ML20141A362
| ML20141A362 | |
| Person / Time | |
|---|---|
| Site: | 05000231 |
| Issue date: | 10/16/1971 |
| From: | GENERAL ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20140G249 | List:
|
| References | |
| FOIA-97-34 NUDOCS 9705140258 | |
| Download: ML20141A362 (17) | |
Text
_ . _ . _ _ . _ _ . . _ _ _ _ _ . . _ _ _ _ _ . . _ . . _ . _ _ _ . . _ . . . . . . . - ._ ..._ _ __.- .. .._. . _ _ ___ .. . . . .
M F 3 .
i e .
h
- t
,' L*Kulag gy, 8 %
i PROPOSED CHANGE NO. 8 FOR THE SOUTHWEST EXPERIMENTAL FAST OXIDE REACTOR OCTOBER 16, 1971
(
naa wua ou 1044 7/
Re: LICENSE DR-15 DOCKET 50-231 1
i i
k 8
I l
l I
t l l
t l
l l
I GENERAL ELECTRIC COMPANY i
310 DeGuigne Drive
{ ,
f Sunnyvale, California 94086 3 i
l- .
l l
~
9705140258 970505 PDR FOIA i PDR ,
VARA.DY97 , l
l l
f I
Proposed Change No. 8 l t
. for the 4
Southwes.t Experimental Fast Oxide Reactor t
I. Introduction Under the authority of License DR-15, General Electric operates l
the Southwest Experimental Fast Oxide Reactor at a site near Strickler, Arkansas.
I A revision of the current Technical Specifications is desired as
! described herein. The applicable revised pages of the Technical Spec:1fications are also included as Attachment A.
II. Proposed Changes l
Pursuant to the provisions of 10 CFR 50.59, General Electric requests that the SEFOR Technical Specifications be changed by substituting Pages 3.3-2, 3.3-7, 3.10-2, 3.12-1, 3.12-1.1, 3.12-3, 3.12-3.1, 3.12-5, 4.3-1, and 4.4-2.1 in Attachment A of this doc-ument for the corresponding pages of the current Technical Speci-fications and adding pages 3.3-7.1 and 3.12-6.
Ill. Purpose of the Proposed Change The changes described in this submittal will accomplish the f ollowing purposes:
A. Permit initiation of the Core II transient test program after completing static tests at 10 MW; revise the limit for the max-imum allowable poison slug worth for transient tests, so that a linear interpolation for slug worth as a function of Doppler coef-ficient can be used for Core II transient tests ;and remove the mandatory requirement for DRL approval to initiate the super-prompt tests in Core II.
i B. Reduce the frequency of fuel curveillance and sodium sampling re-quired during the super-prompt critical test program, based on tb experience gained during Core I testing.
C. Clarify the requirements for static and oscillator' tests durinF the Core II tcst program.
.. - - - . . . -. ... . . - - . . - - - . .. - . - . - ~ _ . . - . - -. -,
4 e ..
b
{ ' -
ents for reactor operation l
D.
Remove an inconsistency in the requirem ,
l p with the head removed. r
~
. 1 l
.IV.- Discussion
- ion changes are discussed individual,y .
- The proposed Technical Specificat as follows:
A. Section 3.12. Excursion Tests leting Static of Transient. Test Program after comp 1, Initiation l Tests at 10 MWt. as originally written for Lbe
.The Technical Specifications pproach to reactor y
[ SEFOR program reflected the conservative afor the safe '
i L operations considered prudent The initial power ascension operation of this test facility.
trated that the reactor per-ted.I ' ' The l program and transient tests demons l formance is predictable, stable, and as expecI test progra !
satisfactory completion of the Core d and measured reactor per-the power f
excellent agreement between predicte oefficient, formance has demonstrated that h the Doppler cal temperat i
l coefficient of reactivity, the isot ermtor stability can be accur-of reactivity at zero power, and reac r levels of 10 MWt and ately determined from operation at powe ts at power t
below.
Therefore, the static and oscillator teside a sufficien !
3.3 and levels up to 10 MWt on Core II will provthe requirenen levels up to of data to demonstrate that 312 B.1 have been met. Thesesubprompt i measurementsand suoer- at ;
l 10 MWt will enable reactor response dur ngly predicted w
~
l i i
prompt transient tests to be accurateIf the reactor is to be o ing measurenents to full power. h n 10 MWt,' additional at steady state power levels greater t alled for in Section 3.10 static tests will be perf ormed as ca Worth 7
2.
Adjustment of Permissible Core 11 table Slugpoison slug wort' j
l , The limits specified for the maximum e ecDoppler coeffi 16 The maximwn permis# f l
4 were previously set at 1.30$ for ahf=-0.005.
[ , Tfh=-0.008,and1.20$ fort
/ .
l l l-j i
reactivity insertion rate and initial power level for these tests are 20$/see and 11 FMt. These limits were based on the !
analysis of effects resulting from the maximum planned transi-l dk ent with a Doppler coef ficient, of Tg = -0.004. (4) The follow-Ing table shows the assumed values for this, transient and other !
l transients used to establish the present limits in tt Techni-cal Specifications.
l Table 1 Comparison of Sunernrompt Critical Transients l r
Analyzed and Technical Specification Limits i Design
- Maximum Tech. Spec.
Transient Planned Transients
- Limits Doppler Coefficient (T ) -0.004 -0.0085 -0.004 -0.008 -0.005 Reactivity Inserted ($) 1.5 1.3 1.2 1.3 1.2 l
Reactivity Rate ($/sec) 50 20 20 20 20 1
l Initial Power (Mut) 7 11.5 9 11 11 )
Core Coolant Flow (%) 100 100 100 90 90 l
- Transients Analy;:ed in Ref. (4). t By comparing values in the above table and referring to the discussion on pages 1-42 and 1-43 of Ref. (4), it can be determined that the transients permitted by the Technical Specifications are comparable to the planned transients pre-viously analvzed.
The v.alue of the Doppler coefficient was determined to be
-0.0081 for Core I.( Thus, the maximum permissible poison slug worth was 1.3$. For Core II, the Doppler coefficient will be reduced by approximately 15% to about -0.0069. The I actual value will be determined from the results of the static tests prior to initiation of the transient test program for l Core II. It is desirable from an experimenta) siewpoint to i
r
\ .
l . perform the superprompt critical transients with the maximum
! permissible poison slug worth consistent with safety considera-t'io ns . Therefore, it is proposed that the present limit of 1.2$
for any Doppler coefficient with a magnitude of less than 0.008, he changed to permit a linear' interpolation between the present
' values for the limiting slug worth as a function of Doppler
- coefficient. This would result in a limiting slug worth of 1.26$ for a Doppler coefficient of .0069. This method of l determining the limit will provide equivalent margins with respect to the design transient for any acceptable value of the Doppler coefficient, as discussed below. The observed values of the dynamic slug worth will be used for comparison to the limits as discussed on page 3 of Ref. (3).
Figure 1 compares the proposed limits for slug worth to the i present limits.
1.3- f g j Proposed Limit -
BC I l l
,8 .c l Wt l Present
$j Limits 3 et In 3.12 xv i 3 ~* J. I 1
- 1.2 J l
l l l
-0.005 -0.008 SODItat-IN DOPPLER COEFFICIENT (T )
Figure 1. Limits for Poison Slug Worths l
f l
i ,
1 l
The calculated val'ues of power and increase of the average i
fuel temperature are given in' Figures 2 and 3 for transients I g with Doppler coef ficient values of -0.005, -0.006, -0.007, and
, ,-0.008, dnd with reactivity insertion values corresponding to i the proposed limit. l l
From these figures it can be seen that the proposed limits for intermediate values of the Doppler coefficient provide an !
envelope for transient tests which is consistent with the I present limits for Doppler coefficients of -0.005 and -0.008.
It should be noted that the values obtained from Figures 2 and 3 represent the maximum values since they are based on the maximum initial power level consistent with the core loading limit of .50$ and on start of reflector movenent (scram) at 0.600 seconds. Thus the proposed limits provide margins with respect to the design transient which are equivalent to the I margins provided by the present Technical Specifications for Doppler coefficients of -0.005 and -0.008.
This conclusion is further supported by the excellent agree-ment between measured and calculated values of reactor power for the superpronpt transient tests with Core I.
The data for superprompt transient No. 2 are plotted in Figure 4.
Non-Doppler feedback coefficients contribute about 12 percent of l the total feedback during the transient. This comparison pro-vides assurance that the results of transient tests for Core II can be accurately predicted by use of the available analytical tools and test data.
- Analysis of the superprompt transients for Core I has provided further verification that the value of the Doppler coefficient l s.
isTfy--0.0081,asindicatedbythecomparisonofvaluesin Tables 2, 3, 4 and 5.
Eight superprompt critical transients were performed at three different initial power levels using reactivity insertion up to 1.28$. Initial and peak power icvels and energy release at
< 0.3 see (iust prior to scram) are compared with predicted values l
l in Table 2 i !
1 I
r The energy release'during transients can he determined from the f' lux measurements of the two U-238 chambers (combined in two
. instances - Test 1 & 2 - with gamma chambers), and from two energy probes, which are essentially calorimetric devices. A comparison of the energy determined from these four instrumen'ts is shown in Table 3.
To obtain an initial estimate of the Doppler effect during the superprompt transients, only the data during the time after (tA 140 msec) the FRED reactivity insertion has ceased, but prior to the time (t 2: 300 msec) of reactor scram, was analyzed.
During this time, the only changes in reactivity are those resulting from feedback. Analysis of the complete power trace requires an accurate knowledge of the time-dependent reactivity characteristics of the FRED, while this procedure does not.
For the appropriate time interval, the observed time-dependent reactivity change, as determined from the kinetics inversion equation, was plotted against the time-dependent energy release 1
obtained from analysis of the fission chanber data.
Over the limited energy range, the data demonstrated essentially a linear dependence of reactivity upon energy release. The slope of this linear relationship, which is the energy coefficient, j as well as the approximate energy interval over which the analy-l sis was performed, is shown in Table 4.
The coefficients in Table 4 were corrected for the calculated non-Doppler (i. e. , the fael and clad axial expansion, sodium and structure effects, etc.) reactivity effect of about 12 per- !
! cent and they were corrected for the fact that only )
l . (0. 932) (0. 94 3) = 0. 88 of the total energy produced is actually absorbed in the fuel during the short time interval under con- l sideration. ( 5) l i
The resultant averaged Doppler energy coefficients, in units of cents per MW-see of energy absorbed in the fuel, are compared in Table 5 with the results obtained from the subprompt tests, and with results calculated for the energy range and initial power level (see Table 4) used for the measurement. As the 1
l I
i
- data of Table 5 indicate, all transient Doppler results show dk good agreement with'a Doppler Tg of -0.0081, which verifies the previously determined value of -0.0081 obtained during the g power ascension tests.
- 3. Removal of the !!andatory Requirement for DRL Approval to Initi'a te Superprompt Critical Tests in Core II The requirement that the prompt critical test program cannot be initiated until DRL completes a review of all test information up through the subprompt critical tests and determines if addf-tional specifications are required was based on a lack of suffic-I ient experimental data prior to Core I testing to support the adequacy of the Technical Specification limits. The successful
~
completion of the Core I experimental program as discussed above has demonstrated the adequacy of these limits and the ability to predict the performance of Core II, which represents only a small extrapolation from Core I. Further, the license requires reporting, of anomalous behavior, which assures prompt reporting of any observed deviations during Core II tests from that pre-dicted. As a result, a mandatory review and approval by DRL should not be necessary prior to initiation of the Core II superprompt tests.
B. Paragraphs 4.3C and 4.40 - Fuel Surveillance jmd Sodium Sample
. Analysis for Sunerprompt Test _s Data obtained from fuel surveillance md edium sample analysis during Core I superprompt trancient tests showed no effects result-ing from the transient tests. In the unlikely event that fuel failure occurred during a transient test, the cover gas monitor and spectral analysis of cover gas samples (required before and after each transient above 90c by Paragraph 4.4P) would provide the most sensitive means of detecting the failure. The fuel rod examination and sodium sample analysis provide back-up information.
l Therefore, it can be concluded that fuel rod examinations and
! sodium sample analysis performed after every other prompt critical transient when combined with on-line cover gas monitor measurements and cover gas spectral analysis performed af ter each prompt criti-cal transient will provide reasonable assurance that Specification 3.12 B.10 will be complied with.
l I r 1 _ __- __. .
1 I
' In addition, the consequences of perf orming a transient test -
with a failed fuel rod which admits sodium have been evaluated (6) and it was determined that such a condition would not lead to
. gross clad rupture.
C. Section 3.10 - Approach to Power .
l The results for conventional oscillator tests performed on Core I have been presented previous 1v in Refs, (1) and (2). These results have demonstrated 'the stability of the reactor. The change to Core II is estimated to reduce the magnitude of the negative Doppler coefficient from the Core I value of .0081 to a value . * .0069 (a change of 15 percent). Analysis has shown that a change in the Doppler coefficient of +20 percent from a value of -0.0085 will yield minimal changes in the SEFOR core transfer function.
For Core I the phase margin at the point where the Nyquist plot i
crosses the unit circle was measured to be between 73 and 83.5*
for power levels from 2 MWt to 17.5 MWt.( ) The estimated change in the Doppler coefficient for Core II will have a minimal effect on the phase nargin determined for Core I, and thus for Core II the phase nargin will be substantially greater than the required mininum of 30*.
D. Specification 3.3M - Inoperable Gross Gamma Cover Gas Monitor The changt to Specification 3.3M and the basis for Section 3.3 will remove an inconsistency which requires that the reactor be shut down if the gross gamma monitor is inoperable (Para. 3.3M) i but permits operation with the reactor head removed (Section 3.9) .
Operation with the reactor head removed results in dilution of the i
reactor cover gas which significantly reduces the effectiveness of l
the gross gamma monitor.
! The proposed change will eliminate this inconsistency by excepting
( operations with the reactor head removed from the requirements of Paragraph 3.3M. Section 3.9 limits the reactor power to less than 500 Kut with the reactor head removed, and the planned tests are per-formed at power levels of 100 KWt or less. At this low power there is no need for on-line monitoring of the reactor cover gas for fission gases since the low fuel and clad temperatures preclude the possibility of anomalous fuel rod performance.
l l
l t
I .
References t
',1. General Electric Breeder Reactor Development Operation, "SEFOR Reactivity l Coef ficients - as Determined f rom Tests Through 10 FN," submitted to the
- g ,
AEC-DRL on December 2,1970, pages 8-10.
- 2. General Electric Breeder Reactor Development Operation, "Results for Core I of Power Ascension and Conventional Oscillator Tests to 20 MWt,"
submitted to the AEC-DRL on June 16, 1971, pages 55-111.
- 3. General Electric Breeder Reactor Department, "Results of the Familiariza-tion and Sub-Prompt Critical Transients for Core 1," submitted to the AEC-DRL on July 16, 1971, page 3.
i 4. SEFOR-FDSAR, Supplement 10, Section 1.3, pp. 1-42 through 1-70.
l
- 5. Op. .it., Ref. 3, page 22.
- 6. General Electric Breeder Reactor Development Operation, " Additional Infor-mation Regarding Sodium Logging of SEFOR Fuel Rods," submitted to AEC-DRL '
l on February 1, 1971.
- 7. Noble, L. D. , and Wilkinson, C. D. , " Final Specifications for the SEFOR ;
Experimental Program," General Electric Advanced Products Operation, January, 1968, (GEAP 5576) , pages 5-67 through 5-73.
- 8. Genero) Electric Breeder Reactor Development Operation, " Revision No. 1 I of Proposed Change No. 4 for the Southwest Experimental Fast 0xide Reactor," ,
submitted to the AEC-DRL on December 11, 1970, pp. 4, 5.
l ,
l l
I I
l
. _n_
i l
t l
l . .
i
- 4 3 2
\V/. .
t l INITIAL
- INSERTION DOPPLER POWER l ,
,j U .005 9.00 MW
^
! 1 1.2000$
li i i n 2 1.2333$ .006 9.67 LN 10.33 IN
, j 3 1.2667$ .007 H 4 1.3000$ .008 11.00 MW 3 '
10 _ 4
~
q
/f2
'f, 1 g
'lil ~
! cl l
r sa 10 2 .
j 1 \ j
! A s
[
/' \
\ j
[
e '\g l l
_l/<, % l
- /
/
~
l l
t ,/
l 10 '
t 0 , , , , , , , _ ,_
i 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
! TIME, (SECONDS)
FIGURE 2: POWER DURING SUPER-PROMPT CRITICAL TRANSIENTS
l l
e t
80n .
700-INITIAL l INSERT 10S ADOP POWER l 1 1.2000$ 005 9.00 W 1.2333$ 006 9. 67 11W 2
60C- .007 10.33 MW 3 1.2667$
C 1.3000$ .008 11.00 MU 4
v 1
rl 2
0 50C.- I 3 '
4 l
y P
e-'
[C .
)
i N .
a ,
- I
$ 400-I I
- l@: 30C-e I u
Ld O
ta !
l
- > l
< i 20C- l l
l' 1
l l
10C- l i
l
' ' l 0 d ' ' '
0.4 0.6 0.8 1.0 0 0.2 TIME, (SECONDS)
FIGURE 3: SEFOR SUPER PROMPT TRANSIENTS - AVERAGE CORE FUEL l
l
. 10,000
[ t l t l .
i CURVES COINCIDE AT Tile PEAR VALUE .
e .
l i
l 1,000 -
X.
~.~' N ' N l
O 100 -
E' i v
M N .
6 ,
l k I
l PRE-CALCULATED CURVE
--- DATA FROM U-238 FISSION l
DETECTOR A
~
INITIAL POWER = 2 MW i
l MAXIMUll SLUG WORTll = $1.28 l
l l
l i i i 1 i
.1 .2 .3 .4 TIME (msec)
.._.,,,-,,,,-n. , , , , . , , ,
TABLE 2 ,
Peak Power Level Energy Release,to.0.3 sec.
Test Initial Power Slug Worth Measured Predicted
- Measure Predicted *
(HW) ($) (HW) (MW-sec) 1 2.0 1.18 5300 5400 82 80 2 2.0 1.28 9000 8800 108 105.-
3 2.0 1.28 8800 8800 106 105 4 5.0 1.18 4900 5300 93 -
95 5 5.0 1.28 8200 8600 120 125 6 8.0 1.18 5100 5300 108 110 7 8.0 1.28 8300 8500 135 140 1 .
f 8 8.0 1.28 8400 8500 136 140
- For a Doppler Coefficient (T ) = -0.0081.
b e
f
TABLE 3- .
COMPARISON OF MEASURED ENERGY RELEASE ,.
FOR SUPER-PROMPI TRANSIENTS ENERGY RELEASE * (MW-sec)
Test SEFOR Energy SEFOR Energy Fission Detector Fission Detector * -
Probe #1 Probe #2 A B 1 43 47 46 43 ,
2 62 65 62 64 3 58 65 60 59
, 4 50 56 49 48 i 5 69 77 66 64 6 62 64 56 56 7 75 84 73 70 8 73 78 74 70
- At 120 msec after lift-off.
TABLE 4 ,
SUPER-PROMPT TRANSIENT ENERGY COEFFICIENTS Measured Test Initial Power Slug Worth Energy Range Energy Coefficient (MW) ($) (IW-sec) (C/MW-sec) 1 2 1.18 50+75 -0.47 - ,
2 2 1.28 70+110 -0.45 3 2 1.28 70+110 -0.45 4 5 1.18 60+90 -0.42 5 5 1.28 75+115 -0.42 6 8 1.18 70+100 -0.39 h 7 8 1.28 90+125 -0.39 8 8 1.28 90+125 -0.38 m
TABLE 5 ,
TRANSIENT TEST RESULTS .,,
Measured Doppler
- Calculated Doppler-Initial Power Slug Worth Energy Coefficient Energy Coefficient (MW) ($) (c/11W-sec) (c/MW-sec) 2 0.97 -0.56 -0.54 .
2 1.18 -0.47 -0.49 2 1.28 -0.45 -0.46 5 0.97 -0.45 -0.45 5 1.18 -0.42 -0.42 5 1.28 -0.42 -0.40 ,
, 8 1.18 -0.39 -0.37 w
T 8 1.28 -0.38 -0.36 10 0.97 -0.35 -0.37
- Calculated with T = -0.0081. i 3
- _ _ . . _ _ . - . _ . - _ . - _ _ _ _ _ _ _ - . _ _ _ _ _ _ _ _ - - _ _ __