W. Twenty cents. 1912. cents. 1912. by Paul Clayton. Fifty cents. Thirty-five cents. 81. Acoustics of Auditoriums. The enough and William Earl Mosher. Steam. 69. Twenty-five cents. Wall Footings and Column Footings. by E. 1914. Bulletin No. R. by Harold H. Properties of Saturated and Superheated 1913. with Special Reference to the Properties and Composition of the Products. by S. cents.

by The Storage Fuel of Bituminous Coal. by A. Flow of Water Through a Submerged Short Effects of Storage Upon the Properties of Coal. Percentage of Extraction of Bituminous Coal with Special Reference to Illinois by C. Circular cents. 1917. by Fred B Seely. 1917. 1917. Fleming and Melvin L. A. Twenty ^Bulletin No. by T. 6. 1916. Twenty cents. Gonnerman. Stoek. ^Bulletin No. MikishiAbe. Economy Hand Fired Power Plants. A. Analysis 1918. Gravity Studies of Illinois Coal. Nebel. Yensen and W. Schmidt. ^Bulletin No. and Tests of Rigidly Connected Reinforced Concrete Frames. 1916. M. Copper. 91. Ten cents. 8. 4. Nickel. C. F. 1918. cents. by Thirty cents. Wilson and H. Some Graphical Solutions of Electric Railway Problems. Magnetic and Other Properties of Iron-Aluminum Alloys Melted in Vacuo. Twenty *BulletinN 0. Parr. 7. Twenty-five cents. 6. by A. *Buttetin No. Talbot and Harrison F. ^Bulletin No. 1918. *Circular No. An Investigation of Twist Drills. by L. 1916. cents. Stoek. *Bulletin No. W..106 PUBLICATIONS OF THE ENGINEERING EXPERIMENT STATION Specific *Bulletin No. No. Young. The Tractive Resistance on Curves Twenty-five cents. Circular No. 92. P. 1917. H. Conditions. Harder. Tests to Determine the Rigidity of Riveted Joints of Steel Structures 1917. 1916. C. Twenty cents. 89. Moore. 1916.105. Snodgrass. Hydraulic Experiments with Valves. Talbot. Nozzles. Lichty. of a 28-Ton Electric Car. Fifty cents. with Special Circular No. by Arthur N. 100. Enger. 1918. C. Thirty *Bulletin No. Comparative Tests of Six Sizes of Illinois Coal on a Mikado Locomotive. 94. Sixty cents. by A. Beyer. Twenty cents. Gatward. 1917. The Effect of Mouthpieces on the *Bulletin No. H. 99. C. 102. Thirty-five cents. Test of a Flat Slab Floor of the Western Newspaper Union Building. J. Holbrook. P. *Bulletin No. by Merle L. Forty cents. Fred B Seely. Twenty-five cents. by Arthur N. ^Bulletin No. *Bulletin No. ^Bulletin No. F. *BuUetin No. Circular No. and Orifice Buckets. M. The Collapse of Short Thin Tubes. 1917. 90. in the Operation of 1918. Lukens. Moore. The Embrittling Action Sodium Hydroxide on Soft by W. The Economical Purchase and Use of Coal for Heating Homes. 95. A Study of the Heat Transmission of Building Materials. W. by S. The Utilization of Pyrite Occurring in Illinois Bituminous Coal. Twenty-five cents. *Bulletin No. A Preliminary Study D. . H. Subsidence Resulting from Mining. Carman. Twenty-five cents. Virgil R. 103. and O. 96. by H. Fifty cents. of *Buttetin No. 1917. 93. by Herbert F. 1917. Benedict and W. 101. Pipe. Tests of Oxyacetylene Welded Joints in Steel Plates. None available. 1917. by E. 97. 106. McFarland and O. 104. of the Alloys of Chromium. E. The Economical Use of Coal in Railway Locomotives. by *A limited number of copies of bulletins starred is available for free distribution. S. M. 107. Buck. Twenty cents. S. 1917. by B. Jr. *Bulletin No. Willard and L. Twenty cents. Parr. Thirty cents. 1917. Hose. Reference to Conditions in Illinois. E. Schmidt and H. D. 1918. 98. *Buttetin No. Orifices. Dunn. Young and H. by E. M. E. and Steel. by 1917. Ten cents.

.

THE PENALTY WILL INCREASE TO SO CENTS ON THE FOURTH DAY AND TO $1.THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW AN INITIAL FINE OF 25 CENTS WILL BE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE.OO ON THE SEVENTH DAY OVERDUE. VC > 30 tQQ7 5 .

00278 UNIVERSITY OF CALIFORNIA LIBRARY .