ML20091M183
| ML20091M183 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/06/1984 |
| From: | Hukill H GENERAL PUBLIC UTILITIES CORP. |
| To: | Stolz J Office of Nuclear Reactor Regulation |
| References | |
| 5211-84-2121, NUDOCS 8406110123 | |
| Download: ML20091M183 (5) | |
Text
e d
GPU Nuclear Corporation Nuclear.
- =rss48o e
e Middletown, Pennsylvania 17057-0191 717 944-7621 TELEX 84-2386 Writer's Direct Dial Number:
June 6,1984 5211-84-2121 Office of Nuclear Reactor Regulations Attn:
John F. Stolz, Chief Operating Reactors Branch No. 4 Division of Licensing U.S. Nuclear Regulatory Commission Washington, D. C.
20555
Dear Sir:
Three Mile Island Nuclear Station, Unit 1 (TMI-1)
Operating license No. DPR-50 Docket No. 50-289 EFH-LOFH Analysis In supplement to our responses of February 4,1983 G3-040), March 22,1983 (83-055) and December 9, 1983 (83-0349) and to our discussions with members of your staff on April 23, 1984, enclosed please find the revised results of our analysis for the determination of the minimum EFH flow for a seismic event with loss of feedwater (LorH) transient.
This revised analysis indicates that 350 gpm EFH will safely shutdown the unit with RC pump heat subject to a 20 minute delay in delivery of EFW to the steam generators.
Tests perfor ned in cold sh!tdown extrapolated to hot conditions demonstrate that 400 gpm is available under worst case seismic events. A summary of the calculations of the extrapolation to hot conditicns is also enclosed.
Detailed calculations, which were reviewed by M. Caruso (NRC) at TMI-1, have been revised to include the friction characteristic of the as-built (as-installed) EFH cavitating venturi. The revised calculations continue to indicate that one (1) EFH pump can deliver 400 gpm to two (2) OTSG's under hot conditions with the pump recirculation line open.
Sincerely, l
D.
kill VP-THI-l cc:
R. Conte pl g
J. Van Vliet
/
04799 - 0015u
'[I
)-
~ GPU Nuclear Corocration is a subsidiary of the General Public Utilities Corporation l
8406110123 840606
-PDR ADOCK 05000289 P
. Seismic Event eith loss of Feedwater
- Assumption a.
Only one (1) EFW pump (motor driven available) j b.
350 gpm.EFH delivered to the OTSGs due to EFW pump recirculation back j
to condensate storage tank (EFV-8 A/B/C locked open) following a 20 minute delay.
c.
Reactor at 102% of 2535 MWt at start of event.
d.
HPI. automatic initiation at 1600 psig.
e.
Reactor. coolant pumps heat included (16 MWt).
f.
1.2 ANS 1972 decay heat assumed.
Results PORV lifts about 7 minutes after the start of the event and operates cyclically for about 40 min.
The RC becomes saturated and the safety valves lift about 15 minutes after the start.of-the event and operate cyclically for about 10 min.
Pressurizer goes solid in 12 minutes.
(Safety valve operation and capacitics for LOFW (Tave - 650*F, Prcs =
2500 psig) are enveloped by EPRI test program for steam / water).
No core uncovery (minimum inventory of 7340 ft').
HPI-is automatically initiated at about 140 minutes into.the event.
No. Departure from Nucleate Boiling (DNB) during any phase of the transient.
RCS Pressure does not exceed 110% of design pressure.
. It should be noted that this analysis for LOFH is conservative in that no
-credit is1taken for the following TMI-l' system features and procedural requirements:
o. 0perator starting the EFH pumps on LOFH o
Manual action on loss of heat sink 04799'- 0015u-
=
1 1
u-u
n.
4'
.An. additional. case-assuming trip of the reactor coolant pumps at loss of 10
.subcooling was--conducted. The results show a much earlier HPI initiation
.(about 60 minutes--.into the event) and greater water inventory (8000 ft' min)
_o 10perator actions toitreat-loss of heat transfer HPILinitiation
. tripping of two RCPs Opening of the PORV at 2300 psi pressure o
HPI initiation upon loss of subcooling margin 1
M*-
- 0479g - 0015u di
~
At 4-
_v:
~
j.
b Emergency Feeedwater (EFH) System Determination of EFH Motor Driven Pump capacity by cold test and calculated extrapolation to " Hot" Conditions (i.e., OTSG at 1050 psig).
Reason for running " Cold" in lieu of " Hot" test
' Heat source du'ing HFT is limited to energy input from 4 RC pumps (16 r
MH).
This heat is removed by supplying approximately 100 gpm EFH Severe transients in RCS due to addition of 400 to 500 gpm EFH is main concern Cold Test Methodology Test procedure selected was TP 233/3
.TP 233/3 Purposest a.
Check EFH Annubar instrumentation and EFH flow 'ndications b.
Check performance of EFH cavitating venturis
-c.
Get data for EFH pump performance Test Procedure:
a.
Feed one (1) OTSG with one (1) motor driven EFH pump Pump recirculation line closed (allows suction flow indicating switch to be used as flow data point)
EFV-30 valve throttled Record flows (annubar & suction flow indication), OTSG level cnange and system pressures-b.
Feed one (1) OTSG with two (2)-motor driven EFH pumps
-.. Pump recirculation line closed EFV-30 valve wide open (in order to measure maximum cavitation flow.through cavitating venturis)
Record flowsL(annubar.& suction flow indication), OTSG level change;and system. pressures c.
Repeat a & b, abcee for other'0TSG Calculated extrapolation 04799 - 0015u-l
.g j
t
e a.
Determine-EFW pump flow from test data 1.
As measured by EFW pump suction flow indication 2.
As measured by annubar 3.
As calculated by OTSG level change (wide & operating iange)
- 4.
As inferred from cavitating venturi expected performance formula.
b.
Calculate average of flows in a.
c.
Plot average flow and pump discharge pressure, i.e., pump system head-flow graph.
Discharge pressure used includes a correction for the difference between the indicated pump discharge pressure, PX-67/PX-68, and the discharge header pressure, PI-476/PI-477 which is higher. PX-67/PX-68 read low because of friction drop due to bearing cooling flow.
-d.
Plot additional pump performance points from TP 273/3 test program (run in 1973-74) e.
Superimpose resulting system pump head curve (actual pump performance) on calculated system resistance curves for condition of one (1) EFW pump feeding two (2) OTSGs under " hot" conditions with pump recircula-tion line open f.
Intersection of actual system pump head curve with calcuated " hot" system resistance is extrapolated flow.
NOTE:
Comparison of actual test data with calculated system resistance for the portion of the system from the EFW pump discharge to the cavitating venturi shows good correlation of data with
. calculation.
Friction drop ~through the cavitating venturi and downstream pipe to OTSG is accounted for by calculation.
Pipe loss is calculated using Crane Tech. Paper No. 410; venturi unrecovered loss is calculated using vendor data for the as-built, as-installed venturi.
Extrapolated performance shows that one (1) EFW pump can deliver 400 gpm to two (2) OTSGs under hot conditions with pump recirculation line open.
04799 - 001Su.
1
.g-.-
t