ML19220C323

From kanterella
Jump to navigation Jump to search
Pseudo Rod Worth Test Procedure Should Not Be Accepted W/O Analysis by Licensee of Consequences of Possible Ejection Accident
ML19220C323
Person / Time
Site: Rancho Seco, Crane
Issue date: 03/09/1978
From: Seyfrit K
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE)
To: Davis A
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
References
TASK-TF, TASK-TMR NUDOCS 7904300395
Download: ML19220C323 (6)
v • d • e


Text

{{#Wiki_filter:.. A.

  • ~

l ~ d.$ 'R 3 1973 ' h I ./ MDORANDUM FOR: I Davis. Chief, Reactor Projects Section No.1 Region I FRCM: K. V. Seyfrit. Assistant Director for Technical Programs, IE SUGJECT: PSEUD 0 ROD WORTH TEST PROCEDtJRE AT THREE MILE ISI.AND UNIT NO. E (AITS F14643H2) r The subject test procedure was reviewed as requested. We support your conclusion that this procedure should not be accepted without an analysis, by the licensee, of the consequences of a rod ejection accident. The Three Mile Island accident analpis provides detailed inforration to show that the consequences of a zero power rod ajection at the beginning of core life are acceptable.for rod wrths up to 0.7% AK/K (Sectica 15.1.19 of the TMI-2 FSAR). In addition, an ejected red wrth of 1% AK/K is calculated to result in a mxicun peak then::al power of only i32 before the transient is turned around by a com-bination of negative doppler ccafficient and reactor trip (Table 15.1.19-3). This analysis is adequate for normal operation since tia rod insertion limits result in maxinsa ejected red worths in this.ange. The s'oject test however, requires the reactor to be critical with 111 regulating red inserted. As the boron conterrt in the coolant is rea'ced to achieve criticality with a larger ntreer of rods inserten. the unrodded regions in the core become core reactive. A result it, that the mximum individual rod worths are greatly increased. This effect is dmonstrated by Figure 4.7-2 from the Rancho Seco startup Report (enclosed). The Rancho Seco, Three Mile Island.2 and Davis Besse (referred to below) cores are ainest identical. Applying the regulating rod wrths given in the subject V.vc, lure to Figure 4,7-2 results in an ejected rod wrth of about 1.9% AK/K. This value is well above the range addressed in the Three Mile Island-E analysis. As shown in Figure 15.1.19-1 (enclosed) Da peak neutron power increases very rapidly as the ejected rod wor.h goes above 0.7% AX/X, which is near the effective delayed a C " TACT: c j;_gigrtpatrick, n 7904300 MS ( o /m ~ su.=Ama > oavu > NRC FORM 313 (M6) NRCM 0240 'A' ua s. eavunwuswr emwvine orrmai s eve - eae.as4

A. B. Davis 2 MAR 9 1978 ceutron fraction at beginning of life. In addition, the Davis Basse rod ejection analysis indicates that rod afections with reac-tivities in exces, of 1.525 AX/K could result in rupture of the reactor pressure vessel (Davis Besse FSAR 15.4.3.2.5). A test procedure similar to the subject procedure was proposed for the Davis Besse startup. This test was discussed with members of the NRR Core Perfarnscce Branch. There was unanimous agreement that the pnscedure was unacceptable without an analysis of the consequences of the higher ejected rod wrth. The Rancho Seco startup test rm)ceduro (description enclosed) was suggestad as a possible alternative. This procedure inycives measurtag the reactivity difference between two rod drops from critical with bank five partially withdrawn. One drop is done with the assumed stuck red inserted, the other with the stuck rod withdrawn. Cognizant NRR members were again contacted regarding the proposed Three Mlle Island-2 test. The consensus recsains that this tast procedure should not be done without further analysis. Hewever, it is also believed that the rare accurate data to be gained frem the less conservative procedure night be justified on a oca time only basis on this type of core. Such a justification woula depend on an analysis to show that the information gained would justify the risk involved. This analysis should be submitted to and approved by NRR prior to tha tast. In suusary, we believe that a zero pcwer test with all requiating rods inserted, should not be approved without further analysis, as stated in earlier positions regarding this matter. Such a test might be justified on a one tina only basis, provided that, an adequate a mlysis is approved. We plan no further action on this matter. Karl V. Seyfrit, Assistant Director for Technical Programs Division of Reactor Operations Inspection

Enclosures:

As stated cc: w/ercls. DISTRIBUTION Regional Directors Docket Files (2) 'A6 P. J. Kallogg, RI IE Files gQ ? H. S. Dunenfel:f. NRR DCX1rkpatrick Rdg Files W. L. Broots, f5t2 TF Rdg Files M M1va; _ 40U DDT Ddn 1"% ICT I:."/. AD- '""*"'* XE ig t, kk:mL LYSeyfri t 7/9/7^ t/ /7't-W u n. eovsmusw mmme omem im-essass N1C 16M 318 (9 76) NRCd 0244

~

  • - * * ~ * * ~ ~ ~ '

,, ', M.s 9. n +i .s. T.',. i

  • *" i i

bfh ' ' E p l. . l.4.L

.+ J.

4. [^' kb 1T b i i i 1 -3 i W I e M "O b * **

  • ~

i H--, - -e.,, ao, w-3, --ye.-- ua. i y i e.e rg C i. i i d , e- - h - M** *-* N* *

  • t-" v 2: r f3 0I r-f-h I h* 1, 4
- T- -

i t l +- r < -r -, G3 r r. -- %.l

  • t r -

o i . iT '- i -.1 ] .1.L: !. ;, D. i, iji I 1 i 1.? F-t - r - ci.. .4 4 4 - - -. i w - m. 6 g , i.., i --t s r-+ +r1 C3 .X i 3 m l-f.,. ., ~1_._h**"* k _.. i * * *

  • l W

g - '.***-*,P

  • -l
  • d

'o g t F.. i. i,. i 4 , i I* ! "W. i .e e i l -* h *T" i--* f?** " I g C as d i t u P4 e . 4e L... . be 4r. . 'i 4,., o 1

.1....g,

T.M 1 %. 1. i - e i. N ed y .t, ,.4 -{-.--.. I y --r.-, g g --- --- j

  • -..t g

y -c =c y h# f *~ P t " i l hT* k"' ,i., 4 =4 [ N 'i.-fl*I--HN- -h ' l' '\\ l . l-i I,. T-"T-*, x i, N 1 ~~TTT - - - - ----N ' ' ' l h h

  • b I t ! ri d. t l ' -

u \\' SJ n4 L.,...i ',.J. '%_ ;.q ..L I a %g 14..{ 4.-. 1 i.4_ 4 t I c .a, ie '.\\

  • U

.L Z.L.. 4'3 i _ s., e i Y i - i.. h ...N' fP h h -h N LN l '. I ki 'l

  • b *'

k w . q , i T d L-- 5 ~ 'c's s 4 'r'--'*' ~A I 4 I-I~4 M ' ; ---"* J' f ,.L' I1 .t_, ! ' ' ' 9 y'

  • 4

)j l* { .i 4 f ' t j L4. L ^ l ^, M,.- o j. 8 . W

g. } e

.} '. j i_{. A w 9- ._,i + t 3. 5 )l_.L i H 4 6..i7.sr,r-r_ '. r. ~ -__h t--j, cy ', j -* -*-' ' t

  • T-* h h h* h -*- -*-h -~.

e. . rr ~ -,-m u H a ~ Ll \\', N 2 ---l-...-- ---y - 4 - -- T ; ---- L - i 7 h. & ~-- 6 L[ p 4 +f-j

i..a

}.ii:4_LJ .'l y -. ; .L.._.. j r a. L- - 4 L ' - J-. J, e il .r 4-.I-L{ H.. .1j. t NJ V ..-.4 ... u _ i.u _ -,. 4...... . g,__ i, .i 2 - 4 y L4..) .,4... N) k ~

  • -..s

,y k & F-{.

1. L 4 6 4

.I _- 4.7..*.. u. -., - 4. ;.--- 7 7

-j l

3Ft I-

l-v k

-. - y.? a ,-.3 C L __- ..s. .- i..}. -. ~. a. J.U._ I .4 h,;. _ . _1. .u..'._' A '.;. L -. H L L'. _. w.. t ' 4 <t t o m n m '*JM7 :: 'T42:oM peu pa2saf2 e 4.7-4 00] O j./ 0 u

tuuuu e. ,g 1000 Ratea Power.EOL / J Ratec Power, BCL E ] 100 O y 10-3 Ratec Power, BOL f E ~ 10-3 Rates Power,EOL 10 / ~ Nomin,al for all Curves l 0.2 0.3 0.4 0.5 0.6 07 0.8 Contral R:a 1orin. "Ak/k PEAX HEUTROH FCWER AS A FUNCTICN OF EJECTED CCNTROL RCD WCRTH THREE MILE LSLAND NL' CLEAR STATION L' nit 2 90;O b .N j W nn i FIGURE 15.1.17-; o7

Qj,. j " '-"MCW) STAO*UP %27~ 4.8 STUCK CONTROL RCD VORTH Technical Speci fications requi re that the shutdown margin be &t least 1% ak/k with the highest worth rod fully withdrawn. The rod calculated to have the highest worth, with all other rods fully inserted, or those symmetric to it. (See Figure 4.8-1). The reactivi ty worth of CRA H-2 and the shutdown m.argin with CRA H-2 fully withdrawn were measured and compared with calculated values. The's tuck rod worth was measured at 532*F with APSR's 36% WD by performing two rod droos. In the first drop, all rods (except APSR's) were dropped from a critical state with CRG-5 at 48: WD. For the existing boron concen-tration, this drop measures the subcri tical reactivity with all rods in. The second drop was made at the sam'e baron concentration but from a critical state with CRA H-2 at 100% WD and CRG-5 at 17% WD. All rods were dropped except the APSR's and CRA H-2 tnus measuring the subcritical reactivity with CRA H-2 at 100% WD and all other rods in. The difference between the two drops i s, there fore, the worth of CRA H-2 with all other rods in. An uncorrected value for the reactivity introduced by each drop was obtained. These values were multiplied by correction factors which had been deterr.ined from measurements on a similar reactor by comparing results from red drop and boron swap measure-The measured resul ts are comparea wi th predicted values in Table 4.8-1. me n ts. Tb.e value thus obtaineJ for the stuck rod is well within test acceptance criteria of 3.91% Ak/k + 30t. Since this value is grea ter than the measured, it is the value used for shutdown margin calculations to insure conservatism is always applied. 4.8-I i n G') 9 [I <O Q L

'""^' " ~~ W

STAArdP L g TABLE 4.8-1 COMPARISON OF MEASURED AND PREDICTED STUCK. ROD WORTH Uncorrected Corrected Predicted Deviation CRG's Measured Tierth Correction Measured Worth Worth frc=

Dropeed I 6k/k Factor A k/k ! Ek/k Predicted 1,2 3,4 4.59 1.35 6.19 6.25 -li 5(48 + 0: tid) 1,2 3,4 2.90 0.93 2.70 2.34 +15 5(17 + 0: WD) CRA H-2100" WD and re=ains cut Stuck Red Uorth 3.49 3.91 -11" e e 4.8-2 ^ 89 250 .}}

Retrieved from "https://kanterella.com/w/index.php?title=ML19220C323&oldid=4905887"