ML20112J501
| ML20112J501 | |
| Person / Time | |
|---|---|
| Site: | Prairie Island  |
| Issue date: | 03/26/1985 |
| From: | Musolf D NORTHERN STATES POWER CO. |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8504050179 | |
| Download: ML20112J501 (6) | |
Text
.
+.
- Northem States Power Company 414 Ncollet Mall Minneapoks. Minnesota 55401 Telephone (612) 330-5500 March 26,1985 Director Office of Nuclear Reactor Regulation U S Nuclear Regulatory Commission Washington, DC 20555 PRAIRIE ISLAND NUCLEAR GENERATING PLANT Docket Nos. 50-282 License Nos. DPR-42 50-306 DPR-60 K(z) Curve
References:
1.
Letter dated June 23, 1983 D M Musolf to the Director of NRR, " License Amendment Request, LOCA Analysis."
(Containing XN-NF-83-38) 2.
Letter dated October 3,1983 D DiIanni to D M Musolf, NRC Safety Evaluation of XN-NF-83-37 3.
Letter dated January 21, 1985 D M Musolf to Director of NRR, " Supplemental Information Related to License Amendment Request dated July 11, 1984."
This submittal is being made in response to a request by the NRC Staff to inform them of our plans for resolving questions related to Exxon Nuclear verification of the Prairie Island K(z) curve. Exxon Nuclear Company has not been able to support the current K(z) curve in Westinghouse PWR reactors. Northern States Power Company will have the K(z) curve reunalyzed for Exxon fuel to resolve this issue.
Our current LOCA analysis was performed by Exxon Nuclear using a chopped cosine axial power shape centered at the 6 foot elevation (References 1, 2 and 3). Exxon Nuclear has indicated that their current methodology is overly conservative in the upper regions of the core (Attachment 1).
Exxon is recorrelating the FLECHT reflood test data to remove some of this conservatism. However, this recorrelation is not expected to be completed until mid-summer.
h
$B84RoldlSISShia fs\\
P 4
o.
Northem States Power Company Dir NRR March 26,1985 Page 2 Another option we are considering, is to have Westinghouse Electric Corporation perform a LOCA analysis on Exxon fuel. Beginning next year, Westinghouse Optimized Fuel will be used for Prairie Island reloads. At the present time we do not have sufficient information to make a decision as to who will perform the reanalysis of Exxon Fuel.
Westinghouse and Exxon Nuclear fuel are similar enough to suggest that they will not behave significantly differently during a LOCA (See Attachment 2).
Westinghouse performs a limited power shape sensitivity study following each major change to their methodology. This same methodology has analyzed the chopped cosine axial power shape for Prairie Island with Westinghouse fuel with similar results as the Exxon methodology. We believe, the present K(z) curve can be rejustified for Exxon fuel with Westinghouse methodology (or possibly the Exxon methodology following recorrelation of the FLECHT data). In addition, recent research has shown that many of the models and assumptions required by 10 CFR 50 Appendix K are unrealistically conservative, which causes Appendix K Evaluation Models to predict artificially high peak cladding temperatures for the large break LOCA.
l Thus, there is a great deal of margin in the peak clad temperatures calculated by current methodology.
Therefore, we believe there is no safety concern with continued use of the K(z) curve currently in the Prairie Island Technical Specifications while this issue is being addressed. The problem is rooted in the basic conservatisms of the Exxon methodology.
However, in order to reconfirm compliance with 10 CFR 50.46, the curve will be rejustified by having one of the vendors mentioned above reanalyze the K(c) curve. We will provide our schedule to the NRC Staff prior to May 1,1985.
David Musolf Manager - Nuclear Support Services DMM/TMP/tp c: Regional Administrator-III, NRC NRR Project Manager, NRC Resident Inspector, NRC G Charnoff Attachments:
1.
Letter dated March 21, 1985 G F Owsley (Exxon Nuclear) to Tom Parker (NSP), "K(z) Verification."
2.
Comparison of Westinghouse and Exxon Fuel i
A int 1 1 of I
- s. ' '
.)If7 fe M
A A
pp"q d,.,
'A ERON NUCLEAR COMPANY,INC.
g g
j 1
pph,7-gas AOUE NE.PO 90X 90m88.mWE.m90000 c
March 21, 1985 "'<,..
GF0:85:009
/,'
n
\\
,u.
Mr. Tom Parker l
Northern States Power 414 Nicollet Mall 1
Minneapolis, MN 55401 SLU ECT: K(z) Verification
Dear Mr. Parker:
As we have discussed previously, the NRC has asked Exxon Nuclear to verify the K(z) curves for some of the Westinghouse reactors to which we supply fuel. Because of excessive conservatism in the current Exxon Nuclear evaluation models when used to analyzse power distributions with peaking in the top of the core, Exxon Nuclear has not been able to support the previously existing K(z) curves for these plants. Currently, we are assessing the K(z) curve for a two (2) loop plant similar to Prairie Island, and based on analysis to date, we expect that the existing curve will not be supportable with the existing Exxon Nuclear evaluation models.
Likewise, we believe that the existing Prairie Island K(z) curve would not be supportable if analyzed with the currently approved Exxon Nuclear evaluation models. We will comunicate the results of this analysis to you when they become available, which should be in early April 1985.
~
~
ENC has initiated an effort to recorrelate the FLECHT reflood test data.
This effort will take into account the conservatism that exists in the present ENC methodology at elevations above the midplane. This effort is expected to be completed by mid summer at which time reanalysis with the revised reflood correlation can be initiated.
If there are questions, or if further information is needed, please contact Larry Federico at (206) A53-4381.
i Si erely f.
Owsley, Ma, ger Reload Licensing iaison l
RAC:naa cc: Mr. L. C. O'Malley (ENC) u.mumonxxo ccmamo l
1 of 3 Comparison of the ECCS Performance of Westinghouse and Exxon Fuel The 14x14 fuel designs that are in the Prairie Island units I and 2 are very similar for both Exxon and Westinghouse. The fuel parameters that are relevant to an ECCS analysis are compared in table 1.
The hydraulic performance of the fuels is designed to be similar. It would thus be expected that the hydraulic characteristics during the blowdown phase of the LOCA transient would be similar for any fuel type. This similarity in blowdown hydraulics would lead to a similar axial temperature distribution for each type of fuel. Also, since the average power rod of any of the three fuel types has the same linear power, the only differences in the magnitudes of the cladding temperatures for the hot rod at the end of blowdown would be due to total peaking factor. In fact, since the smaller diameter rods have a higher surface to volume ratio than the larger diameter rods it would be expected that the cladding temperatures for the smaller diameter rods would actually be lower at the end of blowdown.
The primary difference in ECCS performance between the three types of fuel in Table 1 would occur during the reflood portion of the LOCA transient.
4 The smaller diameter fuel rods have a correspondingly larger core flow area which reduces the reflood rate and results in higher calculated peak cladding temperatures. In addition, since the amaller diameter rods generate the same power as the larger diameter rods, the volumetric heat generation rate of a smaller diameter fuel rod is higher that for a larger diameter fuel rod. This results in a higher heatup rate during the lower plenum refill period when there is no heat transfer from the fuel.
At the time the bottom of the core is recovered, the temperature of the fuel would be inversely proportional to the fuel diameter. The difference would increase further as the core flooded due to the larger core area of the smaller diameter fuel rods. Both Exxon and Westinghouse concur with this as demonstrated by the fact that for a transition core, standard to TOPROD or standard to optimized, only the smaller diameter fuel is analyzed.
1 A comparison of actual ECCS analyses of the Prairie Island Units is presented in Table 2 for the Exxon EXEM/PWR evaluation model and Westinghouse 1981 evaluation model. The parameters that were used in the two analyses are similar. The Westinghouse analysis was performed for the 14x14 optimized fuel design with an FdeltaH of 1.65 which results in a slightly lower Fq than if an FdeltaH of 1.55 was utilized. The Exxon analysis was performed for a 14x14 TOPROD fuel design. The results for i
these two analyses is consistent with the qualitative description given above in that the smaller diameter fuel rods have the higher PCT.
I l
l l
g.
Attachmsnt 2 2 of 3 Both of these ECCS analyses were performed using a cosine power shape.
The K(z) methodology for peaking factor control that is utilized in the Prairie Island units 1 and 2 Technical Specifications was developed from Westinghouse ECCS model sensitivities to axial power shape. Westinghouse performs a limited power shape sensitivity study for each major change in ECCS model. For the 1981 model, the skewed to the top power shape resulted in a lower calculated PCT than the cosine power shape. Based on these results Westinghouse concluded that the cosine was still the limiting power shape. From that study and the comparison of the two Prairie Island ECCS analyses in the previous paragraph it would be expected that if the Exxon TOPROD fuel design were to be analyzed using Westinghouse ECCS methodology for Prairie Island units 1 and 2 the result would be consistent with previous Westinghouse studies in that the cosine power shape would be limiting.
i i
e e
- g.
3 of 3 Table 1 Comparison 'of Exxon and Westinghouse 14x14 Fuel Designs Parameter ENC Standard ENC TOPROD W Optimized Clad O D (in) 0.426 0.417 0.400 Clad I D (in) 0.364 0.358 0.3514 Clad Thick. (in) 0.031 0.0295 0.0243 Pellet O D (in) 0.3565 0.3505 0.3444 Diametral Gap (in) 0.0075 0.0075 0.0070 Pellet Density (%)
94.0 94.0 94.5 Active Fuel Length (in) 144.0 144.0 144.0 Rod Pitch (in) 0.556 0.556 0.566 Table 2 Comparison of Exxon and Westinghouse ECCS Results For Prairie Island Units 1 and 2 Parameter Exxon Westinghouse Fuel TOPROD Optimized Break 0.4 DECLG 0.4 DECLG Fq 2.32 2.28 Fdeltah 1,55 1.65 PCT (degrees F) 2146*
2158*
PCT Location (ft) 8.13 7.50 Time of PCT (sec) 301 198 Hot Rod Burst Time (sec) 45.40 71.0 Hot Rod Burst Location (ft) 6.13 7.00 Steam Gen. Tube Plug (*)
5.0 5.0
- includes UPI PCT I
d
--.