ML20101S345
| ML20101S345 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/12/1996 |
| From: | Dekok D AFFILIATION NOT ASSIGNED |
| To: | NRC OFFICE OF ADMINISTRATION (ADM) |
| Shared Package | |
| ML20099A192 | List: |
| References | |
| FOIA-96-207 NUDOCS 9609030078 | |
| Download: ML20101S345 (2) | |
Text
.-
8 r
/
t David DeKok 113 Conoy St.
Harrisburg, Pa.17104
[
May 12,1996 fBEE00M OF INFORMATION Division of Freedom of Information ACf REQUEST Io I.A-9 4 c7 o 7(
and Publication Services tL g-f M Office of Admi:listration and Resources Management Nuclear Regulatory Commission Washington, D.C. 20555 Re: Freedom of Information request
+
Dear Sir or Madam:
Pursuant to the Freedom of Information Act, I hereby request copies of the following:
Reports of inspection conducted by the Atomic Energy Commission at General Public Utilities Corp.'s Three Mile Island Unit I for the following dates: April 7,1972, July 11-14,1972, and March 26-28, 1973. These are not in the Public Document Room.
--Reports from meetings between AEC and GPU personnel on May 1 and May 8,1973. These are not in the PDR.
-Any other documents in your national or regional files pertaining to Three Mile Island Unit i between 1966 and March 27,1979,which have not yet been placed in the PDR.
--Any correspondence between the AEC and Gilbert Associates, Inc.. of Reading, Pennsylvania, pertaining to TMI Unit i from the period 1966-March 27,1979. Gilbert was the architect / engineer of TMI Unit 1.
96C9030078 960815 OK 07 PDR
i t
--Any correspondence between the AEC and United Engineers and Contractors pertaining to TMI Unit I during the period 1966-March 27,1979. UAE was the general contractor for TMI Unit 1.
Pursuant to the Freedom of Information Act, I also request
" Representative of the News Media" status, which entitles me to a waiver of search fees and 100 pages of free copying. These documents are needed for a book I am writing on the history of General Public Utilities Corp. I am an established freelance writer, the author of one previous book. Unseen Danger:.4 Trxeedy afhople, Goveniment and the Centre //s Mhe/he was published in 1986 by University of Pennsylvania Press. In addition, I have been a newspaper journalist for 20 years.
Thank you for your attention to this request.
Sincerely, a
v 6k l
w_.
)
- -.... - = _. ~....
r l
THREE MILE ISLAND NUCLEAR STATION UNIT I REACTOR CONTAINMENT BUILDING INTEGRATED LEAK RATE TEST i
APRIL 1977 l
i l
l METROPOLITAN EDISON COMPANY SUBSIDIARY OF GENERAL PUBLIC UTILITIES CORPORATION
}.,c.
PREPARED BY k
L..
'R. E. Shirk Gilbert' Associates, Inc.
APPROVED BY M
/
2.e mue R.M.dihama Manager-Gener ion Engineering Metropolitan Edison Co.
b
~~
)
n ~
y e e-p)
V TABLE OF CONTENTS Section Item Title P_ age 1.0 SYNOPSIS 1
2.0 INTRODUCTION
3 3.0 ACCEPTANCE CRITERIA 4
4.0 TEST INSTRUMENTATION 5
4.1
SUMMARY
OF INSTRUMENTS 5
4.2 CALIBRATION CHECKS 7
4.3 INSTRUMENTATION PERFORMANCE 8
4.4 SYSTEMATIC ERROR ANALYSIS 8
4.5 SUPPLEMENTAL VERIFICATION
~12 5.0; TEST PROCEDURE 14 5.1 PREREQUISITES 14 5.2 GENERAL DISCUSSION 15 5.3 TEST PERFORMANCE 17 f
6.0 METHODS OF ANALYSIS 25 6.1 GENERAL DISCUSSION 25 6.2 STATISTICAL EVALUATION 27 7.0 DISCUSSION OF RESULTS 29 i
7.1 RESULTS AT Pa 29 7.2 SUPPLEMENTAL TEST RESULTS 30 i
8.0 TYPE B AND C LEAKAGE RATE HISTORIES 31
9.0 REFERENCES
32 APPENDICIES A.
REDUCED LEAKAGE RATE DATA B.
WEIGHT OF CONTAINMENT AIR AND AVERAGE CONTAINMENT I
TEMPERATURE VERSUS TIME C.
REPORT OF 1976 REFUELING R.B. LOCAL LEAK RATE TESTING 1
D.
REPORT OF 1977 REFUELING R.B. LOCAL LEAK RATE TESTING E.
REPORT OF MISCELLANEOUS LCCAL LEAK TESTING MARCH 1974 to APRIL 1977 Gdhert/Cammenesse
.a
1
~
1.0 SYNOPSIS The Three Mile Island Nuclear Station Unit I reactor containment building was subjected to a periodic integrated leak rate test during the period from April 16, 1977 to April 19, 1977.
The purpose of this test was to demonstrate the acceptability of the
~
building leakage rate at an internal pressure 50.6 psig (Pa)-
Testing was performed in accordance with the requirements of 10 CFR 50, Appendix J and ANSI N45.4-1972.
i The measured leakage rate based on the mass point method of analysis was found to be 0.042, percent by weight per day at 50.6 ps_ig.
The
)
leakage rate at the upper bound of the 95 percent confidence interval is 0.052 percent by weight per day which is well below the allowable leakage rate of 0.075. percent by weight per day at 50.6 psig.
\\
The final leakage rate of 0.042 percent by weight per day was 1
obtained after adjustments were made and the test was restarted.
The initial building Icakage rate indicated was in excess of 0.1 l
percent by weight per day.
The adjustments made consisted of I
tightening mechanical joints and packings.
Since the industrial cooler system was in operation during the integrated leak rate test, addition of the local leakage rate of j
the system isolation valves (RB-V2* and RB-V7) to the measured integrcted leakage rate must be considered. The combined local leakage rate of both these isolation valves was 0.007 percent by weight per day. The addition of this value increases the total integrated leakage rate to 0.049 percent by weight per day.
j Geert/Cammemmesah 1
1
The supplemental instrumentation verification at P, was 1.0 percent, well within the 25 percent requirement of 10 CFR 50, Appendix J, Section III A.3.b.
All testing was performed by Metropolitan Edison Company with the technical assistance of Gilbert Associates, Inc. Procedural and calculational methods were witnessed by Nuclear Regulatory Commission personnel and audited by the Metropolitan Edison Company site Quality Control staff.
_ ~.
2.0 INTRODUCTION
l The objective of the periodic integrated leak rate test was the verification of the overall leak tightness of the reactor containment building at the calculated design basis accident pressure of 50.6 psig.
The allowable leakage is defined by the design basis accident applied in the safety analysis in accordance with site exposure guidelines specified by 10 CFR 100.
For Three Mile Island Nuclear Station l
l Unit 1, the maximum allowable integrated leakage rate at the design basis accident pressure of 50.6 psig (P,) is 0.10 percent by weight i
per day (L ).
a Testing was performed in accordance with the procedural requirements as stated in Metropolitan Edison Company Three Mile Island Nuclear Station Unit 1 Surveillance Procedure 1303-6.1.
This procedure was recommended for approval by the Three Mile Island Nuclear Station Unit 1 Plant Operations Review Committee and approved by the Unit Superintendent prior to the commencement of the test.
The combined local leakage rates from the reactor containment building isolation valves and penetrations required to be tested by 10 CFR 50, Appendix J, was less than 60 percent of the maximum allowable leakage rate (L ) at 50.6 psig prior to the commencement of the integrated a
leak rate test (Refer to Appendix D).
Leakage rate testing was accomplished at the pressure level of 50.6 psig for a period of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period was followed by an 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> supplemental test for a verification of test instrumen-tation. During the 32 hour3.703704e-4 days <br />0.00889 hours <br />5.291005e-5 weeks <br />1.2176e-5 months <br /> period of testing, the reactor containment building internal temperature was maintained at 72.0 1 0.3 F.
ssert/Censuermesm L
i 4-i
3.0 ACCEPTANCE CRITERIA Acceptance criteria e.stablished prior to the test and.as specified are as follows:
by 10 CFR 50, Appe.idix J and ANSI N45.4-1972 The measured leakage. rate (Lam) at the calculated design basis a.
accident pressure of 50.6 psig (P,) shall be less than 75 percent of the maximum allowable leakage rate (La), specified as The 0.10 percent by weight of the building atmosphere per day.
acceptance criteria is determined as follows:
La = 0.10%/ day 0.75La = 0.075%/ day The test instrumentation shall be verified by means of a b.
supplemental test. Agreement between the containment leakage measured during the Type A test and the containment leakage determined during the supplemental test shall be within 25 percent of L,.
l l
i 1-Gast/Cammanssee 4
[.
]
4.0 TEST INSTRUMENTATION 4.1
SUMMARY
OF INSTRUMENTS The sensor locations were the same as those used for the preoperational ILRT in 1974. Test instruments employed are described, by system, I
in the following subsections.
4.1.1 Temperature Indicating System I
Overall system accuracy: i 0.19 F 1
j i
Overall systen repeatability: i 0.19 F l
Components:
- i j
i a.
Resistance Temperature Detectors Quantity 24 Manufacturer Rosemount Type Model 104 AAN, 100 ohm, platinum
- Range, F
60-110
- Accuracy, F
i 0.1 Repeatability, F
i 0.1 b.
Bridge Cards Quantity 24 Manufacturer Rosemount Type Model 440-L3
- Range, F
60-110 Accuracy.
F i 0.25% of span Repeatability, F
i 0.25% of span en este "
5 L.
c.
Digital Indicator i Quantity 1
i Manufacturer Weston Type Model 1230
- Range, F 60-110 Accuracy, F i 0.1 Repeatability, F i 0.1
- Modified for direct digital temperature readout j
4.1.2 Dewpoint Indicating System Overall system accuracy: 1 1.12 F Overall systent repeatability: 1 0.52 F Components:
a.
Dewcell Elements Quantity 10 Manufacturer Foxboro Type Model 2711AG, 18 carat gold Range, F 0-100 Accuracy, F i 1.0 Repeatability, F i 0.5 b.
Dewpoint Recorder Quantity 1
Manufacturer Foxboro Type Model Y/ ERB 12 Range, F 0-100 Accuracy, F i 0.5% of span Repeatability, F
_i 0.15% of span Gert/r -
6 L
a
t 4.1.3 Pressure Monitoring System Overall system accuracy: 1 0.015% of indicated pressure Overall system repeatability: 1 0.001 psia Precision Pressure Gauges Quantity 2
Manufacturer Texas Instruments Type Model 145-01 Range, psia 0-100 Accuracy, psia 1 0.015% of indicated pressure Repeatability, psia 1 0.001% of full scale f
4.1.4 Supplemental Test Flow Monitoring System Overall system accuracy: 11% of full scale Flow meter Quantity 1
Manufacturer Brooks Type Model 1114-08 Range, scfh at 0 psig and 100 F 30.9 - 309 1
Accuracy, scfh i 1% of full scale 4.2 CALIBRATION CHECKS Temperature, dewpoint, pressure and flow measuring systems were checked for calibration before the test in accordance with Metropolitan Edison Company Procedure 1430-Y-23, as recommended by ANSI N45.4-1972, Section 6.2 and 6.3.
The results of the calibration checks are on file at Three Mile Island Nuclear Station Unit 1.
The supplemental test at 50.6 psig confirmed the instrumentation acceptability.
u-7 k
4.3 INSTRUMENTATION PERFORMANCE Prior to the start of the integrated leak rate test, one dewcell began indicating a dewpoint temperature approximately 30 F lower than the other 9 deweells. This dewcell was eliminated from future readings.
The remaining 9 deweells performed well at all times and provided more than adequate coverage of the containment.
The temperature, pressure, and flow measuring systems performed well throughout the test.
4.4 SYSTEMATIC ERROR ANALYSIS Systematic error, in this test, is induced by the operation of the temperature indicating system, dewpoint indicating system and the pressure indicating system.
Justification of instrumentation selection was accomplished, using manufacturer's accuracy and repeatability tolerances stated in Section 4.1, by computing the figure of merit as follows.
The leakage rate, in weight percent per day (%/ day), based on an interval of measurement of 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> duration d s P
T L = 100 [1 P
j %/ day T
o 24 where:
P
=P
-P psia partial pressure of air at start To y
P
=P
-P
- 24. psia = partial pressure of air at finish 24 T24 T, = building mean ambient internal temperature at start, R T
= building mean ambient internal temperature at finish, R 24 Geert/Commanumath 8
k.
5 The change, or uncertainty in L due to uncertainties in the j
systematic measured variables is given by (h
TP24) +(
TP ) +(
TT ) +(
d 6L = 100 o
o 24 24 o
o 24 where T is the systematic error for each variable. The error in L af ter differentiation is 1/2
- P24 24 o "Po 24 "To 24 o "T24 o
,g 4
)-
_ o 24
\\P T
)
\\
24 P, T24 24 where:
= TP e,
p
= TP
- P24 24
- To " "o "T24 " T 24 are essentia1'Ly the same, within Since the values of T, and T24 0.28 F, and P and P are essentially the s.ame, within 0.002 psia, 9
24 and e let T, = T24' o" 24' *Po " "P24 " *p To " "T24 ~ *T' The systematic error in L then reduces to e
2 e
2-f (1)
[
e = 141.4
+
g o
o where the error in pressure (e ) may be e:tpressed as p
2)1/'9 2
c = (e
+e p
p p
a b
i
)
Geert/Commonsesth 9
Li
~
a and e
= error induced by the precision pressure gauges, or p
,, (0.00015)(65.340) pda ep (2)b
~
e
= 10.0069 psia p
a and e
= error induced by the dewcells, or p
1.12 o F e
=
(9)g p
b e
= 1 0.373 F p
From steam tables, at a dewpoint of 65 F, the pressure equivalent to 1 0.373 F is
= 1 0.0039 psia ep Therefore,
= [(0.0069) + (0.0039)2 h,1, 3p ep
= 1 0.0079 psia ep The error in temperature (e ) may be expressed as T
0.19 o e =i F
(24)g T
i
= 1 0.0388 F eT 1
Geert/Commanuesth 10
\\
d Hence, for values at 50.6 psig, i
P, = 65.340 psia T,'=."531.42 R and substitution into equation (1) yields 141,4 [(0.0029)2 0.0388)2j
, g
- L 65.340 531.42
= 1 0.020%/ day e
The maximum expected systematic error (figure of merit) of the test instrumentation is e.
g If equation (1) is solved using previously stated repeatability values, the figure of merit is calcuated to be
=,1 0.011%/ day og Containment leakage rate computations are a function of changes in temperature and pressure relative to each other, not absolute values. Therefore, the repeatability error analysis is more meaningful; A conclusion reached from the above calculation was that the instrumentation selected yielded an error value five times less than the allowable leakage rate value of 0.10 percent per day and that the instrumentation' combination'was of sufficient sensitivity for this test. The e values are not based on a statistical analysis of leakage rate calculations and are used strictly for instrumentation selection.
ss,ste-11
~
L j
4.5-SUPPLEMENTAL VERIFICATION in addition to the calibration checks described in Section 4.2, test instrumentation operation was verified by a supplemental test '
subsequent to the completion of the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> leakage rate test. This test consisted of impocing a known calibrated leakage rate on the reactor containment building. Af ter the flow rate was established, it was not altered for the duration of the test.
During the supplemental test, the measured leakage rate was Lc" v'+
o
- where, L = measured composite leakage rate consisting of the reactor c
building leakage rate plus the imposed leakage rate L = imposed leakage rate 9
L, = leakage rate of the reactor building during the y
supplemental test phase Rearranging the above equation, L,=L
-L y
The reactor containment building leakage during the supplemental test can be calculated by subtracting the known superimposed leakage rate from the measured composice leakage rate.
Genruc.we==m 12
The reactor-containment building leakage rate during the supplemental (L.) was then compared to the measured reactor containment test y
building leakage rate during the preceding 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. test (L,) to determine instrumentation acceptability. Instrumentation is considered acceptable if the difference between the two building l
. leakage rates is within 25 percent of the maximum allowable leakage rate (L,).
t n
er l
GeertICommanussen 13 i
5.0 TEST PR("TDURE 5.1 PREREQUISITES Prior to commencement of reactor containment building pressurization, the following basic prerequisites were satisfied:
a.
Proper operation of all test instrumentation was verified.
b.
. All reactor containment building isolation valves were closed using the normal mode of operation. All associated system valves were placed in post-accident positions.
c.
Equipment within the reactor containment bu-ubj ec t to damage, was protected from external differential pr essures.
d.
Portions of. fluid systess which, under post-accident conditions become extensions of the containment boundary, were drained and vented.
r The penetration pressurization and fluid block systems were f
e.
depressurized. Gauges were installed at penetration pressurization manifolds to provide means for detection of leakage into the system. These gauges were removed and the l
manifolds were vented prior to the start of the test.
f.
Pressure gauges were installed on closed systems within containment to provide means for detection of leakage into such systems.
g.
Local leakage rate testing of containment isolation valves and penetrations was concluded.
Georgt 14
-p-
, fp M
d 6,,5
/ *#,<l f,
- A
/,J' /,#,,#*
e/
.+
y c',"'S':*,/Y '
,# * + /,+ p + s
,*'/
a,,8 e
/.
- e s,>'
<>'/
M$[.s:'+::0,
,#n, -:::~:::,;::::4:::-;:,$
- ~~** #
s,
/,/, ' /
/*
,+
p ** #, s* +/ r>** ##< #,.
/
~ -:
/
/
f, /,~ s -s / +'#r 4+ *
,,*+o+'
p p **,6 *#+
~
+ '#
~
8++
"* /*,,** /.
/*
ac 5n. g:~w s***
~/
~
4A
</+*
I
+
/
% j;
- , s4
. 6. i y&*
L gg-
' b T.
3 p
f6
/
The twenty-fo c lc lated r RTD a
u u
te c.
mpe The atu r
res nin nin e de er w
e wpoint r
e ec v
o ded alu r
v lu This was a
es ad es n
were per co the pr mit nv av rec er ted erted essure to o ded.
age r
corr v
The of ectio apo The air by r
use n
pr av of essur age er of sb the total apo tr u
v e
of total sing ste the u
r a
cting the
\\
pr pr essur pr essur All ess am e (P ) is d e (P,), a ur o
v r
to tab Uni rigin l apo e
pr the le.
T s
a data is t
ess pa ver es 1
c ib age ur.
rtial r
e ed nf in tempe o
The ile mor ratur Thr e detail e (T) a d plot at ho of e
e Mile Islin Se the n
r aver u ly fr (Se age ctio te ad Nu le n61 e Ap mpe n
o m1000 pe d atu r
c April n April n ix B).
re ar o
ad S
n tatio 18, 19 16, 1977 spheric w
Atm eigh n
o t
clo 77 of udy.
to 15 to 18 athe air we was 30 o 30 r
perfo Whe n April n April o
codni n
rmd e
co e ie 19, 197 tio half nv e tr nt, the 7, the 18, 197 ns w
n er were 7
cle ansi G
t av w
Fr m
ar ilbe e
o D
rt ted ailab athe m 19 r
00 As via o si le half-h codni co so on n
mpute ciate, In tio te o
ns r
s a.
- c. ho rtab u ly v po r
w squ pr er ar ogr e
m le alu esfit Co me co es A in l of mpute offic mpute P,, T of f
the data r
e fo ter a
r pr w
co r
n lysis min l ad mpute ogr a
n as P
a a
r
,w am av to T
r ailab ere run esl s
the u
r ts, in l u ing the C le.
was etur Subs m de nd a
e c
to ud LERCAL equ af nt ter the ing a le e
stab the 24 data fo site a
a to r
via st e
l ished ho a fuli the te ur le k er r
fo 24 a
ho minal.
atur, pr r
e an te ur addi st, a period ess ur tio s
nl upe impo e
a ad r
n 8
apo ho v
ur sd e
r period.
le kage pr a
essure Du ing th r
w r
ate er M
e monito is 16 ed tim,
r e
as des cribed
-/
h.
Potential pressure sources were removed or isolated from the containment.
1.
All accessible liner weld channels (approximately 35 percent of the total) were vented to the containment atmosphere.
j.
A general inspection of the accessible interior and exterior areas of the containment was completed.
5.2 GENERAL DISCUSSION r 11owing the satisfaction of the prerequisites stated in Section 5.1, j
o the reactor containment building pressurization was initiated at a j
rate of approximately 2.5 psi per hour.
Building internal I
temperature was maintained at approximately 72 F.
Building pressure e'
and temperature were monitored half hourly and the amperage required O,, l' by the recirculation unit fans (AH-E-1A, 1B and IC) was monitored hourly. Leak rate testing was initiated at the 50.6 psig pressure level. Forty-three hours elapsed between reaching the 50.6 psig pressure level and the recording of official data.
For i
the duration of the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> leak test and the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> supplemental i
test, the average internal containment temperature was maintained within a band of + 0.3 F by varying the industrial cooler cooling water flow rate to the containment recirculation fan unit coolers.
During the test the following occurred at half-hour.4ntervals (See Appendix A):
Pressures indicated by each of the two precision gauges were a.
recorded and the average calculated.
geygfra-nh 15
b.
The twenty-four RTD temperatures were recorded and the average calculated.
c.
The nine dewpoint values were recorded.
The average of the nine values was converted to vapor pressure using steam tables.
This permitted correction of the total pressure to the partial pressure of air by subtracting the vapor pressure.
i J
The use of vapor pressure (P,y), average temperature (T) and the total pressure (P ) is described in more detail in Section 6.1.
T All original data is on file at Three Mile Island Nuclear Station Unit 1.
l l
The plot of average temperature and weight of air was performed half hourly (See Appendix B). Atmospheric weather conditions were clest from 1000 on April 16, 1977 to 1830 on April 18, 1977.
From 1900 on
- %J
,f r
April 18, 1977 to 1530 on April 19, 1977, the weather conditions were cloudy.
When convenient, the available half-hourly values of P,, T and PT were transmitted via on-site portable computer terminal to the Gilbert Associates, 7.nc. home office for analysis using the CLERCAL computer program. Computer program results, including a least squares fit of the data, were returned to the site via the terminal.
A final computer run was made after data for a full 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period was available.
Subsequent to the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> leak test, a superimposed Icakage rate was established for an additional 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period. During this time, temperature, pressure and vapor pressure were monitored as described above.
Start /rW 16
5.3 TEST PERFOPJMNCE 5.3.1 Pressurization Phase Pressurization of the reactor building containment was started on April 15, 1977 at 0500. The pressurization rate was approximately 2.5 psi per hour. When containment internal pressure reached i
12 psig, at 1120 on April 15, 1977, pressurization was secured.
4 An inspection team entered containment to perform the 12 psig inspection. During pressurization to the 12 psig pressure level, the Leak Rate Test System air dryer drain and the cyclone separator drain were not functioning properly. Pressurization was secured while temporary bypasses were installed. While at the 12 psig pressure level, these drains were repaired. The 12 psig internal inspection was completely satisfactorily and pressurization was
]
restarted at 1336 on April 15, 1977.
During pressurization to the 50.6 psig pressure level, the following observations were made:
Several penetration pressurization manifold isolation valves a.
were suspected of leaking.
The main header was then vented to ensure the penetration pressurization system would remain depressurized.
\\
b.
A buildup of pressure on several of the pressure gauges installed on penetration pressurization manifolds indicated a small amount of leakage from the fuel transfer tube flanges, the personnel and emergency airlock door seals, and manifold "J".
GeertICammenweenh 17 1
u-
I l
f I
A small amount of water leakage was noticed from Nuclear c.
Services Closed Cycle Cooling Water valves NS-V4 and NS-V15.
d.
One leak rate test deweell began to indicate a dewpoint temperature approximately 30 F lower than the remaining nine dewcells. This dewcell was eliminated from data collection.
When containment internal pressure reached 50.7 to 50.8 psig, at 0600 on April 16, 1977, pressurization was secured. Temperature was controlled by throttling the industrial cooler pump discharge valve, RB-V18D, which supplies cooling water to the recirculation fan units cooling coils. All penetration pressurization system temporary manifold pressure gauges were removed.
5.3.2 Integrated Leak Rate Testing Phase Af ter waiting 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, leak rate testing was started. Temperature had stabilized at approximately 72 F.
From 1000 on April 16, 1977 until 0500 on April 18, 1977, an excessive leakage rate was indicated by the data collected.
The weight of containment air and the average containment temperature versus time for this time period
)
i are presented in Appendix B, Exhibits 1 and 2.
During this time, I
the following sequence of events took place-a.
At 1200 on April 16, 1977, the leakage rate, based on two hours of data, was 0.151 percent by weight per day.
This established a baseline for the mass point versus time graph.
l Plant auxiliary operators were sent on routine leak detection.
f There was no cause for immediate concern since only a limited amount of data had been collected.
GeertICommonweenh 18
b.
Subsequent mass points were following approximately the same trend as previously reported.
Pressure gauges were installed on manifolds "J", "N" and "O" of the penetration pressurization system for leakage detection. Plant auxiliary operators were again dispatched for leak detection.
c.
At 1930 on April 16, 1977, the leakage rate, based on nine and-one-half hours of data, was 0.199 percent by weight per day.
The pressure gauge on manifold "0" (Fuel Transfer Tube Flanges) was replaced with a flow indicator.
d.
The fluid block line to valve IC-V4 was isolated and vent valve FB-V122 was opened. Leakage through this path was evident.
The purge valves and the acc ess lock doors were soap-checked e.
and no leakage was indicated. A bonnet / packing leak on penetration pressurization system valve PP-V46 and reducer Icaks on the reactor building pressure sensing lines near BS-V37C and BS-V37D wet e found and repaired.
l f.
An investigation revealed that several of the automatic fluid block initiation valves, specified to be open, were closed.
All automatic fluid block initiation valves were opened.
g.
Additional flow indicators were placed on the main steam lines I
from steam generator A (OTSG A) and steam generator B (OTSG B).
At 2400 on April 16, 1977, the following leakages were indicated:
Geet/Cammenuum 19
I
~
i 1
l Location Leakage Manifold "0"
160 seem OTSG A 850 seem OTSG B 0 seem WDG-V4 700 seen i
h.
Since the amount of leakage found was insignificant compared to i
the leakage indicated by the data (250,000 seca), leak detection continued. At 0245 on April 17, 1977, the reactor containment building was repressurized to between 50.7 and 50.8 psig.
i.
Indicated leakage from OTSG A had increased to 1000 seca. The fluid block line to IC-V4 was opened and no pressure buildup in manifolds "N" and "O" was observed.
j.
As leak detection continued, the measured containment leakage rates were as follows:
Date Time Interval Leakage Rate 95% Confidence 4-17 0300-0700 0.120%/ day 0.051%/dey 4-17 0300-1400 0.126%/ day 0.009%/ day k.
A valve lineup verification was performed and no deviations were found. A systematic quadrant by quadrant check of penetrations and isolation valves f ailed to identify any significant leakage. The following adjustments were made on April 17, 1977:
1)
Fittings and connections in the leak rate test panel were l
tightened.
2)
Flanges downstream of LR-V2 and LR-V3 were tightened.
Eieertle 20
~
Minor amounts of leakage were evident at the following locations:
1)
LR-V2 and LR-V3 packing 2)
Purge supply interspace.
3)
Personnel airlock 4)
Purge exhaust interspace 5).
OTSG A 1.
At 2230 on April 17, pressurization of the secondary side i
of OTSG A was begun to determine if a change in the indicated l.
reactor containment building leakage could be detected. With the OTSG A at 16 psig, it was decided to depressurize OTSG A since the data prior to 2230 had indicated an upward trend in the mass points.- At 0210 on April 18, 1977, OTSG A was i
depressurized and it was decided to collect and evaluate a full 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of data.
i m.
At 1030 on April 18, 1977, it was noted that approximately l
5 psig pressure had built-up between the seals of the
{
emergency airlock and the personnel airlock.
Vents were opened, the pressure was bled off and the vents were lef t open to allow a leakage path to exist.
n.
The containment leakage rate measured from 0300 to 1130 on April 18,1977 was 0.097 percent by weight per day with an i
upper bound 95 percent confidence of 0.017 percent by weight per day.
i GastIt "
21
1 I
o.
At 1320 on April 18, 1977, the concrete shield for the equipment hatch was put in place.
i p.
The measured containment leakage rate from 0300 to 1600 on i
April 18,1977 was 0.103 percent by weight per day with an upper bound 95 percent confidence of 0.009 percent by weight
[
per day.
q.
Subsequent to 1600 on April 18, 1977 a shift in the trend of the containment mass points. occurred.
t An acceptable leakage rate of 0.042 percent by weight per day r.
was obtained from 0500 on April 18, 1977 to 0500 on April 19, 1977.
Due to the lack of any local leakage rate determinations prior to the adjustments mentioned in Section 5.3.2.k., the initial f
unsatisfactory leakage rate indications must be assumed to constitute a failed test.
l t
Nevertheless, since an extensive search f ailed to identify any i
significant sources of leakage, it is unlikely that the initial measured leakage rate values, which were in c.xcess of 0.10 percent by weight per day, were true measurements of leakage from the reactor containment building to the outside atmosphere.
Two possible explanations for the initial results are:
j a.
There was leakage into volumes internal to the containment building..The internal volumes may have been (1) the reactor coolant system, since a slow steady decrease'in f
l the pressurizer level was noted throughout the test with Ges,tle-%
22
t no corresponding increase in reactor building sump level, and/or (2) the volume between isolation barriers. Additionally, there may have been air entrainment into the concrete and insulation material inside the containment. However, the length of time that the excessive leakage rate was present and the abrupt rather than gradual change in the leakage rate do not tend to support this explanation entirely.
b.
The apparent leakage was the result of a diurnal effect.
The heating of the containment during the day and the cooling of the containment during the night would cause a change in the containment internal pressure due to the expansion / contraction of the containment without a corresponding detectable change in the containment internal temperature.
However, the data, as presented in Appendix B, Exhibits 1 and 2, does not appear to totally support an explanation based on I
diurnal effects.
f 5.3.3 Supplemental Leakage Rate Test Phase l
Af ter the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> integrated leak rate test data was obtained and evaluated, and the leakage rate found to be acceptable, and a release permit had been obtained, a known leak rate was imposed on the reactor containment building through a calibrated flowmeter for a period of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />.
t l
23
5.3.4 Depressurization Phase After all required data was obtained and evaluated, and the supplemental test results were found to be acceptable, and permission from the health physics department and unit superintendent -
was obtained, depressurization of the reactor containment building was started. A post test inspection of the building revealed no unusual findings.
Geert/Commonween 24
l 6.0 METHODS OF ANALYSIS 6.1 GENERAL DISCUSSION The absolute method of leakage rate determination was employed during teisting at the 50.6 psig pressure level. The Gilbert i
Associates, Inc. CLERCAL computer code calculates the percent per i
day leakage rate using the mass point method of data analysis.
The i
results presented are based on the mass point method.
The mass point method of computing leakage rates uses the following ideal gas law equation to calculate the weight of air inside
]
containment for each half hour:
- where, W = mass of air inside containment, iba
)
6h-
- in.2 K = 144 V/R = 5.3983 x 10 P = partial pressure of air, psia T = average internal containment temperature, R V = 2.0 x 10 ft The partial pressure of air, P, is calculated as follows:
T1 + T2 P=
-P 2
wv
- where, P
= true corrected total pressure from PI-390, psia T1 P
= true corrected total pressure from PI-391, psia T2 P
= partial pressure of water vapor determined by averaging I
the nine dewpoint ' temperatures and converting to vapor pressure with the use of steam tables, psia Geert/Cammennesah j
25 i
The average internal containment temperature, T is calculated as 1
follows:
l f 24 RTD's T = sum
+ 459.69 R 24 j
l The weight of air is plotted versus time for the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test and for the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> supplemental test. The Gilbert Associates, Inc.
CLERCAL computer code fits the locus of these points to a straight line using a linear least squares fit. The equation of the linear least squares fit line is of the form W = W + W t where W is the g
g slope in iba per hour and W, is the weight at time zero.
The least squares parameters are calculated as follows:
2
,Ee 7g
,g 7
g f
i i
i i y
Sxx Et IW net W g 1 g
g
~
1 Sxx where.
S
= net
- (E tg) g The weight percent leakage per day can then be determined from the following equation:
-2400 Wl wt. %/ day =
yo where the negative sign is used since W is a negative slope to 3
express the leakage rate as a positive quantity.
Geert/r 26
t 6.2 STATISTICAL EVALUATION After performing the least squares fit, the CLERCAL computer code t
calculates the following statistical parameters:
Standard error of confidence for the curve fit (S,).
a.
b.
Limits of the 95 percent confidence interval for the curve fit.
Limits of the 95 percent confidence interval for the leakage c.
rate (C ).
g The significance of the measured leakage rate can then be evaluated in view of the number of data points exceeding the limits of the 95 percent confidence interval and by the magnitude of the upper bound j
of the 95 percent confidence interval for the leakage rate.
I Standard error of confidence is defined as follows:
l I
t )]T h
~
E[W - (W
+W g
1 g
S, =
N-2
- where, W = observed mass of air g
t ) = least squares calculated mass of air (W, + Wg g N = number of data points This parameter is an expression of the difference between an observed a:.d a calculated (least squares) mass point. The 95 percent confidence interval of the fit is twice the st.ndard error
' of confidence (2S,). The " degree-of-fit" is evaluated by determining the number of ' data points, W, not f alling in the interval g
(W, + W t) +2S,.
i ge,ur - s 27 a
-w+4 3he+n e
The 95 percent confidence limit for the mass leakage rate is calculated as follows:
2
'5
+
N xx i
C
=t 8
+
g 95 e S
NS xx xx
- where, t
= Student a t distributi n with N-2 degrees of freedom 95 This parameter is an expression of the uncertainty in the measured leakage rate.
4 9
Geert/Cammanussa 28
i 7.0 DISCUSSION OF RESULTS 7.1 RESULTS AT Pa Data obtained during the integrated leak rate test at P, indicated the following maximum changes (highest reading to lowest reading) during the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test period:
Varisble_
Maximum Change P
0.026 psia T
P 0.011 psia 0.47 F T
The method used in calculating the mass point leakage rate is defined in Section 6.0.
The result of this calculation is a mass point leakage rate of 0.042 %/ day.
The 95 percent confidence limit associated with this leakage rate is ll 0.010 percent per day. Thus, the leakage rate at the upper bound of the 95 percent confidence interval becomes L,,= 0.042 + 0.010 L,, = 0.052 %/ day The measured leakage rate and the measured leakage rate at the upper bound of the 95 percent confidence level are well below the acceptance criteria of 0.075 percent per day (0.75 L,).
A comparison of each of the observed weights with the weight =, calculated using the least squares line reveals only one of the forty-nine data points does not lie within the 95 percent confidence interval.
Therefore, reactor containment building leakage at the calculated design basis accident pressure (P,) of 50.6 psig is considered to be acceptable.
Gdbert /CommonwesRA 29
~
7.2 SUPPLEMENTAL TEST RESULTS Af ter conclusion of the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test at 50.6 psig, flowmeter FI-111 was placed in service and a flow rate of 207 SCFH was established.
This flow rate is equivalent to a leakage rate of 0.056 percent per day. Af ter the flow was established, it was not altered for the duration of the supplemental test.
The measured leakage rate (L ) during the supplemental test was c
calculated to be 0.099 percent per day using the mass point method of analysis.
The 95 percent confidence interval associated with this leakage rate is 0.020 percent per day.
None of the i
25 data points is out of confidence.
The building leakage rate during the supplemental test is then determined'au follows:
i i
L,=L
-b I
y c
o L, = 0.099%/ day - 0.056 %/ day y
L, = 0.043%/ day y
Comparing this leakage rate with the building leakage rate measured during the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> test yields the following:
~
l(0.042) - (0.043) l
= 0.01 am v'
=
L, 0.10 1
F The building leakage rates agree within 1.0 percent of L, which is well below the acceptance criteria of 25 percent of L,.
Therefore, the acceptability of the test instrumentation is considered to have l
been verified.
GestICammenessRh 30
f.
8.0 TYPE B AND C LEAKAGE RATE HISTORIES Refer to Appendicies C, D and E for the report on Type B and C testing. performed since the previous Type A test.
f a
1 l
l i
i l
i Geert/Cammenneem 31
i
9.0 REFERENCES
1.
SP 1303-6.1, " Reactor Building Integrated Leak Rate Test",
Metropolitan Edison Company Surveillance Procedure.
2.
Code of Federal Regulations, Title 10. Part 50, Appendix J, (1-1-75).
3.
ANSI N45.4-1972, " Leakage Rate Testing of Containment Structures for Nuclear Reactors" American Nuclear Society, (March 16, 1972).
4.
Steam Tables, American Society of Mechanical Engineers, (1967).
5.
CLERCAL, Computer Code, Gilbert Associates, Inc.
6.
1430-Y-23, " Reactor Building Integrated Leak Rate Test Instrumemt Calibrations", Metropolitan Edison Company Procedure.
1
+
32
4.
J J4 J
.d 4
4 4
5 2--
- 4O aa4Xep----W--.W4 J4 46e-,
o O
,4-4 l
l l
k I
'l t
h, 5
4 i
3 L
I 9
h E
.f i
n d
(
t 4
APPENDICES r
i i
j I
i 1
t 1
1
.i i
k i-t, a
i A i.
e Geert/Cammenmaak
W a
a.-J 6
6 n
m 4
4, w"aM-
-4 m
e e
4 I
?
I I
I APPENDIX A REDUCED LEAKAGE DATA t
e i
a i
U
+
P k
lidhart/Commenuesth t
APPENDIX A REDUCED TEST DATA Average Partial Pressure Partial Pressure Average Weight of Containment of Containment of Containment Containment Containment Pressure Water vapor Air Temperature Air Time (psia)
(psia)
(psia)
(oR)
(ibm) 4/18/77
,0500 65.340 0.294 65.046 531.42 660,753.87 0530 65.338 0.293 65.045 531.42 660,743.71 0600 65.335 0.294 65.041 531.41 660,715.51 0630 65.336 0.295 65.041 531.43 660,690.65 0700 65.336 0.293 65.043 531.44 660,698.53 0730 65.337 0.295 65.042 531.44 660,688.37 0800 65.338 0.297 65.041 531.49 660,616.06 0830 65.340 0.294 65.046 531.52 660,629.56 g
1 0900 65.342 0.294 65.048 531.54 660,625.01 0930 65.344 0.295 65.049 531.57 660,597.88 1000 65.346 0.294 65.052 531.60 660,591.07 1030 65.348 0.295 65.053 531.64 660,551.52 1100 65.350 0.293 65.057 531.68 660,542.44 1130 65.354 0.294 65.060 531.70 660,548.05 1200 65.356 0.292 65.064 531.75 660,526.55 1230 65.354 0.291 65.063 531.76 660,503.97 1300 65.356 0.294 65.062 531.78 660,468.98 1330 65.361 0.298 65.063 531.81 660,441.87 1400 65.358 0.292 65.066 531.81 660,472.33 1430 65.358 0.292 65.066 531.81 660,472.33 1500 65.358 0.293 65.065 531.83 660,437.34 1530 65.357 0.294 65.063 531.84 660,404.62 1600 65.356 0.291 65.065 531.84 660,424.92 1630 65.355 0.292 65.063 531.82 660,429.46 1700 65.353 0.294 65.059 531.81 660,401.27 1730 65.354 0.293 65.061 531.83 660,396.74 1800 65.356 0.291 65.065 531.84 660,424.92 1830 65.358 0.295 65.063 531.87 660,367.37 1900 65.360 0.292 65.068 531.87 660,418.12 m.
m
APPENDIX A (Cont'd)
REDUCED TEST DATA Average Partial Pressure Partial Pressure Average Weight of Containment of Containment
'of Containment Containment Containment Pressure Water vapor Air Temperature Air Time (psia)
(Psia)
(psia)
(OR)
(1ba) 1930 65.356 0.292 65.064 531.85 660,402.35 2000 65.359 0.294 65.065 531.86 660,400.09 2030 65.358 0.294 65.064 531.86 660,389.94 2100 65.360 0.294 65.066 531.88 660,385.40 2130 65.361 0.293 65.068 531.88 660,405.70 l
2200
-65.360 0.290 65.070 531.87 660,438.42 a
2230 65.357 0.294 65.063 531.84 660,404.62
[
2300 65.353 0.294 65.059 531.82 660,388.85
)
2330 65.352 0.294 65.058 531.79 660,415.96 I
- f 2410 65.349 0.294 65.055 531.77 660,410.34 4/19/77
,0030 65.350 0.293 65.057 531.74 660,467.90 l
0100 65.346 0.292 65.054 531.73 660,449.87 l
0130 65.348 0.292 65.056 531.73 660,470.17 0200 65.346 0.291 65.055 531.72 660,472.44 3
0230 65.346 0.291 65.055 531.'i4' 660,447.60
[
0300 65.346 0.292 65.054 531.73-660,449.87 0330 65.348 0.294 65.054
'531.75 660,425.03 i
0400 65.348-0.292 65.056 531.77 660,420.49 0430 65.344 0.292 65.052 531.71 660,454.40
,0500 65.342 0.287 65.055 531.70 660,497.29 SUPERIMPOSED TEST r 0700 65.330 0.294-65.036 531.68 660,329.22 0800 65.333 0.289 65.044 531.73 660,348.34 l
0830 65.334 0.291 65.043 531.76
-660,300.94 0900 65.333 0.292 65.041 531.76 660,280.63 i
,0930 65.330 0.288 65.042 531.73 660,328.04 1000 65.328 0.291 65.037 531.73 660,277.28 1030 65.327 0.291 65.036-531.72 660,279.54 i
I s
,--r
....._.,y.
..._....-.,s,,
-a
.m.
_,s...
....+!
_.....--..~.. - _ _ _ -. -. ~..
APPENDIX A (Cont'd)
REDUCED TEST DATA Average Partial Pressure Partial Pressure Average Weight of Containment of Containment of Containment Containment Containment Pressure Water Vapor Air Temperature Air Time (psia)
(psia)
(psia)
(OR)
(1bs) 1100 65.327 0.291 65.036 531.73 660,267.13 1130 65.330 0.290 65.040 531.74 660,295.32 1200 65.331 0.292 65.039 531.77 660,247.91
'1230 65.327 0.289 65.038 531.77 660,237.76
-1300 65.322 0.294 65.028 531.71 660,210.74 1330 65.320 0.293 65.027 531.70 660,213.00 1400 65.324 0.291 65.033 531.77 660,187.01 1430 65.328 0.293 65.035 531.84 660,120.41 f
'500 65.332 0.290 65.042 531.88 660,141.82 1
1530 65.332 0.291 65.041 531.90 660,106.84 h
l
_e*wmw
.-e
+iv.
e y
e p-o r,
e m
+
ama-,
m-F
.m
,uas 4
o e'
e l
APPENDIX B WEIGHT OF CONTAINMENT AIR AND AVERAGE CONTAINMENT TEMPERATURE i
4 4
+
W 1
A e I Gert/W
9 a
TABLE OF CONTENTS EXHIBIT 1:
1000-0230 HOURS (4/16/77 - 4/17/77)
EXHIBIT 2:
0300-0430 HOURS (4/17/77 - 4/18/77)
EXHIBIT 3:
0500-0500 HOURS (4/18/77 - 4/19/77) 0730-1530 HOURS (4/19/77) l I
i emartle "
i 1
\\
r.
f(['N i 1 NUN UNITED STATES
'\\
ATOMIC ENERGY COMMISS:ON
[,
' :
- b. * }
". t *
~
WASWNCTON. D.C. 2054
=
,g, s
'l
~'h May 26, 1970 R. C. DeYoung, Assistant Director, PWRs t*
Division of Reactor Licensing TilRU:
CharJes G. Long, Chief, PWR Project Branch 2, DRL 4,y INITI AL MEETING WITil M1;T-ED ON TilRE!; HILE ISLAND UNIT 1 POL (DOCKET 50-289)
The initial meeting with the Metropolitan Edison Company representatives concerning the Operating License application for 'three Mile Island Unit No. I was held May 13, 1970.
A list of atterdees is attached.
We discur. sed the proposed review schedule, W.ich calls for our review com-pletion in earJy 1971, and the major revies iams, as follows:
1.
Sodium Thioy,ulfste This is the first PWR-OL application for a plant specifically designed to use sodium thiosulfate as an additive; we stated that this would be a ca ior review item.
Tt.*o additional reports on thiosulfate are forthcoming.
Supplemental information concerning stability and compatibility will be added to topical BAW-10017 by LLW, latcr this iaonth, lo additiva, e.uppleines Lul Infouuativat un iodine removal efficiency will be filed as a Met-l'd amendment, around July 1.
W 2.
Instrumentation We said that we would use the results of the Oconee review where possibic.
Regarding prior Met-Ed commitments we noted ACRS comments on separation of control and safety, scram bus separation, failed fuci detector, and dilution system controls. We asked Met-Ed to show how the final design satisfied these points.
B&W noted that the common mode failure topical was due in August.
- 3. ' Fan Coolers We were informed that the fan cooler test report would _ be available in the 3rd quarter of 1970. We observed that the report was some-what late as compared to earlier commitments by Met-Ed, and compared to actual procurement.
4.
Envi ror. men t al Met-Ed was inf onned that some sort of environmental policy statement would be prepared and that an informatic.. request would be forth-coming.
A
/
-w
'i t e '
7 -
o H. C. DeYoung 2
May 26, 1970 5.
Site Several meteorology data questions were brought out.
A a.
special meeting was set (for May 19, 1970).
b.
Flooding was also considered as a major review ~ item, and a spec; al meeting set (for June 16, 1970).
Environmental Monitoring - We noted several points that were c.
not considered in their program and said that this would be a significant review item.
6.
Raduwste The radwaste system, designed by GA1, is to be a major review iten.
Several inconsistencies in liquid radwaste isotopic discharge esti-mates,1NI #1 POL vs TMI #2 CP were noted.
7.
Safety Analysis We stated our intention to fully review the calculations on steam generator " residual" activity. We also wanted to be assured tnat the 72-hour cooling period assumed in the refueling accident is intended to be an operational limit.
8.
Structures We said all Class I structures, systems, and components would a.
d4 be reviewed, especially in consideration of the aircraf t impact requirements.
b.
We intend a comparative review on vessel thermal shock, material surveillance, seismic and other loads, and rod drives (all B&W topicals reviewed on Oconce).
9.
Miscellaneous
, I asked (as an audit or example) for the detailed analysis or calculations, to be discussed at our next technical meeting, on fuel rod swelling with burnup, on pressurizer stresses following a surge line rupture, fuel pool cooler design, llP pump capacity vjt one stuck primary safety valve, and operation sequence of steam driven emer-gency feedwater pua:p.
'w os~c%
{ cQ d' Denwood F. Ross PWR Project Branch 2, DRL
Enclosure:
Attendance List cc:
See page 3
e
.o-ie j
R. C. DeYoung 3
May 26, 1970 l
cc:
Docket DRL Reading PWR-2 Reading P. A. Morris
(
' F. Schroeder I
T. R. Wilson R. S. Boyd i
R. C. DeYoung D. Skovholt i.
E. G. Case, DRS R.. H. Maccary C<>mpliance (2)
DRL and DRS 13 ranch Chiefs I
D. F. Ross ' (2).
f.
W. E. Nischan l
T. M. Novak i
F. W. Karas R. W. Klocker i
i i
Y 4
4 9
8 9
9
l t
HET-ED MEETING l
May 13, 1970 I
i LIST OF ATTENDEES j
(,,l Flat _Ed
)
Kathy Matt, Project Administrator J. L.- Bachofer,.Tr., Assistant Project Manager George F. Bierman, Project Manager G. Charuoff, Consultant Counsel l
E. G. Itocme, CPU l
Gilbe r_t; C. II. Bitting, Project Manager l
F. W. Symons j
_B &W E. G. Ward, Project Manager.
J. M. Cutchin, Licensing PIA Keith Woodata
).
DRL C..G. Long D. F. Ross T. M. Novak W. E. Nischan I. Van der Hoven (ESSA) j.y 7 P 4
4 e
a
i I
r
~72)0 N'
8'MS, UNlrLD C f/ Yu i
d 'I AT OMI.T. CNCRGY COMMiGSIOh' w.anticTor:, t.c. nso
-g
/f,
l ~4 "w J,*)
' 3 /,,, L L'
1 eiT,7T June 19, 1970 3,
R. C. DeYoung Assistant Director, PWPJ.;
Division of Reactor Licensing g>-@/.'
TIIRU:
Charles G.
Long, Chief, PUR Project Branch 2, DRL liYDROLOGY MEETING ON THREE MILE ISLAND UdIT 1, POL (DOCKET 50-289)
A meeting was held with representatives of Gilbert Associates, the A/E for Met-Ed on Three Mile Island Unjt 1.
The purpose of the meeting was to discuss the PMF calculations for the site. Attending were R. H.
MacLemore and Joel Caves for Gilbert and D. Ross, W. Nischan, and 4
D. Nunn for DRL.
MacLemore demonstrated a plot of flood stage along the river for various discharges up to 1.75 million cubic feet per second.
lie also showed a map where flood contours had been calculated.
lie discussed the procedures used to calculate level versus discharge with the DRL hydrologist. D. Nunn. As a result of our discussions, we notified Gilbert that we contemplated four broad question areas to be included in our next formal list of questions, along the following lines:
1.
U: caid that the applicant cheuld provide a dischsr;;c hydrograph for the PMF, both regulated and unregulated.
2.
We asked for the backwater analyses for the 1936 and 1964 floods and for the calculated discharges of 1.1 million and the PMF.
3.
We requested a discussion of the procedures for calculation of backwater and the significance of overbank flow. This should include, we said, a table showing the Manning-n coefficients, and the discharge values and elevations at each cross section.
We also want the comparison of measured versus computed eleva-tions f or the observed floods incorporating the known high water i
marks, and a map showing the location of the river cross sections used in the computer program.
Finally, we want flood level Con tou rs.
4.
We anticipate a need for a discussion by the applicant of the operating procedures in advance of and during extreme flood events.
This should include the inf ormation that will be avail-able to the operating staf f and the decision levels that the staff must face in terms of river stage or precipitation.
i Y
lv nw m.
~
4 i
L.
-4 I,.
}
i R. C. DeYoung 2
June 19,.1970 t
I I
These comments were given'to the Gilbert representatives. We cautioned i
them to await a formal transmission of these requests before submitting
?4 answe rs. We also notified them of our intention to visit the site next i
Q' mon th.
i k, d V,h
-O'/d 1
u
,,n Denwood F. Ross PWR Project Branch 2 Division of Reactor Licensing
\\
Dis tribution:
Docket
)
DRL Reading PWR-2. Reading i
P. A. Morris i
F. Schroeder T. R. Wilson R. S. Boyd l
R. C. DeYoung D. Skovholt E. C. Cacc, DRS R. R..Maccary i
t g
R. W. Klecker Compliance (2)
I DRL & DRS Branch Chiefs D. F. Ross l
W. E. Nischan D. Nunn I
F. W. Karas
39',0 6
FEB.
f 80
k/
1 Metropolitan Edison Company P. O. Box 542 Baading, Pennsylvania 19603 Attention:
Mr. J. G. Miller, Vice President Gentlemen:
This letter relates to the discussion Messrs. E. M. Howard and D. M. Hunnicutt of this office held with Mr. T. E. Brec::uch of your staff during.the inspection of January 18 and 19, 1971, regarding the construction activities authorized by AEC Construction Permit No. CPPR-40.
As noted during the discussion, apparent deficiencies were identi-fled involving items not in conformance with the Three Mlle Island Unit 1 Final Safety Analysis Report or which may otherwise raise questions concerning the adequacy of construction. These items are as follows:
I Volmse II, Section 5 of the FSAR states in part: "The reactor building has been designed under the following codes:
Building Code Requireme..i.s for Reinforced Concrete, ACI 318 63
. Specification for Structural Concrete for Buildings, ACI j
301-63, except as modified in the design and quality control of this building. '
i Building Code Requirements for Rainforced Concrete. ACE 318-63, paragraph 103, states in part; "(b) Wen the temperature falls below 40*F
.., a complete record of temperature shall be i
kept.
Paragraph 605 states in part; '(a) Concrete shall be maintained j
above 500F and in a moist condition..
Building Code Requirements for Reinforced Concrete, ACI 301-63, 4
paragraph 1202, states in part; '(a) cold Usather* - Wen the mean daily temperature of the atmosphere is less than 40 F, 0
the temperature of the concrete shall be maintained between 50 nnd 700F for the required curing period. W en necessary, ar-J.,...
omer>
DATY > '
}& [>
. nonnicu t t/.id.
nowara_. _xi
=a n._..
.-c 2/11/71 Form AIC-Ste (Reo 9 53) AECM 0240 A~u. s. sovs.wsnt e.dvins eer.cs; iosa o.sao.sov l
3 f [.-
]
i ran",cmento for heating, covering, insulating, or housing the concrete work shall be made in advance of placement and shall be adequate to maintain the required temperature and moisture conditions without injury due to concentration of heat.
- Detailed recomendations are given in 'Reconsnended Practice for Cold Ucather concreting (ACI 306)'.
j ACI Standard 306-66, paragraph 3.1 states in part; 'Before any concrete is placed, all ice, snow, and frost should be com-pictely removed and the temperature of all surfaces to be in contact tith the neu concrete should be raised to as cicae as may be practical to the tc:aperature of the new concrete that is to be placed thereon.
'Mie Three Hile Island Unit 1, Quality Assurance Procedure QC-30, nevicion 2, dated February 16, 1970, states in part: 'If a con-dition arises uhcrein the UELC Field Supervisor-Quality Control l
determinen that project work or taajor portions thereof must bc l
stopped in order to preserve the quality of the project, he l
shall so inform the UE6C General Superintendent and the Home l
Office Quality Control Engincec.
. In the event that the l
General Superintendent, from the total project standpoint, does not agree uith the rococrsendations of the Field Supervisor -
Quality Control then he may decide to continae work. The Field si Supervisor - quality Control than will report the matter to 7 the UE6C Project Manager and his rococumendations to the Hanager of Reliability and Quality Assurance in the Home Offfce for im-i mediate resolution. However, the Mot-Ed Project Manager and/or Met-Ed Site Quality Arsurance Representative are authori;ed to l
initiate additional corrective action including the order to stop work.
Contrary to the above, site records indicate thatapproximately 230 cubic yards of concrete were poured in a fuel handling building wall (Gilbert Ansc ciates, Incorporated, Specification No. SP-5406) from cicvation 331 feet to 346 feet, running north and south 17 feet uest i
of the reactor centerline, at a tino " hen three measured concrete surface tenperaturen uere icnc than 31 F and the ambient temperature i
saa 150F.
e i
Please provjde us, within 30 days, uith your connents concerning t,hesc acms and any steps uh' th have been nr rill be taken to correct theia and to ninimize recitrrence, including any appropriate changes that i
have been or vill be made to your quality assurance program.
U 4
s
'hould you have any :,nections concerning the mattore diccussed in thin letter, you nay cornunicate directly.iith this office.
Very truly yours, Pobert
. P.irlex!n CO 1:DtEl Director l
i I
i l
i l
l l
l j
t I
l
.~
m.
A METROPOLITAN Eo: son COMPANY P. O. Box 542 RsAsues. PanssYLTAMA 19803 JOHN G. MILLER -
Yk= Pmadent ed Chef Endow March 8, 1971 hh,. c,$Y1
'.n, 4
U. S. Atomic Energy Commission 1
Division of Compliance Region I l
970 Broad Street Newark, New Jersey 07102 Attention: Mr. Robert W. Kirkman, Director Re: D/C letter dated February 12, 1971 Three Mile Island - Unit No.1
~- '
- ]
Dear Sir:
This will acknowledge receipt of your letter of February 12, 1971 concerning the pouring of some concrete in the fuel handling building of TMI #1 not in accordance with the applicable codes.
Your letter suggests that this deficiency was identified by your inspectors during their visit of January 18 and 19,1971.
I wish to point out that this deficiency had been identified on the dagr of'the pour (January 8,1971). and that corrective action was under discussion and re-view prior to the visit of your inspectors. Since then the following corrective steps have been taken to (a) determine the acceptability of the concrete that was placed, and (b) prevent a repetition of such an occurrence:
- 1. Cores are being taken from the concrete joint at appropriate locations and these cores will be tested to deter:nine whether the concrete meets specification strength requirements.
' ',Q j
~
- 2. The Inspector's Concrete Check-Out Sheet has been modified to g
require the s.ignature of the UE&C Q/C representative before any concrete placement is allowed to be made.
- 3. UE&C quality control procedure QC-30 covering work stoppage is being modified to clarify and emphasize that significant deficiencies 'noted by the UE&C Field Supervisor of Q.C. shall be brought immediately to the attention of the Manager of Reliability and Quality Assure ce 12 the UE&C home office and the UE&C Project Superintendent for corrective action before proceeding with the work.
The Met-Ed Project Manager and/or-Met-Ed Site Quality Assurance representative will also be notified immediately. Furthermore, the Met-Ed Project Manager has delegated authority to stop work to the appropriate Met-Ed Resident Engineer until corrective action is taken.
/
truly o 9 x. '
L.
/
J. G. Miller i
.% v A5U.
lf*
& nf '
. 0 v. /.. W O N
\\.,
g,,,., 4
....~.a
..a,
.y ;,/
~
-"'.m 9' W.R ; ; myg 4.-
y..
- k. C. SeYoung,,usist.aa 3;:vetor M s
' ; 7 C. G. L>ng,.'niel. Net Branen 2 [f.ddb TH40:
1 KiETSC VITH HETMJOLIT.4 2LISSN Gi 7d.u.'t. NILE ISLG'D NUCLEAR UNIT NO.1 l
We ra wich reprem.cacives of F.etropolitan Edison Company and their ve;,cora and consultants on February 23 ano 24, 1971 to discuss the Three Mile Tsland.s. clear Station Unit No. 1.
A list of attendees is attached.
The purposu of the sa:ecing was to discuss Amendments 15 and 17 which cont.41aed the ar wers to our first qs.estion list. We also discussed the i
itema that are at issue oetweer. us, and the proposed schedule for our
.1rst Acid meeting.
En ma.y ca,um *.ve informed the applicant of our final
{
~
or near final positions, which were discussed at a recent Task Force ne s ting.
, SITE 1.
Meteorologv We informa &tropolitan Edison cta.c the data made available so far did not substantiate the asserted diffusion values.
In fact, we were not able to conclude that the two-hour accident meteorology of g
Pasquill-/ and 1 meter-ppr-second was justified. On the very limited data made available so f ar, it may be that a wind speed of 0.5 meter per second is warranted. However, the applicant has installed a new
.mtaerology cover which has Delta 7 instruments would provide more information on the Pasquill c.,-.ditions actually present. We agreed to have a wereorcaory r.eenn,; su..<.tica during the week of March 15, ato furt't.e discus s tne cate. d...: e.ava been recantly generated from che Delta T in=trument. At cha: 1.ceting. we expect to have r,r. Vanderhovan, sur coc.:culcant from NOAA. Meteorology therefore remains an unresolved issue.
2.
F o:>c We revuvev d.e pM. '.2.e z ma flood calculations with the applicant.
The rn.c
...r t..t dr-C;1:,ar: Associates, Mr. MacImmore, and the 3RL taf f. yr.;;imr... c, ;ut.n:.;;c.:., diacussed the issues in detail at
.:. s, e,. r-:.s r.:. u
- ..z...
. ; c) :=.;:..u:. co the saain weting and z.rar.zc. ;hu a 'd ne....
- .f or;. cicn that we require, in order to cospbts car ecv. -v.
- . Ia;.ct.1. we.
- .te satisfied with the procedures
- ..>t C:it e::t hu ;.
c.:..:.a req. rud information constitutes a 1
i i
rs
~&~'
\\ \\ O 'b U Y D
\\% { Q*
. written verifiestion of our verbal understanding. The applicant agreed to furnish the required information on the basis of the m
understandings reached at the meeting. This item therafore can be considered as informally resolved, subjact to satisf actory documentation.
REACIOR DESIG1 1.
Fuel Design We requested the applicant's reactor vendor, B&W, to discuss:
the fuel design, in particular the high burnup tests that have been performed by B&W; the recently observed fuel pellet abnormalities; and the procedures that B&W uses in calculating fuel swalling and consequent clad strain. Neil Hooker of B&W gave a presentation on this subject. He stated that both the pellet vendor and B&W have QA procedures on the cladding and the fuel pellots. The obaurvations to date show that the fuel pellets have been remaining well within the QA tolerances. He stated that the small amount of chipping and flaking that our reviewers had noticed during a tour of the fuel f abrication plant were only a minor aberration in the pellet design and that the pellet diameters were not becoming excessively large.
In regard to the high burnup tests. B&W personnel stated dhat it was not their intent to prove current designs with the high burnup test; rather they were aimed at advanced designs. They state <' that they g
could not get the same fluxes and enrichmants at the B&W test reactor, therefore the coamercial design and the high burnup test do not have a one-to-one correspondence. They have not cospleted their evaluation l
of the high burnup tests. They do intend to come to DRL with a presentation on this subject when they complete this work, sometime in 1971. At present, they do not plan a formal report.
~
Hooker stated that B&W calculates clad strain in the same manner that he believes the other vendors do.
We asked, and he agreed, and l
Hetropolitan Edison agreed, that the details of how clad strain is calculated be documented in a forthcoming amendment.
l 2.
Burnable Poison Rod Assemblies (BPRA)
Hooker described the design criteria for the burnable poison rod ass emblies. They were:
(1) the zircaloy tube should be free st anding; (2) there should be no clad strain due to diametral growth from thermal or radiation effects on the poison material and (3)" there should oe no clad strain due to axial thermal or irradiation swelling.
I
, l They provide a cwelve asil diametral gap between the pellet and the clad.
At end-of-life this gap will not be filled. For the axial strain they provide a design margin of 13 inches, using corrugated d
spacers.
They predict only 7-1/2 inches of axial growth. The helium b
pressure from the BM reaction will be approximately 600 lbs. at and of life. The clad thickness of the zircaloy tube is 32-1/2 mills.
We agree that B&W had properly assessei the safety aspects of the BPRA's and consider this item resolvea.
3.
Pressurized Fuel The Metropolitan Edison Unit I will use pressurized fuel assemblies.
This will be documented in the next amendment.
PRIMARY COOLANT SYSTEM 1.
Flywheel Inspection We asked B&W to summarize the flywheel inspection criteria for the i
primary coolant pumps. B&W told us that, at the time that increased flywheel inspection was becoming a regulatory requirement, the TMI-1 primary pump motors had already been fabricated, with the flywheels shrunk on.
The motors are manufactured by Allis Chalmers.
In order g
to provide faspection to the extent poss!ble, AC performed an J
inspection on the upper face and outer rim of the upper flywheel on each pump, and took the flywheels completely off one pump.
By drilling calibration holes, they determined that they could measure a flaw size approximately 3/4 of a 5/16 inch hole,1/2 inch deep.
Or, they estiente that a flaw in the general size of a 1/4 inch diameter by a half inch deep is detectable. They have co g uted the critical flaw size for the large flywheel (72-inch diameter);
approximately an 8.4-inch radial crack from the bore out is required before critical strusses are reached. We asked that this information be doctamented and Metropolitan Edison agreed to furnish it.
Based on our inforan1 understanding we believe this item to be closed.
In a related discussion Metropolitan Edison informed us that due to problems that had developed with the Bingham pump, which they intended to use on unit 1, they have decided to switch to Westinghouse pumps.
The changeover is not as severe as it was on Oconee primary system ubich was assembled; TMI-1 welding has r.ot started. We told Metropolitan Edison that they should document the cht ge and verify the stress calculations that might be affected by the switch in pump design.
~
~
2.
Fracture Toughness of Primary System We discussed with Metropolitan Edison the recent changes that have p
taken place regarding the determination of fracture toughness in the primary coolant system, including the vessel. We told them that our position on Oconee was, since certain brittle fracture data were not available, that we would use a conservative pressurization temperature.
This temperature limit was 275*,
Below that temperature the primary system pressure could not exceed 550 psi.
Above that temperature the pressure could go to full system design pressure of 2200 psi.
When more information is realized through operation of the plant and from testing of the surveillance specimens, this temperature limit may be lowered. Metropolitan Edison understands our position. We expect Metropolitan Edison to adopt the same general temperature pressure limit in technical specifications.
3.
Vibration Monitoring Metropolitan Edison proposes a confirmatory vibration monitoring system. Neil Hooker of MW discussed some preliminary vibration monitoring tests that had been performed at the MW shop in Barberton, Ohio. The Three Mile Island internals, weighing some 300,000 lbs.,
were instrumented with accelerators and subjected to shaking action by a vibrator and impact action by a rubber mallet.
In general, the MJ measurements confirmed some preliminary design calculations and also confirmed the ability of the instrumentation to provide data during hot functional tests. We stated that it was our opinion that they i
should do either confirmatory vibration monitoring.or that they should remove the internals for inspection for undue wear, galling, j
etc. af ter the hot functional tests. Note: At a subsequent meeting, we decided that confirmatory vibration monitoring would be sufficient and that the applicant is not required to remove the internals.
However, we do intend to urge him and will so state at our technical specification meeting to visually inspect to the extent possible the j
core internals after the hot functional tests.
4.
Feedwater Ring Header We told Metropolitan Edison that during our Oconee review we required additional inspection of the welds of the primary system in the l
vicinity of the feedwater ring header on the steam generator, since it could not be established on Oconec that a failure of the primary system would not cause a subsequent failure of the secondary system.
However, the Gilbert representative showed us detailed drawings and )
i 1
1 5-referred to the explicit design basis in the FSAR whereby restraints are provided on Three Mile Island that were not provided on Oconee.
They have designed a primary system such that the piping cannot prop-i e
agate a failure at the feedwater ring header area. R ia appears to be sufficient justification for not requiring increased primary system inspection.
j 5.
In Service Inspection We asked the applicant to describe the extent to which the ASME Section 11 code for inservice inspection could be ut11'ized on the primary system. He applicant noted that there would be some areas that he could not inspect to Section 11 standards, due to access.
We asked them to amend the FSAR and to be prepared to incorporate in the technical specification bases the extent they do not comply with Section 11 and why.
6.
Decay Heat System Isolation Valve The MW design provides two isolation valves between the low pressure decay heat system and the high pressure primary system. Between the two isolation valves there is a small tell-tale relief valve which is sized on the basis of only a minute leakage from the high pressure s ide. One of the high pressure isolation valves is provided with an interlock to preclude inadvertent operation. Although this design is 3
not strictly in accordance with our proposed new standard on isolation valves we shall accept it, due to the as-built nature of the design.
7.
Once Through Steam Generator l
We asked MW if they had completed the vibratory measurements that they were taking on the as-built steam generator. They have completed the test; a supplemental report is in preparation.
STRUCTURES WERE COVERED AT THE SECOND DAY WEDNESDAY FEBRUARY 24th AND THEREFORE IS INCLUDED AT THE END OF THIS MEN)RANDUM.
'l ENGINEERED SAFETY FEATURES 1.
Thiosulfate I
We told Metropolitan Edison in November 1970 we had listed for.r conditions relevant to the use thiosulfate and we c Led them to what extent they had been considered. Upon request, we reiterated the four items; they were:
(a) that a pH monitor should be provided, I
4 i
~.-
-s J
o' 4
'(b) that the ability to replenish the sodium hydroxide tank should be provided. (c) that the use of copper and aluminum should be kept at a 5,,,
minimum and (d) that the thiosulfate storage tanks should be monitored y)'
frequently. Metropolitan Edison agreed to decisment in the next amend-ment items (a) ar.d (b). We said that from their design it appeared they had already complied with item (c), with the comment that they should not subsequently add copper or aluminun objects inside con-tainment. As for item (d), we said that that item would be covered with the technical specifications.
1 In regard to the removal credit for thiosulfate, we said that our position had not changed. B&W and Metropolitan Edison are aware of a
how we calculate dose reductions. Gordon Burley described briefly our current model which shows doses slightly above Part 100 for the loss of coolant accident. Bill Nischen pointed out that the exact value was 326 rem at this site boundary. The potential for reduction i
in the meteorology to half-a-meter per-second wind speed could double the dose. Burley said that he was considering, and it was under internal review, minor changes in the DRL evaluation model which could bring their dose from slightly over to slightly under Part 100. We notified Metropolitan Edison that we would commmicate our final j
position with regard to the licensing of this plant at our meeting to be held during the week of March 15th.
This item remains unresolved.
2.
Emergency Core Cooling System Report B&W said that they expected to file this report on schedule next week.
We said that we would try to have an initial evaluation in approxi-mately six weeks. However, this evaluation would not be timely with respect to the ACRS meeting in May., and therefore a subsequent ACES meeting on this and perhaps other subjects is foreseen. Since the report is not in hand and since we are presently committed to evalua-ting the ECCS report before final TMI-l resolution with the ACRS, this item remains unresolved.
3.
Engineered Safety Features Instrumentation We inquired into the design of the level indicators for the borated water storage tank and core flooding tanks. Contrary to what the PSAR shows, two (not one) level instrtaments are provided on the borated water storage tank and th two core flooding tanks. Thus redadancy does exist, although it is not clear what independence exists.
Don Sullivan asked if IEEE-279 was used as a design basis and B&W said I
I m
that it was not.
We said that we wanted Metropolitan Edison to doctament this information in the next amendment. We reserved decisions h
on the adequacy of the redundancy that is provided and the degree to l
which they don't aset IEEE-279. This item remains unresolved at this t
time.
4.
Spray Syster: Actuation Set Point i
We asked why they set the spray system actuation pressure at 30 lbs.
They responded that they saw no reason for turning on the spray system for pressures lower than that, and that inadvertent action of the spray would c wate a housekeeping problem inside containnsat, to say the least. They noted that some f acilities even provide a time delay to preclude inadvertent actuation. We said that this is properly a technical specification discussion, but we thought that advance notice should be given so that B&W would have time for preparation. We suggested 10 poi as a lower value although we admitted that there is very little time dif ference between 10 and 30 psi for large breaks.
This item remains tairesolved in that is is a technical specification item and we do expect to resolve it at that time.
~
5.
Hydrogen Purge We discussed our present plans for facilities such as Three Mile Island Ikiit 1.
Under certain conditions.we espect to approve purging of the containment, when the purge dose should be less than 10% of j
Part 100 guidelines. We told Metropolitan Edison that we had not completed our dose calculations although it appeared that the thyroid purge dose would be within the 30 rem value. We expect to complete.
we said, our calculations by next week using an estimated annual-average meteorology. We told Metropolitan Edison that we would consuni-
- l cate our calculations to them about March 15.
If we can agree with the i I applicant that the whole body and thyroid doses at the site boundary l
due to purging are less than 10% of Part 100 guidelines then we do l
expect to accept the purging concept.
i l
We have additional requests regarding the purge equipment for which Metropolitan Edison must provide additional information. We told i
Metropolitan Edison that we wanted them to document:
(a) the purge procedure; (b) the meteorology instruments that would be available; (c) the long time utilization of the reactor building fan coolers; (d) the details on the hydrogen monitor qualification tests; (e) the 4
ability to extract a grab sample; and (f) the effects of moisture on the hydrogen monitor.
B&W understood the items and agreed that the 4
next amendment would provide these details.
1 l
l 3
i i
-g.
6.
Long Term Cooling We discussed the phenomena that might occur in the vessel folloving a cold leg break whereby boiling would occur as the principal mode
' of heat removal.
Should this happen, the possibility exists for a long-term buildup of solids in the vessel, and subsequent inter-ference with core cooling. We agreed to discuss by telephone the ground rules for this calculation. It was agreed that B&W would instigate the phone call and that Ma6t Taylor and D. Ross would be parties to the DRL end of the conversation.
7.
ESF Pump Performance We asked if Metropolitan Edison or Gilbert Associates had received the first four safety guides that had been published.
They had. We asked if the engineered safety features pumps conform to the net Gilbert's positive suction head requirements of Safety Guide No.1.
answer was that they assumed a containment pressure to be in equilib-rium with the sump water temperature. On inspection of the sump water temperature time relationship, it appeared that for some time after the accident, the sump was up to 220*F.
They referred to figure 14-57 of the FSAR showing 220*F sump water at 15 minutes af ter the accident.
Since the use of an equilibrium pressure equal to saturation pressure at this temperature implicitly assumes that the containment pressure is above atmosphere, then we can state that the design of the ESF pumps is not in conformance with the safety guide regarding i
NPSR. The Gilbert representative stated that should the containment pressure for some reason drop to a lower value than that assumed, then the operator would have to throttle back on the low pressure injection flow or would have to stop one pump. They noted that the low pressure system, nominally rated at 3,000 gpm, could well be delivering in excess of this value due to conservativism in the hydraulic design. We told the applicant that we had not made a final decision on this matter and therefore this is unresolved at this time.
8.
Fan Cooler DesiFn We told Metropolitan Edison that the material submitted as Appendix 6A was generally satisfactory and that we had no further questions on the fan cooler design.
9.
Core Flooding Tank Isolation Valve B&W said that the isolation valve system for Three Mile Island was essentially the same as that provided on Oconea. We asked Metropolitan Edison to provide in the next amendment the following three items:
,+
(a) that there are two independent means of determining valve position; (b) that the condition of not-full-open be alarmed in the control room; and (c) that the power to the motor-operated valve be locked out during normal operation without interrupting power to the valve position indicators or alarm.
It appears that this design information can be supplied and this should not be an unresolved ites.
10.
Zine When asked, the applicant said that there was no exposed sine inside the containment.
INSTRtDIENTATION 1.
Post-Accident Ranges We asked how the range of the gasma instruments compared to the doses that might be expected after an accident. A Gilbert representative had some informal information.
It appeared that this information is satisfactory, and we asked Metropolitan Edison to document it in the next amendment.
2.
Diverse ECCS Signal for Reactor Trip We told Metropol1:an gdison that on Oconee we required a diverse reactor y
)
trip following tha loss of coolant sccident phenousson and stated that on the Oconee reactor a high building pressure of 4 peig was added as a reactor trip input. The applicant understood the problem and it appears that the same reactor scram method will be provided for Three Mile Island.
l 3.
Failed Fuel Detector As on other plants a assma monitor on the letdown line will be used.
the sensitivity of the instrument was discussed. The upper limit of sensitivity corresponds to about 10% failed fuel. We asked and i
Metropolitan Edison had agreed to document the informal information l
presented at the meeting.
I 4.
Use of Duasry Bistables l
We discussed at some length how dusmy bistables were used as a bypass apparatus on the TMI design. The specific designers were not present 4
at the meeting, and information was not readily available.
Our con-cern was that use of dussry bistables should be indicated in a manner i'
t 4
d
_y
,._-.,-.m._,
. readily visible to the control room operator. The B&W people and the Metropolitan Edi6on people were not sure vitether that in f act was the We agreed that this information could be furnished by telephone cas e.
and based on the telephone call, we would decide what additional infor-mation needed to be filed. B&W will instigate this telephone call and will call Don Sullivan directly.
5.
Qualification of Equipment, Topical Esport BAW-10003 B&W intends to file this topical report around April 1st. We stated that the report was essential to our review and it appeared that Sullivan saked the it might be part of a supplemental ACRS report.
His Gilbert people to what extent had tests been run on cables.
question concerned temperature, radiation, humidity. The Gilbert people did not have a ready answer and they will discuss this issue with Sullivan over the phone. At that time we will decide what additional information needs to be filed. In answer to Sullivan's question, the Gilbert people stated that now each diesel has a separate annunciator to indicate the out-of-service condition.
AUXILIARY 1.
Fuel Fool Filters Our soon-to-be-issued Safety Guide states that the fuel pool area Y
should be exhausted through ducts and filters to the unit vent. The Metropolitan Edison design provides for isolating the exhaust system and the supply system and essentially bottling up the fuel pool aus-iliary building. liatropolitan Edison feels that due to the airplane impact design that they have designed or provided a fairly leak-tight building. Therefore, they think that the dose to the public would be less if they simply turned off the fans than if they kept the fan running and have a high radiation signal. We had felt that filtration and ventilation should continue even if a high radiation signal existed downstream of the filters. As a result of discussions with the appli-cant, we are now not so sure. This item remains unresolved on our p art.
2.
Isotopic Analyses We discussed our proposed Safety Guide which would require isotopic analyses on the general order of quarterly, and after startup or unusual changes in activity. We noted that this v a suitable subject for our technical specification meeting. Bill Nischan also had some detailed questions on the Radvaste Systes. He asked how the
applicant would detect and control iodine released through the condenser ejector. They stated that they planned to use Krypton 85
/
as an indicator.
t Their plan is to establish a set-point on the meter on the basis that all of the activity (which should be Krypton 85) is hypo-thetically Iodine 131. If this set-point is reached during normal operation they will extract a sample and analyse it isotopically to determine the proportion of the activity that is Iodine 131. For example, if the ratio of Krypton 85 to Iodine 131 is 50, then they would readjust the set-point higher by a factor of 50. We asked if j
they intended to use an iodine monitor on their waste gas tank, in j
addition to isotopic analyses; they did not. Nischen had an addi-tional question in response to the answer to our question 11.2 con-carning concentrations downstream after a liquid release. He noted that there were four errors in a table 11-14 of the FSAR concerning leC values. He also asked why no Casium 134 was listed. Metropolitan Edison said that in a subsequent amendment they would correct table 11-14. Nischan noted that Molybdenum 99 was the primary isotope i
and wondered is there any procedure for further reducing liquid releases by concentrating on the most prolific emitter. There was no answer readily available.
In regard to our question and their answer to 11.3 we asked about the use of the Rashrasta Treatment equipment. Gilbert Associate representative pointed out that there was no way for high activity release to get to the affluent line
).
without going through bot 6 an evaporator and a domineraliser. They provide for redadancy in equipment. Regarding our question 11.5 i
we noted that Yankee Ron and Connecticut Yankee had experienced different values of release in that corrosion prodsets constituted the principal items, in contrast to the table in the Metropolitan Edison FSAR where the corrosion products are only a minimum. The l
Metropolitan Edison people pointed out that the Yankee core is stain-
]
less steel and that the Three Mile Island mit postulates a certain failed fuel activity.
3.
Fuel Cask i
We asked if the fuel pool could withsted the effects of a dropped fuel cask. They said'that for the portions of the fuel pool over which a cask might be moved (and there are interlocks on the crane to prevent any other movement) the fuel pool concrete is extended all the way to bedrock. Therefore the pool and its liner could with-scand the effects of a cropped cask.
i l
l t
s
,w--
.c
. t i
ACCIDEhT ANALYSIS 1.
LOCA Doses pfq.
We noted that in discussing the meteorology and thiosulfate we had already reasonably well defined our position on' accident doses. As a review, the two hour thyroid dose following the loss of coolant accident is still above Part 100 Guidelines.
If the meteorology gets down to 1/2 meter per second wind speed, then it is not impossible that the fuel pool handling accident would also approach Part 100 Guidelines.
l 2.
ATWS We notified Metropolitan Edison that the subject of anticipated trans-ients without scram would not be a review item for their operating license.
MISCELLANEOUS 1.
Staffing Hetropolitan Edison stated that they were in essential. compliance with AMS-3 standard on training and staffing and thet they would so document in the technical specifications.
2.
Industrial Security We notified Metropolitan Edison that we required a.small amount of additional information on the record concerning industrial security.
We referred them to our Oconee Safety Evaluation, page 75, and to the Duke Power Amendment No.11 for on a guide as to the quantity and type of additional information. They agreed to furnish this information.
t 3.
Technical Specifications i
Metropolitan Edison plans to make the first draf t available about the first day of May. We told them eight copies would be sufficient and to make them available informally.
i i
4.
Restricted Area We asked Metropolitan Edison to define on a large, scale map where the fence would be and what they considered a restricted ares to be.
They showed that essentially a 8-foot chain link fence topped by barbed wire l
13 -
would fcilow the dike around the island. No one can get on to the north end of the island without' authorization, as there is a guard at f.5 the mainland side of the permanent bridge. The south and of the h
island which is accessible via a " temporary bridge" will be continuously available to the general public..However, the road access to the plant from the south will be barred to casual travelers. There will be double fencing from that side separated by open land area which will be useful for spotting interlopers.
5.
Startup Tests We notified Metropolitan Edison that we required additional information to be submitted with the FSAR concerning startup tests. As a beginning ve referred them to what had been made available on Oconee We thought q
7 that the depth of th'e material could be increased in comparison to Oconee. The Oconee acceptance criteria were very short almost to the point of being meaningless. Metropolitan Edison agreed to file eone additional information. We concluded the first day meeting and recon-vened the following morning to discuss the structural design items.
6.
Structures Structures was category 4 on the agenda; agenda item I was the contain-ment design in general. We had asked a ntaber of questions in our September 1970 list., The answers which were made available in January
)
1974 were not fully acceptable. Most of our questions that we asked on structures revolved around the generally deficient area of their January 1971 response. Don Croneburger of Gilbert discussed their contentions that concrete strength under a biaxial stress condition has a higher ultimate strength value than in uniaxial compression. He referred to -
a November 1970 article in the journal of the American Concrete Inetitute proceeding V-67, Page 908. The article of the paper was " Strength of pi Plain Concrete Under Biaxial Stress". It appeared from that article lij that for the-case where concrete was loaded biaxially in compression that the ultimate strength could be increased by approximately a n
l factor of 2.-
If true, then the resistence' of the structure to an air-craft frpact would be considerably increased. Our consultant on air-craft impact design, Jim Proctor of Naval Ordinance Laboratory, was i
very interested in the utilization of that reference.
It was sufficiently recent that it had not been noted by Metropolitan Edison in the January 1971 Amendment.
Mr.. Proctor noted a number of deficiencies in the recent response to 4
i our question area regarding the calculation of the dynamic load factors, in particular, the utilization of a coarse approximation to the load time curve which does not preserve the momentum of the airplane. He also said that a factor of 20% increase that Gilbert assumed on the ultimate e
t
.,v..
e
+
. =.
I i
i strength of concrete in compression would be difficult to approve, in that it aaserted that the strain rate of the concrete was relevant in assessing the proper value of ultimate strength.
Mr. Proctor J
pointed out that, at the time of maxistan strain, the strain rate la
- cf sero, therefore there would be no attendant increase in concrete pro-He also noted some errors in some of the tables in Appendix 5A.
parties.
In each case the Gilbert personnel agreed that mistakes have been made and they agreed to correct these values in the next assadasne.
In discussion of the November 1970 paper in the ACI Journal, Doctor Gluckman said that what really exists in the does of the contain-ment is a triaximi field where there are two compression forces and one i
tensile. He thought that this might reduce the properties of citimats 4,crength, rather than increase. Mr. Chen Chang, a Gilbert employee, said that there would always be radical compression in the done and i "
there would not exist a tension field before impset. However, Doctor Gluckman pointed out that in the vicinity of the tendons there i-would be a tension field and that cracks would have groun. He said that if Gilbert could justify that due to impact there is radial cosqsression, then perhaps we could agree that the ultimate strength properties of concrete could be increased above their nominal value in a multi-axial l
a field. Doctor Gluckman reviewed the difficulties that have been encountered in a Turkey Point done sad said that these in part are responsible for our concern about the response of the Three Mile Island dans to an aircraft impact. We agreed with Metropolitan Edison and l
E l
Gilbert to have an additional meeting during the week of March 15th on the subject of aircraft impact design. At that time, Gilbert espects to have corrected their tables and have drafts of the additional infor-nation which should be available for filtag.
i Our next question area concerned the calculations of thermal gradients and stresses in the vicinity of variable thickness zones, such as the transition from the base to the umil and in the vicinity of the ring girder. They use the finite element method for the calculation of j
stresses in the transition region. They had additional information on moments and shears that were not included in Figure 5B-18 of the l
application. They had calculated the temperature profiles for one-half day, one day, two days, six days, and twenty days after startup. They l
did not do stress analyses for all conditions. It appeared that we were satisfied with their informal response. We asked if they had considered slightly higher temperatures which would give slightly higher stresses should an event such as happened on Dresden 2 also occur at Three Mile Island.
Our next question area was on the use of.85f' as a design basis. They stated that they use ultimate strength value 8f.85' to design to
dimensions, and to size the reinforcing. They did not use it in the final design however. We said that we would like to know the maximum compression in the structure and its relation to f'.
On the next subject of bond and anchorage stresses, Doctor CluEkman said that the recent Ims Angeles earthquake showed that rebars had been pulled out of the concrete rather than destroyed. He wanted to know therefore what are the critical bond and anchorage stresses and the relationship to the code. They stated that sine 18 bars are provided in the base some and at the ring girder sone. They are anctored in the wall on the inside face where compression exists.
We asked aboIst the existence of shear stress and, had they been cal-culated? Their response was that the shear stress was about 24 psi near the ring girder. We asked if the shear stress influenced the allowable ultimate compression stress for the structure. That is, could there be everywhere enough prestressing to handle the slight tensile forces?
We noted that they used load factors equal to 1.0 and asked if this was designing for rupture, in other words, would the stresses always be below.85f' for the concrete, or.9 of the yield strength for the steel?
We asked teen to answer this by giving an example of the margin of safety.
They agreed to provide the figure.
For the same type of information in the anchorage sons we asked what the safety factors would be. That is, would f' be reduced in the presence of tensile stresses? We said that they co61d answer this by giving an
)
example of the high sdress under the bearing plate and the relationship j
of tensile stresses at that value.
On the subject of reinforcing on the inside the concrete near the liner, l
we asked what the actual compression forces would be on the concrete.
l They said that the compression would be approximately 900 psi. Up near the ring girder they do get some tensile forces and they have provided steel there. They also get tensile forces on an aircraft impact.
We brought up the subject of surveillance of the structure and noted that when we discussed this item in the technical specification meetings, we would be discussing such items as the number of tendons, the location, frequency of the test, and how to pull out a sample wire. We said that we did not intend to accept an unstressed wire for surveillance. However, Croneburger of Gilbert said that stress corrosion.has been proven not to be a problem. We asked if they had considered the number of tendons that should be inspected. They had used as a beginning the recommendations of the ACI 349 Committee for surveillance of the tendon anchorage zone. The Committee reconsiendations are 2% of the tendons which for Three Mile would be 13. Metropolitan Edison plans to inspect 15. For lift off tests the i
s Committee suggests 1/4% which for Metropolitan Edison would be 3.
They intend to lif toff 6; they also intend that all 6 of these to be vertical tendons.
Their justification for using a11' vertical tendons in the
,'r lif toff test was that the results would not be impeded by friction of the tendon wires in a curved conduit.
We noted that they had not provided the allowable bearing values for the structures adjacent to the bedrock. We asked what did they actually use, and to give the saan information for dynamic canditions. We said that we had looked at their material on surveillance of the structure during the proof test. We noted that although they were taking three meridional j
measurements, we might prefer as many as six, with a smaller number of points per measurement. They said that this seemed to be excessive in j
cosparison with what had been done recently on other prestressed contain-meats.
We brought up the subject of the seismic instrument to be provided and said that we would like information in the FSAR concerning: how the instrument will be maintained; what will be done when the instrument has recorded a signal; and how the signals will be processed and interpreted.
They said that the sensitivity of the instrument was.01C, and that they would inspect periodically. A local indicator would signify that a record had been made. At that time, the record would be played back.
If the acceleration was greater than 1/2 of the design basis earthquake, that is, if the acceleration was greater than.03G then they would W
digitize the time history from the record to get a response spectra.
This in turn would be cospered to the design.
s l
We brought up the fact that recent construction at their facility had used concrete which was poured in nonconformance with the specifications.
They said that they were going to check the 28-day compression specimena on the concrete that was poured that day. There were some 200 yards I
involved where the pour was interfaced' at a surface below 32*F.
Based
\\
j on the 28-day specimens, they will decide what to do next.
d We next discussed the dynamic analysis of piping. We said that the AEC has not agreed with the Biggs and Roesset method for dynamic analyses.
Chen of Gilbert said that it depends on how the Biggs and Roesset method i-is used. He said that Biggs did not use a single degree of freedom J
system and referred to the 1965 paper entitled " Earthquake Response of Appendage on a Multi-story Building", by J. Penzien and A. Chopra, given at the third world conference on earthquake engineerfag in 1965 at New ealand, Volume 2.
Chen discussed what had been calculated by Stone and Webster on their Beaver Valley calculation. Dave Lange of DRS asked how is resonance handled in the Bigge Method. The answer was that a h
1 t
_-. - _ - _ - - - - =
e.
2-degree of freedom model was used to calculate the response in the resonance region. Lange felt that the time history can be conservative but the superposition of modes which is used in the Biggs method can i
possibly not be conservative. Lange said that the only way to demon-strate conservatism was to do a multimass time history analysis. Gian said that he does not think that the time history method is fully j ustified. Lange's response was that the time history envelopes the.
response spectra for this site. We noted that what ERL has come to refor to as the " Robinson Fix" could be employed at Three Nile Island also. This involves the application of pipe supports at a much more frequent interval. The usresolved items on structural desis include the aircraft impact design, some elements of the static desip involving the presence of tensile stresses, the dynamic analysis of piping, and certain aspects of the tendon and structure surveillance program.
The last structural subject discussed was the cavity desip. Gilbert summarized the final calculations. They said that they had provided in their final design a vent etwa of 141.6 square feet. This corresponds to blowing out of the insulation arossad the primary pipe from the cavity to the steam generator area.
For a 14.1 square foot pipe break the cavity pressure goes to 186 psi. At that point, they have an estimated 45,000 psi in their rebar of the outer fibers of the cavity. They had specified in procurement that the reber should be at yield at 40,000 psi, however, the as-bought materials were somewhat higher. They do not expect therefore gay significant deformation of the cavity. They do expect cracking, b8t no propagation of concreta missiles. The pipe tunnel for the primary piping is lined with a staal liner which served j
as a form for construction. We stated that that information was satisfactory and that no additional information on the cavity would be l
required.
t
.f
.)
D. F. Ross l
Reactor Projects Branch 2 Division of Reactor Licensing l
l i
i i'
t
i
~,
O DISTRIBUTION:
Docket File PWR-2 Reading DRL Reading P. A. Morris F. Schroeder T. R. Wilson R. S. Boyd D. Skovholt E. G. Case, DRS R. R. Maccary Compliance (2)
DRL & DRS Branch Chiefs D. F. Ross F. W. Karas R. W. Klecker W. Nischan G. Burley D. Nunn L. Hulman J. Knight D. Lange D. Sullivan
\\
-5 I
LIST OF ATTENDEES Division of Reactor Licensing Division of Reactor Standards
,9 J. Knight fj W. Nischan D. Ross D. Lange D. Sullivan G. Burley D. Nunn L. Hulman Metropolitan Edison Company C. Long Consultants (Picker & Lowe Associates)
J. Herbein C. Bierman J. Miller F. Schwoerer J. Larizza J. Pickard General Public Utilities Gilbert Associatas_
F. Symons D. Reppert A. Larson J. Thorpe S. Reid J. McConnell C. Bitting Babcock & Wilcox R. Maclamore W. Meek J. Cutchin Chen Otang E. Nodland G. Ward W. Delicate G. Glei W. Smith E. Hooker F. Thomasson 4
s i