ML20012A350

Comments on Three Questions from 900123 Senior Reactor Operator & Reactor Operator Written Exams

ML20012A350
Person / Time
Site:	Neely Research Reactor
Issue date:	01/26/1990
From:	Karam R Neely Research Reactor, ATLANTA, GA
To:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
Shared Package
ML20012A334	List: ML20012A333 ML20012A337 ML20012A350
References
NUDOCS 9003090301
Download: ML20012A350 (7)

Text

-

-3

~

Georgia Institute of Technology M '

NEELY NUCLEAR RESEARCH CENTER OCC ATL ANTIC ORIVE ATLANTA, GC ORGLA 30332-0425 l

t January 26, 1990

)

l Mr. Thomas A. Peebles, Chief j

operation Branch Division of Reactor Safety

(

U.S. Nuclear Regulatory Commission, Region II

'101 Marietta St.,-N.W.

3

~ Atlanta, Georgia 30323 l

Subject:

The SRO and RO Written; Examination for GTRR perssonnel j

of January. 23,,1990.

t

Dear Mr. Peebles:

l I wish to thank the1NRC;for the. opportunity to comment on three l

questions.out of 60 given at the Operator' Licensing, Examination of January 23, 1990'.

l

1. ' Question A-5.

"Enough reactivity is added to a shutdown reactor to double the countirate.

If this same amount l

of' reactivity ~is again added to the reactor, then'the i

reactor will.be' 1

1 a.

subcritical b ~.

critical

.l c'.

super; critical l

d.

prompt critical."

i l

The answer given'on the NRC's list of answers'is.(c),:1.0. super-critical'. ~The reference-to this problem is given as INP-1971 p.110-19. ' Additional material-intsupport'of A-5 was supplied by.

Mr; Richard Baldwin of your staff'on 1/23/90.

This material is appended.

i Comments reactivity, p, may be defined as 1

' C p=

Ikm) 1; ref 1NP 19':!, p. 10-19 (k,)

'A'better definition ist p=

Ikm)' - Ikm) j (k,)

9003090301 900226 D

PDR ADOCK 05000160

$lCi 7

4 V

PDC 1 Teles:: 542507 GTRIOCAATL Fax: 404-8944120 (Verify 404 8944951)

A Unit of the Urwereity System of Georgia An Equal Educaten and Employment Opporanty instiamon

pa ;,.

1 i

=

Where k,,' is the perturbed k,,of the system and where in the second definition it is recognized that reactivity is simply the fractional' change in keff regardless of whether or not the reactor i

is critical.

From the statement of the problem and the normal l

simple multiplication approach, it can readily be shown that 6

1 - I km ).

l w

IrJQ,

=

(CR),

1 - ( k,, ),

For A-5 (CR),/(CR),

1/2

.j

=

Consequently, k,,,

1 + I k_ ),

j

=

2

-l If one assumes that (k,,), = 0.99, one obtains a' value for k,,, =0.995 i

and p = ( km ) - (km).

( k,,),

.995 - 0.990

= 0.00505

=

0.990 if one adds p to (kef f), one gets 0.99500 + 0.00505 = 1.00005 i

I don't believe that a reactor with keff = 1.00005 qualifies to be called super critical.

?

L 2.

Question C-8.

"Which one of the following Nuclear instrument channels is the special instrument channel that will permit the cross checking required in the low flux environment of a subcritical reactor?

j t

I a.

Count rate meter channels i

b.

-Micro-micro Ammeter channel L

c.

B10 proportional detection channels d.

Log-N and period amplifier channels. "

1 y

Comment: The answer to C-8 is (c), in accordance with the statment l

appearing on page 66 of the SAR.

However, the GTRR does not have any.B-10 proportional counters.

The B-10 channels are all l

ionization chambers.

When approaching criticality for the first

[

time back in 1964, the GTRR had B-10 proportional counter for start-up purposes.

Under current condition answer (a) would be more appropriate.

l l

I v

'M

..__.,,.m_.

..a

,o 1

1 3.

Question C-15.

"According to Technical Specifications.

while operation in Mode 2, which one of the following parameters is abnormal for steady state, forced convection mode of operation?

a.

Reactor coolant inlet temperature 124 degrees F.

b.

Moderator level within 10 inches of overflow.

i c.

Reactor coolant outlet temperature 138 degrees F.

)

d.

Reactor coolant flow 1650 gpm."

l Comment:

The correct answer for question C-15 is (a).

But (b) is i

also correct. When moderator level is below overflow, the GTRR will scram.

3 We appreciate your willingness to reconsider the above questions.

1 Sincerely, j

k./9 ' hk" R. A. Karam, Director Neely Nuclear Research Center RAK/arr i

j t

a w.

,-y---

I

.o Se ::untrate :n Se cut-cf-c:re nuclear instrumentanen will increase

ue :: Se :ncrease m subct:ucal muluplicaten.

There :s no reason why 4101 :urve cou!d act be plotted f:r the startup of a fuuy 1

'caced react:r.

n pr:cuce, hewever, it is seldcm done.

First

• nere :s Se ;eeme r:c pr:biem with respect to detec cr locaten.

Th:s ::uid precably be : vere:me, as bef:re, by proper averaging techniques.

A see:nd pr:blem :s chocsing a quanuty against wn:ch is pict 1. M.

The !cq1 cal chetce is agarnst con:r:I red positen.

H: wever, s:nce the rescuvity strength Of different centr:1 rods var:es

ensiderably. Se K,ff Of the c:re depends str:ngly upon which

n:::1 ::ds are withdrawn as well as how many c ntrol reds are wt2-

rswn.

Th:s int:ws an adcitenal ::mplicaung fact:r ints Se pict.

But e.e ma:n reasen wh no 1.M plot :s made is simply because :t g

is net worta de effert.

Cnce the reactor has been br:ught crtucal

• ne first ume, the data item this first criccality (called physics test-mg) :an be used to predict de red positions on subsequent I

isppr: acnes to crtucal.

As for predicung the red pattern on the firs: :r:ucal, the Operster usually uses the results of physics tests T.ade durtng the intual ! cad:ng plus design predictions suppbed by the manufacturer.

This informauen is c mpiled

.n:: s reacuv:r/ balance precedure and the operater uulizes

• nts :ata :: predict :ne crincal red posit!cn.

(Reacuvity

a;an: As w:H te :iis::ssed !ater in Chapter 5. Secuen II.

NCTI:

The Operator must be able to discuss what is happening in the react:r during the approach to cr:ticall:7

-3.

5U3CR:T: CAL MULTIPLICAT!ON " RULES CT THOtB" a.

Scme c:nvenient rules Of dumb that may be used by the

perat:r as an aid in deter aning the point of criucality are as f:1!:w s :

-:* ' e nf. re:::i::ir h.:i ?ee- :isci :: :

s'.a i:: re::::r w i:.i-;e e

R :n~:.~::c, : ;i;;e : :a-: :, re:::i::iry :iii-i:n.: *;; ::.a:e -;.e n::::: :: ?c:: e, raper :: ::i::;.

1

r:ci:

Using equanen (3-40).

Assume an intual K,f*, of 0.3. CR t c

= 20 :ps and CR2 = 40 cps.

)

l b

l-K ri:

CR e-

,r i

=

C3 IS eff

(

t 5.:st:uung:

00 1-K,ff

=

40 1

.3 190

y

,0-And:

(.5) (.2) = 1-K,ff,

.1-1

• Kef, K

= 0,9 eff; 1: sing equatien (3 41)-

p,'eff:

"' e f f i 2=

eff:

eff t

1. * (.9

.8 Subsutuung:

.9)(,8) 2 =.1389

• e eacuvit" added to double the countrate was 0.1389.iK/K.

If

-w e aSd the same amount of reacuvaty again and using equauon (3-41) we 'ind :

E

~

eff; O.1389 =

.9 K,ff,

-Keff,

• 9 0.1250 K,ff, K,gg, (0.1250-1) =-

.9

• 9

-* 1.028 getf,a

. 875 One can see that the reactor is supercritical.

The cperators shcuid convince themselves that regardless of the iniual K,ff the accve relat:ensh:p holds true and the reactor will always ce supercr:ucal under these condiuons.

-hw-; U 9.;; i,' the OFCP::: P 038#8 ;b# 5'{$

f ##".).

'N g

- 'en " '

• L~ C S The TC20;:? $8 EY.' 0'###

,i 4

130 i

-=_-m

- _ _ _ _ _ _ _ = _

.:.o -

i

• 4 Assume de instal concutens as de prev cus prcblem.

l That :s K,,

• 3 and :nital : unts CR i = 20.

Rememcer the sub-Cplicaten formulas to ect held when K'If > ;; therefore Or: cai assu ::e a :r:::si K., Of.999.

et.

Pr:cf:

00 :ps d ubled seven tmes = 2560 :ps Us:ng equaten (3 40) l'E CR eff 1

a CR

-K effa S ub st

tutr.;:

20 c:s 1

.999 l

CR 2 1

.6 90 aas

',335

• 4000 cps

[

CR

=

The estmate: ::::: cal CR of 2560 cps : mpares very favorably with 7.e :3.:u. ate:: Or:t: cal CR cf 40C0 cps and the reacter is indeed verv r

2. 3 e

• ,0 e tr.Q O r10 c al.

[

The :perater should always remember that any critical

rec ct

n.s :nly a guide, and it is generally very fortunate if tne react:r actually gees cridcal at exact!y the point which was pr e::ctec. Cn tne other hand, if the predicted critical' rod pattern

.s puned and de core sell appears to be considerably subertucal, tne :perat:r shculd make an investigauon before pulling additional r:cs t: ver:fy that the discrepancy is simply the result of calculation na::urse:es ender than an indication of mere serious problems such is a malfun:u:n in the red centrol system.

In all cases, the operator T.ust use n:s judgment, c upled with the information ebtained from his e.

.nstrumentat:n to insure that crincality is approached in a safe manner.

t 2

U SECTION IV I

T'.,UX DISTRIBUTION AND REACTOR CONTRCL 94, CONTROL METHODS, CRITERIA, AND MEANS f:

~

82 L

Two ::neellable parameters existing in a nuclear react:r

reaty affe

t de reacuvity balance within the system.

These are

.U

nr:i.ater:3.s pensening and soluble poison ::ncentraten.

D t:

si 1

h

O C-3 s

.w t.

L ENCLOSURE 4 t

NRC Resolution of facility Coments R0/SRO Examination Question A-5 NRC Resolution:

The definition of reactivity given in INP,1971 and documen-tation supplied for Suberitical Multiplication " RULES OF THUMB" are both accepted by Industry and by the Academic Society.

Even though the problem may be calculated to indicate a very small amount of Keff, the reactor would still be considered supercritical. The examination question stands as is.

Question C-8 NRC Resolution:

Comment acknowledged.

Page 66 of the SAR indicated the GTRR utilized B-10 proportional counters, and there was no facility comment during the preexamination review process.

The GTRR presently does not utilize B-10 counters, therefore, answer (a) is the appropriate answer.

The answer key was modified to indicate (a) as the correct response. The facility is reminded to update reference material to reflect current plant equipment, 1

c Question C,

NRC Resolution:

There was no comment generated during the preexamination review process.

A close review of the Technical Opecifications indicates two possible responses could be correct.

NRC policy, concerning questions with more than one possible correct answer, is to delete the question.

The examination and answer key was changed to reflect deletion of the question.

l i

y l

1