Robert E. Englekirk Ph.D., P.E. Adjunct Professor Department of Civil Engineering University of California at Los Angeles; and Chief Executive Officer Englekirk & Hart, Inc. Los Angeles, California

The precast concrete industry is perceived by the author to be at a crossroads - one path leading to an uncertain future, the other to a new in­dustry with ever-expanding horizons. The evolution of the precast concrete industry in seismically active regions of the United States and other parts of the world are dis­cussed. Current innovative system concepts are pre­sented and the technological needs of the industry are identified. A program for in­suring that the industry ex­pands its horizons is outlined.

40

Seismic Design Considerations for Precast Concrete Multistory Buildings

Twenty-five years ago, the use of precast concrete in buildings located in regions of high seis­

micity was more prevalent than it is today. (Reasons for this decline, to­gether with ideas to alleviate this situ­ation, are given in Refs. 1 and 2.)

Currently, most precast concrete used in buildings in high seismic zones, particularly on the West Coast, is confmed to: • Cladding • Topped floor slab systems

• Beams and girders (supporting non­seismic loads only)

• Columns (posts) Today, earthquake bracing systems

composed of precast concrete ele­ments are seldom attempted. This ex­clusion of precast concrete elements from seismic load paths has not al­ways been the case. In the early 1970s, a few precast concrete bracing systems were attempted. Two basic building systems were built: • Ductile frames using precast con-

CONCRETE COLUMN

Fig. 1. Precast concrete shell for a ductile frame beam.

PCI JOURNAL

POUR SOLID POCK AFTER WELDING

ANGLE 2X2X3/16 FIELD WELDED TO

GRO

w: r-

ET "'

BARS-1-

c

~

1-

Ill Ill Ill Ill :u lfl Ill Ill Ill Ill I I

=--

v

~ f:: ~'-r--- f5 I 16'

PANEL REINFORCING

I~

~ I 2-f5 I! BOTTOM OF PANEL

I UT~ t: :::: . . ..~ ... ~ , .. , . ... . .......

c~

(}

Fig. 2. Chord tie between two precast panels.

crete shells (Fig. 1), and • Vertically assembled wall panel

systems. A chord tie for such an as­sembly is shown in Fig. 2.

Unfortunately, the construction of the prototype was more difficult than envisioned and neither approach was developed into a standard building system. Precast concrete seismic bracing systems were abandoned be-

absorption are provided." This "equivalent ductility" option was never used - and is still not used -because a standard methodology for establishing "energy absorption and ductility equivalence" does not exist. As a consequence, all load transfers must be accomplished by a mechan-

u 0

ism developed for cast-in-place con­crete. Additionally, all of the pre­scriptive provisions developed to promote ductility in cast-in-place concrete are imposed on precast con­crete construction.

The concrete details shown in Figs. 3 and 4 were considered "state of the art" design concepts in the 1970s be­cause they exceeded code require­ments. They are taken from plans for a 10-story apartment building located in Whittier, California, which was subjected to an acceleration in excess of 0.6g during the Whittier earth­quake. The building experienced no damage.

State of the practice (code comply­ing construction) details for frames and shear walls are shown in Figs. 5 and 6. Clearly, they are not intended for precast concrete construction.

A recent (early 1980s) attempt at a midrise (six-story) precast concrete building was attempted by Rockwin Corporation. Although all the com­ponents of the building are precast, the seismic technology is really a cast­in-place mentality (Fig. 7). Rockwin used this system to build their cor­porate office building. During the Whittier earthquake, this building was subjected to a full scale test at loads of probably 6 to 10 times code-

B'-4'

cause the connectivity problems were clearly overwhelming. Frame beam shell to column connections and rein­forcing placement requirements were more difficult and time consuming even though the objective was to em­ulate the cast-in-place ductile frame of the 1970s (Fig. 3).

a: ~Ul ZIIJ HH <1-::E 11.111.. a:o

. 5'-B' 0

Connecting panels using welded splices was expensive and the quality control poor. The chord ties and shear transfer details for cast-in-place buildings built in the early 1970s did not include boundary elements and, as a consequence, were simple by today' s standards. Fig. 4 is taken from the end of a shear wall designed to 1970 standards.

Precast concrete construction then, as now, was required to conform to a code developed for cast-in-place con­crete unless "evidence is submitted

. CD . CD . CD . CD . ., . C\1

.. Ill

ct 1--

'J....._r

8 I --,-

2' 4'4'4~12'0R .12'0R .12'0R .12' IR .12'0R .12"0R .1 "OR

1B" 18' 18' 18 18" 18' a· I

'0 '0 ~ fffJ ji) I

__.......J- I

lc 8

h.__ ELEVATION B-B -

that equivalent ductility and energy Fig. 3. Ductile frame reinforcing (1970 vintage).

May-June 1990 41

-ru. '-ADDITIONAL 2-#4 WITH #3 TIES @ 18 f---8" END (OMITTED FROM SHEAR WALL

ELEVATION ABOVE FOR CLARITY) •

r---r- • • ----#4 TI~S

2 @ 4 5 @ 6" REM . @' 10 " 0 . C .

" ~ • •L

6" ARCH'L. . = PROJECTION--· • _ ~ l ta {Jh- ::hl

\__AS ADJACENT FACE

Fig. 4. Boundary element for a shear wall (1970 vintage).

prescribed load levels. No structural damage occurred. The Rockwin head­quarters building is the only building constructed using this system; unfor­tunately, the structural market for precast concrete is rapidly disappear­ing in regions of high seismicity.

Creating a Viable Precast Concrete Industry

If precast concrete is to become a viable construction material, it must address issues that are constraining its use in seismic as well as nonseismic areas. The major constraint facing the precast concrete industry today is the requirement that it comply with a technology developed for cast-in­place concrete and especially those provisions which control the design

-+,

of seismic bracing systems. So long as precast concrete elects to live in the shadow of cast-in-place concrete, it will always remain the ''ugly (unbuild­able) sister."

On the other hand, during the past decade the masonry industry has made enormous strides in developing a seismic design technology that recognizes how masonry is econom­ically constructed. The precast con­crete industry must, and hopefully soon will, develop a seismic design technology that permits its intelligent use as a construction technique.

The National Science Foundation, in conjunction with PCI and the Pre­cast Concrete Manufacturers' As­sociation of California (PCMAC), has recently agreed to fund the first three years of a program entitled

rf4 TIES@ 4 • o.c.

";::>

"

EXTERNAL COLUMN REI~ORCING

LEVEL LONGITUDINAL REINFORCING

1-4 14-#11

4-6 10-#11

6-7 8-#11

7-8 6-#11

8-9 6-#10

9-R 4-#9

PRESSS (Precast Seismic Structural Systems). The objectives of this pro­gram are to:

1. Develop comprehensive and ra­tional design recommendations based on fundamental and basic research data which will optimize the viability of precast concrete construction in seismic zones.

2. Develop new materials, concepts and technologies for precast concrete construction in seismic zones.

Essential to the success of this pro­gram is direction from the precast concrete industry. Initially, the direc­tion must come in the form of ideas. Broad conceptual ideas about where the precast concrete industry should be in the 21st century must be developed. Specific questions that need to be addressed are:

15 TIES @ 4" O.C .

~12-#11 #5 TIES @ 4 • o.c. 17 @ 6" o.c. 42-111

- ~ fl-'-

Fig. 5. Ductile frame reinforcing (1989 vintage). Fig. 6. Boundary element for a shear wall (1989 vintage).

42 PCI JOURNAL

• How must precast concrete build­ings be erected if they are to be competitive?

• What precast concrete products are most versatile and competitive?

• What production techniques will be used to manufacture these precast concrete products?

• What connector concepts stream­line production and erection?

I·

CONFINEMENT ANGLE WITH ANCHORS

DETAIL A BEAM TO BEAM

In order to effectively accomplish this task, manufacturers, constructors and engineers must first purge their systems of all prejudices and per­ceived constraints which could inhibit the creative accomplishment of the aforementioned tasks. The National Institute of Standards and Technol­ogy, under the direction of H. S. Lew and Geraldine Cheok, has already embarked on a program which is

CUTAWAY VIEW

..

OETAI L B CONFINED AREA AT PRECAST COLUMN

Fig. 7. Precast building system developed by Rockwin Corporation.

May-June 1990

prototypical of what should become a standard approach.

The program was conceived and developed in accordance with the fol­lowing processes:

1. A need was perceived for high rise precast concrete buildings whose seismic resistance was to be provided entirely by ductile frames (shear wall free).

2. Erection needs were developed

. ... . . ~ ...

• A·,

-!f+!'""-:.,.;jt~~- ~-.--. ':tttt· f"-1 N S ER T S

f~l: n~j··~· ~~·~~-- ~~+GROUTED JOINT r -STANDARD

tt-tt.,..· ·.-:-''tf:-:"""~h::tttt~ ~~~~~~~~ s .. :· tt-tt,:-..:-,~-'--..,1\'-.....;-tfii-'_ +-- SHIM AS REQ'D.

itft.~~~...:...;--ttlt:· ._____THREAOE 0 .. DOWEL

*lt:--""'--":-: . ..-,~':-ittt~ r-THIN METAL

-'--~-r----~· _• _. ,·,_ ~· ·-~· ~~~~~~VEE T~EBAR FILL WITH GR OUT

DETAIL C

COLUMN TO COLUMN

43

after consultation with a group of in­dustry advisors. The consensus was that the structure must be erected much as a steel frame building is today. Essential elements of this type of erection procedure are: • A two-floor erection process. • Alignment capability after erection. • Early integration of the final brac­

ing system into the construction process so as to minimize tempo­rary bracing costs and constraints. 3. Product versatility dictated:

• The use of rectangular components for beams and columns.

• Prestressing was strongly recom­mended to improve handling, con­trol member deflection, and reduce cost. 4. Connector concepts were iden­

tified as: • A void extensive welding and the

associated embedded hardware. • Incorporate adequate tolerances. • A void large formed wet joints. • Design joints that minimize crane

time. Many engineering details requiring

extensive welding or complex assem­bling procedures were proposed by panel members as yet unable to purge perceptions of probable joint per­formance. The panel decided to at­tempt the 21st century design shown in Fig. 8 because it most closely com­plied with the manufacturing/con­struction criterion previously de­scribed. More imaginative details should and, it is hoped, will be ex­plored. An epoxy bonding (similar to that shown in Fig. 9) should not be presumed unreasonable given the fact that we plan to land a person on Mars.

A connection similar to that pro­posed in the National Institute of Standards and Technology's program was tested successfully by Dr. Robert Park in New Zealand over 10 years ago. Fig. 10 shows the type of con­struction Dr. Park envisioned. A full scale beam-column test program was developed. The moment displace­ment relationship (Fig. 11), though not cycled through intermediate duc­tilities, clearly indicates that a rea­sonable level of ductility is available. The behavior of a cast-in-place joint is shown in Fig. 12 for comparison.

44

L6X6 SIM . TO BELOW W/ 2-1"e BOLTS IN OVERSIZE HOLES

FIBER REINFORCED NON-SHRINK GROUT 10, 000 PSI

4"e CORRUGATED METAL DUCT W/ PT STRANDS

SHIM AS REQ'D

L6X6X1X1'-6" LONG W/ 2-112" THICK GUSSETS W/ 1"e BOLTS @ 6" D.C. (3 TOTAL) IN SLOTTED HOLES TEMPORARY CORBELS TO BE REMOVED AFTER STRANDS ARE STRESSED EACH SIDE

Fig. 8. Precast beam and column connected by post-tensioning.

Behavior at or near ultimate for the precast concrete beam is crushing of the toe. This type of failure is charac­teristic of the post yield behavior of cast-in-place joints. The failure is ini­tiated by cover spalling generated by cyclic yielding (elongation) and buckling of the now overstrained compression reinforcement. This spalling and toe degeneration is fur­ther aggravated by the high level of compression and shear transfer in this region.

v

The introduction of strand as ten­sion reinforcement reduces or elimi­nates the bar buckling problem at the expense of producing higher com­pressive stresses in the concrete. Ul­timately, toe deterioration will always be the limit state for a ductile moment connection. The compression/shear transfer mechanism in the toe must be dealt with in an imaginative way.

A fiber reinforced epoxy grout coupled with a strain limiting bearing pad cast in either the beam or the

~--------~-11-~--~-~--------~

Fig. 9. Precast beam and column connected by epoxy binder.

PC! JOURNAL

- .... - PRETENSIONED TENDONS AND/OR NONPRESTRESSED REINFORCEMENT---

MOR TAR JOINT~' /~MORTAR JOINT

~ I c \. I ~

~ ..

\_POST TENSIONED TENDONS MIDSPAN

NOTE:: ONLY LONGITUDINAL STEEL IN MEMBERS SHOWN

Fig. 10. Structural configuration for constructing continuous frames from precast concrete elements.

BEAM END DISPLACEMENT, INCHES -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14

160

1- 120 UJ UJ LL.

a. 80 1-1 ~

....: 40 z w

0 2: 0 2:

w -40 (!) z .... :I: -eo 2: ~

I I I I I I ,; v ~ -~ NEGATIVE MOMENT AND DISPLACEMENT

CORRESPOND TO DOWNWARD LOAD AT BEAM END /jVI l.._

I / )

L v v v . .

~ I~

v .. .. .,-N

/ .___ ~ I ~4-3/8" DIA. STRA NOS

w -120 m

-160

.. 0!0

~3/8" DIA. STIRR ) I I .. ..

l ~2-12w/0 . 276" DIA v ....- ..........

UP

.. luL!j ~ I~ .. .. 11 x 's .,-N

Fig. 11. Measured beam end displacement vs. beam plastic hinge moment.

May-June 1990 45

U) Q. H ~

u.i u a: 0 I.J...

60

30

0

-30

-60 -1.0

td+l

-.5 0

STROKE, IN.

Fig. 12. Measured beam displacement vs. applied load.

column would better distribute the strain and thereby control concrete deterioration (Fig. 13). The develop­ment of imaginative cures must take precedence over determining the limit state of unimaginative load transfer concepts.

World Experience Precast concrete construction is not

only being used, but it is also being promoted in Japan on high rise build­ings. The Japanese counterpart to the U.S. PRESSS program is charged with developing a national design standard for precast concrete con­struction by 1992. This lack of a na­tional standard has not kept Japanese contractors from using precast con­crete. Taisei Construction Company uses what they call "layered construc­tion" to reduce construction time.

This system consists of precast columns, and a column joint shell developed so that it can receive a solid beam which in turn supports

46

form slabs. Seismic bracing is provided by a ductile frame. All beams contribute to the seismic brac­ing of the building. This bracing sys­tem is designed to emulate cast-in­place construction. Features that are significantly different from U.S. prac­tice include the splicing of all column bars immediately above the floor with mechanical splices.

The technical justification for precast systems in Japan is provided by test. Japanese contractors test full scale subassemblies to a prescribed loading sequence in order to establish that the level of component ductility exceeds that which is required by the design criterion used. The ultimate design strength for precast systems is essentially the same as comparable cast-in-place systems.

The design criterion in Japan ap­pears to accept a prescribed level of subassembly ductility as proof that sufficient building ductility exists. The equivalence of building "duc­tility and energy absorption" current-

5 x's

8/d -2.31

p •1.031

s •3.25"

d •13"

VELOCITY •0 . 1 IN. /SEC .

.5 1.0

ly required by U.S. codes must be reduced to a subassembly behavior criterion as it is in Japan. Compara­tive subassembly tests are the only logical way to determine whether the yield level prescribed for cast-in­place concrete is appropriate for precast concrete construction.

New Zealand codes, for example, require certain precast concrete brac­ing systems to be designed to a higher yield (1.25x) level than comparable cast-in-place systems. Clearly, there has been more focus abroad on the development of precast concrete building systems and design criteria than has occurred in the U.S.

The U.S. precast concrete industry must promote a performance type design philosophy if it is at all inter­ested in the structural system market in regions where seismicity is a con­sideration. Boston, Massachusetts, and St. Louis, Missouri, are only two examples of places which now re­quire seismic considerations. It is ex­pected that many other areas of the

PCIJOURNAL

ELASTOMERIC BEARING PAD

_ ~FIBRE REINFORCED L_GROUT

BONDED P/T STRAND

Fig. 13. Post-tensioned ductile beam-column connection.

IN SITU CONCRETE

Fig. 14. Horizontal bars connected by overlapping loops.

Fig. 15. Horizontal bars connected by welding.

May-June 1990

IN SITU CONCRETE

(a) FLOOR SLAB BARS CONNECTED BY OVERLAPPING LOOPS

(VERTICAL BAAS NOT SHOWN)

(b) FLOOR SLAB BAAS CONNECTED BY WELDING (VERTICAL BAAS NOT SHOWN)

aL !IECTIOII IHI

(c) POSSIBLE METHOD OF CONNECTING VERTICAL BAAS

Fig. 16. Typical horizontal joints between prefabricated large panels.

47

HiiH STRENGTH GROUT

FOAM PLASTIC COLLAR DRY PACK MORT

(a)

OUTLET TUBE

INLET TUBE

PLASTIC BEARING STRIP

Fig. 17. Horizontal joints for prefabricated wall panels (hollow-core slab floor systems).

United States will require such provisions.

The U.S. Scene The U.S. construction industry

needs a viable precast concrete in­dustry if it is to meet the needs of so­ciety. Interest in the precast concrete research necessary to establish a vi­able precast concrete industry is in­creasing, but the application of this research currently awaits code ap­provals.

Precast panel high rise construc­tion, for example, should supply a significant portion of our residential needs. Marketing precast concrete

wall panels for high rise construction in seismic regions is impossible given the design constraints of current codes. The principal obstacles are: • Boundary elements (Fig. 6)3 are re­

quired when PulA+ Mu IS> 0.2 f~ (always the case).

• Two curtains of steel are re~red in shear walls3 when Vu > 2 --Jj(. The new masonry code ( 1988

UBC, Chapter 24)3 does not require boundary elements, nor does it pre­scribe an arbitrary limit for single cur­tain shear reinforcement. The mason­ry industry was able to establish by test, as the concrete industry and the precast concrete industry could, that distributed reinforcement provides

PRE LOAD

&ROUTED JOINT SPACE---...

APPLIED SHEAR LOAD

THICKNESS • 200 mm

0 CD ([) ....

25.4 mm • 1 inch

Fig. 18. Specimen configuration for shear transfer test program.

48

(b)

BOLTED STEEL COUPLING ELEMENT

more ductility than a wall which is over-reinforced by large boundary elements.

The connectivity of precast panels has been studied in many countries which recognize the economic need for precast concrete residential build­ings. The key to innovation in panel construction, assuming that the design hangups of the preceding para­graph can be solved, lie in the devel­opment of connectors which are ef­fective from a cost and engineering perspective. Connections such as those shown in Figs. 14 through 17 are suggested in the literature and should get the job done, but the result­ing system is probably not affordable.

What are the innovative concepts in large panel construction that require development? The transfer of flexural capacity requires continuity of verti­cal reinforcement. Vertical post­tensioning is an obvious candidate as are coupled threaded bars. Post­tensioning systems utilizing strand should have a decided advantage since the introduction of any rigid bar or connector in the end of a thin panel will undoubtedly promote toe failure and, as a consequence, reduce the strength and ductility of the system. The viability of distributed boundary reinforcement must be established.

Shear transfer along horizontal and vertical joints must be provided. Most recently, shear transfer was studied by Foerster, Rizkalla and Heuvel.4

The specimens were tested as shown in Fig. 18. Sliding friction along an unreinforced joint subjected to a con-

PCI JOURNAL

1600 APPLIED COMPRESSIVE 290 0 SP 11 STRESS 580 e SP 21

1400

1200

z 1000 ~

0 800 < 0 600 ...J

400

200

0

700 psi

500 psi

25.4 1111 • 1 inch 1 kN • 0 . 225 kips

-2 2 6 10 14 22 DISPLACEMENT (mm)

Fig. 19. Load-displacement relationship for smooth joint.

stant compression is higher than one might expect [85 percent of the ap­plied compressive load (Fig. 19)]. Whether or not this can be translated into a shear transfer criterion for the compression region of flexing panel needs to be established. Interestingly, a keyed joint (Fig. 20) increased the pre-slip shear transfer by about 50

DIAGONAL STRUT

percent but post-slip shear transfer by only 20 percent. Keyed joints seem hardly worth the effort especially if, as is the usual case, a floor system must interrupt panel continuity (Figs. 16 and 17).

Room for innovation exists. Ad­hesive grouts or grouts which allow for controlled deformability should

improve joint performance. Strain controlling bearing pads might also be the answer to the toe failure mechanism.

Shear wall reinforcement concepts and continuity requirements devel­oped for cast-in-place concrete need to be discarded and replaced by con­cepts which are appropriate for precasting.

The most encouraging step toward developing a viable precast concrete structural system is the initial phase of the NIST program discussed pre­viously. The test program was devel­oped with a stated goal of demon­strating and comparing "equivalent ductility and energy absorption" of comparable cast-in-place concrete subassemblies and precast assemblies connected as shown in Fig. 8. The beam/column assemblies were devel­oped from actual designs of a 15-story structure located in Seismic Zone 4 (Fig. 21).

The results of these initial tests will soon be available,5 but the encourag­ing aspect is that the precast concrete system was able to sustain larger dis­placements and displacement duc­tilities than their cast-in-place compa­rables. Load-displacement curves for

IN SITU CONCRETE PROTRUDING STEEL

WALL PANEL

(b) PLAN SECTION

(a) ELEVATIONS

(NOTE: REINFORCING BARS NOT SHOWN -SEE PLAN SECTION)

Fig. 20. Typical castellated joint between prefabricated large panels.

May-June 1990

en en ~ l­en

~ UJ :I: en

CASTELLATED JOINT

SHEAR SLIP

(c) TYPICAL SHEAR STRESS-SHEAR SLIP RELATION FOR MONOTONIC LOADING

49

r-A

16" ® L

-A

41 3/4" (37") •

47 3/4" (43").

8

L ~

®

8

_j . 9-13 -V'"i w

r 32.

j_ t

26"

l

Loading direction l North-South

1" Post-tensioning bar 1-112" Corrugated duct

8- #3, [0 ·--4-#4

Top & Bottom

0.207" smih. wire 8 - 0.207" ~

@ 3 1/3" O.C. smth. wire

~-0.207" smth. wire

- ,.,?, 5- 13 @ 1 1/3" O.C. 2- #4

MONOUTHIC POST-TENSIONED

• Numbers in parenthesis are for the post-tensioned specimens.

Fig. 21. Test specimen for ductile frame (NIST program).

66.0

39.6

'[ 13.2 ;g, 0 c(

g -13.2

-39.6

-66.0

-2.10 -1.26 -0.42 0.42 1.26

DISPLACEMENT (inches)

Fig. 22. Load-displacement curves (cast-in-place joint).

50

2.10

SECTION A-A SECTION B-B

45

Cii' 15 a. ;g. 0

~ -15 ...J

-45

-75 '-=.._......jl......<:..L.:.... __ ....u.... __ __. ___ ..__ __ __.

-1.75 -0.85 -0.05 0.95 1.85 2.75

DISPLACEMENT (inches)

Fig. 23. Load-displacement curves (precast joint).

PCIJOURNAL

the two subassemblies are shown in Figs. 22 and 23.

The precast concrete system was also capable of sustaining signifi­cantly more cycles to failure than the cast-in-place specimen. Consequent­ly, the total energy absorbed, despite the pinching of the precast beam hys­teresis loops, was equivalent. Many more specimens will be tested in this program so we should not rush to judgment, but rather be encouraged to try new concepts.

Industry leaders continue to ad­vance the innovative use of precast concrete. Lloyd Compton of L.A. Compton Group is in the process of developing a ductile fascia panel to be used to brace midrise (4 to 6 story) buildings. The concept is developed about an adjustable casting capable of dealing with the tolerances which must be provided for in precast con­crete construction.

May-June 1990

Innovation is synonymous with precast concrete. Given some lati­tude, precasters will surely develop innovative systems. It remains then to be able to establish a technology which promotes this innovation.

Conclusion Precast concrete buildings can and

should be constructed in seismic areas. A technology compatible with precast concrete construction needs must be developed. This can be ac­complished only if constructors and precasters take an active, open minded, and yet stubborn part in the development of forthcoming research programs. The thrust must be to develop imaginative systems that im­prove performance and solve prob­lems as opposed to merely establish­ing limit states for poor design con­cepts.

REFERENCES

1. Englekirk, R. E., "Towards a More Effec­tive Use of Precast Concrete Cladding," Proceedings, Architectural Precast Con­crete Cladding - Its Contribution to Lateral Resistance of Buildings, Precast/ Prestressed Concrete Institute, Chicago, IL, 1990,pp. 4-13.

2. Englekirk, R. E., "An Analytical Ap­proach to Establishing the Seismic Resis­tance Available in Precast Concrete Frame Structures," PCI JOURNAL, V. 34, No. l, January-February 1989, pp. 92-101.

3. Uniform Building Code, Section 2625 (f), International Conference of Building Officials, Whittier, CA, 1985.

4. Foerster, H. R., Rizkalla, S. H., and Heuvel, J. S., "Behavior and Design of Shear Connections for Loadbearing Wall Panels" PCI JOURNAL, V. 34, No. 1, January-February 1989, pp. 102-119.

5. Cheok, Geraldine S., and Lew, H. S., "Seismic Performance of 1;3 Scale Post­Tensioned Precast Beam-Column Con­nections;· : to be submitted to the PCI JOURNAL for publication consideration.

51