ML20246C977
| ML20246C977 | |
| Person / Time | |
|---|---|
| Site: | Vermont Yankee File:NorthStar Vermont Yankee icon.png |
| Issue date: | 07/01/1989 |
| From: | Pelletier J VERMONT YANKEE NUCLEAR POWER CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| VYV-89-124, NUDOCS 8907110202 | |
| Download: ML20246C977 (16) | |
Text
%
p
?
v..
VERMONT YANKEE NUCLEAR POWER. CORPORATION.
y
~ P. O. BOX 157 i-GOVERNOR HUNT ROAD VERNON, VERMONT.05354 July 1, 1989 VYV# 89-174 U.S. Nuclear Regulatory Commission.
Document Control Desk Washington, DC 20555
References:
(a) License No. DPR-26 (Docket No. 271)
(b) Letter, USNRC to VYNPC, NVY 89-126, Inspection Reoort 50-271/89-80, dated June 2,1989 Subjects.
Vermont Yankee Response to WC Recuest for Information Regarding-Condition of Drywell Paint (Inspection Report 89-80):
Dear-Sir By lett'er. dated June 2, 1989 beference (b)), NRC requested Vermont Yankee to provide certain information regarding the condition of paint in the Vermont Yankee drywell.
Accordingly, enclosed.please find a copy of our evaluation of drywell paint concerns. This evaluation contains the information you requested.-
We trust that this information is suf'icient to address your concerns, however, should you have any questions or require any additional information regarding this issue, please do not hesitt te to contact us. -
Very truly yours, Veimont Yankee Nuclear Power Corporation p y-
/'
PdX' *X1 aJ James P. Pelletier Plant Manager ccs LUSNRC Regional Administrator, Region I USNRC Resident Inspector, VYNPS h'
8 I
'8907110202 890701 PDR ADOCK 05000271 Q
PDC a
l l
I ENCLOSURE A l
CONTAINMENT PAllfr EVAIJJALTION B ASIS
- 1. Intmduction l
In 1985, the reactor recirculation piping at Vermont Yankee was replaced with new i
recirculation piping. The decision was made at that time to utilize NUKON fibmus insulation on the new piping,in place of the mirror-type insulation that was utilized in the past. As part of this insulation change, a study was performed for Vermont Yankee by General Electric (Reference 1).
This study conservatively ' assessed the potential reduction in Net Positive Suction Head (NPSH) which may occur as the result of a design basis LOCA in the drywell which in turn transported quantities of dislodged NUKON insulation fibers through the downcomers, into the torus, and to the ECCS intake strainers. Inadequate NPSH to the ECCS pumps could lead to cavitation which in turn could lead to pump damage. The analysis in the General Electric study was done in accordance with the procedures outlined in Regulatory Guide 1.82, Revision 1, and NUREG 0897, Revision 1. This Regulatory Guide and NUREG were in turn based upon an extensive amount of analysis and physical testing of insulation fiber transport (References 2 through 5). The physical testing was performed at a facility constructed for this purpose at Alden Research Laboratory.
The conclusion of the General Electric report was that the Core Spray intake strainers possessed sufficient surface area to accommodate a conservatively derived quantity of fibrous insulation from a design basis LOCA in containment, but that the surface area of the RHR strainers I
needed to be increased. The RHR strainers were subsequently replaced with larger strainers for this reason. There is no concern relative to the HPCI and RCIC strainers, since HPCI is designed to cope with a small-break LOCA only (minimal fibrous insulation damage), and RCIC is not an ECCS system and is therefore not required to respond to a LOCA scenario.
1 1
During a recent NRC inspection of plant maintenance at Vermont Yankee, it was noticed that a significant amount of paint topcoat was peeling away from the upper drywell wall. The inspector asked if the effect of paint chips on the ECCS rtrainers had ever been evaluated. The effect of dislodged paint had been considered by Vermont Yankee in the past, but due to the brittle nature of the dislodged paint the tiny fragments that eventually travelled to the toms were deter nined to be of no consequence, since their relatively small size would allow thent to pass through the strainers.
However, with the installation of the NUKON insulation in 1985, the potential exi,ts for the paint chips to combine with the dislodged insulation fibers from a design basis LOCA and further reduce NPSH to the ECCS pumps at the strainers. The purpose of this report is to assess any potential additional effect the paint chips may have on strainer plugging and subsequent reduction of NPSH to the ECCS pumps.
II. Backcround The issue of peeling paint in primary containment dates back to the early 1970's. The attached list of references indicates Vermont Yankee's awareness of this issue and summarizes the variety of evaluat;ons performed. Reference 18 provides a detailed summary of the issue as well as the basis for the cunent evaluation.
References 12 & 13 provided an evaluation of the condition of the drywell coating in response to an NRC Resident Inspector's concem during the 1983 refueling outage. The evaluation found the condition of the drywell coating to be generally good, with some degradation in the area near l
the upper containment spray header, and in the upper drywell area (loose paint was subsequently removed). It also stated that any further degradation would be detected as part of Operations Department Primary Containment Surveillance procedure OP 4115 and would be documented on VYOPF 4115.06.
2
~
In his inspection report of 2/20/86 (Reference 14), the NRC resident inspector noticed during the recirculation pipe replacement outage that peeling of the drywell epoxy coating in the upper elevations had progressed to some extent beyond that observed previously. The resident inspector requested that additional evaluation of the problem be pmvided to ensure that previous conclusions were still valid. The plant subsequently issued a service request (Reference 15), whereby YNSD was requested to perform additional evaluation work. The YNSD evaluation was performed and transmitted to the plant via Reference 16 on March 17,1986. This evaluation recommended that all I
loosely adhering drywell paint be removed by hand scraping prior to startup from the outage, and that samples of the removed paint be retained for long-term evaluation to determine the root cause of the topcoat failure. Reference 17 was subsequently issued on May 21,1986, which provided plant disposition of the evaluation recommendations. Based upon a May 7,1986 Project Meeting, the plant decided to go forward with the recommended paint removal, including reapplication of j
primerin any areas with lost primer.
During the present refueling outage (March,1939), it was found that more topcoat was peeling away in the drywell, predominantly in the upper drywell region (above the upper drywell spray ring header). An NRC maintenance inspection team also noticed the peeling paint, and inquired into the potential effects of the loose paint on ECCS operability. This inquiry prompted a re-evaluation of the condition (Reference 18).
1 l
III. Discussion YNSD was requested to perform an evaluation of the present condition of containment paint.
To supplement the YNSD evaluation, Stone & Webster was contracted to provide a coatings f
specialist for the following services:
3
u l1 1). Review and evaluate Vermont Yankee's containment paint system, 2). Review short-term plans i
3). Provide long-term recommendations.
On Friday, March 17, and Tuesday, March 21,1989, YNSD engineers visited the site to conduct an inspection of the drywell and torus with Vermont Yankee plant personnel and with Mr.
Richard Martin, a coatings specialist from Stone and Webster. Rourine scraping of the torus topcoat was in pmgress at the time of the inspection. The drywell was inspected and it was found that the majority of loose paint was in the upper drywell region (above the upper drywell spray ring header), while the lower drywell region's topcoat had remained mostly intact. (Since the upper drywell region is normally at a significantly higher temperature during plant opemtion, it appears that the failure of topcoat is at least somewhat tempenture dependent).
(
On nursday, March 23, Mr. Martin was contacted to obtain his preliminary findings. The upper drywell exhibited numerous areas of topcoat peeling, while the lower drywell (spherical portion) exhibited little to no topcoat failure. The failed topcoat is extremely brittle, and shatters into very small chips when disturbed. With regard to the cause of topcoat fcilure, Mr. Martin noticed a fair amount of zine powder residue in several locations where there was no topcoat. It is I
possible that the zine base primer was not pmperly prepared before it was applied, contributing to a lack of adhesion. Additionally, the remaining topcoat on the upper drywell shows sigt.s of discolomtion due to thermal aging, which would cause the topcoat to shrink, become brittle, and loose adhesion. On the other hand, Stone and Webster found that the primer (Carbo Zinc 11 inorganic primer) was in very good condition. The primer was measured to be approximately 2-1/2 e.ls thick, while the topcoat consists of two layers, approximately 12 mils in thickness total.
Stone and Webster's preliminary recommendation was to manually scrape for loose topcoat with a metal scraper (plexiglass scrapers have previously been used). Metal scrapers should not 4
i si damage the primer, due to its high durability, and more topcoat should be removed by this method.
)
Any damage to the primer that may occur should be promptly touched-up with approved primer, in accordance with existing plant procedures. For the long term, Stone and Webster recommends the continuing removal of topcoat, with no further action being recommended in terms of a protective coating other than maintaining the primer in good condition. The existing primer prte ides excellent protection of the underlying metal surfaces, and no replacement of topcoat is deemed necessary. These recomraendations were verbally forwarded to the plant, and use of metal scrapers were incorporated into the drywell topcoat removal effon with the goal of removal of all loosely adhering topcoat.
On Tuesday, March 28,1989,i NSD and Vermont Yankee engineers inspected the results of the manual scraping effort, which had been completed. In the opinion of the engineers, the scraping effort was highly succedul, eliminating essentially all loosely adhering topcoat fmm areas which have undergone toperat degradation. The manual scraping effon reduced the amount of topcoat in the upper drywell region to approximately 30% remaining topcoat surface area near the top of the drywell, and approximately 70% remaining topcoat surface area in the lower region of the upper drywell, resulting in approximately 50% topcoat remaining overall in the upper drywell region. The torus scraping effort was inspected during the March 17,1989 inspection, with satisfactory results.
IV.Effect of Dislodged Paint on ECCS Strainers In order to quantify the effect of paint chip transport and their effect on ECCS strainers, several different methods were attempted; however, the lack of physical tesing data for BWR paint chip transpon and the affect of accumulation on a pump suction strainer (as exists for insulation transpon) hampered YNSD attempts at such methodologies. Mr. George He:ker of Alden 5
l t
l ll 1;-
Research Laboratories was contacted on Wednesday, March 22,1989 (Alden Labs did the physical testing of insulation transpon and strainer plugging, upon which the NUREGS and Regulatory Guide 1.82 are based). He confirmed YNSD findings that there is no data available with regard to paint chip transport or plugging of strainers by paint chips, with or without combination with l
insulation fibers.
I However, sufficient basis was found based upon YNSD qualitr/ive evaluation to conclude that paint dislodged from either the drywell or torus surfaces would be of no consequence with regard f
to ECCS strainers and ECCS pump performance. The following summarizes the main points used to form this condusion:
i
- 1. Examination of both drywell and toms topcoat samples retrieved from Vermont Yankee by the Stone and Webster paint consultar reveal by physical test that the paint chips have a idgher density than water. The paint chips were dropped into a basin of water and, upon breaking of the surface
{
tension, they immediately sank. It is believed that insulation fibers would be more apt to remain in suspension in the torus and be swept onto an ECCS intake screen than would paint chips, which would be in suspension only during conditions of heavy turbulence in the torus and which would promptly sink upon a lessening of turbulence. This phenomenon would be due primarily to the relative geometries of the particles, since both have a higher density than water. (Note that the General Electric report of Reference 1 conservatively assumed that insulation fibers would remain in suspension until either trapped by the ring girders in the torus or drawn onto the ECCS stminers.)
- 2. With regard to heavy turbulence, the General Electric tepon of Reference 1 states that the high velocities needed to effect debris transport are substantially reduced by 30 seconds from the I
inception of the design basis LOCA blowdown scenario. From the Vermont Yankee FSAR, the design basis LOCA scenario assumes that Core Spray achieves rated flow at 30 seconds into the 6
_____ a
L
)
i
'1 event, while RHR achieves full flow at 40 seconds into the event. Thus, debris (paint or insulation) transported to the torus in the first 30 seconds would stay in suspension due to the j
violent turbulence of blowdown, but the probability of significant paint transpon to the torus in the l
first 30 seconds is very low due to the fact that the area of degraded paint is the upper portion of the drywell and sut of the flow path. After 30 seconds, transport velocities begin to subside and after several minutes turbulence in the torus would begin to subside, such that entrained debris would begin its descent to the bottom of the ti;us where the physical construction of the torus (see further discussion below regarding primary containment geometry) would also greatly hinder j
1 migration of debris to the ECCS intakes.
- 3. The physical geometry of Vermont Yankee's primary containment (see Figure 1) is not conducive to transpon of significant portions of dislodged paint to the ECCS intake screens.
Failed paint in the drywell would need to traverse a torturous path through equipment and structural members down to the proximity of the drywell floor, and then across the floor to one of the downcomers. The lower lip of the downcomers are approximately 9 inches above the drywell floor,.tnd the downcomer openings have a blast shield at their entance, making the path to the torus even mote tonurous. Since paint chips would not be within the high velocity LOCA flow stream, it would be expected that the majority of failed paint chips would remain on the drywell floor. Any paint chips that continued to be transponed would then enter one of the downcomers and enter the torus through the vent pipes below the torus water level. Due to the relatively 1se arrival of the paint chips into the torus, transport velocities would be diminishing and the majority of the chips would be expected to fall harmlessly to the bottom of the torus, as the density of the paint is higher than that of water. "Ihe RHR intakes are located away from the bottom of the torus at 37-1/2 degrees from venical centerline, while the Core Spray intakes are located 50 degrees from venical centerline (see Figure 3). The intakes also protrude into the torus water volume, away.
l from the torus wall. These design features would preclude drawing suction off the bottom of the torus where paint chips may be deposited. Further, the torus is divided into 16 bays, only 4 of 7
l r,
.t' which have ECCS intakes (see Figure 2). These bays are separated by ring girders, which block 29% of the cross-sectional water volume from bay to bay. Thus, transport of deposited paint chips from a bay without an intake to a bay with an intake would be minimal due to the expected sedimenuaioa rate caused by a reduction in transport velocities within the torus after only several minutesinto the LOCA scenario.
- 4. At Vermont Yankee, NUKON insulation exists only on the recirculation piping, which is located in the lower (spherical) portion of the drywell. On the other hand, the area of degraded topcoat is the upper (cylindrical) portion of the drywell. A pipe break in either region cou'd result in significant debris from either fibrous insulation damage, or the dislodging of a quantity ofloose paint from the upper drywell, but not both. A recirculation pipe break in the lower drywell could result in significant NUKON insulation damage (as analyzed in Reference 1) but would not directly impact the upper drywell paint. Conversely, a break of a main steam line in the upper drywell region could directly affect loose paint in this region, but main steam lines are not covered with NUKON insulation; thus, there could be paint debris but no gross failure of NUKON insulation.
- 5. Existing topcoat degradation could be accelerated by the harsh environment created by the LOCA, but as long as essentially all loosely adhering topcoat is removed by the start of the operating cycle (through manual scraping of wall surfaces during the refueling outage), it would be reasonable to assume that most of the remaining topcoat would remain intact long after the first 30 seconds into the LOCA. High temperature may be a cause of eventual topcoat failure, but would not be expected to occur within the first 30 seconds of heatup. Pressure changes are also of concern, especially depressurization. As per the Stone and Webster evaluation, physical paint system testing of record subjects the coating to a period of pressure increase and " soak-in" (whereby pressurized air and vapor finds its way through any pinholes in the topcoat to void areas) and then allows depressurization, which has the potential to pose a differential pressure acmss the paint and possibly cause separation. Per the Vermont Yankee FSAR, d well pressure spikes up to 8
1 -
approximately 43 psig, and then down to approximately 28 psig, within the first 30 seconds of the design basis LOCA scenario. Pressurization is over a short period of time, thus reducing the " soak in" time to pressurize any void spaces behind the topcoat and thus reducing the subsequent differential pressure during depressurization. Pressure drops off much more slowly after 30 seconds, until a decision is made to take further actions, such as initiation of containment spray, whereby pressure drops still further. In any case, as stated above, the transport mechanism from the drywell to the torus is greatly reduced after 30 seconds, making further paint degradation in the drywell of little cvsequence relative to ECCS pump operability. As for the torus topcoat, the Vermont Yankee FSAR shows that torus pressurization is less dramatic than drywell pressmization due to a design basis LOCA, and there is no rapid depressurization in the early stages of the event.
Any failure beyond several minutes duration would result in prcn:pt settling of dislodged paint chips due to a decrease in torus turbulence. This, combined with the physical construction of the torus (ring girders separating bays and intakes located up the sides of and protruding away from the torus wall) as discussed above, would preclude the possibility of a significant portion of the paint chips frorn ever reaching the ECCS intakes.
J
- 6. The ECCS strainers, even if covered with the insulation mat projected in the General Electric report of Reference 1, still have margin over Net Positive Suction Head Required (NPSHR) by the RHR and Core Spray pumps at full design flow rate. The RHR strainers were replaced in 1985 with much larger strainers (old 19.8 verses new 47.2 square feet per intake). The RHR pumps have a " clean screen" NPSH margin (defined as NPSH available minus NPSH required) of 7 ft.,
which is reduced to a margin of 3.55 ft. due to the addition of.5.9 cubic feet of NUKON insulation. The Core spray pumps have a " clean screen" NPSH margin of 10 ft., which is reduced to a margin of 3.70 ft, due to the addition of 2.3 cubic feet of NUKON insulation (calculated valures fmm Reference 18). Additionally, the General Electric calculations of Reference 1 are held to be conservative, in that the assumptions tock no credit for any settling of NUKON insulation l
l fibers in the torus whatsoever and no credit was taken for fibers that may pass through the stminer.
9
i l,
I
}
i It 1
l l Thus it can be seen that NPSH margin still exists to support RHR and Core Spray design flow 1,
rates, even with some amount of paint accompanying the pmposed amount ofinsulation blockage.
}
i l
- 7. During the present refueling outage at Vermont Yankee (March,1989), loose topcoat in the upper drywell and the torus air space was removed with metal scrapers. Inspection of the rfa es by YNSD and Vermont Yankee engineers reveals that this effort was highly successfun in removing essentially all loosely adhering topcoat. The upper drywell is the only area in primary containment exhibiting significant topcoat failure at present. The lower drywell topcoat appears to be in good condition. The torus air space has been undergoing some form of topcoat failure since early in plant life (as discussed earlier in this report) but such failure is progn:ssing at a much lower rate than the upper drywell. The lower portion of the torus under the water line) does not appear to be a experiencing much failure of top coat, based upon earlier inspection. The only area of concern at present with regard to gross paint failure is the upper drywell, which the most recent manual scraping effort has reduced to approximately 50% remaining topcoat stuface area overall.
This effon reduces the amount of paint that could become dislodged in the highly unlike event of a large-break LOCA. Recommendations are being made as a result of this report to ensure that a similar inspection and scraping eflon become part of each subsequent refueling outage.
I
- 8. As verified by the attached Stone and Webster evaluation, the existing primer in both the drywell and torus is well cured and exhibits excellent adhesion and durability. Additionally, the l
l Stone and Webster evaluation stated that the failure mode of Carbo Zinc 11 primer is in granular l
form, and it does not dislodge from e surface in sheets. Thus, any failure of the primer (considered unlikely due to its high level of observed adhesion and durability in the Vermont i
l Yankee drywell) is considered to be of no consequence to ECCS pump operability. Additionally, I
the primer provides adequate protection for the primary containment interior surfaces, and no I
reapplication of topcoat is required at this time.
l 10 l
_ _ -__m
l V. Conclusions The above evaluation dernonstrates that, based upon removal of all loosely adhering topcoat wherever possible at the beginning of the current operating cycle (manual scraping was performed during the refueling outage to eliminate any loosely adhering paint), no adverse impact will result i
with regard to ECCS operability due to paint loosened during a design basis LOCA scenario.
Further, the lack of topcoat in no way affects the condition of the primary containment interior metal surface. The existing primer,in its present condition, was found to be in excellent condition and provides sufficient protection of these surfaces.
I i
1 l
l
\\
1 l
l l
ll 1
l l
11
4 1
I i- - -w.
l%..
-- _. I
<w I.
l I
.t_
._J sw.
~
4
. r.-
l,, --
/
l),.#.
Y..
- su -m
/
R__
(
e a,amar4 s
~
I
\\
=.
i r
. ~...
- ...n.
__.__.y A
/
- ::.+.)._.
1::.y<'
~=
,l
.4 I
w y.
I
~O
-<--.2_
nu %. 2.E' "?' '
..i.%- - - - - - Na u ewo-B c
+
i, u
- r ~n...
u.waa,,,,
s n
~ ===y.
n,<<
. - - w 23,
e-,.,...,c y
.m.
r 4.
.u 4.
/
p 1 s m / -'~=~ e
\\
w _.,.
S. un:t,,,
.. w./
1 wr i
,L p.
%.... -. :,g n.
.t 4_
r
---.r=t
._c co -.
,,,m m -
3 d C3,.
- t DtfL f f t.8088-
.{',
l 4 gqg g p'
,w,
-==a-Fl* 6TVIP3 0%,
- .gr A6L 8AMCK&D4 0@)
e
- * +,,
ss ga.
- 4
.r
.ew-90'=4" --
Figure 1
)
i 1
j l
_____-___a
g
~.
$ Intake Location o=
FRR FRR ts N COFI SPPAY 14 15 12 COPI SPRAY outzPoot x,
m.
2 3'
3 8
4
,7
'S S
\\
tge l
l Figure 2 l
l 1
?,.
~
,1 4
4 '6' d
.g
]
a s
1 l
/
s
\\
S
~
\\
~
l
\\
\\
\\
A STRAINER O
TORUS A i
\\
'N 37i' RHR
,. 50* CORC SPRAY
-.}
STRAINER
~
i VIEW A-A
/
Figure 3
\\
4 VI. References
- 1. General Electric Report MDE-184-0885, DRF AOO-01713, Revision 1, " Effects of Fiberglass Insulation D:bris on Vermont Yankee ECCS Pump Performance", dated January,1986.
- 2. "Waier Sources for Iong Tenn Recirculation Cooling Following a less-of-Coolant-Accident",
Regulatory Guide 1.82, Revision 1, dated November,1985.
- 3. " Buoyancy, Transport, and Head less of Fibrous Reactor Insulation", NUREG/CR-2982, Revision 1, D.N. Brocard, Alden Research Laboratory, dated July,1983.
- 4. " Containment Emergency Sump Performance", NUREG-0897, Revision 1, A.W. Serkiz, dated October,1985.
- 5. " Hydraulic Performance of Pump Suction Inlets for Emergency Core Cooling Systems in Boiling Water Reactors", NUREG/CR-2772, M. Padmanabhan, Alden Research Laboratory, dated June,1982.
- 6. Memorandum MES 120/72, H.F. Brannan to A.M. Shepard, " Peeling of Paint in the Torus",
dated February 25,1972.
- 7. Memorandum ME 76/75, J.R. Hoffman to L.H. Heider, " Vermont Yankee Torus Paint l
Samples", dated March 3,1975.
- 8. Letter MEG 452/80, YAEC to UQ nors Associates Engineered Product, Inc., dated July 21, 1980;
Subject:
" Carbo Zinc 11 Paint".
- 9. Memorandum MEG 499/80, E.C. Biemiiler to R.G. DiMatteo, " Review of Test Data _
Carboline's Carbo Zinc 11 Coating System", dated August 4,1980.
10.YAEC Report #1409, "Drywell Temperature Evaluation", dated Febmary 1,1984.
- 11. Memorandum VYS 25/84, L.A. Tremblay to A.C. Kadak, dated February 17,1984.
- 12. Memorandum VYB 84/216, R.W. Burke to A.C. Kadak, dated June 7,1984.
- 13. Memorandum File 11.0, R.J. Gianfrancesco to W.L. Wittmer, dated January 9,1984.
- 14. Letter NVY 86-33, USNPC m VYNPC, " Inspection Report No. 50-271/85-40", dated Febmary 20,1986.
- 15. Letter File 2.1, D.A. Reid to R.J. Iodwick, "Ser ice Request-Drywell Paint", dated January 16,1986.
- 16. Memorandum OPVY 289/86, R.L. Smith to R.J. Iedwick, dated March 17,1986, with attached evaluation, Memorandum VYS 46/86, C. Hansen to S.R. Miller, dated March 11,1986.
i
- 17. Memorandum VYB 86/414, R.J. Lodwick to S.R. Miller, " Disposition of SR 86-08, Drywell f
Paint", dated May 21,1986.
- 18. Memorandum OPVY 250/89, R.L. Smith to D.A. Reid, "Drywell Paint Issue Resolution",
dated April 5,1989, with attached Memorandum VYS 32/89, L.A. Tremblay to S.R. Miller, dated March 31,1989, and attached Report, " Evaluation of Containment Paint Degradation Effects at Vermont Yankee", dated March,1989 (YAEC Report #1696).
15