AD-A250 59 MISCEL-LAN EC US PAPER SL-E'2-2

BRICK MODEL TE ST"S -OF 3'HALLOWof EngneersUNDERGROUND MAGAZINESf

Charies E. joacmmr. 53ecoe S. Ruobir oe a Borcoila

*L. DE PAR JMEHT T THE ARM%1YWaterways Experiment Station. Corps of E-igineers

7 3909 Halls Ferry Roao. Vicksourg. Mississippi 39180-6199

V7 *4

T TIi

M'iaron '992__rflnai Recort M

4n:orcvi~c i-or P-,c Rie~ s !2.....M

US Department of Ceferis _xlsysSafety Board

Directorate of Heajth and SaftrMinistry of Defence. United Kingdom

~ Norwegian Defence Construction ServiceLABORATORY

92 5 2~ o

Destroy this report when no longer needed. Do not return

it to the originator.

The findings in this report are not to be construed as an officialDepartment of the Army position unless so designated

by other authorized documents.

The contents of this report are not to be used foradvertising, publication, or promotional purposes.Citation of trade names does not constitute anofficial endorsement or approval of the use of

such commercial proaucts.

REPOT D CUM NTATON AGEForm Approved

REPOT D CUMNTAIONPAG OMB No. 0704-0188

Publt re=ortng burden for this collection of ioformation is stimated to arerage 1 hour per respons, including the time for re-iewing Instructions, searching existing data sources.gatheing and maitaining the data needed, and comleting and reuiewnq the colleon of iformation. Send comments regarding this burden estimate or any other aaeect of thiscollection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1 215 JeffersonDavis Highway, Suite 1204. Arlington, VA 22202-4302, and to the Office of Management and Budget. Paperwork Reduction Project (0704-0188), Washington DC 20503.

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

I March 1992 Final Report4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Brick Model Tests of Shallow Underground Magazines

6. AUTHOR(S)

Charles E. JoachimGeorge S. Rubin de la Borbolla

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) B. PERFORMING ORGANIZATION

USAE Waterways Experiment Station REPORT NUMBER

Structures Laboratory Miscellaneous Paper

3909 Halls Ferry Road SL-92-2

Vicksburg, MS 39180-6199

9. SPONSORING/ MONITORING 61--ENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING

US DOD Explosives oafety Board AGENCY REPORT NUMBER

Hoffman Building 12461 Eisenhower AvenueAlexandria, VA 22331-0600

11. SUPPLEMENTARY NOTESAvailable from National Technical Information Service, 5285 Port Royal Road,Springfield, VA 22161

12a. DISTRIBUTION/ AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

Approved for public release; distribution is unlimited

13. ABSTRACT (Maximum 200 words)

A series of 1:25-scale model tests in a brick test bed were conducted toinvestigate the airblast and debris hazards from accidental explosions inshallow underground magazines in jointed rock.

The results demonstrated that the blast effects produced in a large-scaletest of a shallow magazine could be reproduced with reasonable accuracy in a1:25-scale model. The model tests also provided an experimental basis forextending the results of the large-scale test to account for variations in thethickness and strength of the rock cover and in the chamber loading density.

14. SUBJECT TERMS 15. NUMBER OF PAGESAmmunition storage Underground magazines 81Explosives safety 16. PRICE CODEAccidental explosions

17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACTOF REPORT OF THIS PAGE OF ABSTRACTUnclassified Unclassified I

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)Prescrbed by ANSI Std 13g-i8

298-102

PREFACE

The model test program reported herein was performed by the

Explosion Effects Division (EED), Structures Laboratory (SL), of the U.S.

Army Engineer Waterways Experiment Station (WES), during the period

1 July 1989 through 30 December 1990. The investigation was sponsored by

the U.S. Department of Defense Explosives Safety Board (DDESB), the

Norwegian Defence Construction Service (NDCS), and the Directorate of

Health and Safety (DHS), Ministry of Defence, United Kingdom.

The program monitors were Dr. C. E. Canada, DDESB, Mr. A. Jenssen,

NDCS, and Mr. Jon Henderson, Explosives Storage and Transport Committee,

DHS.

The Program Manager for this study was Mr. Charles E. Joachim.

Mr. George S. Rubin de la Borbolla was the Project Engineer and conducted

the tests, and Messrs. Joachim and Rubin prepared this report. Other EED

personnel assisting in the study were Messrs. D. F. Hale, R. N. Walters,

and G. E. Phillips. Mr. C. N. Thompson, WES Instrumentation Services

Division, provided instrumentation and data recording support.

Mr. Eugene J. James, WES Engineering and Construction Services Division,

assisted with model construction.

The work was performed under the general supervision of Mr. L. K.

Davis, Chief, EED, and Mr. Bryant Mather, Chief, SL.

Dr. Robert W. Whalin was Director of WES. COL Leonard G. Hassell,

EN, was Commander and Deputy Director.

Accesslon ?orNTIS GRA&I

DTIC TABUnannouncedjustification

By

Distributlon/

AvailabilitY odes

/ Avail and/or~ ) Dist Special

CONTENTS

Page

PREFACE............................................................... i.

LIST OF FIGURES......................................................1iii

LIST OF TABLES......................................................... v

CONVERSION FACTORS, SI (METRIC) TO NON-SI UNITS OF MEASUREMENTS ....... vi

SECTION 1 INTRODUCTION...............................................1I

1.1 BACKGROUND.................................................. 11.2 OBJECTIVE................................................... 2

SECTION 2 PROCEDU3RES................................................. 3

2.1 APPROACH.................................................... 32.2 MODEL CONSTRUCTION.......................................... 32.3 INSTRUMENTATION............................................. 42.4 EXPLOSIVE CHARGES........................................... 52.5 EJECTA/DEBRIS COLLECTION.................................... 6

SECTION 3 RESULTS AND DISCUSSION..................................... 7

3.1 AIRBLAST EFFECTS............................................ 73.2 DEBRIS EFFECTS.............................................. 10

SECTION 4 CONCLUSIONS AND RECOMMENDATIONS............................. 14

4.1 CONCLUSIONS................................................. 144.2 RECOMMENDATIONS............................................. 14

REFERENCES..................................................... ....... 16

APPENDIX A: AIRBLAST AND GROUND MOTION WAVE FORMS, TEST 1 ............ 40

APPENDIX B: AIRBLAST AND GROUND MOTION WAVE FORMS, TEST 2 ............ 49

APPENDIX C: AIRBLAST AND GROUND MOTION WAVE FORMS, TEST 3 .............64

LIST OF FIGURESFigureNo. Page

1. Access tunnel cross section for 1:25-scale Brick Model Tests ..... 17

2. Storage chamber cross section for 1:25-scale Brick Model Tests... 17

3. Layout (centerline profile) of 1:25-scale Brick Model ............ 18

4. Testbed design for 1:25-scale tests .............................. 19

5. Tunnel portals and foreground areas of the large-scale ShallowUnderground Test and the corresponding Brick Model Test .......... 20

6. 1.27-kg explosive charge before insertion into chamber ofBrick Model Test ................................................... 21

7. Brick Model Test 3: location of ejecta collection areas .......... 22

8. Scaled arrival time versus scaled distance from the chargeinitiation point ................................................... 23

9. Free-field side-on overpressure scaled to the Shallow UndergroundTunnel/Chamber Explosion Test parameters ......................... 24

10. Peak side-on overpressure versus horizontal distance from thecharge initiation point ........................................... 25

11. Peak side-on impulse versus horizontal distance from the chargeinitiation point ................................................... 26

12. Pressure-distance comparisons from existing data on modeland large-scale detonations in underground magazines ............ 27

13. Airblast Inhabited Building Distance versus total loadingdensity for selected model and large-scale tests ................ 28

14. Photos of debris fields from Brick Model Tests .................. 29

15. Ejecta distribution as a function of azimuth along the6.2 m arc; Brick Model Test 3 ..................................... 30

16. Launch velocities of cover rock ejecta ........................... 31

17. Comparison of relative debris densities from Brick Model Test 3with the Shallow Underground Tunnel/Chamber Explosion Test ..... 32

iii

LIST OF FIGURES (Cont'd)FigureNo. PaRe

Al-A7 Brick Model Test 1: Airblast gage records ................. 41-47

A8 Brick Model Test 1: Accelerometer record ..................... 48

Bl-BIO Brick Model Test 2: Airblast gage records ................. 50-59

Bll-B14 Brick Model Test 2: Acceleration and velocity records ..... 60-63

Cl-C8 Brick Model Test 3: Airblast gage records ................. 65-72

iv

LIST OF TABLES

TableNo. Page

1. Brick Model Tests: Measurement Type and Gage Location Data ....... 33

2. Brick Model Tests: Explosives Charges and Chamber Cover

Thicknesses ......................................................... 33

3. Brick Model Test 1: Airblast Pressure and Ground ShockMeasurements ....................................................... 34

4. Brick Model Test 2: Airblast Pressure and Ground Shock

Measurements ........................................................ 35

5. Brick Model Test 3: Airblast Pressure and Ground ShockMeasurements ........................................................ 36

6. Effects of Magazine Cover Resistance on Airblast ................... 37

7. Brick Model Test 3: Brick Ejecta Distributions .................... 38

CONVERSION FACTORS, METRIC (SI) TO NON-SIUNITS OF MEASUREMENT

SI (Metric) units of measurement used in this report can be converted toNon-SI units as follows:

Divide By To Obtain

cubic metres 0.02831685 cubic feet

kilograms 0.4535924 pounds (mass)

kilograms per cubic 16.01846 pound (mass) per cubicmetre foot

kilopascals 100. bars

kilopascals 6.894757 pounds (force) persquare inch*

metres 0.3048006 feet

metres per kilogram"/3 0.3955977 feet per pound"/3

milliseconds 1000. seconds

radians 0.1745329 degrees (angle)

square metres 0.09290304 square feet

* For pressure, 14.7 psi = 1.0 atmosphere = 1.014 bar = 101.4 kPa

vi

BRICK MODEL TESTS OFSHALLOW UNDERGROUND MAGAZINES

SECTION 1

INTRODUCTION

1.1 BACKGROUND

A considerable amount of research has been performed in the last two

decades to develop data and prediction methods for airblast and debris hazards

from accidental explosions in underground magazines. Much of this work is

concerned with detonations in magazines so deep that venting does not occur.

The effect of cover venting on reduction of external airblast was initially

investigated in small-scale tests (1:25) performed in the United Kingdom

(Millington, 1985). More recently, the Shallow Underground Tunnel/Chamber

Explosion Test (Joachim, 1990), sponsored by the KLOTZ Club*, provided full-

scale airblast and debris/ejecta data for a shallow underground magazine

containing 20,000 kg**, net explosive weight (NEW).

In 1980, the U.S. Army Engineer Waterways Experiment Station (WES)

conducted a series of intermediate-scale (500-kg charges) cratering tests in

open shafts in rock for the Federal Republic of Germany (Davis, 1981). The

results showed a definite effect of rock strength and structure (jointing) and

terrain surface slope, as well as the charge cover depth, on the size and

shape of the crater produced, and on the amount, direction, and velocity of

ejecta thrown out. Earlier 1:75-scale model tests by WES (Joachim, 1988 and

Smith, 1989) of fully contained detonations in long chambers in rock showed

that formulae for predicting strain and spall velocities in adjacent chambers

must include, as a minimum, a characteristic modulus value for the rock

surrounding the chamber.

* The KLOTZ Club is an ad hoc committee, representing the defense

agencies of France, Germany, Norway, Sweden, Switzerland, the United Kingdom,and the United States, which sponsors research to improve the safety ofammunition storage.

** A table of factors for converting SI units of measurement to Non-SI

units is presented on page vi.

I I |1

These results strongly imply that, at large scales where extensive

volumes of rock must be moved during the venting process, the gross (as

opposed to unit) strength and structure of shallow rock covers may be

important factors in predicting the extent of rupture of the cover, and the

ejecta hazard ranges, from site to site. This is in addition to the known

problem of accounting for the variations in cover thickness and surface slope.

The 1988 Shallow Underground Test provided data for a single site,

consisting of a weak, highly jointed rock. In actual practice, however,

magazines of this design have been constructed at sites having a wide range of

rock strengths and cover thicknesses over the magazine chamber. In addition,

the loading densities of the magazines differ from site to iite. To

investigate the influence that these variations would have on the external

blast hazards from an accidental explosion, and the extent that their effect

can be reproduced in small-scale tests, a series of model tests were conducted

in this study to provide a comparative basis to evaluate these factors. In

particular, it was desired to compare the external airblast (beyond the tunnel

portal) with that measured in previous model tests which simulated solid rock

or soil (sand) covers, as well as the airblast measured in the large-scale

1988 Shallow Underground Test.

1.2 OBJECTIVE

The overall objective of the Brick Model test program was to determine

the hazardous effects (airblast and debris) produced by an accidental

detonation of explosive stores which ruptures the overhead cover of an

underground magazine. Specific test objectives were to evaluate the effects

of explosive loading density (kg of explosive per m3 of chamber volume) and

the thickness and strength of the rock cover on the external blast hazards,

and the ability of small-scale models to reproduce large-scale test results.

2

SECTION 2

PROCEDURES

2.1 APPROACH

Models of underground munitions storage magazines were constructed at

1:25 scale and tested to evaluate the effect of cover depth, cover

structure/strength, and loading density on ejecta throwout and venting relief

of the access tunnel airblast pressures. The basic model consisted of a

single storage chamber and access tunnel constructed in a large testbed of

paving brick, simulating a jointed rock mass. The dimensions of the storage

chamber and access tunnel corresponded to a 1:25-scale model of those

constructed for the 1988 Shallow Underground Tunnel/Chamber Explosion Test.

The model access tunnel was 1.0 m long with a cross-sectional area of 84.4 c,.

(9.6 cm in height and width; see Figure 1). The model storage chamber was 72

cm long, with a cross-sectional area of 294.4 cm2 (20 cm wide by 16 cm high;

see Figure 2).

Dynamic measurements included: (1) chamber pressures, (2) access tunnel

pressures, (3) external airblast pressures, and (4) motion (acceleration) of

the simulated rock mass above the chamber. The airblast and ground motion

gage locations are shown in Figure 3 and listed in Table 1. Passive

measurements consisted of post-test surveys of debris distributions for

Test 3, and observations of the extent of cover rupture and debris throw on

all three tests.

2.2 MODEL CONSTRUCTION

All models were constructed with solid paving bricks (without mortar)

inside a reinforced concrete containment structure, as illustrated in

Figure 4. Nominal dimensions of the bricks were 5.8 by 9.3 by 19.7 cm. As

shown in Figure 4, the bricks were laid with the wide face (9.3 by 19.7 cm) in

the vertical plane, and with the long axis rotated 30' from the vertical, in

the direction of the portal. Thus, the overburden surface slope of the models

was 30 degrees. A thin layer of sand was placed over the surface of the

bricks to simulate soil overburden.

3

The chamber and access tunnel were formed around galvanized steel sheet

metal, shaped to the required rross-sections (Figures I and 2). A layer of

mortar approximately 4 cm thick was placed around the top and sides of the

chamber form to fill voids between the form and the bricks, and bricks were

placed around the assembly. The same procedure was used to form the access

tunnel in the model. The chamber was constructed first, and the sheet metal

form removed prior to installation of the tunnel section. Figure 5 compares a

front view of the completed model (for Test 1) with a similar view of the

large-scale Shallow Underground Test magazine.

2.3 INSTRUMENTATION

Two accelerometers were positioned in the overburden above the

tunnel/chamber centerline to measure the motions of the cover material for

each test. Selection of accelerometer gage ranges was based on an

extrapolation of ground shock data from the large-scale Shallow Underground

Test. The accelerometers were mounted in small, WES-designed, two-part

cylindrical canisters. Each gage canister consisted of an outer cylinder with

a tapered, truncated, conical inner surface, and a matching insert with a flat

center section milled parallel to the axis. The gage was mounted on this flat

section before the tapered plug was inserted into the outer cylinder. The

canister was cemented into a hole drilled in a solid brick, which was placed

at the desired gage location in the model. The canisters were oriented to

measure motion perpendicular to the surface slope of the model. Cables were

placed in 6.4-mm-diameter stainless steel tubing, which was routed away from

the chamber.

Four internal airblast pressure gages were installed for each test--two

gages in the chamber wall and two in the access tunnel floor--to record the

internal pressure environment. The gages were mounted in metal fixtures

inserted into 1.9-mm-diameter PVC pipe. The pipes were routed between the

bricks to the side and rear of the test bed. The voids around the PVC pipes

were packed with masonry sand to minimize the density discontinuity.

Six free-field pressure gages were permanently installed in front of the

tunnel portal, along the extended tunnel/chamber centerline. The gage mounts

were cast into a 10-cm thick concrete slab. The concrete slab was 1.8 m wide

4

and extended a distance of 6 m from the tunnel portal. The surface of the

pavement was level out to a distance of 1.5 m, where a downward slope (11

degrees) began.

One total (stagnation) pressure measurement was attempted outside the

tunnel at a distance of 20 cm from the portal. A probe fashioned from 6.4-mm

diameter stainless steel tubing was attached to a gage mount cast into the

concrete slab. The entrance of the tube was positioned 2.5 cm above the

surface-mounted, side-on overpressure gage located 20 cm from the access

tunnel portal.

The gage signals were amplified, digitized and recorded on computer-

controlled, transient data recorders. The gage signals were transmitted to

the recording trailer on 4-conductor, shielded cable (approximately 200 m

long) with a floating ground. Each channel was electrically calibrated at the

digitizer by the equivalent voltage method.

2.4 EXPLOSIVE CHARGES

The explosive charges were assembled from 0.085-kg/m (400-grain per

foot) PETN detonating cord, cut in 48-cm lengths. The detonating cord was

bundled around and taped to a thin wood slat, which was pushed through the

access tunnel into the chamber (Figure 6). The slat was just long enough to

be contained completely within the chamber (approximately 72 cm). The

explosive charge was approximately 5 cm in diameter for Tests I and 3, and 1.3

cm for Test 2. All charges were initiated from the portal end using a

Reynolds Industry exploding bridge wire (EBW) detonator (Model RP-83). Charge

weights, chamber loading densities, and dimensions of the explosive charges

are given in Table 2.

5

2.5 EJECTA/DEBRIS COLLECTION

Both previous test experience and analytical studies have clearly shown

that, while debris throw ranges and relative distributions can be scaled by

model tests, the areal density (impacts per m2) cannot. This is because the

model material which comprises the debris source does not break up with the

same size distribution as does the prototype material. Consequently, for

Tests 1 and 2, only the maximum range of ejecta/debris was recorded. However,

a more detailed ejecta survey was made after Test 3. The boundary of each

sample area was marked on the ground using a square frame covering an area of

I m2 . The location of the sample areas are shown in Figure 7. During

sampling, the frame was centered at the required distance and azimuth and the

number of pieces of brick, either whole or broken, within the frame area was

counted. The distribution data was broken down into the number of pieces

smaller than half of a brick, and those larger than half.

6

SECTION 3

RESULTS AND DISCUSSION

3.1 AIRBIAST EFFECTS

3.1.1 Comparisons of Model and Large-Scale Test Results. The airblast

arrival times, peak pressures, and peak impulse values (from integration of

the time histories) obtained on the Brick Model Tests are given in Tables 3,

4, and 5. Scaled arrival time data are plotted versus scaled distance from

the initiation ends of the explosive charges in Figure 8. A comparison is

shown with the airblast arrival times measured on the Shallow Underground

Test. There is excellent agreement between the model and prototype

measurements for the external gages. The arrival times for the gages inside

the model tunnel are slightly lower (i.e., faster) than occurred inside the

prototype tunnel.

The prediction curves for airblast peak pressure (external) originally

developed for the large-scale Shallow Underground Test are presented in Figure

9. The Shallow Underground Test and the corresponding model (Test 1) data are

included for comparison. The distances from the model portal were multiplied

by the 1:25 scale factor in this plot to match the prototype scale. As shown

here, the model data clearly falls within tue band spread of the predictions.

3.1.2 External Airblast Attenuation - Comparison with Previous

Experiments. In Figure 10, the ratio of calculated exit pressure to measured

free-field overpressure is plotted versus normalized distance from the tunnel

portal for all available data from previous tests of underground magazines.

The Brick Model Tests (WES Model (1:25)) are included, along with six other

model series, and two full-scale tests, including the Shallow Underground Test

(KLOTZ (88)). The exit pressures were calculated from the relation given by

Vretblad, 1988:

Pw - 17.7 ( Q / VT )0.45

where P, is the exit pressure, bars

Q is the TNT-equivalent explosive weight, kg

and VT is the total volume of the underground facility, m3.

7

A reference line through the data in Figure 10 can be expressed by the

equation

P / P. 0 - 1.0 ( R / D )1.35

and D- 4 A /p

where PSO is the free-field overpressure, bars

R is the horizontal distance from the portal, m

D is the hydraulic diameter of the tunnel (for turbulent flow), m

A is the minimum cross-sectional area of the tunnel, m2

and p is the perimeter of the minimum cross-sectional area, m

As shown in Figure 10, the data exhibits considerable scatter, with many

of the points lying above the reference line. Note however, that the results

of both the Brick Model tests and the Shallow Underground Test (solid data

points) fall well within the scatter band, near the reference line.

3.1.3 Effect of Cover Strength and Thickness. Table 6 lists the

Inhabited Building Distances (scaled up to full-scale ranges) derived from

five model tests with similar loading densities, but different scaled cover

depths and cover material strengths. There is a clear trend in the effect of

the overall integrity of the chamber cover on the IBD. With similar cover

thicknesses and loading densities, the heavily-jointed brick model produced

about the same long-range blast pressures as did the Norwegian sand model.

Based on the IBD's, however, the Norwegian concrete model produced a long-

range blast pressure equivalent to a heavily-jointed brick model with almost

twice the cover thickness.

Measured peak pressures from all three Brick Model tests are plotted in

Figure 11. The DDESB airblast prediction equation and the curve fit to the

peak overpressure data of the Shallow Underground Test are included in Figure

11 for comparison. Although there is some data scatter, certain trends are

indicated. When the cover depth was held constant and the loading density was

reduced from 60 to 10 kg/m 3 (Test 2 versus Test 1), the portal pressure was

reduced by a factor of about four, and the long-range external pressures were

about halved. When the scaled cover depth of the brick models was increased

8

from 0.44 to 0.79 m/kg /3 (Test 3 versus Test 1), and the chamber loading

density held constant at 60 kg/m3, the side-on overpressures outside the

tunnel portal increased an average of 30 percent. The peak pressure midway

down the access tunnel increased by about 130 percent. When the scaled cover

depth was held constant at 0.8 m/kg"/3 , an increase in chamber loading density

from 10 to 60 kg/m3 (Test 3 versus Test 2) produced an average of 250 percent

increase in the free-field side-on overpressure outside the portal.

From Figure 11, it is interesting to note the degree to which the

internal and external airblast overpressures measured on the large-scale,

Shallow Underground Test were reproduced in the 1:25-scale brick model (Test

1). In general, the model provided a good representation of the prototype

results. The tunnel exit pressures match very closely, but external

overpressures were low by a factor of approximately three in the free-field.

However, these lower pressures may have been due to the downward slope of the

ground surface constructed for the model (see Figure 3) at the far-field gage

stations.

The peak impulse values from the model tests, obtained by integrating

the overpressure-time histories, are plotted versus distance from the charge

initiation point in Figure 12. The peak impulse data curve from the Shallow

Underground Test are included in Figure 12 for comparison. Although peak

impulse shows more scatter than the overpressure data, the model and prototype

data clearly follow the same trends.

3.1.4 Responding Versus Non-Responding Magazines. Figure 13 is a plot

of Inhabited Building Distance (distance to the 5.0 kPa pressure level) versus

loading density for selected tests, where the loading density is based on the

total volume of the storage facility (i.e., the volume of the chamber plus the

access tunnel). A curved line has been drawn through the data points for the

WES 1:75-scale model test (Smith, et al, 1989). These small-scale tests were

conducted in a model chamber and access tunnel constructed from steel pipe.

Therefore, this model represents a totally non-responding magazine structure.

The data from the large-scale 1987 KLOTZ test at Alvdalen, Sweden (Vretblad

1988) fall very close to the WES 1:75-scale model curve. The Alvdalen tests

were conducted in rock chambers with sufficient overburden to prevent rupture

and cover venting, and therefore also represent non-responding structures.

9

The remaining data presented in Figure 13 are from tests where the

overburden ruptured (responding magazines), allowing release of the detonation

gas pressures in the storage chamber through the cover venting. The full-

scale Inhabited Building Distances (IBD's) derived from the Shallow

Underground Test, the Brick Model Tests, and the Norwegian model tests

(Jenssen and Krest, 1988) all fall well below the IBD curve for the non-

responding magazine tests, by about a factor of four.

While the IBD's for the responding magazines may at first appear

unrelated, certain trends are indicated. For example, the Brick Model Tests

show an increase in IBD of 25 percent (from 212 to 266 m in full-scale) when

the scaled cover thickness was increased from 0.44 to 0.79 m/kg"/3.

Similarly, increasing the total loading density (mass of explosives divided by

total volume of the underground facility) from 7.1 to 42.9 kg/m3 increased the

IBD by 77 percent (from 150 to 266 m), when the scaled cover depth was held

constant at about 0.8 m/kg'/3.

3.2 DEBRIS EFFECTS

3.2.1 Debris Launch Velocities. Successful measurements of the cover

accelerations for the Brick Model Tests were obtained only on Test 2, which

had the lowest loading density. A partial record was obtained for one of the

two accelerometers (EM-I) installed for Test 1, but the peaks of the

acceleration wave form were clipped due to over-ranging of the gage. The wave

form was clipped at 6,000 g's, and the portion recorded indicated that the

actual peak might have been close to 10,000 g's. No useful records were

obtained from the accelerometers on Test 3.

Assuming that the bricks just below the slope surface above the chamber

were launched in an initial direction normal to the surface, a launch velocity

can be roughly calculated based on the measured throw distance. This

calculation indicates a launch velocity of about

4.9 m/sec for Test 1.

The acceleration wave forms obtained for Gages GM-I and GM-2 on Test 2

(see Figure B.11 and B.13 in Appendix B) showed severe "ringing" of the gages,

possibly due to shock reverberations in the bricks containing the gages. It

10

was possible, however, to integrate these acceleration records to obtain

velocity and displacement histories. As shown in Figure B.12, the gage

located 18.3 cm above the model chamber roof experienced a peak velocity of

only 1.1 m/sec, and a total displacement of about 0.5 mm. Because of the

large scaled cover thickness for this test (0.80 m/kg113), the cover material

at this small distance above the chamber was not moved significantly by the

late-time gage pressures, but experienced only a small displacement as the

initial shock wave passed through it on its way to the surface. Gage GM-2,

located about mid-depth in the cover, experienced a similar peak acceleration.

Because it was nearer to the free surface than Gage GM-I, however, it moved at

a somewhat higher peak velocity (2.0 m/sec) and was displaced a greater

distance. The velocities and displacements of the brick layers increased

dramatically with proximity to the cover surface. The bricks at the top of

the chamber cover spalled completely free, but still with insufficient

velocity to constitute long-range debris.

3.2.2 Debris Distribution. The maximum ranges of debris observed on

the Brick Model Tests were 91 m for Test 1, and 32 m for Test 3. Using

W1/6 scaling, the range for Test I is less than half the maximum range

observed on the large-scale test. On Test 3, the explosive charge was larger

than that of Test 2, but the cover thickness was also greater, resulting in

the same (or nearly so) scaled cover thickness of 0.8 m/kg"/3. As shown in

the photos of Figure 14, the upper layers of brick again spalled away from

their in situ positions, traveling to significantly greater distances than

occurred for Test 2. Table 7 gives the surveyed distribution of debris from

Test 3. All of the debris moved outward from the surface slope over the

tunnel and chamber, within a sector extending about 30 degrees to each side of

the extended tunnel axis (see Figure 15). The vast majority of debris pieces

were fragments less than 1/2 brick in size, indicating that the initial shock

shattered most of the bricks near the surface.

3.2.3 Comparisons with Previous Tests. Figure 16 shows a series of

curves (from Helseth, 1982) relating debris launch velocity to the scaled

cover thickness and the magazine loading density. The data sources range from

aircraft shelters, which had very shallow cover thicknesses and loading

densities, to buried cratering charges, which had very deep covers and very

11

high loading densities. Underground magazines would typically fall between

two extremes.

Dimensional analyses show that the ratio of velocities measured in a

model test to those occurring in a full-scale test is equal to the square root

of the model scale factor. Therefore, the peak velocity measured by Gage GM-2

in Test 2 of the Brick Model Tests was multiplied by 5, and plotted in

Figure 16 along with launch velocities recorded on the large-scale Shallow

Underground Test. While this single data point from the Brick Model Tests

appear to almost exactly match the curves and other data of Figure 16, it must

be remembered that the Gage GM-2 was not at the cover surface, but at mid-

depth in the cover. The actual launch velocity for Test 2 (small though it

was, as evidenced by the short debris travel) was no doubt somewhat greater

than at the gage point.

Also shown in Figure 16 is the launch velocity based on the ballistic

calculation for Test 1 (see Section 3.2.1). The value, which was also

multiplied by a velocity scaling factor of 5 for plotting on Figure 16,

appears to be somewhat low in comparison to the full-scale Shallow UndergroundTest.

Figure 17 shows the debris areal density (number of impacts per square

metre), as a function of range from the tunnel portal, for Brick Model Test 3

compared to that of the large-scale Shallow Underground Test. In accordance

with accepted scaling procedures for ejecta/debris (Rooke, 1980), the

distances in the model case have been scaled up by multiplying by the sixth

root of the ratio of the model-versus-prototype charge weights, i.e.,

(20,000 kg/1.27 kg)"t6 .

As stated earlier in Section 2.5, however, it is not possible to

quantitatively compare the debris densities of the Brick Model Tests with

those of the full-scale test, since the number of fragments produced by the

cover breakup does not scale. Therefore Figure 17 should be regarded only as

a comparison of the relative debris densities recorded on Brick Model Test 3

as a function of range and azimuth, with similar relations from the large-

scale test. To provide such a comparison, the entire grouping of the model

data has been arbitrarily positioned with respect to the ordinate scale of

Figure 17. Considering this limitation, the comparison is actually quite

12

good. the attenuation of the model impact densities with distance closely

matches the shape of the curves from the large-scale data. The relative

differences between the debris densities along the extended tunnel axis

(0-degree azimuth) and the densities "off-axis" also compare quite well with

the large-scale results.

13

SECTION 4

CONCLUSIONS AND RECOMMENDATIONS

4.1 CONCLUSIONS

Peak airblast overpressure and impulse values measured on the Brick

Model Tests at the tunnel exit and in the near-field (just outside the portal)

closely match the results of the corresponding large-scale test. The model

pressure data in the far-field was somewhat lower than measured in the large-

scale test, possibly due to the gravity and inertial effects resulting from

our inability to properly scale the overburden. A comparison among the model

test results shows an increase in pressure of a factor of 4 to 6 when the

chamber loading density was increased from 10 to 60 kg/m3. An overpressure

increase on the order of 90 percent was seen when the scaled overburden

thickness was increased from 0.44 to 0.79 m/kg"/3. The Inhabited Building

Distances indicated by the model tests were significantly less than those

found for the large-scale Shallow Underground Test, but this was also

attributed to overburden scaling deficiencies.

Ejecta impact data collected from Brick Model Test 3 demonstrate the

feasibility of modeling the basic nature of overburden ejecta throwout.

Because the breakup of the cover material does not scale, however, ejecta

sizes in the model tests were much too large to accurately define the ejecta

hazard range, in terms of impacts-per-square meter.

4.2 RECOMMENDATIONS

The results of the tests conducted in this study demonstrate that some

explosion effects can be very accurately simulated in small-sc le models,

while others are biased by properties of the test environment that are not

normally scaled with sufficient accuracy. For example, the peak internal

airblast pressures in the brick model closely matched the full-scale China

Lake Test. The external pressures, however, were lower in the Brick Model

Test than in the full-scale event, apparently due to an earlier venting of the

long-duration chamber pressures in the model. This, in turn, was probably due

to the fact that the mass of the cover material, which resists the venting

force, cannot be properly scaled in a normal "lg" model. For the same reason,

14

it is not possible to accurately simulate the breakup of the cover rock in

such a model, nor the range to which the debris is thrown.

It is strongly recommended that, before further model tests of

responding magazines are conducted, a thorough analysis be made of the

fundamental physics of the explosion/response problem for responding magazines

in order to clearly identify the inter-relationships between the

model/prototype properties and the outcome parameters. Two approaches for

such an analysis are recommended. The most inexpensive would be the

development of a single computer model which idealizes the problem, and

predicts outcomes based on simple physical laws. A more revealing approach

involves the use of numerical computer codes capable of treating a structure

"breakup." Some of the recently-developed versions of the Discrete Element

Code are ideally suited to such a problem. These codes have been used

previously to study the formation and throw velocities of debris from simple

cratering explosions in jointed rock, and have provided significant insights

into the actual influence of factors such as rock strength and density, joint

patterns and shear strengths, and overburden surface slopes.

15

REFERENCES

1. Davis, Landon K.; and McAneny, Colin C.; 1981, "Project SPERRE, Report 5,Summary Report: Road Cratering Tests in Rock With Preemplaced Charges,"Technical Report SL-81-5, U.S Army Engineer Waterways Experiment Station,Vicksburg, MS.

2. Helseth, Einar S.; "Underground Ammunition Storage, Model test in scale1:100 in Sand," Fortifikatorisk Notat NR 160/82, Progress Report to KLOTZ ClubMeeting, Norfolk, VA.

3. Jenssen, Arnfinn; and Krest, Ottar; 1988, "Shallow Underground AmmunitionMagazine Trials at Naval Weapons Center (NWC), China Lake, Ad Hoc Model Tests,

Approximate Scale 1:24.8," FORTIFIKATORISK NOTAT NR 192/88, Norwegian DefenceConstruction Service, Oslo Norway.

4. Joachim, Charles E.; and Smith, Dennis R.; 1988, "WES Underground MagazineModel Test," Twenty Third DOD Explosives Safety Seminar, Atlanta, GA.

5. Joachim, Charles E.; 1990, "Shallow Underground Tunnel/Chamber ExplosionTest Program, Summary Report," Technical Report SL-90-10, U. S. Army EngineerWaterways Experiment Station, Vicksburg, MS.

6. Millington, Charles, 1985, "The Ernestettle Report," Preliminary AWREReport UK(ST)IWP 205, O/1/85, AWRE (Foulness), Ministry of Defence, UnitedKingdom.

7. Rooke, Allen D., Jr., 1980, "Crater-Ejecta Hazard Predictions in CohesiveSoils, The Middle Gust i Event," Thesis, Master of Science, Mississippi State

University, Mississippi State, MS.

8. Smith, Dennis R.; Joachim, Charles E.; Davis, Landon K.; and McMahon,Gordon W.; 1989, "Effects of Explosions in Underground Magazines," Technical

Report SL-89-15, U.S. Army Engineer Waterways Experiment Station, Vicksburg,MS.

9. Stromsoe, E.; 1969, "Underground Explosion Trials at Raufoss 1969,Measurement of Airblast Outside The Tunnel," Intern Rapport X-124, FORSVARETSFORSKNINGSINSTITUTT, Norwegian Defence Research Establishment, Kjeller,Norway.

10. Vretblad, Bengt; 1988, "Data from the 1987 KLOTZ Club Tests In Sweden,Report C3:88, FORTIFIKATIONSFORVALTNINGEN, FORSKNINGSBYRAN, Eskiltuna, Sweden.

16

Area = 84.8 cn?

Figure 1. Access tunnel cross-section for 1 :25-scaleWES Brick Model Tests

Area = 294.4 on?

Figure 2. Storage chamber cross-section for I :25-scaleWES Brick Model Tests.

17

.. .. .. .. . ... .

... ........... .. ... ........ ..... .........o... ....

......... ........ .....

c .. .... .... .... ...

o........ ....... .... ...

_ _ _ _*_ _,* ..... .....

........ ..LOl

cp m .........El

.. ..... E

00

18

U)

-00

F-- $4 $

6-4 4

(0

tI-=4p

)-4

4

4.4-4

Wr.

0 34

004-4

4

r-4WoWW

~0

10 0

-4->

C119

a. Large-scale Shallow Underground Test.

//

s/

b. Brick Model Test 1.

Figure 5. Tunnel portals and foreground areas (for externalexternal airblast measurements) of the large-scaleShallow Underground Test and the correspondingBrick Model Test.

20

Figure 6. 1.27-kg explosive charge before insertioninto chamber of Brick Model Test.

21

350 0 00 10 0

34 200

30 30

35 * 45

- 3 i 30 a

DISTANCE, m

Figure 7. WEB Brick Model Test 3: location of eject. collection areas.

22

102 SHALLOW UNDERGROUND

TUNNEL/CHAMBEREXPLOSION TEST

Internal ExternalGages Gages

0,110EuS

U 1

o WESC, MODEL

A 0 TEST IA 0 TEST 2

I A TEST 3-1

10 - a a * miam1 a a I I *emi a 1 2 1 , 1111

101 1 10 102

SCALED DISTANCE, m/kg "A

Figure 8. Scaled arrival time versus scaled horizontal distance fromthe charge Initiation point. A comparison Is shown betweenthe prototype (Shallow Underground Tunnel/ChamberExplosion Test, loading density 80 kg/m 3) and the Brick ModelTest data. Solid symbols are Internal gages In the model tests.

23

10... KLOTZ (1988)... ...... PLANNED GAGING RANGE

... .L I II.....................0. ......

0

~10x

N MNE ORE1 8 NCEL1:1SSCALE

NDCS MODEL (CONCRETE)* KIOZ (I WUX WE$ MOEL (ONCIQK

10*11 a@fn 1 U II IIIIgo

10' 1 10 10 2 10 3 1

DISTANCE FROM PORTAL, m

Figure 9. Free-field side-on overpressure scaled to theShallow Underground TunneV/ChamberExplosion Test (KLOTZ 1988) parameters.

24

10 5

Pw/ Po = 1.0 (R/D)1'=

10'

0

103 P= 17.7 (Q/VT)a45

10

- K .OTZ (98Mo NCEL(1:15)0 NRE (.MUFOSS

* NORWAY DESB 19M10 0 ( SWEDEN 0 Wn

'r WES MODEL (:75_ U WES MODEL (1:25)

+ NDCS (CONCRETE, 1:24.6)X NDCS (SAND 1:24.e)

I iiii i , i ll l t I I t l

10 10 10

R/D

Figure 10. Pressure-distance comparisons from existing data on model andlarge-scale detonations in underground magazines. The ratio ofcalculated exit pressure (Pu) to measured free-field side-onoverpressure (V, Is plotted versus distance (R) from the tunnelportal along the tunnel/chamber centerline. The distance Isnormalized by dividing by the hydraulic diameter (for turbulentflow) of the access tunnel cross-section.

25

10

LARGE-SCALESHALLOW UNDERGROUNDEXPLOSION TEST (DATA)

1 PREDICTION EQUATION

CL10

CL

10~0

MODEL CHAMBER SCALEDITEST LOADING COVER 1

-2NO. DENSITY DEPTH1 - -qm mk /

0 1 so 0.44 1a 2 10 0.90

I A 3 so607

10 11 1 1 1aa a 1 1 lolmu l a I mI a 111

110 10 2 10 3

DISTANCE, m

Figure 11. Peak side-on overpressure versus horizontal distance fromthe charge Initiation point. A comparison Is shown betweenthe prototype (Shallow Underground Tunnel/ChamberExplosion Test, loading density 60 kg/rn3) and the WES BrickModel Test data.

26

103

I . EXLIONTETIO /_,

10 2

0100

MOLAR EL MBRSCALED--

a.W EXLOIO TES

10-1TEST LOADING COVER -

10NO. DENSITY DEPTHkglm 3 m/kg "

(0 1 0 (441 0 2 10 0. 0

10- 2 , =3l_,,, * il 0*i"91 IIl s * II*

*

2a1 10 10 2 10

DISTANCE, m

Figure 12. Peak side-on Impulse versus horizontal distance from the

charge Initiation point. A comparison Is shown between

the prototype (Shallow Underground Tunnel/Chamber

Explosion Test) and the WES Brick Model Test data.

27

103

Numbers Indicate scaled

cover depth, In m/kg l 3 0.4

NDCS Concrete0.33

E

0.79 2040o o rlk

CD Brick

100.80 0.33

IZ U 1988 Shallow Underground Tunnel/Chamber Testz 0 1987 KLOTZ Club Tests (Alvdalen)

-- Norwegian Model Tests (1:24.8)WES Model Tests (1:75)WES Brick Model Tests (1:25)

10 1 1 1 a ,,,,I i 1 1 1 ,m 1 1 1 a m al11

10", 1 10 102

TOTAL LOADING DENSITY (Q / VT), kg/rn

Figure 13. Alrblast Inhabited Building Distance versus total loadingdensity (charge mass divided by total Internal volume) forselected model and large-scale tests. Solid symbols arefor "responding" magazines, and open symbols for"non-respondlng" magazines.

28

.. fft

g 4 4

Ne,

~dd

Figure 14. Photos of debris fields from Brick

Model Test 3.

29

50

0O > HALF A BRICK

o0 0 HALF A BRICK

a: 30

L 20LL0

IL 10C0

0 a 19

DISTANCE FROM PORTAL = 6.1 m

-10 I I a I I I a I a I , I a I a I , I

-50 -40 -30 -20 -10 0 10 20 30 40 50

AZIMUTH, degrees

Figure 15. Ejecta distribution as a function of azimuth alongthe 6.1 m arc, WES Brick Model Test 3.

30

120Legend

0 Shallow UndergroundTunnel/Chamber

p ii Explosion Test' I 20, 000kg100 Iil ElFOULD

3 2500,000 kg100 Wg I A PRE-GONDOLA

1 20, 000 kg3 kg/rn3 0 ESSEXI

10,000kg* 0 Small Concre

-4 0 BunkerE 50kg/m 3 I Arraft Sheltere

WES Brick Model

- 2 kgm~~ WES Brick Model Ejects

60 - ,

>0.54 kg/rn

Z , 1,500kg/n 3

540 0.I gM310 kg/rnm (UET untarnped)

+

20(tamped)

0 - a I - I I a al

10 10" 1 10

SCALED COVER DEPTH, m/kg I

Figure 16. Launch velocity of cover rock ejecta from WES brick modeltest, compared to ejecta velocities from the ShallowUnderground Tunnel/Chamber Explosion Test (Joachim,1990) and other sources (Helseth, 1982) on previousexplosive tests.

31

104

3 010 0 A I U T O

'& LARG E-SCALE TEST

10 20 00

E10 20 0

01 40 0

*1 AZIMUTHS FOR10 MODEL TEST 3

o 0Odegrees IA +/-10 degrees

10*+/-20 degrees+9 +/-30 degrees

10 10 210 3104

DISTANCE FROM PORTAL, m

Figure 17. Relative comparison c' '--bris densities from WES Brick ModelTest 3 with the Shallov. Jnderground Tunnel/Chamber ExplosionTest (KL.OTZ 1988) data curves. Distances In the modei werescaled by the ratio of the charge weights taken to the one-sixthpower.

32

Table 1. WES Brick Model Tests: Measurement Type and GageLocation Data.

Station Horizontal* MeasurementNo. Distance Elevation" Type

(cm) (cm)

AB-1 20 0 Side-on Overpressure

AB-2 20 0 Stagnation Pressure

AB-3 40 0 Side-on Overpressure

AB-4 100 0 Side-on Overpressure

AB-5 300 0 Side-on Overpressure

AB-6 600 0 Side-on Overpressure

AB-7 -10 0 Side-on Overpressure

AB-8 -50 0 Side-on Overpressure

AB-9 -124 0 Chamber Pressure

AB-10 -148 0 Chamber Pressure

GM-1 -115 52.6 Overburden Acceleration

GM-2 -125 34.3 Overburden Acceleration

* Horizontal distance along tunnel/chamber centerline

measured from the tunnel portal.

Elevation measured from the tunnel/chwer floor and/orsurface of concrete airblast pad.

Table 2. WES Brick Model Tests: Explosives Charges, and Chamber CoverThicknesses

Explosive Charge Minimum Scaled

Test Total Loading* Chamber PortalNo. Mass Density No. of** Length Diameter Cover Cover

(ks) (kg/m3) Strands(kg) (cm) (cm) (m/kg

1 /3) (m/kgs

/3)

1 1.27 60 31 48 5 0.44 0.034

2 0.21 10 5 48 0.80 0.062

3 1.27 60 31 48 5 0.79 0.49

* Mass of explosives divided by chamber volume.

** 0.085 kg/m (400 grains/foot) detonation cord

33

Table 3. WES Brick Model Tests: Airblast Pressure and Ground Shock

Measurements; Test 1 (60 kg/m3 PETN charge loading density,

0.44 m/kg /3 scaled chamber cover depth).

Peak Data

Station Horizontal* Arrival Measured FirstNo. Distance Elevation" Time Integral

(cm) (cm) (msec)

AB-l 20 0 0.77 5.37 MPa 2.0 kPa-sec

AB-22 20 2.5 0.67 1.83 MPa 1.1 kPa-sec

AB-3 40 0 0.83 1.92 MPa 0.10 kPa-sec

AB-4 100 0 ---- ---- ----

AB-5 300 0 5.6 0.036 MPa 28. Pa-sec

AB-6 600 0 13.5 0.010 MPa 13. Pa-sec

AB-7 -10 0 ---- ---- ----

AB-8 -50 0 0.17 12.4 MPa

AB-9 -124 8 ---- ---- ----

AB-IO -148 8 0.2 >750 MPa

GM-I -115 52.6 0.24

GM-2 -125 34.3 ---- ----

* Horizontal distance along tunnel/chamber centerline measured from the

tunnel portal. Negative distances are inside tunnel or above chambercover.

** Elevations measured from the tunnel/chamber floor.

Total pressure measurement

34

Table 4. WES Brick Model Tests: Airblast Pressure and Ground Shock

Measurements; Test 2 (10 kg/m3 PETN charge loading density,

0.80 m/kg1/3 scaled chamber cover depth).

Peak Data

Station Horizontal* Arrival Measured FirstNo. Distance Elevation*" Time Integral

(cm) (cm) (msec)

AB-1 20 0 0.88 >5.2 MPa >0.57 kPa-sec

AB-2! 20 2.5 1.35 1.13 MPa 1.1 kPa-sec

AB-3 40 0 1.09 0.45 MPa 78. Pa-sec

AB-4 100 0 2.33 0.125 MPa 64. Pa-sec

AB-5 300 0 7.04 16. kPa 19. Pa-sec

AB-6 600 0 15.3 5.2 kPa 6.0 Pa-sec

AB-7 -10 0 0.64 1.89 MPa 1.9 kPa-sec

AB-8 -50 0 0.37 2.0 Mpa 1.0 kPa-sec

AB-9 -124 8 ---- 114. MPa 3.4 kPa-sec

AB-10 -148 8 ---- 52. MPa 4.0 kPa-sec

GM-I -115 52.6 0.25 1500 g's 1.08 m/sec

GM-2 -125 34.3 0.06 1500 g's 2.0 m/sec

* Horizontal distance along tunnel/chamber centerline measured from thetunnel portal.

** Elevations measured from the tunnel/chamber floor.

Total pressure measurement

35

Table 5. WES Brick Model Tests: Airblast Pressure and Ground Shock

Measurements; Test 3 (60 kg/m3 PETN charge loading density,

0.79 m/kg /3 scaled chamber cover depth).

Peak Data

Station Horizontal* Arrival Measured FirstNo. Distance Elevation* Time Integral

(cm) (cm) (msec)

AB-I 20 0 0.45 4.8 MPa 0.75 kPa-sec

AB-2! 20 2.5 0.77 2.8 MPa

AB-3 40 0 0.60 2.2 MPa 0.16 kPa-sec

AB-4 100 0 1.30 0.78 MPa 0.10 kPa-sec

AB-5 300 0 5.0 0.049 MPa 63. Pa-sec

AB-6 600 0 12.7 0.015 MPa 20. Pa-sec

AB-7 -10 0 0.33 6.3 MPa

AB-8 -50 0 0.16 29. MPa ----

AB-9 -124 8 ---- ----

AB-10 -148 8 ---- ----

GM-I -115 52.6 0.16

GM-2 -125 34.3 ---- ----

* Horizontal distance along tunnel/chamber centerline measured from the

tunnel portal.

** Elevations measured from the tunnel/chamber floor.

Total pressure measurement

36

1-4co inL 0 0 0

LO 0 LA LA) 041C1 CN, C1 C14 0

noL

In 0

.. J Aj C,

"4 LA 4) L

0w

0

bo~

41' -A" bo(n .0F ,z 0 00 0 0

LA 0 - t .0 LA V0O .

004

0 en~ CY 0%aU 0 -b -14 c~r 0

0 0 4 0 0 0 0 C.cn U A

bo

z~ 0

0 E ~ 41U U

0) 0 w~5- 0 w- 0

I44

%.0 .- 4) 0o 0

0 4 LA 41 LA LA b)-00 Q C'J eq. eq. C14 r- -

*00

-4 w4

4.4) 0 0) a)

10 U 10 w- 00 c 0 0.ca 0 . C

CA 0- W U-

37

Table 7. WES Brick Model Test 3: Brick Ejecta

Distributions.

Number of Brick Eject&Horizontal* Size of BrickDistance Azimuth**

W. (degrees) Ful1lilalf cHalf

3.1 0 4 134

4.6 0 3 62

6.1 0 8 50

9.1 0 1 9

12.2 0 2 4

18.3 0 0 4

24.4 0 0 1

3.1 10 0 13

4.6 10 3 24

6.1 10 6 24

9.1 10 0 5

12.2 10 1 16

18.3 10 0 2

24.4 10 1 0

3.1 350 2 84

4.6 350 3 60

6.1 350 1 16

9.1 350 3 2

12.2 350 0 2

18.3 350 0 1

24.4 350 0 0

6.1 2r, 1 14

6.1 30 2 4

6.1 45 0 0

6.1 340 4 20

6.1 330 34

6.1 315 0 0

*Horizontal distance along tunnel/chamberCenterline Measured from the tunnel portal.

**Azimuth referenced to extended tunnel/chambercenterline.

38

APPENDICES A, B AND C

AIRBLAST PRESSURE AND GROUND MOTION WAVE FORMS

Gage Identification Code:

Digital Gage MUST O0-X CBS 0 0000 0.0000Array Size MNFl Low Pass 0000 HZCal Val 00.0Deflection 0000

00 KHZ DD-MM-YY

The first line contains identification code (00-X). AB and GM identify

airblast and ground motion (acceleration) measurements, respectively. In

addition, some time histories require baseline shifting indicated by the CBS

(constant baseline shift) shown on the far right of the first line. The

numbers following the CBS indicate shift starting and ending times (msec) and

the amount that the data was shifted. Negative numbers indicate a downward

shifting of the time history. The second line gives the size of the data

sample. The third line lists the low pass filter cutoff frequency. The

fourth and fifth lines gives the values of the shunt calibration and the

corresponding value of the voltage deflection, respectively. The last line

shows the sampling frequency of the filtered data and the date on which the

plot was generated.

39

APPENDIX A

AIRBLAST AND GROUND MOTION WAVE FORMS

Brick Model Test 1

Explosives Charge: 1.27 kg PETN

Chamber Loading Density: 60 kg/m3

Minimum Scaled Chamber Cover Depth: 0.44 m/kg1 /3

40

0

00

51SX'VW ~~fldrIIU

0 0

01 0 m 0 I~ 0 If

0

Iw 44

V (4-

-o 0

0(x0

1-

-4 N o

01 -w '-4

u rN

o 00 (n -

*0 4.-4

o Nw L-

(0 0 o t ~-> '-

4/1-

0

SXVdW 361fldWI

.4~ c40 0 0o o 0 0 0 0 0o 0 0 0 0 0 00 0 0 0 0 0

0

CY)

coi

0

Co IC

0 0u

0 CI

m 4

0

cuCw1 w

I .

w 410 a 0M 0 mO

x x -- -4

CU)

4)

0 0

0~0

cu C 0ON 0

41N>0 0 ILO 0 U) 0 LO )-4 a C -40 L0 4 q- cu - .- 0 0 0 --4 L. 4J' US 0

VdW x

42

00o

0

SXVdN asifldWI

0 cu 0 02 0clu 00

ou a 0 0 0 0o a 0 0 0

0 0 0 0 0 0o8 0 0 c

1004)

cu

w .

m 00

w (D

ca 02--- -

0 4)

ClD1--

I 0o Cuwd

U~ -44

00

*00

04

En2

00

Ml C 0WN >co 0.4 0 Cm 0 0 C)- - C-

-1 '-1 0.02 -C CD Cu 0u

43

SX~dW 3slfldHI

o 0 0 a a 0o o 0 0 0 0 0 co 0 0 0 0 0 00 0 00 ;4

0

cu 0-41

4)

$4

00

40 01 $4

w CD

44-

4- wco 0

* I------------------------------------------------o00

0 00

o Wto

0 - $4w0

In~~" P --

0 L0

*~ --4

0 0m

'T 4

U2 VO

L0 0 0-C4c

0 0 0 0 C

o - t ---- C

o I44

SX~dN 3SifldWI

U,tn0 ID) 0 ID 0 0

ciJ CU . 40 0 0

o 0 0 0 0 0 0 4o0 0 0 0 0 ;o

00- 4I)

CTQ)

4))W

cu>(nn

-4

1J 0u 0* 0

L-C.3

:3 0 4

0u 414

I-w 0

0

0 Ln 0Su ww

0 i6 0

03Nco 41

40 c u cCa 0.4000

L. *u LI L) ato a

~L4-0 0 0 0 0 0 0o

VdH

45

SXVdW Dslfldwi

o 00 to I- cu 0 0-4 0 0 0 0 0C 0o o 0 0 0 0

o o 0 0 Q o 0

1

in

c o 0

m 0

ww LL

C)0 Cu

02 - -

Vo x

02 LO N

0 0)

x - - -4-4S 450

cu 00

CUCE__r 4-40

a V 0

0 V0

10 C

14 ,.44(D 45 41 05S 00 0 L) 0 L ) Lr 0 L0 clCu CI.j 4-

- 4L 0 0-0

Q2 -Ci fn 0 VdW

46-

SXVdW 3Slfldwicv 1-4

o c co IV cu 0 0*1 0 0 0 0 0

o 0 0 0 0 0 0

.4

0

(00

Cu3

U)- - .U)

V au)a

1- *1-LL 4)

a w co

o o w

wc p

-4 - .

$4 '-4

E-0

-1 1) -) W

o C)0 0

10 2

ON 000

0 0 0 0-

= C - 0

Co L- 4 0 0- 0 o a 0r 0

bD.4-q

VdW i

47

00

4~J

0- w

Clu

L..

:3 o9 r4o C

-c 4400

o w

00

00

0 __Z L N U

C) y e 1)

I-D ii::

0 0 0 0) 0 00)

~~~~ aC.) 00

_ _ 1- -1 - *49:3 C.3C3 0NT Cj N0

_ - - - - .If, .r4C

S-0 N~iVH-1 33V )

484

APPENDIX B

AIRBLAST AND GROUND MOTION WAVE FORMS

Brick Model Test 2

Explosives Charge: 0.21 kg PETN

Chamber Loading Density: 10 kg/m 3

Minimum Scaled Chamber Cover Depth: 0.80 m/kg1/3

49

SX~dH 3sifldNI

C"io v cu 0 0 4-40 0 0 0 0 0 Co a 0 0 0 0 C

co e

CDa)

Co "4ID I Aa a)

0 0._P co 44

'00

0

* -4

CDC

00C

M' w

40

C,

00-

to 0 *04

0 ao

0 0

v -4

In -4c ocu)jo 0 o

Ow dw

~N 050

SX~dW 3sifldI

to C 0-4 0 0 0 a 0

o 0 0 0 0 0 4o 0 0 0 a 0

0

cu

LO 0

0 t0

00

0

L. N Ii

-4 -) -4

w-4

0

(D -1 0-

'--4 44

0- 00 m W C

00

CCu

CD M

-4-4O0tn0Ln0Lo-

-o 0 uc

0d

(~51

SX~dW 3SifldWI

o w V c 0 0.4 0 0 a 0 0 0 .

o 0 0 0 0 0do 0 0 0 0 0o 0 0 0 0 C

0

43

-4

0

'4)

0

4)

0 44

'o 0

1 0I- CM -i <

10

£ cc

4110

cu 00-4

a)..

0) o L 0

'46~9 E4C 05.

040 a 00

w T - 1d0o 40 52

SXVdW 3sifldWI

14 a 0DC 0 0.4 0 0 0 0 0 0

o a 0 0 oP

cu

-. 0

40 0 -1

a0 0

01 , IV-L 0

0

Aj IU3 44'03--

* -

cu *

-4 (YO

1.-00

-1- - - - 0014c

.4) 0-4E- CL-

00rED =-.'. 4 -1

-

004-W) 0U wn01 U )

N.u 0 0

It I I I idN

a a 0LO 0 Lo 0 U) 0 U-53

SX~dW 3sifldWI

tD 0 U, 0 U) 0

o a a 0 0o 0 0 0 00 0 -o a a a 0 0

00

S

-. 41

0

00Lo

w -<

V 0 CL40 c

m z-W 41*r w

CD 00-EH C6

64010

o CT"

4a~~~~~- 0 u c 4 -

ma. 4 4

WN 0W

.4 54

SXVdH 3S~fldWI

o m to cui 0 0.4 0 0 0 0 0 0o a 0 0 00 0 0 a 0 0 0 co 0 0 0 0 0 0

o 0 0 0 0 50 0

- - - -- - - -1cu U)

co 4-)

cu 4)

(.0

0

- mmm U

* 0

13 0 -" -- 0

CD

0 b=

.0 00C

10 Cu

v N C4-

.0 q CV)

0I c

CDN 00v4 I@ go - -- -

3%> a-- 0 0 0 0 0 00 0 0 0 0 0 00; 0 0 0 0 0 0 i

VdW

55

SXVdW 3sifldWI

In a In 0 In 0Mu cu -40 0 0wo 0 0 0 0 0 0 0

o 0 0 0 0 0

-A41)

cu

Q)

1-4

(.0

.0 0 Lo

. 0 Lfln

0 U)

*0 00

0

0 L NF- bl=

U)CuC14-

cu 0

0 '-4

0

ci(a-4

0 lC0

ON 0

6 0U3 0 Lo 0 O

CouL.u-'-64 cu.cu0 0 0-. L 00 bo

VdW

56

SX~dW 3SifldHI

totn 0 In 0 n 0 01cu cu .4 4 0 0 0o 0 0 0 0 0 0 0o0 0 0 0 0 -

0 ; 0) 10

E-4

(1) 1.4tn

Q)

CU

1 0I

CO)

-

0 w 0cf o a

*- - - - - -- - - - - - - - - --4

L- 4).0 bo

* 0 4J

0 0

o w 4

0) -- -U -

00cy -

000t L 9 N

to 0

Cu -v -- - -

-44

C3~4. -c AV

57U

ILD IqC) Ri 11 0 :0 0 0 0 0 0 0o 0 0 0 00

0 :10o 0 0 0 0 0 10 c

0 0

_ Y)4)- -4

0

C\J laI_ C') 4

0-4

_o 0

1 00

0*l I ja

0 0

I Cl) I C

00

N cw

F--4

CC

641000

a)000

0

0

0

C3 N 00Ca 1 .- 4 a 0c

.- 060 0 04 0 0010 CL -4 L 0 LO 0 U) LO)

L.', a' -4 44 I~ ~L 410 0 bto

VdW 4

58

SXVdW Bs-fldwiCd

ED) 'IT C) cm - 0 0o 0 0 0 0 0 0o o 0 0 0 0

0 0 0 0 0 0 1 0

00

c 0

0

-,4

0 Z

*O cu C

0)

*. N ~.- o ow

02~~b - -100

0 I . N O W Ca 0

34E- 0

00

ow0 go. SI

00

CY 0

0N 00

0 coo 0 0 0 0 0 0 00 0-4 0 CD 0 CUj CUmCo.44 -

.4 L 4hU-~C ou -,4

VdW Z

59

0

0 0

o U),-4

0

'-4

0o

C-)-4

0 c cd

0

00

a w

00

-4 w1- 0 4. 0

'

ca co O

u0

00

x~ 00-4

Cuu

o r00

cu

co 0

C3 4 0 0~~4~ 00.4'0

0 o 0 0 0 0 ).4 >.4 0 U) 11)

- NOI1VU3-1903V

60

W - dSia Q

0 0 0 0 0 0a 0 0 0 0 0 0a' 0 00 04

o3 a oa

- -

-41

- --- - -4

(00 4t

C.-)

-- __ ~w 0

x L I.

CM0

0 N

(D 0

0 0

-'- UC-4

LO2 Cu W.~00 - .

0

mmC4fmN 00

" -1 g

.4 -

-4 CJ Cu -4 h- 00

03SIH - AIIOO-1A

61

41)

oc

0

cu

cu 0)

cu0

4.3 0

0 w

Z(0a ID - -LO

0 - W )A_ CD >

0_u C 4

x 0

cu-C b__ Co 4-4

4'O 0

CD En O

z0

00

*u a

col 0 0 cw 41 -4 4

10" .Ui o o 0 0 0 0 00.3 . L " a 0 0 0 0 o

tsL 4- - 0 0 0 0 0 0Q C0. tjA

*-4

S-E0 - NOIJNHD300OV

62

W - ldSI G

LO lqt (Y) cu -~ 0 0 4o 0 0 0 0 0 0 ca

0 0 0 0 0 10 0

cu4-)

cu0r4

- - __ cu >

(00

Zoo

0 Z~

c 0O

a3 Cu I- *

Oh 4-JO

F-4c

-40~ 4) --

0 "

z- 0.0

o) Cr cu0 0

SiEuc c

0-N 0 0CD )L. -4 c- cu 0

OAS/ -J -i 4'lA

~ ina in0 U) 0 63

APPENDIX C

AIRBLAST AND GROUND MOTION WAVE FORMS

Brick Model Test 3

Explosives Charge: 1.27 kg PETN

Chamber Loading Density: 60 kg/m 3

Minimum Scaled Chamber Cover Depth: 0.79 m/kg 13

64

SXVdW 3slfldNI

a to to C 0 0o4 a 0 0 0 0 0 C0 0 0 0 0 0 09 0 1-4o 0 0 0 0 o 0

"-4

1-4

cn 0

4j 0 C~jV

1 00 1. 10 I

a 0 q wCD 0 --

C) C C-) w0 w '-4

0 CU 0

L- w

0 tko

-4 ..-41.- 0 41 oc

Oj Q)

03)

o 0Q1-0

-3:3 14

0-

aU c 0co N 0 1 11 1 1 1- 1

'402-4 -4

0 A u a VdW L

65

SX~dW 3sifldW'I

in2 a In 0 Ln 0 0cl1 cUi -4 -40 00o 0 0 0 0 0

o o 0 0 0 0o

"4J

ti to-4)

_ 0.

41(3 0Cu :

I. I

* 0 C

10

cu -30

02 . N Cu*c C-. x..

-4 41

0 41 0 0_U~

m 0-

00 -41

C33 --4 0

00Cu.N

WN 0

a USO ) cu 048 0 bo=I-~ 0 --4 1.. 41 3

0 .4 w u M66 d

0a0

0 L

0 I 0 10 0 a0 0

o 0 0 0 0 0o a 0 0 0 0 0

* 0o 0 41 0 0

CU n

03

U l C

II0 (0

100

4. 0 -

&- w

*3 0

*4 0 .- .

U3 3 10 cc 0

1% 01 4:

4 $M 00 E_ _

3 -4 bU) cu

C .,

LO 03 LO- O La X- 4 In

- .s9, C.

.4d

02 *'4 L67

LO

00a0

SX~dN 3SifldWI

0 n 0 M1 0 In2 a0

o 0 0 0 0 0 0 co 0 0 0 0 0o 0 0 0 0

.4 0

-- 4

w

100

r44

-4- Q.

0

00

4C, 0* C)

0 00 _

L- w .4ca b

a 00x X .- -1 1

4 W =

CuEU, -r - -0

0 4

cuC

4.44

(0 t0

0 -C 3 A

~~4-4

0068

SXVdW asifldWI

o o cu o o10 0 0 0 0oa 0 0 0o 0 0 0 0 0 0

* . 0

co'-4

00

ca I-- $ 1

0 w 04

0 1 -

0 cuL.

0 0 I

0 3 LON

' 10

M 0~ 0 .xI x I- U0

)0

0 0

01m020

.4

00a3c

0ON

14 4 -

ca 4...>4 0 0 0 0

co - 4 4 -a L0 0- 0 0 0C3 0 0000

VdW

69

SX~dW 3SifldWI

Mn0I 0 ILn 0 0W ~ cui -4- 0 0 01

o 0 0 0 0 ao 0 0 0 0 0a

0 0 0 0 0 0

o 0 0 0 0 0o

- - -S a - - - - - 0 4.)

-4

cu 0-o Cu

00

0*0 (

10 0 cu CL) WL l w wa 14

1 _ 0

L. 5Lk- G

V F- CU

r-4

4U)W

-4 -

4) 0

0.r

Goli

004

CU

toN 0 LIL

CU CU -4 -4 0 0 00 0 0 0 0 0 0 b

0 00 0 0 0 0

VdW

70

SX~dW 3Sfldwif

LOLo 0 Lo L n C)cu cu ~4 . 0 0 0o 0 0 0 a 00

o 0 0 0 0 0 iO0 40

cu 4J

-Ir

co)

UU)0

w>0

100

'-4

00

oJ 0 w

O 0)

0*0 0

Ac0

abri

0 V

-4 LO Nl OW

o0 --4b

00

-4-4

0 480

Z14 -4 Ln~__00

00

04-__ C

00N 0 0If 1 -4 4

0 0 ( 4, C ~ 0

071

SXVdW Ds-fldwi

LI) Iq ) C 4 0 0 4o 0 0 0 0 0 0 Co 0 0 0 0 0

0 2 i0 0 0 0 0 0 1 0 0

m00

to In4n

0

* 00

I _ _ U)

L ow 0

0 0 0

0 0

L4 w

* F'-4

-4I.

00

co S0_ 00E

c 0

0 0

co 41

In-,40 -- - __ - - - - - - - - - - -

8 72