ML19262A331
| ML19262A331 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 02/13/1976 |
| From: | Arnold R METROPOLITAN EDISON CO. |
| To: | Reid R Office of Nuclear Reactor Regulation |
| References | |
| GQL-0229, GQL-229, NUDOCS 7910260656 | |
| Download: ML19262A331 (8) | |
Text
EC Ic e i 196 v2. wCLt An accuta Tons N'
DOCKET NUMC;M
.'C
.1. 5 ' '
um NR'; D:STRIBUTION con PART 50 DOCKET MATERI AL l
i FROM-CATE Cr DQCUMENT
- g.
luid
'!etropolitan z dt se,c
.s su I
2cadin;, PJ.
DATE RECElVE D a,i n
,,,,,,, 3 a 9r
[e76E TT E a
[j N O TO R IZ E D PRCP If.P'J T ( C R*.1 NUMBEH OF COPIES RECEl% ED 3 OR IGIN A L 3 UNC LASSiF IE D gpg 3 ggg q L3 COPY CE3CRIPTICN ENCLOSURE Lt-re their 1-16-76 ler...trans the fc11uwing:
Responses to questions concernir.S revicw te:S specs change requcs: 131..............
(t Cy encl re:'c) 1485 088 PL M NAME.
Three !!ile island 11 SAETY FOR ACTION /INFORMATION ENVIRO 2 7 t>
ASSIGNED AD :
9 g BRANCli CHIEF :
,[\\C l )
{$~
BliNC 1 CUIEF :
i A PROJECT.O. NAGER:
O A / /% C 5 PROJECT MANAGER :
/
LIC. t.S3T. :
17 e f.: en LIC. ASST. :
~
a i
f T
INTERNAL DISTRIBUTION M m' W SYSTJ'JJS SAFETY PD.MT 9YSTrM9 v:p]Tnn y cu
/
NRC PDR FEI E LD TrDESCO ERNST
/
I&E scu tornaq BE!'An0Y.';
EALLA"D
/, OE1D LM'&iS SPANGLER
/
COSSICK & STAFF E!{g;{"E3_EG IPPOLITO L
ing:
_ vAcenv SITE TECH ca3E KNIri!T OPERATING REACTORS CAC ILL i
uAgJ; SIIGiEIL STELLO STEPP i
H_WLESS PAULICKI Hull'AN OPERATING TECI!
PROJECT MANAGEME!.T REACTOR SAFETY
/
EISI'?.1:UT SITE ANALYSIS i
BOYD ROSS
/
SFAO VOLLE R P. C_0LLIUS NOVAK
/
EAER BUdCH I
HOU3 TON ROSZTCCZY
/
SCL'~NCEn J. COLLINS 4
PETERSON CHECK
/
GRIU.S KREGER MEL_T.E UEl.TE'./ S AT & I SIT. S i ETY & ET.'Ind S KOVl!CLT SALTg:_'g1N I A "l.I;'_c ~ S 2
l IllCT"r"';
I I DENT ^' /, 'C 7.L"7 I
EXTEW AL DISTRIBUT!CN CONTR OL NUM BE ii
( {_Lrrg:Mf.'.
- +i.4.A NATL LAD BE L.' ';;.il.;l JAIL lab
/I TIC RCG. V-I"
' ULEI W 'MC'UL) s /y M__iM IR ___
d LA IUR I
O'
l 3' / L
/
l i i ASt3
_.s J CC:an,t. '_.NT '
,I
/
V _isC','3LE__1Wu.?'.1^ !"? " -
t'"!
L.
ry S i t a_ g o o[ a o/
i !
i p-nc,m m.m.
~~ "
7 ~Cd
$m nxuss METROPOLil'AN EDISON COMPANY POST OFFICE BOX 542 READING, PENNSYLVANI A 19603 TELEPHONE 215 - 929 3601 GQL 0229 February I V,M Q,'
M-ihim.D c
Director of Nuclear Reactor Regulation
,.fg],71976 P =
'il Attn:
Mr. R. W. Reid, Chief
"' hy,,pten g
Operating Reactors Branch 4 V
U. S. Nuclear Regulatory Commission p
Washington, D.C.
20555 G
tu Three Mile Island Nuclear Station Unit 1 Operating License No. DPR-50 Docket No. 50-289
Dear Mr. Reid:
Attached please find the responses to questions asked by Mr. Vern Rooney and !k. Carl Berlinger of your office during their review of our Change Request 31 dated January 16, 1976. These responses were also transmitted by telephone to Mr. Berlinger on Feb. 4,1976.
Please note that in addition to the attached responses the following corrections or explanations were also conveyed to Mr. Berlinger on Feb. 4, 1976:
(1) On Figure 1 of " Justification for the Increase in the Maximum Allowable RPS High Pressure Trip Setpoi:
the abscissa (Rod Withdrawal Rate) should be 10-3 at the left hand margin and 10-3 at the right hand margin.
(2) the Loss of Electrical Load (LOEL) test referred to in the Reasons for Proposed Change (page 2) will be simulated by a turbine trip since the turbine trip transient very closely simulates a L0EL. The data from the turbine trip will be extrapolated to evaluate the effectiveness of plant modi-fications.
(3) the one accident referred to in the last sentence page 1 of the Justification refers to case #3 of table 2.
1485 089
)
1510
- fr. Reid Page 2 On February 10, 1976, we informed Dr. D. Bridges of your office that the 215
'lbs/sec assumed relieving rate for the two pressurizer code safety valves was greater than the sum of the vendor certified relieving rates for each safety valve (i.e. 172 lb/sec). Since a relieving rate of 215 lb/sec was not a conservative assumption additional analyses were conducted as explained on attachment 2 to this letter.
Should you have any further questions, please contact me.
Sincerely, 0
l, ;
old RCA: C'4S :pa
Attachment:
- 1) Responses to NRC Questions
- 2) Correction to the Analyses Eapporting the Increase in High Pressure Trip Setpoint 1485 090
ATTACHMENT 1 RESPONSES TO THE NRC QUESTIONS ON THE TMI-l HIGH PRESSURE TRIP SETPOINT INCREASE Re fe rence :
" Justification for the Increase in the Maximum Allowable RPS High Pressure Trip Setpoint,"
Transmitted with TMI-1 Technical Specification Change Request No. 31, January 16, 1976, GQL 0056.
With reference to the feedwater line break analysis transmitted with the referenced justification:
Question 1 What is the time at which peak RCS pressure occurs?
Question 2 What is the time at which the Pressurizer code safety valves open?
Response 1,2 The transients summarized in Table 2 of the referenced letter assume a complete main feedwater line break at time zero. The time of RCS peak pressure and of pressurizer safety valve opening from the beginning of the accident are as follows:
Peak RCS Peak Pressure Time When Code Case (Table 2)
Pressure, psi _a Time, sec.
Safeties Open, sec.
1 2637 7.75 6.30 2
2633 8.10 6.35 3
2655 7.95 6.35 Question 3 What is the relieving rate through the Pressurizer safety valves?
Question 6 At what pressure do the safety valves open?
Response 3,6 The code safety valves open at an assumed value of 2515 psia.
The assumed relieving rate through the two safcty valves was 215 lb/sec.
Question 4 What is the quality of the water through the safety valves?
Response 4 The pressurizer does not become solid with water during this transient. Only saturated steam is relieved through the safety valves.
Question ;
Is high RC pressure the first trip signal?
Response 5 High RC pressure trip setpoint is the first RPS setpoint reached during the transient.
Question 7 Explain the forcing function for the S. G. Heat demand following the break.
1485 091
Response 7 The accident simulation assumes the complete severance of the main feedwater line to one S. G. a t time ze ro.
Since the CADD code cannot simulate the secondary-side transient, the secondary-side response to the feedwater line break is calculated separately. A conservative estimate of the loss of S. G. inventory resulting from the F. W. line break is that the affected S. G. blows dry in 5 seconds, while the unaffected S. G. blows dry in 20 seconds. The feedwater line break is then conservatively simulated in CADD by linearly decreasing the heat demand from steady state at t= 0 to zero in 5 seconds in the affected S.
G., and from steady state at t = 0 to zero in 20 seconds in the unaf fected S. G.
After 20 seconds, the heat demand is maintained equal to zero in both S.
G.'s.
Question 8 What is the pressure vs. time behaviour for the three transients analyzed?
Response 8 See the figures on page 4 With reference to the startup accident analysis transmitted with the referenced justification:
Question 9 For the sensitivity study on Doppler, what was the moderator coefficient used?
Question 10 For the sensitivity study on Fbderator, what was the Doppler coefficient used?
Response 9,10 Both the Doppler coefficient and moderator te=perature coefficient sensitivity studies were based on the nom 4.nal BOL cycle 2 case with a high RCS pressure trip setpoint of 2405 psig and a reactivity insertion rate of 0.00038 AK/K/sec. This case results in a peak RCS pressure of 2579 psia (as shown in Figure 1).
In the Doppler coefficient sensitivity study, only the Doppler coefficient was varied, as shown on Figure 2, over a range which encompasses the values of the Doppler in cycles 1 and 2, both BOL and EOL.
The value of the coefficient for those four points is labeled on Figure 2 to show the extent of the range of the sensitivity study and that all reasonable values of the coefficient have been analyzed for effect on the transient pressure results. Similarly, the sensitivity study on moderator temperature coefficient involves varying only that coefficient.
Thus, the moderator temperature coefficient coefficient sensitivity study is -1.06 x 10~{or the Doppler AK/ m /F.
The Doppler coefficient for the moderator tegperature coefficient sensitivity study is -1.49 x 10 aK/K/F.
1485 092
It should be noted that the feedwater line break accident analysis presented in the referenced Justification is not meant to be an addition to the IMI-l FSAR. The purpose of the analysis is to show through a conservative modeling that the peak transient pressure increase due to the increase in the high pressure trip setpoint is acceptable and does not result in the safety limit of 2750 psia being exceeded. Thus, the difference in peak pressure between the cases 2 and 3 of the referenced Table 2 is the significant result of this analysis.
Question 11 Did the Start-up or Rod Ejection Accident Analyses take credit for the code safety valves.
Response 11 Both the Start-up and Rod Ejection Accident Analyses took credit for two safety valves lif ting at 2515 psia with an assumed relieving rate of 215 lb/sec. 1485 093
HIGH 7E SSURC. TRIP SE7'PcsvT wcgeAsc T MI-l STARTOP Acc.t bE 07 - BC) 3 C N'<.L c 2 H isu Tr<ewne.
SETPon nt -- 2406 vss%. 0.00030 A K/K]sk-COITH. RATC.
,I kvo,T"R1P g
_.....,......y
..p
+
tn 2Sco tJ v1 c
g)
^
cco 230o
~
J-
. u)
.s
_.._ y
. 77 _:.;:......- -.... _..
'. g -
. _ -.m.-...
,_x..,..
u
...;Z8 00 2..
.4 26 29 30
,32.
.M
.3 6 38 40 I6 16.
2 o 22.
7 li+
...t lo. n..,....... _,... _.......T. J M G,
.5 C C. _ _...
.1
.a...:.._
- e
=, *,
.a
=,
.e,...:.; _ _.
..t.
s e
~ ~ ~ ~
"rcEtu;nrE'A -LluC E52:".AK A4citd47"[. BoLi d'fCLC'2 h. P., WGH qsm4-
.... _. _ TRt? SETPotMr= 240 5 ysly..
2
._...:...;.-_ ~2.7ao
. 4._.l '. - -
. _ ;.y
=.. _ :.,'.. :....
~
LL 61 4.
2500 if
_ y....;./..
f_ _:..
u
.. 2 m
._.._._.j_....
g
. - 3..:....
. _. :..};._
. -g..
2300
.. ~.__... _.
-s-a LJ
.M 2l00 o
2 4
6 6
lo 12_. w.
lC
- 1. 8 2o 22.
24-26 2.f!
Bo
.TIMg ICc,
~ ~ ~ - '"- ^ RCD ETECTIDM ICoM O//a FF, - BOL., CYc LE. 2, HiGeiPens sOCC TalP sstrosur = 2aos %y. o,se f Ar;/g s.rcercu t.ederH.
. 27oo 1485-
.. e g __J..:...._....:.. q._.:.. _.....- -
y g
tn
..........i.
..N.
2sco O
u 3,
-.t-:.
y tu
- -i
.. "25co
- ;...___..._a...._;..._.:
.t '
..; 4..>...'.__ ;.. :... p.
l y
--3
. " V_ j42 J
.g E.- i.. ;
=. -
s.. g......-... _
....:. :a.=., -
.e.. :.. I
- .v. - ?:.. L... :: :... L.:. -. --.. :: - t.-. :. !.-. :. -...~-... a
.. j.;
.e
. c:
.. ' - - -;;4..:.-
- y..,
i
- t..
2'*O o 1.
4 6
8 to 82.
"+
16 sc 2.o
- 2. L.
zu 26 2.s do
- e.. ~. e
ATTACHMENT 2 CORRECTION TO THE ANALYSES SUPPORTING THE INCREASE IN HIGH PRESSURE TRIP SETPOINT IN TMI-1.
Reference:
Justification for the increase in the Maximum Allowable RPS High Pressure Trip Setpoint.
Transmitted with IMI-l Technical Specification Change Request No. 31, January 16, 1976, CQL0056.
In the ana'7ses presented in the above reference, the relief flow through the pressurizer safety valves was improperly modeled.
The result was that the flow rate through both safety valves was calculated to be about 215 lbm/
sec of steam, rather than the nominal 172 lbm/sec at 2515 psia; and the CADD-calculated pressure loss f actor for piping between the safety valves and the pressurizer was too large. As a result the safety vcives did not actually open until the pressurizer pressure reached a value higher than the nominal 2515 psia. For examplc, for the Feedwater Line Break accident, case No. 3 in table 2 of reference 1, the safety valves did not open until a pressurizer pressure of 2570 psia was reached. The net result of the two ef fects was an actual overestimate of the peak RCS pressure as calculated by CADD. The Feedwater Line Break accident analysis at 30L, Cycle 2 conditions with the high pressure trip set at 2405 psig was reanalyzed with a correct and conservative modeling of the safety valves. This modeling conservatively assumed the relief flow through both safety valves to be a conservative 156 lbm/sec, or 90 percent of the nominal relief rate at 2515 psia.
For this transient, a high RCS pressure trip occurs at 5.65 seconds from the beginning of the transient. Pressurizer pressure reached 2515 psia at 6.35 seconds, the time at which the pressurizer safety valves open. A peak RCS pressure of 2656, psia is reached at 8.9 seconds, after which the RCS pressure steadily decreases to 2520 psia at about 17 seconds.
Peak RCS pressure is only one psia higher than the corresponding feedwater line break analysis of reference 1.
The RCS pressure vs. time transient curve is nearly identical in shape to the feedwater transient curve shown in Attachment 1.
Because of similarities of this transient with the pressure transients for the limiting startup and rod ejection accidents at BOL, cycle 2, see attachment 1, the same considerations apply to these transients also.
The figures of page 4 of attachment 1 show that for all three transients, the time from pressurizer safety valves opening to peak RCS pressure is less than 2 seconds. The effect of the safety valves relief on the RCS peak pressure valve is therefore similar for the three transients. We therefore conclude that the analyses of the above reference as amplified by the responses in attachment 1 are still valid in supporting the propose.d high pressure trip setpoint increase.
1485 095