ML20214P989
| ML20214P989 | |
| Person / Time | |
|---|---|
| Site: | Three Mile Island  |
| Issue date: | 09/15/1986 |
| From: | Elvin T, Vames G BABCOCK & WILCOX CO. |
| To: | |
| Shared Package | |
| ML20214P980 | List: |
| References | |
| NUDOCS 8609240130 | |
| Download: ML20214P989 (15) | |
Text
-
ATTACHMENT gl g l-0162 BwRP-20440 3 (8/85)
Bah &Mcox ENGINEERING INFORMATION RECORD a McDermott cornparty Safety Related:
Document identifier 51 -1165539-00 Yes G % 0 y itle TMI-2 Incore Nozzle Evaluation, Prepared by G. J. Vames N September 12, 1986 Date Reviewed by T. W. Elvin N September 15, 1986 Date AML Remarks:
Purcose and Summary:
The purpose of this document is to present the results of an evaluation of the load carrying capability of an incore nozzle. The basis for the evaluation is that the recently reported data and observations made during the core bore program are typical of the conditions in the lower head of the TNI-2 reactor vessel. Based on our review of the videotapes of the core bore p rog ram',
environmental and metal l u rgical observations, calculations and engineering judgement it is concluded that virgin material properties can be used for deterTnining the strength of the incore nozzle welds because it is highly unlikely l that the temperature of these welds exceeded 1200 0F.
l 1
l However, since it is not possible to examine the welds, it is recommended that some factor of safety be applied to the calculated strength to obtain permissible loads. Babcock & Wilcox recommends a factor to two.
l 8609240130 860919 PDR ADOCK 05000320 P PDR Page 1 of 10 l
i 1
51-1165539-00 Page 2 Table of Contents EAGE
- 1. Introduction 3
- 2. Temperature Evaluation 4
- 3. Envi mnmental Considerations- 6
- 4. Metallurgical Considerations 8 5 . :- Estimated Nozzle Strength 8
- 6. Conclusions and Recommendations 9
- 7. References 10
51-1165539-00 Page 3 1., Introduction An evaluation of the structural integrity of tne TMI-2 reactor vessel lower heaa was performed by B&W in June,1985. At that time video examination had revealed that a void existed in the upper region of the original core which encompassed approximately one-third of the total core volume and extended to the outermost, partially damaged fuel elements. A debris bed approximately three feet thick lay in the ~ bottom of the icwer reactor vessel head. Efforts to probe through the debris indicated that a layer of hard, impenetrable material lay beneath the surface at about mid-core elevation. Video scans of the lower region indicated that ten to twenty percent of the core material collected in the lower head. It was thought that this material flowed to the bottom of the reactor vessel while it was in a molten condition and that a " chimney" was formed in the center of the Core.
From the data available at that time, the B&W study indicated that the corium formed by the fuel in the bottom head reached temperatures in the range of 4000-5000F, and that the inner surfaces of the reactor vessel head (cladding) reached temperatures in the range of 2100 - 2400F which is below the melting temperature of 2760F for the stainless steel cladding and the melting temperature of 2450F for the Inconel weld metal for the in-core instrument nozzle. Some of the in-core nozzles were sticking up into the corium and it ~was j udged that they probably melted in that area.
Since the observations were made last year, additional information has been obtained. Holes have been drilled down through several of the fuel elements.
Video scans of the drilled holes show that the lower portions cf the fuel s
l l
\
51-1165539-00 Page 4 elements are still intact, ranging in lengths from ~ 11 to 48 inches with the former being located near the center of the core. No " chimney" or void space has been located, although it could conceivably be located at some place that has not been drilled yet. The video scans indicated that the lower end fittings and bolting for the fuel elements are still intact. The scans also indicated that the mate ri al in the region just below the fuel elements (e.g., lower flow di stri butor) did not suffer any noticeable damage. However, no views are available for the regions near the in-core instrument nozzie welds.
In order to take advantage of this new information, a second study was conducted by B&W and documented in this report. The latest information and video tapes f rom the site were reviewed and a new as.iessment of the structural integrity of the lower head was performed. Thermal, environmental, and metall urgical conditions were considered in the eval uation. The resulting conclusions and recommendations are presented in Section 6 herein.
- 2. Temoerature Eval uation A review of the original temperatu re calculations (Reference 1), recent inspection results (References 2 and 3), and the latest video inspection results was performed in order to refine the lower head temperature estimate. Based on this review, it is likely that the temperature of the incore nozzle welds in the reactor vessel lower head did not exceed 1200 degrees F and probably did not exceed 900 degrees F. The following observations were made in making this j udgeme nt:
51-1165539-00 Page 5 A. The calculation performed in Reference 1 were performed before much was known about the conditions in the lower head. Thus these calculations had to be very conservative. Specific conservatisms were:
- 1. All fissions products were included in the heat source term.
- 2. The theoretical density for a UO2 zircalloy and inconel mixture of 8.95 was used. The average of seven measurements reported in Reference 2 was 6.86 gm/ cubic-cm.
- 3. The geometry was assumed axisymmetric around a guide tube nozzle and was insulated at the top which was seven inches from the vessel head.
- 4. The thickness of the corium ring around the guide tube nozzle was four inches.
B. Even though the calculations were very conservative, they showed the essential physics of the problem, which are:
- 1. The stored energy is more important than the decay heat in determining the maximum temperature (assuming initial temperatures in the corium of 5000 degrees F).
- 2. The vessel acts as a very large heat sink or " chill block."
C. None of the instrument guide tube nozzles showed any sign of damage; thus the guide tube nozzle had to be below its melting point (about 2900 degrees F). The calculations performed in Reference 1 predicted temperatures as high as 4100 degrees F.
51-1165539-00 Page 6 D. The only damage observed was to one incore instrument guide tube where a portion of the lower section which is only 1/4 inch thick had melted. A simple finite difference simulation of hot corium striking the guide tube was performed. It indicated that the 1/4 inch thick section would melt in less than a minute, while the thicker section did not melt in the ten minute simulation. This indicates that, for the most part, the corium that fell through the six inch diameter holes in the elliptical flow distributor plate did not drift over to the incore guide tube and by extension to the incore nozzles. Thus; it is highly unlikely that a monolithic ring of corium surrounds the incore guide tube nozzle as was assumed in the analysis. A nonsymmetric analysis where monolithic corium was only present on one side of the nozzle would show much lower temperatures because of the heat sink capabilities of the vessel.
E. The largest piece of recovered corium has a volume of about 15 cubic inches and most of the recovered pieces have volumes of two cubic inches and less.
This must be contrasted to the 176 cubic inch volume which was used in the original analysis. Again, much smaller volumes of monolithic corium will lead to much lower maximum temperatures.
- 3. Environmental Considerations B&W has made many evaluations related to the environmental (chemistry) effects of the RCS material of construction. The most significart ones are reported in
!M References 4, 5, 6, and 7. These references cover the period from the start of I the incident on March 29, 1979 through February, 1986. Subsequently, hydrogen peroxide was used in the reactor coolant to destroy bacteriological growths.
l l 51-1165539-00 i
Page 7 During the incident itself the RCS temperature, pressure, and water inventory went through wide ranges of conditions that made it difficult to provide a proper assessment of the chemistry environment. However, as soon as conditions had stabilized, a basic pH water environment was established by the addition of sodium hydroxide during the drawdown of the borated water storage tank. A basic pH (>7.5 at 77F) has been maintained, as pH is one parameter that can be controlled through the addition of sodium in the form of sodium hydroxide.
A basic pH is a key item in reducing the susceptibility of both austenitic stainless steels and high nickel alloys to stress corrosion cracking. Thus, with a basic pH in the RCS, the possibility of stress corrosion cracking has been reduced or minimized.
Inconel 600 is susceptible to stress corrosion cracking when reduced forms of sulfur are present. During recent periods when hydrogen peroxide was used as a biocide, a strongly oxidizing environment existed which would help oxidize the reduced sulfur species to sulfate, the only oxidized specie of sulfur that can exist. Sul fate has a low potenti al of being involved in any stress corrosion cracking phenomenon. Thus, it can be stated that the presence of hydrogen peroxide can be hel pf ul in reducing stress corrosion of high nickel alloys if.
reduced sulfur species are present.
Based on this evaluation it appears that ro environmental conditions detrimental
. to stainless or high nickel steel have existed in the TMI-2 reactor vessel since the incident on March 29, 1979.
51-1165539-00 Page 8
- 4. Metallurgical Considerations As. stated in Section 2 above, the reactor vessel lower head and incore nozzle temperatu re likely did not exceed 1200 F at any time. This is consistent with the video tapes which show little or no damage to the fuel assembly lower end fittings and lower core support assembly. While the incore nozzle welds to the lower reactor vessel were not visible in the video tapes, it is not likely that they sustained higher temperatures than the lower core support assembly since the debris would have cooled as it fell from the core and since the lower head acts as a heat sink as described in Section 2.
Based on these considerations, in addition to the conclusions in the previous sections, the Inconel 600 weld between the incore nozzles and the lower head can be expected to have its original metallurgical properties.
S. Estimated Nnzzle Strength The an alyses presented herein and in Reference 1 indicate that the critical component in the reactor vessel lower head is the incore nozzle weld. Therefore, only this weld is analyzed. During the remaining defueling program loeds could be hypothetically applied to the incore nozzles by the core bore machine, the impact chisel, by load drop or other accidents. Therefore, tha load carrying ability of the nozzles should be known.
0 51-1165539-0')
Page 9 The ultimate strengths of a nozzle separately in extension, bending and torsion are calculated in Reference 8. The minimum ultimate strength of Inconel 600 of 80,000 psi was used as indicated by the evaluation in the preceding sections of this report. This minimum strength is applicable from room temperature to 1000 degrees F. The nozzle was treated as a twelve inch long cylinder with a two inch outer diameter and a 5/8 inch inner diameter. This cylinder was cantilevered up from the lower head and loaded at the top. The ultimate bending and torsional loads were calculated using conservative moduli of rupture' for alloy steels.
The,results of the analysis show that this configuration can separately withstand 227,000 pounds in tension, 103,000 in-lbs in bending and 97,000 in-lbs in torsion.
- 6. Conclusions and Recommendations Based on the assumption that the recently reported data is typical of the conditions in the TMI-2 reactor vessel lower head, it appears that (1) the incore nozzle weld temperatu re ' di d not exceed 1200 degrees F; (2) there were no detrimental environmental effects; and (3) the w9ld metal maintained its virgin strength.
51-1165539-00 Page 10 As a factor of safety it is recommended that a load of no more than one-half of the calculated strength shown in Section 5, be applied to a nozzle.
- 7. References
- 1. B&W Document 77-1149209-00, Evaluation of the Structural Integrity of the TMI-2 Reactor Vessel Lower Head, dated J une,1985.
- 5. B&W Doc ume nt 86-1137540-00, Chemi stry Soecification Evaluation for Reactor Coolant System Cleanuo, dated April, 1982.
- 6. B&W Document 51-1159985-00, TMI-2 oH Reduction Study, dated Februarf, 1986.
- 7. B&W Document TRG-79-11, TMI-2 Recoverv Proiect BAW Water Chemistry Manual, dated October, 1979.
- 8. B&W Calculation 32-1165525-00, TMI-2 Incore Nozzle Strenoth, dated August 8, 1986.
i l
l l
BWWP 20397 (&85)
DOCUMENT
SUMMARY
SHEET O Babcock &Wilcox a McDermott company DOCUMENT IDENTIFIER 32-1165525-00 TITLr TMI 2 Incore Nozzle Strength PREPARED BY: REVIEWED BY:
NAME_Gul Vames NAMF T. W. Elvin SIGNATURE
_ SIGNATURE N 4 TITLE Sy ngr. DATE 8-25-86 TITLE DATE TM STATEMENT:
COST CENTER 310 3 REF. PAGE(S) REVIEWER INDEPENDENCE PURPOSE AND
SUMMARY
OF RESULTS:
The purpose of this document is to determine the strength of an incore nozzle for various wall thickness.
The results are shown on Page 4.
l THE FOLLOWING COMPUTER CODES HAVE BEEN USED IN THIS DOCUMENT:
CODE / VERSION / REV CODE / VERSION / REV NONE PAGF OF
1 PDS-21036-3 (9 84)
Babcock & WIlcox GENERAL CALCULATIONS a McDermott company Nuclear Power Division Doc. i.D. 32-1165525-00
- 1. Introduction The purpose of this document is to determine the load carrying capability of the TMI-2 Incore Instrument Nozzles in the lower head of the reactor vessel.
1 The loads requi red to sepa rately fail a nozzle in tension, bending, or torsion are calculated assuming virgin material properties. Loads are calculated for various tube thickness.
- 2. Method of Analysis The geometry and material properties used in Reference 1 are used in this analysis. The virgin nozzle is considered as a twelve inch long tube with a two inch OD and a 5/8 inch ID. Strengths are calculated for various wall thicknesses up to that of the virgin tube. The material is Inconel 600 with a minimum ultimate strength of 80,000 psi from room temperature up to 1000 0 F.
The moduli of rupture in bending and torsion from Reference 2 are used in determining the ultimate strengths. It is assumed that the ID remains constant and that the wall is thinned from the outside since this gives the lowest strength for a given thickness.
t
- 3. Analysis The strengths of a nozzle separately in tension, bending, or torsion are calculated in Table 1. The results are shown graphically in Figure 1.
The moduli of rupture in bending and torsion, MR b and MRt , were taken from Figures 2.7.1.1 and 2.7.3.2 (b) respectively of Reference 2. These values are for alloy steel and therefore are conservative for Inconel which
~
PREPARED BY . DATE REVIEWED gy DATE PAGE No.
PDS 21036 3 (9 84) i Babcock & WHcox a McDermott Company GENERAL CALCULATIONS i Nuclear Power Division Doc.i0. 32-1165525-00 exhibits more strain hardening than alloy steel. The moduli are calculated as shown in the following example.
@ t = .125 in. , Davg/t = 6.00 Fb = 129,000 psi (elastic failure bending stress for 90,000 psi steel)
MRb = 129,000/90,000 = 1.43 The modulus of rupture is then multiplied by the 80,000 psi tensile strength of Inconel 600 to determine the equivalent strength to be used in ,
i the elastic equations. '
P = (S u)(A) Axial M = (MR b }(Su )(I)/(OD/2) Bending T = (MR t IISu )(J)/(0D/2) Torsion
- 4. References
- 1. B&W Calculation 32-1157102-01, Max. Incore Nozzle Loads, dated 5-30-85.
- 2. MIL - HDBK - 58, Metallic Materials and Elements for Aeroscace Vehicle l Structures, dated 9-1-71.
l l
l t
PREPARED BY DATE REVIEWED BY D ATE PAGE No.
8WNP-20667 (6-84)
Babcock & Wilcox a naoowmott company 32-1165525-00 a ..o.
t-in. ID in. OD-in. Davy-in. A-in+2 I-in+4 J-in*4
+ m +,++++w+++,wn+++++,+een >>++o ,+*+nn*++wnew++++ m wn,+,n+
0.0625 O.625 O.750 0.68/5 O.13499 O.00804 O.01608 O.1250 0.625 O.875 O.7500 O.29452 0.02128 0.04257 0.1875 0.625 1.000 0.8125 0.47860 0.04160 0.08319 O.2500 O.625 1.125 O.8750 O.68722 O.07114 O.14228 O.3125 O.625 1.250 O.9375 O.92039 O.11235 O.22470 0.J750 0.625 1.375 1.0000 1.17810 0.16797 O.33594 9.43/5 O.625 1.500 1.0625 1.46035 O.24101 0.48203 0.5000 0.625 1.625 1.1250 1.76715 O.334'79 O.66958 0.0625 O.625 1.750 1.1875 2.09849 0.45290 O.90579 O.6250 O.625 1.8/5 1.2500 2.45437 O.59921 1.19842 O.6875 O.625 2.000 1.3125 2.83480 0.77791 1.55582 t-in. Davg/t L/Davg MRb MRt P-lbs M-in lbs T-in lbs ou+ooo m+o uwummm*m* mumumnemommmm 0.0625 11.00 17.45 1.29 0.58 10799 2213 1990 0.1250 6.00 16.00 1.43 0.63 23562 5565 4904 0.1875 4.33 14.77 1.52 0.66 38288 10116 8785 O.2500 3.50 13.71 1.58 0.76 54978 15986 15379 O.3125 3.00 12.80 1.59 0.77 73631 22866 22147 0.3750 2.67 12.00 1.60 0.78 94248 31273 30491 0.4375 2.43 11.29 1.61 0.78 116828 41390 40105 0.5000 2.25 10.67 1.62 0.78 141372 53402 51424 O.5625 2.11 10.11 1.63 O.78 167879 67494 64596 0.6250 2.00 9.60 1.64 O.78 196350 83858 79767 0.6875 1.91 9.14 1.65 0.78 226784 102684 97083 NOIE: (1) Su = 80,000 psi (2) L= 12 inches Table 1 PPEPARED BY DATE h* 2 (" h [
REVIEWED BY N DATE PAGE NO
8WMP-20667 (6-84)
Babcock & Wilcox
. uso.cmoet company Doc.tD.
Figure 1
- x. N l
ix l
.h1 e f_ _ . .. J' . '.X 4__ ___ .. _ _...[ . __
s_. _ . . .]
e g
! % sii ; ; !
i i
! O:
i v, l I 1;
! 'N I i l l \h l ?
_._ M__. . i . .. . { _.1_ p _l_.__ . 1 0 '
n ;
, e't ;
Z .E N s ! i LiJ N . h2 l I k j
(M .s
- y a; L1J e \ i
__1 n N 8 N9 !E N .a 53;
- ~- - -
r--
O -
FT Z ,8 g*
Lij 8' \
5+
Eg y
O* .
O . l Z
=
g qp
._. . , . _ 4._ _ ___ ._ __ fj l '
T
, 4 l
l c O
O O O O O O O O O O O S O = m + m O O e # n n N n - - - - -
(epuosnoq1) 1N3 WOW MO 3OMO.d PREPARED BY DATE [* 2-T - IS a
REVIEWED By W 4 CATE PAGE NO.