ML20148U125
| ML20148U125 | |
| Person / Time | |
|---|---|
| Issue date: | 11/22/1978 |
| From: | Tedesco R Office of Nuclear Reactor Regulation |
| To: | Stroupe E GENERAL ELECTRIC CO. |
| References | |
| NUDOCS 7812060281 | |
| Download: ML20148U125 (10) | |
Text
-
Central Files NRR Reading DSS NOV 2 AB Reading 2 Jg7g i
!!r.' E. P. Stroupe,11anager l
BWR Licensing j
Safety 14 Licensing Operation i
General Electric Company 175 Curtner Avenue San Jose, Califon11a 95125
Dear fir. Stroupe:
The staff is currently reviewing the answers submitted by General Clectric in response to o'ur questions dated June 2,1978 on the ODYli computer program and General Electric methodology for the analysis of transients.
In order to continue our review, additional infomation pertaining to Submittal 3, MFil-324-78 dated August 4,1978 and Submittal 6, MFil-351-78 dated September 1,1978 is required. The attached enclosure presents 0:tr questions requesting this information. These questions in draft form have been sent to you informally on October 31, 1978 and flovember 8,1978.
It is requested that this information be provided by December 132 1978.
If you need any clarification on this matter, please contact Dr. F. Odar (301-492-7911) of my staff.
i Original signed by p L. Tedesco 4
Robert L. Tedesco, Assistant Director for Reactor Safety Division of Systems Safety Office of fluclear Reactor Regulation
Enclosure:
As stated bec:
D. Eisenhut G. Holahan F
P. Check P. Norian K. Kniel J. Voglewede l
T. Novak S. Rubin Z. Rosztoczy L. Phillips R. Satterfield G. Mazetis L. Beltracchi (g-Odah Fv M, Mendonca l
~
~
D. Fieno S: $Ul*
7F/2odo-2.f/
F. Coffman.
DSS:AB A D
AB Dg DS
{
. r.c.,
F0dar:L PNorian goszyczy RTedesco l
f
,11/13/78 11/)#/78 11/7A/78 11/p/78 m
e-
6 ENCLOSURE ADDITIONAL STAFF QUESTIONS ON GF SUBMITTALS MFN-324-78 DATED AUGUST 4,1978 AND MFN-351-76 DATED SEPTEMBER 1,1978
.t i
1.
There appears to be considerable confusion relative to our requirements
'4 for conservatism in the calculation of ACPR for anticipated operational i
occurrences. The cover letter to the sixth GE submittal on ODYN (September 1,1978 letter from E.D. Fuller to D. F. Ross). states, "The staff stated
^
that a statistical evaluation which showed a 95% probability with 95%
confidence (95/95) that no more than 0.1% of the fuel will experience boiling transition was an acceptable measure of the licensing basis conservati sm. " This statement does not accurately reflect our requirements. Question 3 (Enclosure 3) requests that ODYN uncertainties be considered in the determination of the expected function of fuel rods i
.i which would experience boiling transition. Question 6 (Enclosure 3) states that a 5% probability of the actual ACPR exceeding the licensing basis aCPR is an acceptable means of demonstrating the needed margin.
The 5% probability of exceeding the licensing basis ACPR does not imply a 5% probability of exceeding the criterion of 0.1% of the fuel rods experiencing boiling transition.
If you believe that the criterion of a 5% probability of the actual aCPR exceeding the licensing basis t
l ACPR is unnecessarily restrictive and therefore unacceptable to you, then inclusion of the ODYN code uncertainties in the GETAB statistical evaluation is the most direct and appropriate method of resolving this issue.
In the absence of further information the 5% criterion will be used.
2.
Section 3.3 of the sixth submittal on the ODYN code references NED0-20340 as-the basis for the 2% standard deviation on core average power. However, il L
me
^.Aa%-ao p
i-
[,
j 1
NED0-20340 primarily addresses bundle power uncertainties not, core average power uncertainties. NED0-20340 indicates that 99% of the core themal I
power uncertainty arises from the uncertainty on feedwater flow. ASME Performance Test Codes, Test Code for Nuclear Steam Supply Systems (PTC 32.1-1969) p"esents a recommended procedure for detennining the s
thennal output fram a BWR.
Provide the uncertainties on each of the input variables in the recommended procedure and show that, with the
{
exception of feedwater flow these uncertainties have a negligible l
contribution to the overall uncertainty of 2% (at lo).
3.
The NRC requirement that the technical specification be derived from l
the analysis and evaluation included in the safety analysis report and its amendment is clearly stated in 10 CFR 50.36. The use of measured data on control rod drive scram inser'..d. mes, which are significantly less conservative than the technical specification limits, is only acceptable if the technical specificatiors are modified to assure that the assumed performance is being achieved.
Section 3.4 of " Analysis of BWR/3 through 6, Control Rod Drive Scram Insertion Times", NEDE-23884, states "...if the overall scram time mean j
for a specific plant exceeds the currently assessed generic plant scram lf time mean by "K" standard deviations (a), there would be reason to either bring the specific plants scram performance up to the generic level or retain the current scram time technical specification limits with associated end of c/cle derates."
This statement could be used as the basis for an acceptable technical specification on scram times. The technical specification must assure that the specific plant scram times do not i
-l l'
W---e ra som4W>,
exceed the generic plant scram times, assessed with a confidence f
coefficient of 0.95 (i.e., K=1.96//"n").
1 If the specific plar t scram times exceed the generic plant scram times, then the technical specification must provide for an adjustment of the required MCPR to be consistent with the measured scram times.
To accomplish the above, provide a proposed technical specification addressing our requirements that the specific plant scram times be consistent with the assumed scram times in the transient analysis
,i (including the assumed scram times in the ODYN uncertainty assessment).
In the absence of such a statistical specification the present or proposed technical specification value of scram times mest be used in the ODYN uncertainty assessment.
4.
Provide all of the coefficients in the equations for ACPR/ICPR l
(page 4-1 of the sixth submittal) for both end-of-cycle and mid-cycle conditions.
5.
Confirm that the third sentence on page 3-3 of the sixth submittal is incorrect and should read, "...(T-A)n is generally negative for the higher nodes and positive for the bottom nodes...".
l 6.
Tne number of data points for Exposure Index is very limited (11 end-of-cycle and 15 mid-cycle). This is particularly disappointing since there has been considerable operating experience with approximately 30 BWRs. Although we can make a final detemination of the acceptability of the ODYN code based on the data presented, additional data would be extremely helpful and could result in a more favorable conclusion relative to the use of Exposure'Index in the statistical analysis.
a
~
. - ~... _ - _
- ~ -
i I'
l 4
.t i
7.
As in the case of the measured scram times (see question 3 above), we require that a technical specification be developed to assure that
.q specific plant behavior is consistent with the assumed generic,
]
probability density function for Exposure Index.
In the absence of such a technical specification, the Haling distribution must be assumed in l
the assessment of the overall margin.
8.
REDY documentation shows shutoff and opening characteristics of i
turbine stop and bypass valves. Similar documentation for ODYN is not provided.
a.
Complete the documentation; b.
Provide more information than provided in the REDY documentation.
Present fomulations for shutoff and opening characteristics of stop valves, control valves and bypass valves. We understand that this fomulation has an independent variable showing the position of the valve. Present a representative function for a typical l
plant. We understand that ACPR predictions may be sensitive to these characteristics. Provide sensitivity studies using bounding shutoff and opening characteristics for ACPR predictions for turbine trip without bypass transient in a plant similar to Peach Bottom at End-of-Cycle 2.
. -1 e
9.
In the letter from GE, MFN-389-78, dated 10/17/78, GE provided some data for staff audit calculations. Please state whether the void fraction refers to the actual thermal-hydraulic void fraction or neutron effective
- lj void fraction.
i I
l
- 10. Provide a complete listing of input parameters for the ODYN code.
Provide a complete listing of input and output values and associated uncertainties for the GE calculations perfonned in response to staff question 1 (g), Enclosure 3 (Licensing basis calculation for audit I
calculations).
Identify the margin of conservatism in input parameters that GE believes to be conservative.
11.
Provide uncentainty studies and their effect on ACPR calculations for the different heat transfer correlations used in the ODYN code. These i
correlations are Dittus-Boelter, Jens-Lottes and Chen correlations.
Provide the experimental data and the range of the applicability of these correlations. Provide justification that ACPR uncertainties associated with uncertainties in heat transfer coefficients have negligible con-tribution to the total uncertainty in aCPR.
In these studies consider the uncertainties due to application of these correlations to rod bundles 4
at the operating range of BWRs.
I' 12.
In answering Question 28 Enclosure 1, GE did not submit infonnation showing that uncertainties in ACPR calculations due to 1) drift flux parameters,2) subcooled boiling nadel, and 3) pressure drop correlations calculated using the tube data are applicable to rod bundles in reactor operating conditions.
Provide a complete list of ranges of all experimental data (including in tube geometries) used to develop the models in the' analysis. Provide justification that ACPR uncertainties i
associated with uncerta' nties in extrapolating the above models into i
rod bundle geometry have negligible contribution to the total uncertainties in ACPR.
13.
GE's answer to NRC question No. 9 Enclosure 2 does not provide requested infonnation to the staff.
a.
Show how GE's void' fraction model predicts steady-state data; show j
the uncertainties associated with the predictions; present the data range; determine the uncertainties associated with the prediction outside of the data range and determine the uncertainties introduced due to rod bundle geometry if the data are obtained from a different I
geometry.
b.
Show how the void fraction model predicts the void fraction under transient conditions such as:
- 1) a sudden increase in pressure and ensuing void collapse
- 11) a decrease in flow and ensuing void expansion similar to conditions encountered in recirculation pump trip and, iii) sudden increase and decrease in power similar to conditions encountered in transients such as turbine trip without bypass.
c.
Determine uncertainties associated with the predictions of the transient data above, present the data range, determine the uncertainties associated with the predictions outside of the transient data range if the plant transients fall outside of the data range and determine the uncertainties introduced due to rod bundle geometry if the transient data are obtained. in a different geometry.
d.
Provide justification that ACPR uncertainties associated with uncertainties in vapor fraction have negligible contribution to the l,
total uncertainty in ACPR.
. 14.
General Electric should submit licensing topical reports documenting the ODYN computer program, its verification, application and the staff questions and GE answers in a timely manner.
15.
In comparing the GE predictions with the measured core pressures in Peach Bottom turbine trip tests the.. staff notes that there are some discrepancies.
In the beginning, (0.4 - 0.5 sec.), the GE pressure rise rates are smaller than those in the tests; later, (0.5 - 0.6 sec.),
the GE pressure rise rates are comparable or higher; and about (0.6 - 0.8 sec.), the calculated pressure rates and as well as the absolute values are much higher than those measured. The staff calculations indicated that the predictions of neutron power, under certain circumstances, may strongly be affected by the uncertainty in calculations of pressure rise rate and absolute pressure.
a.
Provide a new s'et of neutron power calculations using the measured pressure rates and values in all three Peach Bottom tests.
b.
Show that the ACPR between the two GE calculations (using calculated and measured pressures) is within the ACPR uncertainty claimed for the neutronics model.
~
16.
Our review of the valve flow control model and of the Electronic Hydraulic l
Control (EHC) is incomplete because of lack of information.
a.
The information requested by question 42c in "One-Dimensional Core l
Transient Model", Revision 1, July 21,1978, was not provided.
j-Provide the requested information for staff review.
!L
' b.
In the response to questioni45 in the same reference, it was stated that Figure 4-16 represents a model of a typical BWR pressure regulation system for main steam turbine equipped with Mechanical Hydraulic Control (MHC).
It is our under-standing that current designs incorporate Electronic Hydraulic Control (EAC). Provide and describe the EHC model for staff
)
review.
i 17.
The staff is also concerned with the methodology for evaluating plant specific properties for the control models. Specifically, these are A
the gains, time constants, and response functions that are used in the control models. General Electric has stated that the response functions, gains, and time constants for control systems are determined based on l
detailed computer simulations of the BWR to achieve stable system control and design objectives. Also, the numerical values are fine-tuned based on information acquired during on-site start-up tests.
Stability of the control system is a safety concern. A marginally stable or an unstable control system may cause plant conditions more severe than those for which plant safety systems are designed.
In order for the staff to complete its review of the control models, we will
.j require the following additional information:
a.
Provide a description of the methodology used to establish control system gains, time constants and functions.
In this description, define and discuss the acceptance criteria used to establish stable
~
control systems. This information is to be provided for each control system modeled in the ODYN code.
i i
e
~
9 i
b'.
For each control system-in the code, define and describe the para-meters which have a strong effect on stability.
Discuss how stability margins are maintained as acceptable for the operating or expected ranges of these parameters during the lifetime of the' plant.
i f
i i
' i t
a e
l
.i 4
y.
7.y r..
m.
w--.
4..
,__.m
..