ML20150A783
| ML20150A783 | |
| Person / Time | |
|---|---|
| Site: | University of Missouri-Columbia |
| Issue date: | 03/11/1988 |
| From: | Alger D, Mckibben J MISSOURI, UNIV. OF, COLUMBIA, MO |
| To: | Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20150A786 | List: |
| References | |
| NUDOCS 8803160038 | |
| Download: ML20150A783 (36) | |
Text
Research Reactor Facility UNIVERSITY OF MISSOURI g,3,3,cn p,7g Columbia, Mssoun 65211 Telephone (314) 882 4211 March 11, 1986 Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Attention: Director Standardization and Non-Power Reactor Project Directorate
REFERENCE:
Docket 50-186; License R-103 University of Missouri
SUBJECT:
Response to NRC Request for Additional Information
Dear Sir:
We have completed our response to the questions in your letter dated January 11, 1988 concerning our application dated September 12, 1986 as supplemented in a letter dated September 11, 1987 for an amendment to our R-103 operating license. This amendment would authorize the use in our reactor of a newly developed extended life aluminide fuel (ELAF) element containing higher densities of uranium and a burnable poison.
The six questions and our answers are attached. If you have any questions concerning our responses, please contact me at (314)-882-5204, or Dr. Soon Sam Kim at (314)-882-5246.
' I k kNPp 8803160038 880311 PDR ADOCK 0500 6 ace p.4/ 4 na e COLUMBIA KANSAS CITY ROLLA ST. LOUIS M a, eavai enee..ro f mst tm, g g
Office of Nuclear Reactor Regulation March 11, 1988 Page 2
. Respectfully submitted, h
J. C. McKibben Reactor Manager Reviewed and Approved:
Xh\)3 Don M. Alger Associate Director xc w/ attachment
& enclosure: A. Adams R. Ambrosek K. Brown R. Carter C. Cooper STATE OF MISSOURI COUNTY OF BOONE )
On this li th day of March,1988, personally appeared before me, J. C. McKibben and D. M. Alger, known to me to be the persons who executed the foregoing document.
Subscribed and sworn to before me, Sharon Vandlandingham, a notary public.
My commission expires:.2 M f/
1 J'i v -Notary P W ic b$$!!3$@i% 2 2l.py
~ - .
QUESTION NUMBER 1 In your answer to question 6, dated September 11, 1987 and elsewhere, you refer to a change in the value of the "Shutdown Margin" in your Technical Specifications. . However, no justification was provided for decreasing this margin below the current value in your technical specification. If you want to pursue this action at this time, please provide a specific request for changing -
specification 3.le to a value below the one in your current Technical Specifi-cations. Your request should include a technical justification (safety analysis) bearing in mind such other reactivity-related specifications as 3.19, h, and k.
ANSWER The request to change Technical Specification (T. S.) 3.le is withdrawn at this time. Meeting the current T. S. 3.1.e requirement will be verified by
.l measuring control rod worths and cold clean critical rod height.
A 6
i I
QUESTION NUMBER 2 In your answer to questions 2 and 10, dated September 11, 1987, Tables 10.1 through 10.5 indicate that plate number 7 has the highest power density with the new fuel. However Table 10.6 indicates that plate number 4 is being compared for compliance with the safety limits. Please clarify this apparent inconsistency.
ANSWER:
The "apparent inconsistency" concerns case 4 (F1270+01270) in Table 10.6 in our letter dated September 11, 1987 (Reference 1). This case is a mixed core of four fresh (zero MWD) 1270 gm elements and four depleted (300 MWD) 1270 gm elements with the control rods withdrawn 16 inches from full-in. The peaking factors given were obtained from two different BOLD VENTURE computer models. The two models are 1 a two dimensional 0-R and a three dimensional 0-R-Z. The 0-R BOLD VENTURE model is used to determine the effect of circumferential heterogeneities (the water gap between the side plates, the aluminum side plates, and the homogenized nonfueled fuel plate edge with the coolant channel) on the azimuthal peaking factor within an element. Inclusion of these heterogeneities in 0-R-Z model would greatly increase the computing time and cost. Therefore the 0-R-Z model homogenizes these circum-ferential heterogeneities into the fuel cell.
The 0-R model provides an economical means of predicting azimuthal power distribution within a fuel element but cannot handle overall three dimensional neutron behavior of the MURR core with control rods partially inserted. It should be noted that the 0-R model represents a slice of the core so that hottest channel must be defined f rom the 0-R-Z model. Therefore we use 0-R-Z model to determine the power distribution for the various fuel combinations and control rod heights.
From 0-R-Z output, the hottest plate is defined as the one which contains the hot stripe with the highest average power density. This is determined by summing up all the power along the Z-direction in the hot stripe and dividing by the total
QUESTION NUMBER 2 (cont'd) volume of the stripe. The axial peaking factor and distribution is obtained by dividing the power density of each axial mesh interval in the hot stripe by the average hot stripe power density. Our 0-R-Z result in Table 10.6 shows plate 4 (not plate 7 shown in 0-R results) as the hottest plate for the case 4 The radial peaking factor for plate 4 is 1.416 and includes azimuthal peak-ing due to differences in fuel loading between fresh and depleted fuel elements as was obtained from the three dimensional 0-R-Z model. This was noted in Table 10.6 of Reference 1 as footnote "c". However, as described above, power densities in the 0-R-Z model are still based on the homogenized fuel cell and do not include the circumferential heterogeneities. Therefore, the azimuthal peaking f actor within an element is obtained from the 0-R model. This azimuthal peaking f actor from the 0-R model also includes the azimuthal peaking due to differences in fuel loadings between fuel elements if the 0-R model includes different fuel elements . This c6n be noted by comparing the azimuthal peaking given in Tables 10.2 (8 fresh 1270 elements) and 10.5 (4 fresh /4 depleted 1270 elements) in Reference 1.
To be conservative, in Table 10.6 of Reference 1 the "radial" (1.416) and axial (1.347) peaking factors are for the hot stripe in plate 4 from the 0-R-Z model and are combined with the azimuthal peaking factor (1.33) for plate 7 ("hot spot" plate) from the 0-R model. Both models were the "worst case" 1270 element cores (4 fresh /4 depleted). Combining the nuclear peaking factors this way creates a fictitious hot spot greater than any actual hot spot in the core.
REFERENCE
- 1. Response to NRC Request for Additional Information Regarding the Use of a Newly Developed Extended Life Aluminide Fuel Element, University of Missouri Research Reactor Facility (September 11, 1987).
m
QUESTION NUMBER 3 In your answer to question 7, dated September 11, 1987, Table 7.2 presents values for the highest plate (power density)* in 1.27 kg elements if only fresh elements are in use. Please provide the computed maximum plate (power density)*
for various possible mixtures of some fresh and some nearly depleted 1.27 kg elements, and compare these (power densities)* with values in Table 7.2 and Table 10.5.
ANSWER:
Our 0-R BOLD YENTURE analysis showed that the core mixture of four fresh (0 MWD) 1.27 kg elements with four depleted (300 MWD) 1.27 kg elements has the maximum plate power density of the various possible cores containing all 1.27 kg elements. Using the same methodology described in our answers to questions 2, 7, and 10, in our letter dated September 11, 1987 (Reference 1), the 0-R model power densities of the fresh and depleted 1.27 kg elements were obtained for the same three conditions given in Table 7.2 in Reference 1. The results are given in Tables 3.1 through 3.6.
The three conditions are the two extreme flux trap loadings corresponding to the Technical Specifications reactivity limits (+ 0.006 tK) and the reference case where only water is in the flux trap region (0.00 tK by definition). It should be noted that Table 3.1 is for the same case and condition as Table 10.5 in Reference
- 1. Tables 3.1 through 3.6 show the "hot stripe" is always in the fresh element as would be expected. Table 3.7 summarizes the hot spot maximum power densities and average power data in some plates of interest in the fresh fuel element, and Table 7.2 is reprinted for comparison purposes.
- The question in the NRC letter dated January 11, 1988 inadvertently uses the term "temperatures", but Tables 7.2 and 10.5 are in units of power density (watts /cm 3).
I
QUESTION NUMBER 3 The results in Table 3.7 are consistent with what would be expected from the values in Tables 7.2, 10.2, and 10.5 from Reference 1. The maximum power density in either core (8 fresh or 4 fresh /4 depleted) is for the referenced condition (no flux trap in the reactor) and is on plate 7. The 0-R oistributions for the two cores in this condition were given as Table 10.2 and 10.5 in Reference
- 1. The effect of having the two extreme flux trap conditions installed in the "worst case" 1.27 kg element core are what would be predicted from the results in Table 7.2. The power densities decrease slightly on the fuel plates (4 and 7) nearest the flux trap position and increase slightly on the plates furthest from the flux trap.
Therefore the peaking factors for the flux trap installed at Technical Specification limits (+ 0.006 oK) are covered by the envelope of the worst case peaking factors derived in the answer to questions 2 and 10 in Reference 1 and clarified in the previous answer to question 2 in this letter.
REFERENCE
- 1. Response to NRC Request for Additional Information Regarding the Use of a Newly Developed Extended Life Aluminide Fuel Element, University of Missouri Research Reactor Facility (September 11, 1987).
l t
i
, ~2-
Tabic 3.1 !
. Power Distribution (watts /cm 3) in a Fresh 1270 Fuel Element when loaded l i
with Depleted 1270 obtained from a 45' 0-R MURR Model Reference Case - water only in flux trap region.
Azimuthal Power Densities for mesh intervals of 0.0672 radians !
Chr.nnel from the outside edge of fuel meat to fuel plate centerline ,
Max-Avg. x 100 i Number 1 2 3 4 5 Avg. "9 '
l 1 392.2 393.8 391.5 389.9 389.2 391.3 0.6 J
2 408.9 393.3 385.3 381.6 380.2 389.9 4.8 3 422.1 383.4 367.6 361.3 359.0 378.7 11.5
- 4 512.3 438.7 411.1 400.7 397.3 432.0 18.6 '
5 527.7 428.9 393.6 381.0 376.9 421.6 25.2 t 6 573.8 448.1 405.2 390.2 385.6 440.6 30.2 7 595.2 454.1 406.9 389.4 384,9 446.1 33.4 8 563.9 421.3 375.9 361.3 356.9 415.9 35.6 j 9 543.2 402.8 359.1 345.5 341.6 398.4 36.3 10 528.9 390.8 349.1 336.4 332.8 387.6 36.4 !
11 518.6 382.6 342.7 330.9 327.8 380.5 36.3 i 12 511.6 376.9 338.6 327.7 324.8 375.9 36.1 13 506.9 372.7 335.9 325.8 323.2 372.9 35.9 14 504.5 370.0 334.5 325.1 322.8 371.4 35.8 15 504.1 368.8 334.5 325.8 323.7 371.4 35.7 I
16 505.9 369.7 336.6 328.6 326.7 373.5 35.4 17 511.1 374.3 342.6 335.2 333.5 379.3 34.7 l t
l 18 521.5 385.3 355.2 348.5 347.0 391.5 33.4 l 19 541.8 408.5 380.3 374.2 372.9 415.5 30.4 ,
i 20 551.5 430.5 405.8 400.6 399.5 437.6 26.0
- !1 534.6 438.8 419.4 415.3 414.4 444.5 20.3 j 22 516.4 449.8 436.0 432.9 432.1 453.4 13.9 l
23 507.5 468.5 459.6 457.2 456.5 469.9 8.0 I i l
! 24 483.4 466.0 461.1 459.3 458.6 465.5 3.6 i
e '
---,e-m.,_,-w,,.,--.-,m.- r- wre-**e*-v--URR. Since the FURR fuel channel length ('24") is about one half that of ATR, the use of ATR test results can, in fact, provide consenatism for M1RR because investigators (10) have shown higher or equal burnout heat flux levels for shorter channel lenyth.
Similarly, the shorter channel lengths are less susceptible to the hydraulic instabilities related to incipient bulk boiling.
Other test reactors (l{ FIR, EE) have design and operating conditions that depart further from the FURR conditions, and their test results were not directly useful in developing the M1RR safety criteria.
Preliminary ATR testing (8) indicated that both subcooled boiling bumout and bulk boiling burnout can occur for the range of channel thiebesses then tmder design consideration. Tests were performed at Argonne in 1963 on three chamel thicknesses (0.054", 0.072", 0.094"), and it was found that
for the two thinnest channels (0.054", 0.072") the burnouts were due to hydraulic instability (or autocatalytic vapor binding) when the coolant reached saturation at the channel exit. Presurably, the hylraulic instabilities led to subnormal flow conditions and a lower bumout hea t flux. Subcooled burn-out occurred for the 0.094 inch channel before the coolant reached saturation conditions at channel exit. The subcooled burnout heat flux data obtained in these tests were 0.6 of the burnout heat flux predeted by the Bernath correlatica (7):
'INB b bo *I sat + 6Tsub) where:
D, t hg = 10890. f . 4g D, + Dg j D[.6 Tg = 1.8 57 inP - 54 ;
5
+ 32.
T saturation tenperature at P, *F t
6T g = bulk water tcr:perature, degrees subcooling, *F D, = wetted hydraulic diameter, ft Dg = heated hydraulic diameter, f t V = coolant velocity, fps P = systen pressure, psia Subsequent full-scale ATR testing (9) at Battelle Northwest with a channel thic) mass of 0.070" confimed the earlier test results; namely, that burnout induced by hydraulic instability was the limiting factor for ATR.
In addition, it was established that the hydraulic instakility condition did not correspond to initiation of local boiling, but to the beginning of tulk boiling at the channel exit in the region where the coolant enthalpy was highest. Test results also indicated that lateral mixing (in the channel) was qinte small.
In view of the ATR experience, and in absence of burnout test results for M1RR fuel and at MJRR operating corditions, the following safety limit criteria were adopted for this study:
The coolant exit terperature from the hot channel shall be less than the saturation tenperature at the core exit pressure The local heat flux at any point in the core shall be less than 0.5 of the burnout heat flux as given by the Bernath correlation l st that point.
QUESTIONS 4 AND 5 (cont'd)
The bulk boi1*.ng limitaticn is adopted to exclude occurrence of the in-core hydraulic instabilities related to incipient bulk boiling. The above burnout heat flux limitation is adopted to provide some additional de-sign safety margin by a reduction of the correlated ATR test data by the factor 0.5/0.6 relative to the orginal Bernath correlation. The above criteria are sufficient to preclude the possibility of fuel failure and attendent fission produ:t release due to excessive terperatures.
To "benchmark" the COBRA-3C/MURR to the MURR safety limits, six sets of independent variables (coolant flow, inlet temperature, and pressurizer pressure) were selected and run in a COBRA model corresponding to the 6.2 kg core model run in the BOLERO code to establish the safety limits. Table 4.1 shows a summary of hot channel factors used in this current safety limit analysis. Since the COBRA l input format is different from the BOLERO input, a slight modification was necessary. For the flow related factors, the overall product was calculated to be 1.00 x 1.00 x 1.00 x 1/1.08 x 1/1.05. The channel factor for the effective flow area is taken care of separately in the COBRA input.
BOLERO differentiates between hot channel factors on enthalpy rise and those on heat flux. The overall product on enthalpy rise is 2.72, where the overall l
product on heat flux is 4.35. The difference is the axial nuclear peaking factor and the greater effect fuel thickness / width variation has on local heat flux l (1.15) compared to enthalpy rise (1.03). In the NUS study, BOLERO was used to get l - . _ . . ..
Table 4.1 Summary of MURR Hot Channel Factors On En thalpy Ri se. . . . . . . . . . . . . . . Safety New Limits Fuel Basis Design Power-related Factors Nuclear Peaking Factors Radia1........................................ 2.220 .
1.416 N on-u ni f orm Burn u p . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.112 J Local (Ci rcumf erential ) . . . . . . . . . . . . . . . . . . . . . . . 1.040 1.330 Axia1......................................... 1.000 1.000 Engineering Hot Channel Factors on Enthalpy Rise Fuel Content Variation........................ 1.030 1.030 Fuel Thi ckness/Wi d th Va ri ati on. . . . . . . . . . . . . . . . 1.030 1.040 Overall Product.................................. 2.72 2.02 Flow-Related Factors Core / Loop Fl ow Fracti on. . . . . . . . . . . . . . . . . . . . . . . 1.000 1.000 Assembly Minimum / Average Flow Fraction. . . . . . . . 1.000 1.000 Channel Minimum / Average flow Fraction I n l e t Y e r i .9 ti o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. 0 0 0 1.000 Wi d th Va ri a ti on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. 000 1.000 Thi ck ne s s Va ri a ti on. . . . . . . . . . . . . . . . . . . . . . 1. /1.080 m i
Within Channel Minimum / Average Flow Fraction ?1/1.25 1 Thi cknes s Vari a ti on. . . . . . . . . . . . . . . . . . . . . . 1./1.050 W E f fecti ve Fl ow Area . . . . . . . . . . . . . . . . . . 0.3231/0.3505 0.3231/0.3505 l O v e r a l l P r o d u c t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18. ;81 0.74 i
O n Hea t F1 ux. . . . . . . . . . . . . . . . . . .
l Power-Related Factors Nuclear Peaking Factors Radia1........................................ 2.220 m e 1.416 Non-uniform Burnup............................ 1.112 s Local (Ci rcumf erential) . . . . . . . . . . . . . . . . . . . . . . . 1.040 1.330 Axia1......................................... 1.432 1.347 Engineering Hot Channel Factors on Flux Fuel Content Vari ati on. . . . . . . . . . . . . . . . . . . . . . . . 1.030 1.030 Fuel Thi ckne ss/ Width Variati on. . . . . . . . . . . . . . . . 1.150 1.250 Overall Product..................................... 4.35 3.27 Energy Fraction Generated in Fuel Plate. . . . ..... 0.930 0.930 l
l 1
l 1
l
QUESTIONS 4 AND 5 (cont'd) the power level corresponding to a DNBR = 2.0 = Bernath critical heat flux divided by the "BOLER0" hot spot heat flux. For COBRA the hot channel factor of 2.72 and the axial power distribution can be given in the input, so the heat flux then has the fuel thickness / width variation engineering factor of 1.03 for the hot channel.
The local hot spot engineering factor of 1.15 can not be input to COBRA. There-fore, the "BOLER0" hot spot heat flux = (1.15/1.03) "COBRA" hot spot heat flux.
For COBRA, the local hot spot engineering factor for fuel thickness / width variation can be taken into consideration by changing the maximum acceptable DNBR:
BOLERO DNBR =
Critical Heat Flux = 2.0 BOLERO Hot Spot COBRA DNBR = Critical Heat Flux , Critical Heat Flux COBRA Hot Spot BOLERO Hot Spot therefore:
=
COBRA DNBR BOLERO DNBR = 2.23 i
l I
1 QUESTIONS 4 AND 5 (cont'd)
The "worst case" peaking factors for the new 1.27 kg fuel element are also given in Table 4.1. The engineering and flow-rate factors are justified in a previous section of this answer. The nuclear peaking factors for the 1.27 kg element have been previously discussed in Reference 1, and in the answers to Questions 2 and 3.
Table 4.2 provides a comparison of BOLERO results to COBRA results for the same safety limit basis case (6.2 kg core) and the "worst case" 1.27 kg element core. These values show that applying the current maximum allowable core power levels in MURR Technical Specification 2.1 to the "worst case" all 1.27 kg element core does not decrease the margin of safety. This is true even with the absurdly conservative flow-related factor.
Table 4.2 Comparison of )LER0 and COBRA Maximum Allowable Core Power Level (MW)
Inlet Water Conditions BOLERO COBRA Pressurizer Flow Rate Temperature Safety Limit Safety Limit 1.27 Kg Pressure PSIA GPM Degree F Basis Basis Fuel Design 60 3200 160 12.67 13.48 17.82 75 3200 120 17.39 17.84 23.48 75 3200 160 14.53 15.02 19.84 75 4000 120 19.44 21.40 28.18 75 4000 160 16.14 17.94 23.64 85 3200 160 15.49 15.88 20.98 l
l l
QUESTIONS 4 AND 5 (cont'd)
REFERENCES
- 1. M. M. El-Wakil, Nuclear Heat Transport, International Textbook Co. (1971).
- 2. BOLER0-III Code Manual, NUS-TM-ENG-119, HUS Corporation.
- 3. J. Chao, "COBRA-3C/RERTR, A Thermal-Hydraulic Subchannel Code with Low Pressure Capabilities," Science Applications, Inc. (December 1980).
- 4. R. W. Bowring, et al. , "COBRA-3C/MIT, Code Manual," Massachusetts Institute of Technology (March 1976).
QUESTION NUMBER 6 In your answer to question 5, dated September 11, 1987, it is reported that a reactivity transient was analyzed with PARET/ANL with increased oxide on the fuel plates. What critical heat flux correlation is used in PARET/ANL and has its applicability to MURR been ascertained?
ANSWER:
The PARET/AHL(1,2) program provides three Critical Heat Flux (CHF) cor-relations to the user. These are the correlation in the original PARET(3)
(Westinghouse correlation),(4) Bernath, and Mirshak correlations. We have been using the original CHF correlation in the code because it gives the most con-servative result (worst result regarding the fuel centerline temperature and peak power) among the three correlations. There are no burnout tests for MURR fuel; but fortunately, the MURR fuel element geometry is similar to the Advanced Test Reactor (ATR) fuel element so that ATR experience can be applied to MURR. In addition, extensive analysis on PARET/ANL performed by AHL provides useful infor-mation on the selection of the CHF correlation. The following discussion provides the basis for our selection.
In 1963, the ATR fuel tests were performed at Argonne on three channel thicknesses (0.054", 0.072", 0.094"), and it was found(6) that for the two thinnest channels (0.054", 0.072") the burnouts were due to hydraulic instability when the coolant reached saturation at the channel exit. Presumably, the hydrau-lic instabilities led to subnormal flow conditions and a lower burnout flux. Sub-cooled burnout occurred for the 0.094" channel before the coolant reached the sat-uration conditions at core exit. The subcooled burnout heat flux data obtained from these tests were 0.6 of the burnout CHF predicted by the Bernath correlation. For this kind of reactivity transient, the dominant negative feed-back is the negative void reactivity from the void / density component change due to boiling. The use of the Bernath correlation in PARE.T/ANL will predict a 1.67
QUESTION NUMBER 6 (cont'd) times higher CHF which results in later negative void feedback giving conservative results. However, we do not use the Bernath correlation because the original cor-relation in PARET or PARET/ANL gives even more conservative estimates on fuel and cladding temperatures, and peak power than the Bernath correlation.
In 1982, AHL selected PARET as a standard tool for research reactor transient analysis as part of the Reduced Enrichment Research and Test Reactor (RERTR) project. They modified the original PARET code extensively to include various models and correlations considered more applicable to the low pressures temperatures and flow rates encountered in research reactors. Benchmark calcula-tions performed by AHL against the SPERT I core (highly enriched uranium, plate type fuel) showed that the PARET/ANL results generally agree well with the experimental values. Their PARET/AHL study (2) on the SPERT analysis recommended to use the original CHF correlation in PARET along with Rosenthal and Miller correlation for the single phase heat transfer coefficient, and the McAdams correlation for fully developed two phase flow with the Bergies ana Rohsenow correlation for the onset of nuclear boiling and transition to fully developed nuclear boiling - to give the most realistic estimates for a research reactor transient calculation. The MURR analysis currently uses these options.
Based on the above, our MURR analysis uses the original CHF correlation in PARET/ANL.
QUESTION NUMBER 6 (cont'd)
REFERENCES
- 1. W. L. Woodruf f, "A Users Guide for the Current AHL Version of the PARET Code,"
Argonne National Laboratories (1982).
- 2. W. L. Woodruff, "The PARET Code and the Analysis of the SPERT-I Transients,"
_P r_oceedings of the International Meeting on Research and Test Reactor Core Conversions f rom HEU to LEU Fuels, ANL/RERTR/TM-4 (September 1983).
- 3. C. F. Obenchain, "PARET-A Program for the Analysis of Reactor Transient,"
100-17282, AEC Research and Development Report (January 1969). ,
- 5. F. R. Vaughan, "Safety Limit Analysis for the MURR Facility," NUS-TM-EC-9, nus Corporation (1973).
i
-g $'
i i
l I
SPECIFICATION TRTR-4 l FOR UNIVERSITY OF MISSOURI !
FUEL ELEMENTS ASSEMBLED FOR UNIVERSITY OF MISSOURI RESEARCH REACTOR 3 AUG 1987 REVIEWS MECHANICAL ENGINEERING PRP TECHNICAL SUPPORT QUALITY ASSURANCE, SOUTH BRANCH FUSION SAFETY PROGRAM THERMAL HYOR AULIC OESIGN ANO UNIVERSITY OF MISSOURI ANALYSIS l
APPROVAL TRTSB MANAGER E G 8 G IDAHO, INC.
P O BOX 1625 IDAHO FALLS, IDAHO, 83415 t
Reviews and approval f or this revistore documented on DRR 27300 dated 3 August 1987
._. . _ . .- - ..__ . . - _ _ --