ML20238F340
| ML20238F340 | |
| Person / Time | |
|---|---|
| Site: | North Anna  |
| Issue date: | 09/11/1987 |
| From: | Stewart W VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.) |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| 87-288B, NUDOCS 8709160091 | |
| Download: ML20238F340 (19) | |
Text
VIHOINIA EI.I crHIC AND POWI:H COMPANY It senxonn, Vi noix A cocon w,L. stew ur
% arm Pennipe=y wi cona m ore "'""
September 11, 1987 U. S. Nuclear Regulatory Commission Serial No. 87-288B Attention: Document Control Desk NAPS /JHL Washington, D. C. 20555 Docket Nos. 50-338 50-339 License Nos. NPF-4 NPF-7 I
Centlemen:
VIRGINIA ELECTRIC AND POWER COMPANY NORTil ANNA POWER STATION UNITS 1 AND 2 CONTROL ROOM llABITABILITY SURVEY We have reviewed your letter dated May 4, 1987 pertaining to the control room habitability survey that was conducted by the NRC staff and NRC consultants from Argonne National Laboratory. The enclosures to this letter provide the additional responses to the findings that were not addressed in our letter dated June 8, 1987, Serial No.87-288. Our lutter dated July 30, 1987, Serial No. 87-288A notified you that we would require additional time to submit these additional responses. A subsequent conversation between Leon Engle, NRC and George Pannell, Virginia Power established a submittal date of September 11, 1987.
If you have any questions, please contact me.
Very truly,yours,
)a( :.y g W. L. Stewart Enclosures cc: U. S. Nuclear Regulato'.y Commission 101 Marietta Screet, N. W.
Suite 2900 Atlanta, Georgia 30323 Mr. J. L. Caldwell @
NRC Senior Resident Inspector North Anna Power Station k
P $$f 870922 0500033g PDR
st j I
h, ENCLOSURE 1 RESPONSES TO THE NRC'S ADDITIONAL FINDINGS FROM ENCLOSURE 1 OF THE CONTROL ROOM HABITABILITY SURVEY i
l 1,
NRC FINDING 1:
Control room ventilation- system equipment should be instrumented with gauges j capable of reading the operating parameters of the system. The instruments ;
should not be . pegged, 'off-scale, or in need of repair. The fact that i instruments are on-scale during normal operation does not address the problem that during. the emergency mode of operation, a time when'you need system ,
l status, the instrumentation is incapable of providing the status. (Region II will request licensee to address). ;
l FINDING 1 RESPONSE We have reviewed the requirements for the control room ventilation system instruments and have concluded that'the existing instruments are adequate for i their intended purposes. Further, the instruments in question are periodically ;
, checked for' proper calibration under the Station's instrument calibration program and, at this time, no instruments are in need of repair, j The instruments include various pressure gauges which measure the pressure of the control room envelope in relation to the surroundh g areas._ -The instruments serve two purposes. First, the instruments are used for l surveillance testing an'd'must be accurate over the pressure range required by the surveillance test acceptance criteria ( 05 in. W.G.). Second the gauges .
are used to confirm that the control room envelope is being maintained at a positive pressure during the various emergency modes of operation of the control. room habitability system. The pressure gauges cover a range from'O to 0.5 in W.G. and are considered suitable for satisfying both of these purposes.
NRC FINDING 2:
All pressure boundary doors within the envelope should be marked as such and as technical specification doors. On two separate occasions, once prior to the survey - and once during the survey, control room envelope doors were found open breaking envelope. integrity. Since there are so many envelope doors and since it is. not obvious that some are envelope doors, markings similar to those on fire doors, should be incorporated. (Region II will request licensee to address).
FINDING 2' RESPONSE:
Certain boundary doors are being replaced as part of an upgrade program.
Following door replacement, each door will be marked as a " Pressure Boundary Door." Doors that are not being replaced have been marked as " Pressure Boundary Doors."
l 4
l l
NRC FINDING 3 ,
1 It is questionable whether the administration procedure, which is designed to limit access to the control room envelope through the use of a security guard during the one hour after initiation of the bottled air system, is at all workable. First, several guards, probably 8 or 9, would be required. Second, entrance and exit from the control room may be a necessity. It appears that the licensee should reevaluate the post-accident operation of the control room bottled air and ventilation systems. A new control room dose evaluation may be required if access is not limited. If access is limited, the present procedure must be modified to ensure limited access consistent with the present evaluation. (NRR address revised control room dose evaluation. Region II would request licensee to address the adequacy of the procedure).
FINDING 3 RESPONSE:
An engineering evaluation is being performed to determine the best method for limiting access to the control room following an accident so the control room can be maintained at a positive pressure with the bottled air system operating.
The control room dose vill also be reevaluated, if necessary. The engineering evaluation is scheduled to be completed by March 1988.
NRC FINDING 4:
Drawing 11715-FB-24E-12 shows fire dampers in the ductwork passing from the Unit 2 chiller room to the Unit 2 instrument rack room. However, no fire dampers are shown in the same ductwork which passes from the Unit 2 switchgear air conditioning room to the Unit 2 chiller room. Fire dampers would appear to be appropriate in both locations. Appendix R report requested a waiver for these dampers. Resolution of this request could not be determined. (Virginia Power should supply Region II verification of system design).
FINDING 4 RESPONSE:
Our review of drawing 11715-FB-24E-12 determined that there are no dampers in the ductwork between the Unit 2 air chiller room and the Unit 2 instrument rack room and therefore no drawing revision is necessary.
The Appendix R exemption referred to in the finding dealt with the absence of dampers in the ductwork passing from the Unit 2 chiller room to the Unit 2 emergency switchgear air conditioning room (emergency ventilation system).
This condition was evaluated it. NRC letter dated November 6,1986 and it was concluded that the subject penetration was acceptable for use in the fire boundary as described. An exemption from not having dampers in the ductwork hetween the Unit 2 chiller room and the Unit 2 instrument rack room is unnecessary since the duct is totally closed. The duct itself provides a barrier to prevent smoke from entering either room. In addition, the pressure in the ductwork is higher than the pressure outside the ductwork. Unfiltered air or smoke will not be able to infiltrate into the ductwork.
q.
NRC FINDING 5:
There seems to be a breakdown in the transformation of information in the updated FSAR, the plant's design basis, and the NRC staff's safety evaluations and actual plant operations.. As an example, based upon a review of the updated FSAR. system- and operating procedures it was determined that there.were 96 different combinations of control room and switchgear and relay room air handling = units and their respective' emergency filtration units. However, actual discussions with control room operators revealed that, under certain configurations, chilled water could not be supplied to the air hand 11nFs units without.the-chillers frequently tripping. This'was caused by partially loading the. chillers.. The system description and station procedures were silent on this problem._ They should have addressed this problem. The survey ' team observed. other procedures, situations, etc., where the information did not limit. system operations as it should have or they left it to the individual's experience to know how to perform the intended operation. System descriptions, station procedures, etc. should reflect actual operational configurations that are an . acceptable mode for the plant. Where limitations' exist, it should be noted. (Region II would request licensee to address the adequacy of control room system operating procedures).
FINDING 5 RESPONSE:
Operating procedures will be revised by January 1988 to indicate acceptable operating configurations for the control room ventilation system. Limitations will be noted as. appropriate.
NRC FINDING 6:
Measurements of- flow in the control room and switchgear ventilation system showed that the Unit I and Unit 2 control room emergency filter units had more flow exhausting than was supplied which indicates an inlenkage problem. . In some cases; exhaust was 50% greater than supply. For the switchgear and relay room emergency. filter units similar problems were observed. However, in the case of-Unit 2, supply was less than design. All other cases for emergency ventilation filter showed that supply was greater than design. During normal operation the switchgear room Unit I filter showed outleakage. In all operating conditions, the Unit 2 switchgear air handling units demonstrated backflow. Measurements were not made on the Unit 1 air handling units or the control room ventilation system air handling units since they were in the control room envelope. 'The Unit I damper on the emergency ventilation filter train for the switchgear room leaked about 140-200 cfm during normal operation.
Resolution of finding 2, on page 1 of this report, which requires prompt attention, should eliminate most of the concerns of the finding.
FINDING 6 RESPONSE:
Refer to the response to finding 2 in our letter dated June 8, 1987 (attached) which addresses emergency filter unit inleakage. In order to more accurately measure supply flow to the filter units, new test taps were placed in the suction ductwork'just upstream of the filter housing for all filter units. In addition, all four suction dampers were adjusted to bring the flow rates into compliance with Technical Specifications. After making these adjustments,
l l
tests showed that filter flow rates were not in excess of the Technical Specification limit of 1100 cfm. (The minor leakage has bean repaired). i Excess flow measurements at the exhaust are attributed to minor in-leakage l downstream of the filters and to errors inherent in the measuring method because of the inconsistent nature of flow at the exhaust plenum. In-leakage downstream of the filters is of no radiological concern since infiltration is from within the envelope.
NRC FINDING 7:
Based upon observations of the air handling units and the emergency filtration unit condition it appears that the air handling units may not be operating with the efficiency assumed in their deeign and that system flows are also outside the design basis. The licensee should test these systems and bring them within their operating basis. This may help to determine whether these units contributed to the problems seen at North Anna.
FINDING 7 RESPONSE:
The air handling units and the emergency filtration units have been tested in accordance with the Technical Specifications and are currently operating within their operating limits. See also the response to finding 2 in our letter of June 8, 1987 (attached).
NRC FINDING 8:
During normal operation in three different fan configurations, control room envelope supply measured 2500-2680 cfm while exhaust ranged from 1790-1870 cfm.
Thus, net pressurization flow was 700-820 cfm. In six different measurements of the system operating in its radiological mode of operation, system supply was measured as 2100-2300 cfm. System exhaust (Icakage) ranged from 15-25 cfm.
Thus, net pressurization flow of approximately 2100-2300 cfm was seen. In the normal operating mode the control room envelope was positive with respect to surrounding areas and pressure ranged form 0.032 inches water gauge to 0.215 inches of water gauge. The switchgear room was 0.05-0.085 inches positive with respect to the turbine building. In the radiological mode of operation pressures with respect to the turbine building were greater than 0.5 inches and pressure to surrounding areas were greater than 0.25 inches. Based upon this flow and pressure data it is hard to imagine how a three fold increase in flow results in an eight fold increase in pressure in five areas of the envelope.
We believe that additional makeup air is entering the envelope to produce such high pressure. It would appear that an additional evaluation of this interaction needs to be made by the licensee. (NRR will address).
FINDING 8 RESP 0NSE:
We have evaluated the observed data and have concluded that the increase in pressure is not indicative of additional air entering the envelope. The i observed data are to be expected given the relationship between differential pressure and flow rate (i.e., flow is proportional to the square root of the l differential pressure). Our evaluation of the possibility that additional nakeup air is entering the envelope during the radiological mode of operation
lc V .
was presented in our response to1 Finding 'l in our letter of June;1987 (attached), in which we concluded that no additional makeup air in entering the L. envelope.
'NRC FINDING 9:
Differential pressure should be measured lwith respect.to the two air chiller rooms and all differential pressure gauges should be labeled as to the specific-area that is being monitored.- Then, the control room could demonstrate that it was positive as described in the updated FSAR. (Region II will address).
FINDING 9 RESPONSE:
Refer to the response to finding 3 in our letter dated June 8, 1987 (attached) with regard to the need for differential pressure gauges for the air chiller rooms.
Differential pressure gauges will be labeled by October 1, 1987 to identify the specific area that is being monitored.
NRC FINDING 10:
Technical Specifications need to be modified to include: (NRR will address).
(a)' 30*C laboratory test of charcoal using ASTM D 3803-1979 test method.
(b) Acceptance criteria for laboratory test based upon Regulatory Guide 1.52' formula.
(c) In-place DOP and freon tests acceptance criteria set at 0.05%.
(d) Equipment qualifications temperature based upon the most sensitive equipment or instrumentation within the control room envelope necessary to shut the reactor down and measured at that location.
(e) Demonstration of a positive pressure appropriate for 1000 cfm of makeup flow through the filter unit if the bottled air must demonstrate a positive pressure of 0.05 inches water gauge at a flow rate of 340 cfm.
(f) The technical specification needs to address modes 5 and 6 because the system should be operable during these modes to protect control room operators in the event of a fuel hand 1h g accident.
FINDING 10 RESPONSE:
a, b, and c North Anna Technical Specifications require testing of charcoal in accordance with Regulatory Guide 1.52, Revision 2, March 1978. This procedure involves a representative sample of charcoal adscrber tested at 80*C at a relative humidity of 70%. The acceptable results are a methyl fodide penetration of s 1%. In-place testing of HEPA filters and charcoal adsorbers is also required to be in accordance with Regulatory Guide 1.52, except that a removal efficiency of 2 99% is specified.
._ - _ _ _ = _ _ _ . ._ _ _ _
The: findings propose . laboratory testing of charcoal at 30*C and 70% relative Lhumidity with an associated acceptance criterion.of 2'99% removal efficiency.-
In addition, in-place tecting 'of HEPA filters and charcoal adsorbers would specify an acceptance criterion of 99.95% efficiency for each one.-
Complying with these proposed changes would necessitate a major redesign of the control room emergency filtration system, use of different filter media, and more frequent system maintenance. While we recognize that such changes would-improve the capabilities of the system during a radiological event, we do not
. consider that the proposed changes would substantially increase the overall protection of the public health and safety. We base our views on the analysis contained i. the PSAR which takes credit for a purification factor of 90% for elemental iodine and 70% for methyl iodine. These criteria are conservative as compared to the 99% acceptance criteria in the present Technical Specifications.
Considering the potentially substantial cost of upgrading the system to comply with the proposed testing criteria and the absence of a significant safety benefit, we plan to continue performing surveillance testing in accordance with the current guidance in our Technical Specifications.
d The control. room ventilation system consists of four. one hundred percent
. capacity air handlers and a total of six mechanical chillers capable of providing chilled water to the air handlers. It is highly improbable that a complete loss of control room ventilation / cooling would occur. This event would require loss of six out of six chillers' and/or four out of four air handlers. A complete loss of ventilation or cooling capacity has never occurred at North Anna and such losses are considered to be highly improbable in the. future. In the unlikely event that four out of four air handlers and six out of six chillers were to fail, air circulation would stop and temperature would rise. The control room temperature sensor is located behind the back. boards in the computer room. The location of this instrument, in 'the upper regions of the envelope, is considered to be in a representative location during the event. In contrast the relay room, which is the most sensitive to safe shutdown, is in the lowest part of the envelope and will most likely be at the coolest ambient temperature. Technical Specifications currently require unit shutdown should ventilation and/or cooling be lost or should air temperature exceed 120'F. These requirements are sufficient to protect the temperature-sensitive equipment from extended periods above 120*F. Considering the location of the temperature sensor and the low likelihood of complete loss of cooling, we believe that existing temperature monitoring and controls are adequate.
e We have reviewed the Technical Specifications' surveillance requirements in question for the emergency filtration system and the bottled air pressurization system (Specifications 4.7.7.1(d)(3) and 4.7.7.2(b), respectively) and have concluded that the requirements are appropriate and sufficient for L demonstrating capability of the systems to perform their intended functions.
I The surveillance requirements are intended to demonstrate (1) the integrity of the control room envelope and (2) the operability of each system (i.e., ability of the system to provide a rated flow at a specific differential pressure between the envelope and the outside environment).
o
, 1.
L f.
Regarding envelope integrityg tiie bottled air pre $surization system test is the more rigorous test in that a greater positive pressure must ,he demonstrated i with: a lower pressurization flow. As both tests are perforked on the same j frequency, we see'no advantage of modifying- the emergency filtration system-acceptance- criteria 'as propor,ed, mClearly, if the bottled' air pressurization
- test is. acceptable, the control roem ~onvelope is intact and the emergency.
sfc filtration ' system with its greatet flow capacity will also be capable of !
l maintaining a positive envelope pressure.
Regarding'the demonstration of system operability, the tests- are considered
,g adequate .for demons): rating that the systems are capable of supplying l adequate l'
pressurization flow, and for detecting degradations'iu systbm performance as a result of improper system lineupo, equipment failures, et:4 Therefore, we see no need to. modify _the subject. Technical Specifications' surveillance .)
requirements. l f The Technical Specifications for the control room habitability systems do not address modes 5 and 6 because the systems are not required to protect the j control room . operators ~ in the event of a fuel handling accident. The fuel' l building is maintained at a negative pressure relative to the outside ,
atmosphere via the fuel building ventilation system Juring movement of '
irradiated fuel within the spent fuel pit or movement of heavy loads over the spent fuel pit -(see Technical ' Specification 3.9,17). The fuel building ventilation system- exhausts through a HEPA filter 3 and charcoal adsorber assembly. Therefore, the control room habitability systems.are not. required to i protect the operators in the event of a fuel, handling accident. l f
/
s 1
- v. 1
'*q; 7
NRC C0190!NTS ON 1.-OP-21.7,
[p g Main Control and Relay Room Emergency Ventilation:
p- a,
- 1. Step .3.2 states that a positive pressure must be maintained in the main control room and relay room at all times. Yet step 3.3 states that the emergency supply dampers should be closed if there is high activity or smoke in the turbine building. With these dampers closed it is stated that the alternate suction path is through the manway cover on the suction duct
'for the fan . Since this source is within the envelope there is no way for this envelope to be maintained positive. The licensee needs to address the implication of operation in this manner with respect to GDC-19,
- 2. The ;recautions and. limitation portion of this procedure should include a statement that the control room should be isolated in the event of a toxic gas challenge.
RESPONSE
"J 1. The procedure has been revised to specify that dampers should be closed only in the event noxious gas, smoke, or high activity is present in the turbine building. The latest revision of 1-OP-21.7 retains step 3.2 with regard to maintenance of a positive pressure in the control room and relay i room. The procedure further directs the initiation of the control room bottled air system as required to maintain control room pressure .05" W.G.
above turbine building pressure.
- 2. Although there is no toxic gas challenge that would impact the North Anna control room (Refer to,NUREG-0053 and supplements), the revised procedure, as stated above, does address shutting control room supply dampers in the event of noxious gas or smoke in the turbine building.
NRC COMMENT ON 1-PT-76.10A and B.
Laboratory Analysis for 1-HV-FL-8 and 1-HV-FL-9:
It should be clear from the procedures that the charcoal sample is drawn from the cell that has the oldest date of the three cells.
RESPONSE
1-PT-76.10A and B have been revised to ensure charcoal samples are drawn from the cell that has the oldest date.
l
_ - . _ - _ . . . _ . ~
n'
( ,
c.
l .
p , .
4 .y!
f NRC' CODMENT ON 1-PT-76.12A and B, ' . rf y Post-Maintenance Test on HEPA Filter for 1-HV ' an Q -HV-FL-9:
1 > \ ,\ >
l l ,
Section4.0ofitheprocedure'should.followSaction.10.5ofANSIN510-1980e.p, V '
- readings are to be within f_5%.
+ J
~
It does not. ,
-l:
l JRESPONSE:
i i
l 1-PT-76.12 A and B will be revised by October 15, 19 W to. follow the procedure specified in Section 10.5 of ANSI N510-197% fMeh is referenced in our
- Technical Specifications. t
, L f Based on a review of ANSI N510-1975 and ANSI N510-1980, be" conclude.that' there- '?/
is nosignificantdifferencebetweenthetwostandardsintheareaofconcern; h
/-
therefore, we do not intend to adopt the 1980 vercion. 7e 1 y NRC CODMENT ON 1-OP-21.9, -
i~
Control Ecom Bottled Air Pressurization System: ) e
[i 3
Step 4.2.5 should precede step 4.2.4. The alignment M ? uld be .tverified rdrh to opening the valve header. I ? f' .N v
y j.
l RESPON'.E:
f
\
The purpose of step. 4.2.5 is to independently verify.that step 4,2.4 was
. properly performed. To switch the two steps would , defect the purpose of the steps. No procedure revision is required.
- Other NRC Comments:
(1) If an SI signal is received and the emergency. filter train (s) are '
operating then these trains should be shutdown and bottled air system actuated. )
, o (2) Upon receipt of a low air bottle pressure.,i.e., at 200 psi, then the emergency filter trains should be placed in service. ,,O
/-
RESPONSE: 'l [
b
- 1) Procedures will be revised by October 15, 19$7 to ensure that if the 1
emergency filter trains are operating and an S1 signal is received, the ,
emergency filter trains will be shut down. The bottled air rystem actuates i automatically. / /
- 2) . Annunciator Response Procedure 1D-B8 requires , placing a strandby bank of air bottles in service on receipt of a low air bottle pressure alarm. On receipt of a second low air bottle pressu'e#aldrm, r the emergency filter trains are placed in service. ( A
't
,1 (
11 g
J l
t
, j
- 6 i g 9
(
. ', \
> ",y
- l. 3 9 +
a p
.) j I ENCLOSURE 2 ,
I( RESPONSES TO THE NRC'S FINDI!G FROM I ERCLOSURE 2 0F THE CONTROL RCOM HABI'NITY SURVEY
.,/
h1
,; , NRC FINDING 9.1.1:
r' b Procedures 1-0P-21.7, " Main Centrol Room and Relay Room Emergency Ventilation,"
J step 3.3 states that "dampsrs should be open unless t1'ere is high activity or ((
/ smoke in the turbine building. In this case, operate with the dampers closed.
If fans are operated with the damners closed, an alternate suction path must be provided by opeaing the manway cc cr on the suction duct for the fan." With the emergency ventf1stion system operating in this mode the control room can't be pressurized rcance suction cud discharge points are from the same area. This conflicts with Step 3.2 which requires that a pesitive pressure be maintained in the control room at all times. The procedure should be cicrified to reflect the ddsign basis.
[]
7 FIPDING 9.1.1 RESPONSE:
[> Refer to the response to Enclosure 1, Comment 1 on 1-0P-21.7.
l >
-l
!!RC FFHDING 9 1.2:
. l \
Precedure (-PT-76.4, " Control Room Bottled Air Preussri,ation System," step 4.5 l require? menttoring contzol room pressure differential Crydicators. The list of pressure differential indicators does not contain indicators for all the appropriate areas. The differential pressure from the relay rooms to chiller rooms is omitted. ,
FINDING 9.1.2 RESPONSE:
Differential pressure gauges are not needed to monitor the chiller rooms with respect to the relay rooms. Refer to our response to finding 3 in our letter dated Juar 8., 1987 (attached). No procedure revision is necessary.
NRC 77NDING 9.1.3:
.s Procedure 1-PT-76.4 step 4-6 says to inspect envelope leakage and seal any openings if the average envelope is less than 0.0$'in. water.
Since the indicators are measuring different boundaries, the action should be l req'rired for at>f Indication < 0.05 not the t vert.ge. (Region to address.)-
1 FiNPT# 9.1.3 RESPOEE:
/.
l-PT-76.4 was revised prior to perforecoce of the bottled air test to require inyjection of the envelope for leakage and the sealing of any openings if
, diffazemtial pressure gauge for any monitored boundary ind2ceton less than 0.05
( inch W. G.
l i , _
j f
I . t
NRC FINDING 9.1.4:'
Procedures 1-PT-76.10A and 1-PT-76.10B, " Laboratory Analysis for 1-HV-FL-B2-
.(1-HV-FL-9)" should say 31 days rather than 30 days in step 1.1. The procedure should also ' provide instructions on drawing the sample from the oldest of the
-three cells. (Region II should address this item.) ,
' FINDING 9.1.4 RESPONSE:
1-PT-76.10A and B '.. .e been revised to perform laboratory analysis every 31 ,
days and provide instructions for drawing the charcoal sample from the oldest of the three cells.
NRC FINDING 9.1.5: i Procedure 1-PT-76.12A and .(B), " Post-Maintenance Test' on HEPA Filter for 1-H-FL-B and (1-HV-FL-9)"; Section 4.0 should follow Section 10.5 of ANSI N 510-1975.
FINDING 9.1.5 RESPONSE:
Refer to response'to Enclosure 1,' Comments on Procedure 1-PT-76.12A and B.
-NRC FINDING 9.1.6:
Procedure 1-OP-21.9, " Control Room Bottled Air Pressurization System"; Step 4.2.3 appears-'to close the same valves which were opened in step 4.2.4. The procedure is not clearly written in that the intent of 4.2.3 is to close all
. valves except for the header that is to be in service and step 4.2.4 is to only open the header to be in service. The procedure should be rewritten to clarify these steps.
FINDING 9.1.6 RESPONSE:
Refer to the response to Enclosure 1, comments on Procedure 1-0P-21.9. 2 NRC FINDIEG 9.1.7:
Procedure 1-0P-21.9, " Control Room Bottled Air Pressurization System", ',
Attachment 2, page 1 says that there ere five malfunction lights, one of which ;
comes on when carbon monoxide level reacnes a safe breathing air limit. l The malfunction light should come on when carbon monoxide reaches an unsafe 1 level. (Region II should address this item.)
i 1
1
py
. l j
J FINDING 9.1.7 RESPONSE:
1-0P-21.9 has been revised to indicate that a malfunction light illuminating means carbon dioxide levels are reaching unsafe breathing levels.
NRC FINDING 9.1.8: !
Procedure 1-PT-76.1A and 1-PT-76.1B, " Control Room Emergency Ventilation System l
(1-HV-F-41) for'l-HV-FL-8 and Control Room Ventilation System (1-HV-F-42) for L 1-HV-FL-9'"; and appropriaterprocedures 'or 2-HV-FL-8 and 2-HV-FL-9 should have .l l- caution statements included to secure tw filter trains in'the event of an "SI" !
signal. This would allow the bottled air system to function as intended.
FINDING 9.1.8 RESPONSE: l Refer to response to "Other NRC Comments (1)" on a previous page.
.NRC FINDING 9.1.9: l In the event.of an "SI" signal AP-5-2 requires that the control room envelope j be isolated for the first hour. The procedure as described by station personnel is inadequate since isolation of the control room cannot be maintained. One security guard is stationed to see that the control room is isolated. However, there are at least eight entrances into the control room envelope and one security guard cannot limit access to the entire envelope. -l During a radiological event, access to the control room will be desired by certain station managers as was indicated during the plant survey exit meeting.
Prccedure AP-5-2 should be revised to recognize that access is required and whila it should be restricted it should not be denied. Allowing some access during the first hour will require the safety analysis for a design base accident to be revised in terms of GDC-19 exposure. (NRR and Region II should- i address.).
FINDING 9.1.9 RESPONSE: l Refer to Enclosure 1, Finding 3 response.
NRC FINDING 9.2.1: !
During a design basis accident there is no thyroid dose attributed since the control room is considered to be pressurized from an air bottia supply system during the first hour following the DBA. The bottled air system supplies about 340 cfm during this. time to pressurize the control room. The control room was also assumed to be administrative 1y isolated for the first hour. The licensee assumed no inbress-egress into the envelope and no leakage through dampers, door, etc. It is difficult to demonstrate that there is no inleakage into an envelope. Since it is doubtful that during an actual DBA that the control room could be completely sealed off to all personnel during the first hour and since it is also questionable that there is no inleakage across the control room envelope boundaries due to door seals, and damper leakage and duct inleakage.
A new safety analysis should be developed for North Anna taking into account
.I realistic valves for ingress-egress and envelope inleakage.
FINDING 9.2.1 RESPONSE:
Refer to Enclosure 1, Finding 3 response.
NRC FINDING 9.2.2:
FSAR Section 6.4 assumes that the control room and relay room are at the same pressure and therefore the envelope is positive to all adjacent areas. The way the AHU's are set up it is conceivable that certain portions of the envelope are at different pressures. The survey team did in fact determine that the HVAC rooms in the relay rooms are at a negative pressure with respect to the adjacent chiller rooms while the remainder of the envelope was positive.
Differential pressure gauges should be installed to determine dP between all adjacent areas to the envelope. The only place where such gauges are not located is between the relay room-HVAC equipment rooms, and the chiller rooms.
FINDING 9.2.2 RESPONSE:
Refer to the responses to finding ' and 3 in our letter dated June 8, 1987 (attached).
l l-NRC FINDING 9.2.3:
FSAR Section 9.4.1.3 indicates that the bottled air system will pressurize the control room to 0.10 in. of water. The III.D.3.4, Submittal and Technical Specifications both indicate 0.05 in. of water. The FSAR design basis should be consistent and reflected in the Technical Specifications and Plant Procedures.
FINDING 9.2.3 RESPONSE:
A UFSAR change has been initiated to differentiate between the operational limits and design of the control room bottled air system.
NRC FINDING 9.3.i:
The qualification temperature of control room instrumentation was determined to be 120*F. A Technical Specification limit should be written to verify that the 120*F is ;.ot exceeded. Sensors should be placed to measure the critical ins':rumentation and not control room ambient temperature, ffCDING 9.3.1 RESPONSE:
Refer to the response to Enclosure 1, Finding 10d.
l NRC FINDING 9.3.2:
The technical specifications require that the emergency filter trains maintain the control room at 0.04 in. W.G. while the air bottles are required to maintain the envelope at 0.05 in. W.G. The emergency filter trains, with one control room and one switchgear/ relay room train in service, supplies six times more air flow than the bottled air system. The survey team does not understand
how 340 cfm of bottled air can demonstrate the control room positive to 0.05 in. W.G. if it takes 1000 cfm per filter train to demonstrate 0.04 in. W.G.
The technical specifications should be revised to require the filter trains to l maintain the control room at 2 0.125 in. W.G. positive to all adjacent boundaries. (See Specification 4.7.7.1.d.3 and 4.7.7.2.b.)
FINDING 9.3.2 RESPONSE:
Refer to the response to Enclosure 1 Finding 10e.
NRC FINDINO 9.3.3:
HEPA filter and charcoal adsorber efficiencies are assumed to be 99.5% for DOP and freon testo in the NRC safety evaluation. The FSAR, Section 6.4.1.2 also states that the HEPA filters are capable of removing 99.5% DOP, but does not mention the halogenated hydrocarbons. Consistent with the assumptions utilized in the III.D.3.4. submittal Technical Specifications 4.7.7.1.e and 4.7.7.1.f should be revised to limit penetration during freon and DOP in-place testing to
< 0.05% (99.95% removal).
FSAR Section 6.4.1.2 should be revised to include in-place testing efficiency for the charcoal adsorbers.
FINDING 9.3.3 RESPONSE:
Refer to the response to Enclosure 1, Finding 10c.
NRC FINDING 9.3.4:
Present technical specifications have laboratory tests for charcoal conducted at 80*C. Charcoal in the control room ventilation system would never experience a temperature of 80*C. Charcoal tested at dO*C shows a greater adsorption capability than it would actually have at actual room temperatures of about 30'C. The utilization of the 80*C test to demonstrate charcoal capability is inappropriate because it results in regeneration of the charcoal.
The more appropriate test is 30'C and 70% R.H. provided that the filter train electrical heaters are always activated, otherwise the test should be performed at 95% R.H.
Technical specifications 4.7.7.1.b.2 and 4.7.7.1.c should be revised to perform the laboratory tests of charcoal at 30*C and 70% R.H. or 30*C at 95% R. H.
based on how filter train heaters are operated. Test method ASTM D3803-1979 should be used. Acceptance criteria should be 1% or 0.7% penetration based on relative humidity used in testing.
FINDING 9.3.4 RESPONSE:
Refer to the response to Enclosure 1, Finding 10a.
i I NRC FINDING 9.4.1:
l 1
l Air temperature was measured in the control room with both units operating at '
! full power. Temperature was recorded for both normal and emergency modes of operation. (Refer to Table 2.) The following is a summary of the data:
Overall temperature control seemed to work within design control parameters.
The area inside the main control panels is well ventilated with only a 2.2*F maximum differential temperature to the occupied areas of the control room.
There were areas within the control room that were much cooler than the rest of the control room. This would imply the need for some local balancing in the occupied area of the control room.
FINDING 9.4.1 RESPONSE:
Our review of the temperature profiles measured by the NRC in December 1986 indicates a maximum temperature difference of about three degrees for normal alignments between normally occupied areas of the Unit 1 and Unit 2 control rooms. We believe that this temperature profile is indicative of a properly balanced HVAC system.
NRC FINDING 9.4.2:
Differential pressure data was recorded from installed plant instrumentation.
Differential pressure data as indicated in Table 3, showed the control room more positive than required by technical specifications for all emergency modes of operation. The operation of the emergency air bottle system was not demonstrated during the visit but will be demonstrated during a subsequent visit. Based upon the flow data obtained, the review team believes that the chiller rooms, which are adjacent to the control room envelope, will be positive during operation of the bottled air system and would allow outside inleakage into the envelope. Although there is no installed dP gauges to monitor this boundary of the envelope, when the envelope doors were opened there was a significant in-rush of air into the envelepe during normal modes of operation.
Assessing the flows which pressurized the control room during normal operation and comparing these to the bottled air system flow rates, leads the review team to conclude that pressurizing the entire envelope with the bottled air system may be difficult.
The review team also believes that a larger range dP gauge would be appropriate since installed gauges are pegged during emergency filter train system operation.
FINDING 9.4.2 RESPONSE:
Following adjustments to the system pressure control valves, the bottled air test was successfully conducted on August 2, 1987. Because of concerns expressed by NRC observern, a third bank of 42 air bottles has been lined up to dump on an SI signal to provide additional air capacity margin. In addition, a fourth bank of bottles is available to be manually valved into service if needed. As stated in enclosure (1) in our response to NRC Finding 3, an engineering evaluation is being performed which will also examine the adequacy of the system and recommend modifications as necessary. For a discussion of
o,.-
g l? ,
L
.]
f' iflow rates and dP gauges,' refer to-responses to findings 1 and 3Lin our letter. !
dated June"8, 1987 (attached).
l 4
.i NRC FINDING 9.4.3:
i During the amergency mode of operation, each. door' leading out ~ of' the control 1 l> room envelope was monitored with smoke powder. ' Table 4 shows that the envelope- -
l Lis. positive to all adjacent areas when . the emergency. filter ' trains. are in !
service. Areas -within the envelope vary with respect to one another depending on which equipment is operating.
- FINDING 9.4.3' RESPONSE
- }
No response required.
~
NRC' FINDING 9.4.4:
l '. Filter train ~1-HV-FL-9 has 134-198.cfm of backflow through the filter train during normal modes of operation. ;
The filter train isolation valves should b'e checked to ensure proper seating. Duct' work, filter train housing, fan housing and flexible couplings should be inspected and repaired as needed.
- 2. Flow measurements on the inlets to filter trains 1-HV-FL-8 and 2--HV-FL-8 exceed the 1000 cfm + 10% criterion in the Technical Specif1 cations.- Flow measurements an thI outlet stacks of the filter trains showed excessive flow through all filter trains.
The inlet measurement ports for emergency filter trains are not located where'they will show all air flow through the filter train housings.
- 3. The portion of.the duct work running through the chiller room appears to be~
inducing air from the chiller room into the control room envelope.
Virginia Power has. drilled new test ports to better quantify the flow through this portion of the duct vork. Although they have reduced the differential from 3600-7000 cfm to about 600 cfn. There wuld still appear to be significant inleakage.
The' utility should test the duct by measuring the differential pressure in the duct to the' chiller room at the inside radius -of the duct elbow. to verify that duct pressure is higher than the chiller room ambient pressure.
A further test would be to pressurize the portion of ductwork outside the
- . envelope to determine inleakage.
FINDING 9.4.4 RESPONSE:
'1.
During the bottled air test attempted on 7-30-87, it was demonstrated that the isolation valves provided proper seating. In addition, successful completion of the bottled air test on 8-02-87 demonstrated that the control room pressure was maintained, resulting in verification of back leakage at an acceptable level.
l
= _ _ - _ _ _ - _ _ _ _ - .
- 2. Refer to the response to finding 2 in our letter dated June 8, 1987 (attached).
- 3. Refer to the response to finding 1 in our letter dated June 8, 1987 (attached).
NRC FINDING 9.5:
Virginia Power assumes that there is no thyroid dose due to the pressurized q control room with no outside air for the first hours of a DBA. After the first hour of the DBA, the containment is assumed to be at a sub-atmospheric condition. In order to assume zero filtration, the control room must have no ingress-egress and no duct leakage in the chiller room (Sec. 9.4.4, item 3)'and i all door seals must be tight especially in areas where envelope pressure may be negative due to air flow from adjacent area ventilation systems.
The review team believes that it is difficult for the licensee to demonstrate I zero unfiltered inleakage and zero ingress-egress. An evaluation needs to be conducted to determine realistic values for the inleakage term. Once these values are determined a new dose evaluation needs to be performed.
FINDING 9.5 RESPONSE:
s Refer to the response to Enclosure 1, Finding 3.
NRC FINDING 9.6:
There were no LER's associated with the loss of cooling to the control room envelope. The AHU for both units seem to be more than adequate. However, the technical specification Limit of 120*F needs some modification to ensure ,
adequate monitoring (see item 9.3.1). ,
FINDING 9.6 RESPONSE:
Refer to the response to Enclosure 1, Finding 10d.
ATTACHMENT Virginia Electric und Power Company Letter to NRC Dated Ji w 8, 1987
" Control Room Habitability Survey" I
h f,
. Attachment l !
I June 8,.1987 U. S. Nuclear R.egulatory Commission Serial No.87-288 Attention: Document' Control Desk NAPS /JHL Washington, D. C. 20555 Docket Nos. 50-338 50-339 License Nos. NPF-4 NPF-7 Gentlemen:
VIRGINIA ELECTRIC AND POWER COMPANY NORTH ANNA POWER STATION _ UNITS 1 AND 2 CONTROL ROOM HABITABILIlT SURVEY We have reviewed your letter dated May 4, 1987 pertaining to the control room habitability survey that was conducted by the NRC staff and NRC consultants from Argonne National Laboratory. The attachment to this letter provides the responses to the three findings that required prompt attention. The other findings identified in Enclosures 1 and 2 of your letter will be addressed by July 8, 1987.
If you have any questions, please contact me.
Very truly yours, s
( s\tl CyeD W. L. Stewart Attachment
, r)Jf O
[ hlT* D
cc: U. S. Nuclear Regulatory Cornission 101 Marietta Street. N. W.
Suite 2900 Atlanta, Georgia 30323 Mr. J. L. Caldwell NRC Senior Resident Inspector North Anna Power Station l
l l
l l
l
h -
E
,e AffACHMENT RESPONSES TO THE NRC'S. DOEDIATE CONCERNS FROM THE CONTROL ROOM HABITABILITY SURVEY i
NRC FINDING l
[ The flow' data for the.ductwork which passed from the Unit 2 switchgear and relay. room . air handling area through the Unit 2 chiller room and then back l .into the relay. room at the instrument rack'showed .a considerable amount of l
_inleakage as it passed through the chiller room. Since the. chiller room communicates with the. turbine building and is not part of. the control room it provides a contaminated path into the envelope.
envelope, Flow rate L
measurements and/or differential pressure measurements of the ductwork with respect to the chiller room should be made to determine whether there is inleakage. More extensive tests such as pressure decay tests may be-needed to L verify that there is or'is not inleakage into the ductwork.
RESPONSE
We have reviewed the flow measurement data that was obtained during the survey and summarized in the control room habitability survey report. We have determined that there is no air inleakage into the ductwork. We conclude that erroneous data was obtained because of inadequate test tap locations and spacing of the sampling holes.
An engineering- study . was performed to evaluat$ the possibility of air inleakage into the Unit 2 switchgear supply ductwork from the Unit 2 chiller room. .The study theorizes that air. is being educted from the Unit 2 chiller room through the seams in the ductwork. .It was assumed that. air inleakage of 4800 cfm exists. To accommodate the inleakage plus the difference between the normal supply and exhaust and-to obtain a pressure drop of 0.10 inches of water, a theoretical orifice size needed to be determined. Study assumptions include: 1) all individual leakages within the control room envelope car be approximated-by a single orifice sized to-accommodate the sum of all leakages,
- 2) in order to keep pressure constant, the sum of all leakages must be equal to the. inleakage plus the difference between the supply and exhaust, and 3) the orifice size does not change with a change in pressure.. The calculations in- the study determined that a 10 inch orifice would be needed to account for the overall leakage of the envelope.
Next, a delta-p was calculated using the orifice previously sized with two emergency fans started, delivering 2000 cfm additional flow to the envelope and with the normal supply and exhaust isolated. The calculated delta-p was 0.147 inches water gauge while the actual data showed pressure greater than 0.25 inches water gauge.
l 7 Next, a test was performed in which air handling units 6 and 7 were shutdown so that any eduction may be stopped and the only makeup being supplied by the s
h__________.. _ . _ _ . _ _ . _ _ _ _ . _ _ . - - . _ _ _ _ _ . - - - - - -
i l
l normal supply and exhaust. According to calculations in the study, a change in pressure of approximately 0.08 inches water gauge should have been ob se rved. Test results actually indicated no appreciable drop in pressure.
For eduction to take place, the static pressure (velocity induced) inside the duct must be less than the static pressure outside the duct. Static pressure data were obtained within the duct at two points and in'the chiller room. The pressures .in the duct were measured to be 29.86" Hg and 29.835" Hg. The
-static pressure in the chiller room was measured to be 29.76" Hg. Based on these measured ' values, nc eduction could be occurring. Additional data was taken at the moet probable location for reduced pressure (in accordance with the survey team recommendations), downstream of the elbow in the ductwork.
All measurements show that the pressure inside the duct is significantly greater than the chiller room. !
In addition, flow data were obtained from new test taps that are better suited for uniform flow characteristics. The original test taps that were used 3 during the survey were not ideally located and the spacing of the sample holes ';
was not uniform. The original test taps were located such that with back flow through leaking dampers and the effect of upstream obstructions reliable information could not be obtained. The test results show that no inleakage exists.
The study concludes that measured inleakage by eduction or through a change in :
pressure between the chiller room and the ductwork is not feasible. l NRC COMMENT Flow measurements upstream and downstteam of the filter units for the control room and switchgear and relay room showed flow rates outside the technical specification limit of 1000 cfm plus or minus 10%. Flow rates should be verified and, if outside the specification, brought within it.
RESPONSE
Licensee Event Report (LER)86-019 was submitted on January 9, 1987 to I describe this event. This LER indicates that flow rates were found to be outside their Technical Specification limits. Flow rates were brought back into Technical Specification limits by adjusting the modulating damper installed in the filter inlet ductwork.
NRC CODGENT Pressure data taken when the control room and switchgear and relay room ventilation systems were in their normal mode of operation showed differential pressures in the Unit I cable spreading room and the Unit I cable vault less
4 than 0.05 inches water gauge neg tive with respect to the control room while the chiller rooms, which are outside the control room envelope, appeared positive with respect to the envelope. These conditions occurred with a net pressurization flow of 700-820 cfm. Based upon these observations there is some co- -n as to whether the bottled air system could pressurize the envelope to 0.05 inches with a flow rate of 340 cfm. This needs to be evaluated and differential pressure gauges need to be installed to monitor the chiller rooms in relation to the adj acent switchgear and relay room air handling units area. All other areas adjacent to the control room are nonitored for differential pressure. A test of the bottled air syitem is scheduled for May 1987. The NRC staff intends to be in attendance to ,sensure flow rates and to verify differential pressure. The staff will provide you with the results of this visit. ,
RESPONSE
['
As a measure to ensure that the control room envelope remains at a positive pressure with respect to the chiller rooms, the block walls between the switchgear/ relay room and the fan rooms were removed. Differential pressure testing was performed and the control room envelope was at a positive pressure as compared to the chiller rooms. As an extra measure, the normal room ventilation fan in each chiller room will be reversed.
We believe that the bottled air system can pressurize the control roem envelope to 0.05 inches with a flow rate of 340 cfm. This test is currently scheduled to be performed in July 1987. Your staff will be contacted so they may obtain flow rate data and verify differential pressures.
We do not see the advantage of adding differential pressure gauges to monitor the chiller rooms with respect to the control room envelope and, therefore, do not intend to add them. Various tests have already been performed to ensure that the control room envelope is at a positive pressure as compared to the chiller rooms. In addition, with the removal of the block wall the cause for the low room pressure has been eliminated. Therefore, the switchgear room instrumentation can accurately represent the pressure in the fan rooms.
Likewise, the chiller room itself freely communicates with the turbine building and can be represented by existing turbine building differential pressure instruments.
O t
5
! g 8
l.
' gh4.
,i
() W -
} ,Cf-%n n
.s a
M p
8
__ _ _ _ _ _ _