ML20138C242

Forwards Comments & Opinions Re Outstanding Issues & Unresolved Questions Concerning Hydrogen Control for Facility.Supplemental Info Will Be Provided on Request

ML20138C242
Person / Time
Site:	River Bend
Issue date:	09/10/1985
From:	James Shepherd SANDIA NATIONAL LABORATORIES
To:	Savio R Advisory Committee on Reactor Safeguards
References
ACRS-CT-1820, NUDOCS 8510220385
Download: ML20138C242 (6)

Text

PD A M O Ks ~~~

CT- /$ Qb ya s $

~

Sandia National Laboratories Albu1uerque, New Mexico 87185 September 10,1985 hir. Richard Savi h.

Senior Staff Engineer t r.F 7,M ON U. S. Nuclear Regulatory Commission E.G.Ma...uS, U.S N.R.C.

l Advisory Committee on Reactor Safeguards

,; g gg 1717 H Street NW Washington, DC 20555

Dear Dick:

5 Enclosed are my remarks and opinions concerning outstanding issues and unresolved ques-tions related to hydrogen control for the River Bend plant. As I mentioned in our telephone conversations of Sept. 3 and 9, my previous involvement with this problem has been more concerned with generic combustion issues rather than with plant-specific problems. Ilow-ever,I feel that many of the generic issues that remain unresolved are also present in this case.

Afy opinions have been drawn from my experience as an NRC consultant during the last five years. As part of this experience, I have been involved in discussions with the Hydrogen Control Owners Group (HCOG) and their contractors concerning combustion experiments and modeling of diffusion flames in the wetwell of BWR hfark III plants. I have drawn heavily on this background in evaluating the submittals of Gulf States Utilities (GSU) and the NRC's responses. For this reason, it cannot be said that I am a " detached observer' '

although I have attempted to be as objective as possible.

I will be glad to answer further questions or provide supplementalinformation if necessary.

Sincerely,

/

OY jw-seph E. Shepherd 8510220385 850910 PDR ACRS Fluid hiechanics and Heat Transfer II t

CT-1820 N

Division 1512

Enclosure:

as stated.

(g(Nilwb '

/

p

Air. Richard Savio Septembzr 10,1985 l

River Bend Hydrogen Control Issues

1. Equipment Survival Analysis: The equipment survival analysis carried out by GSU involves a number of assumptions about mode of combustion and hydrogen release rates. The key issues are:

(a) Method of Analysis: Both the 1/20- and 1/4-scale experiments have demon-

.strated that the dominant, perhaps only, mode of combustion is diffusion flames above the suppression pool. The analyses are performed only for repeated de-flagrations, once widely believed to be important but now thought to be less important. The analysis is therefore incomplete and should be extended to a diffusion flame combustion mode.

(b) Hydrogen Release Rate: GSU and HCOG have consistently used mechanistic computations of the core oxidation to predict hydrogen and steam production and release rates. The simulation used (BWRCHUC) typically predicts total oxidation fractions less than 50% Despite the uncertainties in the simulation model and the probabilistic nature of accident sequences, HCOG has refused to consider any parametric variation of hydrogen flowrates in either experiments or analyses. This makes it very difficult to assess the safety margin of equipment or the range of possible accidents that the plant can handle.

Using the GSU/HCOG approach to defining hydrogen flowrates and the recent experimental results in the 1/4-scale facility,it would be difficult to argue that any threat to safety-related equipment exists even without the mitigating effects of sprays. I have personally analyzed data from one of the HCOG scoping tests (S5.01) and the peak gas and surface temperatures are very. low. The flowrates of hydrogen and steam used in those tests are prototypic of those that GSU is using in their analysis.

For this reason,I feel that the most important issue in hydrogen control for BWR's is the method used to specify hydrogen release rates. If the mechanistic simulation results are to be used, a very close scrutiny of the model assumptions and input parameters should be made. Some of the more obvious issues are:

i. Documentation of comparative studies.

ii. Justification of the oxidation cut-off temperature, iii. Consideration of oxidation of the B C in the control blades. Recent simu-lation results by independent researchers indicate that this may be a sub-stantial source of hydrogen.

2. Accident Sequences: The accident sequences used in the analysis of hydrogen events do not consider some factors that may be crucial to determining the severity of the combustion event that occurs following the hydrogen release. The main issues are:

(a) Loss of AC Power: Many of the degraded-core accident sequences examined l

at Sandia involve a loss of AC power. Under these circumstances the igniters l

s L

Mr. Richard Szvio Sept:mber 10,1985 will not function since they are not powered by any of the backup systems. Are l

there any plans for backup power of the igniters?

l I

If the igniters are not functioning during the hydrogen release and power is restored later on, what pressure and temperature transient will occur from the E

resulting deflagration?

i (b),Use of ADS: If the Low Pressure Injection System is not functioning it is unlikely that the plant operator will activate the Automatic Depressurization System. This implies that the hydrogen release will only occur at one point in the suppression pool, at the location of the Stuck Open Relief Valve that initiates the event. What are the consequences for equipment survival at that l

point in the wetwell?

i k

6 1

I

)

i i

1 l

t f

i Y

+

T I

r

~l' 9)t/es A 4/,viesw /

<< >? < < n s assu i n </> re~

c<~ 4 /

r

. I

{e y fuer Agxc/

J f.14 ap4nc/

_Gme//s 4/n f ex n /

Zds h

jll

,4cie J reAen ~<Me 555 Rii cxn> m,

.o c A rn f.

==

ZZZ enn gj

/

H(D 6

}6 ses /e e,v,w n e 7fr suel cas/ysi.)

(f.

hr4 ///r Jar </esizy c a si si s / de s, tsp Net f

- /)no'e I

&>nrf, 'il des / 7fs u.1/<-,., su 4,

/Liefso,sch w'Jc f (L<rp o r / / 4 w s ('y ws)

Du}+

6

//r /ax / -no-s/a,. ~ <f y%e/sswL'cs 5 ding,

is aj,/<c/,

per/ecf saliuf L

tz+c c,, psie / Je nc4res eet, fo, M&

r eo,sie, com/Jun M s on epuipiresi/ sorokobi/,Yy spry 4f hg g ccein,Da Jtitcl dy d /r C//rrt;rs&sJrly erke<<2 Ye e///g,$swdn 4sufys/j

'skrc/ ooc ins /friJr jk c ris r, b.

Orts /75t'cV_..Sfudkr y

i NW 2t4L}

,(5 h4ef' A

l'!ctd/r %

7"

^

/

o/r-s---Jwey/e

{/sa mfe

,(sy4ry ] As

~

44 h 6-hwa..

g,-ia<= 6 ~x

/

haf 7fu< J/-c~

Qny, % fax est

& p fe w f esc <-~f drit n

l s km (d.~rfe q <-u d [y&w

<_ eaj.

ws w <~g iii u$ sad 4 a f~ 44 g se.--4 a ~xa &

vo -

a g;

/4 cs L~M n i

-4 y; s a y

H (s6 ny&,oK fiAV.

sun d m sun

/4 y.ia/s,2y s/ ff-se A ex u l e x 4 y 4 4 fe d

(

//J<d' tus freusfe jsss c rs nm,w.,nss,~~a n d

de a.e</cd n

"/ex4hs4 "

i a/hejk em 6&a r: <~pa n a, a<&-

myN?'.

74 ~e n 7 % p >a w <

wn SA i)

1. < % % A e n 4 - d a s &J p k q no 6

4e a

40 A$

,ffr-Jcx de

/c d ji/

ll uke )j 4.-c (JA%t k t A S / N et k

(

ua ner st

~

& b'6dzr

.a ;. !c JA~f' b

C/hbtwfeb9e.

s' cA r.r.

~

-%_ 8e es a.x q' (A<se p>v 46 9 Wh impn %7' fo an a wwA g

i

/.ss V rn tf w

-Qu h fo w-- s m y

e A 4

1 sy~%

a ~y lll

-pasn44 4 fat a-M.

=in ass 222 een

(.

/

.