Fill and Sign the Bir Form Fillable 1999

I ' DOCKET NO. 1 In the Matter of - IN THE MATTER OF THE FILING BY MONTANA-DAKOTA UTI~ITIESCO., A DIVISION OF MDU RESOURCES GROUP, INC. FOR APPROVAL OF WAIVER Public Utilities Commission of the State sf South Dakota DATE MEMORANDA STATE PUBLIBHINQ.CO.. PIERRE. SOUTH DAKOTA-SMEAD 104 5PlAlJO UTILITIES CO A Division of MDU Resources Group, Inc. 400 North Fourth Street Bismarck, ND 58501 (701) 222-7900 June 3,2002 E Deborah Elofson South Dakota Public Utilities Commission State Capitol Building 500 East Capitol Avenue Pierre, SD 57501-5070 Re: SOUTH DAKOTA PUBLIC UTILITIES COMAAISSIOP( Waiver Request; 49 CFR Part 192, Paragraph l92.59(a)(l) (ASTM D25 13) Plastic Pipe Materials Dear Ms. Elofson: Martin Bettman of your office advised Montana-Dakota Utilities Co. (Montana-Dakota) with an April 23,2002 letter confirming the need to request a waiver for UPONOR pipe received in July 2000. The following letter is Montana-Dakota Utilities Co.'s request for that waiver. In July 2000, a shipping error made by Chevron Phillips Chemical Company resulted in UPONOR receiving non-standard polyethylene raw materials, i.e. TR- 130 resin, that they subsequently converted into pipe. Montana-Dakota received several thousand feet of this pipe that had not been properly qualified to ensure compliance with ASTM D25 13 as required by the referenced code for use in its natural gas system. Montana-Dakota did install that pipe near Rapid City, South Dakota. Since its installation, Chevron Phdlips Chemical Company performed extensive testing and demonstrated the pipe does in fact meet the minimum requirements of ASTM D 25 13. Please see the attached copy of their report. Based on the Chevron Phillips testing, Montana-Dakota proposes to allow this pipe to remain in service. Accordingly, Montana-Dakota is requesting a waiver from the South Dakota Public Utility Commission and the Federal Department of Transportation regulations to allow the pipe to remain in service. Page 1 of 4 INTARIA-DAKOTA UTILITIES CO. In accordance with Mr. Bettrnann's April 23,2002 letter, Montana-Dakota is providing the following information for your consideration: 1. Montana-Dakota Contact: Bruce Nelson, P.E. Gas Distribution Manager 400 North ~ o ' k t Street h ~i&narck,ND 5 850 1 70 11222-7784 2. Waiver requested for: 49 CFR Part 192, Paragraph 192.59(a)(l) (ASTM D2513) Plastic Pipe Materials 3. Description of facilities: The waiver request is for approximately 38,600 feet of two-inch pipe. See attached. 4. Location of facilities: Miscellaneous gas distribution mains scattered throughout the Rapid City and Black Hills Area of South Dakota. Since the pipe installation was part of routine growth for the area and a portion of approximately 75,000 feet of pipe installed in 2000, the exact location of the pipe is unknown. 5. Description of particular operation for which the waiver is requested: Montana-Dakota received several thousand feet of the UPONOR pipe made fi-om TR-130 resin that had not been properly qualified to ensure compliance with ASTM D25 13 as required by the referenced code for use in its natural gas system. Page 2 of 4 WTANA-DAKOTA UTILITIES CO. I 6. An indication of increased risks that the particular operation would create and additional safety measures that proposed to compensate for the additional risk: Based on the attached report, research performed by WONOR and Chevron Phlllips Chemical Corporation, and supported by the Plastic Pipe Institute, Montana-Dakota does not believe there is additional risk to its customers or the public in general. Montana-Dakota expects the pipe in question to perform at least as well as approved polyethylene pipe installed today. 7. Risk Mitigation Measures: Based on the above statement and the attached report, MontanaDakota believes that granting of t h s waiver is not inconsistent with pipeline safety and the replacement of the pipe would be unnecessarily burdensome on Montana-Dakota' s customers since it would involve excavation of streets and yards as well as an interruption of service. Montana-Dakota strives to remain in compliance with the pipeline safety requirements and considers t h s issue a significant matter. Accordingly, Montana-Dakota appreciates your support in accepting and processing t h s waiver request. Please let me laow if you have any questions or concerns. Sincerely, Douglas M. Lee, P.E. Senior Staff Engineer Attachment - UPONOR and Chevron Phillips Chemical Corporation Report Page 3 of 4 INTAM-DAKOTA UTlllTlES CO. CC: Bruce Nelson - General Office Tamie Aberle - General Office Tom Hopgood - Schuchart Building - Legal Department Wanda Dewing - General Office Dan Farmer - Billings Timothy G. Taylor Chevron Phillips Chemical Company 1301 McKinney, Suite 1200 Houston, TX 770 10 Jimmy Collier Uponor Aldyl Company 5062 Allred Road Mariposa, CA 95338 Page 4 of 4 CONFIDENTIALAND PROPRIETARY TREATMENT REQUESTED - JUN 0 6 2002 wwe SOUTH DAKOTA UTlUTIES COMMISSION MEMORANDUM CONCERNING UPONOR MEDIUM DENSITY POLYETHYLENE PIPE RECALL Prepared by: Chevron Phillips Chemical Company LP November 12,2001 TABLE OF CONTENTS Page Section Executive Summary 1 FACTS 3 I. Discovery Of The Use Of TR-130 In Certain MDPE Pipe ~anufacturedBy Uponor Actions Taken By CPChem And Uponor Following Discovery Of The Use of TR-130 11. 111. IV. v. .. Extensive Testing And Analysis Show That TR-130 And TR-4 18N Have Essentially Equivalent Physical Characteristics, And That A Unique Confluence Of Several Factors That Have Not Been Replicated Elsewhere In The Field Resulted In The Keyspan Cracking Incident 6 TR- 130 Has Received A Hydrostatic Design Basis Rating From The Plastic Pipe Institute 9 PIPE MANUFACTURED WITH TR-130 COMPLIES WITH APPLICABLE LAW AND SAFETY STANDARDS IN ALL RESPECTS, AND DOES NOT REQUIRE A WAIVER 9 ANY PERCEIVED NON-COMPLIANCE WITH APPLICABLE REGULATIONS SHOULD BE WAIVED 11 OPERATIONAL CHANGES TO PREVENT FUTURE MISTAKES 13 CONCLUSION 15 EXHIBITS 1-37, DESCRIBED IN THE EXHIBIT INDEX THAT FOLLOWS, APPEAR BEHIND P.17. - - - EXHIBIT LIST . . EXHIBIT TITLE EXHIBIT NUMBER .. Letter Dated January 24,2000 from Uponor Aldyl Company and Chevron Phillips Chemical Company LP Letter Dated March 14,2001 from Chevron Phillips Chemical Company LP TR-130 Meets ASTM D3350 Specifications ASTM D3350 Cell Classification - Samples Remolded from Plastic Pipe ASTM D3350 Cell Classification - Natural Polyethylene Thermal Stability Test ASTM D3350 Pipe Made With TR-130 Meets ASTM D25 13 Specifications Pipe Made With TR-130 Meets ASTM D25 13 Chemical Resistance Specification ASTM D25 13 Squeeze-Off Test ASTM D25 13 Quick Burst Test ESCR (PENT) Test Pipe Made With TR-130 Meets ASTM Dl248 Grade P24 Specifications Dart Impact Test Performed at -40' F Dart Impact Test Performed at Various Temperatures Dart Impact Test (Performed on Chevron Phillips Chemical Company LP TR-130 Pipe) Dart Impact Test (Performed on 3" & 4" Pipe Made from TR- 130) EXHIBIT LIST EXHIBIT EXHIBIT TITLE 18 Cold Temperature Tapping Tests with 60 psig Internal Pressure Fitting "Knock-Off' Test @art Impact Test Performed in Accordance with DOT Regulations) b Tensile Elongation For Department of Transportation Fusion Qualification Arizona Outdoor Weathering Test Shows No Notable Degradation After 2 Months of W Exposure Arizona Outdoor Weathering Test Shows No Notable Degradation After 4 Months of W Exposure Accelerated Weathering Test Results - Series 1 Accelerated Weathering Test Results - Series 2 TR-130 and TR-4 18N Have Essentially Equivalent Physical Characteristics TR-130 and TR-418N Meet ASTM Polyethylene Specifications Molecular Weight Distribution (Molecular Weight and Molecular Weight Distribution Essentially Equivalent for TR-130 & TR-418N) Polyethylene Rheology (Dynamic Melt Viscosity vs. Frequency) Pipe "Bend-Back" Test Cold Temperature Squeeze-Off of 4" Pipe With 60 psig Internal Pressure 23°C Long Term Hydrostatic Strength for TR130 Pipe EXHIBIT LIST EXHIBIT TITLE EXHIBIT . . . TR- 130 Sustained Pressure Testing at Elevated Temperatures S4 Rapid Crack Propagation Analysis of Uponor Pipe by Dr. Walter L. Bradley, Ph.D., P.E. Letter Dated May 4,2001 from Plastic Pipe Institute Granting An HDB Rating for TR-130 ASTM D2513 - 96a ASTM D3350 - 01 CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED MEMORANDUM CONCERNING UPONOR MEDIUM DENSITY POLYETHYLENE , PIPERECALL EXECUTIVE SUMMARY On March 14, 2001, Chevron Phillips Chemical Company LP ("CPChemyy) advised customers of Uponor Aldyl Company ("Uponor") that certain Medium Density Polyethylene ("MDPE") pipe manufactured by Uponor from resin supplied by CPChem was being recalled, because the pipe was made from the wrong CPChem resin. In facilitating the March 14 recall announcement, and implementing its Reimbursement Program for costs associated with the recall, CPChem was acting in a proactive manner while an investigation of all relevant facts continued. Since that time, a substantial amount of testing and analysis has occurred. This Memorandum, and exhibits, are intended to bring Uponor's customers up to date on recent developments and to provide a comprehensive overview of the history of this matter, as well as our analysis of the safety of the pipe in question and its compliance with applicable standards. As set forth in detail in this Memorandum, CPChem believes that the MDPE pipe in question is safe for natural gas pipe applications, and complies with all applicable laws, regulations, and specifications for natural gas pipe applications. Accordingly, it is not necessary to apply with state or federal regulators for a waiver for non-compliance. Nevertheless, should an operator decide to apply. for a waiver for its unique reasons, the data set forth in this Memorandum should suffice to establish that the granting of a waiver would not be inconsistent with - and, in fact, would be wholly consistent with - pipeline safety. See 49 U.S.C. $60118(c) (waiver may be granted provided that the waiver "is not inconsistent with pipeline safety"). CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED The present situation arose after CPChem mistakenly delivered to Uponor, in July 2000, a single shipment of CPChem Marlex@ HHM TR-130 polyethylene ("TR-130") rather than CPChem Marlex@ HHM TR-418N polyethylene ("TR-418Nn). Not knowing of the misshipment,'uponor extruded the resin into MDPE pipe of various sizes during a tkree-day period in July 2000. TR-130 and TR-418N have essentially equivalent physical properties, rendering both appropriate for gas pipe applications. The only significant difference between the two resins is that virgin TR-418N polyethylene contains an UV inhibitor additive to prevent UV degradation after prolonged exposure to UV, whereas virgin TR-130 polyethylene does not contain an UV inhibitor as sold by CPChem. However, all of the pipe made with TR-130 does contain W inhibitor because UV inhibitor was contained in the yellow color concentrate used in the pipe, and it was a constituent in the regrind material used in making the pipe. Accordingly, as discussed more fully herein, the pipe meets applicable regulations concerning the presence of UV inhibitor. Moreover, the installed pipe in question does not present a safety risk, even though it contains less UV inhibitor than pipe made with TR-418N, because the pipe in question was not stored outside for a significant amount of time. There has only been one reported instance of field failure of the pipe made with TR-130. In that case, three different segments of six-inch pipe in one installation, and one segment of six-inch pipe in a separate installation, cracked during pre-installation pressure testing by Keyspan Energy Delivery - New England ("Keyspan") in the Boston area. Extensive investigation of this incident shows that the cracking was likely caused by a unique combination of circumstances, including an impact event and installation in extremely cold temperatures. Moreover, it appears that any potential issue is limited in scope to six-inch pipe made on a single CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED production line when Uponor was unknowingly extruding TR-130 resin, all of which was covered by the announced recall program. Finally, other than the pipe that cracked in Boston, all installed pipe manufactured during the period in question passed pre-commission pressure tests and has performed without incident for close to, or in some cases more than, a year. In sum, we believe that the subject pipe poses no safety or performance risks and meets all applicable standards.' in any case, we recognize that certain owners or operators may feel more comfortable seeking affirmation from their regulatory agencies, whether in the form of a waiver request or otherwise. The facts and data contained in this Memorandum outline in detail the basis for such regulatory cooperation. I. FACTS A. Discovery Of The Use Of TR-130 In Certain MDPE Pipe Manufactured By Uponor In July 2000, CPChem mistakenly delivered to Uponor a shipment of TR-130 resin rather than the specified TR-418N resin. Uponor, in turn, manufactured approximately 520,000 feet of MDPE gas pipe in sizes ranging from one-half inch to six-inch using the TR-130 resin on July 21 - 23,2000. The pipe was then sold to Uponor's customers. CPChem's shipment error was discovered when one of Uponor's customers, Keyspan, experienced cracking of four segments of six-inch pipe, manufactured by Uponor on July 22, 2000. The cracking was discovered during routine pre-commission pressure testing in December 2000 during installation of the pipe in extremely cold weather. Samples of the cracked pipe were sent to CPChem in December. Initial product testing did not reveal anything abnormal &,there was nothing to lead CPChem to believe that the pipe had been made with a resin other than TR-418N). It was 1 Added comfort is derived from the fact that utilities that installed MDPE pipe manufactured with TR-130 presumably will engage in all normal patrolling and leak survey testing of its pipeline as required by 49 C.F.R Part 192.723. . CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED not until subsequent testing perfoded in January showed lower levels of UV inhibitor in the cracked pipe that CPChem was able to determine that the pipe had been manufactured with a resin other than TR-418N. Additional testing revealed that the pipe was made fiom TR-130 .. rather than TR-418N. Other than the issues discovered during the Keyspan installation, neither Uponor nor CPChem have been notified of any abnormal installations, problems with pressure testing, or any performance problems whatsoever with any pipe made with TR-130. B. Actions Taken By CPChem And Uponor Following Discovery Of The Use Of TR-130 Uponor and CPChem immediately informed the U.S. Department of Transportation Office of Pipeline Safety of the discovery of the use of TR-130 in certain MDPE pipe, because TR-130 had not previously been determined to comply with the American Society for Testing and Materials ("ASTM) D2513-96a (one of the applicable regulatory standards). On January 24, 2001, Uponor and CPChem issued a letter to Uponor's customers advising them of the discovery. The January 24, 2001 letter was sent by CPChem and Uponor as soon as the companies learned of the resin mis-shipment, without waiting for the development of additional facts or analysis. The companies only knew that pipe had cracked in the field and that the pipe contained the wrong resin. They had not yet been able to confirm the pipe's safety, nor had they had time to cofirm whether the resin switch was the cause of the cracking incident. Given that it would take time to perform the extensive analysis needed to answer these questions, CPChem and Uponor took what they believed to be the proactive, correct course of action: immediately informing utilities of the facts then known, and stating that details of a product recall would be announced in the near future. .-- -4- CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED The January 24 letter was sent to all of Uponor's customers that purchased pipe . . made at Uponor7sfacilities within a four-day timeframe when it was suspected the TR-130 resin could have been used. The letter advised that Uponor and CPChem were initiating extensive testing to more fully evaluate the situation and to assure that the affected pipe posed no safety hazard. The letter also asked Uponor's customers to immediately refrain fiom further installation of uninstalled pipe made during the suspect timeframe. A copy of the January 24, 2001 letter is attached as Exhibit 1. In the immediate aftermath of that letter, analysis of Uponor7s pipe production information revealed that TR-130 had been used to manufacture pipe only on part of July 21, 2000, all of July 22, 2000, and part of July 23, 2000. Moreover, analysis confirmed TR-130's likeness to TR-418N, and that pipe made from TR-130 should perform safely in the field. Based on this new information, on March 14, 2001, CPChem sent a letter to customers outlining the specific terms of the Recall and Reimbursement Program. The March 14 letter announced the recall of six-inch pipe produced during the operative time period with TR-130 (only five Uponor customers received the recalled six-inch pipe), and all pipe made during the July 20-23, 2000 time period that had not yet been installed. The March 14 letter also indicated that extensive testing protocols aimed at assuring that TR-130 met applicable safety standards were continuing. A copy of the March 14,2001 letter is attached as Exhibit 2. To date, the recall program has been highly successful. A substantial amount of the recalled pipe has been taken out of service andlor returned, and the operators have been fully reimbursed by CPChem. Moreover, other than the original cracking of Keyspan pipe, there have been no reported incidents of cracking or any other field performance problems with the pipe in question. . ... - . . .-. - .. CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED C. Extensive Testing And Analysis Show That TR-130 And TR-418N Have Essentially Equivalent -Physical Characteristics, .And That A Unique Confluence Of Several Factors That Have Not Been Replicated Elsewhere In The Field Resulted In The Keyspan Cracking Incident The initial inference that 'may have been drawn by some upon first learning that the Keyspan pipe that cracked was made with the wrong resin, was that the resin may have been the cause of the failure and that potentially all pipe made with that resin in the July 21-23 timeframe could pose a risk. Because the companies felt it was important to communicate immediately to Uponor's customers and to move forward with a Recall program as soon as possible, the companies initiated contact with Uponor's customers before a fill and complete analysis of the issue could be conducted. Since that time, CPChem has devoted a great deal of effort and resources to determining why, of all the pipe manufactured with TR-130 during the July 2000 time period, only the Keyspan six-inch pipe cracked. This analysis suggests, as explained below, that the Keyspan pipe failed as a result of a unique combination of circumstances, and not because there is something inherent in the TR-130 resin that could cause continued failures. First, results of extensive testing demonstrate that pipe made with TR-130 has essentially equivalent physical characteristics and performance capabilities as pipe made with TR-418N. Accordingly, pipe extruded from both resins should perform in a similar - and safe fashion. Documentation of all such test results may be found at Exhibits 25 to 28. Additionally, among other things: Exhibits 3 and 7 are summary chms that show that TR-130 meets all specifications of the two ASTM standards that apply to the pipe in question: ASTM D2513 and ASTM D3350. The ASTM standards are illuminated in greater detail at pages 10-12 of this Memorandum. Pipe made with TR-130 and TR-418N (including pipe manufactured by Uponor as well as six and eight-inch pipe recently extruded by CPChem from Exhibit.14 -. 17. TR-130) shows similar results in impact testing. CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED The O&J exception is that samples of six-inch pipe made by Uponor on the same date and production line as the Keyspan cracked pipe, exhibited low cold temperature impact resistance in recent laboratory testing. See Exhibits 14 and 15. Six-inch pipe made from the same material on.the same day on Uponor's other production line had excellent cold temperature impact strength. See Exhibit' 14. (CPChem has also tested samples from July 21 and 23 from the same production line that produced the Keyspan pipe. These samples were made from TR-418N, not TR-130, and exhibited good cold temperature impact resistance thus isolating the production window at issue. See Exhibit 14.) rn Testing of pipe made with TR-130 shows excellent resistance to slow crack growth, which is an indication that long-term performance characteristics should be excellent. See Exhibits 3,4,5. Specifically the Polyethylene Notch Tests ("PENT") and sustained pressure testing at elevated temperatures show similar results for TR-130 and TR-418N. See Exhibits 12 and 32. Rapid Crack Propagation ("RCP") was tested on six-inch Uponor pipe using the Small-Scale Steady State (34") laboratory testing equipment, and results were similar for pipe made from TR-130 and TR-418N. Actual pipe from the Keyspan site was tested and confirmed to have good resistance to RCP. See Exhibit 33. Dr. Walter Bradley's work (see the following text for a description of Dr. Bradley and his work) also showed similar RCP results for pipe made with TR-130 and TR-418N; his report states: "Izod testing of razor notched specimens indicates that the resistance to rapid crack propagation of pipe made from TR-418N resin and TR-130 resin is essentially the same." Standard MDPE molded fittings were tested for compatibility with pipe made from TR-130 per the Department of Transportation's regulations for polyethylene fittings for gas applications. This included butt fusion, socket fusion, saddle fusion and electrofusion fittings which were fused to TR-130 pipe and then tested according to the DOT standards with excellent results. See Exhibits 10, 11, 18, 19,20. - In addition to comparative testing of TR-130 pipe and TR-418N pipe, CPChem retained a highly respected plastics expert, Dr. Walter Bradley, a retired member of the Texas A&M University faculty, to examine and analyze the actual cracked Keyspan pipe and to compare performance capabilities of TR-130 pipe with TR-418N pipe. Dr. Bradley's report is. attached to this Memorandum as Exhibit 34 (his report is referred to in the remainder of this Memorandum as the "Bradley Report"). Dr. Bradley's analysis led him to conclude that pipe made with ,TR-130 should perform as well as pipe made with TR-418N, and that the cold CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED installation temperatures on the day of the failure and impact of some type in the handling of the See Bradley pipe undoubtedly contributed to cause the cracking of the six-inch Keyspan pipe. Report at p. 10. Taken together, the available data conclusively demonstrate that the Keyspan pipe did not crack as a result of any inherent unsuitability of TR-130 for gas pipe applications.. To the contrary, it appears that the Keyspan pipe cracked due to a combination of several factors: o 0 First, information provided by Keyspan indicates, and Dr. Bradley's analysis confirms, that the cracked pipe was likely subjected to impact at some point following production. It is unlikely that pressure alone would have been sufficient to cause the cracking. Second, the installation occurred in extremely cold (sub-freezing) weather in Boston. Laboratory testing subsequently revealed that the only abnormality of was its lower than normal impact strength at very low Third, it appears that only six-inch pipe made during a narrow production window, on a single production line when TR-130 was being used, is susceptible to cracking when faced with the unique conditions identified above. The absence of any problem in the laboratory or in the field with any other size pipe, together with the data demonstrating the essential equivalence of the resins, strongly suggests that any conceivable issue was limited to a narrow production window of six-inch pipe.3 2 As the data attached in Exhibit 6 confirms, the pipe Keyspan received had slightly lower thermal stability readings than tested samples of pipe manufactured on other production lines and pipe made prior to the mistaken delivery. However, thermal stability tests performed on all sizes of pipe manufactured with TR130 and on TR-130 resin confirm that all the pipe and the resin comply with ASTM standards for both pipe and resin (i-e., > 220°C). 3 As we have stated in this Memorandum, the only pipe that has cracked at cold temperatures - whether in the field or in laboratory tisting - is six-inch pipe made on a single production line. By contrast, six-inch pipe made with TR-130 on other production lines withstood cold temperature impact. Uponor reports that there were no abnormal operating issues associated with production on that line at the time. Uponor therefore believes that the explanation may lie in the fact that TR-130 and TR-418N have somewhat different stabilization systems, in terms of specific grade of antioxidant and additive level. Both formulations provide a high level of thermal stability which exceeds the applicable requirements of thermal stability set out for both the resin (ASTM D3350) and the pipe (ASTM D2513), i.e. >220°C. Both resins contain a high performance primary antioxidant (hindered phenolic additive) and a secondary stabilizer (phosphite additive). Nonetheless, Uponor reports that the production line from which the pipe that cracked originated, while optimized for TR-418N, may not have been optimized for extruding TR-130 into thick-walled & six-inch) pipe given the different stabilization package. This, together with extreme cold temperature installation and impact, could have .played a contributing role in the cracking of the six-inch CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED D. TR-130 Has Received A Hydrostatic Design Basis Rating From The Plastic Pipe Institute When the resin mis-shipment was discovered, CPChem initiated the required steps to obtain what is known as a Hydrostatic Design Basis, or "HDB," rating fiom the Plastics Pipe Institute ("PPI") for TR-130 resin. As explained more hlly below, one of the regulatory standards applicable to gas pipe, ASTM D2513-96a, requires that the resin used to make such pipe has a "PPI recommended long term hydrostatic stress rating." TR-130 was not offered for sale by CPChem for gas pipeline applications, and thus CPChem had not previously sought a PPI rating for TR-130. Nonetheless, because of the essentially equivalent physical characteristics of TR-130 and TR-418N resins, CPChem was confident that TR-130 would qualify for the required PPI rating. On May 4, 2001, PPI did in fact grant an E-2 Hydrostatic Design Basis ("HDB") rating for TR- 130 resin.4 11. Exhibit 35. PIPE MANUFACTURED WITH TR-130 COMPLIES WITH APPLICABLE LAWS AND SAFETY STANDARDS IN ALL RESPECTS, AND DOES NOT REQUIRE A WAIVER At the time of the January 24, 2001 and March 14, 2001 customer communications, CPChem was not in a position to state that the pipe complied with all applicable regulations for two reasons. First, at that time PPI had not granted the HDB rating to TR-130. Second, CPChem had been unable to confirm that the pipe manufactured with TR-130 contained UV stabilizer - an ingredient required by the regulations. See discussion at pp. 12-14 pipe in question. Without taking a position as to Uponor's explanation, CPChem can confirm that its TR130 resin contains a different stabilization system, and that the only incidents of abnormal laboratory or field performance appear isolated to the six-inch pipe produced by Uponor on this single line - when Uponor was unwittingly extruding TR-130 rather than TR-418N. In any event, the issue should not concern Uponor's customers because all six-inch pipe has been recalled. 4 . On October 25, 2001, PPI extended the E-2 HDB rating to an E-6 HDB rating based on additionaI data submitted by CPChem. CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED of this Memorandum. However, after extensive analysis and fact development, and now that PPI has granted the HDB rating, we are confident that in fact the pipe does comply with applicable regulations. Accordingly, owners/operators do not need to obtain a waiver to lavdully operate .. gas pipeline facilities made with TR-130 pipe. The U.S. Department of Transportation's Office of Pipeline Safety is charged with regulating gas pipeline safety. 49 U.S.C. $60102. The Office of Pipeline Safety has, in turn, promulgated extensive regulations carrying out this mandate. See generally 40 C.F.R. Part 192. The regulations require that materials used to manufacture plastic pipe for natural gas transport meet a "listed specification." 49 C.F.R. Part 192.59. In this case, ASTM D2513-96a, "Standard Specification for Thermoplastic Gas Pressure Pipe, Tubing, and Fittings," is the relevant listed specification. A copy of ASTM D25 13-96a is attached as Exhibit 36. Section 4 of ASTM D2513-96a requires that the plastic used to make gas pipe "shall have a PPI recommended long-term hydrostatic stress rating." As stated previously, "PPI" refers to the Plastic Pipe Institute. TR-130 recently received the required rating. See p. 9, above. Annex A1 to ASTM D2513-96a contains many different materials, testing, and performance requirements for polyethylene gas pipe such as the pipe at issue. Section A1.3.1 requires that polyethylene ("PEW)material be classified in accordance with ASTM D3350 "Cell Classifications of Polyethylene Pipe Tubing and Fittings Materials." ASTM D3350 classifies PE pipe and fitting compounds according to: density, melt index, flexural modulus, tensile strength at yield, environmental stress crack resistance, and HDB at 23OC. These categories describe the physical properties of materials that fall within certain cell classifications. Exhibits 3, 4, and 5 to this Memorandum demonstrate that TR-130 pipe satisfies these requirements. ASTM D2513-96a section A1.3.4 requires either Class C (carbon black) or Class E (colored with UV stabilizer), with Class E the appropriate requirement for yellow pipe. See Section 6.2 of ASTM D3350, Exhibit 37 hereto. Neither ASTM D3350 nor ASTM D2513-96a requires a particular amount of UV stabilizer for qualified PE materials. The pipe in question meets this requirement, even though TR-130 resin does not contain UV inhibitor, because: CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED The yellow color concentrate used by Uponor is made with TR-418N, which cbntains UV inhibit~r,~ Furthermore, the pipe made by Uponor with TR-130 would have gained additional W inhibitor from the introduction of regrind material containing pipe made on previous production runs from TR-418N resin. Therefore, all pipe manufactured with TR-130 contains some amount of UV inhibitor. 'While the amount of UV inhibitor in the pipe made with TR-130 is lower than that in Uponor's TR-418N pipe, the pipe is "colored with UV stabilizer" and therefore complies with the applicable ASTM ~tandard.~ 0 In sum, the pipe in question complies with the applicable regulations and does not require a waiver. 111. ANY PERCEIVED NON-COMPLIANCE WITH APPLICABLE REGULATIONS SHOULD BE WAIVED The foregoing discussion and accompanying data demonstrates the likeness of TR-130 and TR-418N resins. Even more significantly, the data confirm that the pipe made with TR-130 should perform in an equivalent manner to pipe made from TR-418N and poses no threat to health and safety. CPChem believes that pipe manufactured with TR-130 complies with all applicable laws, regulations and specifications and therefore does not require a waiver. Nonetheless, if - presumably because of the lower level of UV inhibitor in the pipe in question any of Uponor's customers prefer to seek a waiver or other affirmation from the appropriate regulators, this Memorandum may be used to demonstrate that granting such a waiver is wholly consistent with pipeline safety. - 5 The yellow color concentrate contains approximately 91% TR-418N and 9% pigment. 6 W stabilizer is also the subject of section A.1.5.7. of ASTM D2513-96a. That section states: PE pipe stored outdoors and unprotected for at least two years from the date of manufacture shall meet all the requirements of this specification. PE pipe stored outdoors for over two years from date of manufacture may be used if it meets the requirements of this specification (emphasis added). The pipe at issue in connection with the waiver applications was only stored outdoors for a short period of time, well less than two years, and thus the foregoing provision would not appear to apply. CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED At the outset, it is important to note that the Recall program addressed all .. uninstalled pipe (by pipe that might have been exposed to the sun for longer periods of time) as . . the issue of obtaining a waiver for any arguable nonwell as all installed six-inch pipe. Thus, , compliance should properly only focus on the remaining installed pipe (the pipe that was not covered by the Recall program), none of which is six-inch pipe. Here, the pipe not covered by the Recall program would have been installed within a relatively short period after manufacture. Specifically, MDPE pipe manufactured by Uponor with TR-130 was installed, at the latest, in January 2001. After that date, Uponor's customers received the January 24 letter and were instructed to stop installation of the Uponor - pipe. Therefore, given the manufacture dates of July 21 23, 2000, the maximum amount of time.the affected pipe could have been exposed to W prior to installation was six months. In reality, much of the pipe was installed within a few months of delivery (before the onset of cold weather), and in any event common sense would suggest that each section of pipe was not fully exposed to the sun for the entire time period between shipment by Uponor and installation. Thus, the reference in the ASTM standards aimed at protecting the structural integrity of pipe that is stored outside for at least two years prior to installation is not implicated.' In fact, given the presence of W inhibitor in the pipe and given the fact that the yellow pigment itself contributes to prevent UV degradation, any such pipe should withstand any modest outdoor exposure that might have occurred. See Bradley Report at p. 1 (the reduced amount of W stabilizer "would only be significant if the gas pipe was exposed for some extended time to radiation from the sun, which was not the case for this pipe"). This is confhned by weathering tests performed by CPChem. Yellow pipe extruded by CPChem with - 7 -- -- See ASTM D25 13-96a Section A1 S.7. - CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED TR-130 was subjected to extreme outdoor exposure at a special facility in the Arizona desert. . . Analysis of those pipe samples exposed for four months - including the intense W summer months - revealed no notable chpnge in tensile strength or elongation of the pipe as a consequence of the outdoor exposure. See Exhibits 21 and 22. Furthermore, in addition to the Arizona testing - in which pipe samples were exposed to actual sunlight, albeit in an extreme "worst case" environment - CPChem also conducted artificial accelerated weathering tests. Exhibit 23 and 24. Not surprisingly, those test results did show differences in the tensile yield strength and ductility at higher numbers of hours of accelerated weathering. However, such results are not by themselves meaningful, because they do not correlate to actual periods of outdoor exposure. In fact, based on the relative performance in accelerated weathering tests of pipe made with TR-418N and pipe made with TR-130, pipe made with TR-130 is safe for use even after exposure for six montJs8 Test data received to date confirm this conclusion. IV. OPERATIONAL CHANGES TO PREVENT FUTURE MISTAKES For the reasons set out previously in this Memorandum, we do not believe that the Keyspan pipe cracked due to any inherent difference between TR-130 and TR-418N. In any case, the resin mis-shipment should not have happened, and CPChem has taken steps to insure that such a mistake does not happen again. The mistaken delivery of TR-130 to Uponor occurred when the hopper car containing TR-130 was shipped to Uponor rather than the hopper car containing TR-418N. CPChem has subsequently implemented three major changes in the operation of its Houston -- 8 - As shown in Exhibits 23 and 24, the TR-130 samples showed a drop in elongation to under 100% at 500 1121 hours compared to 3065 hours for TR-418N. CPChem generally quotes 3-5 year outdoor life for TR4 18N. Thus, if one assumes 4 years of outdoor life for TR-4 18N and a ratio of 500 to 3065 for the relative comparison of TR-130 to TR-418N, that yields a relative outdoor life of 7-8 months for TR-130. . . CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED Chemical Complex ("HCC") aimed at preventing another delivery of the wrong resin to Uponor. First, an additional person is now responsible for monitoring the loadout of the hopper cars used to deliver the resin. The operations shift supervisors now go to the loadout spots associated with each of their plants and record which railcars are on a particular spot and which resin is loaded into each hopper car. This information is then cross-checked with the paperwork that the hopper car loaders (distribution attendants) have previously filled out in order to verify that the information is correct. The Certificate of Analysis that accompanies the shipment is based on the information in the cross-checked paperwork, so the cross-checking procedure insures that the Certificate of Analysis has the proper railcar and resin recorded on it. Second, when the loaded hopper car arrives at the CPChem scales, the scale attendant removes one aluminized tag off one car at a time and verifies that the car number on the tag matches the railcar number. Previously, the attendant pulled tags off all the cars in the string and then matched them with numbers in the computer; this procedure did not provide a cross-&eck against two switched railcars since both would probably be in the string, Third, HCC no longer schedules the production of pipe resins and similar resins on sister reactors at the same time. Thus, any differences would be very obvious and caught before pipe could be made. These operational changes aim to insure that the labels for a hopper car are properly matched with the contents therein, and to confirm that Uponor receives the proper resin. Although Uponor could not have prevented CPChem's mis-shipment of resin, Uponor has informed us that it has implemented new quality control procedures to supplement its procedures in place in July 2000. In particular, Uponor advises that it now samples resin from every compartment of every rail car delivered to its Tulsa facility, and that each sample is then CONFIDENTIAL AND PROPRIETARY TREATMENT REQUESTED subjected to Fourier Transform Infrared Analysis ("FTIR") testing for W level as well as density and melt index9 testing. V. CONCLUSION CPChem believes that MDPE pipe manufactured with TR-130 complies with all applicable laws and regulations and does not require a waiver. Furthermore, the foregoing discussion establishes that MDPE pipe made with TR-130 still in service should perform in the same manner as MDPE pipe made with TR-418N. The two resins possess essentially equivalent physical characteristics, and pipe manufactured using the two resins should perform in an identical manner in field use. Moreover, even if a utility were concerned by the lower level of UV stabilizer in pipe made with TR-130, that condition - as established above - is immaterial to the durability, performance, and safety of the installed pipe in question.. Under these circumstances, we respectfully submit that the pipe at issue poses no safety hazard and is wholly appropriate for use in natural gas transport. 9 Density and melt index testing alone are inconclusive for differentiating between TR-130 and TR-418N because both the density and the melt indexes specificationranges of the two resins overlap. CONFIDENTIAL AND PROPRIETARY . TREATMENT REQUESTED Dated: November 12,2001 CHEVRON PHILLIPS CHEMICAL COMPANY, LP BY: Timothy G. T* Senior Vice President, Olefins and Polyolefins Chevron Phillips Chemical Company, Chevron Tower 1301 McKinney Houston, Texas 77010-3030 Heather M. McPhee Latham & Watkins 555 Eleventh Street, NW Suite 1000 Washington, D.C. 20004 January 24,2q01 Csmpany hhne Xttn: Mr. Customer 123 Main Street City, State Zip Dear Mr. Cusiorner: All of Uponor's quaiii-y c ~ n i r o lrecords verify that the gipe made wiih this diii~rentresin passed all dirn~nsionaiand required tests specified by ASTM 0-2512. EowE\J~:,durifio insiallation, one Uponor cus:omer found that some 3i the siiected pipe aid not p s s s pressure tesiing in temperature conditions srouna 30 degrees F~ihrenheii. Specifically, cracking was discovered in ceriain placas on the pipe. The exact csuse of the crxking is not known. W e have not determined ihat the faiiure was relaied to the type of resin usad in the pipe. Other sampies of the affected pipe de!ivered to the cus:orner passed all standard pressure testing. Chevron Phillips has been provided samples of the failed pipe along with control samples and has conducted significant testing. So far, the analysis by Chevron Phillips indicates that the resin used was probably HHM TR-130 resin, MDPE ("TR-130").The density and meit index of TR-130 are very close to those of TR-4i8N, and Chevron Phillips believes the nominal perforrnsnca cnaracieristics of TR-130 after extrusion should be substantially similar to those of TR318N. We are at a loss to expiain the reported failure of the pipe. The anelysis to date has Seen based on srnsll rings of pipe, and Chevron Fhillips is conduciing more thorough evaluations ?o include seiecied impact and pressure tesiing. We ha\,e discussad commenting UpGn t h ieiaiive ~ seriousness of ihis problen 2nd making a recommendation to U~onor'scustomers. Althougn Chevron Phillips is 2 resin manufacturer 2nd Uponor is well versed on exirusion of polyethylene resins, *we are noi experts on natural ass pipe systern insiallation and operation. Moreover, the conditions of installation and operation of pipelines vary from geographicsi !omtion to locsiion as well 2s from utility to utility. Therefore, we do nor believe it is possible to formulate a comment ar recommendation that would be applicabie to all customers and all locations. As experiencad users of polyethylene as well as many other types of gas piping systems, you ;~iilhave specific knowledge as to the installation procedures and operating conditions for your systems. Your company is the best judge of the seriousness 13i this siiuation ^as it applies to YOU. :Ne are activeiy invesiigating this matter and wiil provide additional teshnicsi information as it becomes avaiisble. In the meantime, if you have any quesiicns at all, piease feel to contact your account representative. Yours very truly, March 14,2001 T. G. Taylor Senior Vice President Ojefins & Polyolefins ~ e i m b u r s e m e nOfice t Address: 1301 McKinney Suite 1299 Houston. TX 77010-3031 Reimbursement Ofice Phone: (866)274-6799 Reimbursement Office Fax: (713)289-4979 Via Facsimile and CRMM# Re: Chevron Phillips' Reimbursement Program for Uponor Pipe Dear Mr. : On January 24, 2001, Chevron Phillips Chemical Company LP (CPC) and Uponor AIdyl Company (Uponor) sent you a letter explaining events that led to the discovery that some medium density polyethylene pipe manufactured by Uponor during the period July 20 through July 23, 2000 may have contained Chevron Phillips' MarlexB HHM TR-130 resin (TR-130) rather than Chevron Phillips' MarlexB HHM TR-418N resin (TR-418N). That letter stated that further details concerning appropriate customer response to this situation would be supplied. This letter updates and supercedes the January 24 letter. CPC is prepared to reimburse you, on behalf of Uponor, for the reasonable and necessary costs of abandoning or removing and replacing six-inch pipe made on July 21 and 22. 2000. At this time, we are not recommending replacement of pipe made on July 20 or July 23 that has been installed and CPC will not reimburse you for such costs. Nevertheless, CPC will repurchase, a t your invoiced cost, any Uponor pipe made between July 20 and July 23,2000 that has been paid for but not yet installed. CPC will also pay the freight charges necessary to return the inventoried pipe. Additionally, to verify the parameters of the reimbursement program, we are requesting Uponor customers with sixinch pipe manufactured on July 20 and 23, 2000 to send product samples to CPC (see Sampling Procedures Attachment). The reasons for limiting our reimbursement program to six-inch pipe made on July 21 and 22 are straightforward. First, w e have only been notified of a problem by a single customer with six-inch pipe made with .TR-130. This customer experienced cracking while pressure testing during cold winter installation. Second, current information indicates that TR-130 was primarily used to produce virgin six-inch pipe . . I . Page 2 on only'July 21 and 22, 2000. Third. egensive technical.informationconcerning the TR-418N and TRI 130 resins shows that both resins' possess essentially equivalent physical characteristics. See Technical Attachment Although TR-130'does not contain UV inhibitor, this does not present a concern with installed pipe because such pipe was not exposed to sunlight for more than a few months. With respect to installed pipe other than six-inch pipe, CPC has commenced a program of testing to obtain a Hydrostatic Design Basis (HDB) rating for TR-130 resin. We are keeping the DOT informed of our testing program and are confident that TR-130 will obtain an HDB rating. Once the HOB rating is granted, CPC will request that you file an application for a waiver covering the installed pipe. , Records show that your company received the following pipe manufactured by Uponor: 1 X" pipe lot nos. T04-072100, T04-072200 and T04-072300, manufactured on July 21, 22 and 23, 2000, respectively; two-inch pipe lot no. T03-072200, manufactured on July 22, 2000; four-inch pipe lot no. T11-072300, manufactured on July 23, 2000; six-inch pipe lot nos. T08-072000 and T08-072100, manufactured on July 20 and 21, 2000, respectively; and %" pipe lot no. T09-072100, manufactured on July 21, 2000. Accordingly, your company is eligible to participate in the reimbursement program a s to six-inch lot no. T08-072100. CPC asks that your company participate in the pipe sampling program with respect to any installed pipe from lot no. T08-072000. Pipe from any of the above lots that has not been installed is also eligible for repurchase. Furthemore, to the extent you incurred reasonable and necessary removal, storage, or shipping costs with respect to any of this Uponor pipe before the date of this letter, we will consider your application for reimbursement Adherence to CPC's Reimbursement Program Guidelines (attached) is a condition to payment. As an administrative convenience CPC will, in certain circumstances, make payment directly to you, on behalf of Uponor. In order to clarify the reasonableness or necessity of expenses incurred, we ask you to submit all removallabandonment plans to CPC before the work is undertaken. The attached Reimbursement Pre-Approval Form should be used for this purpose. We have established a reimbursement office and engaged experts to assist with evaluation of pro~osedreplacement programs. A toll-free hotline has been set-up in order to respond to any questions that you may have. Please feel free to contact u s at (866) 274-6799. We appreciate your patience with respect to this matter. We hope the foregoing and attached information is of assistance. We stand ready to cooperate with you. Very truly yours, Timothy G. Taylor Senior Vice President Chevron Phillips Chemical Company LP Cc: Tom Sheridan, President Uponor Aldyl Co. Attachments: Reimbursement Program Guidelines Sampling Procedures Attachment Technical Attachment . . SAMPLING PROCEDURES Pipe Sampling and Transmittal Instructions for Six-Inch Pipe Manufactured on July 20 and 23,200C 1) Locate pipe and record pipe information. The pipe print line contains sufficient information for documentation purposes and should be recorded. Additionally, if available, provide the coil number andl or foot number. 2) Using a tool such as a carpenter's plane or other scraping device, scrape or shave off a thin sample fiom the outside diameter surface of the pipe. ApproximateIy 0.01-0.02 inch thickness on 2" Outside Diameter (OD) or larger pipe is sufficient. CAUTION: DO PiOT EXCEED TEN PERCENT (10%) OF THE WALL THICK?'4ESS WHEN TAKIXG A SAMPLE. REMOVAL OF MORE THAN 10% OF THE WALL RENDERS THE PIPE UNSERVICABLE, REQUIRING REPAIR. ALSO NOTE: THlS METHOD OF SAL\IPLZNG IS XOT RECOI\/IMENDED FOR TUBmG. The total amount of the sample needs to be approximately the size of a quarter or a large postae sramp, and can be comprised of several pieces. 3) When transmitting the sample to CPC, pleaje include the print line information together with the pipe size, and SDR pipe location by street location, city, and srate. Also include the date of sample collection, the name of the person who collected the sample, and the name of the person transmitting the sample. 4) 5) ' Please identi@ your company by name, address, and phondfa~.In addition, please supply the name of a designated contact person with whom we can communicate if we have any questions. We will convey our test results to your designated contact person. Please ship all samples to: Chevron Phillips Chemical Company LP Houston Chemical Complex 1400 Jefferson Road Pasadena, TX 77501 Attention: Austin Wuu 6) Please direct any questions about this procedure to: Reimbursement Office 130 1 McKinney Chevron Tower Room 1299 Houston, TX 77010 (866)274-6799 (7 13)289-4979 ( f a ) Technical Attachment TR-418N TR-I 30 * Density (glcc) 0.936-0.940 0.938 Melt index (911 0 min.) 0.1 6-0.24 ** 0.25 Catalyst SAME Molecular Weight Distribution SAME Tensile Strength, psi Elongation, % . 2900 >800 Flexural Modulus, psi 86,000 (D3350). PENT, hours SO0 Melting Point (OC) 127 Additive Formulation Primary Secondary YES YES YES YES YES (0.15 0.21) - Pipe Quick Burst Ductile Ductile ASTM D3350 Thermal Stability (OC) >220 * Lot used in pipe Uponor's six-inch pipe ** Historically 0.15 - 0.25 REIMBURSEMENT PROGRAM GUIDELINES March 14,2001 ELIGIBLE PIPE: Six (6jinch diameter pipe produced by Uponor Aldyl Company ('Vponor") on July 21,2000 and on July 22,2000 ("eligible pipe"). REIMBURSEMENT PROCEDURE: CPC will reimburse you for the reasonable and necessary costs you incur for the replacement of eligible pipe. Following the replacement of the eligible pipe, please send us a letter on your letterhead applying for reimbursement. Please attach to.the Ietter sufftcient supporting documentation for CPC to process the application, showing with reasonable detail the various components of the reimbursement application, including: location, identification of the eligible pipe removed, equipment rental, and labor costs. This documentation should take the f o m of invoices, work orders (for internal costs), or, for example, rental contracts for rented equipment. PLEASE NOTE: corporate overhead, capital cost recovery, supervisor or manager time or legal consultations are not subject to reimbursement under this program. Unless pre-approved in writing, CPC shall have the sole authority under this program to determine the reasonableness and necessity of charges. PRE-APPROVAL PROCEDURE SUMMARY: CPC encourages applicants to provide advance notice that replacement of eligible pipe is beginning or has begun at your location(s). You can fax the attached Reimbursement Pre-Approval Form to us at our office (fax: 713-289-4979). CPC is using the information to ensure that we have sufficient resources available to provide any sup?ort you may need. CPC's written approval of a Reimbursement Plan will also confirm CPC's acceptance of the reasonableness and necessity of the activities reflected in it. PROGRESS PAYMENTS: CPC recognizes that certain applicants might require advance or "progress" payments. We will entertain requests that CPC provide progress payments for applicants requiring such assistance. In order to facilitate review of such requests, applicants requiring advance or progress payments should make the req~est using the attached Reimbursement Pre-Approval Form. Please note that prior to issuing an advance or progress payment, CPC will require a certification that the funds will be used only for the replacement of eligible pipe. All other requirements of these guidelines must be met - B C E I P T AND RELEASE: In order to ensure that any and all b d s advanced in this program are used exclusively for the purpose intended, that such disbursed funds have been so used, and that the eligible pipe will not enter any secondary market, CPC will require submission of a receipt, release and warranty sufficient to provide such assurances. In addition, to assure the propriety of a11 payments hereunder, CPC reserves the right to audit the records of any recipient of advance, progress or reimbursement payments hereunder for a period of two (2) years following the fmal payment CPC applicant at any time further reserves the right to request additional information from concerning the purchase, use, replacement or disposition of any eligible pipe, and applicants agree by participating in this program to provide documentation on request. REIMBURSEMENT PRE-APPROVAL FORM CUSTOMER NAME: . . ADDRESS: CONTACT PERSON(S): PHONE: . . . FAX: E-MAIL: REPLACEMENT WORK SITE: DESCRIPTION OF ELIGIBLE PIPE TO BE REPLACED: (Size, Series, Lot #) -- TOTAL NUMBER OF FEET OF ELIGIBLE PIPE: DESCRIPTION OF FIXINGS REQUIRING REPLACEMENT: TYPE OF INSTALLATION: SPECIAL PROBLEMS: (e.g. paving, landscaping, etc.) METHOD OF REPLACEMENT: ESTIMATED COSTS MATERIAL: - - CONSTRUCTION: EQUIPMENT RENTAL: OTHER PROPOSED COSTS: (explain) TOTAL PROPOSED COST FOR THIS SITE: ~ n & s i sof Uponor Pipe Prepared for Chevron Phillips Chemical Company LP 1301 McKinney Suite 3400 Houston, Texas 77010 Prepared by &&&A. * 7 Walter L. Bradley, ~ h . b . ,P I . October 9,2001 Introduction Four sections of six-inch polyethylene natural gas distribution pipe cracked during routine pressure testing after installation in Boston in December 2000 in weather that was below freezing (5F-30F). It was subsequently determined that these four failed sections of pipe had been made from TR130 rather than TR418N, which was the specified medium density gas pipe resin. Further investigation documented that on several days in July 2000, certain gas pipe made at the Uponor plant was made with TR130 resin. I have been asked to compare the impact resistance of pipe made with TR13O resin to pipe made with TR4l8N and also to do a failure analysis on one section of the pipe that cracked, failing pressure testing after installation but prior to being put into service in Boston. This report summarizes the results of this investigation. Background Information Chevron Phillips Chemical Company LP has provided me the following basic information about TR418N and TR130 resins. TR130 and TR418N are very similar polyethylene resins, satisfying D3350 Cell Classification. They both result in medium density extruded polyethylene pipe. The natural TR130 resin at issue had a density of 0.938, compared to a manufacturing specification of 0.936-0.940 for TR418N 0 3 3 5 0 requirement: 0.925-0.940). The melt index of the TR130 at issue was 0.25 compared to a manufacturing specification for TR418N of 0.16-0.24 03350 requirement: 0.15-0.40). This indicates that the TR130 and the TR4l8N are very similar in viscosity, and therefore, should process in a similar manner. The weight average molecular weight for the TR130 was 204,000-208,000 with a ratio of weight to number moleculk weight of 11.1-12.8, while the TR418N values were 207,000 and 11.8. One would expect the mechanical properties for these two resins to be quite similar, and they are. The flexural modulus for each was -86,000 psi (D3350 requirement: 80,000-110,000 psi). The tensile yield strength for both was -2800 psi (D3350 requirement: 2600-3000 psi). The thermal stability temperature in degrees C ranged from 238-258 for TR130 and fiom 243-268 for TR418N @I3350 requirement: 220 minimum). The tensile elongation for both resins was -870% with a brittleness temperature of less than -75C (D3350 requirements: >500% and less than 60C). Finally, the environmental stress cracking resistance (ESCR) as measured by the PENT test gave values of >3000 hours for both the TR130 and TR418N 0 3 3 5 0 requirement: >I00 hours). It is worth noting that the PENT specimen for the TR130 was remolded from a section of pipe that failed the pressure testing and precipitated this investigation. The only significant difference in the two resins is that the TR418N has oxygen and UV stabilizers whereas TR 130 has only oxygen stabilizers. However, the coloring package used by Uponor contains some W stabilizer, though less than that in the TR418N resin. This difference would only be significant if the gas pipe was exposed for some extended time to radiation from the sun, which was not the case for this pipe. Experimental Procedures Materials Provided by Chevron Phillips Chemical Company LP for this Study - The following materials were provided for this investigation: 0 0 Two pieces of pipe (-18 inches in length) that were cut from the several pipes that failed pressure testing in Boston in December 2000; Two-inch and six-inch diameter pipe made fiom TR130 resin and fi-om TR418N resin; Pieces of TR130 pipe that was broken at low temperatures in a Dart test performed at Chevron Phillips. Fractographic Analysis -- Two portions (-15 inches) of 6-inch pipes that failed during pressure testing were provided for inspection. These were carefully sectioned to open up the cracks and permit inspection of the fractured surfaces. The fractured surfaces were carefully examined using a binocular microscope at magnifications from 10X to 70X. Smaller pieces (-1inch cubes) were cut fiom the fracture surfaces of the two pieces of pipe for examination in a scanning electron microscope. These pieces were then sputter coated with a gold-palladium film and examined in the scanning electron microscope. The scanning electron microscopy inspection was performed on a JEOL 6400 SEM in the Electron Microscopy Center at Texas A&M University. Izod Impact Testing -- Since the failure of the pipe during pressure testing was ostensibly by rapid crack growth (as opposed to bulging and rupture, or stress induced slow crack growth), sharply notched Izod impact specimens cut fiom 2 inch and 6 inch pipe made fiom TR418N and TR130 resins were tested. The TR130-T08-072200 pipe used was made on July 22,2000. This pipe is fiom the same line and was made on the same day as the pipe that failed during pressure testing in Boston and the pipe that broke in the Dart testing at Chevron Phillips. The orientations of the Izod specimens cut fiom the pipe and tested are indicated in Figure 1. The specimens were 0.5 inches by 0.25 inches, and 5" long. To introduce notches/cracks in the Izod specimens that were as sharp as possible, a machine that used a serrated razor blade as a cutting tool was utilized. The cracks were cut to a depth of 20% of the dimension of the specimen in the direction of the cut. Cracks labeled A were cut across the 0.5-inch dimension to a depth of 0.05 inches, leaving an un-cracked ligament of 0.20 inches. Cracks labeled B were cut to a depth 0.1 inch, leaving un-cracked ligament of 0.4 inches. The test results fiom the specimens with the cracks in the B orientation were more representative of crack growth in pipes, whereas the specimens with cracks in the A orientation sometimes failed by loading the uncracked ligament beyond its yield strength, giving a very ductile, overload type of failure. Thus, only the comparisons for specimens with B orientation cracks are significant. I Figure 1. Specimen Nomenclature Finally, crack growth in the specimens with orientations 0 (outside diameter), I (inside diameter), and R (radial) is in a plane that cuts through the cross-section of the pipe. Pipe failures for pressurized pipe, however, almost always result fiom cracks that grow along the axis of the pipe, including the pipe that failed in Boston. Thus, the most relevant test results are comparisons fi-om specimens with the H orientation. Unfortunately, Izod test specimens could not be cut fiom the 2" pipe. In fact specimens of only one orientation could be cut from the 2" pipe; namely, through-wall specimens, which are labeled 0+1 in Table 1. All test results are for Izod testing at -40C. Experimental Results Izod Testing Results - The test results for the razor notched Izod tests are summarized in Table 1. First, specimens with A-type crack orientation give results that are often very large compared to specimens with cracks with B-type crack orientation. As previously noted, the mode of failure in the A-type specimens sometimes by tensile overload rather than by crack growth. Thus, these results are not significant for our purposes because the four Boston pressure-testing failures were by crack growth, not by tensile overload. Thus, our analysis of the data in Table 1 will focus on only tests for which the crack orientation is B-type (see Figure 1). Second, the specimens of TR418N with cracks in the B orientation for specimens 0, I, and R had very similar Izod impact energies, implying that the resistance to crack growth (test groups 4,6, and 8) is the same in each direction in the plane perpendicular to the axis of the pipe (zadiaVcircumferentia1plane). Third, Izod impact energies for test specimens cut fiom six inch pipe made fiom TR418N resin and TR130 resin are both 5.7 KT/m2, indicating that the difference in resin did not produce any significant difference in resistance to rapid crack growth in the circumferential direction (test groups 2 and 4). However, the most meaningful comparison of the resistance to rapid crack growth of TR418N and TR130 resin made into six-inch pipe is for H specimens with razor notching in the B orientation (see Figure 1). The results seen in Table 1 for test groups 16 and 18 indicate no difference in resistance to rapid crack propagation. Note that such a comparison for 2-inch pipe could not be made because suitable specimens could not be cut with this orientation in 2-inch pipe. Fourth, two-inch pipe made of TR130 resin by two different manufacturers (see group numbers 10 and 14 in Table 1) indicated a very similar resistance to rapid crack propagation as measured in a razor notched Izod test to two-inch pipe made of TR4l8N. Table 1. Summary of Izod Test Results on Razor Notched Specimens (See Figure 1 to clarify nomenclature in Table 1.) Specimen Crack TYpe Type A R B R A R B R A I B I A 0 B 0 A O+I B 0+1 A 0+1 B 0+1 A O+I B O+I H A B H A H B H 0 Material Pipe Fracture Energy Test Group Dia (in.) KJrn2 Processing Number 6 TR130-T08-072200 24.3" 1 6 TR130-T08-072200 5.7" 2 6 TR4 18N-T08-071900 24.3' 3 6 TR418N-T08-07 1900 5.7* 4 6 TR418N-T08-020601 5.4 5 6 TR4 18N-T08-020601 5.4 6 6 TR418N-T08-020601 4.9 7 6 TR418N-T08-020601 6.0 8 2 TR130-Lot 8 1-9-CPC 25.5** 9 2 TR130-Lot 81-9 CPC 6.0" 10 2 6.1 TR4 18N-T04-0 12901 11 2 5.8 TR418N-T04-012901 12 2 34.0 TR130-TO1-072200 13 2 TR130-T01-072200 5.2" 14 6 9.7" TR130-T08-072200 15 6 TR130-T08-072200 2.8" 16 6 TR418N-T08-07 1900 8.7* 17 6 2.8** TR418N-T08-07 1900 18 *Avg. of 5 tests **Avg. of 3 tests Most important comparisons: 16 to 18 for six-inch pipe, and 10 and 14 to12 for two-inch pipe. Fractography - Boston Pressure Test Failure and Dart Laboratory Failure Macroscopic View Only one of the two pieces of six-inch pipe made of TR130 resin that failed pressure testing in Boston that were provided had the actual crack origin site. My purpose was to see if the cracking started at a manufacturing defect, some impurity in the resin, or possibly at a field weld. The results of the ftactographic examination at low magnification are seen in Figures 2. Figure 2 (top) is from the surface from the failed pipe, showing that the fracture originated adjacent to the weld, not at or in the weld. The smooth region is the region of crack initiation while the regions with lines (so-ca