ML20150A905
| ML20150A905 | |
| Person / Time | |
|---|---|
| Site: | Surry, Farley, 05000000, Trojan |
| Issue date: | 06/30/1988 |
| From: | Beaudoin R NUCLEAR INFORMATION & RESOURCE SERVICE |
| To: | Chilk NRC OFFICE OF THE SECRETARY (SECY) |
| References | |
| FRN-53FR11311, RULE-PR-50 53FR11311-00005, 53FR11311-5, NUDOCS 8807110115 | |
| Download: ML20150A905 (8) | |
Text
.
h
~g D%KET ilUMBER 3D e
PROPOSED RULE 3. IO 2'
(.j'.
g3F4 //3/1) occxtue uwg b
U f
'88 JUL -5 P3 :07 Nuclear Information and Resource Service 142416th Street, N.W., Suite 601, Washington, D.C. 20036 (202) 32ti 0002
- ' '- -[
June 30, 1988 Secretary US Nuclear Regulatory Commission Washington, D.C.
20555 Attn: Docketing and Service Branch
Dear Mr. Chilk,
Enclosed please find the comments of the Nuclear Information and Resource Service concerning the NRC's request for comments on proposing to investigate the safety benefits associated with using leak.-before-break technology for emergency core cooling systems (ECCS) and environmental qualification (EQ) of safety related electrical and mechanical equipment.
Ne thank you for taking the time to consider these comments on the additional application of leak-before-break technology.
Sincerely, Robert Beaudoin Information Servicee 8007110115 e00630 oDR PR PNV go'S3FR11311 y
dedicated to a.ound non uclear enerc polin.
Mj g
On April 6, 1988, the Nuclear Regulatory Commission published in the Federal Register a "Request for Comments" on the additional appli7ation of leak-before-break technology (53 FR 11311). The NRC contends that it has become possible to exclude the double-ended guillotine break from the dynamic structural design basis because it is "unrealistic and overly conservative in certain situations."
The leak-before-break theory claims that pipes will leak before they break, giving operators ample warning and time to shut down a plant before any large-pi.pe break could cause a major loss-of-coolant accident (LOCA). But what if a crack in a pipe doesn't leak enough to be detected before it breaks? Harold Denton, former Director of Nuclear Reactor Regulation, told the Commissioners in August 1983, "Leak-Before-Break is not an established law." He warned that "... if there is really.a pipe out there somewhere that is cracked half-way through or 55 percent of the way through before it begina to leak, then there is very little time to detect that leakage and take prope.
corrective action."
Any basis for the continued and increased application of leak-before-break to detect severe failure in piping must be related to the various problems associated with pipe thinning, quality, stress, and the technological ability to detect pipe cracking before a pipe break could cause an accident. For the NRC to promote leak-before-break as a means to improve safety is questionable at best, because it presumes that the proceedures and equipment in place aren't functioning prope cly.
The present safety requirements in place for containments, emergency core cooling systems (ECCS), and environmental qualification (EQ) are not necessarily under scrutiny, rather the costs associeted with maintaining these safety standards are considered by industry as a severe cost penalty.
This appears to be another case where the NRC has buckeled under industry pressure involving the inspection of piping systems. An example of this is the release of an NRC IE Bulletin 83-02 (March 1983), which required auamented inspections of welds in the recirculation piping systems and called for more stringent requirements for inspectors. Subsequent inspections at three plants revealed widaspread cracking of recirculation and residual-heat-remcval piping. With the accumulation of this evidence, NRC's Director of Nuclear Reactor Regulation,. Harold Denton, decided to accolorate the inspection program. On July 15, 1983 Denton requested that the Commi7sion vote to shut down five units within 30 days The Commission heeded Denton's advice and voted to shut down the plants for inspection; however, on the very next day the Commission revoked its decision in order to give the industry more time to submit its arguments ar.d to arrange a more suitable shutdown schadule.
't 2
The nuclear industry used the leak-before-break theory to influence the Commissioners, even though some NRC staff members expressed doubts about the trustworthiness of this theory.
This reversal by the NRC was followed up with a request by the Commission that it be informed by the Executive Director for operations on what actions should be taken regarding intergranular stress corrosion cracking (IGSCC) in stainless steel piping at uninspected boiling water reactors (BRR). The staff met with Electric Power Research Institute (EPRI), General Electric, and the affected utilities and came up with these conclusions: "
... measurable pipe leakage should occur before pipe structural failure. However, the leak-before-break concent cannot be the sole basis for continued safe operation." The staff also noted that: "There is considerable uncertainty in the ability to determine the depth of cracks and this uncertainty must be accounted for when establishing the extent of the required pipe crack repairs." (SECY-83-350)
Ultra-Sonic Testing (UT) is used in detecting and sizing cracks in pipes, and there is a great deal of disagreement among the NRC, industry, and the scientific community on the reliability of UT. Additional unresolved issues include: the trade-off between radiation exposure incurred by inspection workers and the problems stemming from any attempts to decontaminate portions of the plant; the validity of the industry's aleak-before-break theory; and the rclative values of a wide range of temporary fixes now in use or under consideration by industry. In fact, the NRC's Advisory Committce on Reactor Safegards (ACRS) called reliance on UT a "delusion" to be taken "on the basis of faith."
Further arguments put forth by industry claim that cracked pipes will leak for a long time before they break. ACRS called this an "unproven hope;" and when utilities suggested that pipes could leak for as long as a year before they break, NRC's Denton countered that "
commission experts believe the time between leak and break could be very brief indeed."
Comments on UT would not be complete without mention of the industry's own evaluation of the ultarsonic technique and of the inspectors involved in the pipe crack analysec. In 1983, EPRI conducted a testing exercise seeking to evaluate the industry's capability for detecting cracks and for making valid depth measurements. Only three of fourteen teams correctly characterized 80 percent of the cracks and six teams racked up an "inadequate" score. Industry continued to argue in favor of UT and EPRI announced that a more advanced UT system would be developed. ACRS mitigated its criticism of UT and called the development of new techniques "promising." However, ACRS also warned that "2quipment and procedures that will allow the reliable determination of the depth of the cracks...are not vet available."
3 i
i
l The method of inservice inspections (ISI) to detect cracked piping is used by utilities to justify the continued operation of plants with cracked piping. However, a number of problems with this inspection process have surfaced: the lack of qualified inspectors, EPRI's inability to upgrade the qualifications of i
existing inspectors, and the rapid "burnout" of inspectors due to I
the high radiation exposure involved in the job. The NRC confidently relys on leak-before-break and ultrasonic testing to deal with pipe crack problems regardless of internal and scientific criticism of these methods. The NRC is reluctant to treat pipe cracks as a safety issue; in fact, the NRC's position can be likened to that of a Commonwealth Edison official, who called pipe cracks a "sound maintenance management issue," "not a safety issue."
An article published by the Energy Laboratory at the Massachusetts Institute of Technology ("Predicting Wear in Steam Pipes.") explains how the cycle of corrosion and erosion reduces pipe thickness. Steam-extraction pipes fail after ten years even though they were designed to last for forty. The use of ultrasonic equipment was considered to be limited and inaccurate because "... a single measurement may not be representative of conditions nearby." The most important claim made in this paper is that "Wear is nonuniform and randomly located, with worn spots occuring besides unworn ones. A single measurement may give an inaccurate picture of a pipe's remaining lifetime." What this means is that even though a pipe might pass inspection, it could have a severe crack in an uninspected section resulting lu the pipe breaking before Jt leaked.
Leak-before-br?ak technolvjy in based on the specious assumption that pipes rcmain as they were designed. NRC BULLETIN NO. 87-01:
Thinning of Pipe Walls in Nuclear Power Plants, addresses the problem of erosion / corrosion in carbon steel piping. This bulletin was written because of the catastrophic failure of a main feedwater pipe at Virginia Power's Surry plant in 1986. Some of the piping in the secondary loop thinned from one-half inch in thickness down to the size of a credit card. A feedwater pipe ruptured, releasing superheated steam which injured eight workers. The severity of this accident was not limited to one area of the plant; in fact, steam released from the pipe activated several fire protection systems, which then adversely affected the air in the control room and the plant's security and communications systems.
The eightcau inch diameter pipe that failed was located in the secondary or "non-safety-related" portion and was not subject to inspection. This might suggest that the leak-before-break theory did not fail because it was not applied. However, this line of reasoning is flawed because the pipe, regardless of inspection, 4
never leaked even though its thickness was greatly reduced. Thus, it is clear that the oice did not leak before it broke but it did tragically kill four workers.
As a result of the Surry accident the GAO published a report:
"Action Needed to Ensure That Utilities Monitor and Repair Pipe Damage." This report addressed the Surry accident and the 1987 discovery of widespread pipe deterioration at the Trojan plant.
Although the surry incident involved non safety-related piping, a
follow-up inspection found a significant amount of pipe thinning in the Trojan plant. Portland General Electric reported excessive erosion / corrosion in both safety and non-safety piping. This marked the first time that a utility discovered extensive Camage in both portions; damage was also found in straight sections of pipe far away from where erosion / corrosion would have been expected. In light of this evidence, it is difficult to understand why the NRC permits and wishes to expand the use of leak-before-break, especially when this epidemic of pipe thinning problems was never included in the NRC's design basis.
The GAO report raises questions about the long-term safety of pipe systems at nuclear power plants. Erosion /corriosion in single phase pipe was not anticipated by the NRC or the nuclear industry. Furthermore, the GAO attributes the Surry accident to a continued condition of erosion / corrosion which was neither regulated nor monitored by the NRC or industry standards. Eecause of the significance of the Surry accident, the NRC required utilities to report on the extent of erosion / corrosion damage at their plants. In February 1987, a survey of 91 plants by the NRC discovered a significant amount of secondary plant pipe thinning and noted "... that utilities do not adequately monitor for pipe thinning or ensure that appropriate corrective action is taken when they find it." The NRC also found that Virginia Power did not have an inspection program to examine the thickness of feedwater pipe systems at Surry. The scenario for inspections which are not carried out and leaks which aro ignortd will only leed to more accidents because leaks will not be de.ected in a timely manner.
.\\lthough the Commission sought more information from utilities on the pipe thinning problem, any regulatory action seemed uncertain at best. In fact, the GAO concluded "
that NRC needs a mechanism to ensure that utilities periodically assess the integrity of pipe systems in their plants to reduce the risk of future injury to plant personnel or damage to equipment caused by j
erosion / corrosion." It is foolish for the NRC to call for the increased application of leak-before-break at a time when the GAO j
j is recommending that the NRC adopt the mandatory _ inspection of all nuclear nower olants for nice deterioration.
i 5
1 1
The NRC uses the "defense-in-depth" concept to assure an adequate level of safety. This concept is implemented by using high standards of design, fabrication, inspection and by providing enough back-up systems to protect against accidents. Robert Pollard, senior nuclear safety engineer with the Union of Concerned Scientists, contends that "... the NRC's lax policy on pipe cracks further erodes the defense-in-depth philosophy."
Pollard also noted that "The traditional safety analysis assumes one break but that multiple breaks could result from weaker.3d piping. In the event of an earthquake, even if the emergency core cooling system responds as it is supposed to, it could be a serious accident."
In August 1983 a top NRC official made a similar analogy in regards to a plant's overall risk profile. He said, " If you have a plant with a large number of weakened pipes you are going to get an elevated level of risk from all kinds of things: for instance, earthquakes. Plants are designed so that an earthquake won't hurt normal pipes, but weakened pipes could break, resulting in a large-break loss-of-coolant accident." This o.ficial also went on to say that the NRC would eventually have to face this problem sooner or later.
6
WHY LEAK-BEFORE-BREAK SHOULD NOT BE EXPANDED:
A safety advisory committee to the NRC called the "leak-before-break" theory a "delusion" stating that "we know of no way to demonstrate its validity." The NRC's reliance on leak-before-break is an irrational response to a very real safety problem.
Moreover, for the NRC to even suggest the increased application of leak-before-break points to the continued lip service which the NRC pays to its primary concern, that of fullfilling its safety mission.
In December of 1987 a weld failed in Farley-2's emergency core cooling system resulting in thermal cycling. According to the NRC and some industry sources, thermal cycling may indicate a generic problem that could lead to a double-ended pipe failure within LWRs. Charles Rossi, director of the Office of Nuclear Reactor Regulation's (NRR) division of operational events assessment admitted that a double-ended failure of flawed piping could occur and that an ECCS pipe crack could cause a medium-sized LOCA.
However, Pat Mcdonald, a senior vice president at Farley claimed "The plant was never in any danger," because "... pipe materials verified the existence of leak-before-break." This reaction by Mr. Mcdonald only typifies the "out-of-sight-out-of-mind" mentality used to convince themselves and the NRC that there is no immediade safety problem.
Recently, the NRC identified 34 plants with significant erosion / corrosion pipe damage; and raised concerns about the quality of widely used pipes and flanges, yet the NRC proposes to expand leak-before-break. Rather than investigate this delusionary theory, the NRC should implement mandatory piping inspections of all nuclear power plants and re-evaluate leak-before-break because of its unreliability and tragic failure.
We thank you for considering our comments regarding the additional application of leak-before-break.
June 30, 1988 Robert Beaudoin Nuclear Information and Resource Service 1424 16th Street, N.W. Suite 601 Washington, D.C.
20036 7
I June 30, 1988 Secretary US Nuclear Regulatory Commission Washington, D.C.
20555 Attn: Docketing and Service Branch
Dear Mr. Chilk,
Enclosed please find the comments of the Nuclear Information and Resource Service concerning the NRC's request for comments on proposing to investigate the safety benefits associated with using leak-before-break technology for emergency core cooling systems (ECCS) and environmental qualification (EQ) of safety related electrical and mechanical equipment.
We thank you for taking the time to corsider these comments on the add.itional application of leak-befere-break technology.
Sincerely, Robert Beaudoin Information Services I
1 l
i
-