ML19340D562
| ML19340D562 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 12/23/1980 |
| From: | Johnson W Maine Yankee |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| WMY-80-165, NUDOCS 8012310278 | |
| Download: ML19340D562 (4) | |
Text
-~
^J
\\
a IRAIRE *)HARHEE A10iit;G F02:CGil;23
- 3 e
o!
B.3.2.1 KMY 80-165 December 23, 1980 United States Nuclear Regulatory Commission Washington, D. C.
20555 Attention: Office of Nuclear Reactor Regulation
References:
1.
W. J. Szymczak, Maine Yankee Plant Accident Analysis Model Using FLASH-4, YAEC-1104, dated November 1976.
2.
NRC letter to Yankee Atomic, dated May 27, 2977
Dear Sir:
Maine Yankee is rated at a core thermal power of 2630 MWt.
Due to limitations on the low pressure turbine, Maine Yankee is currently restricted to a core thermal power of approximately 2560 MWt.
To take advantage of the thermal power rating available, Maine Yankee intends to instal'
-ten-drive main feedwater pump, to be active at power levels gre:
Utilization of the steam driven pump will reduce the p. ant Acad by the amount necessary to power the motor driven feedwater pur.p:: and will result "in a net increase in electric power delivered to the New England power grid.
The installation schedule of the steam driven main feedwater pump is such that it will be functionable for Cycle 6 operation.
This letter informs you of the impact of the steam driven feedwater pump on our plans for the Cycle 6 reload analysis requirements.
Cycle 6 operation is scheduled to commence on or about July 1, 1981.
This schedule necessitates an April 1,1981 submittal date (90 days prior to startup) for the relaod, analysis supporting Cycle 6 operation.
The installation of the steam driven feedwater pump will have an impact on the steam line break transient response.
In order to evaluate the impact on Cycle 6 opeation, it will be necessary to characterize the dynamic response of the feedwater system with the installed pump.
The analytical model that is currently used to analyze steam line breaks for Maine Yankee utilizes the FLASH-4 computer code.
This m del is documented in Reference 1 and was e
/
d K.
El;".
p ;-;-
w
~,.
e -
m
.310 777
December 23, 1980 U. S. Nuclear Regulatory Commission Page 2 reviewed and accepted by the NRC for application to Maine Yankee in Reference this code does not have the capability to adequately 2.
In its present form, represent the feedwater system dynamic response with a steam driven pump, exhibiting feedback on the system from the pump characteristics.
'reion that After examining several alternatives we have reached th team the best solution that exists to aciquately evaluate the icp
- e to driven pump on the steam line break transient is to apply ths Jerence our problem.
Alternatives evaluated are listed below in order c:
and feasibility:
application of RETRAN 1) use of FLASH 4 with input for feedwater flow developed from a mini 2)
RETRAN model 3) application of RELAPS 4) application of RELAP4 hire an outside consultant 5) modify FLASH-4 to accept additional noding anc control system logic.
6)
RETRAN is a computer code developed under the direction and financial support of EPRI for the purpose of simulating LWR response to a wide range of RETRAN was developed using the operational transient and accident scenarios.
RELAP-4 based methodology and contains features and improvements enabling realistic respresentation of system transients (control block logic for frodb-rk).
To modeling of control systems and components influenced by ev-te-date more than $6.5 million has been spent on RETRAN devel
.ut2cn to validation.
Documentation describing RETRAN supporting the The control block logic LWRs was submitted to the NRC for generic review.
contained in RETRAN is the primary feature of the code that enables evaluation of the steam driven feedwater pump by coupling pump characteristics to the All other alternatives examined, including dynamic response of the system.
hiring outside consultants, do not avoid the issue of the need to model th(
steam driven feedwater pump and obtain the feedwater flow dynamics as The feedwater flow influenced by the steam driven pump characteristics.
dynamics during the steam line break is extremely important in that the amount i
of feedwater delivered to the ruptured steam generator influences priqary to secondary heat removal and dictates the extent and rate of primary system cooldown.
We feel that RETRAN is the only feasible option for evaluating the impact of the steam driven pump on the steam line break transient without severely impacting the Cycle 6 analysis schedule and needlesly undermining Maine Yankee's commitment to undertstand for itself and defend before the commission RETRAN is the only the safety impact of plant system design changes.
available code, other than RELAP 5, that contains control block logic l
a
December 23, 1980 U. S. Nuclear Regulatory Commission Page 3 ne;essary to analyze the feedback of pump characteristics on feedwater flow Yankee has expended extensive manpower and resources in obtaining a strong experience base in the application of RETRAN to both PWRs and BWRs.
response.
This experience base has been built-up in anticipation of future licensing applications for RETRAN as well as more realistic "off-desir " nnalysis for evaluation input to operator training and emergency procedure revision:
ng term of voiding in upper head during natural circulation cooldowr.
up of planning for use of RETRAN is evidenced by our participation wie utilities seeking generic review of RETRAN (Reference 3).
A second option to the direct use of RETRAN utilizes our currently reviewed and accepted method for analyzing steam line break transients.
However, this approach requires off line a separate mini-RETRAN analysis to Because the feedwater determine temporal feedwater flow data for FLASH-4.
flow response would be determined in a separate calculation a number of iterations between the FLASH-4 analysis and the mini-RETRAN model would be A direct calculation using the control system logic in RETRAN to required.
model the steam driven pump coupled to the plant dynamics makes for a more accurate representation of the system behavior.
RELAPS The RELAP5 option is one approach we looked at quite seriously.
has been recently modified to include trip logic, point kinetics, and control Yankee is system logic somewhat similar to what is available in RETRAN.
currently involved in applying RELAPS to small break LOCA and BWR large break i
Our lack of experience with RELAPS, to date, on modeline transients LOCA.
other than LOCA events, the schedule for developing a modej
{
EELAPS at j
experience base developed in RETRAN makes it impractical te ag,_,RE this time.
to modeling the characteristics of the steam driven pump.
I One of the least attractive alternatives considered is hiring an outside I
i Yankee over the years has developed analytical methods in-house to be able to support plant operation and perform detailed safety evaluations consultant.
This expertise continues to grow in the application I
of plant design changes.
of codes such as RETRAN and. RELAPS and, through increased utilization of these codes, in performing realistic analysis for input to operator training.
Turning to an outside consultant for evaluation of the impact of the steam
- Moreover, driven pump would undermine the capability we've tried to build-up.
Beyond this, however, are schedular this option has no technical advantage.
the consultant's time constraints associated with arranging for a consultant, in generating a Maine Yankee specific model, and the time consuming procedures that would be necessary to transfer commercially protected information on the Furthermore, pump design to a third party that make this approach untimely.
hiring an outside consultant may not avoid the necessity to model the steam i
driven pump with a RETRAN type model.
i l
1 b
T Deccmber 23, 1980 U. S. Nuclear Regulatory Commission Page 4 The final option listed above, the modification of FLASH-4, has severe These schedular constraints and does not avoid the need for NRC review.
changes would only serve to improve the flexibility of our current acceptable method to a level equivalent to RETRAN; this is an illogical choice.
Based on the arguments presented above we have concludt.d that the this time, application of RETRAN represents our only feasible alternat;;
for evaluating the steam line break transient for Maine Yanu We believe -
eingle operation consistent with the schedule required.
application of RETRAN does not represent a significant or unfair drain on NRC We offer full cooperation in providing for that review resources for review.
To including formal presentations of our application efforts for your staff.
demonstrate the capability of the RETRAN code in our application we intend to submit, around the first of February 1981, a comparison of results between RETRAN and FLASH-4 for an analysis that has previously been reviewed and Additionally, we shall provide a comparison of RETRAN approved by your staff.
We trust prediction to plant data collected during a pump coastdown test.
that this information clearly and logically explains our decision to apply RETRAN for the Maine Yankee Cycle 6 reload analysis.
If you have any questions on this matter, or if we can be of further assistance please do not hesitate to call.
Respectively submitted, f Hi#
W. P. Johnson Vice President PAB/smh w
.m-memm. e e-e m -