ML18066A687

Forwards Response to NRC 990908 RAI Re Inservice Insp Program Relief Request 14.Ltr Contains No New Commitments & No Revs to Existing Commitments

ML18066A687
Person / Time
Site:	Palisades
Issue date:	10/19/1999
From:	HASKELL N L CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
TAC-MA5711, NUDOCS 9910290001
Download: ML18066A687 (14)

Text

-,. A CMS Energy Company October 19, 1999 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Palisades Nuclear Plant 27780 Blue Star Memorial Highway Covert, Ml 49043 DOCKET 50-255 -LICENSE DPR-20 -PALISADES PLANT Tel: 616 764 2276 Fax: 616 764 3265 Nathan L. Haslcsll Director, Licensing RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING INSERVICE INSPECTION PROGRAM RELIEF REQUEST NO. 14 (TAC NO. MA5711) On May 17, 1999, Consumers Energy submitted a request for relief from the requirements of the American Society of Mechanical Engineers Boiler and Pressure Vessel Code. This relief request was designated as lnservice Inspection Relief Request No. RR-14. On August 25, 1999, a conference call was held between the NRC, Structural Integrity Associates personnel, and Palisades Plant personnel to discuss a draft NRC Request for Additional Information (RAI) regarding RR-14. The final RAI was issued by the NRC on September 8, 1999. The attachment to this letter provides the Consumers Energy response to the NRC RAI.

SUMMARY

OF COMMITMENTS This letter contains no new commitments and no revisions to existing commitments. U2lo-vt Nathan L. Haskell Director, Licensing CC Administrator, Region Ill, USNRC Project Manager, NRR, USNRC NRC Resident Inspector-:--

Palisades Attachment ATTACHMENT CONSUMERS ENERGY COMPANY PALISADES PLANT DOCKET 50-255 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING INSERVICE INSPECTION PROGRAM RELIEF REQUEST N0.14 (TAC NO. MA5711) 12 Pages

,. Response to Request for Additional Information Palisades Request for Relief from the Requirements of ASME Section XI for Spent Fuel Pool Heat Exchanger Inlet Nozzle to Shell Weld NRC Question 1: Provide equations and input for all parameters appearing in the equations that give a diffusion coefficient of 1.1162 fr /hr of water vapor in air and a leakage rate of 13. 88 lb/year. Estimate the error in the final leakage rate caused by the assumption that the mass transfer is analogous to the heat transfer.

Cite examples from literature showing that the final leakage rate (strongly influenced by free convection) can be conservatively estimated by applying the Nusselt number to the leakage rate by diffusion.

Alternatively, justify the calculated final result by empirical deduction based on observations.

Consumers Energy Response 1: The diffusion coefficient, D, for water in air at 32°F and 1 atm is given as 0.85 ft 2/hr in Ref. 1, Table E-6. It may be adjusted to the conservatively higher assumed temperature 130°F based on Eq. 14.9 of Ref. 1. 3 D _ D x (130+460)2 130oF -32oF 32 + 460 = 0.85 x (1.1992)%

= 1.1162 ft 2 /hr With respect to conservatism in the leakage rate, report SIR-99-032 approached the evaluation of conservatism in two ways. 1. Natural heat transfer effects were used to estimate the ratio between convective effects and diffusion.

A conservatively high T of 50°F was used, whereas there is only a small difference in temperature expected between the heat exchanger and the room. It was assumed that there was no humidity in the room. No credit was taken for the fact that the local vent hole would reduce the effects of natural convection.

2. The predicted amount of liquid leaking in one minute was calculated.

This amount of liquid was estimated to create a hemispherical drop size with a diameter of 0.526 inches. We know from practical experience that this amount of liquid would collect in drops,

• would not evaporate, and would clearly be visible during visual inspection.

We believe this provides ample justification of the conservatism in the leakage rate calculation.

1

• Reference

1. Rohsenow, W. M., and Choi, H., "Heat Mass and Momentum Transfer," Prentice Hall, 1961. 2

• NRC Question 2: Provide information regarding the validation of the computer program pc-LEAK. The staff's primary interest is the benchmark of the crack opening area and the leakage rate using pc-LEAK against results from using other computer codes. Consumers Energy Response 2: The computer program pc-LEAK was written to determine the leakage rate for superheated or saturated steam and saturated or sub-cooled water through axial and circumferential cracks for a variety of stress conditions. -The crack opening area solutions are based on linear elastic fracture mechanics as follows:

• Circumferential Through-wall Crack with Pressure [1]

• Longitudinal Crack with Pressure [1]

• Circumferential Crack with Tension & Bending [1], with modified F functions from Appendices A & B of Ref. 2] To validate pc-LEAK; leakage from test data documented in Reference

[3] was evaluated.

In EPRis validation of the program LEAK-00 [3], the predecessor of PICEP [4], rates were compared to predicted values. As shown in Figure 2-1 the LEAK-00 solutions compared well to. the actual data. The same method was used to validate pc-LEAK. In Figure 2-2 the measured values are compared to the predicted rates of pc-LEAK. As can be seen from Figures 2-1 and 2-2 the predicted flow rate solutions compare well to both the measured and predicted values used by EPRI. In fact, pc-LEAK appears to produce a slightly better prediction than LEAK-00. The trend of prediction for each test is very similar in each case. Validation for the "Crack Opening Area" solutions was accomplished by showing that the equations were producing the correct answers in the verification process based on the methodology of References 1 and 2. Calculations have also been performed to compare the results from PICEP [4] to those of pc-LEAK: The geometry chosen was the 25.375-in diameter, 0.1875thickness heat exchanger shell, subjected to 125 psig (with no external bending moments).

The same material properties were used for both programs (modulus of elasticity

= 28,300 ksi). In PICEP, the Osgood a parameter was set to zero to assure elastic response for a good comparison.

Cases were run for both circumferential and axial through-wall cracks. (The pc-LEAK model was for a cylinder with applied pressure loading [1], the same as utilized for the heat exchanger analysis.)

The results of the comparison are shown in Figures 2-3 to 2-6. There is excellent agreement between the two models. The most significant discrepancy is for small circumferential cracks, where the leakage predicted by pc-LEAK is less than predicted by PICEP. The reason for the discrepancy is that pc-LEAK includes a function for the friction factor (f) that includes the laminar flow relationship for low Reynolds numbers (Re) representative of low flow between parallel plates [5]: 3 f = 96 Re Thus, for the small leakage cracks evaluated for the spent fuel pool heat exchanger evaluation, pc-LEAK probably is more conservative than PICEP. References

1. NUREG/CR-3464, The Application of Fracture Proof Design Methods Using Tearing for Circumferentially Through-wall Cracked Pipes under Axial Plus Bending Loads, Battelle Columbus Division for U.S. NRC, March 1986. 2. NUREG/CR-4572, NRC Leak-Before-Break (LBB.NRC)

Analysis Method for Circumferentially Through-wall Cracked Pipes under Axial Plus Bending Loads, Battelle Columbus Division for U.S. NRC, March 1986. 3. "Calculation of Leak Rates through Cracks in Pipes and Tubes," EPRI NP-3395, December 1983. 4. Norris, D., et al., "PICEP: Pipe Crack Evaluation Program," EPRI NP-3596-SR, August 1984. 5. Blevens, Robert D., "Applied Fluid Dynamics Handbook," Van Nostrand Reinhold, New York, 1984 (Figure 6-6 and Table 6-2 Frame 18). 4 5 I I + +sol ii) 4 I + ..... I (.!) I UJ r'+ + I-3 + < I Q:'. :1£ I < I / UJ / _, 2 I / 0 I / UJ I-/ . so: u I / ..... 0 / UJ / Q:'. 0... / / +/ I/ / 0 0 1 2 3 4 5 MEASURED LEAK RA TE FOR PHASE 1 IKG/Sl Figure 2-1. LEAK-00 Prediction of BCL Test Data !

i 3.00-f-----t-7--1----:7!"---t---t---::.,..i--:::...._--l---t---I 1.00 0.00 F'-----!---l----+---,__---+---+---4----!------!

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 Musur9d Leak Rate (Ibis) Figure 2-2. pc-LEAK Prediction of BCL Test Data 5 0.1 N < rO

<( Cl 0.01 c *c: Q) _..... ,,...,. . --a-pc-LEAK -tr-PICEP

a. " -0 "' --' t-"'"" u 0.001 __,.,.---...-----...-:::::;;.---

0.0001 10 Crack Length (2a), inches Figure 2-3. Crack Opening Area for Circumferential Crack 10 I-r-:----v .....-' """"' -I--E a. Cl 0.1 "' Q'. --I--/I v /?' -/--a-pc-LEAK -tr-PICEP

-" "' Q) ...J 0.01 ---I-,,r --------/ I-./ --,........---I-----0.001 10 Crack Length (2a), inches Figure 2-4. Leak Rate for Circumferential Crack 6 0.1 __,,. N < -.s -,,, cti !!? .,,, <l: C> 0.01 c *2 QJ / v -a-pc-LEAK

-&-PICEP c. 0 -" 0 u 0.001 0.0001 10 Crack Length (2a}, inches Figure 2-5. Crack Opening Area for Axial Crack 100 ./ ./. ./ 10 v -./,,, E c. C> ai iii a:: '../ #' ,v --a-pc-LEAK

-&-PICEP -" <U QJ ...J -5 _,,, ,,, ........... 0.1 ' --/ 0.01 10 Crack Length (2a}, inches Figure 2-6. Leak rate for Axial Crack 7

• NRC Question 3: Your LEFM model in the pc-LEAK program does not appear to be based on the finite element method. Provide the schematic of your model surrounding the crack and demonstrate that LEAK can account for different moduli of elasticity for materials at opposite sides of the crack. Consumers Energy Response 3: The models for computing crack opening areas in pc-LEAK, PICEP, and other codes developed by the national laboratories all compute leakage for cracks in pipes. In this situation, there is growth in both sides of the crack away from the crack centerline.

See Figure 3-1. For cracking adjacent to a nozzle, the relatively stiffer nozzle does not necessarily grow away from the crack. Instead, only the material in the vessel shell is assumed to deform. See Figure 3-1. In pc-LEAK, the crack opening area is inversely proportional to the modulus of elasticity (E) [1]. Thus, to simulate the effect of only one side of the crack opening, the modulus of elasticity was multiplied by a factor of two. This approach effectively assumes that the nozzle side of the crack has an infinite modulus of elasticity.

In addition, no credit was taken for any plastic zones at the ends of the cracks. Reference

1. NUREG/CR-3464, The Application of Fracture Proof Design Methods Using Tearing for Circumferentially Through-wall Cracked Pipes under Axial Plus Bending Loads, Battelle Columbus Division for U.S. NRC, March 1986. 8 Crack Adjacent to Penetration I I I I I I I Ocrack in Pipe 99250r0 Figure 3-1. Schematic of Crack Adjacent to and Remote from a Nozzle* 9

• NRC Question 4: Provide justification for applying the limit load solution for piping with through-wall flaws to flaws in the weld between piping and a shell. Consumers Energy Response 4: The heat exchanger nozzle weld is constructed of stainless steel. Thus, a limit load situation is used to determine allowable flaw size, consistent with the approach taken for circumferential flaws in Appendix C of Section XI of the ASME Boiler and Pressure Vessel Code. However, the weld between the shell and the nozzle is loaded in shear due to the pressure and bending moments applied in the nozzle. See Figure 4-1. For the uncracked pipe, the average shear stress (Pm) between the pipe and the steel is where: F total axial force r 0 = pipe outside radius p m 2m 0 t

• t thickness of intersection of vessel shell and pipe The shear stress attributed to bending is linearly distributed across the cross section with a maximum value of where: M = applied moment M Pb =--2-m0 t In the presence of the crack, the state of stress at limit load is a positive shear stress above the horizontal axis defined by the angle Band a negative shear stress below the axis defined by angle B (Figure 4-1 ). The shear stress is taken as one half of the Section XI, Appendix C allowable membrane flow stress (3 Sm) or 1.5 Sm. The angle B may be defined by force equilibrium in the axial direction:

or 10 The limit moment, Pb' may be determined by moment equilibrium:

or By comparison of these equations with those in Appendix C of Section XI, they are identical except for the fact that the allowable stress has been reduced by a factor of two. 11 Region of Positive Shear Stress Region of Negative Shear Stress I I I I I I I I ...i_:_ ) *\ 99249r0 Nozzle {pipe) Figure 4-1. Schematic ofNozzle in a Shell Loaded by Remote Moment and*Force 12