Document LKRBEO60Dn8Rd5d4aX8wbNEq5

CONFI?ENTIZ.t * Subject to ' . * . f 14th Judicial pil ,c s^S'twes No. i-1 ^ CENTRAL ENGINEERING - RES 532.0 LAKE CHARLES - RES 757 CONFIDENTIAL THE RELIABILITY STUDY OF THE LAKE CHARLES PLANT "B-II" TRI-ETHANE UNIT-FURNACE AND FURNACE FEED SECTION Report by T. A. Bailey J. W. Barton and D. T. Rigler Distribution J. R. Farst (Report) T. G. Brown/O. L. Cromeans (Report) J. E. Fike (Report) R. E. Baker/J. H. Morgan (Report) T. G. Taylor (Report) R. E. Sanders (Report 4 Copies/Appendix 1 Copy) C. R. West - G.O. (Report 1 Copy/Appendix 1 Copy) Authors (Original Report/Original Appendix) Key Word Reliability Study ICI HCL Releases Hydrocarbon Releases Vessel Failure Vinyl Chloride EDC Tri-Ethane II Tri-Ehtane II Furnace Hazard Study Toxic Release Fire Vapor Cloud Loss Prevention Furnace Safety SL 011944 CONFIDENTIAL: Subiect to protective Order o I4th Judicial District Court Mo- 91-1145 ABSTRACT INTRODUCTION DISCUSSION RESULTS RECOMMENDATIONS CONCLUSIONS APPENDIX TABLE OF CONTENTS 1 2 2 7 9 10 12 SL 011945 ABSTRACT ,lvs Order Jrict Court The detailed reliability study (Level III) of the Tri-Ethane II Unit was broken down into three sections; The furnace and furnace feed, the DCE, and the MC section. This report covers the furnace and furnace feed (recycle EDC) section. The study employs the methodology developed by Imperial Chemical Industries, Ltd. of England (I.C.I.) which has been licensed to PPG Industries. This methodology predicts the frequency of undesirable events or possible hazards and the severity of such incidents. The detailed study of the furnace was concerned with three potential hazards. These hazards being: (1) Tube failure with resultant fire and furnace damage, (2) Failure of furnace burner controls allowing introduction of natural gas into a hot furnace with possible explosion, and (3) Ignition by the furnace of a vapor cloud from outside the furnace area. The detailed study of the furnace feed system was concerned with the following potential hazards: (1) Hydrocarbon spills or releases of Vinyl Chloride, Ethylene Dichloride, and Bottoms/Tars, and (2) Toxic gas release of Hydrogen Chloride, Chlorine, or Vinyl Chloride. The results are shown in Tables V and VI. The main recommendation of this study is to properly proof-test existing critical instrumentation according to the schedules in Table III and IV. SL 011946 -l- INTRODUCTION CONFIDENTIAL: Subject to Protective Order of 14th Judicial District Court No. 91-1145 Detailed reliability studies (Level III) of the units within Plant "B-I" and "B-II" of the PPG Lake Charles Chemical Plant began in December 1977 with the Ethyl Chloride Unit. After the completion of the Ethyl Chloride study the Tri-Ethane II Unit was chosen to be studied since it was nearing completion of construction. This unit was divided into the Furnace and Furnace Feed (recycle EDC), the DCE, and the MC sections for ease of study. This report will cover the Furnace and Furnace Feed Section. Studies of the remaining sections of Tri-Ethane are scheduled next. DISCUSSION This report addresses the Furnace Section and the Furnace Feed Section of Tri-Ethane II. To develop a background for Tri-Ethane II Furnace and Furnace Feed System operations, information was obtained from the operating manual, startup and shutdown procedures, furnace control system wiring diagrams, operation of the Vinyl Chloride Furnace and discussions with operating personnel at Vinyl Chloride and Tri-Ethane II. A school for the new alarm and control system, MOD III, was attended by some members of the study team. The operating criteria for the system was presented. A study of this-system was conducted to determine the failure rates and fractional dead time of the controls and alarms. This analysis and the failure rates are presented in the Appendix. The detailed study of the Furnace was concerned with three hazards. Those hazards being: (1) Tube failure in furnace with resultant fire and furnace damage, (2) Failure of the furnace control system allowing natural gas into a hot fire box and possible explosion, and (3) Potential vapor cloud ignition from the furnace fire box. The study of the Furnace Feed System was concerned with two hazards. Those hazards being: (1) Hydrocarbon spills or the releases of Vinyl Chloride, Ethylene Dichloride, and Bottoms/Tars, and (2) Toxic gas releases (fence line concentrations) of Hydrogen Chloride or Vinyl Chloride. A decision was made to develop the frequency rate for the hydrocarbon releases but not predict the possibility of explosion. The methodology for predicting explosions was not developed until the latter part of this study. It will be used in future studies. The methodology employed in the reliability study was developed by Imperial Chemical Industries Limited (I.C.I.). The authors have also attended a "Fault Tree Analysis" School by Gary J. Powers at CamegieMellon University. Other resources were studied including a very good article that appeared in A.I.C.h.E.'s "Loss Prevention" Volume 8. SL 011947 -2- CONFIDENTIAL: Subject to Protective Order of 14th Judicial District Court No. 91-1145 The next step in the study was to prepare "cause-consequence tables", which are line by line reviews of the existing mechanical flow sheets answering questions as to the effects of various situations which can occur such as no flow, high flow, low flow, high pressure, etc. From the mechanical flow sheets, operating manual, and "cause-consequence tables" the fault trees were developed. The fault trees were reviewed with operations before quantification. Procedures were reviewed with Mr. C. R. West, Senior Reliability Engineer, and Mr. G. B. Edwards, Senior Process Engineer, from Central Engineering General Office. The failure rate data for the mechanical, instrument, computer, and alarm systems were obtained from Maintenance and Industrial Engineering files, I.C.I. Manual of failure rates and an article "Some Data on the Reliability of Instruments in the Chemical Plant Environment", by S. N. Anakora, G.F.M. Engel, and F. P. Lees. Failure rate data on the MOD III computer control and alarm system was obtained from Taylor Instruments. The failure rate data obtained was tempered by engineering judgment to fit the conditions the authors thought applicable to the Tri-Ethane II Furnace and Furnace Feed System. The number of occasions/yr that a tube could rupture and cause a fire, that the control system could allow natural gas into a hot furnace fire box, or the possible vapor cloud ignition by the furnace fire box has been calculated for the furnace. The quantity of material necessary for the above events was not calculated.' For the Furnace Feed System the number of occasions/yr that there could be a hydrocarbon or toxic vapor release was calculated. Not only were the occasions/yr determined but the quantity released was determined using I.C.I.'s procedures and the techniques presented in Crane's Technical Paper No. 410. The quantity of hydrocarbon released was categorized as shown in Table I. The releases are arranged in categories from the small release, Category A, to the large release, Category D. An additional Category E was added to indicate the discharge from a pressure relief system. Category E releases could occur from either an over pressure on the system or the pressure relief system operating prematurely. The category basis is the equivalent heat of combustion. Ethylene is taken as the base. A hydrocarbon release can occur from a pressure relief system functioning or a vessel failure. Unless otherwise specified any occasion that has a vessel over pressure and a failure of the pressure relief system will result in a vessel failure. The release of a toxic material was classified as either a Class 0, I, II, or III. The class of release was established by I.C.I. in the computer program old TOXIDISP and defined in Table II. The boundary distances were established as shown on Drawing 32A-602.12 in the Appendix. SL 011948 -3- CONFIDENTIAL: Subject to Protective Order of 14th Judicial District Court No. 91-1145 The weather data for Lake Charles used in the computer program was obtained from the National Oceanic and Atmospheric Administration. The toxic components were Hydrogen Chloride, Vinyl Chloride and Chlorine. Leaks from piping, pumps, pump seals, and sample points were classified according to the severity of the release initially. Later it was decided that although these failures (leaks) occur, they do so in normal operations. We decided to determine the frequency of occurrence of these items, but not the severity (classes and categories). To reflect this change in philosophy we decided to report the hazards as two numbers. The first number was called operational failures. This number included hazards due to the process (SRV release, vessel failure, etc.). Operational failures were divided into both hydrocarbon release categories and toxic gas release classes. It is this number that reflects the effect of the protective systems on the hazard rates. The more all encompassing category is called total failures. This number consists of operational failure plus failures due to piping, pumps, pump seals, and sample points. Total failures were reported as the frequency of occurrence of the hazard, but not divided according to the severity of the hazard. Assumptions used during the study were that systems would function as designed, proof-testing was rigorous and performed as stated, operators would act according to predictions, and slow control loop failures would be revealed and corrected. The effect on the operation of the Furnace and Furnace Feed System resulting from a number of changes in the systems was evaluated. The effect of each change is presented in tabular form. A high pressure nitrogen system is provided to aid in the clearing of sections of the Furnace Feed System. A study of the compressor receiver was conducted. Items in the initial overview pertaining to the Furnace and Furnace Feed System were reviewed. Appropriate action was taken or is in progress on these items. Additional information used in this study came from the operating manual, process flow sheets 69A-10051.2 to 69A-10074.1, MOD III operations manual, MOD III alarm drawings 69A-76003 to 69A-76015, safety relief system drawings 69A-80001 to 69A-80005, furnace flame control wiring diagram 69A-69039, I.C.I.'s toxic gas emissions computer program book, and many discussions with people in operations. The authors wish to express their thanks for the help we received from the many people in Maintenance, Engineering, and Operations who were very helpful in providing the information necessary for this study. A special vote of appreciation goes to A. E. Howerton and F. W. Heinemann for their preliminary work on the cause-consequence tables and fault trees used in this study. SL 011949 -4- CONFIDENTIAL-: Subject to Protective OrA;r of 14th Judicial District Cojli No. 91-1145 TABLE I HYDROCARBON HAZARD CLASSIFICATIONS Hydrocarbon release rates are categorized into five classifications to permit evaluation of an event based on its frequency of potential risk involved. While an attempt was made to categorize the various releases into risk categories, the limits placed on each category are somewhat arbitrary. Release rates for different compounds in each category are based on BTU equivalency. Hazard Category Definition A Vinyl Chloride and EDC release rates are 0 to 105 lbs/1.0 minutes and 0 to 180 lbs/ 10 minutes, respectively, or larger releases which for various reasons would yield a ground level concentration below the lower explosion limit. B Vinyl Chloride and EDC release rates are 106 to 1060 lbs/10 minutes and 181 to 1800 lbs/10 minutes, respectively. C Vinyl Chloride and EDC release rates are 1061 to 10610 lbs/10 minutes and 1801 to 18000 lbs/10 minutes, respectively. D Vinyl Chloride releases in excess of 10610 lbs/10 minutes and EDC releases in excess of 18000 lbs/10 minutes. E Discharge from a pressure relief system from either over pressure or premature release. SL 011950 -5- CONFIDENTIAL: Subject to Protective Order of 14th Judicial District Court No. 91-1145 TABLE II TOXIC GAS HAZARD CLASSIFICATIONS Toxic gas emission hazard calculations were made using the I.C.I. prepared computer program TOXIDISP to determine the probability for obtaining various hazard categories at the plant boundary based on multiple conditions such as release rates, location, weather conditions, etc. Hazard Class Definition 0 No nuisance to the public at the plant boundary. 1 May be smelled and thereby cause a nuisance to the public at the plant boundary - unlikely to cause harm. II May cause some degree of distress to people, or damage to vegetation at the plant boundary which would cause a nuisance and which could lead to claims for compensation. III Could be dangerous or distressing to personal safety, or could be possible risk of life at the plant boundary. SL 011951 -6- CONFIDENT! AL: ___.i, i1 v\ 1e Order '-rice Court RESULTS While the supporting information for this study is quite lengthy, efforts were made to condense it as much as possible and still allow the reader to interpret the results presented. The results are presented in two separate sections: fl) The Furnace Section, and (2) The Furnace Feed Section. For the furnace the hazards are separated into three separate events. These are: (1) Furnace tube failure and resulting fire, (2) Furnace control system fail and potential explosion, and (3) Ignition by the furnace of a vapor cloud formed outside the furnace area. The summary of the furnace hazards indicates the expected occurrence for each hazard. This section shows the alternates that were considered and evaluated. The last alternate is a cumulative effect of all the previous changes and recommendations. The details of calculations and procedures for the alternatives are presented in the Appendix. Alternates considered in the furnace study to decrease potential hazards are as follows: 1. Proof-test existing alarms and control systems on a regularly scheduled basis. Refer to Table III for a recommended schedule of proof-testing. 2. Proof-test the snuffing -steam system every three months. 3. Install a protective cover on the hand switch for the snuffing steam valves. This reduces the occasions that an operator can accidently activate the snuffing steam while the furnace is in operation. 4. Install an interlock such that when the snuffing steam valves are activated the furnace is tripped. 5. Add additional shutdown (Modicom) system. Install additional flow indicators on each pass through the furnace. Tie the new flow indicator into a system with the existing flow indicator and the tube wall temperature. Make this a two out of three voting system. The Modicom System as described was being installed during this study. We decided to treat it as an alternative and show its effect on the hazard rates. The only modification suggested is that instead of both flow signals for a pass coming from one orifice run upstream of the SRV that an additional orifice run be installed on each pass between the present SRV and the furnace entrance. One of the existing flow transmitters could then be installed on this orifice. As presently installed the SRV on a pass could relieve reducing flow through the pass but the Modicom wouldn't react because both flow loops would see the flow going through the SRV and not indicate loss of flow. SL 011952 -7- CONFIDENTIAL: Subject to Protective Order of 14th Judicial District Court Nc. 91-1145 Regular and proper proof-testing of the furnace control and alarm system will reduce the possible occurrence of a tube overheating and rupturing from once every 31.9 years to once every 76.6 years. The potential for a natural gas explosion in the furnace is still 90.9 years and potential fire box vapor cloud ignition is every 18.1 years. The main purpose of proof-testing is to improve the probability that a back-up system will perform whenever it is called on to do so. One negative effect of a back-up system is that occasionally it will be tripped accidently and its performance will have some unwanted effects. In the case of the two snuffing steam flow control valves for the furnace, we predicted that they will be accidently tripped 0.05 occasions per year. If these two steam flow control valves are not proof-tested, we could expect one of them to open only .0077 occasions/year. In this case the furnace would be accidently snuffed out, leading to a potential natural gas explosion once every 90.9 years. If these two valves are proof-tested every three months, their reliability improves so that when they are accidently tripped, at least one of them will open 0.0497 occasions/year leading to a potential natural gas explosion once every 36.2 years, which is more frequent than if not proof-tested. Because it is very desirable to have snuffing steam available when needed, such as when a vapor cloud approaches the furnace, we recommend that additional safeguards be installed to prevent accidental tripping of the steam valves. Installing an interlock that trips the furnace each time the snuffing steam valve was activated would decrease the possibility of an explosion in the furnace. With an interlock the explosion potential was reduced from once every 62.9 years to once every 113.6 years. Adding additional flow indicators and tieing them into the existing flow indicators and tube wall temperatures gives a two out of three voting shutdown system. With this addition the potential of a tube overheating and a fire is reduced from once every 76.6 years to once every 84.7 years. The alternatives mentioned previously would reduce the possible tube rupture and fire from once every 31.9 years to once very 84.7 years, a potential gas explosion from once every 90.9 years to once every 113.6 years, and a vapor cloud ignition from once every 18.1 years to once every 42.7 years. The study for the Furnace Feed System was concerned with two major events. Those were: (1) Hydrocarbon releases, and (2) Toxic releases. The general hazard summary will show the expected frequency of occurrence of a hydrocarbon or toxic release for the present installation. This section will also show the effect of alternatives on the Furnace Feed System. The summary of hazards for the individual sections of the feed system is shown in the Appendix. The hazards are presented as two numbers. The first number is for operational failures and the second number is total failures. Total failures include operational as well as piping, pump, pump seal, and sample point failure. SL 011953 -8- CONFIDENTIAL* t to protective 0r<3ef Sutejec District Court judicial He. 9X-U45 The operational failures are separated into hydrocarbon release categories and toxic classes. Hydrocarbon and toxic release guidelines were presented previously in Tables I and II. Total failures were not separated into categories and classes. The total failure for the system if screwed piping connections were minimized is also presented. The alternate considered in the Furnace Feed System was proof-test of control systems and alarms on the separate groups of equipment on a regularly scheduled basis. Refer to Table IV for a recommended schedule of proof-testing. The alarms and control systems that are provided appear to be satisfactory. It is recommended that the flexible hoses used on the tar buggies be replaced every three months. Initially the piping systems for the equipment in the Furnace Feed System were studied to determine the frequency with which they would fail. These occasions/year were then separated into expected hole sizes ranging from 1/16 inch to 1/2 inch in diameter. From this the hydrocarbon and toxic category and classes were determined using the information from Table I and the I.C.I. toxic gas release computer program. Later discussions brought forth the philosophy that although the piping system may leak it is a part of normal operations and would not be shown or considered as an operational failure. It was decided that piping failure would be shown in the total failure summary. By minimizing the use of screwed piping in flashing flammable liquid service by backwelding or using flanges reduces the total failures from 89.19 occasions/year to 66.36 occasions/year. The pump, pump seal, and sample point failures were considered to be constant with little if any change from one piece of equipment to the next. Results of the high pressure nitrogen compressor and receiver indicate the potential for a hydrocarbon release of the compressor every 4500 years. There is a possibility of destroying the compressor every 32.5 years but this would be from corrosion, mechanical flaw or a leak in the compressor cooling water jacket. RECOMMENDATIONS The following recommendations are made for the Furnace and Furnace Feed System to improve the operation of these sections. These recommendations should reduce the hazard to the equipment and personnel in these areas as well as at the plant boundaries from potential fires, explosions, hydrocarbon and toxic releases. SL 017954 -9- FURNACE 1. Proof-test existing alarms and control systems on a regularly scheduled basis. Proposed schedule is presented in Table III. 2. Proof-test snuffing steam system every three months. 3. Install protective cover on hand switch for snuffing steam valves. 4. Install interlock that trips furnace when snuffing steam valves are activated. 5. Add additional shutdown system that is tied into a new flow indicator to be installed. This would be a modification of the existing Modicom System. FURNACE FEED SYSTEM 1. Proof-test existing alarms and control systems on a regularly scheduled basis. Proposed schedule is presented in Table IV. 2. Replace the flexible hose between the Dopp kettle and tar buggy every three months. 3. Replace with flanged or backweld as much screwed pipe in flashing flammable liquid service as possible. By using a minimum amount of screwed pipe the frequency of leaks is reduced from 89.19 occasions/year to 66.36 occasions/year. CONCLUSION If the proposed recommendations are implemented the following results could be realized. FURNACE Incident Before After Tube Rupture and Furnace Fire 31.9 years/occasion 84.7 years/occasion Control Failure and Natural Gas Explosion 90.9 years/occasion 113.6 years/occasion Vapor Cloud Ignition From Furnace Fire Box 18,1 years/occasion 42.7 years/occasion SL 011955 -10- 'CoOTWENTTW^ oraer "bub-iect to... ,, District ofr i4tb Judi< Do. 91-i 145 FURNACE FEED SYSTEM Incident Before (years/occasion) After (years/occas Hydrocarbon Releases Category A B C D E .32 1.34 17.57 769.23 .93 .32 1.38 18.18 833.33 1.12 Toxic Releases Class 0 I II III .57 6.78 10.20 40.65 .58 13.95 21.50 70.92 TOTAL FAILURES The number of total failures could be reduced from 89.46 to 66.36 occasions/year by minimizing screwed piping in flashing flammable liquid service. &0//po eJ//J8o a/ti/so SL 011956 -11- TABLE III CONFIDENTIALr orttf SAiUd4b-itlhh^CJtt uto2o.jcia] LD-i'pL]tV;-e'rt4,,O- rra- er , N<>. `-J - J L4S .................... Case I Proof-Test Controls Furnace Area Equipment Designation Proof-Test Interval Furnace Individual Burner Shutdowns Inlet Damper Trip Outlet Damper Trip Hi Tube Metal Temperature No. 1 (TI-130) Hi Tube Metal Temperature No. 2 (TI-132) Hi Furnace Outlet Temp. No. 1 (TI-110) Hi Furnace Outlet Temp. No. 2 (TI-114) Hi Tube Metal Temp. (CRT) (UJR-111) Hi Tube Metal Temp. (CRT) (UJR-113) Hi Inlet Pressure No. 1 Pass (PI-112) Hi Inlet Pressure No. 2 Pass (PI-108) Hi/Lo Furnace Natural Gas Flow (FRC-102) Lo Furnace FeedFlow No. 1 Pass (FRC-104) Lo Furnace FeedFlow No. 1 Pass (FRC-167) Lo Furnace FeedFlow No. 2 Pass . (FRC-106) Lo Furnace FeedFlow No. 2 Pass (FRC-165) Furnace Oxygen Analyzer (AE-134) Snuffing Steam Valves (FV-108 8 FV-1099) Main Natural Gas Maxon Valve (FY-1097) Burner Maxon Valves (FY-001 thru FY-024) Pilot Maxon Valves (FY-1001 thru FY-1024) Gas Vent Valve (FCV-1098) Vapor Detectors Deluge System Modicom Trip on Low EDC Feed Flow to Furnace Furnace Shutdown System Furnace Bridgewall Temp. (TI-116) Furnace Stack Temp. (TI-128) Furnace Natural Gas Pressure Controller (PIC-100) Four Months Two Months Two Months Six Months Six Months Six Months Six Months Six Months Six Months Six Months Six Months Four Months Six Months Six Months Six Months Six Months One Month Three Months Four Months Four Months Four Months Four Months Six Months Three Months Four Months Four Months Six Months Six Months Four Months SL 011957 -12- TABLE IV Case I Proof-Test Controls Process Area Equipment Designation Proof Test Interval Quench Tower Inlet Press. No. 1 Pass (PR-118) Quench Tower Inlet Press. No. 2 Pass (PR-136) VC-HC1 Still Pressure Control Loop (PRC-156) VC-HC1 Still Manual Press. Control (HIC-154) VC-HC1 Still Bottoms Temp. Control (TRC-146) Cl^ Addition to Recycle Reverse Flow Trip (FT-171) Low EDC Recycle Flow Trip (FT-166) Chlorine Cylinder Line Check Valve Heavy Still Feed Tank Press. Control (PIC-1750) Heavy Still Feed Tank Press. Alarm (UJR-177) Heavy Still Reflux Tank Press. Control (PIC-217) Heavy Still Bottoms Tank Level Indicator (LI-168) Heavy Still Bottoms Flow Heavy Still Condenser Fan Motor Alarm (UJR-209 & 211) Six Months Six Months Six Months Four Months Six Months Six Months Six Months Six Months Four Months Six Months Six Months One Year Six Months One Year SL 011958 -13- TABLE V C0SFIDEN1IRL- FURNACE HAZARD SUMMARY OCCASION/YEAR ! i j CONDITION AS IS PROOF-TEST FURNACE SHUTDOWN SYSTEM TUBE RUPTURE AND FIRE .0313 (31.9)* .01306 (76.6)* HAZARD POTENTIAL FOR NATURAL GAS EXPLOSION .011 (90.9)* .011 VAPOR CLOUD IGNITION FROM FURNACE FIRE BOX .05521 (18.1)* (.09751-10.3) .05521 PROOF-TEST SNUFFING STEAM SYSTEM .01306 .0276 (36.2)* .02340 (42.7)* INSTALL PROTECTIVE COVER SNUFFING STEAM SWITCH .01306 .0159 (62.9)* .02340 SNUFFING STEAM SWITCH TRIPS FURNACE ADD MODICOM SHUTDOWN SYSTEM .01306 .0088 (113.6)* .02340 .0118 (84.7)* .0088 .02340 *YEARS/OCCASION SL 011959 -14- TABLE VI TRI-ETHANE II UNIT FURNACE FEED AREA HAZARDS SUMMARY lazard Class >erational Failure, esent Equipment and lerating Procedures Total Hydrocarbon Releases Occasions/Year AB C D E Toxic Gas (VC8HC1) Releases Occasions/Year 0 11 III III 5.0007 3.1079 .7463 . .0569 .0013 1.078 1.7477 . 1474 (.2000)* (.3218) (1.3399) (17.5747) (769.2308) (.9276) (.5722) (6.7843) .098 .0246 (10.2041) (40.6504) serational Failure, -oposed Alternates e Implemented 4.7681 (.2097) 3.0846 .7250 .0550 .0012 .8919 1.7233 .0717 (.3242) (1.3793) (18.1818) (833.3333) (1.1212) (.5803) (13.9470) .0465 .0141 (21.5054) (70.9220) >tal Failure, Present [uipment and aerating Procedures 89.4260 (.01118) )tal Failure, oposed Alternates e Implemented 89.1935 (.01507) )tal Failure, Present luipment and >erating Procedures >th Minimum Use of :rewed Piping 66.3651 (.01507) HOi . cW *Numbers in Parenthesis are Years/Occasion SL 011960 -15- O>- 2 O rr 'O DHj - ,!'T<) r.J m 00 oi rc h. ti 2h- f<D o rf O