ML20107G062
| ML20107G062 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 02/22/1985 |
| From: | Hukill H GENERAL PUBLIC UTILITIES CORP. |
| To: | Stolz J Office of Nuclear Reactor Regulation |
| References | |
| 5211-85-2027, NUDOCS 8502260290 | |
| Download: ML20107G062 (4) | |
Text
' ' '., g, GPU Nuclear Corporation g
gf Post Office Box 480 Route 441 South Middletown, Pennsylvania 17057 0191 717 944 7621 TELEX 84 2386 Writer's Direct Dial Number:
February 22) 1985 g
5211-85 202 Office of Nuclear Reactor Regulation Attn:
J. F. Stolz, Chief Operating Reactor Branch No. 4 Division of Licensing U.S. Nuclear Regulatory Commission Washington, DC 20555
Dear Mr. Stolz:
Three Mile Island Nuclear Station Unit 1 (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 Subcooling Margin Indication By letter dated September' 28,1984, NRC requested additional informatic' with respect to the Saturation Margin Monitor Loop Error Analysis for TMI-1. Subsequently, GPU Nuclear Corporation (GPUN) and the NRC staff
.t on October 30, 1984 to clarify the NRC concerns.
By_ letter 5211-84-2291 dated November 30.-1984 (Reference 1), GPUN responded to seven of eleven NRC questions. By letter 5211-84-2001 dated January 16, 1985 (Reference 2),
GPUN addressed the remaining four concerns.
3 The purpose of this letter is twofold. First, we would like to summarize:
the conditions under which the Saturation Margin Monitor is intended for use. and the calculated maximum loop error-associated with those uses. As
-part of this discussion, we would like to address both the assumptions.
presented in our loop error. analysis (Reference 2), and the assumptions presented by Mr. F. Burrows of your staff in telephone conversations of' January 29, 1985 and February 5, 1985,_and February 14, 1985.
. Secondly, we would like to reiterate the conditions under which the margin c to. saturation is determined based on the use of incorr thermocouples and RCS' pressure indication, in conjunction with steam table':.-We'will address the maximum instrument error associated with this method,' based on.the assumptions as presented in' Reference 2,'hs_well as' based onLMr. Burrows' set of assumptions.
8502260290 850222 hl O'
PDR ADOCK 0500028' P
PDe4 h.;
u.
GPU Nuclear Corporation is a subsidiary of the General Public Utilities Corporation
\\
5211-85-2027 February 22, 1985 The following discussions, summarized in the attached Table 1, demonstrata that under all postulated scenarios, the calculated error associated with determining margin to saturation remains under 23.4F which was found acceptable, in conjunction with use of the 25F indicated subcooling criterion for throttling HPI, by the Staff in SECY-84-237, dated June 14, 1984.
Subcooling Margin Determination Based on Saturation Margin Monitor The Saturation Margin Monitor is used to determine subcooling margin when the reactor coolant pumps are operating.
RCP's are not operated below 200 psig, since the RCP seals require this pressure to keep them staged.
Therefore, the Saturation Margin Monitor is not used below 200 psig.
The GPUN loop error analysis (Reference 2, Response to Question 2 and ), assumes that the worst case combination temperature, pressure and radiation errors would be bounded by containment conditions of 0 psig, 212 F and an integrated radiation dose of 4x106 rads.
We evaluated each type of er*or involved in the analysis, and concluded that the only systematic errors are the negative systematic error in the characterizer and the negative systematic error due to containment pressure (combined algebraically.).
All other errors are considered independent vandom errors which can be combined by taking the square root of the sum of the squares.
Based on these considerations, the maximum calculated error associated with the Saturation Margin Monitor is +14.95 F, occurring in the reactor coolant system pressure range of 200 to 400 psig.
To this error is added 1.3F to account for the instrument tap locations below the top of the hot leg U-bends. This results in a total error of 16.25F.
In telephone conversations with Mr. Burrows of your staff, he informed us that he considers errors associated with containment temperature and radiation as dependent parameters that should be combined algebraically.
Also, he thought that the effect of containment pressure, which results in a negative error contribution, should be ignored for added conservatism. GPUN maintains that our methodology, as presented in References 1 and 2, represents a conservative but more realistic set of assumptions.
Nevertheless, we have performed an additional evaluation, algebraically combining the worst case errors associated with containment temperatura and radiation.
However based on the locaticn of the pressure transmitters, (i.e. :n open containment areas which are remote from all potential SBLOCAs and thus not subject to impingement or high localized temperature), the transmitters could not experience containment temperature conditions without experiencing containment pressure. The error cal slated based on this revised set of assumptions (i.e. algebraic combinacion of temperature and radiation, and accounting for the containment pressure) is 18.57F*.
This includes the physical configuration factor.
M I II I
-.s 5211-85-2027 -
February 22, 1985
]
I As further suggested by Mr. Burrows, we analyzed the thermal time lag of the pressure transmitters and determined that it would not be more than 10F* for a small break LOCA while the containment cools down. The negative bias due to containment pressure is sufficient to compensate for the error d te to the 1
containment temperature above 212*F plus a transmitterLtemperature twenty L
degrees above containment temperature; thus, no further consideration is
~
required.
4 Subcooling Margin Determination Based on Use of Incore Thermocouples d
The operator does not use the Saturation Margin Mcnitor to determine subcooling margin if the RCPs are not running. Under those conditions, the margin to saturation is determined by manual calculation, based on reactor.
1 coolant temperature and pressure indications available in the control room, and steam tables. Determination of subcooling margin when the RCFs are not running is discussed in Reference 2, Response to Question 5.
Using the same methodology for combining errors as in the Reference 2:GPUN Saturation Margin Monitor alarm loop error calculation, we calculated an error under harsh containment environmental conditions of 17.76F* at reactor coolant' system pressures down to 175.psig, which is the lowest RCS pressure at which HPI throttling is a consideration (See Reference 2, Respr to
' Question 11). The error decreases _ for higher RC pressure. An err h
reading steam tables of no more than 0.5F* is added to these valur.,
tiong with a physical configuration factor of 1.3F", for a maximum error of 19.56F*.
Assuming the same conservatisms presented by Mr.! Burrows for the Saturation Margin Monitor: loop error analysis, we perfonned an additional analysis under harsh containment environmental conditions, combining = temperature and radiation induced errors' algebraically. With these considerations, and' including an error--of _0.5F* for use of the steam' tables and the'1.3F*-
. physical configuration factor, the ' worst case. combined error is. 22.7F*.
. Conclusion
' Based on the above, we conclude that the capability provided at TMI-1 to Lassess margin to saturation is. appropriate for-all conditions for'which this;
~
detennination is' required.3
' ~
Sincerely','
j
,- o x
~H. D.:Hukilli
~~#
- Director TMI-1 s
~
s HDH/SK/ dis-1026fE
,v.
g 0174At m
cc: fJ. Van.Vliet :
lP. KadambiJ 7.
I^
?
.=
=
l
?'_
j'
~
W w
u
m<en%M. m s bi y
Table 1
SUMMARY
Reactor Method to Worst Case Calculated Error (2)
Reactor. Coolant (l)
Coolant System Determine Margin
- Pumps Status Pressure to Saturation Case 1(3)
Case 2(4)
On
> 200 psig Saturation Margin 16.25F 18.57F Monitor Off
> 200 psig Average of 5 highest 17.74F 20.54F Incore Thermocouples/
RCS Wide Range Pressure Indication /
ASME Steam. Tables
(
Off 175 - 200 psig(5)
Average of 5 highest 19.56F*
22.70F Incore Thermocouples/
RCS Wide Range Pressure Indication /
ASME Steam Tables f
?
Notes:
1.
Reactor coolant pumps do not operate at reactor coolant system pressure less than 200 psig.
N 2.
In all cases, the worst case calculated error corresponds to harsh environmental conditions inside containment.
E-d 3.
Case 1 error-is determined by combining temperature and radiaticn induced errors by the Square Root of the E
sum of the Squares Method. Errors include 1.3F contribution based on physical configuration. Errors g
associated with Natural Circulation also include the 0.5F* maximum contribution due to use of the ASME Steam Tables.
t
' 4.
Case 2 error is determined by combining temperature and radiation induced errors algebraically. Errors
' include 1.3F* contribution based on physical configuration. Errors associated with Natural Circulation y
also include the 0.5F* maximum contribution due to use of the ASME Steam Tables.
B-5.
HPI throttling'is a consideration only above 175 psig in the RCS.
g 2
P E
G e
r
-