ML20206F887

Submits Requested Addl Info for Analysis of Firewater Cooldown for 82% Power Operation,Per

ML20206F887
Person / Time
Site:	Fort Saint Vrain
Issue date:	04/10/1987
From:	Brey H PUBLIC SERVICE CO. OF COLORADO
To:	Berkow H NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation
References
REF-GTECI-B-06, REF-GTECI-PI, TASK-B-06, TASK-B-6, TASK-OR P-87122, TAC-63576, NUDOCS 8704140391
Download: ML20206F887 (21)

Text

.

h Public Service ~ mm 2420 W. 26th Avenue, Suite 1000, Denver, Colorado 80211 April 10, 1987 Fort St. Vrain Unit No. 1 P-87122 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555 Attention: Mr. H. N. Berkow, Director Standardization and Special Projects Directorate Docket No. 50-267

SUBJECT:

Additional Information for Analysis of Firewater Cooldowi for 82% Power Operation

REFERENCES:

(SeeAttachment1)

Dear Mr. Berkow:

Reference I requested that PSC provide information concerning the  !

firewater cooldown from 82% of full power, in addition to that which was presented by PSC in References 2 and 3. PSC provided responses to NRC's first request for additional information in Reference 4. .

PSC provided responses to NRC's second request (Reference 5) for l additional information in Reference 6.

Attachment 2 provides PSC's responses to the NRC's third request (Reference 7) for additional information concerning PSC's original submittals in References 2 and 3. This additional information is provided to enable the NRC to complete its review of PSC's analysis of firewater cooldown from 82% of full power.

0 ott 8704140391 870410 0 gDR ADOCK 05000267 PDR L

h

P-87122

. Page 2

. April 10, 1987 If you have any questions about the responses in Attachment 2, please contact Mr. M. H. Holmes at (303) 480-6960.

Very truly yours, t

hH.L.Brey, Manager Nuclear Licensing and Fuels Division HLB /AW:jmt Attachments cc: Regional Administrator, Region IV Attention: Mr. J. E. Gagliardo, Chief Reactor Projects Branch Mr. R. E. Farrell Senior Resident Inspector Fort St. Vrain

Attachment 1 to P-87122 Page 1 REFERENCES 4

1) NRC letter, Heitner to Williams, dated February 3, 1987 (G-87031)

2) PSC letter, Warembourg to Berkow, dated December 30, 1986 (P-86683)

3) PSC letter, Williams to Berkow, dated January 15, 1987 (P-87002) f 4) PSC letter, Brey to Berkow, dated February 17, 1987 (P-87055)

5) NRC letter, Heitner to Williams, dated March 3, 1987 (G-87060)

6) PSC letter, Brey to Berkow, dated March 20, 1987 (P-87110) 1

7) NRC letter, Heitner to Williams, dated March 30, 1987 (G-87100) <

A I

1 i

)

i l

i 1

I

c.

Attachment 2 to P-87122 '

Page 1 ,

REQUEST FOR ADDITIONAL INFORMATION i

NRC REQUEST 1 The stress evaluation regions in the steam generator during the single cycle cooldown from power using firewater in the EES tube bundles did not consider the effects due to flow induced vibration or dynamic loading due to an SSE.

a. Provide justification for not including the effects due to flow-induced vibration,

b. Provide the basis, method and results of the stress evaluations l '

which include the effects due to SSE loading .for the Reheater tubes and their supports, the EES tubes and the EES tube support structure, the superheater helical tube bundle and its support structure, and the superheater downcomer.

PSC RESPONSE 1 la. Flow Induced Vibration Effects:

The effects of flow induced vibration were not included in the

, evaluation of the steam generator during the single cycle cooldown from power using firewater in the EES tube bundles. In the current safe shutdown cooling and Appendix R cooling analyses, the maximum helium flows are less than 4% of rated values. In testing which duplicated full operational flow velocities described in FSAR Appendices A.13.6 and A.13.7, flow induced vibration stresses were generally less than predicted and small fractions of ASME Code allowables. Therefore loadings induced by helium flows of only 4% of rated flow are considered negligible.

Ib, SSE Loading Effects:

The primary stresses due to an SSE were considered in conjunction with normal and upset operational loadings in the steam generator design report, GADR-9. SSE loadings were not combined with other i

rare events nor was there a requirement for this combination. Fort St. Vrain steam generators were designed to the ASME Boiler and Pressure Vessel Code, 1965 edition as amended by all addenda up to i and including the Winter 1966 addenda. The above ASME code edition did not contain provisions for emergency and faulted conditions. The Summer 1968 addenda to Section III of the ASME Code first addressed emergency and faulted conditions. The first regulatory guidance for

combining SSE loadings with emergency and faulted events was Regulatory Guide 1.48, " Design Limits and Loading Combinations for

Seismic Category 1 Fluid System Components," May 1973. This regulatory guide was not backfitted to older plants including Fort

St. Vrain. Regulatory Guide 1.48 was subsequently withdrawn in 1985, and was replaced by the July 1981 revision of Standard Review Plan i

v.

. Attachment 2 to P-87122 Page 2 3.9.3 "ASME Code Class 1, 2, and 3 Components, Component Supports, and Core Support Structures."

It is therefore concluded that combining the SSE with faulted events is a current regulatory requirement but not applicable to plants constructed prior to Regulatory Guide 1.48 in 1973. This is further substantiated by the NRC Systematic Evaluation Program (SEP) conducted in the late 1970's and early 1980's which evaluated the oldest eleven nuclear power plants to selected current licensing requirements. Load combinations were an SEP topic but were deleted because load combinations were also an unresolved safety issue (USI).

Load combinations remain an unresolved safety issue today, and are USI Item B-6. The following is a quote from Page 2.B.6-2 of NUREG-0933, A Prioritization of Generic Safety Issues, " Reactors constructed prior to 1972 did not have design requirements which included SSE, LOCA, and pipe cooling considerations."

Based on the above information, PSC concludes that the NRC position, expressed in Reference 7, that SSE loadings must be combined with the loadings of other rare events initiating a firewater cooldown (i.e.,

tornadoes, HELB, Appendix R fires) represents a change in a previously accepted staff position. It is not apparent to PSC that a substantial additional safety benefit would result from this new position. However, without accepting that the requested load combination is applicable to Fort St. Vrain, we are responding to your specific request for additional information to facilitate a timely review of PSC's safe shutdown cooling analyses. The responses in fact illustrate that SSE loadings comprise only a small portion of the total loading during normal operation as well as during firewater cooldowns.

EES Tube Bundle:

The effects of the SSE event on the Superheater II tubes are already included in the GA Document 909204, Issue N/C, evaluation, but were not specifically identified as such. The stresses due to dead weight on Page 16 on GA Document 909204 include the seismic contribution.

These stress values were obtained from Section 6.3.4, Page 20 of GADR-9.

See the Attached Table Ib, for a detailed stress evaluation including stress allowables for the Superheater II. From Table Ib it can be seen that seismic induced stress constitutes only a small fraction (6%) of the total stress which in turn is within code allowable limits.

Superheater Downcomer:

Since design temperature for the Superheater Downcomer is 1221 degrees F (GADR-9 Summary Table) and the tube temperature during the firewater cooldown event is 300 degrees F (Page A-5 of GA Document 909204), the restrained thermal expansion stresses will be reduced accordingly. The pressure stresses are lower by a factor of almost three during the firewater cooldown event (Table 5.1 of GA Document 909204). The SSE stress of 614 psi (GADR-9 Section 6.4.5.2, Page 5) 1

_ ___ * ' ~

a-Attachment 2 to P-87122 Page 3 ,

will remain the same. The allowable stresses during the firewater cooldown event are greater than the 100 percent power condition by about a factor of two, so the firewater cooldown event has much higher safety margins than the 100 percent power conditions.

Reheater Tubes:

The critical failure mechanism of a reheater tube has been identified as creep collapse under external pressure. The seismic loading on the FSV reheater tubes produces very low stress levels (550 psi, Section 6.5.4, Page 193 of GADR-9). A low level vibratory bending load superimposed onto an external pressure creep collapse load of a tube has no significant effect. The SSE bending stresses in the tube are in the axial direction, while the collapse stresses are in the hoop direction. Only if the tube bending load was large would the secondary effects from the additional tube ovality caused by bending of an already curved tube require evaluation.

Reheater Tubes and Supports:

Because the reheater tubes and their supports are basically at the same temperature during the EES firewater cooldown event, the interaction loads between the tubes and support structure due to restrained thermal expansion are negligible. The thermal stresses due to tube wall radial gradient and restrained thermal expansion are negligible for the EES firewater cooldown event. The 100 percent load condition evaluated in GADR-9 considered the SSE event in conjunction with the dead weight and helium drag loadings, (3220 psi, Section 6.5.4, Page 96 of GADR-9). This combination of dead weight, SSE and pressure met the ASME Code primary load requirements at 100 percent power. Because the loads and stresses from pressure and thermal effects during the EES firewater cooldown event are lower than the 100 percent loads and stresses, and the allowable loads and stresses for this one time short term faulted event are higher, the SSE when combined with the firewater EES cooldown event will not fail the reheater tubes.

l l

l l

Attachment 2 to P-87122 '

'Ibble Ib.

SiWar II 'Ibbes imriiwy Stress GAIR-9 Section Stress Dir=+im Owiitim calm 1ation 6.3.4 page No. Hoop Iong.

(See Note 1) psi. psi.

Flow Ind. Vib. Insignificant -

0 0 Deadweight Insignificant 19 0 0 Seismic 0 hoop 2156 long. 20 0 2156 Pressure hoop 3291(-613)/1822 28 1107 -

Pressure long. 389(-613)/1822 28 -

131 Bearhug 5702(175)/136 39 0 Wall thermal 7337 (See Note 2) 42916(135)/210 32 27589 27589 (See Note 3) gradient Allowable Stress carhinations Stress, psi Primary Mernbrane stress (See Note 4) 1107 131 17920 (Pressure only)

Primary Maubrane plus Primary Bending stress 1107 2287 26880 (Pressure and SSE)

Total stress, inc. Secondary Bending and Thermal 28696 37213 78000 (Pressure, SSE, Bearhug and Wall thermal gradient)

Notes:

1. This calculation converted full power operation stresses (first number) to firewater cooldown stresses using nultipliers/ dividers obtained frcrn Table 5.1 of GA Document 909204. For convenience a copy of Table 5.1 of GA Doc. 909204 is attached. The full power operation stresses were obtained fran GADR-9 Seton 6.3.4, page nunbers as imilcated.

2. Bearhug stress is a secondary stress caused by restrained thermal expansion of the helical tube bundle by the tube support plates.

3. Tube wall thermal gradient stress is a secondary stress caused by i tenperature difference through the tube wall frcan the heat flux.

4.a Allowable Stress for faulted conditions:

Primary Mesubrane Allowable Stress is S .

Primary Mernbrane plus Primary Bending Y Allowable Stress is 1.5S Primary Menbrane plus Primary Bending plus all Secondary (Tota 1Y stress)

Allowable Stress is 3Sa*

Material properties of SB-163 Grade 2, (Sanicro 31) were taken fran:

ASME B&W Section III Code Case 1342-1. S = 17920 psi 8 385 0F.

ASME B&W Section III Sunmer 1968 Edition. SY = 26000 psi @ 385 F.

8 (See Note'4b) 1 4.b This is the first edition of the ASME B&W Code where limits are provided for Faulted conditions. l

Table 5.1 SUPERHEATER II LOADINGS I.OADING PLATE TEMP. "F TUBE TEMP. "F TEMP. DIFF. PLT.-TBE. TUBE EXTERNAL INTERNAL TUBE CONDITION INLET OUTLET INLET OUTLET INLET DUTLET WALL DT F PRESSURE PRESSURE P.DIFF.

FULL 1099. 1132. 963. 1097. 136. 35. 210. 690. 2512. 1822.

LOAD

. QUARTER 932. 1032. 875. 1032. 57. O. 87.5 589. 2419. 1830. EDN LOAD TRANSIENT 1110. 1135. 995. 1090. 45.

9A @20s.

15.  ? 469. 2713. 2244. $E!

g g r+'

TRANSIENT 1110. 1135 997. 1100. 113. 25.  ? 469, 2713. 2244.

9A @50s. h TRANSIENT 1085. 117.V . 965. 1105. 120. 15.  ? 469. 2713. 2244.

9A @l00s.

TRANSIENT 1015. 1045. 915. 1015. 100. 30.  ? 469. 2713. 2244.

9A @300s.

EES F/W 455, 301. 286. 294. 169. 7. 128. 700. 87. -613.

COOLDOWN 75% POWER EES F/W 468. 302. 293. 296. 175. 6. 135. 700. 87. -613.

C00LDOWN 105% POWER

$# -e Information on full and quarter load and transient 9A obtained from Ref. 7, Section 6.3, p 16. 7N gg (Excerpted from GA Document 909204,ft/A) $$

e-Attachmant 2 to P-87122 Page 4 ,

NRC REQUEST 2 In Attachment 7 of P-86683, it is stated on Page 3 that during the 90 minute period the tube metal temperature is less than 1300 degrees F, while calculation No. 68-02 of Attachment 9 indicates that the maximum of all tube temperatures in this transient is 780 degrees F.

Provide a reconciliation and basis of these two values, indicating which is the correct value.

PSC RESPONSE 2 The maximum tube temperature during the first 90 minutes varies from 1000 degrees F at the start to 780 degrees F at time 90 minutes into the interruption of forced circulation event following reactor operation at 105% power. See Figures 7-1 and 7-2 in PSC Response 7.

The structural evaluation of the steam generator during a delayed firewater cooldown was being performed concurrently with the system analysis. Temperatures and pressures for the structural evaluation were conservatively chosen to assure that they would envelope those from the final system analysis. A more appropriate temperature value for the first 90 minute period would be 1000 degrees F. However, the 1300 degrees F used for the first 90 minute period in the reheater structural evaluation is conservative.

l l

1 I

1 l

o.

l Attachment 2 to P-87122 l i

Page 5 ,

NRC REQUEST 3 Provide the detailed methodology used for performing the creep collapse analysis of the reheater tube, and the User's Manual for the computer program " BUCKLE". Provide assurance that the time to reach yield stress in the maximum stressed point is shorter than the time at which the ovality becomes unbounded, i.e., show that collapse can not occur at a maximum stress which is lower than the yield stress.

PSC RESPONSE 3 The GA Buckle computer code solves the equations used in the Pan method (Ref. 4 of GA document 909204, Issue N/C), which is a published and industry accepted method. The buckling time is defined as the time at which an element of the tube reached a stress level equal to the material yield stress at the temperature of the tube.

The code tracks the distorted shape, and when the combination of shape and loading produces a stress equal to the yield stress, the tube is deemed to have failed. This is conservative as further deformation is necessary to cause the plastic hinge to form through the tube wall, which may allow the tube to collapse. Work hardening which wculd increase the time to failure is ignored in this analysis.

Because of the high temperatures assumed for the analysis and the short times involved, revisions to the computer code were required.

The GA Users Manual is being updated to incorporate these revisions, and will be released and submitted to the NRC by April 30, 1987.

For a long tube to collapse under the influence of external pressure loading without the yield stress of the material being exceeded, the tube must be thin walled relative to its mean diameter. This hypothesis can be shown to be valid by using the critical pressure from Equation 11-12 of Reference 1 to determine the hoop stress due to this critical external pressure for various mean diameter to wall thickness ratios. The tube mean diameter to wall thcikness ratio has to be generally much greater than twenty for external pressure buckling of a tube to occur before the material yield stress is reached. Above a ratio of twenty, collapse can occur without any elements of the tube reaching the yield stress of the material. The Fort St. Vrain reheater tubes are thick walled, having a diameter to wall thickness ratio of eight. Therefore the reheater tubes could not collapse with an element stress lower than the material yield  ;

stress. l i

Reference for PSC Response 3: l

1) Timoshenko, S. P., and Gere, J. M. Theory of Elastic Stability, Second Edition, McGraw-Hill Book Co., Inc.

)

i I 1

i

o-Attachment 2 to P-87122 Page 6 NRC REQUEST 4 Provide a basis for the statement on Page 16 of attachment 7 that "the structural integrity of the steam generator is not likely to be compromised due to excessive tube / plate interaction loads during the EES firewater cooldown event," in particular during simultaneous SSE loading.

PSC RESPONSE 4 The structural integrity of the steam generator is not likely to be compromised due to excessive tube / plate interaction loads during the EES firewater cooldown event because the summation of the tube stresses during the event are less than during the 100 percent power condition. The restrained thermal expansion (Bearhug) loading is classified as a secondary loading, the other secondary category load is the tube wall thermal gradient. Restrained thermal expansion (Bearhug) stresses increase to 7337 psi during the firewater cooldown event from 5702 psi during the 100 percent power condition. The tube wall thermal gradient stresses decrease to 27589 psi during the firewater cooldown event from 42916 psi during the 100 percent power condition. The summation of these secondary category loadings for the firewater cooldown event is therefore less than during the 100 percent power condition. The total longitudinal stress, including primary and secondary stresses which include a 2156 psi SSE load contribution, would be 37213 psi as opposed to a faulted allowable stress limit of 78000 psi. See Table Ib in PSC Response 1 for details of the stress contributions from SSE, deadweight, pressure, flow induced vibration, restrained thermal expansion (Bearhug), and tube wall thermal gradient.

r-j

. Attachment 2 i to P-87122 Page 7 l i

NRC REQUEST 5 It is stated that Sanicro 31 reheater tube material is a European Alloy 800 - type material, which met the required material specifications of Alloy 800 Grade 2 at the time when the steam generators were built. Provide the chemical composition of this material and its strength levels (i.e., yield and ultimate strengths at appropriate temperatures). In addition address the metallurgical treatment of Sanicro 31 in comparison with the treatment of the ASME Code Alloy 800 material. Further, explain why the Sanicro 31 will not degrade to below the minimum levels allowed for the Code material because of the temperature cycles which the steam generator tubes experience.

PSC RESPONSE 5 Clarifying the NRC Request 5: When the Fort St. Vrain steam generators were designed, Ni-Fe-Cr alloys, Grade 1 and Grade 2 (the modern equivalents which are Alloy 800 and Alloy 800H, respectively),

were included in Code Cases 1325-2 and 1342-1. The Fort St. Vrain steam generator tubing was manufactured by Sandvik, to the ASME B&PV Code specification Ni-Fe-Cr alloy, Grade 2 as given in Code Case 1342-1. This material was given the designation Sanicro 31 by Sandvik.

The chemical composition of Sanicro 31 used in the construction of the Fort St. Vrain steam generators is the sama as published in ASME B&PV Code Case 1342 ' (see Figure 5-1). The material properties (i.e., yield and ultima ' strengths at appropriate temperatures) of Sanicro 31 (Ni-Fe-Cr alloy, Grade 2) used in the construction of the Fort St. Vrain steam generators satisfy the requirements of ASME B&PV Code Case 1342-1 (see Figure 5-1). This Sanicro 31 material was annealed to approximately 2100 degrees F, as given in Code Case 1342-1, to meet all the requirements of Ni-Fe-Cr alloy, Grade 2.

The Fort St. Vrain steam generator tubes were designed using the ASME Code Case 1342-1 design stress intensities including consideration for temperature cycling over the design life of the steam generators.

However, in late 1975 it was recognized that potential tube failures may occur in the Fort St. Vrain steam generators during service due to the presence of cold formed (non-heat treated) tight radius bends in the Sanicro 31 tubes. These concerns and programs are discussed in FSAR Appendix A.13.27. The characteristics of concern in the cold formed small radius bends, were recrystallization, low ductility rupture and creep-fatigue damage. Of the three identified concerns, low ductility rupture was shown in Appendix A.13.27 of the FSAR to no longer be a problem. The recrystallization and creep-fatigue damage in the cold formed small radius bends were also subsequently found not to be problems as reported in the Electric Power Research Institute (EPRI) Report EPRI-HTGR 86-03, " Properties of Recrystallized Ni-Fe-Cr alloy,H and Associated HTGR Steam Generator Design Implications," dated July 1986. The tests reported in EPRI-HTGR 86-03 were performed on a FSV archive heat of Sanicro 31 that met all the specification requirements of Ni-Fe-Cr alloy,H.

Therefore, it was concluded that there are no identified

O

Attachment 2-to P-87122 i Page 8 t -

, i metallurgical problems that could degrade the service life because of the temperature cycles which the steam generator tubes experience.

k f

k i

a i

i i

?

4 1

I i

4 i

i 4

)

1 i - < ,. .,,----- -- - .. ..-.---.---- , - - - - . , , - , , . - , . _ . . - , . - - , , , . . - - . , . - - . . . . . - . . ~ . - , - - - - , , - - - , , - , - , . - - , , . -- - - - , , , , . -

ATTuccewr '2-Fi c,u ne 5-1 T N '"

CASE (PAcas I oF ?-)

jgy INTERPRETATIONS OF ASME BOILER AND PRESSURE VESSEL CODE Approved by Council Alarch 14,1966 Case 1342-1 (Special Ruling) tilechanical Propertics:

Requirements for Nickel-Iron-Chromium Grade 1* Grade 2" Alloy (All Product Forms) ,

Section 111 Tensile Strength, psi, minimum , a,000 _0,000 Inquiry: May . nickel-iron-chromium Yield Strength, _

allovs he used in the construction of 0.2 per cent Class A Nuclear Vessels in accord. Offset, psi, and with Section 111, and what re. minimum 30,000 25,000 quirements apply to these materials. Elongation in Reply: It ;is the opinion of the com- cc ;in$u$ 30 30 mittee that nickcl ,ron-chromium i alloys may be used in the construction of

• Anncaled ut opproximately 1800 F Class A Nuclear Vessels and the fol- " Annealed at approximately 2100 F lowing specified special requirements apply in addition to the requirements 2) Fabrication Requirements, for various product forms specified in Welded fabrication shall conform to Section Ill. the applicable requirements of Section

1) Materials conforming to Specifica- 111.

Lions 5B-163 and/or ASTM IM07-63T, 11408-63T, U409-63T and conforming to

" '. pr edure and performanec t,i c 7ollowing requirements as t qualifications shall he conducree ns _

chemical compos,i tion and mechanical prescribed in Section IX, Part B, except that the tensile strength of the reduced P' P " S section specimen shall be not less Chemical Composition: th an th e minimum tensile strensth Per Cent specified for the base material.

Nickel 30.0 -35.0 Chromium 19.0 -23.0 b)E.id.ing c shall be done by any Ir u il lance welding process or combination of Carbon processes capable of meeting the 0.10 max. requirements.

ManFanese 1.50 max.

Sulphur 0.015 max. c) Welds that are exposed to the Silicon 1.00 max. corrosive action of the contents of the Copper 0.75 max. vessel should have a resistance to cor-Aluminum 0.15- 0.60 rosion that is not substantially -less Titanium 0.15- 0.60 than that of the base metal. The use of Meeting of January 14, 1966 Reprinted from MECHANICAL ENGINEERING May,1966, a pub!! cation of the American Society of Mechanical Engineers.

Pages 319 thru 320 replace pages 205 and 206.

319

Arrneumeur '2.

CASE (continuedl/  % NI" A Gorz.e 6-1 1342-1 (pac # rz. or. t)

INTERPRETATIONS OF ASME BOILER AND PRESSURE VESSEL CODE filler metal that will deposit weld Table 1.1 metal with practically the same com- -

position as the material joined is Sir in Limiting Factors recommended. When the manufacturer is 1 of the opinion that a physically-better Strain, Per Cent Factors joint can be made by departure from these limits, filler metal of a different 0.10 0.90 composition may be used provided the 0.09 0 .11 9 strength of the weld metal at the 0.08 0.!!!!

' operatinF temperature is not appreciahly 0.07 0.!!6 less than that of the high-alloy material 0.06 0.!!3

~ ~

to be welded, and the user is satisfied 0.05 . 0.!!0 g that its resistan ce to corrosion is 0.04 0.77 satisfactory for the intended service.

0.03 0.73 i

d) A separate welding procedure 0.02 0.60

and performance qualification shall be 0.01 0.63

made for this innterial.

i e) llent treatment after forming or j fabrication is neither required nor

prohibiicd.

Table 2 i 3) llesi::n stress intensiiv values are Yield Strength values I

listed in Table 1.

4) Yield strength values are listed in 3 Table 2. Grade 1 Grade 2 Table 1 Min. Spec. T.S. 30000 250_00_, ._

Design Stress Intensity, psi' Tem. F Temperoiure not exceeding F Grade 1 Grade 2 100 30000 2i>000 j gg 3 600 21500 100 20,000 16,700 200 20,000 16,700 300 2r>000 19200 300 20,000 16,700 400 25000 17600 r,00 24100 16500 400 20,000 II',000 l 500 20,000 14,900 (>00 23000 1(>200 600 20,000 14,600 799 3.g ,gg g 7,ggg 650 20,000 14,400 800 23000 Irir,00 700 20,000 14,300 750 20,000 14,200 000 20,000 14,000

• The design st re ri a intensity v.il u r is

! controlled by the IcNect value of $ of the  !

j minimum specified yield strength ut ruum I

t e mp era ture, or 0.9 of the 0.27, cffset 1 yield strength ut temperuture.

l 320

~ '

ry

. Attachment 2 to P-87122 Page 9 NRC REQUEST 6 It is implied that Sanicro 31 is equivalent to an Ni-Fe-Cr alloy, H material, e.g., SB-163 or SB-407. If this is the case, justify the value of 23,300 psi at 380 degrees F for Sm. Current ASME Code values indicate Sm values of 16,700 psi for 58-163 and 20,000 psi for SB-407 at 380 degrees F.

PSC RESPONSE 6 The values of Sm on Page 16 of the GA Technologies, Inc. Report 909204, issue N/C were not used in the analysis, and were incorrectly added to the report. The lines that reference Sm should be removed.

The correct Sm value for Sanicro 31 (SB-163 Grade 2) at 380 degrees F is 16,000 psi. This Sm value was obtained by interpolation from the ASME B&PV Code Case 1342-1 (see Figure 5-1 in PSC Response 5). The Sm value was not required nor used in the current analysis, s

]

J

..e, - . -- m-m --~e--a - - - - < - - > . - - - - - - - - - - , - - - - - , - -----.n- ,. - - - - - - -

a-Attachment 2

. to P-87122 Page 10 NRC REQUEST 7 Figure 4.4 is incomplete as submitted. The temperature profile for the time in hours from 0 to 2 is missing. Please supply this information.

PSC RESPONSE 7 The temperatures in the steam generator during the period of no forced circulation (1-1/2 hours) are equal to or less than the normal operating temperatures and thus were not included in Figure 4.4 of GA Technologies, Inc. Report 909204, Issue N/C. As stated in PSC Response 2, temperature / pressure profiles were conservatively established to assure they enveloped the final system analysis. The final system analysis has circulators started at 1-1/2 hours into the transient and not at 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.

A study was performed by GA Technologies in early 1986 to define bulk temperatures in the steam generator during an interruption of forced

, cooling (10FC) following reactor operation at 105% power for periods as long as five hours. The study considered the steam generator to be either dry (empty) or full with retention of the water and steam inventory. The analysis was performed using the RATSAM computer code (Reference 1), and the analysis was documented in Reference 2.

Figures 7-1 and 7-2 (Figures 2 and 3 from Reference 2) show bul k steam generator temperatures varying with time. At time two hours the bulk steam generator temperatures range from about 620 degrees F to 750 degrees F which are well within the normal operating range.

Retention of the water and steam inventory, Figure 7-2, has only a small effect on bulk steam generator temperatures.

References for PSC Response 7:

1) R. K. Deremer and T. Shih, "RATSAM: A Computer Program to I Analyze the Transient Behavior of the HTGR Primary Coolant System During Accidents," GA Technologies, Inc. Report GA-A13705, May )

1977.

2) G. J. Cadwallader and A. Bagierek, "FSV Steam Generator Temperatures During 10FC From 105% Power," GA Technologies, Inc.

GA Document 908784, Issue N/C, March 19, 1986.

- - - , ' - - =r -

FIGU R E 2 FSV STEAM GENERATOR TEMPERATURES - CASE 1 (WET) .

1000-P 7 CURVE IDENT.

n _

l k _

n

_ v e 900 -

REHEATER tti _

Q A v _ ~

i _

SUPERHEATER II

~

E m 3 800 "

3Q - 3 SUPERHEATER I E

An C* "

• EVAPORATOR (D L-A 'A 1 O M "

_' g Ld F

700 < -

A A

7.d -

C

.__ y  :

i NOO o y

~

ECONOMIZER f

d 600 1 1

=

1 a

NM T i

y

_1 I STM GEN OUTLET b_ _

lti -ID y -

o_ ,

I 500 i i i i iiii i i i i i i i i i i i i g@

0 1 2 3 4 5

$ sh-i s

e TIME CHOURS)

IDFC 9 105% POWER "s

s s

FIGURE 3

FSV STEAM GENERATOR TEMPERATURES - CASE 2 CDRY) -

1000-O I CURVE IDENT.

n -

it- -

- O to 900 REHEATER sv

~

A

< i SUPERHEATER II K

11 g 800 c p <

\ - <

SUPERHEATER I c oc w -

O o_ _ -

r 4 --- --

n a e=r? S Ov EVAPORATOR u T p

a ne w 700 <

~

r r* -

y z

x U v 0F V ECON 0MIZER x

600 [^" "^7 g~

.t.

y w STM GEN OUTLET it. -

M - o?

Ps 500 i i i i iiii i i i i i i i i i i i 1 3G

-3 ape j

e 0 1 2 3 4 5 j TIME CHOURS)

P IDFC 9105% POWER i

o-

. Attachment 2 to P-87122 Page 11 NRC REQUEST 8 The possibility of accumulated creep-fatigue damage contributing to tube failure is not explicitly addressed in this submittal. Please provide an evaluation of this potential failure mechanism in the context of the creep-collapse failure mode that was assumed.

PSC RESPONSE 8 Prior creep-fatigue damage in a steam generator tube can affect the

tube's creep-collapse failure mode if the prior accumulated creep-fatigue damage is high. The Fort St. Vrain Steam Generators were designed and licensed to the ASME B&PV Code Section III (through the Winter 1966 Addenda) and eleven Code Cases including Code Case 1342-

, 1. These ASME code rules did not require creep-fatigue damage evaluation. However, the fatigue analysis performed in the steam generator design report, GADR-9, showed low fatigue damage even in those regions of the steam generator where there were significant cyclic loads. Because the heat flux loads in the reheater are low, there is no reason to expect that there will be significant creep-4 fatigue damage in the reheater tubes particularly adjacent to the small radius bends where the maximum ovality is expected to occur.

(It is at these maximum ovality points that creep-collapse would firstoccur.)

Creep-rupture damage (a component of creep-fatigue damage) induced during the transient event analyzed is also small because of the low stress levels involved. The following table gives the creep-rupture damage increments calculated for both the previously defined 105%

power case and the recommended 87.5% power case.

Power Time to Actual Creep-Rupture Level Temp. Press. Stress Rupture Event Time Damage Fraction

% deg. F psi ksi hr hr 105 1660 350 3.5 879 2 0.0022 87.5 1500 350 3.5 34690 2 0.0001 Creep-fatigue damage (obtained by summing creep-rupture damage and fatigue damage) has to reach a value of 1.0 before failure is predicted. Since the creep-rupture damage fraction sustained during the transient event is small and prior accumulated creep-fatigue damage has not been shown to be large, this creep-fatigue phenomenon does not have a significant influence on the predicted times for i

creep-collapse of the tubes.

o.

-Attachment 2 to P-87122 Page 12 NRC REQUEST 9 With regard to the Fort St. Vrain (FSV) Technical Specifications, 5.3.11 - Steam Generator Bimetallic Welds Surveillance and SR 5.3.12

- Steam Generator Tube Leaks Surveillance, describe the results of the surveillance requirements including the number of times and the dates that they have been implemented. Provide any other information that is available concerning degradation of the steam generator tubes.

PSC RESPONSE 9 FSV surveillance test SR 5.3.11, which requires the volumetric examination of steam generator bimetallic welds for indications of subsurface defects at five calendar year intervals, was approved as part of Amendment No. 39 to the FSV Operating License.

Implementation of this testing at FSV is per Inservice Inspection (ISI) Criterion C, which states that this surveillance requirement shall be implemented before the beginning of fuel cycle 5. Since the FSV plant is still in the early life of fuel cycle 4, no examinations per SR 5.3.11 have been performed to date.

FSV surveillance test SR 5.3.12, which requires an extensive evaluation of each steam generator tube leak, was approved as part of Amendment No. 45 to the FSV Operating License. There are no new tube i

examination results available at this time because there have been no new tube leaks since the last (second) leak occurred in the Superheater II tube bundle of module B-2-3. The results of that tube leak metallographic evaluation were reported (GA Report GA-C17042) to the NRC in PSC letter, Warembourg to Collins, dated September 16, 1983 (P-83311). This report found no significant or unusual degradation of the Sanicro 31 subheader tubing that was examined by GA Technologies, Inc.

l Similar results were reported for the first tube leak that occurred

, in the Superheater II tube bundle of module B-1-1 in December of l 1977. This leak and results of GA Technologies, Inc. metallographic l examination were reported in GA Report GA-D15395 (December 1977).

1 4

l l