UNCLASSIFIED

AD NUMBER

LIMITATION CHANGESTO:

FROM:

AUTHORITY

THIS PAGE IS UNCLASSIFIED

AD841792

Approved for public release; distribution isunlimited. Document partially illegible.

Distribution: Further dissemination only asdirected by Office of Civil Defense, ArmyPentagon, Washington, DC 20310., OCT 1968, orhigher DoD authority.

ocd, d/a ltr, 23 nov 1971

·•·

THIS DOCUMENT IS BEST QUALITY AVAILABLE. THE COPY

FURNISHED TO DTIC CONTAINED

A SIGNIFICANT NUMBER OF

PAGES WHICH DO NOT

REPRODUCE LEGIBLY,

i ._

I I I I I I f

-

IIT RESEARCH INSTITUTE Technology Center

Chicago, Illinois 60616

IITR1 Project J6107

FinaL Report

This document may be further distributed by any holder only with specific prior approval of the Office of Civil Defense.

SUMMARY

FRAGMENTATION OF REIlSfFORCED CONCRETE SIABS

Subcontract B-70942(4949A-34)-US OCD Work Unit 3322B

by

Ralph L. Barnett Paul C. Hermann

I I

for

Office of Civil Defense Office of the- Secretary of the Army

Washington, D„ C. 20310

OCD Review Notice

This report has been reviewed in the Office of Civil Defense and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Office of Civil Defense.

I

W

October 1968

I t

NUMMARY

FRAGMENTATION OF REINFORCED CONCRETE SLABS

This report presents the results c - an Investigation of

the debris producing characteristics of reinforced concrete i

wall, flocr and roof panels. The principal conclusion of this

study is that reinforced concrete (R/C) slabs do not constitute

a significant source of debris.

It is convenient for analysis to identify two types of

fr-W"?nts. The first, called p-imary fragments, are caused by

structural action such as bending and membrane tension. These

pieces are prismatic with two sides formed by the Lop and bot-

tom slab faces. All other fragments are called secondary;

secondary fragments may be quite irregular in shape and are

caused by, for example, stress waves or shear stresses at the

steel/concrete interfaces. For primary fragments to occur, all

of the reinforcing steel around their edges must be fractured,

whereas secondary fragments can and normally do arise without

such fracturing of the steel reinforcement.

The first consideration in the study was the loading antic-

ipated on R/C slabs. When used above grade, R/C slabs are

most frequently used as floor or roof panels. The horizontal

orientation for these applications precludes the development

of significant net transverse loads in the Mach region created

by a nuclear detonation. Large transverse loads are found in

ceiling slabs which cover basements or in roof and floor slabs

exposed to high altitude bursts. In these circumstances the

tenacity of R/C slabs operates to minimize their total loading.

Initial cracking or breaching of the slabs allows the overpres-

sure to build up on the downstream side. This diminishes the

net transverse loading which in turn reduces the severity of

the post cracking behavior.

11T RESEARCH INSTITUTE

S-l

I !

•

I I I I

A second consideration of this study was the transport of

R/C fragments which might be formed. The downward pressures

acting on basement ceiling slabs or roof and floor slabs ex-

posed to high altitude bursts produce a downward velocity that

tends to restrict any fragments to the vicinity of their parert

structures. If large concrete fragments were to form, both

field data and theoretical predictions indicate that they would

be transported only short distances from their initial locations.

No tendency to form small primary fragments was observed. Sec-

ondary fragments can be produced which will have significant

horizontal displacement; however, they are formed under large

net transverse forces whose conditions do not favor horizontal

transport. The amount and size of secondary fragments is

small and would not contribute significantly to street conges-

tion, but would be significant in evaluating personnel vulnera-

bility.

The third area of study or. this program was the actual

fragment formation of R/C panels. Simply supported one-way

and two-way slabs can be distorted enough to allow them to

push through their supports. Under these conditions they will

remain intact and form one large primary fragment. Also, when-

ever the slab dimensions make it stronger in bending than in

shear, it is possible to produce forces of sufficiently large

magnitude to cause an entire r>lab to punch out by shearing

along its periphery. In either case, such large fragments

would fall to the ground long before any reasonable horizontal

motion could be imparted to them.

In all the cases observed, secondary fragments took the

form of chips and chunks of concrete broken off of the tension

face of the slab below the tension steel. This layer of con-

crete is usually only 0.75 in. in thickness - the minimum cover

specified by the ACI code. The concrete fbove the tension steel

remains intact. The splitting cracks, caused by wedging of the

MT RESFARCH INSTITUTE

S-2

I 1 I I I 1 i I I I I I I I I I I I I

,"■

. "--..'-

deformed reinforcing bars, occur over the rods on the tension facer This feature limits the maximum size of a secondary fragment to that of a 0.75 in. plate with the dimensions of the reinforcing grid.

The conclusion of this study is that R/C slabs do not con-

tribute significantly to debris accumulation. This conclusion

was unforeseen at the start of the program which was in fact

directed at predicting the amount of debris accumulated from

R/C panels. The justification both analytical and experimental

for this unexpected conclusion, is contained in this final report.

•

11T »SSEA8CH INSTITUTE

5-3

f

IIT RESEARCH INSTITUTE Technology Center

Chicago, Illinois 60616

IITRI Project J6107

Final ReDort

This document may be further distributed by any holder only with specific prior approval of the Office of Civil Defense,

FRAGMENTATION OF REINFORCED CONCRETE SLABS

Subcontract B-70942(4949A-34)-US OCD Work Unit 3322B

by-

Ralph L, Barnett Paul C. Hermann

for

Office of Civil Defense Office of the Secretary of the Army

Washington, D. C. 20310

OCD Review Notice

This report has been reviewed in the Office of Civil Defense and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Office of Civil Defense.

October 1968

I

yp ——»-

.. - - "

I I I r

i

i

IIT Research Institute Project J6107

FRAGMENTATION OX REINFORCED CONCRETE SLABS

ABSTRACT

A reexamination of the blast effects literature from the

point of view of fragmentation leads to the conclusion that re-

inforced concrete slabs do not constitute a significant source

of debris in the postattack environment. Both the initial

orientation and the self-adjusting geometry of slabs minimize

their transverse loading. Also, the horizontal displacement

of potential slab fragments tends to be small because of their

high ballistic coefficients and/or high downward acting loading.

Finally, the steel reinforcing bars tenaciously tie the various

pieces of fractured slab to the supports and to each other even

at pressure levels as high as 100 psi.

. 11

■ ■-. ■

* n i ti

Chapter

I

II

III

IV

CONTENTS

INTRODUCTION AND CONCLUSIONS

LOADING

FRAGMENT TRANSPORT

RESPONSE

A. Introduction

B. Primary Fragments

C. Secondary 1'ragments

DISCUSSION OF FESULTS

BIBLIOGRAPHY

APPENDIX - REVIEWED DOCUMENTS WHICH DO NOT CONTAIN SIGNIFICANT WORK ON SLAB FRAGMENTATION

Page

1

5

13

24

24

44

58

71

75

76

Fi£rre

i

2

3

4

5

6

7

ILLUSTRATIONS

Reinforced Concrete Cubicles after Dynamic Loading

Strengthened Wood Frame House after Exposure to 2.6 psi Overpressure

First Floor Joists of Strengthened Wood Frai.ie House Exposed to 2.6 psi Overpressure

First Floor Joists of Strengthened Wood Frame House Exposed to 4 psi Overpressure

Steel Plate Girder Bridge with R/C Deck and Concrete Facing on Outer Girder Located 270 ft from Ground Zero at Hiroshima

Reinforced Concrete Building Located 0.10 mi from Ground Zero at Hiroshima

IIT Research Institute Aerial Photograph of Property Damage in Elkhart, Indiana by Tornado on April 11, 1965

7

9

10

11

12

14

16

111

i

IJESFARCH

ill

INSTITUT c

.

- - -

________^___^__^__^^__^_^_^^__^^^_^^_^__^^_

'

'

i

1

i

Figure

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

ILLUSTRATIONS

Heavy Wall Bearing Structure Located 0.34 mi from Ground Zero a*- Nagasaki

Machine Tools Behind Masonry Wall in Nevada before 10 psi Overpressure

Machine Tools after Exposure to 10 psi Overpressure in Nevada

Strengthened Wood Frame House after Exposure to 4 psi Overpressure

Wood Frame House after Exposure to 5 psi Overpressure

Ballistic Coefficient vs Range for Concrete Spheres

Longitudinal Cracking over Constant Moment Length (Middle Third) with Single Reinforcing Bar

One Story R/C Building with Sheet Roof Trusses Located 0.26 mi from Ground Zero at Nagasaki

Three Story R/o Frame Building with Brick Wall Panels Located 0.13 mi from Ground Zero at Hiroshima

Circular 60-ft High R/C Stack Located 0.34 mi from Ground Zero at Hiroshima

Motion Picture Sequence of Failure of Unreinforced Masonry Wall under 4.5 psi Overpressure

Preshot and Postshot Photographs of R/C Wall at 7.5 psi Overpressure Range

Effect of Wire Reinforcement on Tenacity of Concrete Slabs Exposed to High Explosive Loadings

Primary Yield Failure of a R/C Block ueam with Secondary Shear Failure through Concrete Fill and Masonry

Tension Steel Failures Near the Edge of a R/C Slab with 0.78 percent Steel Subjected to 11.0 psi Overpressure

Slab with Edges Fixed and Laterally Restrairad/ Dynamic Pressure Load of 101 psi

Multistory R/C Frame Building Showing Doweling Effects in the Beam Reinforcement

Page

17

18

19

20

21

23

25

27

28

30

32

33

35

36

38

40

43

i I T RESEARCH I N S T I TU r

IV

•-ftiflff

. -

I f

I

!

!

i

ILLUSTRATIONS

Figure Page

25 Tension Surface of R/C Slab with 0.78 percent Steel Sub-jected to 7.1 psi Uniform Static Loading 48

26 Tension Surface of R/C Slab with 0.78 percent Steel Subjected to 9.0 psi Overpressure 49

27 Tension Surface of R/C Slab with 1.0 percent Steel Subjected to 12.5 ps Overpressure 50

28 Tension Surface of R/C Slab with 1.17 percent Steel Subjected to 13.2 psi Overpressure 51

29 Reinforced Concrete Frame Building Showing Crushed Concrete Panel Walls on Side Facing Explosion (0.68 mi from Ground Zero at Nagasaki) 52

30 Preshot and Postshot Photographs of Precast R/C Channel Slab Rear Wall at 7.5 psi Overpressure Range 54

31 Compression Surface of R/C Slab with 0.78 percent Steel Subjected to 11.0 psi Overpressure 56

32 Reinforced Concrete Bridge with T-Beam Deck 0.44 mi from Ground Zero at Nagasaki 57

33 Splitting Cracks and Ultimate Failure. Ka) Ty:i^al,(b) and (d) Very Wide Beam (c) and (e) ^ith Closely Spaced Bars 59

34 Splice Failures (a) Violent Failure of No.8 Bars Lapped 36 Diam, (b) Shorter Splice with Cracked Concrete Removed tc Sho? Splitting Surface. Arrows Mark End Slips of Bars 60

35 Tension Surtace of R/C Slab with 1.0 percent Steel Subjected to 9.4 psi Uniform Static Loading 62

36 Tension Surface of R/C Slab with 0.78 percent Steel Subjected to 11.0 psi Overpressure 63

37 Tension Surface of R/C Slab with 1.17 percent Steel Subjected to 11.0 psi Overpressure 64

38 Compression Surtace of R/C Slab with 1.0 percent Steel Subjected to 12.5 p'ii Overpressure 67

39 Compression Chipping Produced in R/C Slab with 1.0 percent. Steel Subjected to 12,5 psi Overpressure 68

40 Cross Section of Mild Steel Plate Showing Multiple Scabbing 70

IIT RESEARCH INSTITUTE

i V

i ~_ - iws**fei«SWje«3j-

!

F

:

CHAPTER I

INTRODUCTION AND CONCLUSIONS

One of the central problems in a nuclear postattack envi-

ronment is the reestablishment of vital transportation links.

Many requirements must be fulfilled in reaching this goal; but,

few are as impressive as the removal of debris accumulations

from important streets and arteries. To perform this task with

the efficiency and dispatch demanded in emergency situations,

planning must embrace the full sc~>pe of considerations involv-

ing attack prediction, target description, fragment generation,

particle transport, and debris accumulation. This report rep-

resents only a small assignment in this overall picture. Spe-

cifically, it investigates the fragmentation of reinforced con-

crete slabs.

The principal conclusion of our study is that reinforced

concrete (R/C) slabs do not constitute a significant source of

debris. This result is based on three major observations that

are developed in the text. Before summarizing these findings

we shall identify two types of fragments which can be generated

from a slab. The first are called primary fragments and they

are caused by structural action such ac- bending and membrane

tension. These pieces are prismatic with two sides formed by

the top and bottom slab faces. They are of the type that might

be formed by a giaat cookie cutter being pressed into the slab

face. All other fragments will be called secondary fragments;

they may be quite irregular in shape and they may be caused by,

for example, stress waves or shear stresses at the steel/con-

crete interfaces. We note that primary fragments require that

all of the reinforcing steel around their edges be fractured;

secondary fragments can arise without fracturing steel.

The first of our three major observations deals with the

loading of R/C slabs. When these structures are used above

grade, they are most often employed as floor or roof elements.

NT RESEARCH INSTITUTE

-.

I Their horizontal orientation precludes the development of sig-

nificant net transverse loads in the Mach region created by a

nuclear detonation. Large transverse loads are found in ceil-

ing slabs which cover basements or in roof and floor slabs ex-

posed to high altitude bursts. In th^se circumstances ehe te-

nacity of R/C slabs operates to minimize their total loading.

Initial cracks or breaches in the slabs allow the overpressure

to build up on the downstream side. This diminishes the net

transverse loading which in turn reduces the severity of the

post cracking behavior. Cases are described in the text in

which slabs with fixed supports are broken into petals which

rotate about the fixed adges to minimize the area exposed to the

dynamic pressure in much the same manner as a weather vane.

Our second group of observations concern the transport of

slab fragments. To begin with, the downward pressures acting on

basement ceiling slabs or roof and floor slabs exposed to high

altitude bursts produce a corresponding downward velocity that

tends to restrict any fragments to the vicinity of their parent

structures. If large concrete fragments were to form, both

field data and theoretical predictions indicate that they would

be transported only short distances from their initial locations.

No tendency to form small primary fragments was observed. Sec-

ondary fragments can be produced whose ballistic coefficients

are sufficiently small to cause significant horizontal displace-

ment; howevei, they are formed under large net transverse forces

whose conditions do not favor horizontal transport. Furthermore,

the amount and size of secondary fragments is small and would

not contribute in a real way to street congestion. Their sig-

nificance i£ in the area of personnel vulnerability.

"ft" Ballistic Coefficient = weight/(drag coefficient) (projected area)

III RESEARCH INSTITUTE

r

-™p-— «■ litisnk

\ • - '-

I I

I

**■

We have indicated that the loading and transport charac-

teristics of R/C slabs conspire to minimize fragment formation

and horizontal displacement. Our final observations deal with

the actual fragmentation resistance of these elements. It is

reasonably clear that simply supported cue-way and two-way slabs

can be distorted enough to allow them to push through their

supports. Under these conditions they tfill remain intact and

form one large primary fragment. Also, whavtevör the slab di-

mensions make it stronger in bending than h\ shear, it is possi-

ble to produce forces of sufficiently large magnitude to cause

an entire slab to punch out by vertical shearing along its pe-

riphery. In either case, such large fragments would fall to the

ground long before any reasonable horizontal notion could be

imparted to them. No examples could be found where primary

fragments were formed; loadings as high as 120 psi were examined.

Only in rare and extreme circumstances were steel fractures even

obtained.

In all the cases observed, secondary fragments took the

form of chips and chunks of concrete broken off ox the tension

face of the slab below the tension steel. This layer 01 con-

crete is usually only 0.75 in. in thickness - the minimum cover

specified by the ACI code. The concrete above the tension sf:eel

remains intact. Prior to the formation of secondary fragments,

extensive transverse cracking occurs in the slab due to both

bending and splitting action. The main bending cracks form a

pattern which is not unlike the familiar static yield line

development in R/C slabs, The splitting cracks, caused by

wedging of the deformed reinforcing bars, occur over the rods

on the tension face. A number of examples clearly illustrate

the rectangular grid of cracks corresponding to the layout of

the reinforcement. This feature limits the maximum size of a

secondary fragment to that of a 0.75 in. plate with the dimen-

sions of the reinforcing grid. A very small amount of chipping

has been observed up to the 10 psi level; about 30 percent of

III H E 5 E A K C H I N S T I f ü T F

"

'-*#*?■' - -

• m

I I I [

the concrete below the tension steel has been found to break

loose at the 100 psi level. This latter case would represent

about 560 lbs of small fragments from a 20 ft by 10 ft shear wall.

The central conclusion of this study, that R/C slabs do

not contribute significantly to debris accumulation, is sup-

ported by three technical chapters dealing respectively with

the loading, transport and response of R/C slabs. This result

was entirely unanticipated at the outset of the present program

which embraced the objective of describing the amount and charac-

ter of debris generated by R/C slabs. Specifically, the program

called for the review of the results of an experimental investi-

gation conducted at the U.S. Army Engineer Waterways Experiment

Station (WES) with a view toward using the findings to develop

a fragmentation model. No fragmentation was achieved in this

experimental study, and indeed, fragmentation was not within the

purview of the investigation. For this reason, the approach of

the present program had to be altered and efforts were finally

directed toward a reexamination of the blast effects literature

from the standpoint of fragmentation.

It is perhaps not surprising that only a small part of

the blast effects literature is useful in the study of slab frag-

mentation. Nevertheless, it is felt that enough results are

available to support the inferences that have been drawn. We

have adopted the case method in our description of relevant

papers and reports and have endeavored to reproduce the sig-

nificant photographs upon which we have based our observation.

We believe that the case method is the most objective way to

justify our conclusions; however, we are aware that this method

characteristically tells a story in a telegraphic style. In

our final chapter, the case method is abandoned in favor of a

rather subjective discussion of results.

i i 1 RESEARCH ' N S T I T I) T '

.

1

i

CHAPTER II

LOADING

The loading on slabs depends upon both their initial

orientation and their post cracking behavior. Several pre-

liminary reiiarks are in order concerning the general employ-

ment of R/C flabs as it relates to their orientation.

Certainly the predominant use of slabs in above grade struc-

tures is in roof and floor elements. We shall see that the

horizontal disposition of these members precludes any

significant transverse loading in the Mach region of a

nuclear environment where the shock front is nearly vertical.

Although R/C curtain walls are almost never used, vertical

R/C slabs are found In special circumstances in the form

of shear walls. We shall describe a rough guide to the use

of shear resisting members.

In R/C buildings of the frame/slab or flat/slab types,

no shear walls are needed for buildings under 10 to 12

stories. From about 10 to 20 stories, R/C core walls may

be used around elevator shafts and stairwells. For buildings

over 20 storias shear is resisted by combinations of:

a. Frames

b. Exterior shear walls

c. Interior shear walls (in place of some partitions)

In such buildings, shear walls are not required in the top

12 to 15 stories. Where earthquake design is employed,

shear walls are required up to about 15 stories. Over this,

shear is resisted either by the frame alone or a combination

of frame and shear walls. Multistory steel buildings up to

25 stories usually have R/C core walls around their elevator

shafts and stairwells« Over 25 stories, shear may be resisted

by X-bracing or by interior shear walls.

0

-.*,:■... -&m*^*--m&MiBvmwii--

«■■■-.

I I I r

.

i.

;,

Title: Model Analysis

Author: Davies, L. LI.

Source: Proceedings of the Symposium on Protective Structures

for Civilian Populations, National Academy of Sciences

National Research Council

Date: April 19-23, 1965

Conclusion I: Complete primary fragmentation may be impossible

in R/C slabs where the reinforcement is continuous

from the slab into the supporting structure. In such

cases, fracture of the steel in the center of the

slab will merely permit the resulting petals to

"weathervane," i.e. to unload themselves by aligning

with the wind.

Remark I: Post-test photographs of R/C cubicles which were

subjected to the blast loading are shown in Figure 1.

In the case of both the scale model and the full size

cubicle, failure was in the nature of the steel frac-

turing along the diagonals with the resulting petals

bending back out of the wind.

Conclusion II: Reinforced concrete walls parallel to the direc-

tion of the blast wind and horizontal slabs above grade

will not fail because, due to their orientation, they

will not experience any transverse loading.

Remark II: None of the side walls or roof slabs in the R/C

cubicles shown in Figure 1 exhibit any damage.

r

en a

m

a

a

CN

i—i cd ü

CG

<!•

•r-(.

Ü a in

<t r~1 m .-t

P fa

a z M Q < O

o M

D a w H fa < a: W —-i

M PQ

U

feJ H fa Cd CJ

o CJ

Q fa o cd o

M W Cd

(U

.-I fa

'«MBBSWK®*^*--- •■ ßt1 wmjikumm&mmmifr^mmmmrgm --mimm^mm^m^)^-^

I

Title:

Editor:

i.

Source:

The Effects of Nuclear Weapons

Glassuone Samuel

U. S. Department of Defense, ü. S. Atomic Energy

Commission

Date: February 1964

Conclusion I: In buildings with exterior wall openings

(windows, doors, etc^> only the ground level floor

is subjected to a significant transverse loading

due to a nuclear blast.

Remark I: Two-story wood-frame houses were exposed _o the

air blast from a nuclear weapon. A postshor photo-

graph of the exterior of the house at the 2.6 psi

overpressure range is shown in Figure 2. The damage

to the first floor joists in this house is shown in

Figure 3 and the even more severe damage to the

first floor joists of a similar house at the 4 psi

overpressure range is shown in Figure 4. However,

even in the house at the 4 psi range, the second

floor showed little damage - thus indicating rapid

pressure equalization above and below the floor.

Conclusion II: Under favorable conditions, R/C structures

may receive virtually no damage from a nuclear blast,

Remark II: Although the reinforced concrete deck bridge

shown in Figure 5 was only 270 ft from ground zero at

Hiroshima, it showed no sign of any structural damage

Thus, due to the fact that the loading was primarily

!,. vertical and the possibility that the magnitude may

have been diminished by reflections of the blast r

wave off the water onto the underside of the bridge,

the bridge apparently just deflected downward under

the blast wave and rebounded, leaving only a slight

net displacement.

I~*

8

m raw» ^'».mmmmmmm

-

0

Figure 2 STRENGTHENED WOOL) FRAME HOUSE AFTER EXPOSURE TO 2.6 PSI OVERPRESSURE

9

*

Cd CO D O X

CK

P O O 3:

O Cd

H O 2 Cd OS H co

Cn O

Cd

1X1 co Cd

CM f*S Cd > O

CO M H co co ex, H O ^O

o o cn

H co

CM

O H

Q Cd CO O

M X Cn Cd

Cn

10

w CO h s

p 8 Q W Z

g bg 2^ w w Hw CO 0Ä

tups o w >

co O H CO H-l M CO OP* '"I

<r 04 oo SH tu Q

W H co oo O 04 CU H X Cm W

0) 'H 3 M

•i-l

En

11

■ ' " ■ ■ ' ■ ■ ■ .' ■ . ■ ■

*•

1 1 r i

!ZM M X U W

fa & M

w ac H w OS u o

H

o

W O

w b °S u o Dd fa

ffiE-i H fa M

Weg O Q Q M H

CQ < U

tf Q

Pi § W H H O

iSps fa fa

w o w H 2 c/3 O

•H fa

12

mmjumtrnm...

1 I

[ i

I

L

!

i f

CHAPTER III

FRAGMENT TRANSPORT

In our next chapter we shall show that potential R/C slab

fragments are either very large or small, i.e., the size of a

full slab with a ballistic coefficient of about 1500 lb/ft2

or the size of a brick with a ballistic coefficient of 50 lb/ft

On this basis we can make several observations that are relevant

to the horizontal displacement of fragments. These observations

are necessarily limited because high rise R/C buildings have

never been exposed to nuclear blast loadings.

Title: The Effects of Nuclear Weapons

Editor: Glasstone, Samuel

Source: U. S. Department of Defense, U. S. Atomic Energy

Commission

Date: February 1964

Conclusion I: High burst heights give rise to smaller fragment

translation.

Remark I: For structures close to ground zero, high altitude

bursts subject their roof and floor slabs to large

downward pressure components. This pressure augments

the gravitational acceleration and shortens the flight

time of the fragments. Since the horizontal component

of the dynamic pressure acts on the fragments for a

shorter time, they undergo smaller translations.

The effects of high altitude detonations is clearly

indicated in Figure 6 which shows a heavily dished

R/C roof slab. This structure was located only 0.10

mile from ground zero at Hiroshima.

Conclusion II: Potential fragments from R/C shear walls or

panel« will tend to remain on the building sita.

13

i

X

9 o w N

§

S

o Cd H

O 2 H

a H

«

i o o

Q W

o

00

I 14

' ^v.'J

- nwwfuffi.TL 'irmrftHitä-.-'"Itll

!

!

SO Remark II: When shear walls are used in the core of a building

they are often located at a distance of two bays or

40 ft from the curtain walls. Thus,.any fragments

generated from these members will have to travel about

50 ft before they threaten to congest the nearest

street. There is considerable evidence to show that

the small secondary fragments which we anticipate can

easily surpass 50 ft in translation. This is demon-

strated by the aerial survey of tornado damage shown

in Figure 7. The tornado produces dynamic pressures

which are not unlike those created by a large hydro-

gen bomb at the 5 psi level.

We anticipate that potential primary fragments will

have dimensions of the same order of magnitude as

the full size slab. To get some idea of the response

of such large fragments we can examine the structures

shown in Figures 8, 9 and 10. The first of these shows

a massive section of wall that was dislodged from the

structure at a height of 60 to 70 ft. We estimate

that the fragment was transported 14 ft horizontally.

The structure was located at a distance of 0.34 miles

from ground zero. The photographs shown in Figures 9

and 10 show two engine lathes weighing 7000 and 12,G00 lb

respectively. The small lathe displaced about 7 ft under

a 10 psi overpressure; the large lathe, with a ballis-

tic coefficient about 7 times greater, did not move

from its anchors. As a final example, primary fragments

w have been produced from the brick chimneys of the two

houses shown in Figures 11 anci 12. In both cases, the

final position of the primary fragments was in the

immediate vicinity of the base of the chimney.

15

r

Figure IF1 i RESEARCH INSTITUTE AERIAL PHOTOGRAPH

OF PROPERTY DAMAGE IN ELKHART, INDIANA BY TORNADO ON APRIL 11, 1965

16

i*-

1 I

:

T

«VI -

\

Figure Ö HEAVY WALL BEARING STRUCTURE LOCATED 0.34 MI FROM GROUND ZERO AT NAGASAKI

I 17

..,*».'..■,.,'. .:.;-•«:.,,, ,.-:-«,;,. ■. ,j*#^ ■ •.' :,.'J' :,,v,|- ' ;täakr^.!i v .«i-.,«-.::*;,. «■ - i > '..-.-.-..■•»...A 'H.mi^M-

I . *9SHV^9^B9H^VMHSME

I

J to CO

a: w

O w ffl

I

s o co

33 W PQ

W ■

►-J o o -■'

H

w 2 M i

; X u

18

I I

I

I 1 I I

I-

mm 'b Wit fV * r • •

<

1 as M

W

Ü3 W

erf

cd w > o M en 0*

O H

o PM X w

W H

o o H

re

0)

Oil •r-4 En

I 19

■ - ■ 3 ■ * :.-. (äSÜ

!

f —

i.

.

I

I

co

tu Pü w i CO PL,

o H

CO

§ X w Di W H

<

W co

Q O O

CO

I •H P*4

20

>*«M I - .*•*-

• 'V ,-.■•: -... • .U-.

1 f

"

I 1 !

w

en w

w

PL,

B

s

13

s

o o s CNi

•H fa

21

'msm^m^j**

I I f

...

i ■

\

Title:

Author:

Source:

Date:

Shelter Evaluation Program

Heugel, W. F. and Feinstein, D. I.

Office of Civil Defense, Contract OCD-PS-64-50

Work Unit 1614A

February 1967

Conclusion: Large fragments tend to remain on the building site,

Remark: In Figure 42 of the subject reference, the effect of

the ballistic coefficient on the debris trajectory

is studied using analytical methods. Focusing oir

attention only on the range or maximum translation,

we have plotted this parameter against the ballistic

coefficient in Figure 13. The dashed line in this

figure represents a conservative extrapolation to

this curve of monotonically increasing curvature.

Based on the extrapolated curve, an entire 20 ft

x 20 ft x 8 in. slab (Ballistic Coefficient = 1439)

will translate less than 16.5 ft; a fragment one-

fourth this size (Ballistic Coefficient = 903) will

move less than 24 ft horizontally. A secondary frag-

mert measuring 12 in. x 12 in. x 3/4 in. has a

ballistic coefficient of about 88 and a range of

150 ft. As a conservative rule of thumb, these

ranges increase as the square root of the height.

n 22

. «tmmamm

P CU e so X « M-l

en U •pli-l r-l«W r-l 4-1

CO 0! CQ O

U

CO

TJ CO

CM

CO X)

ooooo OOOCN IT» <t CM CM r-l

mo ooo CO vOCM vD

i-^ CO

^D -r-l CM <f CM

/

hiil i . o

o o t-H

n i

o o

1

/

/

/

/

CD CQ •H

CO IT) a.

■urn CMC* CO

OCM vO o r-l

« • • • +J •• CU £• CU MO p

•HTH 05 0i 4J CO W cö CU

H >-l 0.3 a oo u rl cu Q CD >

CO O cd x: P-. cu >

•rl W •r-l CO o

PM

J_J L 1 Uli 1 ■ I I I

o o o

#v W c l-l a

4->

e S3 01 CU

•!-! en Ü

•I-l w M-) •-s H •4-1 CM W 0> 4.1 0Ä O M-t o Ü \ ss

CO c 60 Xl u

o nj r-i o H ">-' « o •a o 1-1 I*

4J w x: £5 Ö0/-S Z

•H CO a> cu 21 ? u

ct) en n >

T) 4J g) g cu o W

•H cu H u«o O

o •<-i o H o M-4 U

& r-l M-( a cu "^ w o o o c_> o CJ

T-l M 4J H CO CO

•r-l M r-l ►J r-l (_J cö <J

PQ CQ

3j 'quaiuaoBxdsxa TB3U02T.aoH uinurpxBK

23

CO

cu

•rl

I f CHAPTER IV

RESPONSE

A. Introduction

Studies directed principally to the fragmentation of R/C

slabs are almost nonexistent and, for this reason, we have

chosen to examine the byproducts of investigations concerned

with the analysis and design of conventional and blast resis-

tant R/C structures. The most frustrating problem encountered

in this approach is that the previous investigators were not

motivated to report findings which are relevant to fragmenta-

tion. Also; most experimental work is confined to beams.

Finally, these programs are generally concerned with only

the range of behavior up to severe cracking. Unlike the

homogeneous concrete bending members, this cracking is a

necessary but not sufficient condition for the production of

primary fragments. Indeed, tension cracks are always present

in a R/C bending member under working loads; they become vis-

ible in great profusion under higher loads as shown in Figure 14,

Our studies of s]ab response have indicated that primary

fragments are likely to occur in only two situations; punching

out of the entire slab by shear at the boundaries or by large

deflections of simply supported slabs which cause the entire

slab to push through their supports. The strength and duc-

tility of the steel reinforcement together with the difficulty

of sustaining loads on cracked slabs operate to prevent the

formation of small primary fragments. The problems of primary

and secondary fragments in slabs will be dealt with separately

in this chapter; but, first we shall explore the implications

of high tenacity in other types of R/C members. i. .

24

- ..,. ..

i I I I

I

."

.

L

[

■

I

I I I

§ M

H

W

a Q

W

3

< H CO

O o Pi<

> O CD

55 O M 55 o H psj « O O j* 05 H U W

pi

<: w 55 KJ M Ü P 55 5 M H co M O JC 55 H OM -1^

bC •H

25

NH

[

!

t I

f

::

i

Title: The Effects of Nuclear Weapons

Editor: Glasstone, Samuel

Source: Ü. S. Department of Defense, U. S. Atomic Energy

Commission

Date: February 1964

Conclusion I. Primary fragmentation does not necessarily

accompany general and complete destruction of R/C

frame and panel buildings.

Remark I: The P/C building shown in Figure 15 was located

0.26 mile from ground zero at Nagasaki. Hinge points

can be identified in the beams over the window openings

and hinge lines in the panels. The destruction is

quite complete and yet we note that because none of

the reinforcement has ruptured the building regains in

one piece. There is therefore no primary fragmentation.

Conclusion II: Primary fragmentation does not necessarily attend

the complete and general destruction of R/C frame build-

ings.

Remark II: A three story R/C frame building which is typical of

conventional American construction is shown in Figure 16.

Located only 0.13 mile from ground zero at Hiroshima,

the building suffered enormous damage. The ends oi

many beam and column members have been completely

severed through their reinforcement. We observe, how-

** ever, that there is at most one free end on each member;

P the other is still attached to the structure. There

L is no primary fragmentation of the frame since the entire

structure is tied together by the steel. Much of the

observed fragmentation is composed of pieces of brick

panel which made up the 13-in. thick walls.

26

■mmiwMUMwmw

1 I I I f

I

r ....

I

IW-f

f. ...

»

P w H

o

CO w CO CO O

H

O

£ H W

g CO

M

H<: HCO

H PH iJ< H PO «Pi

w CJN

>HP pJO

wo

m

0) U

I 27

KP

-

I r 4

I f I

CO

H < u o

en ►J w

fa

H a!

X H M

U Z i-i

a H

« O

M X

H

O U Cd -». w PS N

5H Q PA z o 5 H2

o fa" 2 ad o S oi H fa

v£>

0)

3 tsfi

fa

I 28

... . .. ■ .... .

■

****u**mmmmmammmmaHizmmmnwmi»iiM

1

I Conclusion III: When large R/C stacks are severely damaged by

a nuclear explosion, primary fragmentation may still

* be incomplete with the major fragments remaining linked

Y together by the steel reinforcement.

Remark III: A circular 60-ft high R/C stack located 0.34 mile

j from ground zero at Hiroshima was very severely damaged

■ by the blast wave. Referring to Figure 17, it is ob-

• served that the stack remained intact up to a height

of 15 ft above the base. The top 45 ft has been toppled

over and now consists of two relatively intact sections

and the top section which has been completely pulverized,

r

*-Si

I

I

I I I I

The reinforcing rods continue to link all the major

fragments together. The top section of the stack where

the steel is completely exposed probably was denuded

r when that section impacted the ground.

29

" -;*Ji#*Ä&r>*- ■

I I ü

r*

.

I I - I

t&i -• Figure 17 CIRCULAR 60-FT HIGH R/C STACK LOCATED 0.34 MI

FROM GROUND ZERO AT HIROSHIMA

■■-■■-

1. 30

■ ■■vmmimmmmma-vmmmcm.imtmmm&m-,-

[':

L

I.

i i I

Title: Blast Effects of Atomic Weapons Upon Curtain Walls

and Partitions of Masonry and Other Materials

Author: Taylor, Benjamin C.

Source: Operation Upshot-Knothole, WT-741, DDC No. AD-636 766

Date: August 1956

Conclusion: Reinforced masonry is considerably more blast

resistant than unreinforced masonry.

Remark: A front wail consisting of 4 in. of brick facing and

4 in. of clay tile backing was located at the 4.5 psi

overpressure range. Figure 18 shows a motion picture

sequence of the failure of this wall. In the fourth

frame it appears that a primary fracture pattern has

formed which consists of a central rectangular crack

pattern with cracks running diagonally from the corners

of the central rectangle out to the corners of the

panel. Postshot examination revealed that the wall

was 98 percent blown into the cell with the bottom

corners remaining. An 8 in. reinforced grouted brick

masonry rear wall was tested at the 7.5 psi overpressure

range. The resulting damage consisted of some crushing

at the top on the outside face and a 3 ft long crack-

on the inside face which ran from the top down to near

the center of the wall. Even this small amount of

damage may have been prevented, if the wall panel had

been properly tied into the roof slab with continuous

bars. Front and rear walls consisting of 8 in. of R/C

with keyed joints at the top and bottom were also

tested at the 7.5 psi overpressure range. Preshot and

postshot photographs of these walls are shown in

Figure 19. Only the front wall received any observable

damage. It had a 4 ft long horizontal crack on the

outside face near one edge at the ground and a hairline

opening of the construction joint.

31

IE

i

~

I I I I I I

tef^* pwr

.'■Sr*

UÜ,'~u-i?. .,C -

Figure 18 MOTION PICTURE SEQUENCE OF FAILURE OF UNREINFORCED MASONRY WALL UNDER 4.5 PSI OVERPRESSURE

32

•'-

ö io

n

i !

I ei

co

sew P*pi

H CO O co KW to ci

co aJ o w pH > o

<C co P-i

O «o BÖ • co r» w pä H PH<:

ON

0)

So

I I

33

11 !

r

[

Titlei Fibrous Reinforcements for Portland Cement Concrete

Author: Williamson, C. R.

Source: U. S. Army Engineer Division, Ohio River Corps of Engineers

Technical Report 2-40

Date: May 1965

Conclusion: The tenacity of a concrete slab uuder a point high ex-

plosive charge is greatly increased by the addition of a

small amount of reinforcing.

Remark: High explosive tests were conducted in 36 x 36 x 4 in.

concrete slabs. The dynamic pressure loading created by the

2.5 lb charge was in excess of 4000 psi. As illustrated in

Figure 20, the plain concrete slab was completely destroyed

while the R/C slab was only breached. The reinforcement in

the second slab consisted of one percent of steel wire,

1 in. in lengta and 0.01 in. in diameter, randomly oriented

i and 2 in. by 2 in. wire mesh in the center of the slab.

Title: Investigation of the Structural Properties of Reinforced

Concrete Masonry

Author: Saemann, Jesse C.

Source: National Concrete Masonry Association

Date: June 1955

Conclusion: The reinforcing steel does not fracture in R/C beams

under static or impact loads.

Remark: The response of the R/C block beam shown in Figure 21,

is typical of most monolithic R/C beam behavior under either

static or impact loading. Yielding of the bars is not un-

common and it is usually followed by horizontal cracking

which tends to expose portions of the tension steel by

stripping away the concrete from the tension face of the

beam. Literally no beam tests were encountered where frac-

ture took place in the reinforcement. The literature survey

included all of the issues of the Journal of the American

Concrete Institute from 1955 through 1967.

34

I n -

a) Plain Concrete Slab After Testing with High Explosive

b) Wire-Reinforced Concrete Slab After Testing with High Explosive, 1 percent, 0.010 in., 1 in. Lengths

Figure 20 EFFECT OF WIRE REINFORCEMENT ON TENACITY OF CONCRETE SLABS EXPOSED TO HIGH EXPLOSIVE LOADINGS

I n

ES <

o w CO

X H H

g O CO

W p

<!

g fe o o u

z$ tu H

Q W

w 5

Pd

H 2

CM

3

36

MM

I I

ID o

Title: A Dynamic Ultimate Strength Study of Simply

Supported Two-Way Reinforced Concrete Slabs

Author: Denton, D. R,

Source: U. S. Army Engineer Waterways Experiment Station

Final Report, Contract OCD-PS-65-44, Office of Civil

Defense

Date: April 1967

Conclusion: It is possible tc fracture some of the tension steel

in a R/C slab if it is lightly reinforced.

Remark: The steel reinforcement failures shown in Figure 22

were the only ones observed in a series of four static

and 1/ dynamic tests on simply supported R/C slabs.

The slab in which these failures occurred had been

exposed to the highest dynamic overpressure in the series

of slabs with the lowest percentage of steel. Also, it

is of interest to note that the failures occurred in the

transition region where every other bar was bent up.

Title: Model Analysis

Author: Davies, I. LI.

Source: Proceedings of the Symposium on Protective Structures

for Civilian Populations, National Academy of Sciences

National Research Council

Date: April 19-73, 1965

Conclusion: Steel reinforcement failures, whicu are necessary

for primary fragmentation to occur, tend to occur

along the static yield lii_e patterns.

Remark: Steel failures occurred along the classic "x" static

yield line pattern in the front wall of two of the R/C

cubicles shown in Figure 1.

37

«MWHHMIr'l

::

:

i

•

w

Id

00

EC H H

3 CO

u

In « o 5 CO

Id co O W o »J Id Pj

Id Id X > H O

OS H

em d CO Id

H <

o H Q Id

id H Id O Id Id H •") CO PQ 5

J3 co o M td CO Id 3 Id Id H H CO

d

I 33

<M

u 9 M •H Sn

IT II—l II

I

E

[

Title: Resistance and Behavior of Concrete Slabs Under Static

and Dynamic Uniform Loading with Edges Clamped and

Laterally Restrained

Author: Keenan, W. A.

Source: Technical Report, Naval Civil Engineering Laboratory

Port Hueneme, California

Date: To be published in 1968

Conclusion: No primary fragments are formed in R/C slabs sub-

jected to dynamic pressures up to 120 psi.

Remark: The slab shown in Figure 23 has fixed edges which

are laterally restrained. The member is typical of

a series of R/C slabs which have been tested under

uniform dynamic pressures as high as 120 psi. Although

none of the reinforcing bars were fractured in the slab

shown, necking down and fracturing occurred with some

regularity near the edges of the slabs. In the most

extreme case, such fracturing occurred in all the

bars along two adjacent edges; however, primary frag-

ments did not form because the slab segments were held

together by the bars which remained attached to the

intact edges.

When slabs exhibited large deflections (two to three

Limes the slab thickness), their resistance was almost

entirely due to tensile membrane action. In a few

cases this tension caused some of the reinforcing bars

to fracture in midspan. No primary fragments were ob-

tained in the test series.

i I 39

: • ■ v. ■■j^ *wmvimmm m :*■■.-.»

D

-.

Figure 23 SLAB WITH EDGES FIXED AND LATERALLY RESTRAINED/ DYNAMIC PRESSURE LOAD OF 101 PSI

40

......

I I

Title: Behavior of One-Way Concrete Floor Slabs Reinforced

with Welded Wire Fabric

Authors: Atlas, A., Siess, C. P. and Kesler, C. E.

Source: J, Am. Cone. Inst., Proc, Vol.62(5) PF 539-557

Date: May 1965

Conclusion: Slabs reinforced with welded wire fabric fail

suddenly by fracture of the reinforcing wire.

Remark: The longicudinal reinforcing wires in the one-way

slabs investigated in the subject reference were

anchored by rigidly welded transverse wires. In every

case studied, the slabs failed by the sudden fracture

of the steel while the concrete was apparently still

intact on the compression face. It is highly signifi-

cant that these are the only examples that could be

found in the present program where tha reinforcing steel

was actually fractured under static loading. As far

as we could determine, wire fabric is not generally

used in shear walls.

i i

41

! ̂

f -■

.

Title: On the Dynamic Strength of Rigid-Plastic Beams

under Blast Loads

Author: Salvadori, Mario G. and Weidlinger, Paul

Source: J. Eng. Mech. Div., ASCE Proc, pp 1389-1 to 1389-34

Date: October 1957

Conclusion: Under certain circumstances a slab under a uni-

formly distributed blase pressure may fail by

developing plastic slides along the supports.

Remark: Shear failure of a reinforced concrete beam is

accompanied by large changes in the bar geometries

as shown in Figure 24 (Ref. 5). Ultimately, the

shear resistance of the beam is provided by the tensile

strength of the distorted reinforcing bars and, con-

sequently, a plastic slide mechanism precedes rupture.

The circumstances under which this failure mode occurs

in beams is treated in the subject reference. In

our opinion, the extension of these ideas to slabs

is immediate.

If the slab is proportioned so that its bending

strength is greater than its shear strength, it is

possible, under sufficiently large pressures, to

punch out the entire slab along its supports. The sub-

ject reference indicates that once plastic slides

develop and the end slides cannot spread during bend-

ing and shear motion of the member.

I

42

I

CO H U Id

w

o c

8 H

I CO

O z H

iJH HS DW PQS

W wu Sei

In § H

CJ Id

^^

oi w C PQ H WW

HH

£&> £ H

CM

(1)

•H In

43

•

I I

B. Primary Fragments

The formation of a primary slab fragment requires fractur-

ing of the reinforcing steel around the fragment periphery. The

central question then is "what is the fracture mechanism of

the steel?" It is surprising that in the hundreds of static

and dynamic tests of R/C beams and slabs that were examined

for this program, steel fracture was rarely encountered.

Furthermore, no cases were found where these fractures were

sufficient to produce a primary fragment.

To reconcile the fact that steel fracture is feasible i

but very unlikely, we can examine the possibility of obtaining

the ultimate tensile strain in a deformed reinforcing bar.

As loads are brought onto a bending member, the tensile strain

in both the tension reinforcement and the concrete increase

until the ultimate strain is achieved in the concrete. A

flexure crack results and at this crack most of the tension

is carried by the bars; tb» steel stress is maximum at such

cracks. If we continue to increase the loading, the bar

strees at the crack will also increase until either it yields

or further loading is precluded by yielding at other locations

in the bending member. Assuming that the bar yields at the

crack, it would continue to stretch if the external loads

were maintained or increased. If the elongation of the bar

leads to the ultimate steel strain, fracture results; but, I

here we are led to the crux of the matter. How much of the

bar participates in the elongation or over what gage length

do we distribute the stretch?

Now, if only the portion of the bars included within a

finite width crack participates in the elongation, very high

strains will be quickly achieved and steel fracture will ensue.

This is not, however, what happens. Even where sheai is zero

and moment constant■, large local bond stresses exist adjacent

to each flexural crack. These stresses cause a relative slip

I 44

I

\

\

I

between the steel and concrete which represents local loss of

adhesion. This takes place in the pullout test by the time

the bar carries only 2000 to 3000 psi.(Ref. 4). The slip near

the crack gradually brings the lugs on the deformed bar into

(bearing against the surrounding concrete. These lugs tend to

pry the concrete apart creating circumferential tension which

i| in turn causes splitting of the concrete. This splitting en-

** ables a greater length of the bar to accommodate the stretch

I rt and, therefore, lower unit strains are realized.

1 When deformed bars are tested using a pullout specimen,

failure nearly always occurs by splitting the concrete prism

I- into two or three segments rather than by crushing against

the lugs or by shearing on the cylindrical surface which the

lugs tend to strip out. The splitting action can be seen

as horizontal cracks in the constant moment section of the

beam shown in Figure 14. Quoting from Ferguson (Ref. 4) :

"Splitting seems to start at flexure cracks being

most evident where steel stress is largest. Thus

splitting is a progressive phenomenon, working its

way gradually along the length of embedment. Split-

ting may net oe continuous from flexure crack to

flexure crack. A splitting crack often stops short

of the next flexure crack in a bond test beam; in

fact the opening of a new flexural crack usually

occurs beyond the end of a splitting crack, and

added splitting then develops from the new flex-

ure crack. Normally the splitting will eventually

close the gap. Splitting can develop over 60 to

75 percent of the bar length without loss of aver-

age bond strength. Apparently splitting is one

means by which some of the unevenness in bond

stress distribution may be smoothed out. However,

the final failure is sudden (in the absence of

stirrvos) as the split suddenly runs through to

the end of the bar."

45

I

'*—wwawanromwimwae-

- • r -.-rrawWJ

fl

i r

[ r i

[

The implication of the splitting action is to free large

portions of the bar to equalize the bending strain at

values lower than the ultimate strain. There is, of course,

a small chance that the splitting action will not rescue

a strain concentration and then fracture can take place.

In slab elements which employ a wire fabric rein-

forcement, the splitting action can at best enable the

strain to equalize between two nodes of the meshing since

the orthogonal wires will anchor the nodes. One would

expect more fractures of the reinforcement in welded wire

fabric; but, only one reference is cited in support of

this conjecture. It is quite common to me both tension

and compression steel in a slab and, here, th=re is no

possibility of breaching a member by bending alone. With

the small concrete cover normally employed in slabs (3/4 in.)

no suitable moment arm is available to put a tensile load in

the compression reinforcement after severe cracking and

fracture of the tensile steel. Fracture of the compressive

steel must be brought about by membrane response. As a

matter of fact, in the few cases where bars have ruptured

the evidence suggests that fracture was brought about by a

combination of bending and membrane action.

I I

46

1 I

r u

Title: A Dynamic Ultimate Strength Study of Simply Supported

Two-Way Reinforced Concrete Slabs

Author: Denton, D. R.

Source: U, S. Army Engineer Waterways Experiment Station

Final Report, Contract OCD-PS-65-44, Office of Civil

Defense

Date: April 1967

Conclusion: The primary crack pattern produced in R/C slabs

appears to be the same type of patterns predicted

by yield line theory for ductile slabs.

Remarks: Figures 25, 26, 27 and 28 illustrate the crack

patterns produced by uniformly distributed static and

dynamic loads in simply supported R/C slabs. The slabs

had a clear span of 7 ft-6 in. and were 2-5/8 in. thick.

In each case, the primary crack pattern appears to con-

sist basically of an inner square with diagonal cracks

connecting the corners of the inner square to the cor-

ners of the slab.

5

Title: The Effects of Nuclear Weapons

Editor: Glasstone, Samuel

Source: U. S. Department of Defense, U. S. Atomic Energy Commission

Date: February 1964

Conclusion: Reinforced concrete wall panels may develop primary

crack patterns similar to the classic static yield line

patterns for ductile panels.

Remark: The wall panels on the R/C frame building 0.68 mile

from ground zero at Nagasaki, shown in Figure 29, have

primary cracks Lending to occur along the diagonals.

These panels may be considered to have fixed edges and

thus the static yield line pattern would consist of yield

lines occurring along the diagonals and the eaV^es.

I ■n wmmtm nww

47

I

I I I

\

I -

^

I i

r i

i

Figure 25 TENSION SURFACE OF R/C SLAB WITH 0.78 PERCENT STEEL SUBJECTED TO 7.1 PSJ. UNIFORM STATIC LOADING

48

I

!

\

\

I

I I

Figure 25 TENSION SURFACE OF R/C SLAB WITH 0.78 PERCENT STEEL SUBJECTED TO 7.1 PSI UNIFORM STATIC LOADING

48

. ■ ,

A. L..J V -•■'-.. l^^^i. J i ' r f I- ••

Figure 26 TENSION SURFACE OF R/C SLAB WITH 0.78 PERCENT STEEL SUBJECTED TO 9.0 PSI OVERPRESSURE

49

•

V j*m

• -■% i«y- "*8^ .

'*:

» '■'-■

i, , * i „ , |_ - ..j

> •

4 w t

I

Figure 27 TENSION SURFACE OF R/C SLAB WITH 1.0 PERCENT STEEL SUBJECTED TO 12.5 PSI OVERPRESSURE

f 50

I

füfa-p+h- *e ■ A guy.

I -«4

% ■ |

fit

iß.

f

i ■ ■

sf

Figure 28 TENSION SURFACE OF R/C SLAB WITH 1.17 PERCENT STEEL SUBJECTED TO 13.2 PSI OVERPRESSURE

I 51

0}

<

M 2.~s < M fa !^S

fa CO H < W O OS < O 2 2: O H U <

o o W OS cc w CO N

OS Q O 2

O O 2 OS M O

Bs EC O co oS

fa o !Z M H S Q HJ oo M vO E3 • ta o

OS H fa CO

O W r-i H O-i W X OS w u 2 Ü O 2 U M

o < W fa u OS fa O Q fa H Is CO M W 2 OS O

CN

CD U

öß •H fa

52

i r

Title: Blast Effects of Atomic Weapons Upon Curtain Walls

and Partitions of Masonry and Other Materials

Author: Taylor, Benjamin C.

Source: Operation Upshot-Knothole, WT-741, DDC No. AD-636 766

Date: August 1956

Conclusion: Simply supported R/C slabs may undergo such large

deflections under blast loading that they are pulled

off their supports .

Remark: A rear wall consisting of precast R/C channel slabs

was tested at the 7.5 psi overpressure range. Due to

construction details, these slabs may be considered

to have been simply supported along the top and bottom

edges. Figure 30 reveals the extent of the damage

sustained by this wall panel. All of the channel slabs

were damaged with the maximum severity occurring in

the center of the wall panel. The photograph of the

interior of the tesc cell reveals that the flange

reinforcing bars were separated from the channel slabs

and that the permanent maximum deflections were con-

siderable. This allowed some of the slabs to pull off

their supports.

i

!

i 53

KW,

5 ^£§5?

-d 0) 4J cn •\ qj 4J 5-1

■u O O x;-H

4-J w u 0 u CÜ c w -u

0 a Cfi PH M es S .-—s

u pH H 05 5 •u a 0 n

tw

D X! H .. cu

■u o

S3

2 w Eg O oo

W

SC PL.

< w 05 > oo o H M

00 PM

Ln PH

H •

§r oo H H < oo O >J CU _)

SS <3 PS

<2 o 2 3d 00 CQ W <

PH 00

G

QJ U

54

L

Title

Author

Source:

Date

A Dynamic Ultimate Strength Study of Simply Supported

Two-Way Reinforced Concrete Slabs

Denton, D, R.

U. S. Army Engineer Waterways Experiment Station

Final Report, Contract OCD-PS-65-44, Office of C.iviJ

Defense

April 1967

Conclusion: In the case of simply supported R/C slabs, exces-

sive deformations may cause the entire slab to be

pulled off its supports.

Remark: The simply supported R/C slab shown in Figure 31

has apparently pulled off its support along almost

one entire edge. The permanent midpoint deflection

for this slab was determined to be 17 in. while the

corresponding deflection for all the other slabs

tested in the subject reference ranges from 2.0 to

8.8 in.

Title: The Effects of Nuclear Weapons

Editor: Glasstone, Samuel

Source: U. S. Department of Defense, U. S. Atomic Energy

Commission

Date: February 1964

Conclusion: Reinforced concrete members, if they are simply

supported, may fail by undergoing such large deflections

that they fall off their supports.

Remark: The R/C bridge with T-beam deck, shown in Figure 32,

was located 0.44 mile from ground zero at Nagasaki.

One span, 35 ft long, has been pulled off its supports

and has fallen into the river.

55

r

w H W H W u Da w

co

H H fcj

CG CO < CO HJ W

P-, u od --- w 02 > o

O H to

UJ PL, u

co O

!S H O H a co w co H W U DA W

S CQ O D U W

0) S-i

i

56

\

I

I

c w N

C 0Ä C"

o pa

-3-

u w o s < w CO

EC H

O o H Pi ra

W H W

u 12 O u M

U4 < CJ CO oi < o u

i—i UJ H pi <

m

00

57

•

*rm

i

C. Secondary Fragments

Reinforced concrete bending members undergo an enormous

amount of cracking in reaching their ultimate resistance.

These cracks arise from bending, membrane, and splitting ac-

tion and at high pressures they may be caused by tensile stress

vaves. When slab tests are carried to extreme deflections,

the splitting response becomes fully developed and it super-

imposes a regular pattern of cracks on others that may exist.

This pattern provides an upper bound on the size of secondary

fragments. The splitting cracks will be found to follow the

layout of the reinforcement.

Referring to Figure 33 from Ref. 4, we find three modes

of splitting failure in beams. Only the modes shown in

Figures 33b and 33c are appropriate for the slab. The failures

shown in Figures 33b and 33d correspond to a \nde bar spacing

and those in Figures 33c and 33e to a narrow spacing. Evidence

will be submitted to show that these modes also appear in slabs,

however, here, the concrete cover is more difficult to strip

off from the steel. The effect of splices in the reinforcement

is clearly shown in Figure 34 where the stress discontinuity

at the ends of the bars cause flexural and splitting cracks

to occur. We anticipate similar effects in slabs with discon-

tinuous reinforcement.

58

(•*■»- cw?-*==i«»,*Ä.:

1 V CO O to o u

■u w

aco CO 4J

c

CU V4 •a o

3 60 i-<

CO J-l-f-l 0) 4J >^-H c3 C

<U u

O iJ

£ CO

-o •r-(

c o

>

-a

en

Q W

< a. en

<C en U O

PL. O

33 H

ra

W

H <

W H

(!)

Ü

W Q Q <C3: en >-t

T3

H H H

O-I „a en *-«'

en

a» N

00

59

f

-(«

(a)

(b)

Figure 34 SPLICE FAILURES, (a) VIOLENT FAILURE OF NO.8 BARS LAFPED 36 DIAM, (b) SHORTER SPLICE WITH CRACKED CONCRETE REMOVED TO SHOW SPLITTING SURFACE. ARROWS MARK END SLIPS OF BARS

I 60

■• ■ ■ ■

•

!

Title:

Author:

Source:

Date:

A Dynamic Ultimate Strength Study of Simply Supported

Two-Way Reinforced Concrete Slabs

Denton, D, R.

U. S. Army Engineer Waterways Experiment Station

Final Report, Contract OCD-PS-65-44, Office of Civil

Defense

April 1967

Conclusion: The crack patterns produced in the tension surfaces

of R/C slabs under dynamic loading tend to follow the

reinforcement layout. This tendency is greater for

the smaller steel percentages.

Remarks: As shown in Figure 35, the crack patterns produced

in the tension surface of R/C slabs under uniform static

loading appear to be curvilinear and not to follow the

rectangular reinforcement pattern. However, as ob-

served in Figure 36, the pattern produced in a slab,

with 0.78 percent steel, subjected to dynamic loading

almost perfectly coincides with the reinforcement pattern

in the central region of the slab. By way of contrast,

the pattern produced in a s1ab with 1.17 percent steel

under the same loading, Figure 37, is not nearly as

perfect. It is our contention that the splitting

action of the bars becomes more dominant at the larger

deflections associated with the lower steel percentages.

61

i

r

Figure 35 TENSION SURFACE OF R/C SLAB WITH 1.0 PERCENT STEE1 SUBJECTED TO 9.4 PSI UNIFORM STATIC LOADING

I 62

' ■ . ■

s

f

I [

Figure 36 TENSION SURFACE OF R/C SLAB WITH 0.78 PERCENT STFFT SUBJECTED TO 11.0 PSI OVERPRESSURE '

I i

I 63

■ - ' ■ "

I I

i

f

I

i

I t

Figure 37 TENSION SURFACE OF R/C SLAB WITH 1.17 PERCENT STEEL SUBJECTED TO 11.0 PSI OVERPRESSURE

64

■ ■

■ :

«■

I I

Title: Resistance and Behavior of Concrete Slabs Under Static

and Dynamic Uniform Loading with Edges Clamped and

Laterally Restrained

Author: Keenan, W. A.

Source: Technical Report, Naval Civil Engineering Laboratory

Port Hueneme, California

Date: To be published in 1968

Conclusion: Secondary concrete fragments originate from the

downstream face of a R/C slab and are confined to the

layer of concrete covering the steel. Their maximum

size is bounded by the dimensions of the reinforcing

grid,

Remark: A series of R/C slabs were tested under dynamic pres-

sures of up to 120 psi. Extensive crack patterns devel-

oped in these slabs which were similar to that shown

in Figure 23; they tend to follow the outline of the

reinforcing layout. It was observed from these tests that

secondary fragments came from the layer of concrete

cover on the downstream slab face. Concrete stripped

from this layer exposed the steel. All of the concrete

above the downstream reinforcement remained intact.

Furthermore, the largest fragment observed had the

dimensions of the reinforcing grid; most of them are

smaller.

We estimate that about 30 percent of the concrete

cover is stripped away at the 100 psi level. Since the

ACI code calls for a minimum cover of 3/4 in. a 20 ft

by 10 ft panel would produce about 563 lb of secondary

fragments. If the reinforcement is laid out in a 6 in.

by 6 in. grid, the fragments will weigh less than

2.35 lb each.

65

■■ ■

..

Title: A Dynamic Ultimate Strength Study of Simply Supported

Two-Way Reinforced Concrete Slabs

Author: Denton, D. R.

Source: U. S Array Engineer Waterways Experiment Station

Final Report, Contract OCD-PS-65-44, Office of Civil

Defense

Date: April 1967

Conclusion: Compressive chipping will produce some secondary

fragments from R/C slabs.

Remark: Typical concrete compression failures are observed

along the diagonals near the corners of the R/C slab

shown in Figure 38. A detailed view of one of the

corners of this slab is shown in Figure 39. It is

interesting to observe that the secondary fragments

which were produced tended to remain w .1 the slab

on the upstream face.

Title: Behavior of Metals Under Impulsive Loads

Authors: Rinehart, John S. and Pearson, John

Source: Dover Publications, Inc., New York

Date: 1954

Conclusion: When the peak dynamic overpressure is less than the

tensile strength of concrete (about 300 psi), no sec-

ondary fragments will be produced by scabbing.

Remark: The complex phenomenon of scabbing is dependent upon

the detailed understanding of the interaction of stress

waves as they reflect from the various boundaries of an

ob ] ec t.

66

I f

i w w H CO

i w

w

H H W

pa co < CO (J w CO ßS

PM CJ> as Dd >

O fa OM

CO W PH u fn • Pi CM

CO o

Z O MQ co td co H Cd U öS Id Oj i-, S PQ OD CJCO

00 on

M

60 •H fa

67

■■

&!&&,

Figure 39 COMPRESSION CHIPPING PRODUCED IN R/C SLAB WITH 1.0 PERCENT STEEL SUBJECTED TO 12.5 PSI OVERPRESSURE

68

' - -- -

."

A typical example of a scabbing type of fracture is

illustrated in Figure 40. Basically the two most im-

portant factors influencing scabbing are: (1) the mag-

nitude of the tensile stress wave with respect to the

magnitude of the fracture strength of the material,

and (2) the shape of the stress wave. If the magni-

tude of the tensile stress wave is less than the

strength of the material, no scabbing will occur.

The present program is concerned with pressure levels

below 100 psi. Here, the low level stress waves will

help shake stubborn secondary fragments loose from

the slab.

69

[

I

f 55 M PQ pa < CO

w 13

o H

I w H

PH

w w H co

Q

tu O

!2J O M H CJ W en

CO w O

a u 5b

70

- -

11

t -

CFAPTE? V

DISCUSSION OF RESULTS

The occurrence of primary fragme-'-tati.-•« is trie final

stage in the response of a R/'C slab, Before :;his behavior

is realized, enormous damage is entertained by the slab

and certainly its usefulness has long been exhausted.

Studies of R/C members for conventional applications term-

inate when either the maximum resistance is obtained or when

cracks become unacoeptably large. Even in protective con-

struction, interest is not sustained as far as the frag-

mentation stage. In view of this situation it is not sur-

prising that most investigations of R/C sLsbs are not

relevant to the fragmentation problem.

The formation of a primary fragment requires that the

reinforcing steel around its periphery be fractured. With

this in mind, the current study concentrated on situations

which might lead to fracture or which displayed fracture

of the reinforcement. The difficulty and infrequency of

encountering this response has contributed most significantly

to our conclusion that R/C slabs do not develop primary

fragments.

We can expand our insight into the fragmentation problem

by backing up a little and examining the crack patterns

which are formed in dynamically loaded slabs. Taking a much

more subjective approach than chat used in the case studies,

we can identify three basic causes of concrete cracking;

bending, tensile membrane action, and wedging action of

the reinforcing steeL. Our motivation for studying these

concrete cracks stems from the observation that the steel

stress is greatest at the cracks (Ref. 4) and, consequently,

if steel fracture occurs it will do so at such locations.

71

1 H

C

m ■ ■

i ii

u

i

Beginning with the unreinforesd masonry wall described

in Case 6, we find a strong diagonal crack development

which terminates ir a rectangular crack in the center of the

slab. This pattern is typical of those obtained in studies

of monolithic Shcatigihle slabs subjeccsd to uniform dynamic

pressures (Ref. 13 srid 1A). This remarkable pattern is

caused bv beading stresses at small deflections. We note

that fracture does not occur a: the «.anter of the slab where

the stresses are the greatest-, A statistical explanation

has beer; suggested to describe this behavior. The arguments

employed point out thai; th« high stresses ac the center of

the sla:> invulve only a 3;»ali volume of material and, hence,

there is a small likelihood of finding a critical flaw. At

the other extreae, the border of rzhe slab provides a large

volume in which a critic&i flaw may occur with high fre-

quency; but, here the stresses are quite low, Setweer these

extremes we find a combination of applied stress Mid flaw

likelihood which maximizes the fracture probability, namely,

the rectangular crack pattern«

The addition of steei reinforcement to masonry panels

does not affect the elastic bending crack pattern; horever,

it does retard its onset. More important, it provides a pane)

with an elastic-plastic type moment/curvacure relationship.

This enables the panel to develop yield liner- along which

severe cracking will take place. Plastic collapse modes are

found in several of the case studies dealing !rith R/C slabs.

The classic static collapse mode for square Isotropie slabs

is an X pattern which is formed by yield lines -along the diag-

onals. As pointed out in Case 9, this pattern was obtained in

both full scale and 1/24 scale dynamic panel tests. The Naga-

saki experience provides several examples of yield line devel-

opment. In Case 5 we find plattic hinges ir beam elements, in

addition to a well developed slab yield line. A perfect static

yield pattern is illuscrated in Case 14 for a fixed edge slab,

i.e., yield lines form around the border and along the diagonals

72

1

I ....--■ ...

--:.!

Tht v ork of Denton, Case 13, illustrates the similarity

between the static and dynamic crack patterns in slabs with

mmhomogeneous reinforcement. In all cases, the corners of

the slabs have been reinforced more heavily than the central

region. Furthermore, bends in alternate bars were made near

the edges (1 to 2 ft from the edges) to provide bcth tension

and compression steel in the boundary regions. These bends

produced the large cracks shown in Figures 25 through 28

which parallel the slab edges. Such cracks do not arise in

slabs with homogeneous reinforcement, and except for their

presence, the principal crack pattern is very close to the

classic X mode.

Elastic bending failures characteristically lead to cracks

which do not penetrate the entire slab thickness. In plastic

bending the rotations which concentrate at the yield lines

may lead to cracks which completely penetrate the slab. These

cracks arise from tensile bending stresses acting over most

of the slab section and from compressive shear failure of

the remaining compression layer. In both the elastic and

plastic cases, subsequent tensile membrane action of the slab

will cause partial cracks to pass through the entire slab

thickness and will tend to increase the width of all cracks.

Such membrane action becomes significant at deflections of

the order of the slab thickness.

To complete our account of cracking we recall that a

wedging action is associated with the "pull out" motion of

a deformed reinforcing bar. Where the steel stress is suf-

ficiently high, this action manifests itself in a surface

crack which occurs in the slab face directly above the rein-

forcing bars. Such cracking is discussed in Case 18 and the

resulting crack pattern is clearly illustrated in Figures 27

and 36. This effect was also exhibited in Keevan's Study

which is described in Case 10.

73

■MBWUrn»mM»i.tt

I The influence of tensile membrane stresses in severely

loaded slabs is very pronounced and would be quite sufficient

to cause tension cracks which would sever the slab section.

However, the slab is so extensively cracked by other phenom-

ena before the membrane behavior dominates the stress pattern

that membrane action is largely confined to the extension

and separation of existing cracks such as those caused by

wedging action. As we see in Case 10, the membrane tension

can completely isolate small islands of concrete with the

result that the pressure loads easily pass through the slab

to reduce the net transverse forces acting on the various

regions of the slab.

74

-

•

!

.

BIBLIOGRAPHY

1. Atlas, A., Siess, C. P. and Kesler, C. E., "Behavior of One- _ Way Concrete Floor Slabs Reinforced with Welded Wire Fabric"

J. Am. Cone. Inst., Proc, Vol. 62(5) pp 539-557, May 1965

2. Davies, I. LI., "Model Analysis," Proc. of Symposium on Protective Structures for Civilian Populations, tfatl. Ac'ad, of Sci.-Natl. Res. Übuncil, April 1965

»V

3. Denton, D. R., A Dynamic Ultimate Strength Study of Simply Supported Two-Way Reinforced Concrete Slabs'^ U. S. Army Engineer Waterways Experiment Station,"Offfee of Civil Defense Contract OCD-PS-65-44, April 1967

4. Ferguson, Phil M., "Bond Stress - The State of the Art" J. Am. Cone. Inst., Vol.63(11) pp 1161-1190, November 1966

5. Glasstone, Samuel (Editor), The Effects of Nuclear Weapons, U.S. Department of Defense, U.S. Atomic Energy Commission" February 1964

6. Heugelj W. F. and Feinstein, D. I., Shelter Evaluation Program, Office of Civil Defense, Contract 0CD-PS-64-5TI7~Te1>rüary 1557""

7. Keenan, W. A., Resistance and Behavior of Concrete Slabs under Static and Dynastic uniform Loading wj th^d^eF~Cl&rnpeor"ani Laterally ResTrained, Naval Civil Eng. Lab. T^c^InTcaT™Report, to be published in 1968

8. Rinehart, John S. and Pearson, John, Behavior of Metals under Impulsive Loads, Dover Publications, Inc., New York", 1954*

9. Saemann, Jesse C, Investigation of the Structural Properties of Reinforced Concrete Masonry, National ConcreTeTfasonry Assoc., June 1955

10. Salvadori, Mario G. and Weidlinger, Paul, ,r0n the Dynamic Strength of Rigid-Plastic Beams Under Blase loads," J. Exig. Mech. Div., ASCE Proc, pp 1389-1 to 1389-3&, October" 1957

11. Taylor, Benjamin C, Blast Effects of Atomic Weapcns Upon Curtain Walls and Partitions öTTlason:ry__an5^ütner Materials, OperiTIölTUpshot^KnotHöTe, WT-7TT7""5ugust I95F"

12. Williamson, G. R., Fibrous Reinforcements for Portland Cement Concrete, Ohio River Corps of Eng., Tech. Report 2-40, May 1965

13. Barnett, R.L., Costelio, J.F. and Feinstein, D.I., Accumulation and Removal of Blast-Initiated Debris, Final Report IITRI~ Project M61Ö3, Contract B-70942 (4949A-34-US, Civil Defense Technical Office, SRI, October 1966

14. Liber, T. and Barnett, R. L., An Experimental Investigation of Frangible Plate Fragmentation, IITRI Project M6095, Office of Civil Defense, Contract OCD-PS-64-50, Work Unit 3322B, October 1966

75

!

f

I

I

I

I

APPENDIX

REVIEWED DOCUMENTS WHICH DO NOT CONTAIN SIGNIFICANT WORK

ON SLAB FRAGMENTATION

1. Anderson, Ferd E., Jr., Hansen, R. T.s Murphy, H. L., Newmark, N. M., and White, Merit P., Design of Structures to Resist Nuclear Weapons Effects. ASCE, Manuals of Engineering Practice, No. 42, 1961.

2. Albritton, G. E., Description. Proof Test, and Evaluation of Blast Load Generator Facility. Corps of Engineers Waterways Experiment Station, Tech. Report No, 1-707, December 1965.

3. Allgood, J. R. and Shaw, W. A., Test of Concrete Panels Operation Teapot, WT-1130, February 1957.

4. Barnard, Peter R., "The Collapse of Reinforced Concrete Beams," Proc. Int. Symp. Flexural Mechanics of Reinforced Cone♦, November 1964.

5. Baron, Mark T., "Shear Strength of Reinforced Concrete Beams at Points of Bar Cutoff" J. Am. Cone. Inst. pp 127-134, January 1966.

6. Bultmann, E. H., Sevin, E., and Schiffman, T. H., Blast Effects on Existing Upshot-Knothole and Teapot Structures. Operation Plumb bob, WT-1423, May 1960.

7. Byrnes, Joseph B., Effects of an Atomic Explosion on Two Typical Two-Story-and-Basement Wood-Frame Houses. Operation Upshot-Knothole, WT-792, September 1953.

8. Davids, Norman, International Symposium on Stress Wave Propagation in Materials, Interscience Publishers, Inc. New York, 1960.

9. Gamble, W. L., Sozen, M. A. and Siess, C. P. An Experimental Study of a Reinforced Concrete Two-Way Floor Slab, Structural Research Series No. 211, Univ. of 111., June 1961.

10. Gurfinkel, German and Robinson, Arthur, "Determination of Strain Distribution and Curvature in a Reinforced Concrete Section Subjected to Bending Moment and Longitudinal Load," J^_Am. Cone. Inst. , pp 398-403, July 1967.

11. Hatcher, D. S., Sozen, M. A., and Seiss, C. P., An Experimental Study of a Quarter-Scale Reinforced Concrete Flat Slab Floor, Structural Research Series No. 200, Univ. of 111., June 1960.

12. Johnston, Bruce G., Damage to Commercial and Industrial Buildings Exposed to Nuclear Effects, Operation Teapot, WT-1189, October 1958.

76

1 i I I

13. Kar.i, G.N.J., "Basic Facts Concerning Shear Failure," J. Am, Cone. Inst.. pp 575-691, June 1966.

14. Kolsky, H., Stress Waves in Solids. Dover, New York, 1963.

15. Krefeld, William J. and Thurston, Charles W., "Contribu- tion of Longitudinal Steel to Shear Resistance of Rein- forced Concrete Beams", J. Am. Cone. Inst. pp 325-344, March 1966.

16. Krefeld, William J. and Thurston, Charles W. "Studies of the Shear and Diagonal Tension Strength of Simply Supported Reinforced Concrete Beams," J. Am. Cone. Inst. pp 451-476, April 1966.

17. Mattock, Alan H., "Rotational Capacity of Hinging Regions in Reinforced Concrete Beams," Proc. Int. Symp. Flexural Mechanics of Reinforced Cone. Nov. 1964.

18. Mayes, G. T., Sozen, M. A., and Sless, C. P., Tests on a Quarter-Scale Model of a Multiple-Panel Reinforced Concrete Plate Floor. Structural Research Series No. 181, Univ. of 111., September 1959.

19. Morley, C. T., "Experiments on the Yield Criterion of Isotropie Reinforced Concrete Slabs" J. Am. Cone. Inst. pp. 40-45, Jan. 1967.

20. Nawy, E. G., "Cracking of Slabs Spanning in Two Directions," Cone. and Constr. Eng., LX(10), pp 373-378, October 1965.

21. Newmark, N. M., Air Force Design Manual: Principles and Practices for Design of Hardened Structures. Air Force Special Weapons Center, Tech. Doc. Report No. AFSWC-TDR- 62-138, December 1962.

22. Ngo, D. and Scordelis, A. C, "Finite Element Analysis of Reinforced Concrete Beams," J. Am. Cone. Inst.. pp 152-163, March 1967.

23. Ockleston, A. J., "Load Tests on a Three Story Reinforced * Concrete Building in Johannesburg" The Structural Engineer.

pp 304-322, October 1955.

J, 24. Penzien, Joseph, Behavior of Reinforced Concrete Structural Elements Under Long Duration Impulsive Loads: Part II

f Behavior Within the Elastic Range, M.I.T., Contract No. W-19-016-eng-3215, September 1949.

25. Perry, Ervin S. and Thompson, J. Neils, "Bond Stress Distribution on Reinforcing Steel in Beams and Pullout Specimens" J, Am. Cone. Inst.. pp 865-874, August 1966.

f I I I

f

!

77

* .IsMMW

i .

26. Rotz, J., Edmunds, J., and Kaplan, K,, Formation of Debris From Buildings and Their Contents bv Blast and Fire Effects of Nuclear Weapons. URS Corp., URS 651-4, April 1966.

27. Smith, James E., Symposium on Dynamic Behavior of Materials ASTM Special Tech. Publ. No. 336, 1963.

28. Steyn, Keeve, Behavior of Reinforced Concrete Structural Elements Under Long Duration Impulsive Loads. Part III Behavior within the Plastic Range. M.I.T., Contract No. W-19-016-eng-3215, June 1949.

29. Taylor, R., "A Note on a Possible Basis for a New Method

Mag, of Cone. Res., 17(53), pp 183-186, December 1965,

30. U. S. Army Corps of Engineers, Design of Structures to Resist the, Effects of Atomic Weapons; Arches and Dornen. U. S. Army Corps of Engineers Manual, EM-1110-345-420.T" January 1960.

31. U. S. Army Corps of Engineers, Design of Structures to Resist the Effects of Atomic Weapons: Buried and Semiburied Structures. U. S. Army Corps of Engineers Manual EM-1110-345-421, January 1960.

32. U. S. Army Corpf of Engineers, Design of Structures to Resist the Effects of Atomic Weapons: Multistory Frame Buildings. U. S. Army Corps of Engineers Manual, EM 1110-345-418, January 1960.

33. U. S. Army Corps of Engineers, Design of Structures to Resist the Effects of Atomic Weapons: Principles of Dynamic Analysis and Design, U. S, Army Corps of Engineers Manual EM 1110-345-415, March 1957.

34. U. S. Army Corps of Engineers, Design of Structures to Resist the Effects of Atomic Weapons: Shear Wall Structures. U. S. Army Corps of Engineers Manual, EM 1110-345-419, January 1958.

35. U. S. Army Corps of Engineers, Design of Structures to Resist the Effects of Atomic Weapons: Single-Story Frame Buildings, U. S. Army Corps of Engineers Manual, EM-1110- 345-417, January 1958.

36. U. S. Army Corps of Engineers, Design of Struetures to Resist the Effects of Atomic Weapons: Structural Elements Subjected to Dynamic Loads, IJ. S. Army Corps of Engineers Manual, EM 1110-345-416, March 1957

78

^JMSP^g. :■■ tr*m*nm>ML •-,

i S

.

37. U. S. Army Corps of Engineers, Design of Structures to Resist the Effects of Atomic Weapons' Weapons Effects Data Ü. S. Army Corps of Engineers Manual, EM-lllO-345-413, ~" July 1959.

38. U, 3. Strategic Bombing Survey, The Effects of the Atomic Bomb on Hiroshima Japan. Vols. I. II. and III. May 1947.

39. U- S. Strategic Bombing Survey, Effects of the Atomic Bomb on Nagasaki Japan. Vols. I. II. and III. June 1949.

40. Wells, William M., Jr., Behavior of Reinforced Concrete Structural Elements Under Long Duration Impulsive Loads: Part IV Design. Construction and Operation of Slab Machine, M.I.T., Contract No. W-19-016-eng-3215, September 1949.

41. Wiehle, Carl K. and Durbin, William L., Combined Effects of Nuclear Weapons on NFSS Type Structures. URS Corp. URS 658-3, September 1966.

42. Wood, R. H., Plastic and Elastic Design of Slabs and Plates, The Ronald Press Co., New York, 1961.

43. Yitzhaki, David, "Punching Strength of Reinforced Concrete Slabs" J. Am. Cone. Inst.. pp 527-542, May 1966.

The following journals were reviewed for the time

period indicated. It is significant that only a handful

of references were found to be relevant to the fragmentation

problem. The references that were used in the report are

listed in the bibliography.

1. Journal of the American Concrete Institute,

From September 1955 to December 1967

2. Structural Journal, American Society of Civil Engineers

From July 1956 to December 1967

3. Mechanics Journal, American Society of Civil Engineers

From October 1956 to December 1967

79

I UNCLASSIFIED Security Classification

DOCUMENT CONTROL DATA - R&D (Security ilaeeltlcatlon ol Uli», body el aaattaei and tndeatni annotation mi«: be mnimrmd whan the ownll report .. claaalllad)

1 OBIOIHhTING ACTIVITY (Corpora* author)

IIT Research Institute Technology Center Chicago, Illinois 60616

Kit. REPORT SECUrilTV C LA3SIFICATION

26 CROUP

'

J «EPQtT TITLE

FRAGMENTATION OF REINFORCED CONCRETE SLABS

4 DESCRIPTIVE NOTES (Typt ol «port and inchieiva datot)

Final Report 5 AUTHOR^/ (Laal name, lint nama. Initial)

Barnett, Ralph L. Hermann, Paul C.

6 REPORT DATE

October 1968 »a. CONTRACT OR GRANT NO

Subcontract B-7Gy42(4949A-34)-US b PROJECT NO.

OCD Work Unit 3322B

7« TOTAL HO. OF PASES

80 _i-

7b. NO. OF RBTCS

14 9a. ORIOIMATOR'S REPORT NUMBERi'SJ

..6107

iu. GVHtH AktPO^c MO\S) (Arty >her number» that oiay be maligned thin mpoTt)

10 A VA IL ABILITY 'LIMITATION NOTICES This document may be further distributed by

any holder only with specific prior approval of tfot difica of Civil Defense.

II SUPPLEMENTARY NOTES 12 SPONSORING MILITARY ACTIVITY Office of Civil Pefense Office of the Secretary of the Army Washington, D. C. 20310

13 AflSTRACT

A reexamination of the blast effects literature from the point of view of fragmentation leads to the conclusion thai rein^ forced concrete slabs do not constitute a signifiest source of debris in the postattack environment. Both the Initial orientation and the self-adjusting geometry of slabs minimize their transverse loading. Also, the horizontal displacement of potential slab frag- ments tends to be small because of their high ballistic coefficients and/or high downward acting loading. Finally, the steel reinforcing bars tenaciously tie the various pieces of fractured slab to the supports and to each other even at pressure levels as high as 100psL

DD /Ä. 1473 UNCLASSIFIED Security Classification

■■-

UNCLASSIFIED Security Classification

:

KEY WORDS

Reinforced Concrete Fragmentation Debris

LINK A

HOi_E »T

INSTRUCTIONS

LINK C

LE ' WT

1. ORIGINATING ACTIVITY: Enter the name and address of the contractor, subcontractor, grantee, Department of De- fense activity or other organization (corporate author) issuing the report.

2*. REPORT SECXHTY CLASSIFICATION: unter the over- all security classification of the report. Indicate whether "Restricted Data" is included. Marking is to be in accord- ance with appropriate security regulations.

26. GROUP: Automatic downgrading is specified in DoD Di- rective 5200.10 and Armed Forces Industrial Manual. Enter the group number. Also, when applicable, show that options! markings have been used for Group 3 and Group 4 as author- ized.

3. REPORT TITLE: Enter the complete report title in all canital letters. Titles in all cases should be unclassified. i_ a meaningful title cannot be selected without classifica- tion, show title classification in all capitals in parenthesis immediately following the title.

4. DESCRIPTIVE NOTES: If appropriate, enter the type of report, e.g., interim, progress, summary, annual, or final. Give the inclusive dates when a specific reporting period is covered.

5. AUTHOR(S): Enter the name(s) of authors) aa shown on or in the report. Enter last name, first name, middle initial. If military, show rank and branch of service. The name of the principal author is an absolute minimum requirement.

6. REPORT DATE: Enter the date of the report as day, month» year; or month, year. If more than one date appears on the report, use date of publication. 7a. TOTAL NUMBER OF PAGES: The total page count should follow normal pagination procedures, i.e., enter the number of pages containing information.

76. NUMBER OF REFERENCES: Enter the total number of references cited in the report.

8a. CONTRACT OR GRANT NUMBER: If appropriate, enter the applicable number of the contract or grunt under which the report was written.

8b, 8c, & ad. PROJECT NUMBER: Enter the appropriate military department identification, such as project number, subproject number, system numbers, task number, etc. 9«. ORIGINATOR'S REPORT NUMBER(S): Enter the offi- cial report number by which the document will be identified and controlled by the originating activity. This number must be unique to this report. 96, OTHER REPORT NUMBER(S): If the report has been assigned any other report numbers (either by the originator or by the sponsor), also enter this number(s).

10. AVAILABILITY/LIMITATION NOTICES: Enter «ay lim- itations on further dissemination of the report, other than these imposed by security classification,, using standard statements such as:

(1) ' Qualified requesters may obtain cooies of this report from DDC"

(2) "Foreign announcement and dissemination of this report by DDC is not authorized"

(3) "U. S. Government agencies may obtain copies of this report directly from DDC. Other oualifie-* DDC users shall request through

(4) "U. S. military agencies may obtain copied of this report directly from DDC Other qualified u*ers .'hall request through

(5) '■" \11 distribution of this report is controlled. Qual- ified DDC users shall reques'. through

If the report has been furnished to the Office of Technical Services, Department of Commerce; for sale to the public, indi- cate this fact and enter the price, if known. 11. SUPPLEMENTARY NOTES: Use for additional explana- tory notes. 12. SPONSORING MILITARY ACTIVITY: Enter the name of the departmental project off'ce or Imboratory sponsoring (pay- ing for) the research and development. Include address. 13. ABSTRACT: Enter an abstract giving s brief and factual summary of the document indicative of the report, even though it may also appear elsewhere in the body of the technical re- port. If additional space is required, a continuation sheet shall be attached.

It is highly desirable tnat the abstract of classified re- ports be unclassified. Each paragraph of the abstract shall end with an indication of the military security classification of the information in the paragraph, represented as (TS), (St), (C), or (V).

There is no limitation on the length of the abstract. How- ever, the suggested length is from 150 to 225 words. 14. KEY WORDS'. Key words are technically meaningful terms or short phrases that characterize a report and may be used as index entries for cataloging the report. Key words must be selected so that no security classification is required. Iden- fiers. such as equipment model designation, trade narre, -nili- tary project code name, geographic location, may be used as kev words but will be followed by an indication of technical context. The alignment of links, rules, and weights is optional.

UNCLASSIFIED Security Clasjsification

■■

msmsturvnMiCrWi

I 0

! r

i

0 |C

i j..

Ic !

[

*"-\ ? a « <u a •jd V « o -a b

a C «3 **■, i 3 *« « « i-i H • X a • a w ETJ b mjs u a m « i-l «X CUK <" « 1 o U 0 u • 0 "JH II Ob •4 00 C cu-a w > aw *■ a Q H o wo «w a 1 IIIUUHgHtHtll

UvO C C 0 « M «W * 0 * UW C li.M C 5) s^

bCT> «1 il «1 W UHJ^ c ew a u o-j c at», i..«

5 hllK OX > lm9o<<« aa

<u a o a or - 5 £»-ii aooepo u

41 U |U W B W V

« a b • • ctix

0 H« b «i-. 'w^^ :-x

as 3 c !-«■-» o -a « « U 0 «W H M a > o «i-l U BW UW 0

" n aww a 6 «w «x «a 3UJ -i fix £ u «ww a E p u ecu q

W -H o C£it 6 c « w E at« ■ oWl w •

ow a o u -i o u H Ilit^lCH ü J« i

c. j « u A a « a •i o « « o a a e ivs s pi «X OF<ri«H( H m « « at a »< a a si 44 u o a 3

H JX G

g 8 «I <S03 ■fnojoi ■ < U U U 0 <J3 S-H w /N 4-1 WE « 1 (111

P ■»

e bi

as

leads

const

Ironme

ometry

orizon

small

Igh do

a tena

auppo

pal. u en

G « i S u •<

0 o c -3- 2 «1 a x c « > «x «x u «o

K Pä -4 UHOCK x axo a -< C« a bX «w

W CM CO IK M «IX 0 0 o a OJC s « «^ go a

UT3 OW T3 CU « 2 -IM tb OICM

O W ©CO B B B « - > S H [X r-*n oaaumoaa uxx

i-i i JJ « 3 i e u a to «8>aa-r->i«aoww

85 i o pp « M W a a .--» u*a < « «<M i£ H

OS B b a » a c c < it»4 0 i a HI-ITS a

a a a a* ■ u w • i a 3't-i « HKCukli Kovoa c «-H 3 a

95 n X « b £ oS a K A »-S S "-' H « IH

tjim -c SC<M a o « nie » a « a « a > 2 0

pc 38 i-ioSAObOub« m <>UTHJHI»IH»IH

to O

lo I Li

i u

10 §

r

■>< a u a.

a o w IS

«op o« IJO. aw a o u « a

o

o

u g-o

TJ a o (j Ik

0 ll Ui-v

s

«I u

Hi a i-i

.3 «n ■H O w c

e

I 3

y a Si

«

-1 •in u axi u m nena K-i « a o «Ho -o H fli^ > a i-i *> a

u a UH a <u a en arr-i a u o «J a *»< am

C H^Hce • aw ;M>ITI B o-n cass.cs,« hbua^ > 9 i)-S u a • 5 • C£ • 3 aw« coo«o u

9 u i-i d-w • U-H '-«X-rl IJ.C cult -e a « a-i N u a > o U C -rl U-rf O

CO« S 'JHH-J • i-i a on»* o ax aw gnu

uxi a.« a a o o a i >st>

BH-I 313 w c • ail IbM U , Ü 5 3 9 >Wri « > C a ij « o a - c -ri *-'

« Sob ►.en o 1 o

"WO g 0.

PC b

§ o. z a HI a;

o

CM as »CM a* CM

I 50 «

•rl

til 1 b Jt « E

38 tQO

« u o

iS K O 91

tl < >

U > «£ « J= b «O 05« X B.20 C • 9 b.n UH

J «us o o ton eu W-. too> JT3 Oi-l -0 B u « 3 « u -a B-H |i B u a OT) a UXI X 3£U3«C bans o a a —»■< oi a o I-I i-i HUKUUH, a £ a a a fl B

0 i a CiiD a e a.'M • « —; a es a u w e*B o b M c e a E v~* 9 a bjlvl 3 <M r-l U W u -; -o Stfu a o a gm * a « a a >

B£ Okouiiii UilüH1-! <l «ll-liH

/■■»

; is a 91 iu a iH c W o "a 6 BE S) 3 «X 00 3 1-1 a l-l w i-i a a X a ti a. ■rf BtJ b ax u a a « ■H aaa E b a >w a a o a 0 O « O 81 r-l o • 0 b w CO E ctiTj-ri >«-! >\ a 0 H o «a «<-, «

i vmmiG'-'&'-'Vu Xoo XEouaawaua

s** u3 «W ChrlrlC» bCf- a vauuri-iii c »H s b u-H E si» an

5 b b b e O > a a b b « 3 ciw « a a

«ao a ox • 3 X anj &KKO b

a a CS..O

O oi u a« je»«« HU b u-ri iwxw b£

3 B B <-> w 0 Dun ij o a-ri H b a > O MO e-H o c<i «i-i o

« b B-rf-H S B «J r-l 9J.E

UrHllJBfllKlia i-i o CXi^ SB« H ß oc« b ow-i a

fjj CO

Oil a o «i-i o

K H a o ax «r-i o « « u u ux 0.« a i «c uaaoana ^TS £ ax cBi-ii-i 3T3-I B I H <««« Bisabaa

8 i

(/]■«« y a 3

8 pa si o 3 'HHaie« < «««oaxBi-i«

« ilC « Susi aaui >>CH 0 oj o «•o a § b orit 2 a ■ H4BBUN9] V 0,1-1 x « o o «i-i ex « 3 B

r-iiibfibBeo aa « i-l O C i-l B

,—s Q

b i •<

0 CTs 0. t s 91 X 6« > «X «X b «O

M 06 -a- «oca* x axo •H 6 « « bXUr-l § ^

W «M CO lb u MX 0 O o a ox 6u u-~ ceo a

u-e uri TS E« a s •a-tN hi CMN O 15 0<n EC a« »«Bi-i

o«a«BO'Oaoxx -i Sx « 3 o B b a 60

(I, M»S z i O pa u « »a ai-iw S B oPMW tH i-i a afi-l «»j<««i|-l BX s OS

B b M a a B B ■tjü 0 1 B «i-fO B

a a « o><« • IJ i-i m a a Z bja ami rHtCBObb

M 0 « « « C «-I 3 a w b n u 91 b X B i-l SW rH U r-l

• | o « is So*« « o a b a c a a)•« a >

i4 5 SX o b o « b « k EM OiliHiHiuiiaiiH 0T O

■o « £ 41 <H « «4 « « 0-0 b

BE «3 ■W U) 3 W a « « w a« x a a 0. w Et> b «x « a a « •« «J R b a IM « a o a 0 O « O «r-l O • 0 b

rH 03 E C. b 13 i-l > B -rl Si o. 0 M o «a «w a

1 «iMU-i a Gri aw a« XCO x B o « a a i-i a o a

i*^ UM3 «W B b-Mr-l B « b» « « « « U r-l W i-l E aw 8 b O-H CtrMll

0 b b b «X > b«30i-iu B«

a B b a a « a o « ox • 3 USX X awx acoeoo b

a « a « i* e-H « o 8-i« b UM UHXW bX

3SBU«0 T3B« u 5 aw N b a > o as Bi-I O BiH «i-l o

w b B-rti-l B B «w «X a jiM i-i ax XO wwi-i « B i « oou a WOBX-H« E « r-i 6 a« S 0>«-H C

2

to Orl a O «-rl 0

B f-i a u ax aw o « « u « ux o.« a i <- uaaoaca txxi 05 «X COW-rlS-OWS

z P •« « « S5T3 a b a a w IM »j o a 3 g g pq a 0 3 -luw a » o a 2 «««oaxBw« 1 /^ « w B « Sob

a M Ä a a « a ><Ew o a o

at» a a b OriiJ fi 0. • o « 1 wasBwM« a aw

X « o 0 «w gx « 3 B s b < o 0 o> wobSbBoo aa i o. <t « WOO W B a <T. XS«>«X«Xb «O

R OS -* uooeK x axo WE« « bX ww . ^—'

w CM pa •M u 00X O 0 oaoiHuii^aaa 3 -*CM

[J, OICM ut o-H -as«« o »j O C*l 6 c a« -BBW

6 i a « a OTi a uxx w ax u 3 a 6 b a 00

Ex r^ci z I o pa « « Ba «r-|H a B oww r-l •H aaw«-o<««i«BX

B b a a a B B H ti < wiM oi a«wua a a a a1« • « w « « a

SIM U W ope o b b K08««B8W 3a « bX aw HIM w «w « S 0 « T> &fH « O « bus «aa««a>

w 0 BX O b 0 >-■) u a

35 b M u b 2 E 0

Ö 03 O XO

CL, 3 U <>(iwuwu-iueiMw OT O