ML20009E866

From kanterella
Jump to navigation Jump to search
Transcript of DA Hamon Testimony on Behalf of Util Re Tx Public Interest Research Group Addl Contention 55 (Rapid Depressurization - Steam Break).Prof Qualifications Encl
ML20009E866
Person / Time
Site: Allens Creek File:Houston Lighting and Power Company icon.png
Issue date: 07/20/1981
From: Hamon D
GENERAL ELECTRIC CO., HOUSTON LIGHTING & POWER CO.
To:
References
NUDOCS 8107280497
Download: ML20009E866 (7)


Text

f . July 20, 1981

  • ~- .

-1 7 .N

  • c ,4.- -

.,j ~^N

,, _- ys i1 j UNITED STATES OF AMERICA

  • I JUL 3i  ! NUCLEAR REGULATORY COMMISSION .

I

" > d[

2 Ich Oy ;JcidCBEFORIk THE ATOMIC SAFETY AND LICENSING BOARD j e;-

3 - onT3 SI'irl " 4

" . e/

In the Matter of S

$ Qh '3 4 %

HOUSTON LIGHTING & POWER COMPANY S Docket N O 5

(Allens Creek Nuclear Generating r 6

S

'y- JUL g7gggi"

<< gen Station, Unit 1) S 7 *WA v g

DIRECT TESTIMONY OF DAVID A. HAMON ON BEHALF O STON ,[

LIGHTING & POWER CO. ON TEXPIRG ADDITIONAL CONT '

55 RAPID DEPRESSURIZATION - STEAM BREAK 10 11 Q. W uld you state your name and place of employment?

A. My name is David A. Hamon and I am employed as a 13 Technical Leader in the ECCS Engineering Unit at the General 13

. Electric Company.

J. 4 Q. Would you describe your professional qualifications?

la_

A. A copy of my professional qualifications is given in Attachment DAH-1 to this testimony.

17 Q. What is the purpose of your testimony?

18 A. This testimony responds to TexPirg's Additional 19 Contention 55 which postulates a reactivity insertion 20 accident initiated by a guillotine break in a main steam 21 line. TexPirg also mentions a break in a recirculation line, 22 but the governing phenomena is the same in either case for 23 reasons that are apparent in the explanation of the actual 24 tso3 3

8107280497 810720 PDR ADOCK 05000466 T PDR / /

r

m 1 events.

2 TexPirg asserts that the reactor coolant pressure 3 decrease at the point of the break will draw coolant watSr 4 up into the core. This upward flow of water supposedly 5 will push steam bubbles out of the core, collapsing voids, 6 and cause a positive reactivity insertion. TexPirg then

~

7 asserts that reactivity will increase dangerously before the 8 reactor SCRAMS. TexPirg bases this scenario on tests con-9 ducted in June 1970 by the Idaho Nuclear Experimental Labora-10 tories (the so-called Special Power Excursion Tests (SPERT),

11 particularly those reported in IN-1370).

! 12 Q. Will you describe the actual sequence of events 13 which occur after the accident hypothesized by TexPirg?

14 A. Under normal operating conditions, coolant water 15 flowing into a BWR core is at a temperature and pressure

close to the saturation point. Dome pressure for an operating 16 17 BNR/6-238, like Allens Creek, is about 1040 psia; dome temperature is 549*F. Core inlet temperature is approximately l 18 19 533 F; hence, coolant entering the core is approximately 16 F 20 subcooled.

After a guillotine break in the main steam line 21 22 coolant pressure in the reactor core will fall rapidly. (A 23 guill tine pipe break is a pipe rupture in which the break 24 area exp sed is equivalent to two pipe dia! ters.) The s.

1 depressurization rate averages 20 psi per second for the first 2 30 seconds after the break occurs and slowly decreases 3 thereafter. Since the pressure at the break will be signifi-4 cantly lower (initially atmospheric pressure) than the water and steam inside the vessel, the water and steam in the rest 5

f the vessel 'eill move toward the break. Simultaneously, 6

t.he rapid depressurization of the reactor vessel, caused by 7

the escape of steam out of the break, will cause the water 8

in the core to flash to steam rapidly. The rapid change into 9

steam drastically decreases the effectiveness of the coolant as a moderator and, therefore, introduces a large amount of negative reactivity.

Q. How does General Electric calculate the core thermal 13 hydraulics after the main steam line break?

14 A. General Electric's computer code called " LAMB" 15 calculates among other things, the physical state of the 16 reactor coolant after a large pipe rupture. The LAbB code 17 is used to analyze the short-term thermodynamic and 18 thermal-hydraulic behavior of the coolant in the vessel 19 during a postulated loss-of-coolant accident (LOCA). In 20 particular, LAMB predicts the core flow, core inlet enthalpy, 21 and core pressure during the early stages of the reactor 22 vessel blowdown. A more detailed description of the model 23 is given in General Electric Company Analytical Model for 24 l

1

1 Loss-of-Coolant Analysis in Accordance with 10 CFR 50, 2 Appendix K, NEDO-20r36.

3 This cot. has verified tnat after a large rupture, 4 the coolant will flash to steam so rapidly that a surge of 5 cooler water -p through the tortuous paths into the core --

6 generated by differential pressures lasting only seconds --

simply does not occur.

7 The General Electric LAMB code has been verified by g

the TLTA tests described in Mr. G. L. Sozzi's testimony. The TLTA tests were performed for large recirculation line breaks instead of large steamline breaks, but they still demonstrate that the coolant will rapidly flash to steam during a rapid 12 depressurization. For these tests the water level dropped 13 below the elevation of the recirculation line in about 10 14 seconds, uncovering the break. Once uncovered the break was 15 equivalent to a large steamline break - rapid depressuriza-16 tion began and the coolant began to rapidly flash to steam.

17 Q. What do you understand to be the source of TexPirg's 18 contention?

19 A. TexPirg has apparently interpreted IN-1370 to pur-20 portedly show results contrary to the sequence of events 21 described above. This is an understandable mistake. IN-1370 22 (p. 104) contains the untested and unsubstantiated conjecture 23 that:

24 l

1 . . . If a pipebreak occurs, in a BWR, it appears very likely that the resultant 2 pressure relief could cause a significant amount of bubble generation and growth in 3 the water surrounding the core region. At the same time the depressurization could 4 cause the water moderator level in the core region to rise. An increase in water level 5 in the core region would result in a reactivity accident. . . ."

6 The assertion that an increase in moderator water 7 level in the core would result in a reactivity accident is 8 Although the void generation in the core plainly incorrect.

9 may indeed briefly increase the volume of the moderator --

10 thus raising the water level in the vessel -- the creation of 11 large voids will significantly reduce moderator density 12 resulting in a large negative reactivity insertion. This 13 result has been conclusively confirmed by exacting calcula-14 tions and appropriate tests.

15 Q. What are your conclusions concerning TexPirg 16 Additional contention 55?

17 A. Significant negative reactivity is immediately 18 introduced following a steam line break in a BWR, and the reactor begins to shut itself down even before the control 19 r ds are automatically incerted. The Reactor Protection 20 system SCRAMS the reactor seconds after the break occurs by 21 which time the fission rate has already been reduced by the loss of moderator density. Hence, the actual consequences of a rapid depressurization are the opposi'a = what TexPirg P

Q O 9

1 supposes.

2 3

4 5

6 7

8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1

L

r o s ,

Attachment DAH-1 ,

David A. Hamon I received a B.S.E. degree in Mechanical Engineering from Arizona State University in 1976. This degree was  !

accompanied by an Outstanding Graduate award from the Faculty of Mechanical ?.ngineering. Subsequently, I attended the i

University of California at Berkeley where I received an l l

M.S. degree in Mechanical Engineering in 1979.

Following my graduation from Arizona State University in 1976 I went to work for the General Electric Company as l Program Engineer in ECCS System Design. I later worked as a Program Engineer in ECCC Engineering. I have also worked as l a Program Engineer in Fuel Rod Thermal and Mechanical Analysis Unit and in a unit doing unique nuclear analyses. In my present position at General Electric I am involved in the administration of the CHASTE and LAMB / SCAT computer codes, which are used extensively in LOCA analyses performed by General Electric on behalf of all BWR projects. I have also performed several small break analyses and prepared reports for the BWR Owners' Group in response to post-TMI concerns.

l I have also been involved in the recirculation flow valve closure analysis, and in responding to NRC concerns about l GE's fuel clad swelling and rupture models.

i l

l l

l i

i