rna consulting, incorporated report 1.pdf · 2020-02-17 · 5. ross pelz v. worthington industries....

RNA Consulting, Incorporated Specializing in forensic materials engineering and sciences

27820 Saddle Court Los Altos Hills. CA 94022·]810 USA

June 25,2008

Mark D. Epstein

Robert N. Anderson,. Ph.d.,. P.E., President

Alborg, Veiluva & Epstein LLP 200 Pringle Avenue, Suite 410 Walnut Creek, CA 94596-7380

Re: Shalaby v. Irwin, et al.

Dear Mr. Epstein:

Bus: 650·<)49-1092 Fax: 650.9495641

email: [email protected]

At your request, I have examined the 4/21106 incident where Mr. Andrew Shalaby was injured while using a 16 oz cylinder Bernzomatic MAPP gas torch. MAPP gas is a trademark of the Dow Chemical Co. and is composed of extremely flammable methylacetylene-propadiene-propane.

It is my understanding that Mr. Shalaby was in the process of igniting logs in a fire pit, using a TS4000 torch head and Bernzomatic MG9 MAPP gas cylinder when the cylinder vented and he was burned. In the incident, the center valve housing, attached to the canister by brazing material, ruptured.

Materials Reviewed: I have reviewed the following documents:

I. Deposition of Michael Ridley, Senior engineering manager, Irwin Industrial Tool Co., taken 11113/07.

2. Deposition of Steven T. Gentry, Quality Control Department Worthington Cylinder Corp., 11114/07.

3. Deposition of Andrew W. Shalaby volume I and II, Plaintiff, 10/24/07 and 10/25/07.

4. Deposition of Warren L. Ratliff, Jr., Park ranger supervisor, Campland, 4/17/07. 5. Deposition of Randy T. Stephens, Ranger at Campland, 4117/07. 6. Deposition of Joe Russo, Paramedic, 4/18/07. 7. Miscellaneous manufacturing drawings of torch parts. 8. Health & Safety Laboratory report 2006/121. 9. Bernzomatic instruction manuals 96001, 97090 10. Bernzomatic catalog. 11. Investigative Report with reference to interview of Anne Carrol and David Borger

by Howard Felder 10/6/07. 12. Consumer Product Safety Commission Release # 78-088. 13. Transcribed statement of Andrew Shalaby by Joe Tancredy, 6/1/06 14. Material Safety Data Sheet for MAPP Gas. 15. Supplemental Response to Request for Production of Documents (Set No. One). 16. Worthington Industries Expert Witness Disclosures.

Epstein_ShalabLRNAreport.doc - 1 -

17. Third-Party Defendant/Cross Claimant Western Industries, List of Expert Witnesses.

18. Disclosure of Experts by Defendants, Bernzomatic. 19. Protective Order. 20. Defendants' Initial Disclosures. 21. Drawing 304432. 22. MAPP gas cylinder specifications.

Analysis: The failure of the Bernzomatic MAPP gas torch Mr. Shalaby was using appears to be at the collar or threaded area between the center valve housing and the cylinder. The center valve housing is fastened to the cylinder by a copper-nickel brazing material. The Rangers in the park that examined the gas torch confirm that the fuilure was in that location.

• Park Ranger Supervisor Warren Ratliff, in his deposition, comments were that there "appeared to be a crack in the cylinder at the bottom thread level of the cylinder" page 25, lines 21-25; page 26, lines 24-25.

• Ranger Randy Stephens comments on the f~lure in his deposition on page 42, linesl5-25; page 43, lines 1-11; page 73, lines 9-25; page 741ines 1-15.

• Also, the recollection of Andrew Shalaby in his transcribed statement to Joe Tancredy on 611106.

Mr. Shalaby was using the torch to ignite firewood in a fire pit and his torch would have been partially inverted in that situation. Health & Safety Laboratory report 2006/121 (report included in test results CD) determined that the torch orientation was important and confirmed that when the cylinder was inverted, explosion could occur. The directions do indicate "Use upright to prevent flare-ups or flashes" caused by the liquid entering the torch. However, this orientation is impossible in some situations.

CPSC Release # 78-088 issued a notice of a recall for fuel cylinders from another manufacture (Cleanweld Products) that separated "where the threaded connector meets the cylinder". A flaw in this area is very serious.

A review of other MAPP gas torch failures involving lawsuits fIled since January 2002 and supplied by the Defendant in their Supplemental response to request for production of documents (Set One) had listed 7 lawsuits identified below.

1. Thomas Segrest, Jr. v. Bernzomatic. (Date of injury 2/9/04) 2. Richard Gleen v. Newell Operating Co. (Date of injury 113/06) 3. Andrew Gelzer v. Thermadyne (Date of injury 2/13/04) 4. Melvin Wilfredo Bonilla Carranza v. Bernzomatic. (Date of injury 6/13/05) 5. Ross Pelz v. Worthington Industries. (Date ofinjury 5/29/05) 6. Mark Loewes v. Worthington Industries. (Date of injury 3/27/05) 7. Timothy Welch v. Newell Rubbermaid (Date of injury 7/3/06)

The Glenn v. Newall is a Ventura California case in which a Bernzomatic cylinder failed at the braze material (Photographs 1,2).

In addition, I have reviewed photographs for a Minnesota case called Venderlinde v. Ace Hardware Corp., where a ''TurboTorch'' failed in the braze material between the center

Epstei"_ ShalabL RNAreport.doc -2-

valve housing and the cylinder. There are two Lenox cylinders (John Barrett v. Lenox and Lemaralejo) that also failed in the same location.

Tests Conducted: l. Metallography of the brazing material in MAPP gas cylinders. 2. Microhardness of the brazing material in MAPP gas cylinders. 3. Energy Dispersive Spectrum (EDS) ofMAPP brazing material. 4. Metallography of corrosion test of the MAPP brazing material. 5. Metallography of brazing material inMAPP PRO gas cylinder. 6. Microhardnes ofthe brazing material in MAPP PRO gas cylinders. 7. EDS of brazing material in MAPP PRO gas cylinders.

Test data are contained in a compact disc (CD) included with this report.

The braze material is a copper nickel alloy. The composition was determined to be approximately 61 % (atomic) Cu and 39 % Ni by Energy Dispersive X ray. This is 63% Cu by weight and the specifications on Drawing 32600-23 call for a maximum of 60% by weight Cu. It is possible that the brazing material is off specifications.

Steven Gentry in his deposition (page 89, line 7) states that the brazing temperatnre is between 2000 and 2100 degrees Fahrenheit, which is 1093 -1149 degrees centigrade. According to the Cu-Ni phase diagram (See Figure 1) from the reference "Hansen, Constitution of Binary Alloys", that temperature is too low to melt the brazing alloy.

It should be noted that the brazing material used in the Bemzomatic MAPP PRO cylinders has been changed to all copper without the addition of nickel. Metallography of the MAPP PRO brazing material is shown on the CD.

Conclusions/Findings:

Based on the facts of failure in the brazed area, I have examined three exemplar MAPP gas cylinders, (WIOG57E, WIIG152W and W8G230E), with respect to the Cu-Ni braze between the center valve housing and the cylinder. Microhardness testing of the brazing metal gave values of23 HRC for WIOG57E; 33 HRC for WllG152W, and 97 HRB for W8G230E.

The cylinders have been sectioned in half and four sections have been cut from each cylinder to show a portion of the neck piece and cylinder wall and the brazing material between. These sections have been mounted in plastic and polished and photographically documented. Representative examples of the microphotographs from each cylinder are shown in Photographs 3, 4, and 5. The brazing materials have large voids in the bulk and smaller voids in the interface between the cylinder walls and the center valve housing as shown in Microphotographs 6, 7, and 8. The brazing on the outer surface of the cylinder is undercut in all three cylinders rather than forming a meniscus. The undercutting is a sign oflack of wetting and penetration of the brazing material with the cylinder and valve. A good meniscus shows that wetting has occurred. The brazing defects shown in photographs 3-8 reduce the strength of the joint and make it more likely that the valve will partially separate from the cylinder and release gas when the torch is used as intended.

Epstein_ShalabLRNAreport.doc -3-

The outside surface of the brazing material that is undercut represents a lack of wetting and penetration of the brazing material with the cylinder and valve housing. 1bis flaw is sufficient to reject the cylinder. This flaw should have been picked up by the manufacturer with a simple visual inspection of the cylinders.

The Bernzomatic one pound MAPP gas cylinder is manufactured per Federal specification published in 49 CFR 178.65 (D.O.T. 39). The specification for non-reusable (nonrefillable) cylinders states "brazed seams must be assembled with proper fit to ensure complete penetration of the razing material throughout the brazed joint." The brazed joints shown in microphotographs 6, 7, and 8 lack complete penetration.

Corrosion tests show that the brazing material is strongly cathodic to the cylinder and valve and will cause the steel to corrode in a suitable moist atmosphere. The interior walls, of the sectioned cylinders, also showed signs of corrosion. See Photograph 9, Interior View of Cylinder WlOG57E Showing Corrosion.

In my opinion, the braze material between the center valve housing and the cylinder is the weak element in the assembly, and subject to failure when the torch is attached to the cylinder. The brazing material has voids and lacks sufficient fusion to the cylinder wall and valve housing to resist stresses placed on it when used in a normal manner. This problem with the brazing material is due to a combination of poor cleaning of the brazing area, contamination of the brazing material and improper process parameters such as furnace temperature and time. For these three cylinders that were examined to be offered on the market clearly establishes the failure of Bernzomatic inspection and quality control procedures.

The MAPP gas torch and cylinder is unsafe and unreasonably dangerous as designed and manufactured.

Please call me if you have any questions.

Sincerely,

~;ij:~ Robert N. Anderson, Ph.D., P.E.


Photo 1: Overall view of Glenn v. NewaU;.eern~Qm~t'c Torch.

Epstein_ShalabLRNAreport.doc - 5-

Photo 2: Close up of Glenn v. Newall c<ylinder.aHhevalve housing, showing a brazing material failure. .

Epstein_ShalabLRNAreportdoc - 6-

Figure 1: Copper - Nickel Phase DiagrCim from Hansen.

10

.. 140 0

0

110

,..-/ /" ---o~ __

10830

1000

0

10 0

0

-10 0

-20 0

-27 3 o c.

10

20

---------

'flOG"T PEt! 'CENT tH£I<El 30 40 SO 60 70 90 , , , , I

145)0

~ -~ -f--~

~ v_ -- ----

------

---------------

I 1

(CV~ Nil .... ,/

/ /

/ ,

/ .. MAGN. TRft.NSF.

/

/ ,/

/ ,

// 30 40 50 6G 70 80 90 100

H;' ATOMIC PER CENT HltKEl

Fig. ;~47. Cn-Ni


!

Photo 3: W10G57E. @ 13X. Valve housing on the top and cylinder on the bottom with brazing material inbetwe.en.

Epstein_Shalaby-RNAreport.doc -8-

(a) 50X

(b) 100X

Figure 3 Representative micrographs of the braze at section 1 (Figure 2b).

Photo 5: W11G152W. @ 13X. Valve hous,ing on the top and cylinder on the bottom with brazing material inbetween.

Epstein_ShalabLRNAreport.doc - 10-

Photo 6: W11G152W. @ 100X.

Epstein_ShalabLRNAreport.doc - 11 -

Photo 7: W8G20E. @ 13X. Valve housing on top. and cylinder on the bottom with brazing material .

Epstein_ShalabLRNAreporl.doc -12 -

Photo 8: W8G20E. @ 200X.

Epstei"_ShalabLRNAreport.doc - 13-

Photo #9: Interior View of Cylinder W110Cli51ES,hOlwin1Q

Epstein _ Shalaby-RNAreport.doc - 14-

(a)

(b)

Figure 1 Photographs of the MAPP gas cylinder W10G57E in (a) the as-received condition, and (b) prior to sectioning for metallography.

(a)

(b) Box in 2(a) 13X

Figure 2 Section 1 images of (a) overall mounted cross section, and (b) braze cross section.

Outside Surface of Cylinder

Inside Surface of Cylinder

(a) 200X

(b) 500X

Figure 4 Representative micrographs of the braze at of section 1 (Figure 2b).

(a)

(b) Box in 5(a) 16X




(a) 50X

(b) 100X


(a) 200X

(b) 500X


(a)

(b) Box in 8(a) 13X




(a) 50X

(b) 100X


(a) 200X

(b) 500X


(a)

(b) Box in 11(a) 13X




(a) 50X

(b) 100X


(a) 200X

(b) 500X


EDS OF BRAZING MATERIAL IN

MAPP GAS CYLINDERS

·'

. SEMQuant results. Listed at I I :30:50 AM on 6i12/200S Operator: Ryan Client: none Job: MAPP Gas Cylinders Spectrum label: Mapp Gas

System resolution = 135 eV

Quantitative method: ZAF ( 3 iterations). Analysed all elements and nOffilalised results.

1 peak possibly omitted: 0.04 keY

Standards: P K GaP 29/11193 Fe K Fe 01112/93 Ni K Ni OU12/93 Cu K Cu 01112/93

Elmt Spec!. Element Atomic Type % %

P K ED 4.27 7.93 Fe K ED 29.31 30.19 Ni K ED 23.77 23.28 Cu K ED 42.65 3S.60 Total 100.00 100.00

* =<2 Sigma

SEMQuant results. Listed at !l :31 :56 AM on 6il2!2008 Operator: Ryan Client: none Job: MAPP Gas Cylinders Spectrum label: Mapp Gas

System resolution = 135 eV

Quantitative method: ZAF ( 2 iterations). Anal ysed all elements and normalised results.

3 peaks possibly omitted: 0.04, 2.00, 6.40 keV

Standards: Ni K Ni 01112/93 Cu K Cu 01112/93

Elmt Spect. Element Atomic Type % %

Ni K ED 36.68 38.53 eu K ED 63.32 61.47 Total 100.00 100.00

• =<2 Sigma

HEALTH & SAFETY LABORATORY REPORT

HSLl2006/121

Harpur Hill, Buxton Derbyshire, SK17 9JN T: +44 (0)1298 218000 F: +44 (0)1298 218590 W: www.hsl.gov.uk © Crown copyright (2006)

The Behaviour of 'Bernzomatic' MAPP and

Propane Cartridges When Exposed to Heat and Flame

HSL/2006/121

J E Fletcher Project Leader:

Author(s): J E Fletcher

Science Group: Hazard Reduction

CONTENTS

1 INTRODUCTION......................................................................................... 1

2 BACKGROUND INFORMATION................................................................ 2

3 EXPERIMENTAL ........................................................................................ 5 3.1 Test Program........................................................................................... 5 3.2 Test Cartridges ........................................................................................ 5 3.3 Test series A, Non-uniform heating, diffuse flame ................................... 6 3.4 Test series B, Uniform heating in a Waterbath ........................................ 7 3.5 Test series C, Non-uniform intense heating, pre-mixed flame ................. 8 3.6 Test series D, Single cylinders in the open.............................................. 8 3.7 Test series E, Multiple cylinders, pool fire................................................ 8

4 RESULTS ................................................................................................. 10 4.1 Test series A, Non-uniform heating, diffuse flame ................................. 10 4.2 Test series B, Uniform heating in a waterbath ....................................... 11 4.3 Test series C, Non-uniform intense heating, pre-mixed flame ............... 12 4.4 Test series D, Single cylinders in the open............................................ 12 4.5 Test series E, Multiple cylinders, pool fire.............................................. 15

5 DISCUSSION............................................................................................ 17

6 CONCLUSIONS........................................................................................ 19

7 FURTHER WORK..................................................................................... 20

8 REFERENCES.......................................................................................... 21

i

EXECUTIVE SUMMARY A number of concerns were raised by HSE following an application to store large quantities of MAPP gas in 450 g cartridges. The fire risk and mitigation control measures appear based on the assumption that cartridges of MAPP behaved in a similar way to propane and comparable to aerosols. It has been reported that MAPP can spontaneously detonate when heated. HSE therefore commissioned a programme of work, comprising a series of experiments on both single and multiple cartridges of MAPP and propane to establish the hazards.

Objectives a) Determine if MAPP detonates on heating,

b) Develop an understanding of the failure temperature/pressure of cartridges under uniform and non-uniform heating.

c) Direct comparison of fireball diameters between LPG and MAPP cartridges,

d) Projection distances for ejected cartridges, and

e) Knowledge as to whether spontaneous failure of multiple cartridges is induced under fire engulfment.

Main Findings Based on the evidence of the tests undertaken, it was concluded that:

a) MAPP gas contained within a ‘Bernzomatic’ cartridge does not detonate upon heating under the conditions of the tests carried out here.

b) When heated uniformly, cartridges vent at 80±2°C.

c) Similar MAPP and propane gas cartridges behave in the same way.

d) Intensely heated cartridges present an explosion hazard by a pressure burst failure.

e) The likelihood of explosion depends upon cartridge orientation.

f) Projection distances for ejected cartridges and fragments were a maximum of 13.9m and 26.4m respectively.

g) No fireballs were seen, but flame heights of 1-2 m are achievable from venting cartridges.

h) There was no evidence of induced or spontaneous failure of multiple cartridges.

Recommendations

There is the potential to extend the scope of this work to include:

• Multiple cartridge tests with increased confinement to ensure cartridges stay in close proximity to each other

• Scale up to several transport packs

• Cartridge systems other than Bernzomatic.

• Aerosols containing flammable propellants.

ii

1 INTRODUCTION

A number of concerns were raised by HIDCI2G following an application for a warehouse occupier to store 48 tonnes of propane, including 9.5 tonnes of methyl acetylene-propadiene-propane mixture (MAPP gas) in 450 g cartridges. The company’s fire risk and mitigation control measures appear based on the assumption that the behaviour of cartridges of these gases is repeatable and comparable to aerosols containing propane. However, it has been reported (see Section 2) that Methyl Acetylene can spontaneously detonate when heated. HSE therefore has an urgent need to address this issue to establish the hazards of MAPP gas in storage.

Large-scale test data on the fire behaviour of such cartridges are not available. Consequently, HIDCI2G requested that HSL undertake a series of experiments on both single and multiple cartridges containing both gases to identify the hazards. The main objectives/deliverables of this work were to:

a) Determine if MAPP detonates on heating.

b) Develop an understanding of the failure temperature/pressure of cartridges under uniform and non-uniform heating.

c) Provide a comparison of fireball diameters with propane and MAPP cartridges.

d) Measure projection distances for ejected cartridges.

e) Determine if spontaneous failure of multiple cartridges can occur under fire engulfment.

f) Take a video record of all tests.

1

2 BACKGROUND INFORMATION

MAPP gas is a complex mixture, comprising mainly methyl acetylene (35 mol%), propadiene (27 mol%) and propane (18 mol%) with other trace stabilising compounds. Taken individually these first two components are considered flammable and reactive. The most important properties of these constituents are parameterised in Table 1.

Methyl acetylene (also known as propyne) is a flammable, colourless liquefied gas with a boiling point of –23.1°C and a freezing point of -101.5°C. It is self-reactive and can decompose explosively at 4.5 to 5.6 atmospheres pressure. It has the chemical structure CH3C≡CH [1].

Propadiene is also flammable with a critical temperature and pressure of 120°C and 54 bar. It is an isomer of propyne but is not acetylenic, containing instead, a carbon atom carrying two double bonds. Like propyne, it decomposes explosively when a hot-spot is created inside it. It has the chemical structure CH2=C=CH2 [2].

Table 1: Properties of MAPP gas constituents [3,4].

Constituent Gas Property Methyl Acetylene Propadiene Propane Molecular weight 40 40 44 Boiling Point (°C) -23.2 -34.5 to -32.0 -40 Melting Point (°C) -102.7 -146 to –136 -190 Flammable limits (vol%) 1.7 to 11.7 1.7 – 12.0 2.1 to 9.5 Critical Temperature (°C) 127.8 120 96.8 Critical pressure (kPa) 5349 4417 4255 Vapour pressure (kPa) 517 (at 20°C) 873 (at 21.1°C) 853 (at 21.1°C) Specific gravity (water = 1) 0.70 (at -50°C) 1.787 0.5853 (at -45°C) Vapour density (air = 1) 1.4 1.4 1.55

Mixtures of methyl acetylene, propadiene and propane are produced as a C3 fraction during steam cracking treatments of heavier hydrocarbons at oil refineries. Other trace stabilising compounds are then added to produce the commercially available MAPP.

Unlike acetylene, commercial MAPP is not distributed in pressurised cylinders, dissolved in acetone in a porous mass, but as complex mixtures stored in the liquid state in gas cylinders. Components added to stabilise the propyne-propadiene mixtures are typically C3 and C4 alkanes and alkenes. By changing the relative proportions of each component, a wide range of mixtures is produced. The levels of stability of these mixtures vary - a diagram given by Medard [2] and shown in Figure 1 explores this and gives a limit line for stability.

2

Figure 1: Stability diagram for MAPP, according to Medard.

An example MAPP gas composition given by Stull [5], is shown in Table 2, and includes some thermodynamic data. One of the principle uses for MAPP gas is for blowtorches or for welding, when used in conjunction with oxygen, as a safer alternative to acetylene. The flame temperature data highlight this, as MAPP provides a (lower) flame temperature close to that of acetylene, yet is significantly hotter than that of propane.

Table 2: Decomposition of MAPP gas.

Mole% Gas Heat decomp. J.g-1

Flame Temp. K

Peak pressure kPa.m-1

35.4 Methyl acetylene -4628 1816 1236 27.4 Propadiene -4791 1864 1266 0.7 1,3 – butadiene -2213 1086 1043 1.5 Cyclopropane -1958 936 831 3.3 Propylene -1628 866 689 1.9 Iso-butane -1331 800 760

18.0 Propane -795 628 456 11.8 Iso-butane -762 623 567 100 MAPP -2259 1034 932 100 Ethylene -2259 1005 1074

Evidence for the detonation of methyl acetylene is given by Bretherick [6] who states that the liquid material in cylinders is not shock sensitive, but a wall temperature of 95°C accompanied by pressures of about 3.5 bar, will cause a detonation to propagate from a hot-spot. Induced decomposition of the endothermic hydrocarbon leads to flame propagation in the absence of air above minimum pressures of 3.4 and 2.1 bar at 20 and 120°C, respectively.

3

Both Stull and Bretherick present the opinion that although pure methyl acetylene is highly endothermic (∆Hf = +185.4 kJ.mol-1) the diluted MAPP gas can be handled safely by treating it in the same way as ethylene.

Previous work has been undertaken at HSL on cartridges of MAPP gas. In 1984, flame impingement tests were conducted on several gas brazing and welding kits [7]. Of relevance here, one of the kits, under the ‘Bernzomatic’ trade name, comprised steel cylinders of oxygen and fuel. The cylinders were fitted with a pressure relief device and contained either 400g of propane or 450g of MAPP gas. Photographs in the report show that these cylinders are very similar to the cartridges tested in this work (see Section 3.2 for a description).

In these tests the flame from a torch (powered by MAPP/oxygen) impinged on the steel cylinder under test, held lightly upright. Tests were performed on single full MAPP cylinders with flame impingement below liquid level and on single almost empty cylinders, so that impingement was above liquid level.

A range of effects was observed including gas escaping from the pressure relief device and small jets of flame/gas released from the sidewall at the point of impingement. During testing, the cylinders bulged 4-16mm at the point of observed impingement. No cylinders were seen to fail.

4

3 EXPERIMENTAL

3.1 TEST PROGRAM

All tests involved heating either 400 g propane cartridges or 450 g MAPP cartridges with a propane flame until they burst. Tests performed were:

a) Non-uniform heating of single cylinders in a protective bunker with a diffuse flame, to establish if MAPP could be made to detonate on heating.

b) Uniform heating of single cylinders in a water bath to measure cylinder failure temperatures.

c) Repeat of a) using a pre-mixed flame from several burners to give more intense heating.

d) Heating single cylinders in the open air to study the projection distance of cylinders and fragments and fireball diameter/lift-off.

e) Heating multiple cylinders: 1 box (12 cartridges).

Multiple cylinder tests were intended to provide information on fireball diameter; projection distances and whether failure of one cylinder induced failure in neighbouring cylinders. All tests were repeated for both gas types to allow a qualitative comparison. Tests were recorded to videotape for subsequent review.

3.2 TEST CARTRIDGES

Cartridges for test were sourced from two suppliers. Two boxes of each gas were supplied by the company concerned, whilst a further two boxes of each were obtained from an independent supplier. Each box contained 12 cartridges of gas. Cartridges from both sources carried the ‘Bernzomatic’ trade name/branding. Cartridges of both gases appeared identical in construction but were readily identifiable/separable by their colour: yellow for MAPP and blue for propane. The cartridges were of pressed steel construction, in two parts with a central circumferential crimp or weld, and complied with the U.S. DOT 39 standard for non-refillable gas cylinders. The upper section of the cartridge contained the main filling or ‘end use’ valve and a pressure relief valve (PRV). A white plastic cap covered the main valve. Some measurements of the cylinders are given in Table 3 and example photographs of the cartridges are shown in Figure 2. Cartridges were labelled in compliance with DOT-39 with the following information:

• Bernzomatic Propane

PW 16 PH 25 bar 0.4 kg 1.0 L .7 mm EN12205 0035/2005/11 UN1978 Propane –20°C + 50 °C DOT-39 NRC228/286 M1003 USA Chiltern SN W11E22E

• Bernzomatic MAPP PW 13 PH 22 bar 0.45 kg 1.0 L .7 mm EN12205 0035/2005/11 UN1060 Methyl acetylene and propadiene mixture, stabilised –20°C + 50 °C DOT-39 NRC188/235 M1003-E6686 USA Chiltern SN W11E55E

5

Where: PW is the normal service pressure, PH is pressure at which the PRV operates, 0.4/0.45 kg is the nominal weight of the contents and L is the wall thickness.

EN provides cylinder type/batch/manufacture date information.

UN is the United Nations designation for the article.

DOT-39 NRC gives the normal working pressure and test pressure (in psi). M is the manufacturers registration number.

The final line gives additional manufacturer information.

Table 3: Measurements of test cartridges.

Measurement MAPP Propane Height to shoulder 250 mm Height to top 275 mm Diameter 75 mm Gross weight 900 g 850 g Mass of gas 450 g 400 g

Figure 2: Photographs of the test cartridges.

3.3 TEST SERIES A, NON-UNIFORM HEATING, DIFFUSE FLAME

In this test series, single cartridges were placed horizontally on a stand, unclamped, and heated by the impingement of a diffuse flame from the open end of a propane hose. Figure 3 shows a schematic diagram of the test arrangement while Figure 4 shows a photograph of the apparatus. Tests were conducted in an enclosed bunker and the gas flow to the heating flame controlled remotely via a solenoid valve, with a small tray of solvent providing the ignition source (for convenience, this was later changed to a gas pilot flame). Cartridges containing both gas types were tested.

6

Gas Cartridge

Open ended propane hose

Flame arrestor

10 m

10 kg propane cylinder

with flame arrestor

1 m

Small tray (5 cm diameter) of solvent as pilot flame

Sandwich of corrugated iron sheets and plywood fastened to stakes

Solenoid valve

Figure 3: Schematic diagram of test apparatus for non-uniform heating - Test Series A.

Figure 4: Photographs of the set-up for Test Series A.

3.4 TEST SERIES B, UNIFORM HEATING IN A WATERBATH

In this test series, cartridges were heated using a water bath. The water bath comprised a steel tank with dimensions 425mm width x 740mm length x 295mm height, fitted with four electric heating elements with a power rating of 2.2 kW. The bath was filled with water to a depth of 200mm (approximately 60 litres).

Test cartridges were submerged under the water and held in a horizontal position by steel wire wrapped loosely around a retort stand. The water and cartridge temperatures were monitored using 1.5 mm diameter, stainless steel sheathed type K thermocouples. The latter was fixed to the cartridge body, positioned against the central seam, using steel wire. Temperature data were recorded to computer via a Microlink data logging system at one second intervals.

7

3.5 TEST SERIES C, NON-UNIFORM INTENSE HEATING, PRE-MIXED FLAME

The test apparatus used was as in Test Series A. However, for these tests, a twin burner arrangement was connected to the previously open-ended hose. This allowed for pre-mixing of the propane fuel with air producing a significantly hotter, bluer and more intense flame.

The burners were simply placed on the floor underneath the metal stand, approximately symmetric either side of the test cartridge and directed upwards, to impinge on to the central seam of the cartridge. Again, cartridges containing both gases were tested.

3.6 TEST SERIES D, SINGLE CYLINDERS IN THE OPEN

The burner system and stand used for Test Series B were modified to incorporate three burners instead of two, with the aim of giving a greater coverage of flame around the circumference of the test cartridges. Test cartridges were placed on the stand in either upright or inverted orientations, supported loosely by a retort ring and then heated rapidly by the burners, see Figure 5. In total, 12 cartridges of both gas types were tested with 6 in the upright and 6 in the inverted positions.

Figure 5: Burner assembly for Test Series D.

3.7 TEST SERIES E, MULTIPLE CYLINDERS, POOL FIRE

Complete boxes of cartridges were subjected to a pool fire for a sustained time period and observations recorded.

The test boxes were standard supply/transport cardboard cartons, as received from the supplier, containing 12 cartridges of gas in a 2 x 6, upright arrangement, see Figure 2. These boxes were

8

placed on a stand in the centre of a steel tray with a base 1 m square and a wall height of approximately 300 mm. The tray was partially filled with 50 litres of n-heptane, to a depth of 50 mm and then ignited, remotely, using a hot wire ignition source.

The resulting pool fire lasted for approximately 10-20 minutes.

One box of MAPP cartridges was subjected to the first trial, E1, and one box of propane cartridges was used in the second trial. Both tests were recorded for subsequent review.

9

4 RESULTS

4.1 TEST SERIES A, NON-UNIFORM HEATING, DIFFUSE FLAME

Two MAPP cartridges and one propane cartridge were tested in this manner. The same behaviour was observed for all three cartridges:

After a short period of time, venting of gas was heard. The heating flame ignited this release and a large jet of flame was observed issuing from the top of the cartridge. Due to the relative orientation of the cartridge and video camera it was not possible to distinguish from which of the cartridge orifices the gas was being released. The venting/jetting continued until all gas in the cartridge had been consumed. No cartridges were seen to fail. It was noted that the flame from MAPP appeared hotter, more intense and yellower in colour compared to that from the propane flame, which was orange in appearance.

Due to a technical fault, no video footage of Test 1 was recorded. Masses of recovered cartridges were recorded. Some results are summarised in Table 4, while Figure 6 shows the recovered cartridges.

Table 4: Results of Test Series A, non-uniform heating, diffuse flame.

Test Gas Time to jetting, Duration of jetting

Mass After (g)

Observations

A1 MAPP N/a 437.0 Cartridge remained intact and complete. Not bulged, contracted or cracked. Entire surface covered in black deposits.

A2 MAPP 2 mins, 3mins 438.7 As A1 A3 Propane 2 mins, 4 mins 436.5 As A1

Figure 6: Photograph of cartridges post Test Series A.

10

4.2 TEST SERIES B, UNIFORM HEATING IN A WATERBATH

Three tests were performed in total. The first test was discounted because an electrical fault caused a malfunction in three of the heating elements part way through the test, but some useful data were still obtained. In all three tests, gas was observed venting from the PRVs, which formed a fog/mist on the water surface. Venting was observed between 78 to 82°C and continue up to the point the water began to boil. Actual heating rates were calculated from the collected temperature data, see Table 5. The recovered cartridges were intact, but some of the labels were peeled back. Cartridges appeared to be slightly bulged below the central join. Vernier calliper measurements gave the diameter of each cartridge tested as 76 mm below the join and 75 mm above.

Table 5: Results of Test Series C, uniform heating in a waterbath.

Test Gas type Temperature for start of venting

(°C)

Overall heating rate (K.min-1)

B1 MAPP 79-81 1.3 B2 MAPP 82 1.8 B3 Propane 78-80 1.6

Figure 7: Photographs of cartridges post Test Series B.

11

4.3 TEST SERIES C, NON-UNIFORM INTENSE HEATING, PRE-MIXED FLAME

A total of five cartridges were tested, three MAPP and two propane. The observations recorded are given in Table 6. In summary, cartridges behaved in a similar manner to those in the previous Test Series A, with jetting from the top of the cylinder. In this instance, it was possible to identify from where the gas was venting and that the PRV had failed.

In these trials one cartridge failed catastrophically, with an audible bang, bursting into several large pieces. Three pieces were retrieved, one comprising the top section of the cartridge with the valves. Photographs of the test cartridges are shown in Figure 8.

Table 6: Results of Test Series C, non-uniform intense heating, premixed flame.

Test Gas Observations Time to Venting

(s)

Time to Explosion

(s) C1 MAPP Gas released from cartridge PRV after a short period

of heating. Visible as ignited jet. Subsequently, with further heating the cartridge failed.

28 56

C2 Propane Gas released from cartridge PRV after a short period of heating. Visible as ignited jet.

35 N/a

C3 MAPP As C2. 39 N/a C4 Propane As C2. 29 N/a C5 MAPP As C2. 34 N/a

Figure 8: Photographs of cartridges post test series C.

4.4 TEST SERIES D, SINGLE CYLINDERS IN THE OPEN

Following on from Test Series C, a further 24 cartridges were tested by exposure to an intense flame. The effect of cartridge orientation and retaining the white plastic cap or otherwise was examined. Initially, all cartridges behaved in a similar manner. As in previous tests, after a short period of heating, gas was released from the test cartridge PRV. This was visible as an ignited

12

jet. Subsequently, with further heating, some cartridges were seen to fail with an audible bang and ejection of fragments, while other cartridges were either ejected from the test burner assembly or continued venting until the gas was exhausted. Details are given in Table 7, including the time to start of venting (or failure/explosion) and cartridge masses before and after testing. Additional data for tests where cartridges either exploded or were ejected are given in Table 8; this includes fragment projection distances.

Table 7: Results of Test Series D, intense heating of single cylinders in the open.

Test Gas Orientation Cap Observations Time to venting

(s)

Time to explosion or

ejection (s)

Mass before

(g)

Mass after (g)

D1 MAPP Upright Off Venting N/a - 909.3 439.9 D2 MAPP Upright Off Venting N/a - 900.7 439.2 D3 MAPP Upright Off Venting 21 - 901.6 438.6 D4 MAPP Upright On Venting 21 - 890.3 437.8 D5 MAPP Upright On Venting 19 - 912.6 437.5 D6 MAPP Upright On Venting 25 - 919.8 439.9 D7 MAPP Inverted Off Vent then

explosion 29 98 914.8 -

D8 MAPP Inverted Off Venting 28 - 902.5 436.8 D9 MAPP Inverted Off Vent then

ejection 18 40 901.0 442.4

D10 MAPP Inverted On Vent then ejection

21 33 912.8 441.9

D11 MAPP Inverted On Venting 19 - 885.6 440.3 D12 MAPP Inverted On Vent then

explosion 20 57 906.2 -

D13 Propane Inverted Off Venting 24 - 881.2 455.3 D14 Propane Inverted Off Venting then

explosion 22 54 849.8 -

D15 Propane Inverted Off Venting 21 - 845.1 434.2 D16 Propane Inverted On Venting then

ejection 16 31 867.7 457.3

D17 Propane Inverted On Venting then explosion

20 70 868.0 -

D18 Propane Inverted On Venting 21 - 860.5 436.5 D19 Propane Upright Off Venting 13 - 878.0 451.4 D20 Propane Upright Off Venting 17 - 877.2 451.8 D21 Propane Upright Off Venting 16 - 869.1 451.4 D22 Propane Upright On Venting 15 - 863.6 447.4 D23 Propane Upright On Venting 12 - 863.2 446.6 D24 Propane Upright On Venting 16 - 871.3 440.7

13

Table 8: Additional data for failed or ejected cartridges from Test Series D.

Test Effect Fragment mass (g)

Distance (m)

Sound Level (dBa, LcpKmax)

D7 Exploded, into 3 pieces 382.9 17.7

7.7

5.2 3.8 1.7

155.9

D9 Ejected 7.5 D10 Ejected 3.8 D12 Exploded, into 3 pieces 344.0

47.6 48.1

1.0 26.4 19.7

144.7

D14 Exploded, 2 pieces recovered, valve assembly missing

222.3 74.6

13.0 2.5

153.4

D16 Ejected 13.9 D17 Exploded, into 1 piece 398.0

1.0 153.4

Figure 9: Photographs of failed cartridges from Test Series D.

14

4.5 TEST SERIES E, MULTIPLE CYLINDERS, POOL FIRE

A timeline of observed events is given for the MAPP cartridge trial in Table 9 and for the propane cartridge in Table 10, with photographs of the retrieved cartridges in Figure 10 and Figure 11, respectively.

In the MAPP trial, a total of three cartridges failed explosively with one cartridge ejected from the fire. Retrieved cartridges were weighed and all those intact were found to be empty/spent. One cylinder and one large fragment were retrieved from outside the tray at distances of 3.7m and 2.4m respectively.

In the propane trial, a total of four cartridges failed explosively. No cartridges were ejected from the fire.

The average heptane fuel consumption rate was calculated as 3.5 ± 0.2 mm.min-1.

Table 9: Timeline of events for Test E1, MAPP pool fire.

Time (Min:Second)

Observations

00:00 Fire started. 00:00 – 00:37 Cardboard box is consumed by fire and contents fall into metal tray. 00:55 First venting of gas heard. 01:00 onwards Venting witnessed from cartridges as a jet of flame varying in height from 1 to 2 m. 01:33 Audible report from cartridge failure. 01:34 Audible report from cartridge failure. 08:45 – 09:02 More jetting increasing in intensity, culminating in explosive failure of a cartridge. At

the same time a single cartridge or large fragment is ejected from the fire. 11:53 More jetting, decreasing in intensity 13:10 Pool fire extinguished

Table 10: Timeline of events for Test E2, propane pool fire.

Time (Min:Second)

Observations

00:00 Fire started. 00:30 – 0:39 Cardboard box is consumed by fire and contents fall into metal tray. 00:58 First jetting heard and seen. 01:43 Audible report from cartridge failure. 04:01 Audible report from cartridge failure. 11:39 Audible report from cartridge failure. 12:37 Audible report from cartridge failure. 15:00 Pool fire extinguished.

15

Figure 10: Photograph of retrieved cartridges from Test E1.

Figure 11: Photograph of retrieved cartridges from Test E2.

16

5 DISCUSSION

Test Series A and B provided an initial opportunity to assess the hazards posed by these cylinders when heated or exposed to flame. All cylinders behaved in a similar manner; i.e. venting after a short heating period. All cartridges remained intact, no detonation of contents was observed and it was considered viable to proceed with the subsequent intense heating test series. In early tests (Series A and B) the only observed difference between propane and MAPP was the colour of the jetting flame: orange for propane and (more) yellow for MAPP.

It was not possible to see the origin of venting in Test Series A and B, but in Test Series C it was evident that venting was occurring from the pressure relief valve, indicating either a PRV failure or deliberate operation/release due to elevated internal pressure. The first apparent difference in the behaviour of MAPP and propane was observed in this test series with one of the three MAPP cylinders tested failing (exploding), while neither of the two propane cartridges failed. From a review of the video footage it was noted that the failed cartridge had a lower PRV than all the other cartridges, possibly below the liquid level in the cartridge. This may have caused a cartridge failure due to a ‘liquid locked’ PRV or a PRV unable to control the rate of internal pressure rise due to passing liquid rather than gas. With the cartridges positioned horizontally, the PRV from the failed cartridge was at ‘2 o’clock’ compared to ‘12-1 o’clock’ for the other cartridges. In view of this and to give a wider understanding, Test Series D was conducted with more cartridges, orientated vertically, both upright and inverted.

The results from Test Series D confirmed a dependence of cartridge orientation on probability of cartridge failure but cartridges of propane and MAPP behaved in a similar manner. In these trials, two cartridges of each gas failed when in an inverted position, a failure rate of 33%, yet no cartridges failed when in an upright position.

Besides the failed cartridges, a number of cartridges were ejected from the test apparatus – two MAPP and one propane. These cartridges were self-propelled by venting gas and travelled distances of up to 13.9m. A good example of this is Test D16 where video footage shows the venting cartridge being propelled randomly around the test area, before leaving the view.

Failed cartridges comprised a single large section with two or three ejected fragments. Generally, the large section comprised both the base and top shoulder linked by the main body that was burst or split, typically of mass 200 to 400g. The ejected fragments typically weighed around 50g (measured range 7.7 to 74.6g) and were retrieved at distances up to 26.4m. The edges of the retrieved fragments were smooth indicating a pressure burst rather than a detonation. If the contents had detonated then it would be expected that many small fragments would be found with torn or jagged edges.

No fireballs were observed during testing. However, flame heights from venting cartridges were estimated at between 1 and 2 m in length.

Following on from Test Series D, the multiple cylinder pool fire trials (Series E) provided an opportunity to examine the behaviour of cartridges in a more realistic fire scenario1. As in Test Series D and previous tests, cartridges were observed to vent and several were seen to fail. Three MAPP and four propane cartridges failed, a failure probability of 25% and 33% respectively. Cartridges and fragments were again ejected and were retrieved at distances up to 3.7m from the fire. There was no mass explosion, and no evidence of induced or spontaneous cartridge failure, with cartridges venting and/or failing linearly in sequence, although in the E1 1 While Test Series E gives a reasonable simulation of an engulfing fire, the intense heating from Series D can be likened to the scenario where the flame from a jetting cartridges impinges on to another neighbouring cartridge. 17

MAPP test there were two failure events in close succession (1s apart). In these tests, the cardboard box burned away early on in the test and cartridges fell away from each other into the fire pool. It is likely therefore that the cartridges would not be in close enough proximity to each other for an induced failure to occur.

18

6 CONCLUSIONS

Based on the evidence of the tests undertaken, it was concluded that:

a) MAPP gas contained within a ‘Bernzomatic’ cartridge does not detonate on heating, under the conditions of tests performed here.

b) When heated uniformly, cartridges vent at 80±2°C.

c) Similar MAPP and propane gas cartridges behave in the same way.

d) Intensely heated cartridges present an explosion hazard by a pressure burst failure.

e) The likelihood of explosion depends upon cartridge orientation.

f) Projection distances for ejected cartridges and fragments were a maximum of 13.9m and 26.4m respectively.

g) No fireballs were witnessed, but flame heights of 1-2 m are achievable from venting cartridges.

h) There was no evidence of induced or spontaneous failure of multiple cartridges.

19

7 FURTHER WORK

There is the potential to extend the scope of the tests carried out here:

• Multiple cartridge tests with increased confinement to ensure cartridges stay in close proximity to each other.

• Scale up to several transport packs.

• Cartridge systems other than Bernzomatic, including aerosols containing flammable propellants.

Further work could seek to expand upon the multiple cartridge tests performed here. Video footage showed the carton containing the cartridges disintegrating and the cartridges falling away from each other into the fire. Additional tests could be designed to prevent this happening, holding cartridges with increased confinement in close proximity to each other. Such a test would be better able to determine if a failed cartridge can induce failure in neighbouring cartridges. One method of increasing confinement would be to scale-up to include several transport packs, perhaps even to a pallet-sized scale.

This work has focused solely on the ‘Bernzomatic’ cartridge system. It could be extended to other similar systems, perhaps containing other gases, either fuel or oxygen. For example, aerosols are similar in design, incorporating a fuel gas (propane) used as a propellant confined in a thin walled steel canister. A comparative study between these and the Bernzomatic cartridges would confirm the validity of assuming these behave similarly for risk assessment purposes.

20

8 REFERENCES

1 T F Rutledge, “Acetylenic Compounds.” Reinhold, NY, (1968).

2 L A. Medard, “Accidental Explosions: Volume 2: types of explosive substances.” Ellis Horwood Ltd., Chichester, UK (1989), pages 698-700.

3 Material Safety Datasheets, Matheson Tri Gas. http://www.matheson-trigas.com

4 R C Weast, “CRC Handbook of Chemistry and Physics.” CRC Press Inc., Cleveland, Ohio, 58th edition, (1977).

5 D R Stull, “Fundamentals of Fire and Explosion.” Dow Chemical Company, (1976), pages 14-16.

6 L Bretherick; “Bretherick’s Handbook of Reactive Chemical Hazards.” Butterworth’s, London, UK, 4th Edition, page 354.

7 D K Pritchard and A M Nicol, “ Flame Impingement Tests on Bernzomatic, Clean welder and Solidox Gas Brazing and Welding Kits.” HSL Internal Report IR/L/FL/84/16 (1984).

21

http://www.matheson-trigas.com/

,

Name:

Address:

Education:

VITA: Robert N. Anderson, Ph.D., P.E., P.J. President, RNA Consulting, Inc.

Robert Neil Anderson, Ph.D., PE, P.1. President, RNA Consulting, Inc. Specializing in forensic materials engineering and sciences Consultant, expert witness in materials failures, accident analysis

27820 Saddle Court, los Altos Hills, CA 94022-1810 Office: 650-949-1092 Fax: 650.949.5641 CeIlNM: 650-464-1620

Ph.D., Metallurgy, Stanford University M.S., Chemical Engineering; Minor: Petroleum Engineering, U.C. Berkeley B.S., Chemical Engineering, U.C. Berkeley B.S., Chemistry, University of San FranciSCO Post Doctoral Research Associate in Metallurgy: Stanford University.

Patents in high temperature reprocessing of nuclear fuels, nuclear reactor, and extraction of reactive metals from their ores.

Licenses: Registered Metallurgical Engineer, California No. 1490 Registered Nuclear Engineer, California No. 91 Private Investigators license, CA No. 14033

Consultant and Expert WItness: More than 20 years experience in forensic materials and metallurgical consulting in accident analyses, industrial materials applications, including asbestos, thermodynamics, and design failures.

Consultant to legallirms, including plaintiff and defendant, and court appointed expert. Extensive deposition and courtroom experience in U.S. and Canada.

Expert in electrical lire source determination, using Auger arc residue analysis.

Published in professional journals.

Designed and led workshops for attorneys, and other experts in the presentation of evidence and findings.

led teams and presented professional papers on unique archee-materials investigations. Leader of 1995 of scientific team that performed research on the cause of the 1913 fire that destroyed Jack London's famed Wolfhouse. "

Expert in firearms and ballistics.

Experienced in multiple engineering disciplines and in the potential impacts of materials, such as metals, composites, polymers, and ceramics.

Past Academic Positions: SJSU, Emeritus Professor, Department of Materials Engineering.

FOlmer depariment chair, full professor. Stanford University, Associate Professor.

Past Industrial Positions: U.S. Naval Radiological Defense laboratory, San Francisco

Operations Research Analyst, Research Engineer Arabian American Oil Company, Dhahran, Saudi Arabia, Chemist

Prior Consulting, Industrial: FMC, accident simulation. EPRI, nuclear reprocessing. SRI, materials research. IBM, intellectual property. Memorex, hard disk technology.

Prior Consulting, Government: Brookhaven National lab California State Energy, Resources & Development Commission Executive Office of the President of the United States,

Council on Environmental Quality; Office of Science & Technology Policy

United States Congress, Office of Technology Assessment Lawrence Livermore Lab Smithsonian Institute, Washington, D.C. Suffolk County, New York, consulting in nuclear plant failure Center for Scientific and Industrial Research, Government of India,

Consultant in the surface degradation of Taj Mahal. At request of the government of India, investigated, advised on procedures to slow the surface degradation.

Professional Associations: Offices: American Academy of Forensic Sciences (AAFS): Fellow. Director.

Chairman (2004-2005; 1996-97) of the Engineering Sciences Section. Awarded 2005 Founders Award by the ESS Section.

National Academy of Forensic Engineers (NAFE): Fellow. Past President 1998. Intemationallnstitute of Forensic Engineering Sciences: Diplomate and Director. Society of Forensic Engineers and Scientists (SFES):

Founding member, Past PreSident; Past Secretary/Treasurer. Forensic Expert Witness Association (FEWA), San Francisco Chapter. American Institute of Metallurgical Engineers: Past Chairman, Northern California Section. American Nuclear Society: Past Student Affairs Chairman.

Other Professional Affiliations: National Society for Professional Engineers National Associate of Corrosion Engineers Society of Plastics Engineers Society for the Advancement of Material and Process Engineering Society of Automotive Engineers Society of Wood Science and Technology American Society for Metals American Chemical Society American Instijute of Chemical Engineering American Ceramic Society American Welding Society American Insmute of Archaeology The Materials Research Society Professional Honor Organizations: Alpha Chi Sigma, Tau Beta Pi, Sigma Xi

Special Interests: Archeo-materials: Developed university courses.

Presented resealCh on archeological investigation and materials study. F'irearms, ballistics, and arson investigation: Published professional papers.

Detennine cause of fires through analysis of electrical arc residues, using Auger spectroscopy.

High temperature research and consultant on nuclear fuels. 2005: Hong Kong. Presented paper on Bullet Design at

International Association of Forensic Sciences (IAFS) 2002: Montpelier, France. IAFS. Presented paper and poster session. 1996: Tokyo, Japan. Conference chairman at IAFS. Presented and published paper. Speaker:

• Rotary Clubs International chapters: Topic - forensic science and its relationship to business and industrial materials conditions.

• High schools, elementary schools: Topic - the profession offorensic engineering and sciences.

Community volunteer: Current service. Member: Los Altos Hills - Emergency Communications & Emergency Planning Member: Board of Directors - Purissima Hills Water District.

RNAINCcv_fee2008.doc Page 2 612512008

Consulting Fee Schedu,leSummary ~ Updated 912712007

~NA Consulting, Inc.

Specializing in forensic materials engineering and sciences

Corporate Tax 10#: 77-0428512

27820 Saddle Court, Los Altos Hills, CA 94022-1810

Office: 650-949-1092. FAX: 650.949.5641. Cell:·650464-1620. Email: [email protected]

Professional Time:

Depositions:"

Trial:

Retainer:

Travel Expenses:

Automobile/Air Travel:

Support Team Expenses:

Storage Charges:

Professional Services Agreement:

RNAINCcv_fee2ooB.doc

Robert N. Anderson, Ph.D., P .E., P .1. President, RNA Consulting, Inc.

$420.00/hour, including travel.

$420.00/hour "Consistent with the Code of Civil Procedure, pre-payment will be required. -The retaining attorney shall be responsible for notifying the attorney requesting the deposition of these policies.

$420.00/hour, to include courtroom standby. All outstanding invoices are to be paid before trial.

$3,360.00 (8 hours) to be paid at the initiation of a case. - No listing as an expert without advance payment of the retainer and completion of Professional Services Agreement. (See Professional Service Agreement stipulations)

Reimbursement for related travel expenses. Charge for professional time as per case need

Associated expenses, including bridge tolls and parking. Allocated time included as part of professional fee.

Engineering Associate - Range: $175.00 - $250.00/hour Research, Photo-documentation, Technical Associate - $160.00/hour. Lab and other technical and related support - as billed.

$50/month, or by arrangement.

Commences with completion of RNA Consulting, Inc. Services Agreement and submittal of retainer.

Page 3 6/2512008


27820 Saddle Coun Los Altos Hills. {"..A 94022-J810 USA

1/09/08

2127/08

2115107

3113107

3123107

415107

4/9/07

4126107

6/21/07

9/05107

11/05107

12118107

1110/06

1117106

5117106

RolJert N. Anderson. Pb.d., P.E., President

·.1·

Sus: 650-949-1092 Fax: 650.949.5641

email: robenNA@aoLcom

Chris Rice V Genentech. et al

Edward Harrison/Carol Harrison v

Jack Zhao, a minor, and Guardians v American XTAL

Candy Negrete-Schwinn v Costco

Jack Zhao, a minor, and Guardians v American XTAL

Roberto v Harris Rebar. eta!

Blair v Shin Yea

Campbell v Polyguard

Allstate Ins. V Heil

DomvBNSF

Mitchell v Rick Case Cycle, et. at

Tagoai v SMG Moscone Center,

8/15/06

10102106

11121/06

4111105

4120105

5123/05

6121/05

7122105

7/27/05

9/14/05

9126/05

10110/05 1

10/26/05

11129105

12120/05

San Court

Fifth Circuit State of Hawaii No. 02-1 91

Court

San Diego Superior Court

Sutter County Superior Court

Division K, Judicial District Court for the Parish of Jefferson,

State of louisiana

Contra Costa Superior Court No. C01-05111

Third Circuit, State of Hawaii

San Diego Superior Court

Marin Superior Court

Antelope Assoc. v louisiana

Megison v GM et al

liu v KB Homes

Morales v. Delaurentis

Furushima v. Suzuki

Calfee Design

Bridge Common v

Morton v. EBMUD

Cox v. Pacific States Cast Iron

Kellner v True Temper

Mitchell v Rick Case Cycle, et. at

May-Carmen v Wal-Mart

Silveira v Burlington Northern


Robert N. AnderSon, Ph.d . ., P ..E.t President 27820 SaddJe Court I...os AJtos Hins. CA 94022-1810 USA

Bus: 650-949~1092 Fax: 650.9495641

email: robenNA@aotcom

COURT APPEARANCES: Robert N. Anderson, Ph.D., P.E., P.I.

514107& Santa Cruz Superior Court Megison v GM et al 519/07 No. CV150825

9111/07 Stanislaus Superior Court White v Back

No. 343650

5/4/05& Sacramento Superior Court Bunch v Fischer 616105 No. OOAS03600

519/05 San Diego Superior Court Morales v Delaurentis No. GIC 821322

11/16/05 Amador County Superior Court, People v Shoemaker CA. No. OHCR 5952

1219/05 & Marin Superior Court May-Carmen v Wal-Mart 12116/05 Case CIV-050622

7130104 OSHA Hearing Testimony, Docket S025A D.C.

6/9-10 Santa Clara County Superior Court lick Mill Creek Apts. v et. at

4/14/04 Alameda County Superior Court (Criminal case) vloidaCruz

2124-25/04 Los Angeles Superior Court, Griggs v. Caterpillar, BC 216425.

10/02103 Fifth Judicial District Court of Idaho, MorriS v. Renegade, CV 03-00274

04/11/03 San Francisco Sisk v.

COURT APPEARANCES_update-04-07-Q8.doc -1-

Industries, No. 321538.

01128-29/03 Santa Clara Superior Court, Morgan Hill Unified School District v. Minter & Fahy Construction, et at. No. CV 772368.

COURT APPEARANCES_UpdatEKl4-07-08.doc - 2-