ML20151U746
| ML20151U746 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 08/09/1988 |
| From: | Tucker H DUKE POWER CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NUDOCS 8808190238 | |
| Download: ML20151U746 (9) | |
Text
- _ _ _ _ - _ _ -
g DUKE POWER GOMPAhT P.O. BOX 33188 CitAHLOTrE. N.C. 28242 liALB. TUCKER mrs, sown (704)073 4531 ms enesamt moot:4m emoovenox August 9, 1988 U.S. Nuclear Regulatory Commission ATTENTION: Document Control Desk Washington, DC 20555
Subject:
McGuire Nuclear Station Docket No's:
50-369 and 50-370
Dear Sir:
Flow balance testing performed on the Unit 2 Hydrogen Skimmer System (VX) during the recently completed refueling outage determined that individual compartment ventilation flow rates stated in the FSAR could not be achieved.
The purpose of this letter is to document Duke's conclusion that the VX System remains capable of performing its design basis function and to describe planned actions to improve the performance of this system.
Background
The VX System has a single header that draws air f rom each of the containment subcompartments by the use of redundant fans located on eLher end of the header as shown on the atte.ched schematic. Either fan is capable of drawing the total design basis flows from the areas served.
However, the physical arrangement of the system (i.e. a fan on each end) is such that balance of the individual compartment flow rates is difficult to achieve when operating each fan independently.
Unit 2 Testing 2
Near the end of the Unit 2 EOC 4 refueling outage, a flow balance test was performed in an attempt to verify distribution and balance as needed while running either 2A or 2B f an.
Flow measurements were performed using a hand held velometer (air velocity meter), scanned across the intake. When the.FSAR values for individua1 compartment flow could not be achieved, Duke's Design Engineering Department was consulted to assist in evaluating the test.results.
On their recommendations, a more accurate flow hooi was brought in to repeat the tests. When test results did not demonstrate uniformly acceptable results and because the planned justification for continued operation (JCO) did not rely on flow numbers, the flow measurement tests were terminated This decision was based, in part, on the evaluation of thq operability of the VX System (see Attachment).
In the JC0 it was recognized that the maximum hydrogen concentration assumption used in the design of the VX System was more conservative than required by NUREG-0800, that the hydrogen generation source term had been decreased by 10CFR50.46 and subsequerit Edq analyses and that a f
Hydrogen Hitigation System had been added and was operable per Technical f Specificatlon.
8808190238 880809
{
PDR ADOCK 05000369 I
P PDC
Document Control Desk
]
Page 2 August 9, 1988 The method used for flow balancing VX was to first take "as found" flow measurements while running one fan, then the other.
Data was then evaluated.
to determine optimum damper positions.
It was observed that the "as left" damper positions were essentially the same as the pre-operational positions as.
evidenced by paint shadows on the damper actuators. Two campers (both serving the reactor head area) were initially found closed but were left in -a throttled (approximately 50% open) position.
Unit 1 Testing The Unit 1 VX System flow distribution was verified prior to fuel loading (Jan
- FEB '81).
The test method was different from the one used most recently on
?
the Unit 2 System. On Unit 1, both fans were run at the same time and dampers adjusted to provide approximately twice the 7SAR flows for each ccupartment.
Each manual actuaior was then secured in the "as left" position.
At some later time (assumed to be prior to initial criticality), painting on each valve body and actuator handle - lef t a shadow mark on the valve position indicator.
Each valve was verified to be in the same painted mark position during the Fall '87 refueling outage.
Current Operability L
As d,1scussed above, the current design of the VX System does not lend itself to flow balancing on an individual fan basis.
Attempts to balance the system on both units have demonstrated +. hat even though the total fan flow is higher than the minimum FSAR value, it is not possible to achieve the current minitaum FSAR flow for emch compartment for each fan.
For bath units there is reasonable assucance that all dampers are curre,ntly throttled to a
near-optimum position. Flow measured on Unit 2 established that only a few compartments have less than the currently analyzed minimum flows rates.
Duke considers that the additional analyses now being performed using a lower metal-water reaction 1,5%. (0.3% times a factor of 5 for conservatism) and a higher allowable concentration, 4% versus 3.5%, will demonstrate that the current flow rates ara acceptable.
Although it is recognized that the Hydrogen Hitigation (EHM) System cannot he relied upon as a substitute for the VX System, the EHM Joes provide added assurance that in the unlikely event of a design basis accident, tbc concentration of hydrogen gas Vuld not build up to detonatable concentrations.
A recent operabilfty test verified the EHM was fully l
operable on both units.
Planned Corrective Actions Analysis Analysis using revised hydrogen source term and allowable concentration will be completed by Augen 15, 1988.
Results of the analysis vill be made available to th.2 NRC.
I i
}
i Doc *ument Control Lesk Page 3 August 9, 1988 Additional Flow Balancing Unit 1 is scheduled to begin its EOC-5 refueling outage on October 14, 1988.
Prior to restart, a flow balance will be conducted using one fan, then the other. Flow measurements will be conducted using the more accurate flow-hood technique.
Flow measurements will be done for Unit 2's B train during the next outage involving cold shutdown conditions.
A hardware toodification is under evah. tlon that would address the current difficulty of individual train flow bahneing.
A.a envisioned, such a modification would cross-connect the suction p2 ping to each fan.
If possible, this modliication will be implemented on eech enit during its next refueling outage.
Summary It is Duke Power's conclusion that the VN Syr, tem for each unit is currently balanced in a near-optimum position ond that revised analysis will demonstrate that currently measured flows are sufficient to prevent the buildup of excess quantities of hydrogen. Therefore, the YX System ean be considered operable.
i Very truly yours, et I V.
Hal B. Tucker RLG/600/ bhp Attachment xc:
Dr. J. Nelson Graco Regional Administrator Regi.on II U.S. Nuclear Regulatory Commission 3b3 a
rg Mr. W. T. Orders NR0 Resident Inspector
{
McGuire Nuclear Station 1
I
)
e
@ ox t'0 0 Ec @ m P2tw y N - - - - -
MCGUIRE_ COMPLIANCE GROUP TEL: 875-4000-2523 Aug 08,88 14:50 No.008 P.02 l
I e
l I f 84 fla*i lis 8 1 m
lg=l$! I l !
gl81 I lIlINIl ! ! !
"l i
~
< ! n lx..
_ _,l
=.
i/
.m i mis
'iD"R i i
i M !l g!
l I-o<h 7
E lJ
~
ij p-oo-y l - ll 4
-*4 l l
.s p$
w +.
3%
ll M
ss mI" y
.g q' s sw g k" M
A
%t-e.
m J
-m g
-.a.
l
\\
V v
MCGUIRE NUCLEAR STATION UNITS 1&2 Hydrogen Skinner System -- VX VioW Balance
[UGTIFICATION FOR CONTINUED OPERATION Summary flow balance testing performed on individual trains of the Unit 2 hydrogen skimmer system determined that compartment ventilation flow rates stated in the FSAR cannot be gehieved.
Even with the system in a degraded condition, adequate design basis features are in place to mitigate the consequences of hydrogen produced during a design J
basis accident.
Although the hydrogen skinner system was designed to reduce hydroger.
concentration in the dead-ended compartments during a design basis several developments concerning hydrogen have occurred since event,ime which permit safe station operatic 1 with a degraded that t These developments include installation of hydrogen skimmer system.
a hydrogen mitigation oystem, which is redunde.nt to the skimmer This mitigation system will prevent accumulation of system.
hydrogen, thereby providing adequate protection for continued safe operation of both station units.
References
- 1) Probism Investigation Report PIR 2-M88-0186
- 2) Test Procedure TT/2/A/9100/217
- 3) F5AR Section 6.6
- 4) 3ER Supplement 2
- 6) Calculation Mcc 1211.00-00-0027 Evaluation of Flow Requirements Testing of the Unit 2 hydrogen skimmer system for flow balance determined several compartment flow rates to be less than values stated in Table 6.6.2-1 of the FSAR.
The problem was documented on reference 1, and a design evaluation was initiated.
Reference 6 contains the original design basis re @ irements and calculations for the shirmor system.
The values for compartment ventilation flow rates were determined from a conservative analysis '
based on Azc safety Guide 7, dated March 10, 1971.
The original calculations of hydrogen evolution were based on a five(5) percent sirconium-water reaction, while current regulatory
t V
V 4
McGuire Units 1&2 VX System Flow Balance Jc0 Page 2 in conjunction with plant-specific ECCS analyses, guicance (ref. 5),
The prescribes a much lower olad reaction of only 1.5 percent.
prescribed value contains the required margin of safety of five(5) for added conservatism.
The values of compardnent flow rotes presented in the FSAR are also based on a dead-ended ecmpartment hydrogen concentration upper limit of 3.5 volume percent.
AEC Safety Guide 7 considered 4.0 volume percent as an acceptable upper limit, but 3.5 volume percent used in the original design for a margin of safety.
Based on use of the hydrogen recombiners to limit the upper containment waa minimum compartment j
hydrogen concentration to 3.0 volume percent, steady state ventilation flow rates were determined for maintaining The 4.0 i
an upper limit of 4.0 volume percent in the compartments.
volume percent criterion is also consistent with current regulatory i
Revised values do not take guidance contained in reference 5.
credit for reduction of the hydrogen source terms, and thtrifore Minimum flow rates have been revised only to remain convervative.
increase the allowable compartment concentration from 3.5 to 4.0 volume percent.
Comparison of Test and Design Data System flow balance data for Unit 2 does not meet the required minimum flow rates for an upper limit hydrogen concentration of 4.0 Although in a slightly degraded condition, the volume perceAt.
system will perform its fundamental safety function, and in conjunction with the hydrogen mitigation system, will limit hydrogen concentration in the dead-ended compartmants following a design basis less-of-coolant accident.
The attached tabulation compares compartment flow rates for
- 1) the nominal system design values, 2) minimum four(4) cases:
flow rates for 3.5 volume percent, 31 minimum flow rates for 4.0 volume percent, and 4) measured flow rates for system 2A, The compartment numbers correspond to FSAR Table 6.6.2-1.
1
_ _. ~, - _.. _ -... _... -,
i V
%/
21cGuire Unita 1&2
)
VX System Flow Balance JC0 Page 3 Compartment Ventilation Flow Rates i
compartment Desian 3.5 v/o 4.0 v/o Train 2A 1
85 85 43 171 2
85 85 43 60 3
64 64 32 206 4
14 14 7
83 5
14 14 7
65 6
64 64 32 68 7
52 52 26 60 1
g a
e a
e 9
564 477 239 278 10A 441 355 178 223 1
10B 441 355 178 763 100 442 355 178 1112 j
100 442 355 174 167 11 442 355 178 160 Total 3150 2630 1319 3416
- The flow requirement for corsartment number 8 was evenly distributed among compart: cents 9, 10A-D, and 11 in the original design due to the : physical configuration.
A specific vale.e cannot be measured.
4 i
j Q
%/
McGuire Units 1&2 VX System Flow Balancc JCO Page 4 Eval _uation of Safety Significance 10CFR50.46 requires that the amount of fuel element claddin that chemically reacts with watar or steam shall not exceed one(g) 1 percent of the total amount of Eirconium alloy in the reactor.
In accordance with this requirement, the McGuire design basis analysis indicates that the total metal / water reaction is less than 0.3 percent for all breaks.
Standard Review Plan (NURZG-600), section 6.2.5, requires that the hydrogen control and mitigation system be designed for five times the amount of hydrogen released in the F8AR analysis of a DBA assuming at least one train of ECCS operable.
For an accident of this type, virtually no radioactivity would be released from the fuel.
As stated in NUREG-800, a lower flammability limit of 4.0 volume percent hydrogen in air or air-steam atmospheres is well established and is adequately conservative.
Research has shown that relatively low levels of turbulence in containment promote sufficient mixing such that stratification of hydrogen will be minimised(EPRI NP-2669).
Based on the FSAR analysis of hydrogen Production in a postulated DBA, with no hydrogen control measures at all, it would be approximately eight days before the lower flammability limit was exceeded.
The McGuire hydrogen skimmer system was included as part of the original plant desgn basis due to a concern that hydrogen created in containment by accident conditions would tend to accumulate at the high points in lower containment.
There have been several developments concerning contcinment hydrogen which occurred since that timet
- 1) The containment hydrogen mitigation system was installed at McGuire.
This system consists of redundant sets of hydrogen i
igniters located throughout containment, including the tops of the steam generator and pressuriser enclosures.
These igniters prevent accumulation of hydrogen by causing it to burn at low concentrations.
Because the ignitors in the pressurizer and steam generator er. closures are 1ccated clone to the hydrogen skimmer fan suction, the presence of the ignitsra eliminates concerns over hydrogen skimmer fan eporability by performing the same function.
- 2) Sensitivity analyses of containment response to hydrogen burning performed using the CLAS:X code showed th&t the presence of the hydrogen skimmer fan had no effect on containment roeponse.
t
McGuire Units 1&A /
v VX System Flow Balance JCO Page 5 Conclusion The hydrogen mitigation system, consisting of hydrogen igniters and the hydrogen skimmer system in its current condition, are sufficient to insure that the design basis for applicable hydrogen producing events are met and therefore the hydrogen skimer system is considered operable,
)
........,,.Ad &
.t..sda
.. i..d 2(?
..t., r. v.a f%
VU y
Date: 7-TE'N Approved By:
b 4
e 9