ML20093K308
| ML20093K308 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 10/31/1984 |
| From: | OGDEN ENVIRONMENTAL & ENERGY SERVICES (FORMERLY MULTI |
| To: | |
| Shared Package | |
| ML20093K307 | List: |
| References | |
| DECO-12-2191, DECO-12-2191-R03, DECO-12-2191-R3, NUDOCS 8410170273 | |
| Download: ML20093K308 (34) | |
Text
. .,
" REPORT NO. DECO-12-2191 um REVISION 3 OCTOBER 1984 ENRICO FERMI ATOMIC POWER PLANT UNIT NO. 2 EVALUATION OF CONTAINMENT COATINGS eggogggg;aggg8 g MULTIPLE DYNAMICS CORPORATION A
., ,_ g
l i
THE DETROIT EDISON COMPANY FERMI 2 NUCLEAR POWER PLANT EVALUATION OF CONTAINMENT COATINGS FOR FERMI 2 REPORT NO. DECO-12-2191 REVISION 3 PREPARED BY MULTIPLE DYNAMICS CORPORATION 29200 SOUTHFIELD, SUITE 103 SOUTHFIELD, MICHIGAN 48076 (313) 557-7766 OCTOBER 1984
MULTIPLE DYNAMICS CORPORATION SUIAJECT: DATE ISSUED DOC. NO. ,
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE i TABLE OF CONTENTS PAGE 1.0 Introduction 1 2.0 Containment coatings 3 2 .1 Drywell Interior 3 2.2 Concrete Surfaces 3 2.3 Structural Steel 4 2.4 Suppression Chamber 5 3.0 Equipment Coatings 6 3 .1 Galvanized Surfaces 6 3.2 Hangers and Supports 6 3.3 Piping 6 3.4 Unqualified Coatings 7 4.0 Coating Survey 7 5.0 Coating Qualifications 9 6.0 Failure Modes of Unqualified Coatings 11 7.0 Safety Evaluation of Coatings 14 7 .1 Corrosion Protection 15 7.2 Decontamination 16 7.3 Hydrogen Evolution 16 8.0 Effects of Coating Debris 17 8.1 Debris Transport 17 8.2 ECCS Performance 20 8.3 Containment Sprays 21 8.4 RPV Core Spray and Feedwater Spargers 21 8.5 Normal Operations 22 9.0 Additional Considerations 23 9 .1 Surveillance and Inspection 23 9.2 Additional New Coatings 23 10.0 Conclusion 24 11.0 References 25 12.0 Addendum (Responses to NRC Questions) 26 3
m er MULTIPLE DYNAMIC 3 CERPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE i
1.0 INTRODUCTION
An evaluation of the Fermi 2 primary containment coatings was performed in response to comments expressed by the Nuclear Regulatory Commission and Duke Power Company as part of its Construction Assessment Inspection at Fermi 2 in June 1984. The comments raised primarily pertained to the repair and touch-up of damaged coatings, the amount of unqualified coatings and assessment of uncoated surfaces. It was further implied that failure of certain coatings could lead to degraded plant performance in accident conditions.
In response to these comments, Fermi 2 Engineering initiated a review of primary containment coating types and qualifications. This review encompassed the following subjects:
. An analysis of the specific coating substances used f or containment surf aces and components, and the rationale for their use
. Original design and current regulatory standards for application and testing of coatings
. Definition of qualified, safety-related coatings
. Proper application of non-qualified coatings
. Confirmation of test results
MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 2
. Detailed as-built survey and inspection of primary containment coatings and determination of coating parameters and quantities
. Analysis of failure modes of ungualified coatings and evaluation of safety concerns
. Transport of coating debris and potential effects on:
plant systems performance
. Additional test, surveillance and inspection programs to be implemented by Detroit Edison
. Application of new coatings before and after fuel load
. Response to additional NRC questions (see Addendum) 3 This evaluation determined that the applied containment coatings do not in any way degrade or affect the safe operation of the plant under normal or accident conditions.
The Fermi 2 PSAR and the coating specification will be appropriately revised to reflect the conclusions and as-built conditions discussed herein.
k i
MULTIPLE DYNAMICS CORPORATION
- l
SUBJECT:
DATEISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 3 4 2.0 CONTAINMENT COATINGS 2 .1 Drywell Interior The original contianment specification issued in 1969 included a requirement for coating of the interior ,
pressure boundary surfaces in the drywell, with the ,
primary objective being long-term corrosion protection.
The industry standard at that time was to apply Carbo-Zinc 11 as manufactured by the Carboline Company.
in accordance with the manufacturer's recommendations.
This type of coating has been successfully applied to most operating BWR's, and has stood up well over the years, even under a variety of adverse conditions.
Most of the CZ-11 coating was originally applied to the , \
Fermi 2 surfaces before the issuance of Regulatory i Guide 1.54 and ANSI N101.4. However, the Fermi 2 QA i Level 1 criteria was applied in the absence of definitive coating criteria.
2.2 poncrete Surfaces Following erection, pressure testing and coating of the containment drywell, the drywell floor and RPV pedestal were poured to accept the reactor vessel. Erection of the sacrificial shield and drywell primary steel structures followed. The concrete surfaces of the il ,
drywell floor, and the exterior and interior of the pedestal, were coated with an Ameron Nu-cl'ad surfacer '
110AA and a finish coat of Ameron polyamide epoxy 66.
This coating was applied in accordance with ANSI N101.4, and met the pull test requirements (200 psi)
y 7
j MULTIPLE DYNAMIC $' CORPORATION
SUBJECT:
, DATE ISSUED DOO. NO.
EVALUATION.OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERM1' L n REVISION: 3 PAGE C
, per ANSI 5.12, Section 6.2. Required DBA testing was c 's , g1
)
a i conducted for these coating meterials. T.(1,c, prime , .
objective of the concrete coadings is to effectivsly<
seal the porous surfaces, to inhibit intrusion of 'c -
radioactive contaminants and to facilitate easy /,
washdown and decontamination if required.
2.3 Structural Steel , .
2 The primary structural stecl within;Nie drywell and the exterior steel surface of the sacrificial shield ver_e ,
coated with Carbo-Zine 11. Surface preparation of these surfaces' included blasting or hand power tooling to near-white metal. The purpose.of the C2-11 coating
[
is to provide long-term protection against'exc.essive '
(
corrosion and rusting, and to facilitate e$sy, 3 decontamination if require 61 / 4 s ,
As a result of the new Mark 1 Program containment LOCA loads, new Safety,,-
Relief Valve discharge
- loads, and'a' .
general containment steel load reevaluation, substan-tial modifications were required to be implemented in ,
two different phases. The two different timp periods
~
resulted in varying degrees of steel surfaEe'prepara-tions. Approximately 250 tons of structural steel were ;
added by means of welding to the existing structures.
Welding and NDE operations required removal of.the -f existing CZ-11 coating at the tie-in and welded connec-I tions. Due to the increased . complexity of component placementandshrinkingvorbspace,aswellas E ,
completed installation o2>much of the mechanical 1
I i MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATEISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 5 eauipment, sandblasting and painting became more and more impractical and time concuming, affecting construction and preoperational testing progress.
Therefore, coating of new structural steel and recoating of modified steel components was not routinely completed. The mill scale present on the uncoated and unblasted surfaces of the steel members is discussed in the debris evaluation within Section 6.0 of this report.
2.4 Suppression Chamber The entire interior of the suppression chamber, includ-ing the vent system, ring girders, structures, mono-rail, piping and supports, is coated with Plasite 7155 manufactured by Wisconsin Protective Coating Company.
The coating system is considered safety-related, QA Level 1, and was applied in accordance with Regulatory Guide 1.54 and ANSI N101.4. Some small areas (on the order of 1 inch 2 ) which were subject to mechanical damage, newly-installed vacuum breaker flanges, and two RHR orifice flanges require repair and touch-up. In bays 15 and 16, the Plasite 7155 has been removed in 1 x2 inch spots to facilitate installation of test instrumentation and strain gauges for the scheduled SRV test. These areas are to be repaired during the first refueling outage, whan the instrumentation is removed.
MULTIPLE DYNAMICS CCRPORATION =[
+
3UBJECT: DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 a REVISION: 3 PAGE 6
i 7
3.0 EQUIPMENT COATI _NGS 3 .1 Galvanized Surfaces h The drywell cooling system ducting and dampers are completely galvanized without any further coatings. At _
welded joints, the galvanized surface was ground off to _
clean metal, and in some locations these ground are>- [
were touched up with Galvanox V, a zinc-rich coating similar in properties to CZ-11. In addition, all electrical conduit, terminal boxes, cable trays and supporting unistruts are galvanized. The only exceptions are some large flexible conduits made of :
stainless steel. J 3.2 Hangers and Supports -
Hanger and support components, including clamps, rods, E spring cans, snubber attachments, pipe whip restraint ;_
components and secondary support steel, were originally I coated with CZ-11. Significant changes in the hanger and support design resulted in the addition of secondary support steel and change-out of hanger ;
components and welding of attachments. Coating repair and touch up of these areas is scheduled to be y accomplished as time permits, to facilitate ?
decontamination and provide long term corrosion protection. .
3.3 Piping Most of the piping within the drywell is insulated with reflective metallic insulation panels (Mirror _
i l _ _. _ _ . . .
h MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATE ISSUED DOC. NO.
=
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 7 Insulation), consisting of removable sections having an outer cover of stainless steel. No fibrous insulation has been used within the containment. Normally cold fluid system piping is not insulated or coated. Th'e uninsulated carbon steel piping was shop coated with a protective varnish. Tight mill scale and some rust is apparent on the piping surfaces. The varnish and mill scale are considered unaualified coatings for the purpose of this evaluation and per Standard Review Plan 6.1.2. The design of the piping pressure boundary :
included a corrosion allowance of 0.125".
3.4 Miscellaneous Coatinos As part of a detailed coating survey.within the containment, miscellaneous unidentified coatings were surveyed. These coatings consist largely of manufacturers' shop coatings and primers such as red -
lead, aluminum base, enamels, polymer and phenolic paints and yellow safety paint. These coatings are present on valve bodies, yokes and bonnets, motor and air operators, handwheels, handrails, electric motors, etc. Another category consists of identification marking and banding of electrical conduit, terminal boxes and trays. Only very small cuantities of these coatings are present as shown on Table 1.
4.0 COATING SURVEY A detailed coating survey was conducted of all surfaces within the primary containment, to assess type of coating, surface areas, and dry film thickness (DFT). ,
l (P
C TABLE 1 E nm o
SUMMARY
OF PRIMARY CONTAINMENT COATINGS Oh U ee MC z>
Q8 Approx. mw Total Dry Film Total Total y@
Density Volume O Type of Qual. Average DFT* Surface Mass Coating Coatings (mils) (ft2) (#/ft3) (ft3) (lbs) *O m E
. wn C F
Carbo-Zinc 11 No 7 125,000 217 73 15,84i i g@
Me Plasite 7155 Yes 12 67,000 150 66.9 10,035 w3z
- r D Ameron 66 and Yes 1/16" plus 7,380 125 44.6 5,575 m .c Surfacer 10 mills $ 2 Galvanox V No 5 775 202 0.36 73 I g
n Mill Scale No 3.4 89,000 350 25.22 8,827 n and Varnish a m
o y
o
< 3
-< 3 Unqualified No 0.7 to 3 1,782** 90 to 150 0.48 72 @ m o g
Paints O o en 2 2
re ta
. c >
q m
O O
- 2
- DFT (Dry Film Thickness) measurements were taken with a Positector 2000 w gauge, Serial #30531, calibrated with NBS shims #22272.
- For details see Item 5 of Addendum. $ 8 o
m e o m -
n E I
w OOg s
i M
I i 1 N
.,,,e
~
l l
W
. . ., J -
MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 9 This data was used to calculate total quantities of ,
qualified and ungualified coatings. The results of this survey are summarized in Table 1.
5.0 COATING OUALIFICATIONS ~3 The Plasite 7155 and Ameron 66 coatings have been j, applied in full compliance with the provisions of Reculatory Guide 1.54 and ANSI N101.4 (1972). Pull . -
tests have been performed in accordance with ANSI 5.12, Section 6.2 on the original coatings and the repaired areas. DBA testing was performed by the paint manufacturers as required under ANSI N5.9 and ANSI N101.2. These coatings are therefore considered fully cualified per the regulatory criteria of Standard Review Plan 6.1.2.
The Carbo-Zinc 11 coatings applied to the containment _
surfaces have also been subjected to extensive DBA testing for a variety of application techniques, and were found to be acceptable for use in BWR environments under LOCA conditions. The test results are contained
- in Report No. 56878 issued by the Carboline Company (Reference 1).
As discussed earlier, most of the CZ-11 on the -
containment pressure boundary was applied prior to the issuance of R.G.1.54 and ANSI N101.4. Even though a specific effort was made to document the CZ-11 application in accordance with the Fermi 2 OA Manual, a current inspection of the existing documentation
MULTIPLE DYNAMICS CORPORATION SUSJECT: DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 10 reveals deficiencies when gauged against current criteria. These deficiencies in documentation, however, do not adversely affect the primary function of corrosion protection. The coatings have already successfully withstood more than 10 years of _
construction environment, wear and tear, and have undergone a substantial degree of thermal cycling, from direct summer sun exposure to near freezing winter conditions, without deleterious effects. For the p,f purpose of this assessment, however, the CZ-11 coatings ,... p/.1.; l.
- y. 1 are evaluated as unqualified coatings. ..o p ...
- -Qf
- .;
. e.1-. ,
Those containment surface areas which are severely ' t ; p. .:. ?
damaged or worn (visible bare metal) are repaired by OP;l;'
. 1.:
touch up, to ensure the continued long-term function of M7;;4;..
c...
corrosion protection for the pressure boundary and te a8
-E.
- y...
facilitate potential decontamination. The conseguences 3.3 ;
. c, 9.
- of failures are evaluated in Section 8.0 of this f9 j" report. These coatings are maintained under the Fermi Ihj 2 QA Level 1 criteria, to assure the long-term
((ffk corrosion protection for the pressure boundary.
(N(.).0:
f.a.
I s The remaining coatings listed in Table 1 are Galvanox h
V, Unqualified Paints, Mill Scale and Varnish. All of
.Y.. %3.f;h these coatings are non safety-related, and are !fm.;6 ~.
n considered uncualified by the definition of
[./[y D>E ,
SRP-6.1.2-III. They are assumed to form solid debris .[: -}.j
^ " ' "
under LOCA conditions for the purpose of this evaluation, though complete failure is not expected.
M
-Y MULTIPLE DYNAMICS C@RPORATI@N _.
'i e
SUSJECT: DATE ISSUED DOC. NO.
4 EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 . _ _ _
COATINGS FOR FERMI 2 '-
REVISION: 3 PAGE 11
' ?
=-
.c -
6.0 FAILURE MODES OF UNQUALIFIED COATINGS Containment coatings are postulated to fail in one of two modes under extreme environmental conditions.
Inorganic metallic type coatings, such as CZ-11, 3 u T4 m Galvanox V, Keeler and Long #7720 and aluminum based
~
paints, without any top coating, have low tensile
~
strength and are very brittle, such that any sizable -
flakes, separating from surfaces due to lack of proper = i adhesion, crumble into small particles. This failure .
mode has been identified as the most likely by the 4 (
largest manufacturers of inorganic type zinc coatings .
(Mobile, Ameron, Carboline and Napko). When directly .
impinged by steam or water, these coatings separate from the surface in small particles as a result of the scouring action, j Phenolic and polymer based paints when applied in .
thicknesses of approximately 3 mils and over are likely I to separate from the surfaces in the form of peels, *]
blisters, and flakes as a result of chemical breakdown _
and extreme temperature exposure. z Coatings of less than 3 mils generally are too thin to sustain the strain of peeling or blistering and the _
adhesion force, and are likely'to separate in or . ; i disintegrate into small particles. -; E m
For the purpose of thic evaluation and in estimating y the total quantities of paint debris, 100% failure of : )
L t,
F S -
- _=
MULTIPLE DYNAMICS C@RPORATION h r
SUBJECT:
DATE ISSUED DOC NO.
{
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 )
COATINGS FOR FERMI 2 REVISION: 3 PAGE 12 -'
=
7 all the unqualified coatings was conservatively [
assumed. /
l l
The DBA test report No. 56878 issued by Carboline ;
(Reference 1) demonstrated that Carbo-Zinc 11 is not Y _
l lost in flakes, but rather in particles of a size less than 20 microns. The report further states that the particles do not dissolve in water and do not clog M screena. The density of CZ-11 dry film coatings is _
between 3 to 4 times that of water, and particles are ; _
expected to settle to the bottom of the drywell and the suppression pool, concentrating as sludge in low (
velocity areas. The particle separation mode is a result of continuous scouring action of steam and water spray as simulated in DBA testing programs. In a _
typical BWR containment, direct scouring occurs only in ,
the immediate vicinity of the postulated pipe break and '
within a few feet of the containment spray headers (if used). The total affected surface area due to steam c and water scouring is less than 10% of the CZ-11 coated -
areas. Temperature resistance of CZ-11 up to 750*F is =
e considered excellent by the manufacturer, as described in Reference 2.
- The failure mode for Galvanox V is considered similar to that cf CZ-11, based on the physical properties provided by the manufacturer (Reference 3). However,
~
no detailed test data exist to Quantify the rate of deterioration. The tensile strength of Galvanox V dry ?
I film coating is such that if flaking occurs, the flakes
=-
MULTIPLE DYNAMICD CCRPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 13 will break up into small particles within a turbulent fluid or when subjected to gravity impact. The resulting debris can be classified as solid hydroxide as defined in NUREG-0897, Reference 4, in respect to ECCS pump and strainer performance.
Very small quantities of unqualified miscellaneous paints within the drywell are assumed to fail under LOCA conditions. In the absence of substantiating test data, these paints are assumed to fail in the form of flaking and peeling. The resulting debris is considered similar in behavior to fibrous insulation material with a near neutral buoyancy. The detailed coating survey determined DFT (Dry Film Thickness) measurements between 0.7 and 6.1 mils, with 70% of the coatings less than 3 mils. A DFT of less than 3 mils is not expected to result in flaking or peeling , and the coating will disintegrate into small particles as 3
discussed previously.
Even though not considered a coating under standard definition, the mill scale present on some uncoated carbon steel surfaces was included in this evaluation.
Mill scale exists on uncoated structural steel and on uncoated, uninsulated carbon steel piping. The piping has a thin varnish applied over the mill scale. To determine the failure mode under LOCA conditions, short-term simulation tests were conducted by the Engineering Research Department of Detroit Edison.
These tests concluded that the mill scale remained
MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATEISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 14 [
~
adherent to the steel in a hot (210*F) dry nitrogen atmosphere. In a distilled water immersion test in a nitrogen environment, the mill scale spalled off within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> at 210*F. The mill scale particles ranged in size from 4 to 60 microns and settled on the bottom of the test container. In addition, a LOCA simulation test was conducted, simulating the BWR LOCA conditions for an SBA event. The mill scale specimen was subjected to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of steam atmosphere at 340*F, followed by 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> at 250*F. Mill scale did not --
separate, removal of rust was observed on the surfaces.
Particles found in the container were in the form of sludge and water discoloration. Based on these results, it is postulated that some of the mill scale will spall off during postulated LOCA conditions. The density of mill scale is approximately 350 lbs/cu ft,
~
and the scale is expected to settle in the drywell bottom and suppression pool as a sludge. The sludge is classified as hydroxide as defined in NUREG-0897.
Temperature resistance tests were also conducted for the varnish present on uncoated, uninsulated steel piping. Specimens were baked at 340*F in a dry oven, and the varnish was observed to sof ten but did not run or separate from the surfaces. No other deterioration was identified.
7.0 SAFETY EVALUATION OF COATINGS The application of primary containment coatings has inherent positive and negative aspects. As discussed
MULTIPLE DYNAMICS CORPORATION q: ..g W.y h
SUBJECT:
DATEISSUED DOC. NO. h.
n .q e EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 L' Q )
COATINGS FOR FERMI 2 4i.;d e REVISION: 3 PAGE 15 h. y 4:
n?n N 35 7 L-Q l ip.:
.a:4 earlier, the desired positive qualities of long-term {:
corrosion protection and decontamination must be .g..C; .1 balanced with the potential negative aspects of e,14 ' .". ;.
hydrogen evolution and debris generation during or f,..j)fl following a postulated LOCA. This section, therefore, individually examines these aspects to assure that the d[r[:
y .$.4 - <
, t r(7g6 g applied coatings will not degrade or affect the safe Q.'Q ;
operation of the plant. v .;.;%4
..;p y .
7 .1 Corrosion Protection [.'.:.*'pk Under normal operation, the Fermi 2 containment is "y ~ c.i 5 nitrogen inerted and maintained at an operating temper- ff.'f ature between 135 and 150*F. The drywell cooling (.' k system continuously removes excess moisture from the .; .
$.'.. 1:%,
environment to maintain a dew point near the EECW/RBCCW
,:3 . x water temperature (approximately 95*F). The drywell
./.r..: -
F.M atmosphere is therefore substantially below saturation. D ?.3 3 .. N Short-term, substantial corrosion during normal 6/@-
operation is therefore not expected, nor has it been [.)-y +
observed in operating BWR's. The existing CZ-11 >/ g . .I e ., ,
coating on the pressure boundary, together with repair 9Rrs and touch up coating, will adequately protect the T.3 2 containment from corrosion. Long-term effects, if any, . .
will be monitored via visual inspections during T. @ '.
~.;_ ,
refueling outages, and will be repaired in those areas J;f:23r N ..: .
where rusting or discoloration is apparent. Deteriora-
{ f5 tion of the CZ-11 coating is anticipated during a postulated LOCA. Following the LOCA and containment
((
. . i . ,,,
depressurization, the onset of corrosion can be 1!
... g.- r expected and will progress over time at a predictable q. j. . ,
f.I. g. g S;
fI1 y : k *_
b, *_
.id _ f
i.,. - - .r- . g- i eiwi . i n- : .M n - i 9' %.' n. "4 -ri i q ,1 . m, in -. i F-M n MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATEISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 16 rate. In the absence of significant concentrations
(>25%) of acids and chlorides in the containment following a LOCA, a maximum corrosion rate of 0.02 to 0.05 inch per year is given in the literature, Reference 5. With a minimum shell thickness of 0.75 inch, containment integrity and leak tightness can be assured for years following the LOCA.
7.2 Decontamination The coatings within the containment also serve as a surface sealer to prevent contaminated fluids and particles from penetrating into porous materials and crevices. Washdown and other physical decontamination methods are made more effective and less time consuming, thereby promoting ALARA personnel exposure considerations. Decontamination itself is not considered a safety-related activity, but more specifically a normal plant maintenance item. It is an objective of Fermi 2 Nuclear Production to coat interior containment surfaces and maintain these ,
coatings during the life of the plant for economic and ALARA reasons. Further discussions regarding ALARA considerations are provided in the Addendum.
7.3 Hydrogen Evolution The evolution of hydrogen from the corrosion of ]
aluminum, zinc and zinc-base paints has been previously j assessed in the Fermi 2 FSAR Section 6.2.5.3.1. The ]
two hydrogen recombiner systems have been adequately '
I
=
J MULTIPLE DYNAMICS CORPORATION :
1
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 4 COATINGS FOR FERMI 2 REVISION: 3 PAGE 17 sized to comply with the provisior.; of Regulatory Guide I 1.7.
s1 1
8.0 EFFECTS OF COATING DEBRIS To evaluate the effects of coating debris generated by the postulated failure of unaualified coatings, bcth solid hydroxide type and fibrous debris (paint peels) are assessed separately in respect to their transport mechanisms and affect on ECCS performance. The assumptions used are extremely concervative for the purpose of this analysis.
8.1 Debris Transport The debris generated from the failure of unqualified
^
coatings and mill scale during and following a LOCA will be transported to the suppression pool by the flow of water, steam and noncondensible gases from the drywell. After initial surface separation, the coating particles will tumble to the drywell floor or other horizontal intervening surfaces by gravity. Pipe break flow or containment sprays will then flush most of the debris through the vent cystem into the suppression pool. High vent system flow velocities occur only during the initial 20 to 40 seconds following the DBA-LOCA while the reactor vessel depressurizes. i Coating debris which remains in the drywell after this initial transient can be transported to the pool by the ECCS flow or containment spray flow. Drywell floor and vent line/ header flow velocities are very low and
MULTIPt.E DYNAMICS CORPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE yg subject to gravity flow only. A pool of approximately 2 feet depth will develop on the drywell floor before excess water spills over to the vent lines into the vent system. Debris particles with densities significantly higher than water (such as CZ-11 and mill scale) will settle out in the drywell floor bottom.
Particles which are swept over into the vent system will flow through the vent header and exit the downcomers near the center of the torus.
Average water velocities within the torus following the initial transient are below 0.25 ft/sec, but approach 0.3 f t/sec within a 4 foot hemisphere of the largest 3-suction strainers at maximum RHR pump capacity. Refer to Figure 1 for the dimension and arrangement within the torus. The maximum surface velocity at the strainer surface was calculated to be 2.5 ft/sec. The strainer intakes are located 9'-8" from the center of the torus and 5'-9" above dead bottom. For these pool velocities hydroxide type particles are therefore not postulated to reach the strainers in a realistic situation.
Ungualified paints with DFT's of less than 3 mils that 3 f ail in a peeling or flaking mode are expected to float on the pool surface or remain suspended in water, where they could ultimately migrate toward the ECCS suction strainers. The particles of these thin coatings are expected to be small enough to pass through the 3
strainers as discussed in Section 6.0.
MULTIPLE DYNAMICO CORPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 -
REVISION: 3 PAGE 19 l
VENT SYSTEM .
19' ~' 7 ~ ~ W ATER'LR - ~
=2 dY ~ ,e
.M a
'<f ~ s,X ECCS SUCTION . ~~ " '
, SUCTION STRAINER t
NOZZLE
-- \ '
. . e-
- 3. .
s
- J. . .
TORUS SHELL' 9 '- 8** .,
ECCS SUCTION STRAINER LOCATION FIGURE 1 Y
3 MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATEISSUED DOC. NO. -
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 i COATINGS FOR FERMI 2 _
3 REVISION: PAGE 20 3-F 7-8.2 ECCS Performance ,
To conservatively evaluate the postulated performance _
of plant systems and equipment, it was assumed that the --
hydroxide type debris, consisting of CZ-11 and mill T scale particles, are completely suspended in the pool water during the early turbulent phase of the LOCA.
Under this assumption, the total hydroxide particle 7-concentration in the pool is calculated to be less than O.35%. A more realistic assumption, using a 10%
fraction of the debris as discussed in Section 6.0 would produce a hydroxide debris concentration of ;
0.035% in the pool water. In accordance with the -
conclusions and guidelines given in NUREG-0897, Section , - .
3.2.2.4, a solid hydroxide concentration of less than -
1% of mass does not affect pump performance. The solid particles are less than 60 microns in size, and will therefore freely pass through the 1/8" holes in the -
suction strainers of the RHR, Core Spray and HPCI lines. __
n The fibrous debris, generated by failure of unqualified 3 thin coat paints, is also transported to and ,
distributed in the pool water volume. The total volume of 0.5 cu ft (72 pounds), as shown on Table 1, results in a volumetric concentration of 4.1 x 10-4%, well below the acceptable limits of 4% given in NUREG-0897 for fibrous debris. The particles are expected to be 3 small enough to freely pass through the strainers. In __
addition, the ECCS pumps and system piping has been h
g g
MULTIPLE DYNAMICS CORPORATION - g p]S. ...i f
'T J.
?O
SUBJECT:
DATEISSUED DOC. NO. , t.9
-( %
EVALUATION OF CONTAINMENT COATINGS FOR FERMI 2 Oct. 1984 DECO-12-2191 <
.? %
y.' .
l REVISION: 3 PAGE 21 U d
- no
?
, designed for a worst case assumption of 50% strainer Ef-lh 62,gQj
- - . , . - v f blockage for the Design Basis Accident. 4t 1 7 ' i%.
I 8.3 containment sprays The Fermi 2 containment spray headers are equipped with % + ..- l a , 1- >e
"=
a 1-1/2-7G25 fog nozzle as manufactured by Spraying -"? .? ke.
= s
.. y Systems Company. The fog nozzle has a fleet passage of . ".' '.; i . '
n 0.125" which is of equal size to the ECCS strainer c
,;[. p.h, '
passage. .p (y 4..j' t
3 . .;s . c. , ,
Solid and fibrous particles which pass through the 3 fi i '..s' k strainer are therefore expected to also pass through .G., . >,
e .
=
the fog nozzle, and no clogging or performance &,* . u i degradation occurs. E. 2-). / . ':
3 v. .
i.r:' .
h
.s g
The plant design basis does not require containment g ~ *:
~
spray operation. This function, however, is desirable ^ .:
v, .
to mitigate the consequences of a LOCA.
'.9i ;f+; >
.1 v
Er j ', - j. ,-
8.4 RPV Core Spray and Feedwater Spargers 6:
'i r....
r ECCS flow is injected into the reactor vessel directly : " ('/ j . .
- via the RHR and Recirculation System piping and/or Core
).' . .
- Spray piping. No intervening obstructions are located 7..;.( . ..
f
=
in the RHR flow path. Core spray flow is injected via [.~./.,
. c .
f the core spray spargers located directly above the N i-f fuel. The minimum flow passage through the sparger is [D more than 0.5 inch, which is larger than the strainer f-I flow passage. Similarly, HPCI and RCIC flow is [ : tg : .
injected into the vessel via the feedwater system g* " ..,ny 5
s piping and spargers. The minimum flow passage in the .[ C
- ' . ,l.y.. .2
.b: N . . e.
[ . Vj; .
3 2
=-
yy .
ni.+
'i ,
' . i
MULTIPLE DYNAMICS CORPORATION
SUBJECT:
DATEISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 22 feedwater spargers is more than 0.5 inch, and no flow ;
blockage or performance degradation will occur. 1 The potential ultimate depositioning of debris within 3
the reactor vessel was evaluated by General Electric ._
Co., as discussed in the Addendum (Item 3).
The recirculation pumps are equipped with a .
demineralized water seal purge system, thereby '
preventing particle intrusion into the pump seal and bearing assembly.
8.5 Normal Operations During normal plant operations, and as a result of operational vibrations, some coating and mill scale debris is expected to be generated. In the absence of any fluid flow between the drywell and torus, the particles will accumulate on the drywell floor and other horizontal surfaces. Some airborne particles may be transported via the vent system into the torus during purging and inerting operations, where they will settle to the bottom. The Torus Water Management System (TWMS) is designed to continuously process torus water through a demineralizer system to maintain water cuality and to prevent gradual build-up of particles, contamination and sludge. The TWMS takes suction from the bottom of the torus. Normal operations, including ECCS testing, is therefore not impaired.
MULTIPLE DYNAMIC 3 CORPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 23 9.0 ADDITIONAL CONSIDERATIONS 9 .1 Surveillance and Inspection Following completion of the repair and touch-up of the qualified coatings and the CZ-11 coating on the con-tainment pressure boundary, a thorough visual inspec-tion will be conducted by Detroit Edison Nuclear Quality Assurance. Af ter commercial operation is achieved, these same surfaces will be visually inspected during refueling outages to determine onset of corrosion, blistering or peeling, and coating discoloration. The extent of these inspections will be commensurate with the number of affected areas found.
Suspect areas will be cleaned, including manufacturers' recommended surface preparations, and new coatings will be applied in accordance with the original application criteria.
9.2 Additional New Coatings As discussed earlier in this report, uncoated carbon 3
steel surfaces may be coated in the future as much as practical to reduce corrosion and facilitate decontamination, if necessary. Surfaces will be prepared with hand power tool cleaning, and CZ-11 coatings will be applied in accordance with manufacturers' recommendations. These coatings will be applied on a schedule not interfering with the critical path. As time permits, coating activities will
- continue during refueling outages, when the containment is accessible.
MULTIPLE DYNAMSCS CORPORATl3N
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 24 Unaualified coatings with DFT's over 3 mils (excluding inorganic zinc primers) will be removed from equipment 3 surfaces and recoated with CZ-11 where appropriate.
10.0 CONCLUSION
Af ter detailed evaluation of the coating qualities, cuantities and potential failure modes, it is concluded that the Fermi 2 coatings currently applied within the primary containment do not adversely affect the safety of the plant, and will not impair normal or abnormal operation. The coatings are therefore classified as follows:
Suppression Pool -
Plasite 7155, qualified and safety related Drywell -
Carbo-Zinc 11, unqualified, safety related, see Section 5.0, page 10 Concrete Surfaces - Ameron 66, qualified and safety related Structural Steel -
Carbo-Zinc 11, unqualified, not 3
and Equipment safety related*
Mill scale, unqualified, not-safety related*
Ducting -
Galvanox V, unqualified, not safety related Carbon Steel Pipe - Mill scale / varnish, ungualified, not safety related*
~
ER -
MULTIPLE DYNAMICS CORPORATITN
SUBJECT:
DATEISSUED ' DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2
~
REVISION: 3 PAGE 25 Misc. Coatings -
Various, unqualified (DFT of 3 3 mils or less), not safety related*
- Surfaces to be coated and/or repaired per manufacturers' recommendations as time permits.
The Fermi 2 FSAR and applicable specifications will be revised to reflect the conclusions of this assessment.
11.0 REFERENCES
- 1. The Carboline Company, Report No. 56878,
" Carboline /ORNL Round Robin DBA Testing - Test II"
- 2. The Carboline Company, Carbo-Zinc 11 Product Data Sheet of February 1981
- 3. The Carboline Company, SUBOX Division, Product .
Bulletin No. 29 of June 1979, Galvanox Type V
- 4. NUREG-0897, Revision 1 Draft, USNRC, " Containment Emergency Sump Performance"
- 5. Perry's Chemical Engineers Handbook, Table 23-3, 4th Edition 7 i .,
.'d .
- g. -
- e
~
J['-E I'5.. .
. . 3
MULTIPLE DYNAMIC 3 CCRPORATION
SUBJECT:
DATE ISSUED DOC. NO.
EVALUATION OF CONTAINMENT Oct. 1984 DECO-12-2191 COATINGS FOR FERMI 2 REVISION: 3 PAGE 26 12.0 ADDENDUM i Additional information and responses to NRC questions in 3 accordance with Detroit Edison-NRC meeting on September 14, 1984.
I L.
l
- 1. ALARA Considerations (Reg. Guide 8.8)
[
The NRC requested that Detroit Edison provide additional :
information describing the primary containment radiation protection provisions which will ensure that occupational -
radiation exposure will be as low as reasonably achievable "
(ALARA). -
The following provides a discussion of the plant operating and -
administrative procedures, the design provisions and surface ;
coating considerations which, in combination, ensure that -
occupational exposure will be maintained ALARA. -
(a) The primary containment is classified as a restricted T area with limited access. Access will normally only be required when the plant is in cold shutdown and j7 refueling. Health Physics will conduct detailed surveys _g 3
to map area radiation levels and sources, identify n minimum protective measures, and strictly control personnel access and the duration of access. Because the jj -
drywell is a highly congested area, work packages and _
inspections will be thoroughly planned and the personnel __- _
indoctrinated and trained to skillfully perform their required activities so as to minimize the time spent in the drywell and the radiation exposure levels. -
(b) The drywell equipment drains systems is designed to f collect the leakages from all anticipated leakage - .
sources, such as valve stem packings and recirculation pump seal leak-offs. The leakages are collected in a .
piping system and routed to the drywell equipment drain 'e sump. Leakages from unidentified sources are collected i in the drywell floor drains sump. The floor drains sump -
level instrumentation will detect, and alarm in the main =
control room, leakages greater than one (1 ) gallon per if minute. As such, with the drywell equipment and floor drain systems, significant surface contamination buildup _
due to primary pressure boundary leakages is very -
unlikely. Operating BWR's are equipped with very similar "-
drywell drain collection and leakage detection systems.
Historically, significant surface contamination at operating BWR's has not been a problem and decontamina-tion procedures to control radiation exposure from ',
general surface contamination have not been required. 7;
( c) The surface coatings which have been applied in the 3 primary containment complement the procedural methods and plant design features for radiation protection by further minimizing locations where contaminated fluids and particles could accumulate. The major surface areas in ?
the primary containment, including the porous concrete _
surfaces, major portions of the structural steel and 7
=
3
- 1. (Cont'd) piping , and equipment and component surf aces , have been coated and sealed to prevent any accumulation of contaminated substances and subsequent higher radiation exposure levels.
(d) It is Detroit Edison's objective to properly coat surfaces in the primary containment and to maintain these coatings during the life of the plant. Problems with surface coatings can lead to new sources of contaminated particles, increased decontamination problems and higher exposure levels. Radiation exposures associated with inspection and maintenance of coating systems must also be properly considered in determining the need to apply coating to the remaining uncoated surfaces. Detroit .~ l Edisor. will evaluate the radiation sources and levels .
g inside the primary containment during each outage and ;
apply additional surf ace coatings, as appropriate, to ensure that occupational exposures are as far below the limits specified in 10CFR20 as is reasonably achievable.
- 2. SRP Review Regarding Coating Requirements ,
Detroit Edison was requested to review all applicable provisions of the SRP in respect to primary containment coating criteria and acceptance provisions. .
A general review of the SRP was conducted to identify guidance provided by the NRC in regard to the application of coating ;
systems in the primary containment. Section 6.1.2 of the SRP ;
provides specific regulatory criteria concerning this subject.
Supplemental criteria is included in Section 3.8.2. No other ,
sections in the SRP were found to discuss the application of 2 coating systems.
The specific criteria of Section 6.1.2 and the supplemental criteria of Section 3.8.2 has been considered in the Fermi 2 coating evaluation report. Therefore, this evaluation, which 1 considered the qualified coatings, unqualified coatings and the uncoated surfaces has addressed all the acceptance >
provisions of the SRP.
y
- 3. Evaluation of the Effects of Hydroxide Debris in the Reactor f vessel f The question was raised whether the effects of the hydroxide type debris given in Table 1 (page 8) of Report No.
DECO-12-2191 were evaluated in respect to the reactor vessel.
The debris, consisting of CZ-11 and mill scale amounts to less than 100 cu ft, in the form of small particles that readily traverse the ECCS suction strainers.
- 3. (Cont'd)
Even though the probability of the postulated amount of hydroxide debris reaching the RPV is extremely remote, we requested the General Electric Company to evaluate the potential concern.
The effect of up to 100 cu ft.of hydroxide debris (20-60 microns in diameter) being deposited in the RPV by the ECC systems during post LOCA cooling is expected to be negligible.
All critical flow paths inside the vessel (jet pump nozzle, core spray nozzle, top and bottom of fuel bundles) are larger than the strainer flow passage. As a result, most of the debris will be deposited in the lower plenum where it does not +
affect the ability of the ECC system to provide adequate core y cooling. 4 Each fuel channel forms an essentially independent flow path connecting the upper and lower plenum in the core bypass region. In order to maintain adequate cooling, less than one gallon per minute must be provided to each assembly. Each fuel assembly has 3 independent inlet flow paths, the top and bottom of the fuel bundles, and the flow paths between the bundle and the bypass. Calculations have been performed which show that all 3 flow paths have to be greater than 99% blocked before any fuel damage will result. It is highly unlikely that debris of 20-60 microns in size could produce this amount of blockage in any channel under the flow conditions present in the vessel following a LOCA. Thus, depositing hydroxide debris from paint inside the RPV will not impair the ability of the ECC systems to maintain adequate core cooling.
- 4. Purge Valve Operability Confirmation was requested that inlet screens to the purge valves have been provided and that the valves will be capable of operating with potential paint debris passing the screens.
The two 24-inch drywell purge inlet and outlet valves (VR3-3012 and 3023) are protected from potential paint debris by stainless steel screens. The screens are made of perforated plate with 1/8" diameter holes, 33 holes per square inch. The maximum particle size that could traverse the screens is therefore 1/8 inch in size with a thickness of less than 15 mils. The purge valves are normally closed and are opened for only short periods during normal operation to control containment pressure and just prior to and following refueling when inerting or deinerting is performed. In the extremely unlikely event that a LOCA coincides with purging operation, the valves are capable to close in less than 2 seconds.
?
- 4. (Cont'd)
While the valves are closing, during the initial stage of the LOCA, flow velocities across the valve seat can be in excess of 600 ft/sec, particularly during the latter phase of closing when critical orifice flow is approached. It is therefore highly improbable that small paint particles can be trapped between the seats. Only very small quantities of paint debris c are expected to be generated within the short tirae and transported to the screens.
We therefore conclude that the operability of the purge isolation valves is assured.
- 5. Evaluation of Potential Strainer Blockage from Unqualified Coatings
~
A more detailed evaluation was requested, to substantiate that the unqualified coatings will not excessively block the ECCS g suction strainers and that ECCS flow is not degraded.
ThetotalsurfaceareaogtheECCSsuctionstrainersis calculated to be 92.1 ft Unqualified coatings within
[ .
I the primary containment constitute a total of 1,785 f t 2 as j shown in Table 1 of Report No. DECO-12-2191. This quantity
- was derived from the following
P E 37 Valves (yokes, bonnets, bodies) - 460 ft2
! Valve Operators (M.O., A.O.) -
90 ft2 j Recirc. Pump Motor Flanges (2) - 220 ft2
- MSIV Top Works (4) - 100 ft2
! 2790 ft of Conduit Labeling - 350 ft2
- 500 ft2
- Miscellaneous 62 Terminal Box Labelling -
62 ft2 In an effort to further quantify the surface areas, a more detailed survey was conducted, including additional
- dimensional verification and DFT measurements. Based on this updated survey, the unqualified coatings were categorized into
a A. Coatings with average DFT of 3 mils or less B. Coatings consisting of zinc primers r C. Coatings thicker than 3 mils DFT
^
The coatings with DFTs of 3 mils or less do not produce flakes or peels capable of blocking the strainers, as discussed in Section 6.0 of Report No. DECO-12-2191.
I i Coatings consisting of inorganic zinc primers, such as Keeler
- & Long #7720 on the Recirc. Pump Motor Flanges, are
=
categorized as hydroxide debris together with Carbozine 11 and Galvanox V. These coatings do not produce debris capable of
___h_m . . .
- 5. (Cont'd) blocking the strainers as discussed in Section 6.0 of this report.
The remaining thicker coatings are identified and tabulated as follows:
25 Valve Operators (3/4" to 28" valves) - 142 ft2 Recirc. Pump Motor Holst -
18 ft2 Recire. Pump Motors (2) - 440 ft2 Ventilation Fan Housing (14) - 256 ft2 These unidentified coatings will be removed prior to fuel load and the surfaces will be recoated with Carbozine 11. This action will effectively remove the potential generation of fibrous debris from unqualified coatings, such as flakes and peels, capable of plugging the ECCS strainers.