IR 05000320/1978020
| ML19220B423 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/26/1978 |
| From: | Caphton D, Higgins J, Wong H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML19220B417 | List: |
| References | |
| 50-320-78-20, NUDOCS 7904260013 | |
| Download: ML19220B423 (10) | |
Text
-
..
U.S. NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT Region I Report No.
50-320/78-20 Docket No.
50-320 License No.
OPR-73 Priority
--
Category C
_
Licensee:
Metrocolitan Edison Ccmpany P. O. Box 542
.
Readina, Pennsylvania 19603 Facility Name:
Three Mile Island Nuclear Station, s.;it 2 Inspection at:
Middletown, Pennsylvania Inspection conducted:
May 10-12, 1978 Inspectors:
AM re'-7
[
Reactor / Inspector
/date signed J.C.HigAg's
.47 sierM
- f
-
H. J. Wong, Engineer / Intern dateesigned-g)
date signed 7 3h/7[
Approved by:
t/f 1e G D. L. CapTiton, Chief, Nuclear Support
/date' signed Section No. 1, RO&NS Branch Insoection Se= mary:
Insoection on May 10-12,1978 (Recort No. 50-320/78-20)_
Routine, unanno"nced inspection of containment integrated leak Areas Insoected:
rate test report, local leak rate testing and containment systems surveillance.
The inspection involved 41 inspector-hours and 5 accompaniment-hours on site by
,
one NRC inspector and one intern.
Of the three areas inspected, no Items of Noncompliance were found in Results:
one area, three. Items of Noncompliance were found in the second area (infraction -
failure to perform aieloc',: surveillances per the Technical Specification - para-graph 5.a; infraction - failure to implement surveillance procedures - paragraph 5.b; and, infraction - failure of an individual to monitor himself upon leaving a controlled area - paragraph 5.c) and one Item of Noncompliance was found in third area (infraction - failure to verify the equipment hatch closed and t%
. led - paragraph 3).
.
Region I Form 12
.
(Rev. April 77)
7r001
,
.
19u42600/3
.
-
DETAILS 1.
Persons Contacted The balow techni al and supervisory level personnel were con-tacted.
M. Bezilla, Engineer
- J. Floyd, Operations Supervisor F. Hurve, Health Physics Foreman
- T. Mackey, Lead Mechanical Engineer W. Marshall, Operations Engineer S. Poje, Startup Engineer
- J. Seelinger, Unit Superintender.t of Technical Support R. Sieglitz, Maintenance Engineer R. Toole, Test Superintendent
- denotes those present at the exit interview.
The inspector also talked with and interviewed several other licensee employees including reactor operators, auxiliary operators and health physics technicians.
2.
Containment Integrated Leak Rate Test (CILRT)
a.
_ General TP 150/3
" Reactor Building Initial Leak Rate Test", dated December 28, 1977 TP 151/1
" Reactor Building Isolation Vaive Leak Test",
dated July 29, 1977
" Reactor Containment Building Integrated Leak Rate Test -
January 1978" - Report to NRC
"GPUSC Audit Report No. 77-21 - Audit of Startup and Test Program"
" Shift Test Engineer's Log Book" - December 31,1977 to January 5,1978 P
i
'ib 002
.
_
-
The above referenced procedures, reports, and logs were re-viewed for proper resolution of test changes and deficiencies, accurate data transcription, cal.ibration corrections, meeting of acceptance criteria, correctness of the analytical methods, and the recording of significant events during the test.
The
.
inspector also reviewed the results of containment displacement measurements taken during the perfomance of the CILRT, in an at;empt to quantitatively measure diurnal variations in the leak rate.
Due to the weather conditions during the CILRT, neither the displacement measurements nor the leak rate measure-ments reflected noticeable diurnal effects.
The data, however, was useful and fulfilled the comitments made by Metropolitan Edison to the NRC in this area.
The inspector has no further questions at tnis time, b.
Satisfactory CILRT
'
The inspector independently calculated instrument calibration corrections contained air mass values, leakage rates and confidence intervals.
The results of these calculations sumarizea in Table 1, agree substantially with the licensee's values anc indicate that the CILRT was satisfactorily completed.
3.
Containment Integrity Technical. specification (TS) 4.6.1.1 requires that primary contain-ment integrity be demonstrated at least once per 31 days by verifying certain penetrations aTid valves are closed and~ that the equipment hatch is closed and sealed.
Surveillance Procedure 2301-M8, Rev.
4, dated April 18, 1978, " Containment Integrity Verficiation" performs this verification for penetrations and valves, howevcr, not for the equipment hatch.
The last verification of equicment hatch closure took place on February 9, 1978 using Rev. 2 of 2301-
.
M8. Failure to conduct this surveillance from February 9,1978 to May 12,1978 is an Item of Noncompliance (320/78-20-01).
The licensee stated that, althougn the surveillance had not been con-ducted, the equipment hatch had not been removed 6ering this time period.
'
TS Table 3.6-1 permits some non-automatic containment isolation valves (CIV) to remain open on an intemittant basis provided they are under administrative control.
The licensee stated that his adninistrative control for these valves when opened was via the locked valve list, which is maintained in the control room, and
& 003
.
r
-
--
requires signatures opening and closing and has periodic shift supervisor review.
inspector noted that three valves (CF-V 145, CF-V 146 and MU-V 330, were not included on the list.
The licensee took action to add these valves and the inspector had no further questions at this time.
4.
'
a.
General The inspector reviewed the licensee's surveillance procedure for conducting Tyoe B and C local leak rate tests (LLRT's), SP 2313-R7 titled " Reactor Building Local Leak Rate Testing".
Rev. O, dated January 10, 1978.
The inspector also reviewed the preoperational test procedure, TP 151/l, used to conduct the initial LLRT's and the results.
The TS and Appendix J to CFR 50 limit on Type B and C tests is 0.60 La or 0.078%/ Day.
The licensee's preoperational test recults indicated a total Type B and C leakage of 0.19 La.
The inspector had (;mments delineated in paragra-S _ b through h below on SP 2313-R7, which are collectiv., aesignated as unresolved item (320/78-20-02).
b.
Boundary Valve Leakage During the conduct of Type C LLRT's indications of high leak-age may be the result test boundary valve leakege vice CIV leakage. The licensee's procedure attempts to quantify this leakage using one of two methods.
In Method A inaccurate results could be obtained if the two readings are not taken under identical conditions.
Method B is non-conservative unless the difference between the upstream flowmeter and down-stream flowmater obtained in the first step (when positive) is added to the CIV leakage obtained in the second step.
c.
Stabilization Inaccurate and possibly nonconservative results could be ob-tained if flowmeter readings are taken prior to both pressure and flow stabilization.
When large test volumes are involved, stabilization can be difficult to obtain and the average of multiple readings may need to be used.
7U CCd
.
-
.
d.
Tyoe B and C Leakage Total Both TS and App J require that the combined leakage rate of all penetrations and valves subject to Type B and C tests, when pressurized to Pa be,< 0.60 La.
The licensee's procedtre does not include all required Type B and C tests in the total
_
for comparison with the acceptance criteria, as delineated below:
(1) The overall airlock leakage from SP-2303-SA2 is not incl uded; (2) The airlock seal leakage form Appendix B to SP 2311-5 is not included.
Note, this leakage is typically measured at 10 psig and must be converted to equivalent leakage at Pa before addition into the total; and, (3) The leakage of any valve or penetration measured after maintenance must be used to update to Current Type B and C total.
e.
Instrumentation The licensee's procedure does not call for instrumentation to be traceable to nationally recognized standards (e.g. NBS standards) as required by 10 CFR 50 Appendix B (Criterion XII)
and the Metropolitan Edison QA Program for Operations (FSAR paragraph 17.2.17).
In fact the procedure implies that flow-meters will only have generic calibration curves supplied by the manufacturer.
f.
Rotameter Operation The procedure gives no guidance for operation of the test rotameter or the LLRT test panel.
Also, if vent va7ve V-9 in the bypass line around the flowmeter is not opened and the bypass valves leak, nonconservative readings would result.
g.
Superseded Items SP 2313-R7 currently refers to several items which appa :ntly have been superseded:
i h
.
_
-
_
_
(1)
Step 5.2 refers to a drawing which is not included; (2)
which are not in use currently; and, Step 7.3 o (3)
Enclosure 3 is labelled with TCN's that do not h.
Minimum Sensitivity
.
minimum sensitivity values for flowmeters.The pr urer's measured by other methods also have minimum sensitivities Leakage rates should be used in the Type B and C total to ensure cons results.
For the pressure decay method, minimum sensitivities can be calculated using the test volume, test time interval and one-half the smallest division on the pressure instrume 5.
Airlock Tes_ ting
.
.
a.
Seal Leakage Ts 4.6.1.3.a states that:
"Each containment air lock shall be demonstrated a.
used for multiple entries, then at least once hours, by verifying 10.01 La seal leakage by pressure decay when the volume between the door seals is pressur
'
ized to 2,10 psig, for at least 15 minutes."
-
2311-5 and is contained in Appendix B.The licensee's February 24, 1978 and TCN 2-78-292 was issued on March 11 Rev. 1 was issued on 1978.
Rev. 2 was thac issued on A parated the TCN items permanently.pril 18, 1978 and incor-
,
the test by maintaining pressure on tacBoth revisions performed
'
decay to O psig.at 10 + 1 psig and then measuring the time for p entries,,this pressure decay time was measured as follo
!
I
.
!
V6 006
,
.
_
l l
-
-
.
3/19 cuter personnel door 23 sec outer equipment door 47 sec 4/7 both outer doors 60 sec 4/22 both outer doors 30 sec This is an item of nonce tpliance (320/78-20-03) with TS 4.6.1.3.a in that the seal leakage was measured with the volume not pressurized to > 10 psig and using a time interval of less than 15 minutes.,
The licensee further informed the inspector that the seal manufacturer's maximum recommended test pressure for the seals is 10 psig.
Hence the only possible successful pressure decay test with the current TS would be one that demonstrated zero leakage or perfect seals.
The licensee modific.d his test procedure to more fully meet the intent of the TS and initiated a TS change request.
The current test being used to demonstrate seal integrity uses a retameter type flowmeter, is run at 10 psig for 15 minutes and has a leakage acceptance criterion of
< 0.01 La seal leakage.
b.
Procedure Imolementation On several instances, as detailed below, the licensee failed to implement his surveillance procedures for airlock door seal testing, as required by TS 6.8.1.
ance (320/78-20-04).
This is an item of noncompli-(1) SP 2311-5 was performed on March 13, March 18, March 22, March 26 and April 13, 1978 without utilizing TCN 2-78-P92.
(2) SP 2311-5 was performed on March
'9, April 10 and April 18 without recording the data required by step 3 of Appendix "B", namely the time for the seal pressure to decrease to'0 psig.
c.
Test Witness On May 11,1978, the inspector accompanied an operator while performing three airlock dcor seal leak tests per SP 2311-5, Appendix "B", Rev. 2.
The test successfully passed on both inner airlock doors and had a pressure drop time of 12 seconds from 10 psig to 0 psig on the outer personnel door, indicating a leakage slightly less than 0.01 La.
i i
IU TC7
'
-
_
.
The inspector stated that had pressure been maintained at > 10 ps;2 for the duration of the test, leakage would clearly have been > 0.01 La.
Repairs were made to the door seals and a pressure decay test run from 10 psig to 5 psig with a measured leakage significantly < 0.01 La.
Since this test still did not fully meet the intent of the TS, the licensee again modified his procedure, as described in paragraph 5.a above.
In order to perform the seal leak test on the outer equipment hatch door, the operator went from the Reactor Building (a controlled area), through the two equipment hatch doors and then outside of the outer door (a clean area). The operator did not monitor either himself, or the test air battle and hoses for contamination upon leaving the controlled area.
This monitoring is required by step 5.3.2 of Health Physics Procedure 1612 "Monitorina for Personnel Contamination" and by T5 6.11, which states the Health Physics procedures must be adhered to, and is therefore an item of noncompliance (320/
78-20-05).
When the pressure decay test is used to measure airlock seal leakage, the pressure source is isolated from the seals by a single root valve.
If this root valve leaks by (as as observed during the May 11, 1978 testing) nonconservative results will be obtained.
This item is unresolved pending procedural corrections (320/78-20-06).
This comment is applicable whenever pressure decay is' used as a measurement of leakage rates.
Additionally, the inspector noted that for every door tested, the operator opened the door and passed through immediately af ter completing the test, thus defeating one of the test purposes (to ensure satisfactory door closure).
Due to test plug arrangement not all four doors can be shot and then tested, but three of the four doors can b, so tested.
T'li s item is unresolved pending procedural guidance in this area (320/78-20 07).
-
d.
Intearated Airlock Test The integrated airlock leak test is performed at Pa = 56.2 psig using Sp 2303-SA2.
Ti.is procedure measures leakage
.
's t CC4
_
-
-
.
using a flowneter rig, with a bypass line around the ficw-meter for tast fill.
If this bypass valve leaks during the test, the readings of the flowmeter will be nonconservative.
This item is unresolved (320/78-20-08).
6.
Miscellaneous Containment Systems Surveillance The inspector reviewed the licensee's method of complying with Technical Specifications for containment purge valve openings, containment air telperature, containment pressure and CIV opera-bility.
This review included:
SP 2301-51, Shift and Daily Checks completed-on April 17,
--
1978 to April 21, 1978.
Reactor Building Purge Valve Time Log.
--
Draft AP 1042 for surveillance of CIV cycle time and LLRT
--
after maintenance.
No items of noncompliance were identified.
7.
Unresolved Items Items about which more information is required to determine acceptability are considered unresolved.
Paragraphs 4, 5.c and 5.d of this report contain unresolved items.
8.
Exit Interview At the inspection's end the inspectors held a meeting (see Detail 1 for attendees) to discuss the inspection scope and findings.
The Items of Noncompliance and unresolved items were identified.
's(5 009
.
O
.
.
TABLE 1
.
-
January 1978 - CILRT Conducted at 56.2 psto at THI Unit 2
.
Item Acceptance Criteria Reported Results Inspector's Findings 0.0887%/ Day
'
O.0887%/ Day 1.
CILRT teak Rate by
--
Mass Point Technique 2.
Local Leak Rate Test
--
0.0000009%/ Day 0.00009%/ Day Additions (Type B&C)
0.0887%/ Day 0.0888%/ Day 3.
Total Containment Leak nate
--
(Mass Point)
4.
Upper 95% Confidence Level 0.75 La = 0.0975%/ Day 0.0947%/ Day 0.0952%/ Day on Leak Rate (Mass Point)
5.
Supplemental Verification
< 25% La 10%
10%
Test Difference
,
.
i O
r*)
,
e
.