ML20031H452

From kanterella
Jump to navigation Jump to search
Testimony of Rf Stirn Re Doherty Question 24 on Drop Rod Accident & ASLB Question on Rod Worth Limitation.Ge Rept Aped 5756 Does Not Affect Conclusions in 800729 Affidavit
ML20031H452
Person / Time
Site: Allens Creek File:Houston Lighting and Power Company icon.png
Issue date: 09/18/1981
From: Stirn R
GENERAL ELECTRIC CO., HOUSTON LIGHTING & POWER CO.
To:
Shared Package
ML20031H319 List:
References
NUDOCS 8110270492
Download: ML20031H452 (5)
v • d • e


Text

September 18, 1981 l1 UNITED STATES OF AMERICA i

NUCLEAR REGULATORY COMMISSION 7

_BE? ORE TFE ATOMIC SAFETY AND LICENSING BOARD 3

In the Matter of S

4 S

HOUSTON LIGHTING & POWER COMPANY S

Docket No. 50-466 S

(Allens Creek Nuclear Generating S

Station, Unit le S

g j l

i DIRECT TESTIMONY OF RICHARD C.

STIRN REGARDING DOHERTY ' CONTENTION NO. 3 (1)

ROD DROP ACCIDENT (2)

BOARD QUESTION - ROD WORTH L!M TATION c.

'~O Stirn, have you reviewed your prior affidavit Q.

Mr.

11 on Doherty Contenticn No. 24, which affidavit is attached 12 hereto as Attachment RCS-l?

T31 A.

Yes, I have.

Q.

Are the statements contained therein still true and 94 correct?

A.

Yes, they are.

16 Q.

In its Order of September 1, 1981, the Licensing 17 Board noted that your affidavit did not address GE Report, 13 Would you please explain why that report does not APED 5756.

~

13 affect the conclusions stated in your affidavit?

20 A.

Prior to developing the methodology described in 21 NEDO-10527 GE employed the use of.a one-dimensional (1-D) excursion model for evaluating the control rod drop accident.

--ss The model described in NEDO-10527, which is GE's current

..42 licensing basis model, is based on a two-dimensic tal (2-D) l 24I l

i 4

8110270492 810910' PDR ADOCK 05000466 T

POR - _ _ _ _ _ - __-_- _

model.. This 1-D model predicted lower fuel enthalpies than the 1

2-D model.

A detailed explanation for these differences is given in APED-5448.

2 In addition to 'he above, changes in the reactor 4

core designs since 1969 have resulted in changes to the

=

control rod drop accident characteristics.

NEDO-10527

~

5 and supplements 1.and 2 of tbf, report discuss all of

'T these changes and their impac. on the control rod drop 3

accident in detail.

A caraful examination of these reports will reveal that depending on core design, operating state, g

rod drop velocity, scram insertion time, core exposure, etc.,

many different control rod worths may lead to achieving 280 11 For one core design and condition a 1.9% A k rod cal /gm.

to may result in a peak fuel enthalpy of 280 cal /gm whereas for

' ~~

1 3 a second set of conditions a 1.4% A k rod may reach 280 14 cal /gms.

There is no inconsistency on this issue, and in 15 fact NEDO-10527 and supplements 1 and 2 were specifically written to document the sensitivity of the control rod drop

~

lg accident (CRDA) to various design parameters and core at operating conditions.

,,o It is sufficient to state that t.2e Allen's Creek 19 core design in conjunction with the banked position with-20 drawal sequence (NEDO-212 21), as enforced by the dual channel rod pattern control system, will result in CRDA peak enthalpies c~

~~

22 which are well below the 280 cal /gm licensing limit.

The 22 evaluations and sensitivity studies presented in NEDO-10527 I

are directly applicable to the Allens Creek design.

24 I,

1 Q.

Has GE considered the effects cf residual reactivity and destructive pressure pulses in its analysis?.

2 A.

The licensing criteria of 280 cal /gm precludes the 3

existence of destructive pressure pulses.

This issue was 4

addressed in detail by testimony given by R.

J. Williams 5

~

and K. W. Holt = claw and will not be reiterated here.

6 The concept of " residual reactivity" as described 7

in BND-50584 is only applicable to calculational models which use a weighting factor concept to calculate various 9

reactivity components.

GE's model as described in NEDO-10527 10 does not use the weighting factor approach and hence the 11 concept of residual reactivity is not applicable.

As described 12 in NEDO-10527 GE's model calculates reactivity using an 13 eigen value difference of a upatial core physic representation i

14 of the direct neutron flux iterations..

I' sing this approach 15 eliminates any weighting factor approximations and hence, 16 any potential errors that would result from " residual 17 reactivity".

t 18 Also, great care has been taken to ensure that the 19 reactivity and point kinetics parameters as used by the point model kinetics are defined on a consistent basis.

This 20 consistency has been demonstrated by the comparison of this gy m del to the SPERT I and lII coce 7.xcursion tests as 22 discussed in NEDO-10527.

Q.

Is there any requirement for ccmmitment by 1

Applicant to a control rod design that limits incremental 2

rod worth to 0.8% ek/kT 3

A.

There is no licensing cr design requirement that the 4

control rod worth be limited to 0.8% A k/k.

The only licensing limits for the control rod drop accident are that 5

the peak fuel enthalpy cannot exceed 280 cal /gm and that the 6

radiation release rates he within 10 CFR 100.

Since the a

sensitivity studies presented in NEDO-10527 have conservatively demonstrated that 280 cal /gm.will not be achieved or exceeded for rod worths less than 1% A k/k, the NRC does not require a plant specific analyses for rod worths less than 1% Ak/k.

11

~

For rod worths in excess of 1% A k/k a plant specific 12 analysis is Lacformed to demonstrate compliance to the 280 13 cal /gm licensing' limit.

Hence, there is no specific require-14 ment to restrict control rod worth by the design but only to 15 maintain or demonstrate compliance to tre 280 cal /gm and 10 16 CFR 100 '3 icensing limits.

17 The ACNGS plant design is committed to the rod 18 pattern control system and core design which limits control 19 rod withdrawal sequences such that the peak fuel enthalpy 20 resulting from a design basis CRDA will be much less than 21 the stated licensing limits.

Hence, the total system design 22 more than meets the licensing requirements for the CRDA.

23 0

Does your last answer also address the questior.

raised 'sy the Board at page 61 of the September 1 Order :'

24

, ~.

1 A.

Yes, it does.

In my view there is no inconsistency 2

between the rod worth limitation (1% A k/k expressed b-j 3

the Staff and the value of.0083 A k/k used in GE's analysis).

4 5

6 7

8 9

10 11 12 13 14 15 16 17 18 19 20 !,

21 lli i

22 23 24

Attachment RCS-1 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION N

BEFORE THE' ATOMIC SAFETY AND LICENSING BOARD i

In the Matter of HOUSTON LIGHTING 4 POWER COMPANY)

Docket No. 50-466 (Allens Creek Nuclear Generating Station, Unit No.1)

)

)

)

L AFFIDAVIT OF RICHARD C. STIRN d.

State of California County of Santa Clara

(!G I, Richard C. Stirn, Manager of Core and Fuel System Design within the Nuclear Power Systems Engineering Department of the General Electric Company, of lawful age, being first duly

.p) sworn, upon my oath certify that the statements contained it' in the attached pe.ges and accompanying exhibits are true and

=-

correct to the best of my knowledge and belief.

(W I

Executed at San Jose, California Julyg2ff, 1980

.4 0. W i

ti Subscribed and sworn to before me this J f day of July, 1980.

~'

/

3 OFFICIAL SEAL h

/r //X)\\

lM7#f 6 i mcecececececm ececececococece<

S KAREN 5. VCGELHUBER G

NOT//BY PUBLIC IN AND PpR SAID '

J' noun nisuc.CAUFCRNIA 3 COUNTY MD STATE g'"

3 U

3 O

SANTA CLARA COUNTY g 3

My Commissien Expires Dec. 5,1980 g

m occo:cescocc cc>:o:ocancncno:6 0

My commission expires

/J-5 of 19[_d

.:e

.fl T) poM

.=

6 39

Attachment RCS-1 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD p.-

2 In'the Matter of 5-

?"

S 3

HOUSTON LIGHTING & POWER S

COMPANY 5

Docket No. 50-466 5

(Allens Creek Nuclear S

Generating Station, Uni t No. 1)

S A.ffidavit of Richard C.

Stirn My name is Richard C.

Stirn.

I am employed at c.

a),

General Electric Company as a Professional Nuclear Engineer, I have been so eraployed for fifteen years.

A statement of my f'.[l.

experience and qualifications is set out in Attachment 1.

.('

This affidavit addresses the issues raised in li; Doherty's Contantion No. 24 which states that the Applicant has not provided a basis for showing that the reactivity insertion from any dropped control rod will be sufficiently 5-small to prevent the peak er.argy yield from exceeding 280

.["

calories / gram of fuel.

^ :

I.

Introduction g_

For gross control of reactivity in the Allens Creek reactor, cruciform control blades are inserted between the 3-fuel assemblies. 'These control blades, or rods, enclose T

~JA y.J

... ~

~

y 4

smaller rods filled from boron carbide, a neutron absorbing material.

The reactor is controlled by driving these control f*

blades-(177 for ACNGS) into the reactor core to reduce reactivity (and thus pover) and withdrcwing the-rods to increase reactivity

~

(and power).

Tne control blades are moved by hydraulic drives which insert or retract the blades in small increments and cortinuously drive the blades in on a shutdown signal.

t The hydraulic drives are attached to the bottom of the pressure vessel.

' ch drive is attached to the bottom of a control blade bl a special coupling.

f

(_

II.

The Control Rod Droo Event There are many ways of inserting reactivity into a iyl' boiling water reactor.

However, most of them result in a-

!L relatively slow rate of reactivity insertion and therefore Lg,.jj pose no threat to reactor control.

It is possible, however, y~

that a rapid removal of a high worthl/ control rod could.

e

?

result in a potentially significant power excursion.

The f

design basis accident dealing with rapid removal of a control rod is the rod drop accident.

1 c.r S;

The worst-case credible control rod drop accident i

I' for the ACNGS design is described as follows:

(a)

Reactor is operating at 50 percent control rod

,g density (half of the rods withdrawn in a checkerboard 1v pattern).

This pattern resulte in the highest incremental

!H

- 1/

Control rod worth is the measure of reactivity which will Ee added to'the reactor if the rod is moved.

I w 1 4

~

n i

rod worth for the' inserted rod since the four adjacent control blades are withdrawn.

I (b)

A fully-inserted control rod drive must sustain I

a complete break or disconnection from its cruciform control blade:at or near the coupling.

L (c)

The blade must stick in the fully-inserted position as the rod drive is withdrawn.

~

IB (d)

The blade fal1s by gravity to the position b

occupied by the rod drive after it is withdrawn.

Without regard to how the control rod blace drcps, the worst case result is a blade falling unimpeded 4

by its drive under the influence of gravity.

The

I mechanism which causes a rod to drop ':Us not a concern provided that'the.naximum rate of fall (reactivity

?;

insertion) is used in the analysis.

The sequence of h.c.

events and the approximate times of occurrence are as r;

follows:

J Approximate Event Elaosed Tirae (a)

Control rod wh2 :h provides maximum incremental worth becomes uncoupled.

t (b)

Operator selects and with-draws the control rod drive of the uncoupled rod such that proper : ore geometry for maxijmum control rod worth exists.

(c)

Uncoupled control rod sticks in the. fully inserted position.

l f -

.j s 1-

I Approximate Event Elaosed Time (d)

Control rod becomes unstuck I'

and drops at a nominal measured velocity */

plus,_for conservatism, three standard deviations.

0 s-(e)

Reactor goes on a positive period and initial power burst is terminated by the Doppler reactivity feedback.

1 sec.

(f)

APRM 120 percent power sig~tal scrams reactor.

(g)

Scrams terminates accident.

5 sec.

This rod drop event sequence is the worst case because no other credible sequence of events can add positive reactivity at a faster rate.

~ '

III.

Rod Pattern Control System The purpose of the Rod Pattern Control System

~

(RPCS) is to limit the worth of any control rod such that no unacceptable effects will result from a rod drop accident.

P 11

  • /

The velocity is limited by the control rod velocity b;-

. limiter.

The velocity limiter is in the form of two nearly mated conical elements that act as a large piston inside the control rod guide tube.

The vclocity limiter is provided with a streamlined profile in the scram (urward) direction Thus, when the control rod is scrammed, water flows over the smooth surface of the upper conical element ines ohe annulus between the guide tube and the limiter.

In the dropout direction, however, water is trapped by the lower conical element and discharged through the annu us between the cuo conical sections.

Because this water is jetted in a' partially reversed e

direction into water flowing upward in the annulus, a severe turbulence ir created, thereby slowing the descent of the control rod assembly to less than 5 feet /second at l-70 F.

-/

- i

-_~

j-d

. U"

The RPCS will apply rod blocks before any rod motion can produce high worth rod patterns.

The RPCS is a dual channel,

]^

scfety-related system.

These electronic circuits will have, in permanent storage, the identification of all rod groups y

and logic control information required to prevent movement of rods into unacceptable rod patterns.

The RPCS, hence, limits the maximum rod wor th of any rod which might uncouple and drop as discussed above.2/

IV.

Consequences of Rod Droo Event With the RPCS operational, the maximum incremental worth of any control rod is limited to approximately.8 percent esK.

This limit is derived from the analysis in

" Banked Position Withdrawal Sequence," NEDO-21231 (January, 1977).

This very low incremental rod worth will produce a specific enthalpy well below the NRC design limit of 280 3,

calories / gram.

The peak enthalpy from a dropped rod given the conditions described above are less than 135 calories / gram.

This result is computed using an adiabatic approximation of a super-prompt, critical large co.re and a two-dimensional multi-group flux representation, as discusced in " Rod Drop Accident Analysis for Large Boilin~g Water Reactor," NEDO-a 10527 (March, 1972).

2/

The design and functioning of the RPCS system is more fully described in the Affidavit of Mr.

J.

F.

Lesyna filed concurrently in this docket.

A 1 -

-2>

~

v J

.e.

11 f'

r.

-V.

Conclusions The design basis control rod drop event is the

.')

worst case reactivity insertion accident.

To minimize its te -

effect, a highly reliable Rod Pattern Control System has been t,

~

designed for Allens Creek which will maintain individual li incremental control rod worth less than.8 percent.

As a result, a rod drop event cannot produce a specific fuel enthalpy greater than 135 calories per gram.

4

'83 E{",

G' S4 *

.N' T

w ed a-

-dE&

ATTACHMEN'-' 1

['

QUALIFICATIONS Richard C. Stirn, Manager Core and R:e1 Systems Design General Electric, NEBG r

My name is Richard C. Stirm by business address is -175 Curtner Avenue, bhil Code 'i40, San Jose, rnWornia, 95125.

I Pz. a registered Professional hbclear Engineer in the State of California (hU 630). As Fhnager I have the responsibility cf direccing core and. fuel systecs design for the General Electric C% ny, NE3G.

2 I grnebted from Tennessee Technological Univercity in 1962 where I re-ceived a Bachelor of Science regree in Engineering Science. ~During the Sunner of 1962 I worked 5ar the Arnold Engi.eering Development Center m Tullahoma, Tennessee as -a:rEngineer.

In the Tall of 1962 I entered Purdue University cn an AEC Fellowship, and in August of 1964 T. received a Master of Science Degree in hbclear Engi-neering. Upen cc pletica of cy studies at Purdue I entered the University

~

of Arizona to work tcward a.PhD degree in Riclear Engineering; however, I left schec1 in February of 1965 to work for the General Electric Company, NEBG before % eting the-PhD requirements.

P 1

Upon joining General Electric I entered the Eagineering Training Program and had assignts hmg with light water coderated. thermal reactors, steam cooled. fast reactors, and sodium cooled fast reactors.

After cca-pleting my training assigenent in October 1967, I was appointed to the position of Technical leader of Core Dynamics and Reactivity.

In June

]

1972 I was appointed to the position of bbnager of the hbclear Safety Analysis Component of the-Core Nuclear Engineering Unit.

In this position I co-authored or contributed to three papers and three reports on the topic of nuclear reactor excursion analysis. I also participated'in the development. of the control rod drop accident boundarf value approach for the reload licensing submittal.

In September of 1974 I assumed my present responsibilities as hbnager, Core and Fuel Systems Design.

In this capacity I an responsible f'.c the development of system requirementc for the~ Core and Fuel Perfomance, Core Performance Transient, and Fuel Mechanical Systems.

Additional responsibili-ties include co:e thermal hydraulics evaluatiers, the development and issuance of core physics design requirements for reactivity control systems, the performance of criticality analyses of the uel storage and handling facili-e ties, and the development of core physics design bases for plant transient;.

RCS/16 7/8'O

,.e

- ms

\\

-

Retrieved from "https://kanterella.com/w/index.php?title=ML20031H452&oldid=4595730"