ML20235H757

Submits Response to NRC 870424 Request for Addl Info Re PTS & Info Requested During 870701 Telcon Re Satisfaction of 10CFR50.61 Requirements Concerning Pts,Including Neutron Fluence on 860123

ML20235H757
Person / Time
Site:	Millstone
Issue date:	07/06/1987
From:	Mroczka E NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
A06530, A6530, TAC-59964, NUDOCS 8707150250
Download: ML20235H757 (16)

Text

, _ . -

l l

4 q

NORTHEAST UTILITIES cenerai Orrices . seiden street. seriin. Connecticut J $l75 $E[cN *I

,a os mponia ***

P.O. BOX 270 HARTFORD. CONNECTICUT 06141-0270 L t j [""g{"l"',*","", (203) 665-5000 July 6,1987 Docket No. 50-336 A06530 Re: 10CFR50.61 i

U.S. Nuclear Regulatory Commission

. Attn: Document Control Desk Washington, D.C. 20555 Gentlemen:

Millstone Unit No. 2 Pressurized Thermal Shock Response to Request for AdditionalInformation Northeast Nuclear Energy Company (NNECO) hereby submits its response to the NRC Staff's April 24,1987 re pressurized thermal shock at (PTS) quest Millstone (l)No.

Unit for2.additional information By that letter, the concerning Staff posed several questions regarding NNECO's projected values of neutron fluence over Millstone Unit No. 2's plant life. In addition, NNECO provides information requested during a July 1,1987 telephone conference between NNECO and the Staff. NNECO originally submitted information regarding satisfaction of 10CFR50.61 requirements concerning PTS, including neutron fluence, on January 23, 1986.(2)

The Staff's present questions concern the assessment performed by NNECO for Millstone Unit No. 2 to determine appropriate values of RTPTS, and specifically with respect to projected neutron fluence values, for comparison with the screening criteria set forth in 10CFR50.61(b)(2). For the most part, the Staff's request seeks further detail regarding the fluence estimates performed for Millstone Unit No. 2 in accordance with 10CFR50.61. However, the Staff also requested that NNECO address two specific matters (use of P3 and Sg approximations and ENDF/B-IV cross-section data)in its fluence calculations.

Response to Staff Requests NNECO responds to each of the Staff's requests in Attachment A. A description of the PTS and fluence evaluation performed for Millstone Unit No. 2 is provided (1) D. H. Jaffe letter to E. 3. Mroczka, regarding pressurized thermal shock at Millstone Unit 2, dated April 24,1987.

(2) 3. F. Opeka letter to C. I. Grimes, Haddam Neck Plant, Millstone Nuclear Power Station, Unit Nos. 2 and 3,10CFR50.61 Compliance, dated January 23, 1986.

Q7150250B70706 P ADOCK 05000336 fO PDR ll l

U.S. Nuclear Regulatory Commission A06530/Page 2 July 6,1987 as Attachment B. With respect to the two specific requests noted above, and without waiving any position regarding the application of 10CFR50.61 with respect to such information, NNECO summarizes below information provided in the attachments. .

First, concerning the use of the P3 and Sg approximations, NNECO has confirmed that the fluence calculations for Millstone Unit No. 2 did address anisotropic scattering cross-section with a P3 expansion. In addition, angular quadrature was approximated using an Sg approximation.

~

Regarding the . cross-section data, NNECO used the DLC-23 CASK, 22 energy group, neutron cross-section library. As noted in Attachments A and B, use of that library rather than the ENDF/B-IV library proposed by the Staff is appropriate. Nevertheless, NNECO notes that variations in predicted RTPTS from the use of one or the other library should not be significant. Further, even if one assumes a material difference in RTPTS values would result, Millstone Unit No. 2 has a wide margin between predicted RTPTS and the screening criteria of 10CFR50.61(b)(2). Thus, there is no significant safety impact resulting from the'use of the DLC-23 CASK 22 library.

Conclusion For the reasons set forth above, and in the attachments hereto, NNECO requests that the Staff approve NNECO's analysis of PTS for Millstone Unit No. 2.

Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY

? W 2<t /

E. 3. '1Virf'czl<a' //

Senior Nice President cc: W. T. Russell, Region 1 Administrator D. H. Jaffe, NRC Project Manager, Millstone Unit No. 2 T. Rebelowski, Resident inspector, Millstone Unit Nos. I and 2

-4 l

l l

. . 1

.l Docket No. 50-336 l A06530 l

l j

!a l

Attachment A RESPONSE TO REQUESTS FOR ADDITIONAL INFORMATION MILLSTONE UNIT No. 2 i

July,1987

I Attachr.ient A A0%30/Page 1 Response to Staff Requests for AdditionalInformation i I

NNECO responds below to each of the Staff's requests for additional information. For this purpose, NNECO ]

fluence calculations for Millstone Unit No. prepared 2, the results a detailed description of which of the were originally s

j presented in Reference 2. That description is provided as Attachment B. Each l Staff request is addressed separately below. NNECO's responses reference I Attachment B, as appropriate. In responding to these requests, NNECO does not I waive any position regarding the application of 10CFR50.61 to these requests.

Staff Request

1. To m rl" justify the proposed value of the fast neutron fluence to the pr~ r assel, the licensee must describe the method, codes, cross sectior mimations, input parameter values, error analysis, etc.

NNECO Re Attachment a pra es a detailed explanation of the methods, codes, cross sections, approx.ma. ns, input parameter values, error analysis, and other information relevan, to the neutron fluence calculations for Millstone Unit No.2.

Staff Request

2. NRC Staff and Staff consultant experience with fluence estimates and projections indicates that as a minimum a two-dimensional discrete ordinates bench-marked code must be used (such as DOT).

NNECO Response As described in A6tachment B (page 1), NNECO employed SAND-II and DOT-III computer codes to calculate the fast flux and fluence at the surveillance capsule assembly location and at the reactor vessel. The DOT-III code is used to provide both the initial flux estimate for SAND and to determine the azimuthal flux distribution on and in the vessel, bench-marked relative to surveillance data.

The DOT-III code is a two-dimensional discrete ordinates code.  :

Staff Request .

3. The P3 and Sg approximations and an ENDF/B-IV based cross-section set with a sufficient number of energy groups are required.

NNECO Response NNECO addressed anisotropic scattering cross-section in its Millstone Unit No.

2 flux calculations with a P3 expansion. Also, angular quadrature was approximated with an 53 approximation.

With respect to the cross-section library, as described in Attachment B (page 3),

NNECO employed the DLC-23 CASK, 22 energy group neutron cross-section library. NNECO notes, as the Staff is aware, that regardless of the

Attachment A A06530/Page 2 particular cross-section library employed, the resulting fluence calculations 1- should differ only slightly so long as appropriate bench-marking is performed I

using available surveillance ' data and applicable bias factors are properly evaluated.

In addition, even if a difference in fluence values is assumed to result from using different libraries, the effect on the comparison of calculated RTPTS values and the screening criteria in accordance with Section 50.61(b)(2) is insignificant -

First, even relatively large differences in fluence predictions do not significantly impact the RTPTS result in view of the fractional power applied to the fluence term in the equations.

Second, a significant margin exists between the l calculated RTPTS (196.880F at 32 EFPY) and the screening criteria (2700).

(Attachment B, Page 3.) Thus, there would be a minor impact on the overall safety evaluation even if higher fluence values were assumed to result from use of a different cross-section library.

Staff Request

4. The source term for the outer row of assemblies must be expressed for each pin or if the input is on an assembly average basis the result should be adjusted.

NNECO Response As described in Attachment B (page 3), NNECO calculated the power source term for the DOT calculation based on a pin by pin distribution adjusted by the assembly average power to correspond to an average distribution over time from start-up to the end of Cycle 3.

Staff Request

5. The projected source term should account for future loading schemes.

NNECO Response At this time it is anticipated that future loadings will reflect past experience, f with allowance for the power output of 2700 MWt and thermal shield removal.

Staff Request

6. If future loadings include low leakage loadings and the use of once, twice or thrice burned assemblies in peripheral locations, the effect of the increased amount of plutonium on the neutrons / fission and the neutron spectrum must be accounted for.

NNECO Response NNECO has no firm plans to change prior loading schemes, either with respect to low leakage loadings or multiple-burn assemblies. In the event different schemes are employed which could produce a significant change to projected values, .

NNECO will update its assessment in accordance with iOCFR50.61(b)(1).

l

Attachment A A06530/Page 3 Staff Request

7. If there exist dosimetry measurements from surveillance capsules, the results should be compared with those results.

NNECO Response NNECO has employed surveillance data in confirming the adequacy of existing flux calculations as well as adjusting those calculations to account for actual data. As explained in Attachment B (page 4), the flux spectrum is adjusted by an iterative technique until calculated and measured activities agree with a standard deviation of five percent. Surveillance capsule results were also used to adjust the fast neutron flux calculation to assess the impact of removal of the thermal shield in 1983 (Attachment B, pages 4-5).

NRC Request

8. Finally, the fluence should have an error estimate.

NNECO Response Error estimates are provided in Attachment B (page 4) with respect to predicted fluence values and the principal components thereof. The overall uncertainty of fluence at the vesselID is predicted as +30 percent.

i l

I 4

l 1

l 1

1 I

i

Docket No. 50-336 A06530 i

Attachment B FLUENCE CALCULATION FOR MILLSTONE UNIT NO. 2 PRESSURIZED THERMAL SHOCK ISSUE i

}

i 1

1 I

l July,1987 l

Page 1 FLUENCE CALCULATION FOR MILLSTONE UNIT NO. 2 PRESSURIZED THERMAL SHOCK ISSUE I

The existing analysis of the Millstone Unit No. 2 (MP2) (Reference

1) vessel made use of SAND-II (Reference 2) and DOT-III (Reference 3).

The SAND-II and DOT-III computer codes were used to calculate the fast flux and fluence at the surveillance capsule assembly location  :

and at the reactor vessel. The SAND-II computer code is used to calculate a neutron flux spectrum from the measured activities of the flux monitors. SAND-II requires an initial flux spectrum estimate; this is calculated using DOT-III. This analysis also employed a P , legendre polynomiac expansion of the scattering cross-sectiond. The discretization of the angular flux was represented by an S 8 quadrature. The measured activities must be adjusted before they can be put into SAND. The various steps of the procedure are described below.

The measured activities were decay corrected to reactor shutdown.

Before being used by SAND, the foil activities must be converted to saturated activity with units of disintegrations per second per target atom (dps/a). The following equation was used for the conversion:

M A 16.67

^ sat NIS where A = Saturated activity (dps/a) sn M = Measured activity at shutdown (dpm/mg)

A = Atomic weight N = Avogadro's number I = Isotopic abundance of target isotope S = Saturation factor, explained below For U 238 fission product activities, g required SAND input has dimensions of fissions per second per U atom (fps /a).

I L___ -_- .

Page'2 FLUENCE CALCULATION FOR MILLSTONE UNIT NO. 2 PRESSURIZED THERMAL SHOCK ISSUE l

l This is obtained by dividing A by the fractional fission yield of the fission product whose acti6fky was measured.  !

The saturation factor, S, converts the. measured activity to a j saturated activity. The actual reactor operating history was used to calculate the saturation factor. The reactor was assumed to operate for several periods of constant power. Then, for each isotope, S was calculated.

S = -

exp (- RTg) [1- exp (-htg)]

l l

where i

Pi = Power of ith interval ,

j Po = Full Power 3 = isotope decay constant Tg = Time between end of ith operating period to reactor shutdown t = length of ith operating period t

The combination of a cadmium shielded and unshielded U-238 foil is included in the flux monitor set. The activity of the unshielded foil is used to correct for U-235 fissions in the shielded foil. As a result of this calculation, the U-238 fission rate was determined to be 94% of the shielded uranium foil activity. No correction for photo fission of U-238 was included.

SAND requires an initial estimate of the neutron flux spectrum.

This initial estimate was calculated using DOT-III, a two-dimensional discrete ordinate code.

1 l

l l

l f

I

I Page 3 L FLUENCE CALCULATION FOR MILLSTONE UNIT NO. 2 l PRESSURIZED THERMAL SHOCK ISSUE The DLC-23 CASK, 22 group neutron cross-section library was used.

The use of this library, with appropriate benchmarking and bias factors, would produce predicted fluences which differ only slightly from fluences calculated using other libraries, including ENDF/B-IV library. This, however, would not present a significant variation based on the fact that the impact of the overall 20% increase in fluence on the calculated RT would be just 5%. This value is resultant from the equation M ch defines RT as a function of fluence. This equation, as presented in 10CFP[d.kl, is as follows:

0.270 RT PTS

=I +M+ (-10 + 470 Cu + 35- CuNilf ,

where M is the margin term.

As defined by this expression, raising fluence by 20 percent results in an increase of 5 percent in the RT values. This amounts to a 4 degree Fahrenheit increase in the RYb va is more than ten times accounted for in the error Ngrm.lueGiven whichthe allowance still available, i.e. a calculated end of life RT value of 196.88'F of versus the screening criteria of 270

• F ,PT t bis slight variation has no significant impact.

The reactor geometry is shown in Figure 1. The distribution of pin power for the DOT calculation was based on a pin-by-pin distribution adjusted by the assembly average power to correspond to an average distribution over the time from startup to the end of cycle 3.

Figure 3 shows the surveillance capsule detail used in the DOT model.

SAND uses an iterative technique to calculate the neutron flux spectrum. The activities of the set of f?ux monitors and an initial flux spectrum are the input required by SAND. Activities are cal-culated for each __'l for the flux spectrum using the following equation A =

T (Eg) Eg gQ where (Eg) is reaction cross section at energy Eg, barns.

$(Eg) is the flux at Eg, n/cm s, mev AE g is width of energy band at Eg, mev.

Page 4 FLUENCE CALCULATION FOR MILLSTONE UNIT NO. 2 PRESSURIZED THERMAL SHOCK ISSUE The flux spectrum is adjusted by an iterative technique until the calculated and measured activities agree with a standard deviation of five percent. The result of this is a 620 group neutron flux.

l In addition to DOT-III being used for the initial flux' estimate for SAND, it was also used to determine the aximuthal flux distribution on and in the vessel relative to that at the surveillance capsule.

With the DOT-III results, Lead Factors were calculated as the ratio of the peak fast flux (En > 1.0 MeV) at the surveillance capsule assembly to the peak fast flux at the vessel-clad interface and 1/4 and 3/4 vessel thickness locations. These Lead Factors were then multiplied by the fast flux generated by SAND to give us the fast flux and fluence at the vessel-clad interface and 1/4 and 3/4 thick-ness into the vessel respectively. The fluence was calculated for the end of cycle 3 (3.0 Ef f ective Full Power Years at 2700 MWt) and end of life (32 EFPY at 2700 MWt).

The SAND code will give fluxes that are accurate to within + 10% to

+ 30% if the errors in the measured activities are within~ similar limits. The uncertainties in flux spectrum monitor activity levels are at a 2-Sigma level and are determined from counting statistics.

+ 20% for uranium monitors and + for all An otheradditional error ofis metal monitors ~eatimated from volumetric and gravimetric operations and from the certified uncertainties of calibration isotopes. The additional error associated with the sulfur monitor results in 18%. Therefore, it is estimated that the uncertainty in the measured flux at the surveillance capsule location is about 120%

to -+30%. The extrapolated flux in the vessel will be slightly higher, so a reasonable value to use for the uncertainty of the fluence at the vessel ID is 130%.

A calculation was performed by Northeast Utilities in 1983 (Calcula-tion No. 83-060-241CP) to assess the impact of the thermal shield removal on the fast neutron flux to the reactor vessel. The impact ,

of that change was evaluated by Combustion Engineering for Florida I Power and Light (FP&L). This resulted in a fluence multiplication factor of 1.74 for Millstone Unit No. 2 (Reference 4).

f new = 1.74 f old i

l l

i l

m

l.

Page 5 FLUENCE CALCULATION FOR MILLSTONE UNIT NO. 2 PRESSURIZED THERMAL SHOCK-ISSUE The fluences used in the RT calculation were based upon-the Capsule W-97 results, adjustegTg y the thermal shield removal factor  ;

and revised for the appropriate effective full power date. Addi-tionally, the fluence attenuation for 1/4T and 3/4T locations was calculated via the following relation:

p , .24x Pg This "dpa equivalent" equation was presented in Reference 5 where:

F = Inner Surface Fluence, as discussed above, and F = Fluence at a 18 cation "x" inches radially into the RV wall.

This results in a fluence for the limgingg lpcation, the inter-mediate cpgrse pfate C505-1, of 4.32 NO /cm at 28 EFPY and of 5.0 x 10 "/cm at 32 EFPY. The corresponding values of RT are 192.09'F and 196.88 F, respectively. (See Table 1 of J. F fTbpeka letter to C. I. Grimes, "Haddam Neck Plant, Millstone Nuclear Power Station, Unit Nos. 2 and 3, 10CFR50.61 Compliance", dated January 23, 1986)

In response to the possibility of low leakage loadings and multiple burn loads, there are no firm plans for such core loadings. Should this- loading type be impismented, it would represent a change in core loadings and consequently provide a significant change to projected values. At that time, an assessraent of the impact will be performed as required by 10CFR50.61 (b)(1),

i

l 1

REFEEENCES:

1. Combustion Engineering, Inc. " Northeast Utilities Service Company, Millstone Nuclear Unit No. 1; Evaluation of Irradiated Capsule W-97; Reactor Vessel Materials Irradiation Surveillance Program", TR-N-MCM-008, dated April 1982.

2. SAND Users Manual, AFWL-TR67-41, September 1967.

3. DOT-III Users Manual, ORNL-TM-4280, September 1973.

4. Combustion Engineering letter N-PH- 3 6 8, L. B. Tarko to T.

Natan, " Millstone Point 2 Fast Neutron Fluence After Thermal Shield Removal", dated Nov^mber 14, 1983.

l

5. P. N. Randall, " Pressurized Thermal Shock", presented to the l

ACRS; 5/11/82; pg. 9.

l l

4 l

. - _ _ _ - _ _______________-_a

7  :-

\

\ g ,( L U

2 A

A S

P T t EH H E T M 6

S

\

N \A/ / \ - y=~

\

m st, s

N &

9p

, Lp j

A s #2 5* _"

Se *c x\ o

, /

N -

g

\ x\ v \\ \ [ -

)

x \

N p

x \

x v \

os S 3\ \ \

1 x\

x

\

o e '

r u

Fig i

F 0 F "Ss SS x

e \

2A 05I A 4 T 4I c c 6S D

Oi 5SP/ gc 5P/cg 0 = 0 4 N

I

=55 E 29 E 26 55 D R25 R23 E

S U=7 TU=7 E=

Ui T E=

A A 4 RY 4 Y R RY UT E SI R

E UT SI R P S S P S S T

E MEN MEN ERE ERE M

O TPD TPD E RRR RRR G EEE EEE TTT TTT 2 AAA AAA T

I N

WWWWWW SSS U RRR EEE SSS AAA E MMMPPP N OOOYYY O CCC BBB T NNN EEE WWWRRR

_ S L

_ OOOOOO ILi DDD CCC M

_ 0 0 0 0 0 0 0 0 8 6 4 2 8 6 4 2

_ 1 1 1 1 1 5[?Hb

E L

U S

P O

A' C

e ot 0 T

O D

N I

P.,

D $' $ U!U D

E g A A $%

S

• gT)#

i UPO U e S 94- 5 3 S N E 5 9

9 7 R O 6 8 E I

T T 1 E U $m.E M 3 I p-O l

2 2- T R 9 1 8 8 1 8 N T 2

~Y 4 6 7 1 9 E S 7 0 0 1 9

I 9 C D g-9 6 1 1 R

E EO e Um.S

/g 9 1 7 W 'SpO 5 1 2 2 1

2 8 2 O n- 7 9 8 4 5 1

0 6

e r D P

'e 0 4

7 0 0 9 1 1

1 7

uN g

1 1 U 6.U F A '

i S 3 N 5 5 2

. 9 4 0 O 3 0 0 8 I

S 7 0 3 4 N

E M

1 N1 2 7

0 1

7 0

1 I

0 D 0 2

T I #

9 N U 7

E N N O I T S S N L

OS

L S R I

I e

M N ET E E

MM I

DIT LN LE AC s

,l l Il l lfIfll

Figure ' 3

~

MILLSTONE UNIT 2 SURVEILLANCE CAPSULE LOCATION CAPSULE W.97 i

ALL DIMENSIONS IN CENTIMETERS CAPSULE AND CLAD ARE STAINLESS STEEL VESSE L CLAD i CAPSULE WATER 3.81 1

70 1 2 3 4 5 6 RADIAL DISTANCE (CENTIMETERS) i 1 214.65 2 2 215.69 3 217.57 4 218.46 5 220.0475 6 221,0 1

• l l

l