ML20062J178
| ML20062J178 | |
| Person / Time | |
|---|---|
| Issue date: | 02/11/1992 |
| From: | Miraglia J Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20059F318 | List: |
| References | |
| FOIA-93-230 NUDOCS 9208280219 | |
| Download: ML20062J178 (140) | |
Text
. _ _
l,.'
' :N
. ?*
kj v......s nye
.s,
...mm
-,-.....-.--..,r.
n
--x
- v 12 v.
w
,,.w
,y),.
..c
,x
.0:,:
- j.
D 'U. o.. p%.4 -*
,c y
,3;;2/ FINAL REPORT Hjc m:.it,Vs.qg,b bhf NN OFFICE OF.: NUCLEAR REACTOR REGULATION i
%j,4 dS$Qd@&.QAt SPECIAL REVIEW TEAM FOR -r.
~A@THE REVIEW OF,THERMO-LAG FIRE BARRIER PERFORMANCE 3
riAA a
hl5C
- nl' D1?N f
%\\,3 ') 1f6f.,'f'96,,
Oy@Q.5'iN?4]W k
QQj
%f QMfWpl3['
7 s;
$j
~ ^O
. : 'YCKgh tR, W. M -
9 %fi e.
.M.
. y.4 b :, ' ' hf '.y.!y 2, :
f.Q.p.{Q i )13.t,,(-f hjf k'.W,p.m.. -
.. x&'k h.
.,s jij ~ ~
2-, ' al; 7' j -
o t
. :9 a
.er;.wp ;.>i
-: ; rv
~q y %3[r qp?.k',%)ql,y;%,d; <,\\W:
e(, V: ::::Qs S&,4 t.y r %. ypf!, Tiff.:'p.fe%;a.:=c
.. A. %g % /k.h. l, p.h
.;a',, e m
?:-
o c
.a
,e aps.
~
m,
. c, e v: y y,4$d*jJ ~ W E.. t (+
it d IM
,1 '.
q Xtx ngy m
. a.,.
Fobruary ;11;.1992 W,%m. -E.. %, M.e we m.
'V '.*-
I
.Q i4 s e n
q W.
{.p gyd.".
C,: ;
~-.,
'gr 1
-.ew. m,
.. ry 1..
.~
4 i
m x
-s.
,s s.r "c
I.
..';;, y, % :, Q b lj, W Q h.5h Prepar'oN h b
h, )h3@!$
' Dhh,IM, f
g-
?f..:d i
[
.F.
J. Miraglia,fL.1'R.'P1 sco,.K. S. West, NRRl l
Bp
' Y ?$
h Ikh J>
4
.a 1
4 a
' s,"'
g-a,A.rp
..e.,
i s
- k,..;h [ h.
,)'V l f !'
N J4
.i,*
f :: W ??? 7'",
s'Q$,f-
,a
.. t, rd %>] j.gs$g;f%g,n;A.?
.M P
[,
L c
- Isht$fl.c..ut.y kNM j
., n
. :. y.
- m 9,., A A,',\\,$ksk 5.Y
,,y G '%l,$: !l:5, kWh 55
'y y
1,: -
6, c Ng,W%. t k.w.+,:,
~.q
.o n.
i I
- i
\\
I r
s
,., ~ ~. u-a ny rm en e-, e v n.n,__
. m i;...--7----m..
,,,.. - - - - - -,,-..~,,,,,
_s. n
.m im m.
- . m.., - _
m.
I J
+
. V 'A y,f, Q.i..
.W:
a t ".=.% &,r,- 1;
,,)
, r,
. w. <p,,,
. v,
. > ~
.i
~..
U..
f,
- h WWly,.
- 3. '
~c
~'
t he T
mr 7,T Itif(,
- 920B280219.92 804 ifggy Q.?~,,'..%;g.> p; j
2 yrks,,
j
- o
, w,c A.
e
. f...f.
p
.d.s.
u':
7
.sh..M C
. r +N.s Y.* c" w.
n
+<
.n o
$.v.
.~
a l
FINAL REPORT OFFICE OF NUCLEAR REACTOR REGULATION SPECIAL REVIEW TEAM FOR THE REVIEW OF THERMO-LAG FIRE BARRIER PERFORMANCE Table of Contents i
EXECUTIVE
SUMMARY
1 BACKGROUND.....................................................
6 1
ACTIVITIES, ACCOMPLISHMENTS, TECHNICAL ISSUES.................
16 NRC REQUIREMENTS AND GUIDANCE.................................
24 SAFETY SIGNIFICANCE...........................................
30 EVALUATION OF FIRE AND AMPACITY DERATING TESTING OPTIONS......
36 TEST SPONSORSHIP..............................................
43 CONCLUSIONS...................................................
47 RECOMMENDATIONS...............................................
48 LIST OF ATTACHMENTS...........................................
49 t
t t
-w
s
~ N EXECUTIVE
SUMMARY
Many U.S.
Nuclear Regulatory Commission (NRC) licensees use the Thermo-Lag 330-1 (Thermo-Lag) fire barrier system to meet the NRC's requirements for the protection of redundant safe shutdown equipment.
In response to industry operating _ experience, allegations that Thermo-Lag fire barriers may not provide the level of fire protection claimed by Thermal Science, Incorporated (TSI, the vendor), and allegations that the ampacity derating factors for Thermo-Lag are greater than claimed by the vendor, the Director of the Office of Nuclear Reactor Regulation (NRR) established a special review team to assess the safety rignificance and generic applicability of the technical issues regarding the use of Thermo-Lag.1 A joint Office of the Inspector Genera'l (OIG), Office of Investigations (OI) task force was also established to investigate wrongdoing allegations.2 The special review team reviewed the NRC's requirements'and guidance regarding fire barriers; reviewed docket information for six operating reactors and information provided voluntarily by i
three licensees; reviewed vendor documentation and about 50 fire endurance and ampacity derating test reports; visited five operating reactors; and assessed the safety significance of the technical issues.
The team also reviewed information obtained by the OIG/OI task force through subpoenas and interviews.
The special review team found:
The fire resistance ratings and the ampacity derating factors for the Thermo-Lag 330-1 fire barrier system are indeterminate.
i 2 Memoranda from T.E. Murley, NRR, to F.J. Miraglia, et al, NRR, July 12, 1991 and August 15, 1991 ( Attachment 1).
2 Memorandum of Understanding Concerning the conduct of the 4
Thermo-Lag Investigation, signed by B.B.
Hayes, OI, and D.C. Williams, OIG, August 7, 1991.
1
~
-~.....
c.
Some licensees have not adequately reviewed and-evaluated fire endurance test results and ampacity derating test results to determine the validity of the tests and the applicability'of the test results to their plant designs.
Some licensees have not adequately reviewed installed fire barrier configurations to ensure that they either replicate the tested configurations or provide an equivalent level of protection.
Some licensees used inadequate or incomplete installation procedures during the construction of their Thermo-Lag barriers.
Although there is a history of raceway fire barrier qualification and installation problems,'the staff's licensing reviews and inspections of the barriers may not-have been commensurate with either the problems or the importance of the barriers.
Based on these findings, the special review team recommends the following:
Advise industry of the staff's concerns regarding Thermo-Lag fire barriers through a public meeting with the Nuclear Utilities Management and Resources Council (NUMARC).
Encourage NUMARC to coordinate a response that verifies'the fire resistance rating of the Thermo-Lag fire barrier configurations installed by the licensees to meet the NRC's requirements and the ampacity derating factors for those configurations.
Issue a generic letter (Attachment 2) that' discusses the concerns and requires the licensees to provide information needed by the staff to verify compliance with 10 CFR 50.48, GDC 3, and GDC 17.
2
i e
If new tests are needed to verify the fire ratings and the b
ampacity derating factors, provide staff oversight of the industry's test program and review the licensees' corrective.
action plans for resolving any plant-specific fire barrier design, evaluation, and installation issues.
Terminate the special review team and assign SPLB the lead (as primary review branch) for prioritization and resolution of the open items identified in Attachments 4 and 5, for generic letter follow-up activities, and for continuing i
interface with the OIG/OI task force.
Assess the NRC's fire protection review and inspection programs to determine if adjustments are needed.
Ensure that all allegations received by the OIG/OI task force are handled in accordance with commission policy.
s (The NRC Allegations Program Manager will follow-up.)
Formally advise other government agencies and foreign-nuclear industries of the staff's concerns.
l Although the special review team considers the fire resistance ratings of the Thermo-Lag fire barriers indeterminate,.there is evidence that the barriers will provide some level of fire protection.
In addition, most fire areas have low fuel loads, controlled ignition sources, and are equipped with other passive and active fire protection features to alert and assist plant operators in the event of a fire.
The review team considers the relative safety significance of the fire protection concerns to be low.
The team considers the concerns regarding the fire resistance ratings of the Thermo-Lag barriers compliance issues.
Cables in electrical raceways are derated using the ampacity derating factors for the materials,used to enclose the cables, for example, cable trays, conduits, or duct banks.
Cables enclosed in raceways protected by fire barrier materials require 3
t P
additional derating because of the insulating effect of the fire barrier materials.
Ampacity derating calculations based on inaccurate or nonconservative derating factors could result in the installation of undersized cables and raceway overfilling.
This could cause higher than design operating temperatures within the raceways thereby reducing the design' life of the cables.
In extreme cases, cable jacket insulation failures could occur.
If inaccurate or nonconservative derating factors are used, design changes could be required to extend cable life and to restore safety margins.
The special review team did not identify any immediate public health and safety concerns regarding the-c ampacity derating issues.
The review team recommends that the Electrical Systems Branch (SELB) further assess the safety significance of the ampacity derating issues, particularly any plant-specific issues that arise following issuance of the generic letter.
The OIG/OI task' force has informed the special review team that it has allegations and testimonial evidence.that the vendor falsified fire endurance and ampacity derating tests that are being used by licensees as technical bases for using Thermo-Lag fire barriers to meet the NRC's requirements.3 Substantiation that the tests or the test results were falsified would.be convincing justification for conducting the tests recommended by the review team.
However, the OIG/OI task force has advised NRR not to inform the industry of the test falsification allegations at this time because it has yet to substantiate the allegations.
Therefore, the review team has based its conclusions and recommendations solely on the results of its evaluations of the technical issues raised during the review, and not on any results of the OIG/OI investigation.
The industry may attempt to resolve the concerns regarding the validity of the procedures and results of the fire endurance and ampacity derating tests with 3 Memorandum from E.T.
Pawlik, Region III/OI, to L.R.
plisco and K. S.
West, NRR, December 24, 1991 and memorandum from K.E.
Walker, OIG, to X.S.
West and L.R.
plisco, January 29, 1992.
4
engineering analyses instead of performing new fire endurance and
~
ampacity derating tests.
Resolution of the issues by analyses will not be acceptable if the allegations regarding falsification of tests are substantiated.
It is important, therefore, that the i
investigation be vigorously pursued and concluded as promptly as practicable so that appropriate corrective actions can begin.
Final resolution of the Thermo-Lag concerns cannot be made without conclusions regarding the allegations of falsification of tests.
Declaring a final resolution before the investigation has been completed has the potential for the NRC to be criticized for withholding information from the public and the industry that may have ensured more timely resolution of the concerns raised regarding Thermo-Lag.
l I
l i
5 i.
. 4.
=
r
.~
BACKGROUND The Thermo-Lag 330-1 (Thermo-Lag) fire barrier system is l
available from its manufacturer and supplier, Thermal Science, Incorporated (TSI, the vendor), St. Louis, Missouri, with vendor claimed fire resistance ratings of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 3, hours.
Many NRC licensees use Thermo-Lag fire barriers to satisfy the NRC's requirements for protecting safe shutdown capability from fire and to achieve physical independence of electric systems.
On February 21, 1991, NRC Region IV received confidential allegations that Thermo-Lag materials release toxic gases when exposed to fire and that Thermo-Lag fire barriers will not provide the fire resistance ratings claimed by the vendor.'
On February 22, 1991, Region IV notified NRR of the allegations.
NRR assumed responsibility for the resolution of the allegations on February 27, 1991, and assigned the review to the Plant Systems Branch (SPLB).5 On March 5, 1991, NRR's Allegation Review Board (ARB) recommended that the allegations be referred to OI for the investigation of possible wrongdoing by the i
vendor. '
After consideration of the safety significance,.the ARB assigned the allegations a priority level of 3.'
Following l
the ARB meeting, the SPLB staff continued its review of the o
allegations and interviewed several individuals that the alleger stated could substantiate the allegations.
Some of the information obtained by the staff during the interviews supported i
' Letter from confidential alleger to R. Wise, Region IV, February 21, 1991 (Allegation RIV-91-A-0022).
i 5 Allegation NRR-91-A-0011.
ARB Summary approved by B.K.
Grimes, NRR, March 5, 1991.
Memorandum from T.E. Murley, NRR, to F.J. Miraglia, et al, 7
NRR, " Priority Ranking System for Review Efforts - Revision 1,"
March 24, 1989.
6 l
a
= ~
s the allegations.
The staff provided the results of the interviews to OI and the ARB.a l
On April 12, 1991, NRR requested that OI initiate an investigation to determine if the results of fire tests of Thermo-Lag fire barriers were falsified by the vendor.'
on April 23, 1991, OI initiated an investigation.20 The Region III 1
OI field office was assigned the lead for the investigation.
Later, OIG joined the investigation and a joint OIG/OI task force, which is led by OIG with an OI deputy, was established."
In addition to the allegations, Gulf States Utilities (GSU), the licensee for River Bend Station (RBS), informed the NRC of 12 Thermo-Lag fire barrier installation problems and of the failure of an "as-designed" Thermo-Lag fire barrier to pass a 3-hour fire endurance test conducted by the Southwest Research Institute (SWRI)."
Later, GSU reported discrepancies in the installation of Thermo-Lag fire barriers at RBS.2' On
- Notes prepared by D. Notley, NRR, May 10, 1991.
' Memorandum from T.E. Murley, NRR, to B.B. Hayes, OI, April 12, 1991.
OI Response to Request for Investigation from E.T. Pawlik, 10 RIII/OI, to T.E. Murley, NRR, April 23, 1991 (Case No. 3-91-006).-
Memorandum of Understanding Concerning the Conduct of the 22 Thermo-Lag Investigation, signed by B.B. Hayes, OI, and D.C.
Williams, OIG, August 7, 1991.
GSU Licensee Event Report (LER)89-009, " Inadequate Thermo-12 Lag Coverings as Fire Barrier Per T.S. 7.7.7.a," April 17, 1989.
" Letters from J.E.
Booker, GSU, to U.S.
- NRC, December 20, 1989 and January 9, 1990.
GSU LER 90-003, " Inadequate Thermo-Lag Fire Barrier 2'
Envelopes Surrounding Safe Shutdown Circuits Per T.S.
3/4.7.7,"
March 8, 1990 and Rev.
1, July 12, 1990, Rev.
2, February 4,
- 1991, and Rev. 3, June 28, 1991; and GSU LER 91-008," Lack of Fire Wrap -
Inadequate Fire Barrier Caused by Inconsistency in Design Bases Documentation, May 15, 1991.
7
e u
May 29 and 30, 1991, the staff visited RBS to review with GSU the circumstances surrounding the failed fire test and the installation discrepancies.
The staff found that the results of the SwRI fire endurance test raised questions regarding the j
ability of Thermo-Lag to provide a fire rated barrier."
In response to the allegations and the RBS operating experience T. E. Murley, Director, NRR, directed, at a June 27, 1991 I
meeting, that a review of the safety significance and generic applicability of the technical issues regarding the use of Thermo-Lag be conducted.
In a July 12, 1991 memorandum," which was superseded by a August 15, 1991 memorandum" (Attachment 1),
T. E. Murley confirmed this direction and established a special review team to consider the issues and provide recommendations for their resolution.
The NRR line organization was informed of the review team's responsibilities on July 30, 1991.28 The alleger has had frequent contact with the OIG/OI task force and has made additional allegations regarding Thermo-Lag.
The OIG/OI task force has referred all of the technical concerns it has found during its investigation to the special review team, but it has not referred the allegations it has received directly from the alleger or during interviews with other parties to an NRC allegation coordinator to ensure that they are entered into the NRC Allegations Management System (AMS)
One significant new-allegation is that the vendor falsified ampacity derating tests and that the ampacity derating factors for Thermo-Lag are greater than those claimed by the vendor.
The NRC Allegation Program
" Memorandum from J.J. Petrosino, NRR, to L.J. Norrholm, NRR, June 11, 1991.
" Memorandum from T.E. Murley, NRR, to F.J. Miraglia, L.R.
Plisco, and J.F.
Stang, NRR, July 12, 1991.
" Memorandum from T.E. Murley, NRR, to F.J. Miraglia, L.R.
Plisco, and K.S.
West, NRR, August 15, 1991.
la Memorandum from F.J.
Miraglia, NRR, to B.K. Grimes and A.C.
Thadani, NRR, July 30, 1991.
8
Manager will follow-up to ensure that all allegations are handled in accordance with Commission policy."
The vendor provided a list of licensees that use Thermo-Lag fire barriers to the special review team.zo The OIG/OI task force has received similar information from the alleger.
The information provided by these two sources indicates that as many as 100 commercial nuclear power plants may use Thermo-Lag barriers.
These plants are identified in Attachment 3.
The amount of Thermo-Lag installed by the licensees varies j
widely.
For example, a Northern States Power fire protection I
engineer informed the special review team that only two conduits are protected with Thermo-Lag fire barriers at Monticello.
Conversely, Texas Utilities Electric Company (TU) has installed I
more than 20,000 square feet of Thermo-Lag in Comanche Peak Steam
-l Electric Station (CPSES), Unit 1, and plans to install at least this amount in Unit 2.
Some plants, such as the callaway Plant, also use Thermo-Lag to achieve physical independence of electric systems in accordance with the guidelines in Regulatory Guide 1.75, " Physical Independence of Electric Systems".
A 1-hour fire rated Thermo-Lag panel (4 feet by 6 feet by 1/2-inch thick) costs about $925.
A 3-hour fire rated panel (4 feet by 6 feet by 1-inch thick) costs about $2500.
Trowel-grade Thermo-Lag material, which is used to seal the joints between individual i
panel sections when the barriers are assembled around the rameway, costs about $300 per 5-gallon pail.
Installation typically takes about two man hours per linear foot of raceway.
During a December 20, 1991, visit to the TSI factory, the vendor informed members of the special review team that Thermo-Lag products are used by the National Aeronautics and Space
" NRC Manual Chapter 0517, Manaaement of A11ecations, April 3, 1990.
2 Appendix VII to letter from R.
Feldman, TSI, to F.J.
Miraglia, NRR, October 5, 1991.
9
e Administration, the' Department of Defense (for nuclear weapons shipping casks), foreign nuclear plants (Laguna Verde), and the petroleum industry.
According to the vendor, a Thermo-Lag product (not Thermo-Lag 33 0-1) is also installed in every console television sold in the United States to obtain a ten minute fire rating for the back panel of the console.
Many fire endurance tests have been conducted on electrical raceways protected with Thermo-Lag fire barriers (Attachment 6).
However, the special review team found only three fire endurance test reports for Thermo-Lag fire barriers that were formally reviewed and approved by NRR.
The first was a 1-hour fire endurance test conducted by SwRI for CPSES.21 TSI built the test specimen for this test, which was the licensing basis for the CPSES fire barriers.
Pennsylvania Power & Light Company (PP&L) later informed the staff of its intention to use the CPSES test as its technical basis for installing Thermo-Lag fire barriers at Susquehanna Steam Electric Station (SSES), Unit 1.22 However, PP&L changed the design that the staff approved for CPSES.
The staff found the use of the Thermo-Lag barrier acceptable, but did not approve the design change and requested that PP&L submit test data to-demonstrate that the modified design would provide a 1-hour fire rating.2 PP&L submitted a TSI test report to support the design change.2' The staff found that the test was not 21 SwRI~ Report 03-6491, " Fire Qualification Test of a Protective Envelope System," October 27, 1981, approved in letter from R.L.
Tedesco, NRR, to R.J. Gray, Texas Utilities Generating Company, December 1, 1981.
22 Letter from N.W.
Curtis, PP&L, to A.
Schwencer, NRR, February 9, 1982 2' Memorandum from W.V. Johnston, NRR, to R.L. Tedesco, NRR, April 16, 1982.
2' Letter from N.W.
Curtis, PP&L, to A.
Schwencer, NRR, May, 12, 1982 and TSI Technical Note 8232-1, " Engineering Test Report - One Hour ASTM E-119 Fire Simulation Facility Fire Test 10
)l
^'
s performed in accordance with American Society for Testing and Materials (ASTM) Standard E119, " Standard for Fire Resistance of Building Materials," and, therefore, was not -acceptable.25 Later, PP&L submitted a report for a 1-hour fire endurance test conducted at SwRI for SSES to justify the design change.2' This test showed that conduits protected by the proposed method provided a 1-hour fire rating, but that cable trays and air drops-exceeded the temperature rise acceptance criteria.
The staff approved-this test for conduit fire barriers, but not for cable trays or air drops.27 The staff subsequently reviewed and approved a 3-hour fire endurance test for Washington Nuclear Project, Unit 2 (WNP2) that was conducted by TSI at its test facility.2s The special review
.l team did not find any other fire endurance tests of Thermo-Lag fire barriers that were reviewed and approved by the NRC in safety evaluation reports.
With the exception of SSES, Unit 1 and limited applications at WNP2, the review team did not find-any field installations that match the assembly methods used to construct the specimens tested in the three tests'that were reviewed by NRR.
The review team also found evidence that licensees had submitted fire test reports on the docket that were not officially reviewed Followed by a Short Term Water Hose Stream Impact Test on a Class 1E Cable Tray and Air Drop Assembly," June 1981, 25 Memorandum from W.W.
Johnston, NRR, to R.L. Tedesco, NRR, June 28, 1982.
26 Letter from N.W.
Curtis, PP&L, to A.
Schwencer, NRR, August 25, 1982 and SwRI Report 01-7163, " Qualification Fire Test-of a Protective Envelope System," August 1982.
27 Memorandum from W.V. Johnston, NRR, to T.
Novak, NRR, September 15, 1982.
2e ITL Report 82-5-355B, "Three-Hour Fire Endurance Tests on Thermo-Lag 330-1 Subliming Coating Envelope System for WPPSS Nuclear Projects," July 1982, approved in memorandum from W.V.
Johnston, NRR, to T.
Novak, NRR, December 27, 1982.
11
=
by the staf f.2' The licensees may have interpreted this inaction as tacit NRC approval of the reports.
The special review team also found that tests have been conducted to determine the ampacity derating factors for various Thermo-Lag fire barriers (Attachment 7).
Several NRC guidance documents (see the "NRC Requirements and Guidance" section of this report) state that cable derating due to use of fire barrier materials should be considered during design.
However, the review team did not find any evidence in the docket files for the six plants reviewed that the staff reviewed any ampacity derating test reports.
The review team recommends that these matters be considered during the assessment of the NRC's fire protection licensing review program (see Attachment 5, Section 7).
The review team found that both the NRC and the licensees have addressed similar concerns regarding fire barrier qualification' and testing in the past.
For example, in SECY 83-269, " Fire Protection Rule for Future Plants," July 5, 1983, the staff informed the commission of generic issues arising from its review-of almost 600 requests for exemptions from the requirements of Appendix R to 10 CFR Part 50 including problems encountered with testing and installation of fire rated cable wraps.
The staff reported that although the materials performed adequately in laboratory test furnaces, field installations introduced uncertainties due to variations in the training and abilities of installation personnel.
The staff considered these uncertainties to be of marginal safety significance.
The staff also reported that some facilities had installed fire barriers without a basis for their fire rating such as a Underwriters Laboratories, Incorporated (UL) listing or testing conducted by a nationally recognized testing laboratory for the configurations installed in the plant.
The NRC informed the industry of these problems in Office of Inspection and Enforcement (OIE) Information Notice (IN) 84-09, " Lessons Learned From NRC Inspections of Fire 29 Letters from G.D.
Bouchey, WFPSS, to A.
Schwencer, NRR, September 20, 1982 and October 4, 1982.
12
o 5
)
l Protection Safe Shutdown Systems (10 CFR 50, Appendix R),"
February 13, 1984.
This IN provided supplemental NRC guidance for conducting analyses or making modifications tx) implement Appendix R.
The IN stated:
"At some of the facilities inspected, fire barriers were installed without basis for their j
fire rating (such as UL listing or testing conducted by a nationally recognized testing laboratory for the configurations installed in the plant).
Fire barriers installed to meet the requirements of Section III.G.2 of Appendix R must have such a rating."
The special review team also found examples of problems with Thermo-Lag fire barriers experienced by the licensees.
During a 1986 Appendix R inspection of WNP2,30 the staff found that the supports for electrical raceways protected with Thermo-Lag were not protected in accordance with the tested configuration.
The licensee resolved this finding, which was also the subject of an' investigation, by conducting fire endurance testing to verify i
that the installed support fire protection was adequate.
- Later, during a 1987 WNP2 inspection,32 the staff found that although redundant circuits were protected with Thermo-Lag material, "the protection provided did not qualify as a 1 or 3-hour fire barrier because the licensee applied the material improperly and in untested configurations.
Preliminary results of tests conducted by the licensee indicate that the 1-hour application may provide
.a 32 to 50-minute fire rating.
The 3-hour application may provide 150 to 160-minute fire rating."
The licensee resolved this issue by removing and replacing some of the fire barriers and by qualifying others by fire endurance tests conducted at TSI.
(This issue is discussed in detail in the review team's I
report documenting its fact finding visit to WNP2.s2)
- Finally, Inspection Report 50-397/86-05, April 2, 1986.
22 Inspection Report 50-397/87-02, April 17, 1987.
i Memorandum from L.
R. Plisco and K.
S.
West, NRR, to a2 F. J.
Miraglia, NRR, December 11, 1991.
13
r F
m in March 1991, as part of a scheduled annual inspection, the licensee for WNP2 found two deficiencies with a Thermo-Lag _ fire barrier protecting a safety-related cable tray.as The licensee
~
attributed the defects--inadequate Thermo-Lag thickness on the side rail of the cable tray and an incomplete transition between a prefabricated Thermo-Lag panel and a Thermo-Lag spray application--to installation errors and inadequate inspection practices.
In another example of a plant specific problem, CPSES originally _
built its fire barriers using panels that had been fabricated by the constructor with trowel-grade Thermo-Lag material at the CPSES construction site.
Subsequently, the licensee identified deficiencies in the thicknesses of the panels during inspections of the installed fire barriers.
The licensee removed about 12,000 square feet of fire barriers constructed with site fabricated Thermo-Lag panels and replaced them with fire barriers constructed of panels manufactured by the vendor at its facility.
This issue was reviewed in detail by Region IV.35 The special review team also found that the staff and the licensees have addressed concerns with ampacity derating.
For example, in 1986, NRC' inspected the Minnesota Mining and Manufacturing Company (3M), Saint Paul, Minnesota.8' The purpose of the vendor inspection was to review the circumstances WPPSS LER 91-004, " Inadequate Fire Protection (Thermo-Lag) 33 of Division II Safe Shutdown Cables Due to Inadequate Installation and Inspection," April 28, 1991.
Letter from W.J.
Cahill, TU Electric, to U.S.
- NRC, 3'
October 12, 1989.
'5 Letters from C.I. Grimes, NRR, to W.J. Cahill, TU Electric, December 1, 1989 and January 31, 1990.
Inspection Report 99901038/85-01, March 11, 1986.
14
1
~
t surrounding an NRC concern identified during an inspection at
~ !
Fort Calhoun Station.87 The specific NRC concerns were:
(1) Omaha Public Power District (OPPD), the licensee for Fort Calhoun Station, did not verify the validity of the ampacity derating values supplied to them by the 3M, and (2) OPPD or other NRC licensees could misinterpret the derating_ values because of the way they were presented by 3M.
The NRC vendor inspector informed-the special review team that at the time of the 3M inspection, he believed that licensee failure to verify the validity of vendor supplied ampacity derating factors could be a generic concern.
A 3M representative informed the inspector that TSI was a competitor and that they also provided ampacity derating information to the industry.
The inspector subsequently contacted TSI (telephonically) and requested its ampacity derating information for possible NRC review.
In response to this request, TSI provided copies of several ampacity derating tests to the inspector.'8 Due to competing priorities, the ampacity derating test reports provided by TSI were not reviewed by the staff.
As stated above, the review team recommends that this matter be considered during the assessment of the NRC's fire protection review program (see Attachment 5, Section 7).
After the vendor inspection, 3M informed its nuclear customers that its ampacity derating information should "be used only to assess the. general impact of our fire protective envelope systems with respect to ampacity derating.
Ampacity critical situations must be evaluated by performing actual tests and are the responsibility of the utility.3'"
The review team believes that this inspection may have prompted Bechtel Power Corporation to
Item D.5.2-1, NRC Inspection Report 50-285/85-22, December 13, 1985.
Letter from R.
Feldman, TSI, to J.J.
Petrosino, NRC, April 13, 1987, and letter from M.E. Grau, TSI, to J.J. Petrosino, NRC, April 14, 1987.
Letter from R.R.
Licht, 3M, to R.
Clemens, OPPD, February 18, 1986.
15
t 1
perform ampacity derating tests of Thermo-Lag fire barriers for South Texas Project.
These tests (UL Project 86NK23826, File R6802, January 1987), which were performed under the vendor's control at UL, resulted in derating factors'that exceeded those previously reported by the vendor (for example, ITL Reports 82-355-C, 82-355-F, and 82-355-F1).
The vendor informed the NRC and its customers of the preliminary results of the UL tests.
ACTIVITIES, ACCOMPLISHMENTS, AND TECHNICAL ISSUES The major generic safety issues identified and reviewed by the special review team were:
Are the licensees using the correct ampacity doratings for the Thermo-Lag fire barrier configurations installed in the l
plants?
Will Thermo-Lag fire barriers provide a fire rated barrier if installed in accordance with the tested configurations and the vendor's installation procedures?
Are there wide-spread installation problems with Thermo-Lag fire barriers?
Are toxic gases given off by Thermo-Lag when it is exposed to fire?'1 Detailed discussions of these four major issues are presented in.
The specific technical and programmatic issues found. by the special review team, and those issues referred by
'O Mailgram from R.
Feldman, TSI, to U.
S.
- NRC, October 2, 1986.
'2 NRR reviewed a previous allegation that Thermo-Lag releases toxic substances when exposed to fire and concluded that it did not pose an unacceptable risk (letter from C.E. McCracken, NRR, to the alleger, June 20, 1989).
Results of toxicity testing arranged by the review team confirms this conclusion.
16
the OIG/OI. task force are identified in Attachment 5.
These
~
attachments identify each-issue and the references associated with the issue, and provides the disposition and status of each issue, including the extent of the team's review of the issue and its recommendations for resolving the issue, if it was not closed by the team. is a chronological listing-of the review team's major activities.
The "NRC Requirements and Guidance," " Safety Significance," " Evaluation of Fire and Ampacity Derating Testing Options," and " Test Sponsorship,"
sections of this report also describe significant activities and I
accomplishments.
In summary, the special review team reviewed docket information for RBS (Attachment 9), CPSES (Attachment 10), WNP2 (Attachment 11), the Perry Nuclear Power Plant (PNPP), the Enrico Fermi i
i Nuclear Power Plant, and SSES (Attachment 12); reviewed j
information provided voluntarily by the licensees for the Palo Verde Nuclear Generating Station, the Callaway Plant, the Donald i
C.
Cook Plant, and industry consultants; and reviewed fire endurance test reports, vendor technical documentation, and a large volume of information obtained by the OIG/OI task force through subpoenas and over 20 interviews.
Several of the review team's more significant activities and accomplishments are-discussed below.
Following the May 1991 site visit to RBS42 and the review of the RBS operating experience, the special review team prepared IN 91-47, " Failure of Thermo-Lag Fire Barrier Material To Pass Fire Endurance Test," August 6, 1991.
The notice alerted NRC licensees to problems that could result from using or improperly installing Thermo-Lag fire barriers.
Following its initial review, the special review team identified a nunber of concerns regarding the adequacy of Thermo-Lag-fire 42 Memorandum from J.J. Petrosino, NRR, to L.J. Norrholm, NRR,.
June 11, 1991.
17
=-
~
e barriers.
To help resolve these concerns, the team submitted questions regarding the Thermo-Lag material, installation, and-test procedures to the vendor."
The vendor answered the questions," and, on October 17, 1991, met with the staff to discuss its response and to other technical issues arising during the review assignment.'s At_the request of the OIG/OI task force, this meeting was closed to the public and staff not directly involved in the review."
Following the meeting, the review team issued follow-up questions," which the vendor answered."
While conducting its review of IN 91-47, the Cleveland Electric Illuminating Company (CEI), the licensee for PNPP, found discrepancies in the installation of Thermo-Lag fire barriers that could adversely affect the safe shutdown capability at PNPP.
Specifically, CEI found that the distances between mechanical fasteners exceeded the vendor's recommendations."
On November 4
19, 1991, members of the special review team visited PNPP to
)
s f
" Letters from F.J. Miraglia, NRR, to R.
Feldman, TSI, September 10, 1991 and September 18, 1991.
" Letters from R.
Feldman, TSI, to F.J. Miraglia, NRR, September 12, 1991 and October 5, 1991.
)
Official Transcript of Proceedings, " Meeting with Thermal
'5 Science, Inc., to Discuss Issues Involving Thermo-Lag 330,"
October 17, 1991.
" Note from K.S. West, NRR, to Thermo-Lag File, j
September 24, 1991.
" Letters from F.J.
Miraglia, NRR to R.
Feldman, TSI, October 31, 1991.
Letters from R.
Feldman, TSI, to F.J. Miraglia, NRR, 1
November 8, 1991, November 12, 1991, and December 3, 1991.
CEI LER 91-020, " Cable Tray Raceways Found to be Impaired as a Fire Barrier Adversely Affecting Safe Shutdown Requirements,"
November 19, 1991.
18 i
+
s review with CEI the event and the as-built Thermo-Lag fire
~
barriers.55 Members of the special revies team also visited RBS,51 CPSES,52 WNP2,55 and the Callaway Plant' to obtain information on the 5
use of Thermo-Lag fire barriers by the industry.
As a result of these visits the review team found evidence of common failures in the use of Thermo-Lag fire barriers by the licensees.
These included (1) failing to review and evaluate fire endurance and ampacity derating test results to determine the validity of the tests and the applicability of the test results to the plant design, (2) failing to review and analyze fire barrier configurations to ensure that they either replicate the tested configurations or provide an equivalent level of protection, (3) failing to install the barriers in accordance with the vendor's installation procedures, (4) failing to review the.
revised ampacity derating factors provided by the vendor.55 The review team also identified plant specific issues which it 55 Memorandum from L.R.
Plisco and K.S. West, NRR, to F.J. Miraglia, NRR, December 18, 1991.
52 Memorandum from L.R.
Plisco and K.S. West, NRR, to j
F.J..Miraglia, NRR, October 31, 1991.
32 Memorandum from L.R.
Plisco and K.S. West, NRR, to F.J.
Miraglia, NRR, December 24, 1991.
55 Memorandum form L.R.
Plisco and K.S. West, NRR, to j
F.J. Miraglia, NRR, December 11, 1991.
5' Memorandum from L.R. Plisco and K.S. West, NRR, to F.J. Miraglia, NRR, January 7, 1992.
55 Mailgram from R.
Feldman, TSI, to U.
S. NRC and the licensees, October 2, 1986.
19
e a
y referred to the region offices" and investigative issues which it referred to the OIG/OI task force."
On December 6, 1991, NRR issued IN 91-79, " Deficiencies in the Procedures for Installing Thermo-Lag Fire Barrier Materials."
The notice informed NRC licensees of installation problems that the vendor acknowledged durir.g the October 17, 1991, meeting with the special review team, and that the team had found during its visits to RBS, CPSES, WNP2, and PNPP.
In letters to licensee fire protection engineers" and the NRC," the vendor stated that the 1989 test conducted by SwRI for GSU' was flawed and invalid because of installation deficiencies.
The vendor believes that the test assembly failed and collapsed because the seam at the interface of the Thermo-Lag and Promat interface, which covered the horizontal support, separated under fire exposure and allowed fire gases to enter the interior of the cable tray.
The special review team interviewed the GSU employees that built and inspected the test specimen and that witnessed the test.
The team also evaluated SwRI's test report.
NIST provided technical input to the review team.62
" Memorandum from F.J.
Miraglia, NRR, to R.D.
Martin, Region IV, November 12, 1991; memorandun fr Miraglia to J.B. Martin, Region V, January 9,
- "92;
-nd memoranda from F.J.
Miraglia, to A.B.
Davis, Regit IP, January 10, 1992 and February 6, 1992.
" Memorandum from L.R.
Plisco, NRR, to R.C.
- Paul, Region III/OI, November 22, 1991.
" Letter from R.A. Lohman, TSI, to F. Garrett, Arizona Public Power, et al, August 23, 1991.
" Letter from R.
Feldman, TSI, to F.J.
Miraglia, NRR, October 5, 1991.
SWRI Draft Report 01-2702, "Three-Hour Qualification Test on Two Protective Envelope Systems for Class 1E Electrical Circuits and Five Penetration Seals," May 1991.
'l Letter from V. Babraskas and K.D.
Steckler, NIST to K.S.
West, NRR, January 21, 1992.
20
The review team reviewed the temperatures recorded at locations on the bare copper wire adjacent to thermocouple (TC) 82, as well as temperatures along the horizontal support between the insulation interface and the center or the cable tray, and concluded that the vendor's explanation of the test failure is possible but not probable.
Temperatures recorded at points along the copper wire adjacent to TC 82 (TC 81 and TC 83) do not show evidence of impinging furnace gases before or at the time of excess temperature at TC 82.
Although it is possible that a narrow jet of furnace gases entered the cable tray and struck TC 82 but not the nearer TC 83, the more plausible explanation of the temperature recorded during the test is that the Thermo-Lag fire barrier failed in the area of TC 82 independent of any failure at the interface of the Thermo-Lag and Promat fire barrier materials.
(This finding was a significant contributor to the review team's finding that the fire resistance ratings for Thermo-Lag are indeterminate.
Additional testing problems identified by the team are discussed in the " Evaluation of Fire and Ampacity Derating Testing Options" section of this report and in Attachment 5.)
The special review team also requested a vendor inspection of the TSI facility and assisted the vendor inspectors prepare for the inspection and develop the inspection plan."
The inspection was conducted during the week of December 16, 1991.
Members of the team attended the inspection exit meeting.
The inspection report has not been completed.
However, the NRC inspectors informed the team that they identified one nonconformance regarding inspection procedures.
On December 9, 1991, the special review team informed OIG of the results of its review and proposed a plan to transfer the Memorandum from R.C.
Wilson, NRR, to L.J. Norrholm, NRR, December 12, 1991.
21
. remaining review activities to the NRR line organization.'8 The OIG approved the proposal," the line organization was informed of the plan,'S and on January 21, 1992, OIG members of the OIG/OI task force met with the special review team and the NRR line organization to discuss the transfer of the review.
The special review team's final activities have included the preparation of a generic letter to notify the industry of its technical findings and concerns (Attachment 2), coordination of a meeting with NUMARC to discuss the findings and concerns discussed in the proposed generic letter, and preparation of a memorandum to inform the Commission of NRR's plans for resolving the findings and concerns through NUMARC."
The proposed generic letter, which was provided to the Commission for information by the EDO, has been reviewed by the NRR Executive Team (and appropriate staff), the Generic Communications Branch, the Office of the General Counsel, the OIG/OI task force, and the technical editor.
The Advisory Committee on Reactor Safeguards, Subcommittee for Auxiliary and Support Systems, has requested briefings on IN 91-47 and IN 91-79.
Members of the special review team are scheduled to give the briefings on February 14, 1992.
Technical Assistance To support its review of Thermo-Lag fire barriers, NRR established an interagency agreement with the National Institute 63 Memorandum from F.J. Miraglia, NRR, to L.J. Norton, OIG, December 9, 1991.
" Memorandum from L.J. Norton, OIG, to F.J. Miraglia, December 30, 1991.
'5 Memorandum from F.J. Miraglia, NRR, to A.C.-Thadani, NRR, January 17, 1992.
]
" Memorandum from J.M. Taylor, EDO, to The Chairman and the
)
Commissioners, February 7, 1992.
22
1 of Standards and Technology (NIST)."
The objective of the agreement is to obtain specialized technical expertise from NIST's Building and Fire Research Laboratory (BFRL) to support l
the special review team in its review and evaluation of the technical issues regarding Thermo-Lag fire barriers.
The tasks assigned to NIST include (1) review the NRC's requirements and guidance documents (complete), (2) review technical documentation, (3) perform physical and chemical analyses of Thermo-Lag 330-1 materials, (4) prepare input to testing options (complete), (5) prepare input to an ampacity derating and fire endurance test program, (6) examine in-plant conditions, (7) identify testing laboratory candidates, and (8) conduct toxicity tests.
Detailed discussions of the review team's technical direction and of the work completed by NIST are documented in the team's reports of meetings with NIST.
The special review team arranged for NRC to purchase Thermo-Lag prefabricated panels and trowel-grade material from Union Electric Company (from Callaway Plant stock) for analysis by NIST."
NRR recently issued a request for proposal to the Department of Energy (DOE) to obtain additional technical assistance from the Sandia National Laboratory."
Specific tasks to be assigned to Sandia include (1) review the NRC's requirements and guidance documents, (2) review technical documentation, (3) develop technical input to a test program, (4) help NRR and NIST assess testing laboratory candidates, and (5) help conduct the test
" Interagency Agreement NRC-03-91-032, " Technical Assistance for the Evaluation of Fire Barrier Performance," October 21, 1991 (FIN L2063).
6a Notes from K.S. West, NRR, to File (FIN L-2063),
December 30, 1991, December 2, 1991, October 22, 1991, September 27, 1991, and September 5, 1991.
" NRC Order DR-91-13 68, September 27, 1991, and Union Electric Company Miscellaneous Sales Ticket 8355, September 27, 1991.
" Letter from W.S. Schwink, NRR, to R.Y. Lowery, DOE, "Sandia National Laboratory Technical Assistance to NRR - Evaluation of Fire Barrier Performance," FIN L2414, December 16, 1991.
23
i program and analyze the test results.
The review team is awaiting Sandia's proposal for this project.
In setting up the technical assistance contracts with NIST_and Sandia, the review team assumed that the NRC would be conducting If this activity is conducted by the industry, a test program.
which the review team recommends, these contracts should be reevaluated and modified to specify technical assistance for the review of the industry's effort, if needed.
NRC REOUIREMENTS AND GUIDANCE Amoacity Derating General Design Criterion (GDC) 17, " Electric power. systems," of Appendix A to Title 10 of the U.S.
Code of Federal Reculatio.D2, l
Part 50 (10 CFR Part 50), " General Design Criteria for Nuclear Power Plants," states:
"The safety function for each [ electric power] system shall be to provide sufficient capacity and capability to assure that (design conditions are not exceeded]."
To meet this GDC, the licensees must derate power cables enclosed in electrical raceways using the ampacity derating factors for the materials surrounding the cables, such as, cable trays, conduits, or duct banks.
Cables enclosed in raceways protected by fire barrier materials require additional derating because of the insulating effect of the fire barrier materials.
In addition, 10 CFR 50.55a(h), " Protection systems," requires that protection systems meet the-requirements set forth in the Institute of Electrical and Electronics Engineers (IEEE)
Standard: " Criteria for Protection Systems for Nuclear Power Generating Stations," (IEEE-279).
Section 4.3 of IEEE-279 identifies the requirement for derating of components.
i Guidelines for achieving physical independence of electric systems are provided in Regulatory Guide 1.75, " Physical i
Independence of Electric Systems".
24
Branch Technical Position (BTP) Auxiliary and Power Conversion Systems Branch (APCSB) 9.5-1,
" Guidelines for Fire Protection for Nuclear Power Plants," and Appendix A to BTP APCSB 9.5-1,
" Guidelines for Fire Protection for Nuclear Power Plants Docketed Prior to July 1, 1976," (see below) state:
"Possible cable derating owing to use of (fire retardant coatings] must be considered during design."
In addition, Section 8.3.1, "A-C Power Systems (Onsite)," of NUREG-0800, " Standard Review Plan,"
Revision 2, July 1981, states:
"PSB (Power Systems Branch) will review cable derating and raceway fill to ensure compliance with accepted industry practices."
Fire Protection Federal Recister, Volume 45, Number 105, " Fire Protection Program for Nuclear Power Plants Operating Prior to January 1, 1979,"
May 29, 1980, and Generic Letter (GL) 86-10, " Implementation of Fire Protection Requirements," April 24, 1986, contain detailed discussions of the NRC's fire protection requirements and guidance.
This section of this report provides an overview of these requirements and guidance regarding fire barriers.
" Fire protection," of Appendix A to 10 CFR Part 50, states that structures, systems, and components important to safety shall be designed and located to minimize, consistent with other safety requirements, the probability and effect of fires and explosions."
The licensees rely on their fire protection programs to satisfy GDC 3.
Following a major fire at the Browns Ferry Nuclear Station on March 22, 1975, the NRC evaluated the need for improving the fire protection programs at all nuclear power plants."
Following this evaluation, the NRC issued guidance for implementing GDC 3 in BTP APCSB 9.5-1.
This guidance did not apply to plants docketed at that time.
The NRC later provided guidance for operating reactors in Appendix A to
" NUREG 0500, Recommendations Related to Browns Ferry Fire, February 1976.
25
BTP APCSB 9.5-1.
These BTPs include guidance for fire barriers and protection of electric cable constructions.
By.early 1980, most operating plants had implemented most of the guidelines in BTP APCSB 9.5-1 or Appendix A to BTP APCSB 9.5-1.
However, several licensees refused to adopt staff guidance related to several specific generic issues.
On May 29, 1980, the NRC published a proposed rule for nuclear power plant fire protection, Section 50.48, " Fire protection," (10 CFR 50.48) and Appendix R to 10 CFR Part 50, " Fire Protection Program for Nuclear Power Facilities Operating Prior to January 1, 1979,"
which proposed minimum requirements for the unresolved issues.72 On February 17, 1981, 10 CFR 50.48 and Appendix R to 10 CFR Part 50 became effective.7' Section 50.48 requires that each nuclear power plant licensed to operate before January 1, 1979, meet the requirements of Section III.G, " Fire protection of safe shutdown capability," of Appendix R to 10 CFR Part 50 even if the NRC had previously approved alternative fire protection features in these areas.
In a separate action the Commission required that all plants to receive their operating license after January 1, 1979, also satisfy specific requirements of Appendix R to 10 CFR Part 50, including Section III.G.
The objective of Section III.G of Appendix R is to ensure that at least one means of achieving and maintaining safe shutdown conditions will remain available during and after any postulated fire in the plant.
Licensees can satisfy Section III.G by separating one safe shutdown train from its redundant train with fire-rated barriers.
The fire resistance rating required of the 72 U. S. NRC, " Fire Protection Program for Nuclear Power Plants Operating Prior to January 1, 1979," Federal Reaister, Vol. 45, No. 105, May 29, 1980, pp. 36082-36090.
73 U.
S.
NRC, " Fire Protection Program for Operating Nuclear Power Plants," Federal Reaister, Vol. 45, No. 225, November 19, 1980, pp. 76602-76616.
26 1
barrier, either 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> or 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />, depends on the other fire protection features provided in the fire area.
Many NRC licensees use raceway protective envelopes, such as Thermo-Lag fire barriers, to satisfy the separation requirements of Section III.G of Appendix R to 10 CFR Part 50.
The NRC has provided guidance for implementing NRC fire protection requirements through information notices, regulatory guides, standard review plans, branch technical positions, and generic letters.
BTP Chemical Engineering Branch (CMEB) 9.5-1,
" Guidelines for Fire Protection For Nuclear Power Plants," July 1981, included the acceptance criteria identified in a number of these documents, including Appendix R to 10 CFR Part 50.
BTP CMEB 9.5-1 states:
"This document supplements Regulatory Guide 1.75 in determining the fire protection for redundant cable systems."
In the Spring of 1984, the Commissicn held regional workshops with the applicants and licensees on the implementation of NRC fire protection requirements.
At those workshops, the staff distributed NRC staff responses to industry questions, including those regarding fire barriers.
In a September 13, 1984 memorandum," the Executive Director for Operations (EDO) established a steering committee and directed it to review fire protection issues to assure consistent levels of fire protection safety at all plants.
One of the issues assigned to the committee for review was the adequacy of the NRC guidance to the industry.
In its final report," which was provided to all licensees and applicants," the steering committee concluded
" Memorandum from W.J.
Dircks, EDO, to H.R. Denton, NRR, 31 al, September 13, 1984.
" Memorandum from Fire Protection Policy Steering Committee to W.J.
Dircks, EDO, October 26, 1984.
" GL 85-01, " Fire Protection Steering Committee Report,"
January 9, 1985.
27 s
that the NRC had provided adequate technical. guidance to the industry, but that there were areas where confusion could arise.
The steering committee concluded that new guidance was not needed and recommended that the existing guidance be brought together in one generic letter and that the Standard Review Plan, technical.
specifications, and licenses be made consistent.
The generic letter recommended by the steering committee was approved by.the EDO and the Commission, and was issued as GL 86-10 on April 24, 1986.
The guidance provided during the regional workshops was appended to GL 86 10.
To the extent that the guf. dance in GL 86-10 was inconsistent with prior guidance, the gu!. dance in GL 86-10 took precedence over prior guidance.
Fire Barrier Oualification The NRC provided its technical basis for requiring 3-hour fire barriers in the Federal Reaister notice that issued 10 CFR 50.48 and Appendix.R to 10 CFR Part 50 (see Footnote 70).
The NRC stated that although nuclear power plants have low fire' loads,.
the potential consequences of fire are serious.
Therefore, 3 bours was selected as the minimum fire resistance rating for fire barriers used to separate redundant safe shutdown systems.
Where plant conditions preclude the installation of a 3-hour fire barrier, the NRC considers a 1-hour fire barrier with automatic fire detection an suppression systems to be equivalent to a 3-hour fire barrier.
Therefore, fire barriers relied upon to protect shutdown systems need to have a fire resistance rating of either 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> or 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />.
The NRC also noted that fire barriers are rated for fire resistance by being exposed to the standard test fire defined by the ASTM E119, " Standard for Fire Resistance of Building Materials.""
BTP APCSB 9.5-1 and Appendix A to BTP~
APCSB 9.5-1 reference NFPA 251 and define fire rating as: "the endurance period of a fire barrier or structure; it defines the
" National Fire Protection Association (NFPA) Standard 251,
" Standard Methods of Fire Tests of Building Construction and Materials," is identical to ASTM E119 and is referenced in some NRC guidance documents.
28
period of resistance to a standard fire exposure before the first
~
critical point in behavior'is observed."
In GL 86-10, the NRC provided guidance on its fire barrier acceptance criteria, the documentation required to verify the fire barrier ratings, and the evaluation of deviations between tested configurations and field installations.
GL 86-10 states that the fire resistance rating of each fire barrier used to satisfy the NRC's requirements should be determined by a standard fire endurance test.
This standard, NFPA 251 (or ASTM E119),
requires that a test specimen representative of the construction for which a fire rating is desired, as to materials, method of assembly, dimension, and configuration, be exposed to a standard test fire.
The NRC applies the NFPA 251 acceptance criteria for non-bearing fire barriers to electrical raceway fire barriers.
These criteria specify-that the transmission of heat through the barrier "shall not have been such as to raise the temperature on its unexposed surface more than 250 *F above its initial temperature."
It is generally recognized that the ambient air temperature is 75
- F at the beginning of a fire test.
The resulting 325 *F cold side temperature criterion is used for raceway fire barriers because they function to preserve the.
integrity of the cables and keep them free of fire damage.
GL 86-10 stated that cables that begin to degrade at 450 *F will'be free of fire damage at 325
- F.
When Appendix R to 10 CFR Part 50 was implemented, the NRC did not require licensees to replace conduit and cable tray enclosure materials that were previously accepted by the NRC as 1-hour barriers and already installed if they did not meet the 325 *F criteria.' However, GL 86-10 states that newly identified conduit and cable trays that required fire barriers should meet the 325 *F criterion, or justification should be provided for use of the materials.
This justification may be based on an analysis that demonstrates that the maximum recorded test temperature is 29
sufficiently below the cable insulation ignition temperature.
Although GL 86-10 was issued in 1986, the special review team found evidence that the staff has applied the 325 *F cold side temperature criterion since at least 1981 (see Footnote 167).
GL 86-10 also states that the licensees should have the design description of the barrier and the report that documented that a test assembly passed a standard fire endurance test (NFPA 251 or ASTM E119) to verify the fire resistance rating of each fire barrier.
This assembly should represent materials, method of assembly, dimensions, and configuration for which a fire resistance rating is desired.
Construction variations may substantially change the performance characteristics of the assembly.
Therefore, to ensure that the fire barriers can provide the level of fire protection required, the licensees should either install barriers that replicate the configurations that were tested or justify, by engineering evaluation, that fire barriers that deviate from the tested configurations provide an equivalent level of protection.
BAFETY BIGNIFICANCE Ampacity Deratina cables enclosed in electrical raceways are derated using the ampacity derating factors for the materials surrounding the cables, for example, cable trays, conduits, or duct banks, to ensure that systems provide sufficient capacity and capability to perform their intended safety functions.
Cables enclosed in raceways protected by fire barrier materials require additional derating because of the insulating effect of the fire barrier materials.
Cable derating calculations that are based on J
inaccurate or nonconservative derating factors could result in the installation of undersized cables and raceway overfilling.-
This could cause higher than design operating temperatures within the raceways thereby reducing the expected design life of the cables.
In extreme cases, cable jacket insulation failures could 30 l
1 occur.
Therefore, undersized safety-related cables reduce plant safety margins.
If inaccurate or nonconservative derating factors are used by the licensees for the electrical system design basis, design changes could be required to extend cable life and to restore safety margins.
The special review team did not identify any immediate public health and safety concerns regarding the ampacity derating issues.
The review team l
recommends that SELB further assess the safety significance of the ampacity derating issues, particularly any plant-specific issues that arise following issuance of the generic letter.
Fire Protection The licensees use the defense-in-depth concept of echelons of safety systems to achieve the high degree of safety required for nuclear power plants.
This concept is also applicable to nuclear power plant fire safety.
The defense-in-depth principle applied to the fire protection program is aimed at achieving an adequate balance in preventing fires from starting, detecting and suppressing fires quickly to limit damage, and designing plant safety systems so that a fire that starts in spite of the fire prevention program will not prevent essential plant safety functions from being performed.
No one of these echelons can be perfect or complete by itself, but each echelon should meet certain minimum requirements.
However, strengthening any one echelon can compensate in some measure for weaknesses, known or unknown, in the others.
In the Federal Reaister notice that issued the proposed Appendix R to 10 CFR Part 50 (see Footnote 69), the NRU stated:
"The phenomenon of fire is believed to be sufficiently well understood to permit evaluation of existing and potential fire hazards and probable extent of damage should a fire occur.
Such evaluations are useful in assessing the possible consequences of fire in a given area.
However, the phenomenon of fire is so unpredictable in occurrence and development that measures to prevent unacceptable consequences may not be omitted on the basis of low 31
The minimum fire protection probability of occurrence.
requirements for nuclear power plants must be established not only to identify fire hazards but also to protect against unacceptable consequences of fire."
The primary objective of the fire protection program is to minimize both the probability and consequences of postulated fires.
In spite of steps taken to reduce the probability of Therefore, means are needed fire, fires are expected to occur.
to detect and suppress fires with emphasis on providing passive and active fire protection of appropriate capability and adequate capacity for the systems necessary to achieve and maintain safe Generally, in plant shutdown with or without offsite power.
plant areas where the potential fire damage may jeopardize safe plant shutdown, the primary means of fire protection should consist of fire barriers and fixed automatic fire detection and suppression systems.
Also, a backup manual fire fighting capability should be provided throughout the plant to limit the extent of fire damage.
The licensees determine the adequacy of fire protection for plant safety systems and areas by analysis of the effects of the postulated fire relative to maintaining the ability to safely shut down the plant and minimize radioactive releases to the environment in the event of a' fire.
The variables used to evaluate the level of fire protection needed in a given fire area include:
Fire loading and fire hazards.
Automatic fire detection and suppression capability.
Layout and configurations of safety trains.
Reliance on fire barriers including, the quality of the materials and system, and the quality _of the installation.
Fire area construction.(walls, floor, ceiling, dimensions, volume, ventilation, and congestion).
Location and type of manual fire fighting equipment and J
accessibility for manual fire fighting.
Potential disabling effects of fire suppression systems on shutdown capability.
32 l
Availability of oxygen (for example, inerted containment).
~
Amount of cable insulation and other combustible materials.
Alternative or dedicated shutdown capability During NRC reviews of the licensees' fire protection programs, the staff evaluated these variables to ensure that each licensee provided an adequate level of protection.
In addition, the NRC has granted licensees requests for exemptions from specific Appendix R requirements and has approved deviations from staff guidance.
The staff performed safety evaluations and granted the exemptions and deviations based on fire hazards analyses performed by the licensees that demonstrated that an alternative provided an equivalent level of protection.
The NRC provided criteria for evaluating exemptions in GL 81-12, " Fire Protection Rule," February 20, 1981.
The fire areas at greatest risk if raceway fire barriers do not provide the fire resistance intended by the licensees, are the areas where NRC granted exemptions from the requirements of Section III.G 2 of Appendix R to 10 CFR Part 50, because reduced defense-in-depth was approved based, at least in part, on the assumption that the installed fire barriers would perform as claimed.
The following configurations are ranked from the most safety significant to least safety significant.
(These are general rankings.
The true safety significance of any particular configuration is also dependant on the nature of the protected equipment and its relative importance to safety.
Factoring in this aspect would require licensee input and a case-by-case analysis.)
Exemption from Section II.G.2.c of Appendix R.
When one train of a system required to achieve safe shutdown is separated from its redundant train by a 1-hour fire rated barrier and automatic fire detection and suppression capabilities are installed,Section III.G.2.c of Appendix R is met.
If the licensee has been granted an exemption from the requirement to provide automatic detection and 1
33
suppression systems, the ability to achieve safe shutdown is largely dependent on the ability of the 1-hour fire barrier to maintain the protected train free of fire damage.
If the fire barrier cannot provide this level of protection, safe shutdown capability may be adversely affected.
Exemption from Section III.G.2.a of Appendix R.
When one train of a system required to achieve safe shutdown is separated from its redundant train by a 3-hour fire rated barrier,Section III.G.2.a of Appendix R is satisfied.
Automatic fire detection and suppression capabilities are not required.
Therefore, the ability to achieve safe shutdown is dependent on the 3-hour fire barrier to maintain one train free of fire damage.
The technical bases for approving this type of exemption may have included fire test results that, in combination with a fire hazards analysis, demonstrated reasonable assurance that safe shutdown could be achieved if a fire barrier of less than 3-hours fire resistance is installed.
If the fire test results are not valid, then the fire hazards analysis would not be valid and the level of fire protection needed to protect safe shutdown capability may not be provided.
Exemption from Section'II.G.2.c of Appendix R.
When one train of a system required to achieve safe shutdown is separated from its redundant train by a 1-hour fire rated barrier and automatic fire detection and suppression systems are installed,Section III.G.2.c of Appendix R is met.
If' the licensee has been granted an exemption from the requirement to provide either automatic fire detection gr suppression systems, the ability to achieve safe shutdown is-dependent on actuation of the installed system and the 1-hour fire barrier to maintain one train free of fire damage.
If the fire test results are not valid, the level of protection needed to protect safe shutdown capability may not be provided.
34 i
I
Compliance with Section III.G.2 of Appendix R to 10 CFR Part
~
50 based on invalid fire test results or improper fire barrier installation.
The special review team considers the concerns regarding the fire resistance ratings of the Thermo-Lag barriers compliance issues.
The review team considers the relative safety significance of the fire protection concerns to be low for the following reasons.
Fire resistive barriers are evaluated in a testing furnace by exposing a test specimen to a fire whose severity follows a time varying temperature curve--the standard time-temperature curve--
for the duration of the fire rating required.
The NRC has conservatively selected 3-hours as the minimum fire resistance rating for fire barriers used to separate redundant safe shutdown systems.
(1-hour barriers with automatic fire detection and suppression systems are considered equivalent to 3-hour barr iers. )
The fire test standard maximizes fire severity by subjecting the barrier to a fire of rapid temperature rise in a confined space that totally engulfs the test specimen.
In an i
actual fire situation, the fire resistance required of a barrier depends on the expected severity of the-fire to which it may be exposed.
Typical nuclear plant fire loads are not great enough q
to produce a fire approaching the severity of a test fire.
In j
addition, an actual nuclear power plant fire would have a much slower temperature rise than the test fire.
In large open volumes, such as most nuclear plant fire areas, a fully developed fire may occur in one part of the area, but it is not probable that the entire volume (fire area) would flashover.
Unless a fire reaches this stage, it is not likely to present.a credible challenge to any nuclear power plant fire barrier.
Although the j
fire resistance ratings of the Thermo-Lag fire barriers are considered indeterminate,.the special review team believes the j
barriers will provide some level of fire protection.
In addition, most plant areas have controlled ignition sources, which helps reduce the occurrences of fires and are equipped with other passive and active fire protection features, and are continuously occupied by plant operators, security, and other 35
~
personnel, which contributes to early fire detection and suppression activities.
Review Priority Although the special review team did not identify any immediate public health and safety concerns regarding the ampacity derating issues and considers the relative safety significance of the fire protection concerns to be low, the concerns are important and will require staff action over the long term.
The review team assessed the overall review effort in accordance with NRR's priority ranking system (see footnote 7) and recommends that it be ranked as priority level 3.
However, the team recommends that the primary review branch review the priorities for the remaining activities as they develop, such as those required to issue the proposed generic letter and to work with industry to develop corrective actions, for assignment of higher priority, if warranted.
In addition, plant-specific issues that develop after the generic letter is issued may require higher review priority.
EVALUATION OF FIRE AND AMPACITY DERATING TESTING OPTIONS The special review team reviewed about 40 of the fire endurance test reports identified in Attachment 6 and found that the validity of the tests and the acceptability of the test results as technical bases for establishing the fire resistance ratings of Thermo-Lag fire barriers are questionable.
The problems identified by the review team included inadequate documentation of test procedures and results, incomplete or inadequate test procedures, unqualified test personnel, inadequate test equipment, questionable methods of assembly and quality assurance, and failure to meet NRC acceptance criteria.
The special review team also reviewed nine ampacity derating tests (Attachment 7) and found conflicting test results.
For example, the vendor has reported derating factors for cable trays 36
that range from 7 percenta to 28 percent" for 1-hour fire 7
barriers and from 16 percenteo to 31 percent" for 3-hour barriers.
In addition, ampacity derating tests of Thermo-Lag materials conducted for 3M" found the ampacity derating to be 37 percent for a 1-hour barrier.
There are similar inconsistencies for conduit barriers.
l-l l
The special review team concluded, based on the tests found by the team, that the fire resistance ratings and the ampacity-derating factors for Thermo-Lag fire barriers are indeterminate.
The review team recommends that industry be informed'of this conclusion and be asked to provide test results, if available, that demonstrate the fire resistance ratings and the ampacity derating factors for Thermo-Lag.
If industry cannot produce existing tests, the review team recommends that additional testing be performed to demonstrate the adequacy of Thermo-Lag fire barriers for satisfying the NRC's requirements.
Based on the assumption that valid fire endurance and ampacity derating tests do not exist, the special review team evaluated the following options for determining the fire resistance ratings and the ampacity derating factors of Thermo-Lag fire barriers:
(1) small-scale tests, (2) full-scale confirmatory tests, (3)
TSI Technical Note 11171, " Engineering Report on Ampacity Test for 600 Volt Power Cables Installed in a Five Foot Length of Two Inch Conduit Protected with Thermo-Lag 330-1 Subliming Coating Envelope System," November 1981.
" UL Test 86NK23826, "Special Services Investigation of Ampacity Ratings for Power Cables in Steel Conduits and in Open Ladder Cable Trays with Field Applied Enclosures,"
January 21, 1987.
ITL Report 82-5-355F, "Ampacity Derating Test for 1000 Volt Power Cables in a Ladder Cable Tray Protected With a._Three Hour Rated Design of Thermo-Lag 330-1 Subliming Coating Envelope System," July 1982.
" SWRI Project 01-8818-208/209-a, "Ampacity Derating of Fire Protected Cables in Cable Trays Using a One-Hour Subliming, Rigid Board Provided by 3M, Incorporated," August 11, 1986.
l 37 l
I modified full-scale confirmatory tests, and (4) original full-scale tests.
The special review team recommends that original full-scale fire tests be performed in accordance with the NRC's requirements and guidance to determine the fire resistance ratings of the Thermo-Lag fire barrier configurations installed by the licensees to meet the NRC's requirements, and that original full-scale ampacity derating tests be performed in accordance with an industry accepted procedure to determine.the ampacity derating factors for those configurations.
Discussions of each of the testing options evaluated by the review team follow.
Small-Scale Tests:
GL 86-10 states that the fire resistance rating of each fire barrier used to satisfy the NRC's requirements should be determined by a standard fire endurance test.
This standard, NFPA 251 (or ASTM E119), requires that a i
test specimen representative of the construction for which a fire rating is desired, as to materials, method of assembly, dimension, and configuration, be exposed to a standard test fire.
There are no small-scale test standards or procedures for determining fire resistance ratings.
NFPA 251 (Section 4-1.2) allows the dimensions of the test specimens to be reduced only when the field installations will be limited to such use.
An improvised small-scale test would not accommodate a test specimen representative of the construction for which the fire rating is desired.
Therefore, the results of small-scale tests could not be used to determine the fire resistance ratings of Thermo-Lag fire barriers.
j i
The special review team believes that it may be possible to improvise a cost-effective small-scale fire test procedure in a short period of time to obtain insights into the fire endurance capabilities of Thermo-Lag, for example, a scaled-down NFPA 251 type test of a small-diameter conduit.
However, unless the test failed catastrophically, which would necessitate reassessment of the safety significance of the issues, the results of a such a 38
I d
test would not be dispositive and could not used to determine the fire resistance rating of Thermo-Lag.
The following are some of the factors that affect ampacity:
cable temperature rating, cable jacket material, conductor material and size, cable loading, the number of cables in a group, raceway components and construction, fire barrier materials, and the ambient temperature.
The review team does not believe that these factors are amenable to scaling.
Therefore, the team does not believe that ampacity derating factors can be derived from small-scale testing.
The special review team has concluded that the fire resistance ratings of Thermo-Lag fire barriers and their associated ampacity derating factors cannot be determined by small-scale tests.
The review team recommends, therefore, that small-scale testing be eliminated as an option for resolving the safety and technical issues associated with the fire resistance ratings and ampacity derating factors for Thermo-Lag fire barriers.
Full-Scale Confirmatory Tests:
Confirmatory testing would involve repeating or duplicating previously performed full-scale ampacity derating tests and NFPA 251 (or ASTM E119) fire endurance tests.
The results of confirmatory tests could be compared with the results of the tests that were duplicated to either substantiate or refute the fire resistance ratings and ampacity derating factors claimed by the vendor.
The special review team reviewed about 40 fire endurance test and nine ampacity derating test reports and found that without exception, the test reports lacked the level of detail needed to duplicate the tests.
For example, information on the test specimen dimensions, materials of construction, construction procedures, curing, and thermocouple placement were not documented in sufficient detail to allow true duplication of the test specimens.
None of the reports reviewed by the review team included dimensioned drawings.
In addition, it is not possible 39
to verify that some basic requirements of the test procedures, such as equipment cal.Abration, were performed by the testing laboratory.
Moreover, some elements of many of the test procedures,_such as furnace size and thermocouple size, do not meet the requirements of the test standards (NFPA 251 or ASTM E119).
The special review team is also concerned that the configurations of the test specimens for many of the previously performed tests are atypical of the field installations observed during the plant site visits.
The results of a fire endurance test are generally applicable only to the specific configuration tested.
Therefore, a confirmatory test of an atypical configuration would be of limited value in resolving concerns regarding the use of the results of previously performed tests to justify the field installations.
The special review team believes that these concerns render mest of the previously performed tests invalid for demonstrating compliance with the NRC's requirements and guidance and preclude confirmatory testing as an option for resolving the safety and technical issues associated with Thermo-Lag.
The review team believes that duplication of one of these tests would only provide another invalid test.
In summary, repeating questionable or invalid fire endurance and ampacity derating tests may provide limited insights into the accuracy and repeatability of the previous tests, but would not be useful for resolving the safety and technical concerns on fire resistance ratings and ampacity derating factors.
In view of the expense and effort required'to perform tests, the special review team does not believe that confirmatory tests would be an I
effective or efficient use of resources.
Finally, the team did not identify any advantages to performing confirmatory tests over original full-scale tests (discussed below).
40
.l
Full-Scale Modified Confirmatory Tests:
Full-scale modified confirnatory tests would be similar to the confirmatory testing.
discussed above.
However, the test procedure would be designed to compensate for unknowns and to correct errors or weaknesses in the test being duplicated.
The special review team believes, however, that compensating for the weaknesses with previous test reports, which were addressed above, would reduce the usefulness of confirmatory test results for comparisons with previous test results.
Moreover, the review team believes that the number of unknowns and problems associated with the previous tests would require compensation which, in effect, would result in such tests being original tests rather than confirmatory tests.
Modified confirmatory testing could also include, for example, side-by-side tests of Thermo-Lag fire barriers and another fire barrier product with known fire test results.
In this case, if the other product performed as expected, the overall results of the test would be more conclusive.
That is, if the non-Thermo-Lag product performs in accordance with known parameters, the staff would have fairly compelling evidence that the test was conducted properly and that the results of the Thermo-Lag barrier (pass or fail) are valid.
Before conducting side-by-side fire tests, it would be necessary to obtain baseline fire performance data to ensure that one of the products would not effect the performance of the other product during the side-by-side test.
This would require that the products be tested individually, which would add time and expense to the testing program.
Moreover, side-by-side tests would result in direct comparisons of competitive-products which could give one or the other product an unfair competitive l
advantage.
The review team believes that the problems associated with i
modified confirmatory testing are similar to those associated with confirmatory testing as discussed above, and that there is no clear advantage to conducting such tests to resolve the 41
t concerns with Thermo-Lag fire barriers.
The special review team does not believe that modified confirmatory tests would be the most direct and efficient means of resolving the safety and
' technical issues regarding the fire performance'and ampacity derating of Thermo-Lag fire barriers.
Oricinal Full-Scale Tests:
Original full-scale tests would be designed to determine the fire resistance ratings of the Thermo-Lag fire barrier configurations installed by the licensees to meet the NRC's requirements and the ampacity derating factors for
-t those fire barrier designs or configurations.
The fire tests would be conducted in accordance with NFPA 251 (or ASTM E119) using the NRC's acceptance criteria in GL 86-10.
The ampacity i
derating tests would be conducted in accordance with a standard or procedure recognized by the industry.
The following factors should be considered in the design of the program:
i Cable trays of various configurations (horizontal runs, vertical runs, stacked trays) sizes, materials (steel, _
aluminum), designs (solid bottom, ladder back), cable loadings (minimum fill, maximum fill); types of cables (power, control, and instrumentation) and types of cable insulation materials.
Conduits of various configurations, sizes, and materials (steel and aluminum) with accessory components such as couplings, pull boxes, and junction boxes.
Air drops.
Bus ducts, HVAC ducts, and dampers.
{
Structural supports.
Hatch covers, manhole covers.
Vaults, partitions, and walls.
The range of fire barrier system components and installation methods and techniques used by the licensees.
The special review team believes that absent the availability of complete and valid test results, which it did not find, the most practical means of addressing these factors is through an original full-scale test program conducted in accordance with NRC
- guidance, t
+
42 e
m a
The results of fire endurance tests are generally only. applicable to the specific fire barrier configurations that were tested.
Therefore, a series of tests that either addresses the. worst case j
fire barrier configurations or bounds the possible field' installations and configurations should be conducted to resolve the identified concerns.
For example, from the perspective of heat transfer through the barrier and into the raceway, a small fire barrier enclosure, say a 6-inch wide cable tray, is generally recognized as the worst case configuration.
The ability of the fire barrier material to limit the transfer of I
heat into the enclosure during exposure to the standard test fire is a major acceptance criterion.
Therefore, it stands to reason that a small enclosure should be tested.
From the standpoint of l
structural integrity, larger configurations, say a 36-inch wide cable tray, is the worst case configuration.
The ability of the test specimen to maintain its structural integrity is also required.
Therefore, a large enclosure should also be tested.
The special review team believes that if both the small and the large enclosures pass the tests in accordance with the NRC acceptance criteria, this would bound the range of conditions found in the field and would be acceptable as technical bases for the installation of the barriers by the licensees.
Similarly,
)
raceway materials, cable type and loading, and each of the other
)
factors identified above may effect the cutcome to=the tests.
Therefore, the test program should consider all of these factor.
For example, it may be necessary to test both aluminum and steel trays.
The fire resistance ratings of both 1-hour and 3-hour Thermo-Lag fire barriers are suspect.
Therefore, both 1-hour and 3-hour fire endurance tests of each configuration should be j
conducted.
l There are no consensus standards for ampacity derating tests.
However, the vendor and 3M have used procedures developed by 43
4
- ~
s ULaz and Bechtel.83 These procedures should be considered in the design and conduct OF the ampacity derating test program recommended by the review team.
TEST SPONSORSHIP The special review team considered the following sponsors for i
determining the fire resistance ratings and the ampacity derating factors for Thermo-Lag fire barriers:
(1) the NRC, (2) the vendor, (3) individual licensees, and (4) an industry group such as NUMARC.
NRC Soonsored Test Procram:
The review team believes that an NRC-testing program would have the following problems:
(1) the NRC would assume the vendor's and the industry's technical and financial responsibilities for determining the acceptability of a product for satisfying the NRC's requirements, (2) industry input into the program may be limited or nonexistent, (3) significant costs and staff resources would be required to develop and carry out a testing program, (4) the NRC does not have ready access to the materials required to construct the test specimens (cable trays, conduits, fire barrier materials, etc.), qualified Thermo-Lag fire barrier installers, or a test facility, (5) the r
appearance that NRC is validating one vendor's' product for'end use by the industry over other products could be interpreted as NRC endorsement of the product and could give the vendor an unfair competitive advantage if the NRC tests are successful, and (6) in this case, the licensees may rely on the results of NRC tests to justify the use of Thermo-Lag to meet the NRC's requirements.
The special review team believes that it is unlikely, absent significant industry involvement, that the NRC a2 UL Outline of Investigation, " Tests for Ampacity of Insulated Electrical Conductors Installed in Fire Protective Systems," Subject 1712, July 1984.
Draft Bechtel Specification, " Proposed Standard for the 83 Determination of Ampacity Derating of Fire Protected Cables,"
May 27, 1986.
44
could design and conduct a test program that would bound the 1
various configurations and installation procedures used by the licensees.
Therefore, even if the NRC sponsors a test program, the industry may have to conduct additional tests to completely resolve the issues.
Vendor Soonsored Test Proaram:
Failure of Thermo-Lag to provide the fire resistance ratings and to limit ampacity derating to the levels claimed by the vendor could adversely impact the vendor.
During the October 17, 1991 meeting at NRC Headquarters and again during the December 20, 1991 vendor inspection exit meeting, the vendor informed the staff that it would conduct additional testing if the staff believed such testing would resolve its concerns.
In light of the wrongdoing allegations discussed above, the special review team does not recommend this option at this time.
However, if the allegations are not substantiated by the OIG/OI task force, the staff should reconsider this option.
Individual Licensee Sponsored Test Procram:
Most of the licensees use Thermo-Lag fire barriers to satisfy the NRC's requirements for protecting safe shutdown capability from fire.
Failure of Thermo-Lag to provide the fire resistance ratings and to limit ampacity derating to the levels claimed by the vendor could result in financial and regulatory impacts on those licensees.
Therefore, the licensees should be involved early in the process to avoid second guessing of the NRC's testing program and criticism in the timely release of potentially safety significant information.
However, in terms of depth of understanding of the problem, as well as timeliness and cost effectiveness, the review team does not believe that it would be advantageous for individual licensees to investigate the generic concerns with the fire performance capabilities and ampacity derating of Thermo-Lag materials.
However, individual licensees may have to conduct confirmatory testing and perform engineering analyses to resolve plant specific issues after the generic concerns are resolved.
45
. = ~
Industry Soonsored Test Procram:
If the testing program is designed and carried out by an industry group, (1) the cost of testing will not be incurred by the NRC, but will be spread'out among the licensees, (2) the licensees can provide the materials required to perform the tests from stock (raceways, cables, Thermo-Lag) and the qualified installers (licensee employees) required to construct the test specimens with minimal effort and expense, and (3) the burdens of reviewing and resolving the technical, safety, and regulatory issues are placed largely with the end users--the nuclear industry.
The activities required to resolve the identified concerns should have been carried out by the individual licensees during the licensing process.
Given the number of licensees relying on Thermo-Lag to meet the NRC's requirements and the potential adverse regulatory and financial impacts on those licensees should Thermo-Lag fire barriers fail to provide the levels of fire protection claimed by the vendor and intended by the licensees, the review team recommends that industry group, such as NUMARC, develop and perform a test program.
A coordinated nuclear industry solution with NRC involvement would provide the most well thought out approach to solving the problems identified by the review team.
Therefore, if additional testing is needed to resolve the concerns and this recommendation is adopted, the NRC should be involved at various stages of the program.
For example, the staff should (1) work with industry to ensure that all of the problems are identified, (2) review the test program and procedures at several stages and before the industry conducts any fire endurance or ampacity derating tests, (3) witness the construction of the test assemblies, (4) witness the conduct of the tests, and (5) review the test results.
The staff should also monitor the vendor's involvement, if any, in the test program.
NIST has determined that only a limited number of testing laboratories in the United States and Canada have the facilities and equipment required to conduct full-scale ASTM E119 and 46
ampacity derating tests.
The NRC generally' accepts the use of UL listed products to meet NRC fire protection requirements.
The Thermo-Lag 330-1 fire barrier uystem is not UL listed.
- However, UL has a listing service for these types of fire barriers and has
' tested and listed 3M's electrical circuit protective systems.s5 UL also has experience in conducting ampacity derating tests.
The special review team recommends that testing by UL be given serious consideration.
In summary, the special review team recommends that industry be required to substantiate the fire resistance ratings of the Thermo-Lag Lire barriers used to satisfy the NRC's requirements and the ampacity derating factors associated with those barrier dec!gns.
If industry cannot produce existing test results that sub antiate these parameters, the team recommends that an indust;y group, such as NUMARC, take the lead for developing and carrying out a test program acceptable to the staff.
It is also recommended that the NRC be involved in the review and oversight of the program at specific milestones.
Although the special review team recommends that industry perform any new testing needed to resolve the safety and technical issues identified by the team, the OIG/OI task force has expressed an interest in new tests for investigative purposes.
If the task force can define and justify such tests, the review team recommends that the staff consider supporting the tests, to the extent that the support will not impede resolution of the identified safety issues.
CONCLUSIONS UL Outline of Investigation, " Fire Tests for Electrical 8'
Circuit Protection Systems," Subject 1724, May 1984.
85 UL Buildino Materials Directory, " Electrical Circuit Protective Systems (FHIT)," December 14, 1990, pp. 202 - 215.
47
o The special review team found the following regarding the use of Thermo-Lag by the licensees:
The fire resistance ratings and the ampacity derating factors for the Thermo-Lag 330-1 fire barrier system are indetermini e.
Some licensees have not adequately reviewed and evaluated fire endurance test results and ampacity derating test results to determine the validity of the tests and the applicability of the test results to their plant designs.
Some licensees have not adequately reviewed installed fire barrier configurations to ensure that they either replicate the tested configurations or provide an equivalent level of protection.
Some licensees used inadequate or incomplete installation procedures during the construction of their Thermo-Lag barriers.
Although there is'a history of raceway fire barrier qualification and installation problems, the staff's.
licensing reviews and inspections of the barriers may not have been commensurate with either the problems or the importance of the barriers.
RECOMMENDATIONS The special review team recommends the following:
Advise industry of the staff's concerns regarding Thermo-Lag fire barriers through a public meeting with the Nuclear Utilities Management and Resources Council (NUMARC).
Encourage NUMARC to coordinate a response that verifies the fire resistance rating of the Thermo-Lag fire barrier configurations installed by the licensees to meet the NRC's 48
44 w
requirements and the ampacity derating factors for those configurations.
Issue a generic letter (Attachment 2) that discusses the concerns and requires the licensees to provide information needed by the staff to verify compliance with 10 CFR 50.48, GDC 3, and GDC 17.
If new tests are needed to verify the fire ratings and the ampacity derating factors, provide staff oversight of the industry's test program and review the licensees' corrective
. l action plans for resolving any plant-specific fire barrier design, evaluation, and installation issues.
Terminate the special review team and assign SPLB the lead (as primary review branch) for prioritization and resolution of the open items identified in Attachments 4 and 5, for
{
generic letter follow-up activities, and for continuing interface with the OIG/OI task force.
Assess the NRC's fire protection review and inspection programs to determine if adjustments are needed.
Ensure that all allegations received by the OIG/OI task force are handled in accordance with Commission policy.
(The NRC Allegations Program Manager will follow-up.)
Formally advise other government agencies and foreign nuclear industries of the staff's concerns.
LIST OF ATTACHMENTS 1.
Memorandum from T. E. Murley, NRR, to F. J. Miraglia, L.
R.
Plisco, and K.
S. West, NRR, August 15, 1991 2.
Draft Generic Letter 92-XX, "Thermo-Lag Fire Barriers,"
February 11, 1992 3.
List of Nuclear Power Plants That May Use Thermo-Lag 330-1 Fire Barriers to Satisfy NRC Requirements 49
ATTACHMENT T
/
]o, UNITED STATES
(
NUCLEAR REGULATORY COMMISSION
,.e t
WASHINGTON. D. C 20555 a
E
,/
August 15, 1991 MEMORANDUM FOR:
Frank J. Miraglia, Deputy Director Office of Nuclear Reactor Regulation Loren R. Plisco, Section Chief Performance and Quality Evaluation Branch Division of Licensee Performance and Quality Evaluation, NRR Steven K. West, Project Manager Technical Assistant Management Section, NRR FROM:
Thomas E. Murley, Director Office of Nuclear Reactor Regulation
SUBJECT:
SPECIAL REVIEW ASSIGNMENT AND PLAN FOR RESPONSE TO ALLEGATIONS OF POTENTIAL SAFETY CONCERNS REGARDING THERMAL SCIENCE, INC.
REFERENCE:
Memorandum dated February 21, 1991, from confidential alleger to Russell Wise This memorandum supercedes my July 12, 1991 memorandum on this subject.
You are hereby assigned the responsibility to conduct a review of the allegations contained in the reference memorandum. Your primary objective is to identify any issues of safety significance that could affect the continued safe operation of those plants using "Thermolag " Mr. Loren Plisco is to be the leader of the review process.
Your guidelines for the conduct of this review are as follows:
1.
Based upon your review of the available information, identify any safety-related issues and determine whether these issues had been previously addressed by the NRC. Appropriate documentation should be identified.
2.
Determ.ine, based upon your experience and knowledge of NRC requirements and programs, whether there is a reasonable basis for determining that the safety issues associated with these matters were adequately addressed.
3.
For those issues that, based upon your experienc:: and knowledge of NRC requirements and programs, the available information would not provide an adequate basis for resolving the safety concerns, recommend appropriate actions.
4 Your recommendations for resolving the safety concerns, if any, should be discussed with Frank J. Miraglia prior to completing your final report.
P
5.
During the conduct of the review, you shall protect the identity of the alleger. The identity of the alleger should be treated as privileged information and should not be divulged to any persons in or outside the NRC except as authorized by Frank J. Miraglia, NRR, in consultation with OIG/01 Joint Task Force.
6.
The OIG/0I' Joint Task Force has an investigation on this matter underway, with technical support from the EDO.
It is important that technical support of this activity be provided.
If additional support is required, Mr. Miraglia will be nottfied so that appropriate arrangements can be made.
It is equally important that the technical interface between the review team and the OIG/0! Joint Task Force be maintained. Loren Plisco will coordinate daily activities with the Chairman, Joint Task Force.
7.
The review team will develop plans for independent testing of the material in question, if deemed necessary. The plans and conduct of the testing will be coordinated with the Office of Nuclear Regulatory Research.
8.
In carrying out the assignment, the review team will report directly to Frank J. Miraglia, Deputy Director, Office of Nuclear Reactor Regulation.
9.
Frank J. Miraglia will coordinate with the Chairman, OIG/0! Joint Task Force and keep the EDO informed as appropriate.
- 10. To review the issues contained in the reference document, it will be necessary to collect and review many documents and will require contact with NRR personnel and representatives from various vendors and laboratories.
All plans regarding discussions with personnel inside or outside the agency should be reviewed with and authorized by Frank J. Miraglia.
- 11. On completion of your review, provide a report of your findings and recommendations to the Director, Office of Nuclear Reactor Regulation.
The Director, NRR, should notify the EDO of actions he plans to take.
The EDO will keep the Comission informed as appropriate.
Thomas E. Murley, Director Office of Nuclear Reactor Regulation cc:
J. Taylor, EDO B. Hayes, 01 J. Sniezek, EDO D. Williams, IG
o I
3 ATTACHMENT 2 i
4 DRAFT GENERIC LETTER February 11, 1992 TO:
ALL HOLDERS OF OPERATING LICENSES OR CONSTRUCTION PERMITS FOR NUCLEAR POWER REACTORS.
SUBJECT:
THERMO-LAG FIRE BARRIERS (GENERIC LETTER 92-XX)
Purpose The U.S. Nuclear Regulatory Commission (NRC) is issuing this generic letter to require licensees to provide information to verify that Thermo-Lag 330-1 fire barrier systems manufactured by Thermal Science, Incorporated (TSI, the vendor), St. Louis, Missouri, comply with the NRC's requirements.
The NRC reviewed Therr -Lag 330-1 fire barrier systems after receiving reports from Gulf States Utilities (GSU) that these systems had failed qualification fire tests and had installation problems.
The NRC reviewed fire endurance and ampacity derating test reports, installation procedures, and as-built configurations and identified the following concerns regarding Thermo-Lag fire barriers:
test results that are incomplete or indeterminate, installations that are not constructed in accordance with the vendor's installation procedures, incomplete installation procedures, and as-built fire barrier configurations that may not be qualified by a valid fire endurance test or justified by an engineering analysis.
The NRC is concerned that licensees may not be meeting the requirements of Section 50.48,
" Fire protection," and General Design Criterion (GDC) 17,
g GENERIC LETTER 92-XX " Electric power systems," of Appendix A,
" General Design Criteria for Nuclear Power Plants," to Part 50 of Title 10 of the Code of Federal Regulations (10 CFR Part 50).
The NRC is requiring information on compliance with 10 CFR 50.48, GDC 17, and associated license conditions under the provisions of 10 CFR 50.54 (f).
Qualification Recuirements for Fire Barriers Section 50.48 requires that each operating nuclear power plant have a fire protection plan that satisfies GDC 3, " Fire protection."
GDC 3 requires that structures, systems, and components important to safety be designed and located to minimize, in a manner consistent with other safety requirements, the probability and effects of fires and explosions.
Systems associated with achieving and maintaining safe shutdown conditions are of major importance to safety because damage to these systems can lead to core damage.
Fire protection features required to satisfy GDC 3 include features to ensure that one train of those systems necessary to achieve and maintain hot shutdown conditions be maintained free of fire damage.'
one means of complying with this requirement is to separate one safe shutdown train from its i
See Appendix R to 10 CFR Part 50, " Fire Protection Program for Nuclear Power Facilities Operating Prior to January 1, 1979."
J GENERIC LETTER 92-XX redundant train with fire-rated barriers.
The level of fire resistance required of the barriers depends on the other~ fire protection features provided in the fire area.
The NRC provided guidance on acceptable methods of satisfying the regulatory requirements of GDC 3 in Branch Technical Position (BTP) Auxiliary and Power Conversion Systems Branch (APCSB) 9.5-t 1,
" Guidelines for Fire Protection for Nuclear Power Plants;"
Appendix A to BTP APCSB 9.5-1; BTP Chemical Engineering Branch (CMEB) 9.5-1,
" Fire Protection For Nuclear Power Plants," July-1 1981; and Generic Letter (GL) 86-10, " Implementation of Fire Protection Requirements," April 24, 1986.
In the BTPs and GL 86-10, the staff stated that the fire resistance ratings of fire barriers should be established in accordance with National Fire Protection Association (NFPA) Standard 251, " Standard Methods of Fire Tests of Building Construction and Materials," by subjecting a test specimen that represents the materials, workmanship, method of assembly, dimensions, and configuration for which a fire rating is desired to a " standard fire exposure" at a 2
nationally recognized laboratory.
In GL 86-10, the staff also provided guidance on the acceptance criteria for fire tests and on evaluations of deviations from tested configurations.
2 American Society for Testing and Materials (ASTM) Standard E119 was adopted by the National Fire Protection Association (NFPA) as NFPA Standard 251.
i
i GENERIC LETTER 92-XX 3DC 17 requires that onsite electric power systems be provided to permit functioning of structures, systems and components important to safety.
The onsite electrical power system is required to provide sufficient capacity and capability to ensure that vital functions are maintained.
The Institute of Electrical and Electronics Engineers (IEEE) Standard 279, " Criteria for Protection Systems for Nuclear Power Generating Stations,"
provides guidance regarding acceptable methods of satisfying GDC 17.
IEEE 279 states that the quality of the protection system components shall be achieved by specifying requirements, such as for the derating of components.
Areas of Concern Fire Endurance Testina and ADolication of Test Results Many fire endurance tests have been conducted on electrical raceways protected with Thermo-Lag 330-1 fire barrier systems.
Although many of the test reports document results that meet the NRC's temperature acceptance criterion discussed in GL 86-10, some test assemblies have failed, such as the assembly tested.in October 1989 at the Southwest Research Institute (SwRI) and discussed in NRC Information Notice (IN) 91-47, " Failure of Thermo-Lag Fire Barrier Material to Pass Fire Endurance Test."
The NRC has reviewed approximately 40 1-hour and 3-hour fire
3 GENERIC LETTER 92-XX endurance test reports involving Thermo-Lag 330-1 fire barrier systems and has found that the test assemblies met the NRC's temperature acceptance criteria when the test article protective envelope was constructed by TSI personnel using TSI's installation procedures.
However, the NRC has found other Thermo-Lag 330-1 fire barrier test assemblies that failed to meet the NRC's temperature acceptance criterion.
In most cases, the test assemblies that failed were either constructed by the licensee's or contractor's qualified installers, or did not follow TSI's installation procedures.
In the fire endurance test conducted in October 1989 at SWRI, the test article that failed was constructed'by TSI-certified licensee personnel using TSI's installation procedures.
The NRC reviewed fire test reports from various testing facilities and found that testing methods and procedures used during some of the qualification tests did not meet the NRC's guidance.
NFPA 251 advises that the test conditions should be j
evaluated carefully because variations from the construction or conditions that are tested may substantially change the performance characteristics of the assembly.
The test reports reviewed did not contain sufficient details of the construction methods used for the test article, did not contain details of'the materials used, did not contain dimensioned drawings, and documented test configurations that were atypical of as-built configurations.
i GENERIC LETTER 92-XX In GL 86-10, the NRC provided guidance on deviations from tested fire barrier configurations.
While reviewing the Thermo-Lag fire f
barriers, the NRC staff found several instances in which l
licensees installed fire barrier configurations that may not have t
been qualified by fire endurance testing or justified by engineering analysis.
For example, when the NRC conducted its review, some licensees could not justify their practice of j
extrapolating test results from small barrier enclosures to i
significantly larger enclosures, or installing barriers ucing procedures and materials that were different from those tested.
i The NRC visited site after issuing IN 91-47, and also found
)
i several licensees that had constructed fire walls, partitions, and vaults using Thermo-Lag as a component.
These licensees could not provide qualification tests or engineering analyses of deviations from tested configurations to demonstrate the acceptability of these fire barriers.
Aroacity Deratino Desian Basis Cables enclosed in electrical raceways protected with fire barrier materials are derated because of the insulating effect of the fire barrier raterial.
Other factors that affect ampacity derating include cable fill, cable loading, cable type, raceway construction, and ambient temperature.
The National Electrical Code, Insulated Cable Engineers Association publications, and other industry standards provide general ampacity derating
i, g
GENERIC LETTER 92-XX factors, but do not consider the effects of passive fire barrier systems.
Although a national standard ampacity derating test method has not been established, ampacity derating factors for raceways enclosad with fire barrier material have been determined by testing.
TSI has documented a wide range of ampacity derating factors that were determined by testing.
For example, TSI provided test reports to licensees that document ampacity derating factors for cable trays that range from 7 percent to 28 percent for 1-hour barriers and from 16 percent to 31 percent for 3-hour barriers.
On October 2, 1986, TSI informed its customers by Mailgram that, while conducting tests in September 1986, at the Underwriter Laboratories (UL) facilities, TSI found that the ampacity derating factors for Thermo-Lag barriers were greater than previous tests indicated.
However, the test procedure and test configuration differed from previous tests, and the results from the different tests may not be comparable to each other.
The NRC is concerned that licensees may be using nonconservative ampacity derating factors since the tested configurations may not represent as-built configurations.
The NRC learned during its review that testing conducted at SwRI found the ampacity derating as 37 percent for a 1-hour barrier.
Deficiencies in the Installation and Inspection Procedures 1
GENERIC LETTER 92-XX.
While conducting site visits after issuing IN 91-47, the NRC staff observed a number of installations that were not in accordance with TSI's installation procedures and some installations that did not appear to be qualified by fire endurance testing or an engineering analysis.
In IN 91-79,
" Deficiencies in the Procedures for Installing Thermo-Lag Fire Barrier Materials," the NRC staff discussed installation problems resulting from TSI's incomplete installation procedures, licensecs' inadequate installation procedures, installer errors, incomplete or incorrect design documents, and inadequate quality control oversight.
In IN 91-79, the staff listed the installation details in which it found dif ferences.
Actions Covered by this Generic Letter The NRC's regulations require that safe shutdown equipment be protected from fire.
The NRC has found qualification test failures, test results that are indeterminate, installation problems, and differences between reported ampacity derating factors.
Therefore, the licensees should confirm that Thermo-Lag 330-1 fire barrier systems have been qualified by representative fire endurance ano ampacity derating testing and that these qualified barriers have been installed with appropriate quality controls to ensure that they comply with the NRC's requirements.
Reportina Recuirerents
GENERIC LETTER 92-XX All addressees are required, pursuant to Section 182a of the Atomic Energy Act of 1954, as amended, and 10 CFR Section 50.54(f), to provide a written report within 120 days from the date of this generic letter.
In this written report, the licensee shall:
(1)
State that it has identified all fire barriers using TSI's Thermo-Lag 330-1 to meet 10 CFR 50.48 or that it does not use Thermo-Lag 330-1 at the facility to meet this requirement.
~
(2)
State that it has qualified the Thermo-Lag 330-1 fire barriers by conducting a fire endurance test in accordance with NFPA 251 or in accordance with previous licensing commitments.
(3)
State that it has constructed the as-built Thermo-Lag 330-1 fire barriers in accordance with the procedures used to assemble the qualification test article and that the as-built fire barrier configuration represents the materials, workmanship, method of assembly, dimensions, and configuration of the qualification test assembly configuration or that the licensee has analyzed the deviations from the tested configuration.
(4)
State that the design basis for the ampacity derating
RGENERIC LETTER 92-XX.
factors used for all raceways protected by Thermo-Lag 330-1 is consistent with the as-built configuration and that representative ampacity derating test results have been-reviewed for applicability.
(5)
List any necessary corrective actions and a schedule for any deficiencies identified while conducting the actions described above and describe any compensatory measures taken in accordance with technical specifications or administrative controls.
(6)
List any Thermo-Lag 330-1 fire barriers that cannot be verified in accordance with reporting requirements (1) through (5), provide a justification for continued operation until such time as the identified barriers can be verified, and provide a schedule for completing the verifications.
The licensee should retain all documentation of any reviews performed to satisfy the reporting requirements for any future NRC audit.
If the addressee cannot provide the information required or meet the reporting deadlines, it shall include in the response a justification for alternative approaches and schedules.
The NRC i
encourages licensees to work together to develop acceptable generic solutions to the problems addressed in this generic
i J
s I
GENERIC LETTER 92-XX letter.
The written reports required shall be addressed to the U.S.
Nuclear Regulatory Commission, ATTN: Document Control Desk, Washington, D.C. 20555 under oath or affirmation.
A copy of the report shall also be submitted to the appropriate regional administrator.
Backfit Discussion The NRC is requiring information that will enable the NRC staff to determine if licensees are complying with 10 CFR Section 50.48.
The staff is not establishing a new position regarding compliance in this generic letter.
Accordingly, this generic.
letter does not constitute a backfit.
Thus, 10 CFR 50.109 does not apply, and no backfit analysis need be prepared.
Reauest for Voluntary Submittal of Imoset Data This request is covered by Office of Management and Budget Clearance Number 3150-0011, which expires May 31, 1994.
The estimated average number of burden hours is 200 person-hours for each addressee's response, including the time required to assess the requirements for information, search data sources, gather and analyze the data, and prepare the required letters.
This estimated average number of burden hours pertains only tc the
~
a GENERIC LETTER 92-XX identified response-related matters and does not include the time to implement the actions required to comply with the applicable regulations,_ license conditions, or commitments.
Comments on the accuracy of this estimate and suggestions to reduce the burden may be directed to Ronald Minsk, Office of Information and Regulatory Affairs (3150-0011), NEOB-3019, Office of Management and Budget, Washington, D.C.
20503, and to the U.S. Nuclear Regulatory Commission, Information and Records Management Branch, Division of Information Support Services, Office of Information and Resources Management, Washington, D.C.
20555.
Although not required, the following information would assist the NRC in evaluating the cost of complying with this generic letter:
(1) the licensee staff's time and costs to perform requested inspections, corrective actions, and associated testing (2) the licensee staff's time and costs to prepare the required rannrta and documentatjnn (3) the additional short-term costs incurred as a result of the inspection findings such as the costs of the corrective actions or the costs of down time (4) an estimate of the additional long-term costs that will be
g g
s GENERIC LETTER 92-XX incurred in the future to implement commitments such as the estimated costs of conducting future inspections or increased maintenance If you have any questions about this matter, please contact one of the technical contacts or the lead project manager listed below.
Sincerely, James G. Partlow Associate Director for Projects Office of Nuclear Reactor Regulation 4
Enclosure:
List of Recently Issued Generic Letters Technical contacts:
Pat Madden, NRR 301-504-2854 i
Ralph Architzel, NRR i
301-504-2804 i
Lead Project Manager:
4 e
e.
ATTACHMENT 3 List of Nuclear Power Plants That May Use Thermo-Lag 330-1 Fire Barriers to Satisfy NRC Requirements l
Plants Identified by TSI as Usina Thermo-Laa Arkansas Nuclear Hatch San Onofre Beaver Valley Indian Point Sequoyah Braidwood LaSalle Shearon Harris Browns Ferry Limerick Shoreham Brunswick Maine Yankee South Texas Byron Millstone Summer Callaway Nine Mile Point Surry Clinton North Anna Susquehanna Comanche Peak Oyster Crcek Three Mile Island D.C.
Cook Palisades Trojan Cooper Palo Verde Turkey Point Crystal River Peach Bottom Vermont Yankee Davis Besse Perry Vogtle Diablo Canyon Prairie Island WNP2 Duane Arnold Rancho Seco Watts Bar Enrico Fermi River Bend Wolf Creek Grand Gulf St. Lucie Zion Additional Plants Identified by the Allecer as Usina Thermo-Lac
'l Bellefonte McGuire o
Calvert Cliffs Monticello i
Catawba Oconee Dresden Pilgrim Fitzpatrick Point Beach Ginna Quad Cities Haddam Neck Waterford Kewaunee Yankee Rowe Lacrosse Appendix VIII to letter from R.
Feldman, TSI, to 2
F.J. Miraglia, NRR, October 5, 1991.
. c; ATTACHMENT 4 HUMMARY OF GENERIC SAFETY IDSUES This attachment identifies each major generic issue identified by the Special Review -Team.
The table provides a summary of ' the information collected and recommendations for follow-up corrective action.
The recommended corrective actions are based on the information known to date, and may need to be revised upon receipt of additional information.
A detailed discussion of all of the specific technical issues is included'in Attachment 5.
The issue numbers of this table correspond with the sections in Attachment 5.
i i
t
SUMMARY
OF GENERIC SAFETY ISSUES ISBUE FACT 8 CORRECTIVE ACTION
- 1. Are the correct ampacity 1.
Utilities are using a 1.
Conduct further deratings for Thermo-lag range of ampacity deratings evaluation of importance of being used by the industry?
provided by TSI.
ampacity derating.
2.
The ampacity derating 2.
Notify industry of the investigation at UL was issue in generic letter.
conducted by TSI and 3.
Conduct an ampacity Bechtel.
derating test of THERMO-3.
SwRI (3M) conducted LAG.
tests that show a
much higher derating.
4.
Some licensees have not reviewed ~ the revised TSI derating factors.
2.
Will THERMO-LAG provide 1.
The tests conducted to 1.
Notify industry of a 1-hr or 3-hr fire barrier date without TSI issue in generic letter.
if installed properly?
Has involvement have failed.
2.
Identify any other test the material been properly 2.
There has not been a reports not reviewed.
l tested?
test that passed without 3.
Conduct a NFPA 251 fire l
involvement of TSI.
endurance test of THERMO-3.
ITL did not provide LAG.
third party oversight of the TSI tests.
l 3.
Are there a wide-spread 1.
Two Information Notices installation problems with as-built
' configurations have been issued to discuss.
THERMO-LAG?
failed.
issues (91-47 and 91-79).
2.
Only TSI installed test 2.
Notify industry of the l
rigs have. passed.
Issue in generic letter.
3.
A number of plants have 3.
Verify licensee had installation problems.
corrective action by on-4.
The vendor has stated site inspections as that the procedure needs necessary.
i revised.
s O
e e
m
.. - c-4.
Are toxic gases given 1.
SwRI testing shows high 1.
Review and evaluate the
(
off by THERMO-LAG wl.. ~
levels of toxic gases, final NIST test data.
exposed to fire?
2.
NIST conducted toxicity 2.
Provide data and testing of Thermo-Lag and allegation information to found toxic gases such as OSHA.
HCN and CO.
3.
NIST preliminary review of test data found Thermo-Lag was toxic, but no more so than other plant materials.
m aw.
~
y ve-a m
e-
-m,~
w w m
ws.
u---
.. g t
. ATTACHMENT 5 SPECIFIC TECHNICAL AND PROGRAMMATIQ ISSUBS.IDENTIFIND BY.TER dPECIAL RBYIEW TEAM i
This attachment identifies each specific technical and programmatic issue found by the.Srncial Review Team, and those issues referred.
?
by the OIG/OI Investi stive Task Force.
The' discussion of each i
issue includes a descr. '. ion of the issue, appropriate references, information gathered to date, the conclusion of the review team based on the current information, and recommendations for any necessary-corrective action.
The Special -Review-Team's recommendations for corrective action are based on information available to date and may need to be revised if additional information becomes available.
'i i
e
.t 5
i h
n
SPECIFIC TECHNICAL AND PROGR WW4 TIC ISSUES IDENTIFIED BY THE SPECIAL REVIEW TEAM 1.
AMPACITY DERATING FACTORS
1.1 Issue
Ampacity Deratina Testina and Factors Used How are the ampacity derating factors advertised by TSI being used by the licensees?
What factors are being used by the licensees? What are the correct ampacity derating factors for Thermo-Lag? Were the tests conducted by TSI valid? What test standard should be used?
References:
TSI October 2, 1986, Mailgram UL Test Report of January 21, 1987 ITL Ampacity Test Reports SwRI Ampacity Test Report Information: South Texas had a confirmatory ampacity derating test conducted at UL, under the control of TSI in September 1986.
The results of this test were sent to the NRC and all of TSI's customers in a mailgram.
The mailgram stated that a test was conducted, that the results were different that his previous numbers, and that an analysis of the results was forthcoming.
Apparently, the mailgram was sent the licensee procurement departments.
Site visits have found that the mailgram may not have been reviewed by. licensee engineering organizations.
Since there is no national standard for the
- test, the results of the test can be debated.
3M, a
competitor, conducted testing at SwRI and found even higher derating factors.
The UL test article was constructed by TSI personnel in a manner that was not consistent with the TSI procedures.
TSI initially claimed an ampacity derating factor of 10-17 but 3M representatives have alleged their tests
- percent, indicate a factor of 30-40 percent for 1-hour and 45 percent for the 3-hour system.
==
Conclusions:==
There is a wide variation in the numbers reported, as well as the panel thicknesses, which may affect the derating.
Some licensees are using the revised numbers, and others are using the numbers originally provided by TSI.
Testing independent of the vendor has not been conducted, with the exception of the 3M test conducted at SwRI.
The potential impact of this data is significant since it may involve replacing cable for safety related equipment.
There is not a national standard test method and the test results may be highly dependent on a number of variables that are not well controlled or documented during the test.
Recommendation / Status:
Conduct ampacity testing of the material using a recognized testing procedure.
1
a w
1.2 Falsification of Ampacity Test'Results It has been alleged by a former TSI employee that ampacity tests at TSI were altered by cutting holes in the-barrier and putting an air cooling system in the tray assembly.
References:
Interviews.
UL ampacity test results 3M ampacity test results Information:
Testimony indicates that the ampacity derating factors advertised by TSI were developed from tests of non-representative test articles.
The original TSI (ITL) test reports, and the test conducted using UL facilities, are suspect (see 1.1).
==
Conclusions:==
None to date.
Recommendations / Status:
OI investigation has not been i
completed, and the allegation is still being reviewed.
Results of ampacity derating tests could provide additional information regarding the allegation.
2.
FIRE ENDURANCE TESTING AND APPLICATION OF TEST RESULTS
2.1 Issue
Fire Test Acceptance Criteria What are the test acceptance criteria being applied by the NRC to fire endurance tests?
Is the NRC interpretation of ASTM E119 correct?
Has the NRC consistently applied the position stated in GL 86-10?
References:
Generic Letter 86-10 ASTM E119 and NFPA 251 LPL letter to NRC dated 2/14/83 Information:
During the site visit to RBS, the licensee stated that they were aware of GL 86-10, but were not familiar with the 250
- F above ambient temperature acceptance criterion.
The task force found that most licensees used the.
250
- F criterion during their testing.
Waterford asked the NRC to address the difference between the NRC position and ASTM E119 in 1983, but the task force could not determine if the NRC ever responded to Waterford in writing.
GL 86-10 appears to clarify the position by stating that a maximum individual temperature of 250
- F above ambient should be applied, although this is more conservative than the average temperature criterion in ASTM E119.
==
Conclusions:==
Generic Letter 86-10 describes the NRC acceptance criterion. and the basis for selection of the i
temperature.
In
- addition, it provides guidance for engineering analyses if the temperature is exceeded.
{
2 i
u a
Discussions with former NRC reviewers found that the single point criterion was applied during licensing reviews, recent reviews may have accepted the average temperature.
Recommendation / Status:
Formalize the NRC staff position in writing prior to conducting fire endurance testing to resolve allegations.
2.2 Issue
ANI Bulletin #5 Why do many licensees consider the ANI Bulletin as the sole test standard for the fire endurance tests rather than NRC guidance and ASTM E-1197
References:
ANI Bulletin #5 GL 86-10 Information:
During the meeting with the vendor on' october 17, 1991, the vendor stated that the utilities required the use of the ANI standard for its testing.
The NRC has never officially endorsed the ANI test standard.
Review of vendor test reports found that all of the tests were based on the ANI test standard.
Interviews with ANI personnel found that since 1986 ANI is no longer involved in the approval of fire barrier tests.
==
Conclusions:==
The NRC has no difficulty with licensees supplementing the test requirements of ASTM E119 with additional requirements. However,- the ANI standard alone does not appear to provide for a test acceptable to NRC, since ANI does not incorporate temperature acceptance criterion, nor provide sufficient details on conducting the test to be a stand-alone document.
Recommendation / Status:
No further action needed since ANI is no longer involved.
However, this issue is the cause of some of the other issues listed below.
2.3 Issue
Flamina of Thermo-La.g Is the flaming of Thermo-Lag noted in qualification testing a concern?
Does the material have a flame spread of less than 25 as required by NRC?
References:
Interview with CTL staff Observations of NIST testing Information:
Pictures from the January 1987 Thermo-lag test conducted at CTL show the exposed surf ace flaming. Two of the three maximum cable temperatures measured during the 1-hr test were 209'F and 251*F.
The toxicity testing conducted at NIST also noted that the material would flame for 7-15 minutes after the heat source was removed.
The NRC requires that all 3
e s
materials have a flame spread rating of less'than 25.
==
Conclusions:==
Insufficient information available to draw conclusion. The flaming of the material appears abnormal, but does not necessarily mean that the product is unacceptable.
Recommendation / Status: Observe material during fire endurance testing.
Verify the flame spread rating of less than 25 and non-combustible rating.
2.4 Issue
ExtraDolation of test results What are the limits on the use of extrapolation?
Can the results of tests conducted on 12" cable trays be extrapolated to 30" trays?
Can small enclosures be extrapolated to large vaults and walls?
Were tested configurations sufficiently representative of all in-plant configurations as required by ASTM E1197
References:
ASTM E119 GL 86-10 Information: Tests conducted at SwRI in October 1989 on large (30") aluminum cable trays failed.
Large vault and wall configurations have been observed at plants, but only small scale tests have been performed.
GL 86-10 provides guidance on conducting evaluations for configurations that are different from the tested configuration.
Test -reports reviewed do not encompass the minimum and maximum type configuration enclosures.
==
Conclusions:==
In some cases, licensees have not tested configurations representative of the as-built configurations.
NIST has been requested to provide technical guidance on this-issue, but no conclusions have been reached. However, several licensees that were visited do not have evaluations for their configurations.
Recommendation / Status:
Conduct testing of the Thermo-Lag material using small and large com Lqurations.
Have NIST provide technical input on extrapola.clon.
2.5 Issue
Testina Variables To what extent does cable fill ratio, cable size, and cable jacket insulation need to be considered in qualifying tested configurations to as-installed configurations?
References:
ASTM E119 GL 86-10 Information:
Certain test reports specify only the minimum cable fill or don't specify cable jacket material ignition temperature.
Generic cables were used in some tests, but the 4
-results were used by a large variety of plants.
Some tests were conducted with one bare conductor in the tray.
==
Conclusions:==
None to date.
Some licensees have not evaluated these variables to determine if test results are applicable to as-built configurations.
Recommendation / Status:
Have NIST provide technical input on test article construction.
Resolve issue in the test program development process.
2.6 Issue
Optional Products What effect do the optional products have on fire endurance capability and ampacity derating?
Have they been adequately qualified by testing?
Are the installed configurations consistent with the tested configurations?
References:
ITL Test Reports Information: The vendor sells several optional products, such as topcoat and cure accelerator, for the construction of the fire barriers.
Several plants use these optional products.
However, the task force was not able to identify that these products have been tested with the barrier in a fire endurance test to determined its effect on the rating of the barrier.
The vendor provided some test reports related to the optional products, such as a UL test on the cure accelerator, but they were not conclusive.
==
Conclusions:==
The optional materials have not been fire tested.
Recommendation / Status:
Conduct fire endurance testing of configurations with these materials, if they are used.
2.7 Issue
Thickness Tested Should the as-installed fire wrap enclosure material have a specified minimum and maximum thickness within the same range of the tested enclosures? What range of thicknesses have been tested?
Were the thicknesses adequately documented in the test reports?
References:
ITL Test Reports CTL Test Reports Perry Trip Report River Bend Trip Report Information:
Inconsistencies exist between the installation procedures,
- drawings, and test reports regarding the acceptable variance in thickness of the Thermo-lag barrier.
Several test reports only specify a minimum thickness, and some specify a nominal thickness.
However, site visits have 5
v confirmed that the licensees have received panels for installation that have a very wide range.
For example, a 1-hour panel was observed to have thicknesses measurements of 3/8 to 1 1/4 inches.
==
Conclusions:==
The ITL test reports do no provide sufficient detail for a licensea to determine if its installation is similar to the tested configuration with regard to thickness.
Recommendation / Status:
Determine, by testing, the impact of thi Wss differences using panels typical of those installed.
2.8 Issue
Fire Test Failures l-Did the test failures at SwRI, CTL, 3M, and Warnock-Hersey provide sufficient information to consider the Thermo-Lag material inadequate?
Were these tests valid?
Did the affected licensees properly evaluate the test data?
References:
SwRI Test Report l
CTL Test Reports 3M Test Reports Warnock Hersey Test Report Information:
A number of test failt.t es have occurred when i
Thermo-Lag barriers have been constructed by licensee or
'l contractor personnel and the test has been conducted at a third party laboratory.
Thermo-Lag used in a cable tray configuration has not passed a 3-hour fire test at a facility other than at TSI.
However, each test has had some test anomaly, or is suspect in some way.
For example, some of the 3M tests did not use certified installers.
Other tests, such as the Warnock Hersey are so poorly documented that it is difficult to derive any conclusions.
-l
==
Conclusions:==
The test reports, although flawed, provide sufficient information to consider tha. ability of Thermo-Lag to provide a rated fire barrier, if installed by the people who install it in the plant, as indeterminate.
Recommendation / Status:
Conduct testbg of the material in accordance with ASTM E119.
2.9 Issue
ITL Test ReDorts Do the ITL test reports on Thermo-Lag provide a valid basis for accepting the material?
Were the results properly documented?
Is ITL qualified to oversee and conduct fire testing? Are the application methods and details sufficiently documented in the reports?
References:
ITL Test Reports Vendor Inspection Report 6
4 Information:
The majority of the qualification tests that have been performed on Thermo-Lag were conducted at the TSI facility, using ITL as the third party.
Most utilities use these reports for their design basis.
The test reports are poorly documented and do not appear to meet the guidelines of ASTM E119.
In addition, the vendor inspection found that ITL does not observe the test article construction, or even write the test report, but only certifies the test data.
==
Conclusions:==
The ITL test reports do not provide sufficient information for a
licensee to justify a
Thermo-Lag installation, nor do they meet the NRC guidance of having a nationally recognized laboratory test the product.
Recommendation / Status:
Conduct testing of the material to verify the fire rating of Thermo-Lag.
2.10 Issue:
Time-temoerature curve It has been alleged that temperature variations from the standard time-temperature curve from E119 were used during the fire endurance tests to allow proper buildup of the char layer on the Thermo-Lag. Were the temperature variations within the allowable band?
References:
ITL Test Reports Information:
According to testimony from a former TSI
- employee, to be effective as a fire barrier Thermo-Lag requires the formation of an outer crust.
This crust could only be formed early in the test by having temperatures below the curve.
The ex-employee said TSI manipulated the test temperatures to achieve this.
==
Conclusions:==
There is some information that shows the time-temperature requirements were not always met to the letter, but NRC review of the test data could not substantiate the allegation.
The time-temperature curves for the tests reviewed varied greatly, and no trend was observed.
- However, it is possible that the temperatures documented in the test report were non-conservative, or that the thermocouples were not properly calibrated. The vendor inspection found that the vendor had not calibrated the furnace thermocouples since the furnace was constructed.
Recommendation / Status:
Conduct fire testing of the material to verify the fire rating of Thermo-Lag.
2.11 Issue:
Falsification of test results According to testimony from a former TSI employee, the fire endurance tests were f alsified, in that the test articles were altered to facilitate passing.
7
References:
Interview records ITL Test Reports Information:
A former TSI employee has alleged that the Thermo-Lag test articles were altered to facilitate passing the fire endurance tests.
Methods included overlapping panel seams, and using an extra coat of trowel grade.
TSI employees constructed the test articles for most of the tests that passed.
Tests conducted by 3M failed to substantiate TSI's claims on the fire protection ability of Thermo-lag.
NRC review of ITL test reports noted that in many cases, the test article construction description was not in accordance with TSI procedures.
For example, many of the test articles.had an extra coating of Thermo-Lag applied just prior to the test.
Also, the UL description of the test article construction 'or the ampacity test was inconsistent with the TSI procedures.
Testimony indicates that TSI built their test rigs at TSI and shipped them to the test facility, so the constructan was never verified.
On several occasions, TSI requested revision of CTL test reports that altered the test engineers observations of barrier thickness.
In one test at CTL, testimony and photographs indicate TSI covered the test rig with saran wrap.
==
Conclusions:==
It is clear that the documentation for the tests does not provide sufficient information to determine what the test article looked like, and even provides some information contradictory to the TSI installation procedure.
Although no documentation was reviewed that would substantiate falsification of test data, the data was so vague that it would also not substantiate qualification.
The strongest evidence is the fact that the barriers failed whenever TSI did not build the test article.
Recommendation / Status:
Conduct fire testing of the material.
3.
DEFICIENCIES IN THE INSTALLATION AND INSPECTION PROCEDURES j
3.1 Issue
Installation Procedure Chanoes Have changes. in the TSI approved installation procedures adversely affected the qualification of Thermo-lag systems?
1
References:
TSI Installation Procedures Information: TSI has revised its installation procedures many times over the last 10 years, and also provided many site-specific installation procedures.
Many of the qualification tests were conducted with installation procedures that are no longer used by the facilities.
In addition, installation methods are not well documented in the procedures, and a large portion of the certification of installers is based on 8
information received in the installer training sessions. Site visits have confirmed that large number of installation variations have been used.
==
Conclusions:==
The TSI installation procedures have changed, but normally because of the addition of other configurations that can be installed.
The basic procedure for installation of cable trays and conduits has not changed.
Discussions with the vendor and licensees has confirmed that the current procedures are deficient, and that a number of licensees did not install the barriers in accordance with the procedures.
Recommendation / Status: Information Notice 91-79 was issued to discuss the installation problems.
The proposed generic letter will require licensees to verify their installation against the vendor recommendations and their design basis test reports.
3.2 Issue
Joint Installation Method What is the proper method for installing the joints'at the prefabricated panels and conduit sections?
Have all of the methods been qualified by tests?
References:
GSU LER Interview with TSI Information: Many facilities use an installation method where the ends of the panels are butt joined to one another and the seam is filled and covered with Thermo-Lag trowel grade.
This method is referred to as " dry-fitting" by some licensees.
- However, a TSI letter sent to licensee fire protection engineers on August 23, 1991, states that " dry-fitting" is not allowed by TSI procedures.
Another method of filling the joints is " pre-buttering" the edges of the panel prior to installation.
Based on NRC review of test reports and installation procedures, it is uncertain whether both n hods of joint installation have been tested.
The initial. st reports reviewed by the NRC in 1981 and 1982 for Comanche Peak (SwRI Report No. 03-6491),
Susquehanna (SWRI Report No. 01-7163),
and WNP2 (ITL Report No.
82-5-355B) were for qualification of the spray-on application method for Thermo-Lag where the underlying stress skin was mechanically fastened at the joints.
This joint configuration is no longer discussed in the vendor installation manual, since the current Thermo-Lag product has the stress skin integral with the-prefabricated panel.
In some qualification tests the vendor has covered joints with an additional layer of stress-skin, whici.
is another method not documented in the vendor instructions.
l
==
Conclusions:==
The vendor's installation procedures do not clearly describe the method for sealing joints.
Discussions 9
l l
i
s with the vendor and licensee installers has provided contradictory information.
Review of test reports found that it was often impossible to determine the method that was used for construction of the test article.
The vendor has stated that the procedure needs to be revised.
Recommendation / Status: Information Notice 91-79 was issued to discuss the installation problems.
The proposed generic letter will require the licensees to verify their design basis.
Fire endurance testing should provide additional insights.
3.3 Issue
Ouality Control Durina Construction What level of quality control existed during the fire barrier installation?
Were thickness measurements taken?
Was the installation checked against the installation procedure?
References:
Trip Reports Licensee Procedures Information:
Site visits have found that a wide range of quality control was used during installation of Thermo-Lag fire barriers.
At some sites, the licensee hired a TSI field representative to assist them in oversight of the installation.
- However, there were many instances of installations that were not installed in accordance with the TSI procedures.
In addition, since critical steps were not included in the vendor installation procedure, quality control may not have been aware of the critical steps and measurements unless they also attended the installation certification training provided by TSI.
==
Conclusions:==
It is difficult to determine at this late date what level of quality control was applied to specific installation, but it is clear that barriers were installed that did not meet guidance provided by the vendor.
Recommendation / Status:
Conduct testing of.the barriers to determine the acceptable configurations. The proposed generic letter will require the licensees'to review its installations and compare them with the vendor procedures and the qualification tests.
t
3.4 Issue
Trainina of Installers-What training was provided to licensee installers by TSI? Was adequate training provided to account for the gaps in the procedure? What assumptions were made regarding the knowledge level of the trainees?
What is ' certified by the TSI certificate?
References:
GSU Internal memo on inadequate training TSI response to NRC questions 10 l
e Vendor inspection report Information:
Based on the wide variation in installations observed on site visits, the installation methods used by individual sites were different.
As discussed in IN 91-47, RBS identified a number of installation deficiencies during inspections in 1987, and subsequent followup walkdowns.
The licensee found that " stress-skin" and structural ribbing had been removed from prefabricated panels during initial installation, holes and cracks in the barrier had developed since installation, and unfilled seams and joints existed.
Other facilities have also reported identification of unfilled seams, inadequate protection of supports, and site-specific configurations which are not installed in accordance with vendor instructions.
The majority of the reported installation problems were caused by errors made by contractor installers and inadequate quality control oversight during initial construction.
The task force reviewed the training records at Comanche Peak for a 2 day training session that the licensee conducted in September 1989 after identifying many installation errors.
The training was conducted in groups of up to 60 installers at a time.
The licensee also stated that the installers were not skilled insulators, but consisted of roofers, framers, and drywall hangers that spoke very little English.
A translator was used for the training sessions.
The vendor inspectir:,a conducted at TSI found that the TSI training of installers was conducted without lesson plans or a
syllabus.
A two page outline was provided to the inspectors.
Licensee documentation also did not indicate the-information that was covered.
==
Conclusions:==
Based on the quality of the installations
- observed, either the training and/or the procedure are inadequate.
The vendor is revising the procedure to include all necessary application steps.
Recommendations / Status:
Conduct an interview of a former TSI field representative to better understand the-training provided.
Installation deficiencies were discussed in IN 91-47 and 91-79.
The generic letter requests confirmation that installations are consistent with the vendor's instructions.
f
3.5 Issue
TSI On-site Representatives What service was provided by the TSI on-site representatives?
Did the rep approve configuration changes?
Did the rep approve procedure changes?
Did the rep provide pseudo-QC l
services?
l
References:
Trip Reports 11
p
~
Information:
Several of the plants visited noted that they has a
TSI field representative on-site throughout the installation of Thermo-Lag.
The reps provided certification training, and observed installations.
They also provided approvals on installation deviations. During site visits, the 1
licensees informed the review team that the TSI representative approved configurations that were different from the vendor procedures.
==
Conclusions:==
It is unclear what role the representatives actually were assigned.
Rubin Feldman stated that TSI employees did not perform QC roles onsite.
However, the licensees feel that they were paying for full-time oversight of the installation process.
It is clear that although the TSI representatives were on-site, it did not prevent the installation of configurations that were not in accordance with TSI procedures. TSI informed the NRC inspectors that the site field representative only conducted training and inventory control.
Conclusions / Status: Interview former TSI field representative at WNP2 to determine if TSI provided installation oversight.
3.6 Issue
Receiot Insoection What receipt inspection requirements are needed by the licensee to verify the acceptability of the material received?
Should thickness and density be verified? How should moisture content be checked? What are the acceptable thickness ranges?
References:
10 CFR Part 50 Appendix B Administrative Requirements Information:
The receipt inspection requirements varied greatly between the sites visited.
Significant characteristics such as thickness and moisture content were not measured, and shelf-life and storage requirements were not checked.
Fire protection materials are purchased as commercial grade items but typically have " augmented QA" requirements.
==
Conclusions:==
It is unclear what requirements actually apply.
The variation in thicknesses could impact the operability of safety related cable and equipment if the ampacity derating factors are significantly affected.
Recommendation / Status:
Have this issued reviewed by vendor branch and regional procurement inspectors.
4.
TOXICITY ALLEGATIONS i
12
y Issue:
Toxicity of Off-Gases from burnina Thermo-Lac 4.1 Could the level.of Thermo-lag off-gases provide a hazard to plant personnel (i.e. control room) prior to the personnel being 'able to sense the condition? Could the level of Thermo-lag off-gases provide a hazard following a fire during overhaul activities when personnel may not be required to coating to wear SCBAs?
Is the material excessively toxic?
'\\
References:
Allegations Chemical Analysis from SwRI David Taylor data and interview NIST toxicity test data Information:
The NRC does not have any regulations regarding toxic materials, but we are concerned with the potential impact on plant operators.
This allegation was previously reviewed and closed out (NRR-89-A-0016).
The alleger raised toxicity concerns to the NRC in March 1989.
The issue was addressed in a letter to the alleger dated June 20, 1989.
The justification provided for closing out the allegation was that a large number of materials in the plant will give off toxic and Thermo-Lag will not make it any worse, gases in a fire, In addition, plant operators will be wearing SCBA's during a fire and therefore they will be protected from the fumes.
==
Conclusions:==
The SwRI report, if validated by NIST data, does conclude that the material off-gases are toxic, but the importance of this information is not definitive. Preliminary NIST toxicity testing found that the material does off-gas some toxic gases, including HCN and CO, but not at an excessive rate.
Recommendation / Status:
Complete chemical and toxic analysis at NIST.
Provide final data to OSHA.
Physical Droblems with installers.
Issue:
4.2 It was alleged that Thermo-lag caused physical problems to the installers (i.e., skin peeled off).
References:
WNP2 Trip Report Allegations Information: During investigative interviews and site visits, there has been no corroboration from laboratory personnel (where several fire tests have been conducted) or site personnel that physical problems to installers have occurred.
The documentation provided with the material contains warning labels stating that gloves and aprons should be used.
Interviews with installers at Perry and River Bend indicated number of that they did not use protective clothing over a years and had never experienced any problems or discomfort.
13
,~
c During the WNP2 site visit, a Bechtel employee stated that a court case had been filed by the insulators union concerning physical problems because of the presence of formaldehyde in the Thermo-Lag, but the case was thrown out based on performance of a chemical analysis.
A letter from the Heat and Frost Insulators and Asbestos Workers was provided by the alleger that states that some union members had become sick after working with Thermo-Lag.
The letter states that special protection is required when using Thermo-Lag, such as organic particulate respirators and adequate ventilation.
It states test results show that the material contains formaldehyde.
The alleger provided a copy of a
NIOSH report (HETA 84-172-1573) regarding an investigation at Limerick that concluded that a health hazard did not exist.
==
Conclusions:==
Interviews with installers at a number of facilities did not identify any cnrroboration of the information.
The toxicity testing conducted by. NIST found that toxic gases are produced by the Thermo-Lag when burning, but the levels are no more toxic than other materials typically in the plant.
The NIOSH report specifically reviewed the issue alleged and concluded a health hazard did not exist.
Recommendation / Status:
Complete the chemical analysis of the material at NIST, and provide the results to OSHA.
5.
OTHER MISCELLANEOUS ALLEGATIONS AND ISSUES
5.1 Issue
Fire seal testina Thermo-lag is also used as a fire seal and fire stop material.
If tests conducted at TSI are unacceptable, inadequate fire i
barrier penetration seals may also exist where Thermo-lag acts as a fire seal.
References:
Allegations Information:
Specific test information or site reviews have not yet been conducted on this issue.
==
Conclusions:==
None.
1 Recommendation / Status:
Conduct a fire test of the Thermo-lag j
material.
If the material fails the test, further review of this issue may be necessary.
5.2 Issue
Bechtel Knowledae of Deficiencies It was alleged that 3M contacted Bechtel and advised them of the 3M test results which disputed TSI claims.
It has also 14
'l
been alleged that TSI was the only material that Bechtel would install in power plants and that' other companies were not permitted to bid on the installation.
References:
Allegations Information: A 3M representative, provided a copy of a letter dated August 8, 1986, from T.E. Sheehan, 3M, to Bob Chreate, Bechtel, providing SwRI Amapacity test Report no. PJ-19, dated July 30, 1986, covering the ampacity tests of TSI's 1-hr panel.
This was apparently the reason for the testing conducted at UL for South Texas.
The review team has also seen internal procurement documents at WNP2 that stated that the insta13ation of the fire barriers would not be put out for bid because Thermo-Lag was the only ANI approved fire barrier material. Several other Bechtel documents reviewed also state that Thermo-Lag is the only approved material for raceway fire enclosures.
==
Conclusion:==
Bechtel was advised of TSI deficiencies found by It has not 3M, and decided to conduct an ampacity test at UL.
been determined if they had any other information to suspect that the ampacity deratings were suspect.
Recommendation / Status:
OI will continue interviews to determine if any improprieties occurred. Site specific issues regarding reportability are discussed in section 6.
5.3 Issue
Chloride Content Does Thermo-Lag contain chloride levels that may be harmful (stress corrosion) to stainless steel components in the plant, contrary to Regulatory Guide 1.367
References:
Allegations Regulatory Guide 1.36 Vendor Inspection Report Information:
It has been alleged that Thermo-Lag contains chloride levels in excess of NRC requirements.
The vendor inspection found that the requirements of licensees vary regarding chloride content.
However, the vendor inspection also found that prior to 1989, the vendor tested the Thermo-Lag for chlorides, and the results were always less than 200 ppm.
RG 1.36 is only applicable to thermal insulation on stainless steel fluid system piping.
No purchase order was found that imposed the chloride requirements.
The vendor inspection-found that the vendor no longer certifies the chloride content.
==
Conclusion:==
Preliminary results of the NIST chemical analysis indicates only trace amounts of C1.
The RG does not appear to l
be applicable to Thermo-Lag.
15
c y
Recommendation / Status: Complete chemical analysis of material samples.
T.efer results to Chemical and Materials Branch for closeout following receipt of NIST final report.
5.4 Issue
Other fire barrier materials Based on the deficiencies identified.
during this j
investigation, should a review of all fire barriers be considered?
References:
Trip Reports Information:
Many of the generic issues involve the failure of the licensees to adequately review and evaluate qualification test data and installation practices.
These areas could very well apply to other fire barrier materials.
==
Conclusion:==
Although the generic concerns could be applicable to the other fire barrier materials, the NRC does not have any a
information to suspect that they are deficient.
In addition, most of the other materials have been UL tested and listed, which has been previously acceptable to the NRC for any fire rated assembly.
Recommendation / Status:
The task force recommends that this issue be closed, unless other information is received.
5.5 Issue
Stainless Steel Stress Skin It has been alleged that stainless steel stress skin was used in the qualification tests by the vendor, but galvanized stress skin is actually sold to the licensees.
References:
Interviews Information:
The vendor has been asked on several occasions to provide detailed information to the NRC on the material makeup of stress skin, but no information to resolve this issue has been received.
The-test reports-reviewed provide insufficient information to determine the material makeup of the stress skin.
The site visits have.found that the stress skin does not appear to be stainless steel.
==
Conclusions:==
None to date.
Recommendations / Status: Conduct fire qualification test using licensee in stock materials.
5.6 Issue
Seismic Desian Basis What is the basis to qualify Thermo-lag seismically? Were the correct thicknesses and weights assumed in the analysis?
References:
TSI Technical Notes Comanche Peak Trip Report 16
i Information:
TSI provides a generic seismic qualification package to licensees, based on calculations.
Most licensees conduct their cable tray support seismic qualification certain panel weight.
However, because of the assuming a variation in panel thickness, density, and moisture content, it is possible that the weight being assumed for the calculations is non-conservative.
==
Conclusions:==
The task force has only determined that a wide variation in panel weight is possible, but not determined the magnitude of the range or scope of the potential impact.
Recommendation / Status:
Have NRR technical staff review the TSI calculation, and several sample licensee calculations, and determine if the calculations and assumptions are valid.
6.
SITE-SPECIFIC ISSUES 6.1 RIVER BEND (October 7-8, 1991) 6.1.1 Receiot Insoections Inconsistencies were found in the thickness tolerances for the pre-fabricated panels between two specifications.
Also,. inconsistencies were noted in the inspection attributes assigned to Thermo-Lag purchases, such as the maximum storage temperature.
This issue was referred to Region IV on November 12, 1991.
Region IV plans to conduct an inspection to followup this issue.
6.1.2 Uncualified Installations Three fire barrier configurations were observed during the plant tour that the licensee could not justify by either fire tests or engineering analyses.
These were (1) a large horizontal barrier separating Fire Area PH1 from Fire Area PT1 in G tunnel, (2) a large cable tray enclosure in F tunnel, and (3) an instrument rack enclosure at elevation 98 of the control building.
In addition, structural steel forming parts of the barriers were not protected to provide fire resistance equivalent to that required of the barriers.
This issue was referred to Region IV on November 12, 1991.
Region IV plans to conduct an inspection to followup this issue.
6.1.3 Acceptance Criteria 17
s.
4 During fire tests in November and Dece.mber 1990 intended to qualify proposed upgrades to Thermo-Lag fire barriers, the 1 Acensee did not apply the NRC acceptance criteria to the test results.
The licensee used 325 'F above ambient ins +.ee.d of 250 *F above ambient.
The licensee could not previde the basis for the acceptance criteria.
This issue was referred to Region IV on November 12, 1991.
Region IV plans to conduct an inspection to followup this issue.
In addition, PSB was requested to clearly document the NRC position on this issue.
6.1.4 ReDortina Recuirements In July 1988, the licensee conducted fire tests on the as-installed fire barrier configurations and the test articles failed.
Although the licensee identified significant additional nonconforming conditions and declared additional barriers inoperable, the licensee did not submit additional LERs or_ supplemental LERs to report the nonconforming conditions.
This issue was referred to Region IV on November 12, 1991.
Region IV 6 plans to conduct an inspection to followup this issue.
6.1.5 Installation Procedures During initial construction, ANCO installers deviated from the installation procedures by removing stress skin and ribs from the prefabricated panels.
The licensee installed the panels using the dry-fitting method of joint sealing.
- Also, degraded barriers caused by moisture and wear were observed during the plant tour.
The installation problems noted at RBS are generic, with the exception of removing the stress-skin and ribs.
These problems were discussed in Information Notice 91-47 and 91-79 and will be addressed in the proposed generic letter.
No further action is necessary.
6.1.6 Amoacity Deratina GSU could not determine if they had received TSI's October 2,1986, mailgram that discussed revised ampacity deratings. The licensee was using much less conservative numbers at the time of the site visit.
The licensee stated that the revised figures would be reviewed.
This issue is generic and will be addressed by the proposed generic letter. No further action is necessary.
18
l' t
w 6.1.7 Part 21 Reportina During the Promatec 3-hour qualification testing conducted at SwRI in October 1985, the Thermo-Lag portion of the test assembly burned away.
Testimony indicated that RBS personnel discussed filing a Part 21 report, but concluded it was not required.
The alleger states that this was proof that Thermo-Lag would burn.
The alleger also stated that the sponsor had Promatec conduct another test without Thermo-Lag so the. failure would not be included in the report to the NRC.
During the site visit, the licensee stated that the j
Thermo-Lag was installed on a conduit that was open at
{
one end.
Therefore, the Thermo-Lag was exposed to fire on both sides of the material.
This configuration was not installed in the plant.
The task force recommends l
this item be closed.
\\
6.1.8 Unaualified 3-hour Barriers Following the October 1989 SWRI test
- failure, the licensee developed upgraded configurations to be installed to correct the problem.
However, the tests of the upgrade failed in November and December 1990, Following issuance of the generic letter and resolution i
of item 6.1.3, this issue should be referred to Region IV for followup review.
i i
6.2 COMANCHE PEAK (November 4-5, 1991) 6.2.1 Fire Barrier Desian Basis During the site visit, the licensee stated that it was using the 1981 SWRI test as the design basis for using Thermo-Lag.
The licensee provided a list of 16 other fire endurance test reports that it had received from the vendor as backup information.
However, during the site f
visit, the licensee could neither identify the specific fire testo used for the design basis nor provide any evaluations performed to verify that the Thermo-Lag fire barrier configurations installed at CPSES either replicate the tested configurations or provide an equivalent level of protection.
After the site visit, the licensee provided the review team with a copy of Impell Calculation 0210-063-0046,
" Review of TSI Protective Envelope Fire Test," November 13, 1987.
Impell developed this calculation for evaluating the SwRI Project 03-6491 and ITL 82-11-80 test reports.
Impell found that the material configuration 19
a:
a described in SWRI Project 03-6491 did not represent the materials installed at CPSES and therefore could make no conclusion as to the fire resistance of the material installed at CPSES.
However, ITL test report 82-11-80 was found by Impell to be acceptable as evidence of a 1-hour rated fire barrier envelope system at CPSES.
This issue is generic and is addressed in the proposed generic letter.
No further action is necessary.
6.2.2 Recelot; Inspectipa The licensee's receipt inspection procedure used panel density and moisture content as acceptance criteria.
Moisture content is checked by the licensee using a moisture meter if the panel has not cured for at least 30 days.
The licensee stated during the site visit that during the construction phase some panels were still wet (uncured) when received from the vendor, but the problem was corrected. The density is calculated by dividing the panel weight by the panel area (length x width).
The licensee established maximum and minimum acceptance criteria for the density. This method of determining the proper curing of the panels using density does not appear valid because of the number of variables involved:
thickness, void content, moisture content, and material composition.
This issue is generic and should be resolved following performance of the fire endurance testing.
6.2.3 Installation Procedures While performing a walkdown inspection of the plant, the review team observed a number of installation details that deviated from TSI's procedures.
Banding straps installed within 2 inches of each panel joint.
This technique is more conservative than the. current TSI manual but is consistent with the special TSI installation manual for aluminum cable trays -(TSI Technical Note 20689-AL).
The task force's concern is that this strap location may be necessary.for all installations, but the TSI procedures do not require them.
The licensee could not recall the reason.
for requiring this installation detail.
For cable trays greater than 24-inches wide, the licensee's procedure requires installing banding straps around the tray before' installing the panels.
These bands support the top layer of 20
c c,
Thermo-Lag and present sagging into the tray.
The panel joints are all prebuttered with approximately 1/4 to 1/2 inch of Thermo-Lag trowel grade material, although this detail is not listed as an option in the TSI manual.
The licensee stated that TSI informed it during training sessions that this was an essential step of the installation process.
The raceway supports were covered with 1/2 to 3/4 inch of Thermo-Lag trowel grade based on a "9-inch rule,"
which the licensee derived from a
calculation.
The vendor manual recommends covering the first 18 inches of each penetration into the fire barrier
- system, and all structural steel supports.
On October 4,
- 1983, the licensee submitted a position paper to the NRC providing an engineering basis for not protecting cable tray supports.
In
- 1988, the NRC approved this installation.
The licensee applied topcoat to all Thermo-Lag installations but could not provide qualification test documentation that demonstrated acceptability of this product.
The licensee's installation procedures specify that all seams, joints, and scored edges shall be filled with Thermo-Lag trowel grada to slightly above the level of the surrounding prefabricated sections to ensure 1/2-inch minimum thickness is maintained after the trowel grade material shrinks approximately 25 percent during curing.
Although this appears to be a good practice it is-not in the TSI procedures manual and has not been observed during other site visits.
The licensee's installation specification required that the prefabricated panel "V" stiffener ribs be oriented perpendicular to the cable tray on the top section to prevent sagging, and parallel with the tray on the bottom section to ensure the joints are tight.
The task force is not aware of any guidance from the vendor-that requires this technique.- The task force is concerned that this may be necessary for seismic considerations, but has not been communicated to the licensees by the vendor.
Weaknesses in vendor and plant-specific installation procedures and practices are a generic concern and have been identified at each facility visited by the team.
21
e Installation problems are, therefore, being addressed in IN 91-79 and the proposed generic letter.
No further action is necessary, pending results of the fire testing.
6.2.4 IDptaller Trainina A
TSI fielo representative conducted certification training of the installers in September 1989.
The two day training session was conducted for over 60 people, and was supposed to include hands-on training.
It is questionable that adequate training could have been provided to all of the attendees.
In addition, very few of the installers spoke English, and an interpreter was used during the training sessions.
The adequacy of training is a generic issue that is coupled with the adequacy of the vendor's procedures.
The vendor is revising its procedures.
6.3 WNP2 (November 6-7, 1991) 6.3.1 Amoacity Deratina WPPSS could not determine if they had received TSI's-October 2,1986, mailgram that discussed revised ampacity l
deratings.
The licensee was using deratings from its site specific testing conducted by TSI (ITL 82-355 series). The licensee stated, after a short review, that the ITL tests were more applicable to WNP2, since site l
specific cable and materials were used to conduct the test. The licensee also stated that the WNP2 design does not include sufficient margin to accept additional cable i
derating.
j
~l This issue is generic and will be addressed by the proposed generic letter.
The issue of proper evaluation of vendor information was forwarded to Region V on January 9, 1992.
No further review is necessary.
1 6.3.2 Uncualified Installations During the plant tour, one Appendix R fire-barrier was observed that the licensee indicated was not substantiated by either fire test or ' engineering analysis.
This barrier, a wall separating the radwaste building from the turbine building at elevation 471, is constructed of concrete blocks and coated with Thermo-Lag material on'one side.
The task force is not aware of any ASTM E119 fire endurance test on this wall configuration.
This issue i
was forwarded to Region V on January 9, 1992 for followup 22 j
s action.
6.3.3 Receint Inspection The licensee does not have any procedures or guidance concerning specific inspection attribute.s important to Thermo-Lag, such as shelf-life requirements, temperature limitations, and thickness tolerances.
The shelf-life I
was not included as an inspection attribute in the receipt inspection.
This issue was forwarded to Region V on January 9, 1992 for followup action.
6.3.4 3M/TSI Interface Test Failure on August 19, 1986, 3M performed an interface test for the licensee to qualify a method for joining 3M Interam E-50D series flexible 3-hour fire wrap with the 3-hour rated Thermo-Lag 330-1 fire barrier system.
The Thermo-Lag portion of the assembly failed the test.
On May 7, 1987, TSI repeated the test for the licensee at TSI, and the test passed.
The licensee informed the task force that TSI advised it that the Thermo-Lag failed the test because it had not cured for the recommended 30 days.
(Records show it cured for only 13 days.)
However, the task force reviewed other ITL test reports in which the Thermo-Lag did not cure for 30 days and the test passed.
Therefore, the reason for the initial failure remains unclear.
This issue can only be resolved by conducting testing on the Thermo-Lag fire barriers.
In addition, this test failure is indirectly referenced in the proposed generic letter.
6.3.5 Installation Problems During construction, the licensee used a method of installation called low pressure extrusion.
- However, after construction the Region V identified during an inspection that this method had never been adequately In March 1987, the licensee tested qualified by testing.
the in-situ configuration at TSI, and the test assembly failed.
During the 3-hour test, temperatures inside the conduit exceeded 325 'F after 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 12 minutes. The cause of the failure was not-documented in the report.
licensee informed the NRC that the failure was The attributed to voids created in the Thermo-Lag material by the injection process.
The licensee also stated that TSI's field engineer observed, but never questioned, the use of this technique.
This material was removed from 23
u the plant and replaced with prefabricated panels and sections.
The 1-hour rated barriers installed with this j
method passed a test, and are still installed.
No i
further action is necessary, unless testing indicates
]
that the material is not adequate, or test falsification i
is substantiated.
)
6.3.6 Use of Failed Test from Susouehanna Region III (OI) referred an issue to the task force on August 20, 1991, regarding the suspected use of a test report received by WPPSS from PP&L.
The. PP&L test, conducted at SwRI documented the failure of configurations of 1-hour cable trays.
The licensee 1
confirmed during the site visit that this test report was not being used as the design basis.
This issue is considered
- closed, and a
closeout memorandum was forwarded to Region III-OI.
6.3.7 Fire Barrier Desian Basi _s During the site visit, the licensee provided the task-force with copies of seven fire endurance test reports held by WPPSS.
However, the licensee could not provide its specific design basis.
The licensee had never reviewed the test reports and compared them with the as-built configuration to determine if they could justify their specific installation. The licensee stated that it would review its design basis.
l This issue is generic and will be addressed in the proposed generic letter. No further action is necessary.
6.3.8 Installation Procedures The task force reviewed-the licensee's current installation procedure and concluded that it was inadequate to ensure proper installation and QC inspection of Thermo-Lag fire barriers.
It did not include all of the sequential steps involved in applying Thermo-Lag.
During the plant tour, the team could not identify the specific installation techniques for some of the configurations observed. Several items were noted to deviate from the vendor's procedures including gaps between conduit preshaped sections, _three layers of stress-skin, hardware cloth substituted for stress skin, and panels without structural ribs.
IN 91-79 discussed some of the installation problems.
This issue is generic and will be addressed in the proposed generic letter. No further action is necessary, i
24
i 6.4 PERRY (November 19, 1991) 6.4.1 Fire Barrier Desian Basis During the site visit, the licensee informed the task force that it had established as its technical bases for installing Thermo-Lag fire barriers the fire tests identified in TSI Technical Note 20684, "Thermo-Lag 330 Fire Barrier System Installation Procedures Manual Power Generating Plant Applications."
- However, the licensee did not eva3uate (1) fire test reports to validate the fire resistance ratings of Thermo-Lag and their applicability to the PNPP design or (2) the fire barrier designs installed at PNPP to ensure that they either replicated the tested configurations or provided an equivalent level of protection.
The licensee stated that the problems identified in IN 91-47 had prompted it to plan to establish and document its technical basis.
This will include performing technical evaluations of the fire test reports to ensure that they are valid and apply to PNPP, finding any deviations between the tested configurations and the field installations, and evaluating any identified deviations for acceptability.
This issue is generic and is addressed in the proposed generic letter.
No further action is necessary.
6.4.2 Cable Trav SuDoort Protection The licensee protects raceway supports with a 1/4-inch dry film thickness of trowel-grade Thermo-Lag material, which deviates from the vendor's recommendation.
The licensee applied the material to each support from the point of attachment to the raceway to a point 18 inches away from the point of attachment.
To assess the performance of its raceway supports during fire exposure, the licensee performed a small-scale 1-hour ASTM E-119 fire test at Construction Technology Laboratories (CTL) on September 28, 1986 (Bisco Fire Test Report 748-218).
The licensee constructed the test assembly in accordance with Bisco Procedure SP-116 to represent a typical PNPP installation.
Portions of the unprotected supports were exposed to the fire.
Temperatures during the test.
exceeded 325
'F.
The review team concluded that the fire test could not be used to determine the fire resistance rating.of Thermo-Lag because portions of the Thermo-Lag test assembly was not cured for 30 days before the fire test and the test report lacked the. information and detail needed to reach a conclusion about the fire performance of the Thermo-Lag test assembly.
Furthermore, the task force could not determine if the temperatures inside the protective envelope exceeded 325 25
o
'F because of a failure of the Thermo-Lag fire barrier or as a result of heat transmission through the support system.
In a letter of December 15, 1986, the licensec asked the vendor to review the test report.
In a June 29, 1987 letter, the vendor provided its comments on the test results to the licensee.
The-vendor discussed the lack of cables in the test assembly, the placement of the thermocouples, and the cure time, but did not state a conclusion as to the acceptability of the test results.
The licensee informed the task force that it continued to protect raceway supports with a.1/4-inch thick layer of Thermo-Lag rather than a 1/2-inch thick layer on the basis of the vendor's review of the test report and its analysis of the test report and the vendor's comments.
The licensee's analysis was not available during the site visit.
Following performance of fire testing, this issue may need further review.
6.4.3 Recolot Inspection The review team found that the licensee's receipt inspection procedure did not provide guidance regarding storage temperature limitations for trowel-grade Thermo-Lag materials.
The review team found that the licensee included the temperature strip charts in the receipt inspection file but did not check the temperature data against inspection acceptance criteria or record the data in the receipt inspection report.
The licensee agreed that this was a flaw in its receipt inspection procedures and that it would revise the procedures to ensure that the temperature limits are included in future receipt inspections of the trowel-grade material.
The review team found this response acceptable.
j i
On January 10, 1992, this issue was referred to Region III for followup.
6.4.4 Bandina Soacing While reviewing of IN 91-47, the licensee identified that the spaces between some mechanical fasteners (banding straps) exceeded the vendor's tested configuration i
without technical justification. The licensee concluded that this discrepancy was caused by inadequate design, could adversely affect, the ability to meet. the safe shutdown requirements, and was reportable as a-defect under 10 CFR Part 21.
The licensee reported the installation discrepancies in LER 91-020 of November 4, 1991.
PNPP procedures would have allowed up to 36 inches between bands, while the vendor recommended 12 inches.
26
During followup to IN 91-47, the licensee also found supports that were protected.with less than 1/4 inch.
The licensee concluded that 1/4 inch had been applied, but shrinkage had not been adequately considered.
These issues are typical of the generic installation problems that were identified by the review team, and they will be addressed by the generic letter.
6.4.5 Solit Conduit Barrier While performing the plant walkdown inspections, the review team observed a split in a Thermo-Lag fire barrier on a conduit at the 638-foot elevation of the control complex division 1 cable spreading room.
This split was the full depth of a joint between preshaped conduit sections.
This damage appeared to have been caused by individuals stepping or walking on the conduit.
On December 17, 1991, the licensee informed the review team that it had issued a work order to repair the conduit fire barrier.
Fire watches have been established in the cable spreading room because of the banding strap spacing deviations reported in LER 91-020.
The split will be repaired when the banding strap spacing is improved.
On January 10, 1992, this issue was referred to Region III for followup.
6.4.6 Ampacity Deratina on December 4, 1991, the licensee informed the review team that its purchasing department had received the vendor's Mailgram during October 1986 and forwarded it to the quality department.
The licensee subsequently reassessed the PNPP ampacity derating based on the information provided in the Mailgram.
The licensee stated that it would provide copies of letters that detail the results of its assessment to the review team for information.
This issue is generic in nature and is addressed in the generic letter.
6.4.7 Fire Test Failure During testing of Thermo-Lag by 3M in March 1986, the Thermo-Lag exceeded the 325 'F acceptance criterion.
3M notified Perry (also Bechtel, Stone & Webster, and Beaver Valley) of the test failures.
On March 21, 1986, 3M sent a letter to PNPP reporting the results of the TSI/3M side-by-side ASTM E119 fire tests.
27
=
v s
F In two tests, the TSI board material failed a 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> test.
The installation,
- however, was not done by certified installers, nor did the material have a
Certificate of Conformance.
The review team concluded that because of the 3M disclaimers, the test results could not be used to question the fire rating of _ Thermo-Lag.
No further action is necessary.
6.5 CALLAWAY (December 19, 1991) 6.5.1 Fire Barrier Desian Basis Prior to the site visit, the licensee informed the review team that ITL Reports 82-11-80 and 82-11-81 were its technical bases for installing Thermo-Lag fire barriers at Callaway Plant to meet the requirements of Appendix R to 10 CFR Part 50.
During the site visit, the licensee informed the review team that it did not evaluate (1) the ITL fire test reports to validate the fire resistance ratings claimed by the vendor or to determine the applicability of the test reports to the Callaway Plant design or (2) the fire barrier designs installed at callaway Plant to ensure that they either replicated.the tested configurations or provided an equivalent level of protection.
This issue is generic in nature and will be addressed by the generic letter.
No further action is necessary.
6.5.2 Uncualified Fire Barriers During the walkdown inspections, the review team observed Appendix R fire barriers that the licensee indicated may not have been substantiated by either fire test or engineering analysis.
These barriers included (1) hatch.
covers on elevation 2026 of the auxiliary building and (2) large encapsulated. valve hatch covers (13 feet by 11 feet) on elevation 2000 of the auxiliary building. These barriers form parts of fire area boundaries
- and, therefore, need 3-hour fire ratings to satisfy Appendix R requirements.
The barriers are constructed of checker plate hatch covers that have been covered-on one side with 3-hour Thermo-Lag prefabricated panel sections.
The review team is not aware of any ASTM E-119 fire endurance test on this type of configuration. The review -
team recommended that this issue be referred to Region III for review.
6.5.3 Installation Problems 28
=
Installation anomalies observed by the review team included (1) a junction box barrier with the stress skin on the wrong side, (2) band spacing greater than 12
- inches, (3) structural supports and base plates not protected, (4) cable tray barriers where the stress skin had been removed, (5) conduit barriers constructed with prefabricated panels that had been scored and wrapped around the conduit with the gaps formed by the scores filled with trowel grade material, and (6) 3-hour panels i
installed on cable trays with the ribs on the outside.
The panels were apparently installed by the dry-fitting method. since an unfilled seam was observed at the site.
Weaknesses in the installation procedures and practices of the vendor and the licensees is a generic concern.
Thus, the review team is addressing the installation problems generically.
The staff addressed most of these problems in IN 91-79.
6.5.4 Amoacity Deratina During the site visit, the licensee informed the review team that based on limited research it could not find a record of having received the October 2,
1986, TSI Mailgram.
The licensee stated that as a result of the review team bringing the existence of the Mailgram to its attention, it had incorporated its data into an ongoing study of derating at the Callaway Plant.
The licensee could not provide the derating factors currently used in the plant design to the review team during the site visit.
The review team has identified ampacity derating as a generic concern. The review team is continuing to review this concern, including reviewing ampacity derating test methods and analyzing and using ampacity derating test results for plant design. This issue is addressed in the generic letter.
6.6 SAN ONOFRE In November 1991, the review team had a telephone conference with the fire protection engineer at San Onofre (Carol Cowser 717-368-9229).
The licensee discussed several issues that they had identified when they reinspected their installations against TSI Technical Note 20684, Revision 5.
They stated that they-were not aware of tests for some of their configurations, and that some barriers had deteriorated over time, especially barriers located outdoors with no topcoat.
They also identified cable tray support problems.
29
a A-
=g These issues are generic in nature, and will be addressed by the generic letter.
6.7 WATERFORD On November 1, 1991, the review team returned a phone call to Barry Collier of Westinghouse.
Mr. Collier informed the review team that IN 91-47 was recently brought to his attention and that he was calling in response to the NRC request for information.
He stated that there were two partial height Thermo-Lag fire barriers installed at Waterford 3 between redundant safe shutdown components with the stress skin removed.
Mr.
Collier stated that the barriers were not qualified by.
fire tests. He also stated that the last time he saw one of the barriers, the Thermo-Lag was badly deteriorated.
This allegation was documented by the review team and provided to the NRR allegations coordinator.
The allegation was referred to Region IV for followup.
7.
NRC PROGRAMMATIC ISSUES 7.1 Licensino Reviews The review team found that many of the SERs reviewed did not clearly describe the details of the fire barriers that were approved to be used, or the technical basis for the fire barrier material.
It appears that the licensee only stated that the fire barriers were " approved 1-hour barriers" and no further details were' provided.
The licensee submittals were approved on that-basis.
The level of review that was conducted on licensee submittals during this period does not appear to have been well' documented.
Only three very early Thermo-Lag fire tests appear to have been reviewed and approved.
After the material installation procedures were significantly changed there is no record of further NRC review.
7.2 Clarity of Recuirements The initial handling of fire protection and Appendix R
~
issues in the 1980's was not well documented or consistent.
The program had to deal with a large volume of licensee submittals, developing staff. positions for the new regulations, and a major Differing Professional
.)
Opinion.
In August 1984, William Dircks, EDO, directed that a Fire Protection Policy Steering Committee. be formed to conduct a review of fire protection issues and-provide recommendations for resolution.
The committee considered the adequacy of current guidance to industry, interpretation of Appendix R requirements, and adequacy 30 l
j
i 1
of inspection practices.
In their final report of October 26, 1989, they stated that they were aware "that not all parties will be fully satisfied with these actions."
The result of their endeavor was Generic Letter 86-10.
However, other actions recommended by the committee, such as revising the SRP, appear to have never been completed.
Even today, there is disagreement among individual reviewers of the interpretation of NRC requirements.
7.3 Fire Protection and Appendix R Insoections The site visits conducted by the review team found many apparent problems with fire barrier installations and qualifications.
In these cases, the team could not find any specific information that would indicate that the NRC had previously reviewed and approved specific barrier configurations.
The acceptability of Appendix R
inspections was previously reviewed by a review team in 1984, and some corrective action was taken.
In addition, the review team had a large volume of information and knowledge collected over several months that would not have been available to the previous inspection teams.
The review team recommends followup of plant specific issues as discussed in section 6,
and inclusion of lessons learned into a TI that will be used to review licensee actions in response to the GL.
A followup inspection conducted at River Bend Station in response to review team findings found a
large number of deficiencies.
The inspection team met with the review team prior to the inspection.
7.4 Amoacity Part 21 Report On October 2,
1986, the vendor notified the NRC by Mailgram of revised ampacity derating factors.
At the time, Part 21 followup belonged to AEOD.
It was later transferred to the Vendor Branch.
VIB closed out the Part 21 in December 1990 based on administrative reasons.
Therefore, the ampacity derating issue that is being pursued by the generic letter was known by the NRC since 1986, and it appears that no action was taken.
l l
7.5 LERs/Information Reports / Inspection Reports i
During the review, the review team found that a large volume of information existed on the docket that would raise questions regarding the ability of Thermo-Lag to 31 l-
8
...u perform as a fire-rated barrier.
Much of the information the review team used to develop the technical issues was already available to the NRC in
. NP2, RBS, and CPSES had several technical the docket.
W issues that were related or indicative of the generic issues that existed.
32
v-
~,
e 4
ATTACHMENT 6 LIST OF FIRE ENDURANCE TEST REPORTS INVOLVING TRERMO-LAG 330-1 FIRE BARRIER SYSTEMS KNOWN TO THE NRC This attachment provides a listing of the fire endurance test reports involving Thermo-Lag 330-1 Fire Barrier Systems that are known by the NRC to be available.
Tests have been conducted by Industrial Testing Laboratory (ITL), Southwest Research Institute (SwRI), Construction Technology Laboratory (CTL), Minnesota Mining and Manufacturing (3M), and Warnock-Hersey International.
The table includes the date the test was conducted, test report title and description, and notes on items of interest documented during the test.
The left column also includes a notation if the test failed to meet the NRC acceptance criteria of 325
- F cold side temperature.
-LIST OF FIRE ENDURANCE TEST REPORTS INVOLVING THERMO-LAG 330-1 FIRE BARRIER SYSTEMS KNOWN TO THE NRC February 13, 1981 Fire Endurance Tests dated June 1981 Submitted by PP&L for information on 10/26/81 1-hr E-119 test on Class 1E cable trays, Failed conduit and air drops.
Passed based on 14" cable tray /4" conduit circuit integrity.
Some temperatures exceeded 325F.
April 27, 1981 TSI Technical Note 8275-1 dated G Se 1981 Submitted by PP&L 5/12/82.
1-hr E-119 test on Class 1E Cable conduit.
Test passed circuit continuity.
Onsite 5.5" conduit furnace used.
Report not signed.
Dates of drawings inconsistent.
April 28, 1981 TSI Technical Note 8232-1 date June 1981 Submitted by PP&L 5/12/82.
Same test later 1-hr E-119 test on Class 1E Cable tray and Air revised (dated June 1982)
Drops.
Test on 14" cable tray.
Test passed with Feldman's signature.
circuit continuity.
Onsite furnace used.
Report not signed.
No independent QA ev102nt.
14" cable tray-Dates of drawings inconsistent.
August 6, 1981 Enaineerino Report dated Aucust 1581 Package included in literature provided to 1-hr E-119 test on Class 1E Cable
- tray, NRC on 10/26/81 by PP&L.
Conduit, and air drops.
Conducted at TSI, but reported by Wesson and Associates.
Test passed based on circuit integrity.
Max temperature recorded 315F.
1 Revised February 7, 1992 l-
___.___.,_______m.__
e september 17, Southwest Research Report No. 03-6491 dated Test approved as 1981 October 27. 1981 acceptable 1-hr barrier by NRC in December 1,
" Fire Qualification Test of a
Protective 1981 letter to TU.
Test Envelope System" conducted for Texas Utilities report submitted to NRC Services (Comanche Peak).
Used TU test December 7, 1981.
procedure dated 9/9/81.
1-hr test conducted.
Spray on application Passed based un circuit integrity, but cable 18" solid tray /18" ladder temperatures met 325F.
Installed by TSI.
tray /5" conduit Additional coating applied prior to test.
8 0 18 1
( 9 / 8 1 )
Fiberglass armoring used. ANI #5(79) used.
Installation.
Prairie Island holds March 1,
1982 ITL Report No. 82-3-2 copy.
3-hr E-119 Fire Endurance Test to qualify the thermo-lag conformable 3-hr stress skin fire wall system.
May 17-28, 1982 ITL Report No. 82-5-355A dated June 1982 6"
steel trays tested.
Report submitted to NRC 1-hr E-119 fire simulation tests, water hose 9/20/82 but no record of Failed stream impact tests and electrical circuitry approval on docket.
Test continuity test on Class 1E cable trays and conducted for WNP2.
conduit test assemblies. Conducted at TSI by TSI.
Temperatures exceeded 325 on one tray, 12" ladder tray /6" solid.
but " passed" on circuit integrity.
Bechtel tray /4" conduit /2.5" and Burns & Roe witnessed test.
conduit June 18-25, 1982 ITL Report No. 82-5-355B dated July 1982 Report submitted to NRC by WNP1 10/5/82.
NRC 3-hr fire endurance tests on Thermo-lag for approved 3-hr barrier in WNP2.
All SSER 3.
configurations passed, and temperatures did not exceed 325F.
Direct spray on flats, then 12" ladder tray /6" solid formed.
tray 2
Revised February 7, 1992
M.
E A
8-1-
August 10, 1982 Sout;hvest Research Report No. 01-7163 Test submitted by PP&L on 8/25/82 to demonstrate 1-1-hr E-119 test on Class 1E Cable
- tray, hr barrier for "as-Failed Conduit, and air drops.
Test conducted at installed" TSI material.
Southwest for PP&L for "as-installed."
CP Test was only accepted test procedure used.
Circuit continuity for conduits.
325F remained intact.
Shortened cure time (13 criteria applied by NRC.
days).
Spray-on application.
18" cable trays /5" conduit.
Copy exchanged w/APS and WNP2.
September 9-28, ITL Report No. 82-11-80 dated November 1982 Submitted to Stone and Webster (River Bend) on 1982 1-hr fire endurance tests conducted on test 8/9/84 as background articles containing " generic" cables protected information.
with thermo-lag.
Passed based on circuit Prairie Island holds integrity, but maximum temperature was 288F, copy.
and no visual fire damage noted.
" Cure 6"
solid tray /12" ladder accelerator" used.
Application method not tray /4" conduit specified.
~
september 10-ITL Report No. 82-11-81 dated November 1982 Submitted to Stone and Webster (River Bend) on October 12, 1982 3-hr fire endurance tests conducted on test 8/9/84 as background i
articles containing " generic" cables protected information.
~
with thermo-lag.
Passed based on circuit Prairie Island holds integrity, but maximum temperature was 320F, copy.
with no visual fire damage.
" Cure 6"
solid tray /12" ladder accelerator" used.
Application method not tray /4" conduit specified.
November 1982 ITL Report No. 82-11-240 1-hr fire endurance tests conducted on thermo-lag applied by the direct spray on to 4"
conduit for WNP2.
3 Revised February 7, 1992 4
9
~.m.
s e
June 1985 ITL Report No. 85-1-106. Revision 1 Prairie Island holds copy.
3-hr fire endurance test conducted on flex conduit connected to a ladder cable tray.
June 1985 ITL Report No. 85-2-382. Revision 1 1-hr fire endurance test conducted on air drop cables and unistrut section connected to a 4" diameter conduit.
June 1985 ITL ReDort No. 85-4-377. Revision 1 Prairie Island holds copy.
1-hr fire endurance test conducted on a 4"
diameter aluminum conduit assembly with a condulet, and air drop cable installed in a 3/4" fluid type flex conduit.
June 1985 ITL ReDort No. 85-3-314 1-hr fire endurance test conducted on a 4"
conduit protected with flexi-blanket.
June 1985 ITL ReDort No. 85-6-283 3-hr fire endurance test conducted on 4"
conduit protected with flexi-blanket.
January 1986 ITL Report No. 86-1-143 1-hr fire endurance test-conducted on 3
thinwall steel conduit sections mounted adjacent to a concrete wall.
l 6
Revised February 7, 1992 l
l
March 10-17, 1986 3M Fire Test Report 86-42 and 86-43 2" conduit /5" conduit Side by side tests of 3M 1-hr fire endurance test.
TSI exceeded 325'F.
and TSI in preparation Failed Conducted at 3M facility. Cured for 3-5 days, for interface test.
Incorrect installation.
Conducted for Perry, June 13, 1986 3M Fire Test Report 86-73 (PJ-10)
Conducted for Beaver-Valley.
1-hour fire endurance test of 3M/TSI Failed interface. TSI exceeded 325'F.
5" aluminum conduit.
August 19, 1986 3M Fire Test Report 86-92 (PJ-16) (Twin Cities Received from 3M and Testina #414186-1119)
WPPS.
Conducted for WNP2.
Failed 3-hr fire endurance test of 3M/TSI interface.
(See ITL 87-5-76)
Circuit integrity maintained.
but maximum 2.5" steel conduit.
l i
temperature of TSI exceeded 325*F.
Thermo-Lag 3 layers of stress skin installed by certified installers, under WPPS used.
2 layers of 1-hr board.
supervision.
August 1986 ITL Report No. 86-8-207 1-hr fire endurance test conducted on 4"
conduit assembly protected with pre-shaped thermo-lag.
September 17, 3M Fire Test Report 86-102 (PJ-21)
Conducted for Beaver Valley.
1986 1-hour fire endurance test of 3M/TSI 5" aluminum conduit.
t l
Failed interface.
Duplication of 86-73 test.
October 21, 1986 3M Fire Test ReDort 86-112 (PJ-24)
Conducted for Perry.
1-hour fire endurance test of 3M/TSI 4" conduit.
interface.
7 Revised February 7, 1992 i
i e
4 W
u..
.n
-4
(
October 22, 1982 ITL ReDort No. 82-11-241 dated November 1982 Received from GSU.
Sent to GSU by TSI-to 1-hr fire endurance tests conducted on thermo-justify 1-hour design lag applied by direct spray on 4"
conduit without stress skin, containing
" generic" cables.
Met circuit integrity and temperature criteria.
Maximum temperature was 315
- F.
June 1, 1983 ITL Report No. 83-5-472A dated July 1983 Provided to GSU in response to questions on 1-hr fire endurance test conducted on thermo-1-hr barrier without lag applied by direct spraying, rolling and stress skin.
troweling methods to Class 1E cables installed Includes
" shrinkage in a modified ladder cable tray test article.
test."
Half of ladder back tray modified to simulate solid'botton tray.
12" cable tray.
June 7, 1983 ITL Report No. 83-5-472 dated' September 1983 Report received from (Revision 1)
WPPS.
~
Witnessed by Bechtel and 1-hr fire endurance test conducted on thermo-ANI.
lag applied by direct spraying, rolling and trowelling methods to Class 1E cables 12" steel ladder tray.
+
installed in a modified ladder cable tray test article for WNP2.
Tray modified to simulate solid tray.
Cure accelerator used.
Test passed circuit integrity and temperature.
July 1984 ITL ReDort No. 84-6-109 I
l 30 minute fire endurance test. conducted on a ladder cable tray protected with thermo-lag.
4 Revised February 7, 1992
December 13, 1984 ITL ReDort No.
84-12-181 dated June 1985 Received from WPPS and LRevision 2)
APS.
Prairie Island holds copy.
3-hr fire endurance test conducted on a ladder cable tray with a P1000 unistrut attachment 12" steel ladder tray and transition section.
Unitstrut covered for 18" with 1.2" dry thickness Thermo-Lag.
Test met circuit integrity and temperature.
April 1985 ITL ReDort No. 85-4-235 3-hr ASTM E119 fire endurance test conducted on a fire wall test assembly protected with thermo-lag.
June 18, 1985 SwRI Proiect 01-8305-040B (CTP 1092A)
Test conducted for GSU.
3-hour fire endurance test conducted on 4"
Failed flex conduit conducted for Promatec and River Bend.
Report states installed by certified installer.
Thermo-Lag burned up during test.
l Test rerun without Thermo-Lag.
June 1985 ITL ReDort No. 84-12-294 Prairie Island holds copy.
i 3-hr fire endurance tests conducted on two conduit sections mounted immediately adjacent to a concrete wall.
June 1985 ITL Report No. 85-2-382. Revision 1 3-hr fire endurance test conducted on air drop cables and a unitsrut section connected to a 4" diameter electrical conduit.
I 5
Revised February 7, 1992 l
L
m November 21, 1986 Construction Technoloav Proiect No.
Report received from CTL.
CRE134/4324 dated January 1987 Test paid for by BV.
Thermo-lag cured for 3-4.
Failed 1-hr fire endurance tests of 3
3M/TSI days.
interface configurations.
Each vendor installed own barrier system.
Test plan 4"
aluminum conduits w/
included 325F acceptance criteria (S&W test preformed Thermo-lag.
plan).
Configuration A TC exceeded 325F.
Circuit integrity maintained for all 3M questioned
,TC configurations.
TSI installed own product.
placement.
Saran wrap on CTL QC observed construction of test' articles.
Thermo-Lag noted by 3M.
January 1987 ITL Report No. 86-11-155. Revision 1 1-hr fire endurance test conducted on a 18X30" air duct test assembly.
February 1987 ITL Report No. 86-10-49. Revision 1
~
1-hr fire endurance test conducted on a 4"
aluminum conduit test assembly containing generic cables.
March 31, 1987 ITL ReDort No. 87-3-606 dated ADril 1987 Received from WPPS 3-hr fire endurance test conducted on 2" and Failed 1.5" "in-situ" conduit-sections.
Applied by WPPS using low pressure extrusion procedure.
Conduits failed 325'F criteria at 1 hr 15 min.
j and circuit integrity at 2 hrs 5 min.
April 1,
1987 ITL Report No. 87-4-3 dated ADril 1987 Received from WPPS.
1-hr fire endurance. test conducted on 4" "in-situ" conduit sections.
Applied by WPPS using low pressure extrusion procedure.
8 Revised February 7, 1992
April 13, 1987 Warnock-Hersey International (WHI-495-PSV-Conducted for APS.
Test received from APS.
0543)
No hose stream test.
Failed 1-hr fire endurance test of 2"
conduit assembly cover with box enclosure of Thermo-Lag with a 3/4" diameter pipe passing through.
Test failed temperature criteria.
May 7, 1987 ITL Report No. 87-5-76 dated June 1987 Received from WPPS.
Conducted for WPPS.
3-hr fire endurance test conducted on 2"
Repeat of 3M 86-92.
conduit protected with Thermo-Lag interfacing with Interam E500/E-54A Flexible Wrap System.
3M installed by WPPS.
May 7, 1987 ITL Report No.
87-5-77 dated June 1987 Received from WPPS.
(Revision Il WNP2 gave copy to APS.
1-hr fire endurance test conducted on ladder 12" steel ladder tray tray with a
P-1000 unistrut attachment.
Unistrut temperature measured for 9 inches.
March 9, 1988 No Test Report Test run for GSU.
No test report.
Temperature 3-hr fire endurance test.
Circuit integrity records received from Failed acceptance criteria.
18" steel cable tray.
TSI.
Ribs and stress skin removed (as-installed).
Received from TSI.
Test at TSI, conducted by TSI. ' Test failed.
i Control of furnace lost.
i l
l l
9 Revised February 7, 1992 i
a e
,m m
m
July 30, 1988 ITL Report No. 88-07-5982 dated September 29.
Test received from TSI 1988 and GSU.
Test run for GSU.
Rerun of March 1988 Failed 3-hr fire endurance test with circuit test.
integrity and 325'F above ambient as acceptance criteria.
12" steel cable tray with ribs and stress skin removed.
Test at TSI, conducted by TSI.
GSU QC witnessed.
Test failed at 115 minutes.
April 14, 1989 Construction Technoloav ReDort dated October Test received from GSU.
1989 (Revision 1) (240056/824-63)
Another CTL Report dated September 1989 was 3-hr fire endurance test with circuit provided to TSI.
Same integrity as acceptance criteria. Constructed test, but was revised.
by
Technology Lab.
CTL witnessed construction.
GSU provided the cable Test passed both circuit integrity and trays.
temperature.
R.
Hall questioned thickness Joints pre-buttered.
numbers.
Surface coat applied by TSI prior to Developmental test.
entry into furnace.
Cure time of 10 days.
May 5, 1989 Construction Technoloav Report dated October Report received from CTL 1989 (Revision 1) (240056-824/824-59) and RBS.
Conducted for GSU.
Failed 3-hr fire endurance test on 30" aluminum ladder back tray.
Maximum temperature reached 30" aluminum ladder bac 't 363F.
Average of T/C was 316
'F.
Test cable tray.
One layer of conducted for GSU.
GSU questioned extra cable.
Other systems coating that had been applied.
Question tested but not in report.
raised on thickness of material (1" vs. 1.5").
CTL witnessed installation.
10 Revised February 7, 1992 w
May 5, 1989 Construction Tecanoloav ReDort dated November Report received from CTL.
1989 (240056-824/u24-75L Developmental test for TSI.
3-hr fire endurance test on a
" modified" 12" steel ladder back design.
Maximum cable temperature was 320F.
cable tray.
WNP2 supplied the cable tray.
May 5, 1989 Construction Technoloav ReDort dated November Report received from CTL 1989 (240056-824/824-77) and RBS.
Developmental test for Failed 3-hr fire endurance test on 30" aluminum cable TSI.
tray.
Installation by TSI.
CTL witnessed 30" aluminum cable tray.
installation and test.
Max cable temperature Tray supplied by GSU.
reached 331F.
Average T/C was 271
'F.
Thickness ranged from 1"
to 1.5".
TSI installed barrier.
October 26, 1989 SwRI Proiect No.
01-2702 dated May 1991 Test conducted for GSU.
(Draft)
Test not submitted for NRC review.
GSU 3-hr qualification test on two protective submitted Information envelope systems for Class 1E electrical Report to NRC.
circuits.
GSU installed material.
Thermo-lag Prefab panels /prebuttered covered tray failed in 47 min.,
30" aluminum cable tray.
catastrophically at 82 min.
Promat passed.
(1-hr test not
.TSI stated not installed properly.
GSU documented) declared test inconclusive.
~
11 Revised February 7, 1992 t
I
Enoineerino-Test ReDort dated ' January 1991 Received from :RBS visit.'
7
November.
20 December 17, 1990 (Preliminary)
Draft test report.
Some j test results reported in _-
1-hr and 3-hr fire endurance tests with LER 90-003.
and circuit integrity as acceptance criteria.
supplements.
GSU 'QC witnessed.
In-situ and upgraded 18" aluminum ladder, tray configurations tested.
As-installed tests failed.
Upgrades passed except for 3-hr tray.
(Acceptance criteria used not IAW NRC requirements) t L
12 Revised February 7, 1992 i
jj ATTACHMENT 7 LIST OF AMPACITY DERATING TEST REPORTS INVOLVING THERMO-LAG 330-1 FIRE BARRIER SYSTEMS KNOWN TO THE NRC This attachment lists the ampacity derating test reports involving' Thermo-Lag 330-1 Fire Barrier Systems that are known to the Special Review Team. Tests were conducted by Industrial Testing Laboratory (ITL),
Thermal
- Science, Incorportated (TSI),
Underwriters Laboratories (UL), and Southwest Research Institute (SWRI).
The table includes the test report number, date of test report, description of the test, and the ampacity derating factors.
It should be noted that a national standard test procedure does not exist and the results. of the individual tests should not be compared without the detailed review of the test article configuration.
b
LIST OF AMPACITY DERATING TEST REPORTS INVOLVING THERMO-LAG 330-1 FIRE BARRIER SYSTEMS KNOWN TO THE NRC Report Number Date of Report Description of Test DeratinG Factor TSI Technical 9/81 14" steel cable tray 7%
Note 92981 600 volt power cable 10%
210'
- 00 AWG 562" D l
l 2820' #10 AWG 215" D l
TSI Technical 2/85 (Rev 5) 1-hr 2" steel conduit 7.47%
Note 111781 11/81 (Orig) 600 volt power cable 21' #00 AWG.562" D ITL 82-355-F1 1/85 (Rev 1) 1-hr solid bottom tray 2.48%
7/82 (Orig) 600 volt power cables 12.39%
210' #00 AWG 0.0562"D 2820' #10 AWG 215" D ITL 82-5-355C 7/82 3-hr ladder back tray 16.15%
1000 volt power cable 16.86%
1485' #08 AWG 0.286"D 17.68%
870'
- 04 AWG 0.377"D 300'
- 2/0 AWG 0.617"D ITL 82-5-355F 7/82 3-hr ladder back tray 16.15%
1000 volt power cable 16.86%
1485' #8 AWG
.286" D 17.68%
870'
- 4 AWG
.377" D 300'
- 2/0 AWG.617" D ITL 84-3-275A 3/84 3-hr ladder back tray 20.55%
1000 volt power cables 19.24%
1485' #8 AWG 286" D 20.07%
870' #4 AWG.377" D 300' #2/0 AWG.617" D 2
Revised February 7, 1992 3
___.______.._.-.m
_________m_
___.m._
O ITL 84-10-5 10/84 3-hr 2" steel conduit 9.72%
J-600 volt power cable 21' #00 AWG.562 D SwRI Project 8/86 1-hr 24" steel ladder back tray 37.4%
0 1-8 8 18-
- 6 AWG 0.75" D 208/209a (3M PJ-19)
SwRI Project 10/86 1-hr 4" steel conduit 1.27%
0 1
-8 8 1 8 -
- 6 AWG
.75" D 208/209c (3M PJ-19)
UL 86NK23826 1/87 1-hr 24" ladder back tray 28.04%
File R6802 3-hr'24" ladder back tray 31.15%
(South Texas) 2000 volt power cable 2980' #6 AWG 1" D UL 86NK23826 1/87 1-hr 4" steel conduit 0%
File R6802 3-hr 4" steel conduit 9.4%
(South Texas) 2000 volt power cable 294' #6 AWG 1" D 3
Revised February 7, 1992
~
a.,.sg
.r va a
m_
+
p, g.
+
1 4.
- i
'4 Y g.
LIMITED DISTRIBUTION - NOT FOR PUBLIC DISCLOSURE REFERENCES FOR FINAL.. REPORT OFFICE OF NUCLEAR REACTOR REGULATION.
SPECIAL REVIEW TEAM FOR THE REVIEW OF THERMO-LAG FIRE BARRIER PERFORMANCE J
r
' I i
l
- i LIMITED DISTRIBUTION - NOT FOR PUBLIC DISCLOSURE i
l i
1
e 1.
Memoranda.from T.E. Murley, NRR, to F.J. Miraglia, et al, NRR, July 12, 1991 and August 15, 1991 (Attachment 1).
i 2.
Memorandum of Understanding Concerning the Conduct of the Thermo-Lag Investigation, signed by B.B. Hayes, OI, and D.C. Williams, OIG, August 7, 1991.
3.
Memorandum from E.T.
Pawlik, Region III/OI, to L.R.
Plisco and K.S. West, NRR, December 24, 1991.
Memorandum from K.E. Walker, OIG, to K.S. West and L.R. Plisco, January'29, 1992.
4.
Letter from confidential alleger to R. Wise, Region IV,.
February 21, 1991 (Allegation RIV-91-A-0022).
t 5.
Allegation NRR-91-A-0011.
6.
ARB Summary approved by B.K. Grimes, NRR, March 5, 1991.
7.
Memorandum from T.E. Murley, NRR, to F.J. Miraglia, et al, NRR, " Priority Ranking System for Review Efforts - Revision 1," March 24, 1989.
8.
Notes prepared by D. Notley, NRR, May 10, 1991.
9.
Memorandum from T.E. Murley, NRR, to B.B. Hayes, OI, April 12, 1991.
10.
OI Response to Request for Investigation fro,n E.T. Pawlik, RIII/OI, to T.E. Murley, NRR, April.23, 1991 (Case No.
3-91-006).
11.
Memorandum of Understanding Concerning the Conduct of the Thermo-Lag Investigation, signed by'B.B. Hayes, OI, and D.C. Williams, OIG, August 7, 1991.
(See Reference 2 for copy.)
12.
GSU Licensee Event Report (LER)89-009, " Inadequate Thermo-Lag Coverings as Fire Barrier Per T.S.
7.7.7.a,"
April 17, 1989.
13.
Letters from J.E.
Booker, GSU, to U.S. NRC, December 20, 1989 and January 9, 1990.
14.
GSU LER 90-003, " Inadequate Thermo-lag Fire Barrier Envelopes Surrounding Safe Shutdown Circuits Per T.S.
3/4.7.7,"
March 8, 1991 and Rev.
1, July 12, 1990, Rev.
2, February 4, 1991, and Rev.
3, June 28, 1991.
1
g.;
GSU LER 91-008," Lack of Fire Wrap - Inadequate Fire Barrier Caused by Inconsistency in Design Bases Documentation, May 15, 1991.
15.
Memorandum from J.J. Petrosino, NRR, to L.J. Norrholm, NRR, June 11, 1991.
16.
Memorandum from T.E. Murley, NRR, to F.J. Miraglia, L.R. Plisco, and J.F. Stang, NRR, July 12, 1991.
(See Reference 1 for copy.)
17.
Memorandum from T.E. Murley, NRR, to F.J. Miraglia, L.R. Plisco, and K.S. West, NRR, August 15, 1991.
(See Reference 1 for copy.)
18.
Memorandum from F.J. Miraglia, NRR, to B.K. Grimes and A.C. Thadani, NRR, July 30, 1991.
19.
NRC Manual Chapter 0517, " Management of Allegations,"
April 3, 1990. (Copy not included.)
20.
Appendix VII to letter from R.
Feldman, TSI, to F.J. Miraglia, NRR, October 5, 1991.
21.
SWRI Report 03-6491, " Fire Qualification Test of a Protective Envelope System," (Title page only),
October 27, 1981.
i Letter from R.L. Tedesco, NRR, to R.J. Gray, Texas Utilities Generating Company, December 1, 1981.
22.
Letter from N.W.
Curtis, PP&L, to A.
Schwencer, NRR, February 9, 1982 23.
Memorandum from W.V. Johnston, NRR, to R.L. Tedesco, NRR, April 16, 1982.
24.
Letter from N.W. Curtis, PP&L, to A. Schwencer, NRR, May, 12, 1982.
TSI Technical Note 8232-1, " Engineering Test Report - One Hour ASTM E-119 Fire Simulation Facility Fire Test, June 1981.
(Title page only.)
l l
25.
Memorandum from W.W. Johnston, NRR, to R.L. Tedesco, NRR, June 28, 1982.
26.
Letter from N.W. Curtis, PP&L, to A. Schwencer, NRR, August 25, 1982.
SwRI Report 01-7163, " Qualification Fire Test of a Protective Envelope System," August 1982. (Title page only.)
2 l
t
. ~.
27.
Memorandum from W.V. Johnston, NRR, to T. Novak, NRR, September 15, 1982.
28.
ITL Report 82-5-355B, "Three-Hour Fire Endurance Tests on Thermo-Lag 330-1 Subliming Coating Envelope System for WPPSS Nuclear Projects," July 1982.
(Title page only.)
Memorandum from W.V.
Johnston, NRR, to T. Novak, NRR, December 27, 1982.
29.
Letters from G.D. Bouchey, WPPSS, to A.
Schwencer, NRR, September 20, 1982 and October 4, 1982.
30.
Inspection Report 50-397/86-05, April 2, 1986.
(Relevant pages only.)
31.
Inspection Report 50-397/87-02, April 17, 1987.
(Relevant pages only.)
32.
Memorandum from L. R.
Plisco and K.
S. West, NRR, to F.J. Miraglia, NRR, December 11, 1991.
33.
WPPSS LER 91-004, " Inadequate Fire Protection (Thermo-Lag) of Division II Safe Shutdown Cables Due to Inadequate Installation and Inspection," April 28, 1991.
34.
Letter from W.J.
Cahill, TU Electric, to U.S.
- NRC, October 12, 1989.
35.
Letters from C.I. Grimes, NRR, to W.J. Cahill, TU Electric, December 1, 1989 and January 31, 1990.
36.
Inspection Report 99901038/85-01, March 11, 1986.
(Relevant pages only.)
37.
Item D.5.2-1, NRC Inspection Report 50-285/85-22, December 13, 1985.
(Relevant pages only.)
38.
Letter from R.
Feldman, TSI, to J.J. Petrosino, NRC, April 13, 1987.
Letter from M.E. Grau, TSI, to J.J.
Petrosino, NRC, April 14, 1987.
39.
Letter from R.R.
Licht, 3M, to R.
Clemens, OPPD, February 18, 1986.
40.
Mailgram from R. Feldman, TSI, to U.
S. NRC, October 2, 1986.
Mailgram from R. Feldman, TSI, to Consumers Power Company, October 2, 1986.
3
?
n
~,
\\
,4 41.
Letter from C.E. McCracken, NRR, to the alleger, June 20, 1989.
42.
Memorandum from J.J.
Petrosino, NRR, to L.J. Norrholm, NRR, June 11, 1991.
(See Reference 15.)
43.
Letters from F.J. Miraglia, NRR, to R.
Feldman, TSI, September 10, 1991 and September 18, 1991.
44.
Letters from R. Feldman, TSI, to F.J. Miraglia, NRR, September 12, 1991 and October 5, 1991 (without enclosures).
43.
Official Transcript of Proceedings, " Meeting with Thermal Science, Inc., to Discuss Issues Involving Thermo-Lag 330,"
October 17, 1991.
(Title page only.)
46.
Note from K.S. West, NRR,
'm Thermo-Lag File, September 24, 1991.
47.
Letters from F.J. Miraglia, NRR to R.
Feldman, TSI, October 3.1, 1991.
48.
Letters from R. Feldman, TSI, to F.J. Miraglia, NRR, November 8, 1991, November 12, 1991, and December 3, 1991.
49.
CEI LER 91-020, " Cable Tray Raceways FoOnd to be Impaired as a Fire Barrier Adversely Affecting Safe Shutdown Requirements," November 19, 1991.
50.
Memorandum from L.R. Plisco.and K.S. West, NRR, to F.J. Miraglia, NRR, December 18, 1991.
51.
Memorandum from L.R. Plisco and K.S. West, NRR, to F.J. Miraglia, NRR, October 31, 1991.
52.
Memorandum from L.R.
Plisco and K.S. West, NRR, to F.J. Miraglia, NRR, December 24, 1991.
53.
Memorandum form L.R. Plisco and K.S. West, NRR, to F.J. Miraglia, NRR, December 11, 1991.
(See Reference 32 for copy.)
54.
Memorandum from L.R. Plisco and K.S. West, NRR, to F.J. Miraglia, NRR, January 7, 1992.
55.
Mailgram from R.
Feldman, TSI, to U.
S. NRC and the licensees, October 2, 1986.
(See Reference 40 for copy.)
56.
Memorandum from F.J. Miraglia, NRR, to R.D. Martin, Region IV, November 12, 1991 Memorandum from F.J. Miraglia to J.B. Martin, Region V, 4
p January 9, 1992.
Memoranda from F.J. Miraglia, to A.B. Davis, Region III, January it, 1992 and February 6, 1992.
57.
Memorandum from L.R. Plisco, NRR, to R.C. Paul, Region III/OI, November 22, 1991.
58.
Letter from R.A. Lohman, TSI, to F. Garrett, Arizona Public Power, et al, August 23, 1991.
59.
Letter from R.
Feldman, TSI, to F.J. Miraglia, NRR, t
October 5, 1991.
(See Reference 44 for copy.)
60.
SWRI-Draft Report 01-2702, "Three-Hour Qualification Test on Two Protective Envelope Systems for ' Class 1E Electrical Circuits and Five Penetration Seals /" May 1991.
'(Title page-only.)
61.
Letter from V.
Babraskas and K.D.
Steckler, NIST to K.S. West, NRR, January 21, 1992.
62.
Memorandum from R.C. Wilson, NRR, to L.J. Norrholm, NRR, December 12, 1991.
i 63.
Memorandum from F.J. Miraglia, NRR, to L.J. Norton, OIG, December 9, 1991.
64.
Memorandum from L.J. Norton, OIG, to F.J. Miraglia, December 30, 1991.
65.
Memorandum from F.J. Miraglia, NRR, to A.C. Thadani, NRR, January 17, 1992.
66.
Memorandum from J.M. Taylor, EDO, to The Chairman and the Commissioners, February 7, 1992.
67.
Interagency Agreement NRC-03-91-032, " Technical Assistance-for the Evaluation of Fire Barrier Performance,"
October 21, 1991 (FIN L2063).
68.
Notes from K.S. West, NRR, to File (FIN L-2063),
December 30, 1991, December 2, 1991, October 22, 1991, September 27, 1991, and September 5, 1991.
69.
NRC Order DR-91-1368, September 27, 1991.
Union Electric Company Miscellaneous Sales Ticket 8355, September 27, 1991.
70.
Letter from W.S.
Schwink, NRR, to R.Y. Lowery, DOE, "Sandia National Laboratory Technical Assistance to NRR - Evaluation 5
w e?L of Fire Barrier Performance," FIN L2414, December 16,_1991.
71.
NUREG 0500, Recommendations Related to Browns Ferry Fire, February 1976.
(No copy provided.)
72.
U. S. NRC, " Fire Protection Program for Nuclear Power Plants Operating Prior to January 1, 1979," Federal Reaister, Vol. 45, No. 105, May 29, 1980, pp. 36082-36090.
73.
U. S. NRC, " Fire Protection Program for Operating Nuclear Power Plants," Federal Reaister, Vol. 45, No. 225, November 19, 1980, pp. 76602-76616.
74.
Memorandum from W.J. Dircks, EDO, to H.R. Denton, NRR, at al, September 13, 1984.
75.
Memorandum from Fire Protection Policy Steering Committee to W.J. Dircks, EDO, October 26, 1984.
76.
GL 85-01, " Fire Protection Steering Committee Report,"
January 9, 1985.
77.
American Society of Testing and Mateials (ASTM) Standard E119, " Standard methods of Fire Tests of Building Construction and Materials."
National Fire Protection Association (NFPA) Standard 251,
" Standard Methods of Fire Tests of Building Construction and Materials."
78.
TSI Technical Note 11171, " Engineering Report on Ampacity Test for 600 Volt Power Cables Installed in a Five Foot Length of Two Inch Conduit Protected with Thermo-Lag 330-1 Subliming Coating Envelope System," November 1981.
79.
UL Test 86NK23826, "Special Services Investigation of Ampacity Ratings for Power Cables in Steel' Conduits and in Open Ladder Cable Trays with Field Applied Enclosures,"
January 21, 1987.
80.
ITL Report 82-5-355F, "Ampacity Derating Test for 1000 Volt Power Cables in a Ladder Cable Tray Protected With a Three.
Hour Rated Design of Thermo-Lag 330-1 Subliming Coating Envelope System," July 1982.
81.
SwRI Project 01-8818-208/209-a, "Ampacity Derating of Fire Protected Cables in Cable Trays Using a One-Hour Subliming, RigidBoard Provided by 3M, Incorporated," August 11, 1986.
(Title page only.)
82.
UL Outline of Investigation, " Tests for Ampacity of Insulated Electrical Conductors Installed in Fire Protective 6
I
P
~
.Y.
,n Systems," Subject 1712, July 1984.
83.
Draft Bechtel Specification, " Proposed Standard for the Determination of Ampacity Derating of Fire Protected I
Cablec," May 27, 1986.
84.
UL Outline of Investigation, " Fire Tests for Electrical Circuit Protection Systems," Subject 1724, May 1984.
85.
UL Buildina Materials Directory, " Electrical Circuit Protective Systems (FHIT)," December 14, 1990, pp.
202 - 215.
w e
O 7
i o
ATTACHMENT 8 Chronoloov of Review Team Activities February 21, 1991 Allegations sent to Region IV by alleger.
February 27, 1991 NRR assumed responsibility for resolution of allegations.
April 12, 1991 NRR requested OI investigation.
April 23, 1991 OI initiated investigation.
.i May 1 & 30, 1991 Alleger interviewed by OI.
May 13,22, June 20, 1991 Former TSI employee interviewed by OI.
May 29-30, 1991 VIB and RIII conducted site visit to RBS.
I j
June 11, 1991 VIB Trip Report from RBS issued.
June 27, 1991 NRR Review Team members assigned.
July 9, 1991 Review Team met with OI Region III.
]
July 12, 1991 Review Team Charter established.
July 23, 1991 Prior involvement memos provided to OIG.
July 27, 1991 Review Team membership revised.
.Tuly 30, 1991 All NRR followup activities for TSI issues transferred to Review Team.
August 6, 1991 Information Notice 91-47 issued.
August 15, 1991 Revised Review Team Charter issued.
August 23, 1991 TSI letter to licensees regarding IN 91-47.
August 26, 1991 NIST RFPA issued.
September 5, 1991 Kickoff meeting with NIST/OI/OIG.
September 10, 1991 Letter to TSI requesting response to technical issues.
i I
September 25, 1991 NIST/NRC Interagency agreement authorized.
September 26, 1991 Meeting with NIST.
October 7-8, 1991 Review Team site visit to River Bend.
l l
__________________________________________ _ _ __________g
1 0
4 f.
October 10, 1991
.Thermo-Lag panels provided to-NIST.
October 17, 1991 NRC meeting with TSI.
October 18, 1991 Meeting with NIST.
November 4-5, 1991 Review Team site visit to Comanche Peak.
November 5, 1991 Perry 50.72 notification on banding.
November 6-7, 1991 Review Team site visit to WNP2.
November 19, 1991 Review Team site visit to Perry.
November 26, 1991 Statement of Work issued to Sandia.
November 27, 1991 Meeting with NIST.
December 6, 1991 Information Notice 91-79 issued.
December 16-20, 1991 VIB inspection at TSI.
December 17, 1991 Meeting with NIST.
December 19, 1991 Review Team /NIST site visit to Callaway.
December 20, 1991 TSI Vendor Inspection Exit meeting.
January 16, 1992 Review Team briefing of RBS inspection team.
January 21, 1992 NRR/OIG meeting in preparation for transfer of issues to line organization.
January 21, 1992 Meeting with NIST.
January 22, 1992 Review Team briefing of NRC Fire Protection Inspector Counterpart Meeting.
January 24, 1992 Final draft generic letter provided to ET.
January 31, 1992 Memorandum sent to Commission informing them of plans for generic communications.
l January 31, 1992 Review Team activities completed, i
l
.o
..g.
adequacy.
States that fire tests to -be conducted by April 1987 of untested configurations.
Also states ampacity issued closed, but discusses 9.4% and 17.7% derating.
May 7, 1987 3M/TSI interface conducted by TSI for WNP2.
Test passed.
June 6-10, 1988 Fire protection inspection (88-16).
Closes intervening combustible issue based on TSI letter.
(Missing para. 5 of report)
February 14, 1990 SAIC TER on WNP 2 re-evaluation.
March 28, 1991 Application deficiencies noted on Thermo-lag barriers.
Inadequate thickness and incomplete transition from spray to board.
August 23, 1991 Letter to J.
Kittler from TSI responding to the Information Notice 91-47.
e e,
ATTACHMENT 9 RIVER BEND HISTORY AND DOCUMENTATION REVIEWED BY SPECIAL REVIEW TEAM (50-458)
June 30, 1982 GSU submitted comparison of fire protection program to Appendix R with FSAR amendment.
April 6, 1983 Draft SER on fire protection.
Safe shutdown and fire barriers remained open items.
May 1984 Safety Evaluation Report (NUREG-0989) issued.
Safe shutdown still under review.
Page 9-42 stated that all cable trays are steel construction.
August 9, 1984 TSI letter to Stone and Webster provided 1-hour, 3-hour, ampacity derating, and seismic qualification information.
September 24, 1984 First purchase order for TSI from GSU.
April 4, 1985 Onsite audit of fire protection by NRR.
Fire barrier wrap installation not completed.
May 10(?), 1985 3-hour qualification tests of "as-installed" internal and external aluminum conduit seals conducted at SwRI by B&B/Promatec and GSU.
(Thermo-Lag does not appear to have been used on test articles.)
Reference SWRI Project Report 01-8305-040a.
June 18, 1985 3-hour fire test of flex conduit penetration seals conducted at SwRI by B&B/Promatec and GSU.
Thermo-Lag on conduit disintegrated during the test.
Reference SwRI Project Report 01-8305-040b dated July 1985 and Promatec Report CTP-1092a dated October 31, 1985.
(According to Promatec Report CTP-1092b, GSU concluded that this test provided conclusive evidence that Thermo-Lag protection is not required to maintain flex conduit integrity.)
j June 20, 1985 June 18, 1985 test rerun without Thermo-lag.
Reference SwRI Project Report 01-8305-040c dated August 1985 and Promatec Report CTP-1092b dated November 1, 1985.
July 26, 1985 GSU submitted SwRI Report 01-8305-040a dated June 1985 to NRC to justify Appendix R 1
i i
1
~
b%
.l deviation for aluminum conduit seals.
August 1985 SSER 3 issued.
Deviation granted for sealing inside aluminum conduits based on SwRI test report submitted to NRC on July 26, 1985.
Also granted deviation on delay of installation of fire wrap on spent fuel cooling system and service water.
February 23, 1987 Condition Report 87-0236 initiated following review of February 3, 1987 surveillance test results.
Sixty percent of the barriers failed the acceptance criteria.
This CR superseded by CR 87-0236A.
i March 12, 1987 Condition Report 87-0236A initiated.
Identified specific examples of deficient fire barriers.
Cracks, wear conditions, unsealed joints, unprotected supports, and 18 inch criteria not' met.
Breakdown of vendor QC inspection program identified as root cause.
March 25, 1987 LER 87-005 submitted oy GSU.
Described fire barrier' deficiencies, including incomplete construction, found during surveillance test procedure.
Hourly fire watch patrols established in accordance with Tech Spec 3/4.7.7.
GSU committed'to make'100 percent visual inspection.
April 30, 1987 Internal RBS memo summarized the results of 100 percent fire barrier-inspections committed to in LER 87-005.
May 20, 1987 Internal memo from D. Beauchamp (GSU QC) on insufficient training given by TSI on May 18-19, 1987.
June 30, 1987 Condition Report 87-0698 initiated. Inspections required by CR 87-0236A and LER 87-005 found Thermo-Lag installations that did not conform to GSU requirements.
July 1, 1987 Condition Report 87-0705 initiated. Inspections required by CR 87-0236A and LER 87-005 found removal of ribs from Thermo-Lag panels during installation.
This resulted in gaps in stress skin where ribs were cut.
July 13, 1987 Condition Report 87-0745 initiated. Inspections required by CR 87-0236A and LER 87-005 found 2
stress skin and ribs removed from fire barriers.
During discussions with GSU, TSI stated that removal would have little or no effect.
July 24, 1987 GSU issued purchase order (7-ON-72055) to TSI to conduct 3-hr fire endurance tests with stress skin removed.
August 20, 1987 TSI prepared engineering test plant for 3-hr-fire endurance test for GSU.
January 6, 1988 Condition Report 88-0012 initiated.
During installation of additional cables, control building cable tray found to have stress. skin and ribs removed.
January 19, 1988 GSU letter to TSI asked for confirnation that use of 1-hr Thermo-Lag panel is acceptable with ribs and stress skin removed.
(GSU referenced CR 88-0012.)
January 22, 1988 TSI letter to GSU submitted ITL Report 82 241 (direct spray application on conduit) and ITL Report 83-5-472A (direct spray, etc. to
)
cable tray) to justify installation of 1-hour i
barriers without stress skin.
i February 1, 1988 GSU provided comments on test plan to TSI.
GSU stated that acceptance criteria should be 325'F above ambient.
(NRC criteria is 250*F above ambient.)
l February 4, 1988 Interim disposition prepared by GSU for CR 88-0012.
Stated that "as-installed" 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />.
barriers were " qualified for use as is" based on TSI's January 22, 1988 letter.
See August-29, 1990.
February 8, 1988 Revision I to TSI test plan for 3-hour test.
February 23, 1988 TSI letter to GSU provided Revision I to TSI test plan for comment.
j March 9, 1988 3-hour fire test conducted on "as-installed" configuration (12 inch wide steel tray, stress-skin and ribs removed) at TSI for GSU.
Test terminated due to loss of control of the furnace.
Circuit integrity lost at 2 hr 15 min.
No test report drafted.
- However, thermocouple records provided to NRC by TSI attorney.
(July 29, 1988 test was intended as 3
~-
e i
e '
repeat of this test.)
P March 15, 1988 GSU initiated audit of RBS fire protection program.
April 21, 1988 Internal GSU memo addressed results of March-1988 fire protection program' audit.
The effectiveness of fire barrier inspection surve' 'ance test procedure (STP) was que d (pg 8).
LER 88-009 is referenced.
June 22, 1988 Revision II to TSI test plan for 3-hour test.
July 22, 1988 CR 88-0594 initiated.
July 29, 1988 3-hour fire te' conducted on "as-installed" configuration nch wide steel tray, stress i
skin and ribs ved) at TSI for GSU.
Test specimen failed on temperature, circuit integrity, and structurally.
Reference ITL Report 88-07-5982 dated September 29, 1988.
July 29, 1988 Condition Reports 88-0607 and 88-0608 initiated.
Addressed 3-hour test failure at TSI.
October 11, 1988 TSI letter t.
JU transmitted final test report ITL 88-07-5984 dated September 29, 1988.
325'F exceeded at 90 minutes, circuit integrity failed at 115 minutes.
December 12, 1988 SwRI letter to-GSU transmitted Addendum 1 to proposal for 3-hour fire endurance tests of "as-installed" and proposed " repaired" configurations.
4 January 1989 GSU initiated test program with SWRI to qualify "as-installed" Thermo-Lag fire barriers.
February 3, 1989 GSU letter informed TSI of plans to conduct fire tests at SwRI during March 1989 and provided copy of SWRI test plan.
March 3, 1989 TSI letter to GSU discussed developmental tests of proposed design modifications conducted at TSI and tests scheduled for April 1989 at CTL.
TSI offered to exchange new 3-hour panels for used RBS materials and stated that tests at SwRI would not be cost effective.
Letter acknowledged " prerequisites of government-regulatory agencies," but stated that temperature measurements are not required and 4
are for information purposes only.
Spring 1989 GSU put testing at SWRI on hold in favor of-testing with TSI.
March 16, 1989 Condition Report 89-0204 initiated.
Stress skin and ribs found removed from Thermo-Lag fire barriers.
March 31, 1989 GSU letter to TSI discussed upcoming testing at CTL.
GSU asked to observe construction of test specimen and to witness the test.
LER 88-009' referenced.
April 7, 1989 TSI letter to GSU provided details of upcoming (May 5, 1989) tests at CTL.
April 14, 1989 3-hour fire test conducted at CTL on 30-inch wide aluminum cable tray.
Test passed. Rig constructed by TSI.
CTL witnessed installation.
Reference CTL Report 824-63 and Revision 1.
April 17, 1989 LER 89-009 submitted to NRC describing removal' of ribs and stress skin from 1-hour Thermo-Lag fire barriers during construction.
Fire watches established.
100 percent inspection planned.
May 5, 1989 3-hour fire endurance tests conducted at CTL on "as-designed" 30" aluminum cable trays.
GSU witnessed initial construction of article at TSI.
Test passed.
However, GSU questioned
. application of extra coating on test article by TSI, which GSU could not reproduce.
Reference CTL Reports 824-59 and 824-77.
Summer 1989 Additional questions raised by GSU on original qualification of Thermal-Lag.
August 1989 GSU re-established test program with SwRI.
October 12, 1989 RBS QC issued memo describing construction of 1-hour and 3-hour fire barrier test assemblies-for October 1989 tests at SwRI.
1-hour barrier I
had ribs and stress skin removed, 3-hour 1
barrier was "as-designed."
October 13, 1989 GSU letter invited TSI to inspect cable tray test assemblies prior to October 26, 1989 fire endurance tests at SwRI.
5 l
1
4 October 23, 1989 TSI letter informed GSU that they could not inspect test articles at SwRI.
October 26, 1989 3-hour fire test of 30-inch wide "as-designed" aluminum cable tray conducted at SwRI.
Promat and Thermo-Lag tested.
GSU constructed test articles.
Thermo-lag failed at steel support and tray radius in less than one hour.
TSI claimed TSI/Promat interface and installation problems caused the failure and claimed test not valid.
(Final report not issued as of September 1991.)
October 26, 1989 1-hour fire test at SwRI?
Documentation not available as of September 1991.
October 26, 1989 Condition Report 89-1144 initiated.
" Qualification testing of 3-hour fire rated material fails to meet requirements of 10CFR50 App R."
All 1-hour and 3-hour Thermo-Lag barriers considered inadequate.
October 31, 1989 TSI letter to GSU transmitted CTL Report 824-59 (05/05/89 test) and CTL Report 824-63 (04/14/89 test).
Stated thickness of panels was 1.250 inch +/- 0.250.
Enclosed Tech Note 20684, Revision V with addendum (Tech Note 20684-AL) for aluminum cable trays.
November 8, 1989 TSI met with GSU on test results.
November 9, 1989 GSU letter informed TSI of October 26, 1989 test results at SWRI and asked for evaluation.
Letter states that thermocouple data for 1-hour and 3-hour SwRI tests were previously submitted to TSI.
November 9, 1989 TSI letter to GSU discussed test results.
TSI claimed that Thermo-Lag /Promat interface and incorrect installation (gap. widths, stress skin separation) caused the October 1989 SWRI test failure.
TSI concluded that test was not valid.
November 10, 1989 GSU memorandum summarized November 10, 1989 meeting at RBS with TSI to discuss test failures.
GSU memo stated that Part 21 reportability was being evaluated.
November 14, 1989 TSI letter to GSU confirmed plan for plant walkdowns by TSI.
6 i
4 November 17, 1989 GSU letter to TSI provided construction details of fire barriers installed at RBS for TSI review.
December 7, 1989 GSU met with TSI at RBS to discuss corrective actions for CR 89-1144.
December 7, 1989 TSI letter to GSU discussed upgrades and qualification testing.
December 18, 1989 TSI letter to GSU provided meeting summary from December 7, 1989 meeting.
Provided TSI proposed upgrades and stated that ampacity reduction analysis had been performed.
December 18, 1989 GSU memorandum provided results of December 6/7, 1989 meeting with TSI at RBS.
December 20, 1989 GSU submitted Informational Report to NRC regarding testing of Thermo-lag fire barriers.
Discussed questions concerning tray materials, joint sealing methods, and size of test configurations.
CR 89-1144 referenced.
GSU stated that their investigation revealed generic concerns and that the " test results are indeterminate."
January 9, 1990 Revision to December 20, 1989 Informational-Report.
Revised fire watch status.
October 1989 SwRI test results still under evaluation by GSU. Committed to update report by March 31, 1990 (LER 90-003, dated March 8, 1990?).
January 11, 1990 GSU letter to TSI stated that October 26, 1989 test failure would have occurred on tray without barrier interface problem.
Also questioned gap widths and joint sealing.
January 26, 1990 NRC safe shutdown inspection (90-02) at RBS.
Violation identified concerning FRA requirement not implemented by procedure.
February 7, 1990 Condition Report 90-0095 initiated. Thermo-Lag fire barriers in pipe tunnel found to have surface cracks on visual inspection.
Fire watches established.
Cracks to be patched.
February 8, 1990 Condition Reports 90-0101 and 90-0106 initiated.
During surveillance inspection, deficiencies (small holes, cracks and unfilled i
seams) were found in Thermo-Lag fire barriers.
l i
u 9
March 8,-1990 LER 90-003 submitted.
Stated that surveillances conducted in February identified barrier deficiencies.
References CRs 90-0094, 90-0095, 90-0101, and 90-0106.
Committed to supplemental response to this LER and the January 9, 1990 Informational Report by July 15, 1990.
March 9, 1990 GSU met with TSI at RBS to discuss tests required to qualify "as-installed" barriers and potential repairs and upgrades.
March 19, 1990 TSI telecopied preliminary test procedure-for 11/90 and 12/90 tests to GSU for comment.
(No copy.)
April 1990 Test plan completed by TSI.
July 12, 1990 LER 90-003 Rev. 1 submitted to NRC.
Stated that 1-hour and 3-hour fire tests are planned to (1) evaluate "as installed" barriers and (2) develop repair methods.
July 17, 1990 GSU issued purchase order (90-G-70456) to TSI.
to conduct 1-hour and 3-hour fire tests.
August 20, 1990 GSU and TSI agreed on schedule for construction and testing (for 11/90 and 12/90 tests at TSI).
August 27, 1990 TSI began assembly of test articles for 11/90 and 12/90 tests.
August 29, 1990 GSU memorandum stating that fire tests (october 1989 tests at SWRI?) failed to verify use of Thermo-Lag panels with ribs and stress skin removed for 1-hour fire rating.
Reference TSI's January 22, 1988 letter.
September 10, 1990 GSU QC inspector arrived at TSI facilities to examine and monitor installation of test articles for 11/90 and 12/90 tests at TSI.
November 19, 1990 1-hour and 3-hour fire test program started at TSI facilities on "as-installed" and " upgraded" a
barrier configurations.
i December 18, 1990 Testing completed at TSI.
1-hour and 3-hour j
"as-installed" conduits and cable trays failed.
Some " upgrades" (addition of 1/4-inch thick panels) passed based on GSU acceptance criteria.
However, the 3-hour cable tray upgrade failed.
8 I
~ p December _22, 1990 ITL letter to GSU summarizing results of 11/90 and 12/90 tests performed by ITL at TSI lab.-
February 4, 1991 LER 90-003 Rev. 2 submitted to NRC.
Discussed ITL/TSI November and December 1990 fire tests.
"As-installed" 1-hour and 3-hour tests failed.
Additional 1-hour and 3-hour tests of
" upgraded" configurations passed (according to GSU criteria) except for 3-hour cable tray.
April 26, 1991 Irternal RBS memo provided information necessary to evaluate methods for installing fire barriers.
May 15, 1991 LER 91-008 submitted to NRC.
Fire wrap was not installed on electrical cables per the FHA since plant startup.
Found in response to review of FHA by contractor following up January 1990 NRC inspection.
Corrective action not yet determined.
References NRC Inspection Report 50-458/90-02.
May 24, 1991 SWRI letter'to GSU transmitted draft final report 01-2702-for october 26, 1989 3-hour fire test to GSU for comment.
June 11, 1991 NRC memorandum documenting NRC visit to RBS to review the circumstances surrounding LER 87-005 and LER 90-003.
June 26, 1991 GSU letter to SwRI provided comments on SwRI Report 01-2702.
June 28, 1991 LER 90-003 Rev. 3 submitted to NRC.
Acceptable upgrade for 3-hour cable tray protection has not been found.
Several alternatives being explored by GSU.
Fire watches remain in place.
Committed to supplemental response by' December 31, 1991.
August 6, 1991 Information Notice 91-47, " Failure of Thermo-Lag Fire Barrier Material to Pass Fire Endurance Test," issued.
9
i T
i ATTACHMENT 10 I
COMANCHE PEAR HISTORY AND DOCUMENTATION REVIEWED BY SPECIAL REVIEW TEAM (50-445)
September 17, 1981 One hour fire endurance test conducted at SWRI.
Test assembly built by TSI.
ANI i
standard used.
18" cable trays.
Passed i
based on circuit integrity.
Highest cable surface temperature 260*F.
Air drop reached 554*F.
October 27, 1981 Test report SwRI Project No.
03-6491,
" Fire Qualification Test of a Protective Envelope System."
October 29, 1981 Copy of SwRI test report hand delivered to NRC.
November 16, 1981 Memo from Johnston to Tedesco approving SwRI test report.
November 18, 1981 Letter from TU (Schmidt) to NRC (Tedesco) confirming that report was hand delivered on October 29, 1981, and requesting NRC to review the test report.
December 1, 1981 Letter from NRC (Tedesco) to TU (Gary) stating that on October 29, 1981, a copy of the test report was received by NRC.
The test report was evaluated and approved.
December 7, 1981 Test report SwRI Project No.
03-6491 docketed in letter to NRC.
October 4, 1983 TU letter to NRC regarding need for protection of cable tra.y supports.
October 19-23, 1987 Fire protection program inspection conducted (87-22).
September 6, 1988 Technical Evaluation Report for CPSES fire protection program (SAIC).
September 15, 1989 TU verbally notified NRC concerning 50.55e on Thermo-Lag.
October 12, 1989 TU letter to NRC issuing 10CFR50.55e on site fabricated Thermo-Lag panels.
Site fabricated Thermo-Lag panels were found with localized dry film thickness of less than 1/2 inch.
Also seams and joints were found less than 1/2 inch.
3-October 26, 1989 TSI letter to TU stating that localized thickness variations to 3/8 inch are acceptable for less than 2% of the panel.
November 3, 1989 TSI letter to TU stating that localized thickness variations are acceptable for conduits.
December 1, 1989 Resident _ Inspection Report 89-71 which discusses Thermo-Lag thickness issue.
January 3-30, 1990 NRC Inspection 90-05 conducted.
Reviewed Thermo-Lag allegations associated with receipt inspections.
May 17, 1990 Civil Penalty ($25,000) issued concerning Thermo-Lag.
June 1, 1990 TU response to Civil Penalty.
July 13, 1990 TU letter to Region IV discussing acceptance criteria for Thermo-Lag panels.
July 20, 1990 Region IV letter to TU acknowledging July 13, 1990, letter.
- ' i.
ATTACHMENT 11 e
WNP2 HISTORY AND DOCUMENTATION REVIEWED BY SPECIAL REVIEW TEAM (50-397)
December 28, 1981 Letter from WNP2 to NRC summarizing telecon of December 11, 1981 between NRC (Harrison),
WNP2, Bechtel, Burns & Roe on 3-br barriers.
NRC stated that a barrier w/E-119 rating and.
ANI acceptance would be acceptable to NRC.
May 17-28,1982 1-hr fire endurance test conducted at TSI.
ITL report no. 82-5-355A.
Temperature reached
- 325F, however TSI indicated acceptance of barrier because no loss _of circuit integrity occurred.
June 18-25, 1982 3-hr fire endurance test conducted at TSI.
ITL report no. 82-5-3558.
All configurations passed circuit integrity and temperature.
August 2, 1982 WNP2 letter concerning fire endurance testing.
September 20, 1982 WNP2 submitted ITL report no. 82-5-355A for NRC review.
No approval was formally transmitted to WNP2.
September 30, 1982 Telecon between WNP2 and NRC on test report data.
October 4, 1982 TSI letter to NRR (Kubicki) responding to questions on tests conducted at TSI.
October 4, 1982 WNP2 response to NRC questions (Kubicki) of September 30, 1982.
October 5, 1982 WNP2 submitted ITL report no. 82-5-355B for NRC review.
November 15, 1982 Additional copies of 355B transmitted to NRC.
December 27, 1982 Draft SER on fire protection.
April 26, 1983 Revision II to TSI Engineering Test Plan for direst spray on cable-tray.
May 1983 SSER #3 issued for WNP2.
Page 9-3 approved the 3-hr TSI Thermo-lag as fire rated barrier.
States that material is capable of limiting temperature rise to not more than 250'F above ambient.
June 7, 1983 ITL Test 83-5-472 conducted for 1-hr direct application method.
9
~ 7 January 17, 1984 Draft SSER on fire protection.
March 3-7, 1986 Fire protection inspection (86-05).
Unresolved items identified concerning protection of cable tray supports,. intervening combustibles not covered, and unwrapped cable trays.
April 15, 1986 Enforcement Conference on fire protection.
May 9, 1986 Letter to RV from WNP2 responding to enforcement conference on Thermo-lag installation on supports.
States problem was judgemental error.
July 29, 1986 WNP2 letter revising committment to correct cable tray support protection to be completed in August 1986.
August 19, 1986 3-hr fire endurance test of 3M/TSI interface conducted at 3M.
TSI failed 325'F.
Installed by certified installers.
September 23, 1986 Revision I to TSI Engineering Test Plan for 3-hr "In-situ" test of extrusion procedure.
October 8, 1986 WNP2 determined that cable derating calculations for service water did not consider routing in duct banks (LER 86-003).
Further review identified Thermo-lag not taken into account.
31.5% derating used.
No cables replaced.
October 20, 1986 3M letter to WPPS enclosing test report for 3M/TSI interface test.
January 12-15, 1987 Fire Protection Inspection (87-02).
Ampacity reviewed and closed.
Violation concerning application method of Thermo-lag in untested configurations.
March 25, 1987 Revision II to TSI Engineering Test Plan for 3-hr "in-situ" test of extrusion procedure.
Revision I to 1-hr test plan.
March 31, 1987 3-hr fire endurance test of 2"
conduit with Thermo-Lag applied by extrusion.
Conduit failed temperature and circuit continuity, j
April 1, 1987 1-hr fire endurance test'of 4"
conduit with Thermo-Lag applied by extrusion.
April 13, 1987 WNP2 submittal of Fire Protection Re-evaluation.
Discusses Thermo-lag fire barrier i
5-ATTACHMENT 12 SUSOUEHANNA HISTORY AND DOCUMENTATION REVIEWED BY SPECIAL REVIEW TEAM (50-387)
March 25, 1981 Meeting of PP&L discussing fire protection issues.
March 26, 1981 SER input, with several open items, including the battery room.
March 26, 1981 Licensee letter committing to install 2-hr fire rated ceiling and 2-hr enclosure in battery room. (PLA-683, not found in NUDOCS)
April 15, 1981 SSER with only open item alternate shutdown system.
June 19, 1981 Licensee informed NRC of testing of Quelcor cable wrap in July 1981.
October 26, 1981 Licensee letter submitting two test reports for cable wrap material.
Includes 4/81 and 8/81 TSI test reports.
February 9, 1982 Licensee letter with proposed modifications to the fire protection program including a 1-hr rated cable wrap for battery room and remote shutdown panel with only manual suppression and detection.
Letter referenced previous NRC acceptance of Comanche Peak barrier in December 1981.
March 11, 1982 Meeting with PP&L discussing 2/9/82 submittal'.
NRC requested a copy of the TSI test results, as installed, and installation procedure.
April 16, 1982 SSER accepting 2/9/82 proposal, but requiring an as installed test or automatic extinguishing-system.
"The-proposed cable wrap was previously accepted based on independent laboratory test data.
. The applicant verbally stated that the manufacturer's recommended installation procedure had not been followed when the cable wrap was installed.
We do not accept the material as a 1-hr-fire barrier when installed differently."
May 12, 1982 Licensee provided TSI Technical Reports 8232-1 and 8275-1 (June 1981) and Bechtel installation procedures.
... ~ ~-
~7 June 9, 1982 Licensee provided TSI letter dated June 3,
1982, in. response to NRC. questions on tests submitted on May 12, 1982.
TSI report l
" revised" to correct some noted errors.
June 28, 1982 SSER stating that tests submitted by PP&L were not performed in accordance with adequate QA procedures and therefore not acceptable.. The application procedure was also not clearly l
specified.
Problems. were noted with the accuracy of the furnace temperature control and thermocouples. The licensee should conduct an ASTM E-119 test att "an approved testing laboratory,"
or install automatic fire extinguishing system.
July 6, 1982 Licensee commits to perform a 1-hr test of TSI material using
.the Comanche Peak-test procedure.
Material to be installed per PP&L procedures submitted on May 12, 1982.
July 23, 1982 Licensee submits copy of Test Specification August 25, 1982 Licensee submits test report from SWRI.
Test conducted on August 10, 1982.
August 31, 1982 Licensee commits to modify the installation of cable wrap on cable trays by adding. fiberglass armoring in accordance with TSI Technical Note 80181, Revision I, dated 8/81.
September 15, 1982 SSER.
approving the.
1-hr fire barrier installation.
August 1982 test was only accepted for conduit.
The cable trays exceeded 325F ' criteria, which was applied.
The approved method of installation is based on CP submittal of October 27, 1981, and is referenced.
SUMMARY
PP&L originally committed to install a 2-hr rated ceiling l
and enclosure in the battery room, but later changed to a
1-hr cable. wrap.
They considered two materials (Thermo-lag.and Quelcor), and submitted information on both to the NRC, and also. referenced' NRC's previous acceptance of Thermo-lag at Comanche Peak.
NRC reviewed' the TSI test reports submitted by PP&L.- (6/81 and 8/81),
but found the test reports not acceptable, and also concluded the--
Susquehanna. "as installed" was not represented.in the test, reports.
PP&L then conducted an as-installed test in August 1982 at Southwest =Research.
The cable tray test f ailed the-250F criteria, and the NRC l
required PP&L to modify their installation to that previously accepted in the Comanche Peak: October 1981 test installation procedure.
_.,..~.
-e
4.
Summary of Generic Safety Issues 5.
Specific Technical and Programmatic Issues Identified by the Special Review Team 6.
List of Fire Endurance Test Reports Involving the Thermo-Lag 330-1 Fire Barrier Systems Known to the NRC 7.
List of Ampacity Derating Test Reports Involving the Thermo-Lag 330-1 Fire Barrier Systems Known to-the NRC 8.
Chronology of Review Team Activities 9.
River Bend History and Documentation Reviewed by Special Review Team 10.
Comanche Peak Steam Electric Station History and Documentation Reviewed by Special Review Team 11.
Washington Nuclear Project, Unit 2 History and Documentation Reviewed by Special Review Team 12.
Susquehanna Steam Electric Station History and Documentation Reviewed by Special Review Team 1
1 50 i
ST. LOUIS POST-DISPATCH 4/l6 ini nNNmVANIA ANNE, N.W.19 fit M8 3AMRK2p 2Wpst Bill Lambrecht Washington Correspondent TREEDOM 0F INFORMATION Donnie H. Grimsley ACT REQUEST t
Freedom of Information Office
{C N Mb Nuclear Regulatory Commission M {cy'.4.2/-S,3 t
Dear Mr. Grimsley; I write to make a Freedom of Information request on behalf of the St. Louis Post-Dispatch. We have done articles about Thermal Science Inc., a local company, and it's product Therma-Lag, and we are requesting these documents as part of our research. We also have an ongoing interest in two nuclear plants in our circulation area:
Callaway and Clinton.
-We.would like to receive copies of f'mdings of violations from inspection repons-at Callaway and Clinton in 1993; l
-We would like copics of correspondence, memoranda, reports or - any l
information related to Thermal Science's foreign clients, i.e. nuclear plants
-l L
outside the U.S. to whom the company has sold Therma-Lag;
-We would like to obtain copies of correspondence to or from F.J.
Miraglia since Jan. 1990 about Thennal Science or Therma-Lag;
-We would like to get copies of correspondence or memoranda to or from David P. Notley since -Jan. 1990 about Thermal Science or Therma-Lag;
-We would like to receive a copy of a report done by David Taylor Research in the early 1980s related to the use of Therma Lag by the U.S. Navy;
-We are interested in receiving copics of correspondence or i
memoranda to or from Conrad E. McCracken of the Plant. Systems Branch since July 1992 related to test and acceptance critera for fire endurance testing:
-We would like to receive a letter dated February 10,1993, with -
enclosures from Chairman Ivan Selin to Rep. John Dingell related to changes in existing fire barrier testing; If there is anything we can do as far as streamlining our requests or making-it-easier to obtain documents, please telephone-
- j me at our Washington office.
q We' would ask for immediate attention to our request and we.
thank you for your consideration.
Sincerely, i
kb 9H olok(o %T
'