ML20032C357
| ML20032C357 | |
| Person / Time | |
|---|---|
| Site: | Midland |
| Issue date: | 10/30/1981 |
| From: | Jackie Cook CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| References | |
| 10CFR-050.55E, 10CFR-50.55E, 14598, 79-12-#10, NUDOCS 8111090709 | |
| Download: ML20032C357 (8) | |
Text
..
CORSumBIS POW 8f
,,,,,,, m 0
Vice President - Projects, Engineering and Construction c.n.res offices: 1945 West Parnell Road, Jackson, MI 49201 e (517) 7880453 79-12 #10 October 30, 1981 ff)
Mr J G Keppler, Regional Director r8
((
I Clfice of Inspection & Enforcement C
gLU g.
y 7,S Nuclear Regulatory Co=iission li NOVO O /
D
]$,g g %
ggg C legion III
)
'99 Roosevelt Road g
7 G en Ellyn, IL 60137 ff, I ^n s
MIDLMID PROJECT -
'b / '
p DOCKUT NOS 50-329 AND 50-330 CONTAINMENT INTERNAL STRUCTURES COATING DEFICIENCY FILE:
0.h.9 37 SERIAL:
1h598
Reference:
CPCo letters to J G Keppler; Same
Subject:
- 1) Serial Hove-309-79; dated December 13, 1979
- 2) Serial Howe-26-80; dated February 7,1980
- 3) Serial Howe-TS-80; dated April 15, 1980
- 4) Serial 9125; dated June 11, 1980
- 5) Serial 8823; dated September 9,1980
- 6) Serial 10051; dated October 31, 1980
- 7) Serial 11199; dated February 23, 1981
- 8) Serial 12009; dated May L9, 1981
- 9) Serial 12068; dated July 31, 1981 The referenced letters were interim 50 55(e) reports on the containment internal M uctures coating deficiency. This letter in the final report l
on this sub ect. The attached MCAR-35 provides a summary of the actions J
l which have been taken to correct this deficiency and prevent its recurrence.
WRB/1r
Attachment:
MCAR-35, Revised Final Report, " Containment Internal Concrete Structures Coating," dated October 21, 1981 l
l CC: Director, Office of Inspection & Enforcement l
Att Mr Victor Stello, USNRC (15) l Director, Office of Management l
Information & Program Control, USNRC (1) l 8111090709 811030 S
PDR 80V 2 1981 i
PDR ADOCK 05000329
-_f, h-)74 L
2 Serial 1h598 79-12 #10 CC: RJCook, NRC Resident Inspector Midland Nuclear Plant (1)
CBechhoefer,-ASLB Panel RSDecker, ASLB Panel FPCowan, ASLB Panel AS&L Appeal Panel
!WCherry, Esq MSinclair BStamiris CRStephens, USIGC WDPaton, Esq, US'EC FJKelley, Esq, Attorney General SHFreeman, Esq, Asst Att'orney General WE4arshall GJMerritt, Esq, TNK&J
Attachment to Serial 14598 Bechtel Associates Professional Corporation 79-12 010 04620i MCAR 35 0 4 614 EUBJECT:
Containment Internal Concrete Structures Coating REVISED FINAL REPORT DATE:
October 21, 1981 PROJECT:
Consumers Power Company Midland Plant Units 1 and 2 Bechtel Job 7220 Introduction This is a summary of investigations and findings concerning MCAR 35.
It includes analysis of information from reports prepared by independent laboratories. A description of the corrective action for repair of the deficiency is included, and steps taken to prevent recurrence are discussed.
Description of Deficiency Coating System 9 is a decontaminable coating system applied to concrete walls in containments. As applied, System 9 consists of two coats of Ameron NuKlad 117 and two finish coats of Ameron Amercoat 90. NuKlad 117 is a 100% solids epoxy coating used to fill and seal the concrete surface.
Amercoat 90 is a solvent base phenolic epoxy coating used to provide a decontaminable finish. Delamination has occurred between two coats of NuKlad 117. The failure pattern is random.
Investigation Investigation of the delamination has included the following items.
1.
Evaluation of the Extent of Coating Failure A knife cst adhesion test was developed to measure delamination on a preselected grid pattern. One or more tests were made on sample grids measuring approximately 100 square feet. Special sketches of these grids were prepared to permit mapping of Units 1 and 2 containment coating failure. On this initial test, approximately 6% of the coating in Unit I and 18% in Unit 2 failed. The failed grids were eliminated from further testing and the delaminated coating has been removed.
2.
Review or Documentation Specification 7220-A-15(Q) requires daily documentation according to the applicant's QA program. All variable information was compared with the maps described above for pass / fail areas. Deviations which became apparent from this review are as follows.
Bechtel Associates Professional Corporation MCAR 35 6l$@
Revised Final Report 04620,i October 21, 1981 Page 2 a.
Two full coats of NuKlad 117 were applied in most cases, with isolated areas having three coats.
b.
The time between applying coats of NuKlad 117 exceeded the manufacturer's limits in several areas.
c.
In some areas, Nef. lad 117 was applied over Amercoat 90.
d.
Ameron product information was revised to permit two-coat application. The revision was not submitted for review as required by the applicant's QA program.
e.
The applicant failed to record relative humidity in most cases, although the approved QA program indicates that this requires daily documentation.
3.
Determination of Root Cause of Delamination Investigation of principal factors which may cause delamination has been performed under laboratory programs by KTA Associates, IMCO Laboratories. Consumers Power Company laboratories and Bechtel have also conducted some testing. The following potential causes were evaluated extensively:
a.
Time interval between applying coats of NuKlad 117 b.
Intercoat contamination c.
Batch and mixing variation and packaging d.
Solvent additions e.
Variable time between mixing and application f.
Relative humidity and temperature during cure 4.
Root Cause of Delamination Delamination occurs only where two coats of NuKlad 117 are applied.
The following factors can cause delamination.
a.
At 80% relative humidity (or higher), delamination occurs at 75F surface and air temperature. This appears to be due to exudation of the convertor. The delaminated interface is highly alkaline.
i Bechtel Associates Professional Corporation.
-l
> 6148
" CAR 35 04620i Revised Final Report October 21, 1981
- Page 3 t
b.
Application of.NuKlad 117 near or after the end of the effective pot-life causes delamination in the following ways.'
1.
Material is warm from exothermic heat. This maintains fluidity in the mix.
2.
'When applied to a-cool surface, the interface material thickens or solidifies and does not properly wet the first coat of NuKlad 117.
The following factors appear to increase the probability of delamination.
a.
Addition of Amercoat 6 thinner (not authorized) produces immediate (exothermic) heating - probably due to a reaction with the amine convertor. This abnormally extends the apparent pot life and changes the bonding properties of the mix.
b.
Adhesion between coats decreases as cure time between coats increases.
c.
A variation of application and cure conditions changes the working time, cure time, and the influence of the other factors defined above. For' example, at 70% relative humidity or less, delamination may occur if the temperature during cure is 60 to 65F.
The following factors do not appear to be significant in the.delanination between two coats of NuKlad 117:
a.
Contamination between coats b.
Batch dif ferences or weight discrepancies due to component packaging errors noted c.
High film thickness influences the ease of delamination, but is not the cause Corrective Ac tions 1.
Selection of Materials and Procedures for Corrective Action During the course of this investigation it became apparent that all coating residues could not be removed. It was necessary to develop w-s w--,+--m-e,..r,m r,
-t r e vs
---w -w * +
---nim--e~-
-v-w.-r-y e-
Bechtel Associates Professional Corporation -
-4 7f k b 04620l-nCAR 35 Revised Final Report October 21, 1981 Page 4
- a repair system which would adhere to the residue of NuKlad 117 and the cementitious surface. Specification 7220-A-57(Q) was prepared
-to control the application of coating systems for_ test-and the collection of core samples. Design accident tests at Oak-Ridge National Labs performed in accordance with Specification 7220-A-45 (ANSI N101.2 criteria) verified'that repairs could be made with any one of several coating. systems from four coating suppliers.
2.
Requalification of the Applied System 9 Coatings An adhesion test procedure has been-developed. This test is performed i
by cutting-a 1/4-inch wide strip in the coating system and attemping to remove it with a knife. Core samples which passed the knife test have been tested at Oak Ridge National' Laboratories and hav.
met DBA and irradiation requirements of ANSI N101.2. Specification 7220-A-55(Q) was used to control sample preparation.
1 6
Where delamination has been found using the knife test, the coating has been removed. - The remaining coated surface has been divided into test areas which are then subdivided into sampling locations or grids. Each area will be tested with the knife test on a statistically significant number of grids selected at random. If the allowed 4
number of failed grids within an area is exceeded, the entire area j
is deemed unacceptable. Portions of the area.may be retested using a revised sampling plan or the entire area will be repaired.~
Acceptable areas will be considered qualified and will be allowed j
to remain.
l 3.
Procedure Qualifications for coat.ing System 9 Repair To establish and qualify repair procedures, test coating applications vere made over surfaces where the delaminated coating had been a
removed. Samples of the test coating were submitted to Oak Ridge National Laboratories for irradiation and DBA testing. Test results showed that tested repair coatings are qualified in accordance with
!~
ANSI N101.2. This work was completed under Specification 7220-A-57(Q). Procedures for System 9 ccating repair are included in j
Specification 7220-A-56(Q) and are based upon procedures established by the test application program. Repairs are underway, and will be completed by September 1, 1982.
i l
Prevention of Recurrence j
1.
Elimination of Multiple Coats of NuKlad 117 It has been determined that NuKlad 117 is too sensitive to mixing and application conditions to be used as a two-coat system.
L i
Bechtel Associates Professional Corporation MCAR 35 04620:-
RevisedFinc.1Reportg4g,4g October 21, 1981 Page 5 Amercoat 90 applied as a " tie coat" will be utl'Ized in any work requiring repair or overlap of NuKlad 117 over a previous coat.
This system will be used for new work or repairs of the delaminated area.
2.
Control of Application Conditions Although relative humidity control is not recommended by Ameron for surfacers, a maximum of 80% relative humidity has been specified for all surfacer systems in Specification 7220-A-56(Q). Ia addition, the time allowed between coats has been specified on a more conservative basis than permitted by Ameron.
3.
Documentation Control Section 13.0, Inspection and Testing, Technical Specification 7220-A-56(Q) requ'res that daily inspection reports and tests be documented in accordance with Section 2.0, Quality Requirements. This program is more extensive than the previous subcontract requirement and has been prepared to eliminate the causes for delamination described in Section 4.4.
- 'W Submitted by:
_M.A.
Hughes Architectural Group Supervisor Approved by:
Projedt En'g'ineer i
f Concurrence by:
Bngineeri g Manager
y 1.
- ^
Bechtel Associates _ Professional Corporation
'MCAR 35 04b f k 0-04620I Revised Final Report-October 21, 1981 Page 6 REFERENCES Specifications 1.
-Technical Specification for Subcontract for Epoxy or Phenolic' Decontaminative-Surfacer, 7220-A-15(Q), Rev 8,-2/23/80 4
2.
General Requirements for DBA Environmental Testing,'7220-A-45, Rev 4, 5/1/81 l-3.
Technical Specification for Adhesion Testing, Collection, and Preparation of Concrete Core Samples for Design Basis Accident Testing of Coatings, 7220-A-55(Q), Fev 1, 4/24/81 Technical Specification for Application.of Decontaminable Coatings 4.
to Concrete, 7220-A-56(Q), Rev 0, 4/14/81 5.
Technical Specification for Application of Coatings to Concrete for j,
Evaluation and Design Basis Accident Testing, 7220-A-57(Q), Rev 1, 4/30/81 Reports Oak Ridge-National Labs 1.
January 21, 1981 2.
March 4, 1981 3.
May 21, 1981 KTA-Tator Associates, Inc.
1.
Final Report on Test Panel Preparation and Adhesion Testing of NuKlad 117 Applied Over Steel snd Concrete Substrates, Purchase Order 7220-A-48, 7/23/80 2.
Final Report of NuKlad 117 Failure Investigation, Midland Nuclear Station, Midland, Michigan, Purchase Order 7220-A-48, 8/12/81 IMCO Laboratories, Inc.
1.
Project 164 Report, 5/28/81 2.
Project 164 Supplemental Report 1, 7/8/81