ML20027C597

Applicability of Arching Theory to Unreinforced Block Masonry Walls Under Earthquake Loading

ML20027C597
Person / Time
Site:	Haddam Neck File:Connecticut Yankee Atomic Power Co icon.png
Issue date:	08/31/1982
From:	Harris H DREXEL UNIV., PHILADELPHIA, PA
To:
Shared Package
ML20027C596	List: ML20027C595 ML20027C597 ML20027D709
References
NUDOCS 8210190160
Download: ML20027C597 (10)

Text

.

Madiud 4

-.. ~

I 4

APPLICABILITY OF ARCHING TEEORT TO UNREUPORCED BLOCK MASO'NRY WALLS UNDER RARTBqUAKE LOADING 2

submitted to Franklin Research Center Philadelphii, Pennsylvania by Dr. Earry G. Earris and Dr. Ahrad A. Hamid l

Departmut of Civil Engineering l

Drexel University Philadelphia, Pennsylvania 19104 August 1982 i

c w h 0 I6 6 g3iot

~. %A

IN11t0 DUCTION 3

j Unreinforced masonry valls have shown, under certain conditions, h'igh-er resistance to out-of-plane loads than would be predicted based on con-ventional bending analysis.

B is additional capacity is developed by'creat-

~

ing Isrge in-plane clamping forces, thereby forming a three hinged arch l

^

mechanism after mid span and support flexural cracking has occurred (6,13,

' 14). ne most important conditions for the arching mechanism to develop are the existence of rotational restraint at the boundaries and the prevention i

of gross sliding of the vall at support sections.

l Two different types of arching can occur depending on the boundary l

conditions:

rigid arching if the vall is tightly fitted between supports l

and gapped arching if the vall is separated from one support by a small gap due to poor construction and/or vall shrinkage (6). n e mechanisms of the l

two types of arches are shown in Fig. 1.

j For rigid arching under increasing uniformly applied lateral load, the vall deflects and flexural cracks are. developed at the support sections and at mid-span where maximtue bending moments are present (based on an elastic analysis assuming a fixed and condition). De vall segments on bcth sides of the mid-span crack tend to rotate about the edge supports. as the vall deflects under load, see Fig. 2.

Because of the finite vall thickness, this rotation results in the vall ja=%g against the edge supports and a wedging action is developed.

h is mechanism induces an arial compression force in the vall.

D e compression forces acting on the contact areas at the edges and mid-span sections provide internal coupling (P(u). r(u)) to resist l

higher external loads, see Fig. 2.

D e effectiveness of the developed com-pression force, P, depends on the depth-to-span ratio of the vall, contact t

i 1

.l n-,,--,m, n

,a--.

..,-.m, w,m-,,..,,..,,

-..+, -

- - - ~ -

nninnnnnp//// ////////

. ///////0///////////////////

or

_q-3 v

Q g

j k

a.

H i

I I

l Direction of 1.eading,

yCl Direction of 14adia C

2 2

j S

I T

I 7s I

l 1

I 8

///////// ////ti t////// ////

.////////////// ////////// / / / /

A A

R!GIO ARCw!NG GAPPG ARCH!NG Fig.1-Mechanisin of Rigid and Capped Arching Masonry Walls ( )

2L z'

' APPLIED LOAD IIIIIIIIII44444IIIIIIIIIII E

RIGID RI

/ GID SUPPORT Ca-SUPPORT INg

.g~ x.m.

D,r a

c 4

I.?q N s.3.

p CRuSHitjg Y

gges#

E' O

e Fig. 2-Arching Action Between Rigid Supports ( }

~

2

l area, maximum compressive strain of masonry, distribution of compressive strain and size of initial gap between the edge support and the val 1.

Static test results of an experimental program conducted at MIT (1) indicated that unreinforced solid brick masonry beams tha't ars confined between rigid edges could resist lateral loads many times higher than those acting as a cantilevf,r or simple beau valls. Failure initiated by local crushing of masonry at the hinges where rotation continued until contact area diminished, thereby internal couple (resisting lateral load) vanished and vall destroyed..

A static test program was conducted at URS (7) to de-termine the resistance of hollow concrete block masonry beams confined be-tween rigid edges (rigid arching beams). A web-shear failure mode was 'ob-served which indicates the relatively low resistance of hollow block mason-ry under combined shear and compression loading compared to either solid or fully grouted block masonry (4,5). R ese test results indicate the sensitivity of the failure mode and consequently the resisting load of rigid arching beams to the material properties and geometry of cr s sec-tion.

The Arching neory can be extended to a vall supported on four sides by assuming a reasonable cracking pattern due to out-of-plane loads (13).

g.

As a result of their experimental study, Ctbrielsen and Kaplan (6) conclud-ed that rigid arching valls under static loading can resist 6 to 8 times the loads that gapped arching valls can. Also the energy absorbed by gap-i t

ped arching valls was only about one-thirteenth that absorbed by a sfwflar l

rigid arching vall.

Gapped arching valls are still, however, considerably.

stronger than either cantilevered valls or valls mounted as simple beams.

The British code (2) nilows the use of arching theory to deterv.ine.

3

.n_,_.-r._er.

_%,.wv._ _, _.. -, -..

e

the resistance of solid brick masonry valls to' lateral loads. Two' eases "

are specified; vertical arch between floor slabs and horisontal arch be-tween vertical supports such as cross-valls and frames. C,apped arch is not allowed and for non-rigid supports, =fnf== axial dead load is required for vertical arching.

It has to be noted that the code provisions are applied to solid masonry untler static or blast loading and is not necessarily appli-

. cable to cyclic dynamic loading due to earthquakes.

MASONRY WALLS UNDER BLAisT LOADING An extensive dynamic test program (6.7,16) sponsored by both the De-fense Civil Preparedness Agency and the Veterans u=fnf stration was con-ducted at URS research company to study the resistance of brick walls to uniform blast loading.

Full scale valls mounted on a steel frame vere sub-jacted to blast vaves in a large shock tunnel. These valls acted as simple

^

plates and exhibited modified arching behavior once flexural cracks began.

The steel frame' became a perimeter restraining ring causing archin's thrusts to develop.

The valls had a lead carrying capacity approximately twice as strong as conventional simple plates.

Some valls were loaded and reloaded many times.

The pulses had a positive loading pulse of about 0.1 see.

Iong, followed by a negative loading phase of about the same duration, but of much lower intensity.

These valls resisted the multiple pulses without failure which indicated the ability to withstand mildly reversed loading cycles.

It has to be noted, however, that this loading condition was not a complete reversed one because of the decayed intensity of succeeding

' cycles.

Therefore, it is not a true representation of the fully reversed nature of earthquake loads.

4

o SEISMIC TESTS ON SINGLE-S'EORT MASONRY B00SES AT UC, BERKZLET A series of shaking table experiments were conducted at the University of California, Berkeley Earthquake SM1stor Laboratory du ing the period 1976-1979 (3,9,10,11).

This experimental investigation was aimed at de,-

termining the reinforcement requirements for single-story dwellings in Uni-form Building Code Seismic Zone 2 areas of the United States. Tests b re

}

conducted on four ma'sonry houses, with both unreinforced and partially re-inforced vall pan'els, assembled to form 16 ft. square models of typical masonry houses.

The masonry units utilized in the construction of all test i

structures were full-sized now h 1 6 in. concrete masonry and clay units.

1 Each hease was provided with a timber truss roof structure to which wei' hts g

vere attached so as to obtain realistic loads on the bearing valls.

The out-of-plane unreinforced vall panels, which may have some relevance to

\\

arching action, were 8' - 8" high and ranged in vidth from 3' - 4" solid piers to 16' valls with a variety of door and vindow cutouts. Although the unreinforced out-of-plane valls were able to sustain ground accelerations as high as 0.45g, cracking occurred in th'e 50 to 70 percent range of the height region with no apparent consistent true arching action mobilized. - This type of mechanism was probably precluded from materializing due to the relatively flexible top and bottom vall supports provided in. the tests. r-

~ ~

The lovest intensity shaking that caused cracking of an out-of-plane vall was found to be 0.25g in the tests of House 1.

According to the authors (3):

"Unreinforced out-of-plane valls continued to perform satis -

factorily af ter cracking for several tests of increased intensity, but the displacements of these valls generally became excessive during tests with peak accelerations in excess of 0.4g. ' These large displacements involved 5

s =y.

me sa ee --

• 7+ p* g w., y +.

• ye p..a+_e-o, f+-g.=,gg.9-3 9g,y +,,. g g e 4 g

.g,g y

p,.-,em..es, ma se

,+4g is a s ee saw

. i up 's

~

s.

s hinging at.the horizontal crack line and exhibited poten'tial inhtybility,

\\

s s

although actual collapse did not occur in any test."

In addition, these researchers found that the test procedule used 1+1 d

the four single-story houses tested has one important deficiency in tha.t v

only one horizontal conponent of the ground acceleration was inv'estiga ed.

Quoting from their donclusions' (3):

"Because the unreinforced valls sub-,,

y y

w

.jected to intense out-of-plane motion responded by signifies;st cracking"and

" hinging" out-of-plane, it is probable that the capacity of such walls to i resist simultaneous in-plane shear forces would be greatly. diminished. For J

this reason, the second horizontal component of input motion would be ex-pected to have a detrinental effect on unreinforced masonry walls, and

~

based on judgment, it seems reasonable to assume conservatively that this effect is equivalent to an increase in the intensity of single horizontal component motion of between 30 to 50 percent. On the other hand, the partially reinforced masonry valls exhibited little distortion regardless 7

of the intensity of in-plane or out-of-plane swag; thus, there would be little tendency for coupling of the in-plane and out-of-plane effects."

A study of the dynamic displacement patterns of tbs out-of-plane valls indicates that the arching mechanism was not present (9). Kariotis, et al.

(12), reported that the reason for the increased resistance of the 'out-of-plane valls under earthquake loading is not known and suggested a testing program to investigate this behavior. Results from such a testing program are not available.

.N MASONRY WALLS IN UCLEAR PDFER PLANTS Unreinforced, ungrouted or parHally grouted, and solid masonry valls -

i in nuclear power plants represent a large variation of configurations,

\\

I 6

,_i7

_ ~ _ -

p s

u...-,.

i pA+"

.s r

c y

c

.,.w t

constructio3 details and meth6ds of support and restraint a[kheir boda-daries.

Both dingle vythe and undti-vythehstruction is present in such s

l n-1[edditions ahtach$catsSf v'arious types introduce 'pentric structures.

y f.

~

inertial masses sad points of local' stress concentration in sonie applica-j'

,,,.v'

. tions.

Thus y:estions as to the, applicability of the arching theory for s

n, f

w -

t these valls trise %

M,

.\\s i; The supporting'syiscerns forathese out-of-plane valls ebow a large 6.s i,

varietz of botAdtry con 41tiorg fr'on -ci,ose to; full fixiQ. to frea. edges.

's

^

When c.oncrete or steel frs'me's tre.previded 'the masonry valls act essen-s

.\\,,M e'

g i.

'tf ll? in su in-fill es.pacity'. Masonry dalis can spd.n in either a predomi-a Q

y,

'G u,

g na;:tly vs?tical or predominantly horizontal direction or can behave as 'true E

lates depending on the amngement and fixity of their supports.

Surveys of existing power plants reveal that mortar joint shrinkage s

c frequently 1 caves a gap at the top of nasonry valls which would imply that

.a gapped arch behsvior is more applicable.

Static tests on valls with a

,T,,

s gap have been performed by Gabrielson and his associates (16) in which he has coricluded that:

" Walls with a gap, however, were only slightly strong-er than non-arche.d valls."

l r

l An important consideration in these types of valla is the reversed

.- load nature of the out-of-plane earthquake inertia forces. Except for the very limited Berkeley tests (3), no data is presently available on the per-formance of unreinforced masonry valls under earthquake loading. In these tests the earthquake ground motion was applied with components in either the in-plane or the out-of-plane directions. It has to be noted that even if valls are not designed in nuclear power plants to' carry in-plane shear, 1

both in-plane and out-of-plane-inertia loads vill be developed due to the

~

random direction of earthquabte motion.

7

..er

.yy--

,. =

.v.,,

+-

y G

CONCLUSIONS AND R.LCO F2NDATIONS A review of the available information on the applicability of arching theory to unreinforced, ungrouted and partially grouted, and solid rEsonry valls under out-of-plane earthquake loading has been presented. 'The docu-mented data in the literature does not give enough insight for understand-ing the mechanics apd performance of unreinforced ' masonry valls under cyclic fully reversed dynamic loading. Therefore, the applicability of arching theory to masonry valls in nuclear power plants is questionable.

It is evident that more data is needed to quantitatively detezuine the effects of different vall geometry, material properties and boundary conditions on unreinforced masonry vall resistanca to earthquake loading. It is recomend-ed that a testing program be undertaken to investigate the applicability of arching theory to unreinforced block masonry valls in nuclear power plants.

O 9

e G

0 C

m 0

e 8

+

RIFERENCES 1.

Ano=ymous, " Behavior of Vall Panels Under Static and Dynanic Leads ~ II,"

Department of Civil and Sanitary Engineering, MIT, January 1954.

2.

British Standards Institution, " Code of Practice for Structural Us.e of hsonry," BS5628:

Part 1. 1978.

3.

Clough, R., hyes, R. and Gulkan, P., " Shaking Table Study of Single Story hsonry P.ouses, Volume 3:

Suzanary, Conclusions and Reconsnenda-tions," Earthquake Engineering Research Center, Report No. UCB/EERC-i 79/23, College of Engineering University of California, Berkeley, CA, September 1979.

4.

Eanid, A. and Drysdale, R., " Concrete hsonry Under Combined Shear and Compression Loading," Proceedings of the American Concrete Institute, Volume 77, No. 5, September-october 1980.

5.

Eamid, A. and Drysdale, R., " Proposed Failure Criteria for hsonry.

Under Biaxial Stresses," Journal of the Structural Division, Proceed-ing of ASCE, volume 107, ST8,. August 1981.

6.

Gabrielsen, B. and Kaplan, K., " Arching in hsonry Walls Subjected to Out-of-Plane Torces," NBS Building Science Series 106, National Work-shop on Earthquake Resistant Easonry C5'nstruction, Septe:nber 1977.

7.

Gabrielsen, B., V11 ton, C. and Kaplan, K., " Response of Arching Walls and Debris from Interior Walls Caused by Blast Leading," Report No.

7030-23, URS Research Company, San Mateo, CA, Febntary 1975. '

8.

Gabrielsen, B. and Wilton, C., " Shock Tunnel Tests of Arched Wall Panels," Report No. 7030-19, URS Research Company, San hteo, CA, December 1974.

9.

Gulkan, P., h yes, R. and Clough, R., " Shaking Table Study of Single Story Houses, Volume 2: Test Structures 3 and 4," Earthquake Engineer-ing Eesearch Center, Report No. UCB/EZRC-79/24, College of Engineering, University of California, Berkeley, CA, September 1979.

(

10.

Gulkan, P., h yes, R. and Clough, R., " Shaking Table Study of Single Story h sonry Houses, Volune 1:

Test Structures 1 and 2," EERC Report No. UCB/EERC-79/23, College of Engineering, University of California, Berkelev. CA. Sentember 1979.

2 11.

Culkan, P..and hyes, R. and Clough, R., " Strength of Timber Roof Connections Subjected to Cyclic Loads," EERC Report No. UCB/EERC 78/17, College of Engineering, University of California, Berkeley,,

l CA, September 1978.

l

~

12.

Kariotis, J., Eving, R. and Johnson, A., " Methodology for Mitigation of Seismic Eazards in Existing Unreinforced Masonry Buildings,"

9

~

Proceedings Conference o'n Research in Progress on, Masonry Construction, '-

Marina Del Rey, CA, March 1980.

13.

Lefter, J. and Colville, J., " Evaluating the Earthquake Resistance of Existing Masonry Construction," Proceedings of the First Canadian Ma.sonry Sy=posium, Calgary, Alberta, June 1976.

14. McDowell, E., McKee, K. and Sevin, E., " Arching Action Theory of -

Masonry Walls," Journal of the Structural Division. Proceedings of ASCE, ST2, Paper 915, March 1956.

15. McKee, K. and savin, E., " Design of Masonry Walls for Blast Ioading,"

ASCE Transaction, Paper No. 2988, Volume 124, 1959.

16. Wilton, C. and Gabrielsen, B., " Shock Tunnel Tests of Preloaded and Arched Vall Panels," Report No. 7030-10, URS Research Company, San Mateo, CA, June 1973.

5 8

e 9

ao G

10

.