ML20136H500
| ML20136H500 | |
| Person / Time | |
|---|---|
| Site: | Three Mile Island  |
| Issue date: | 11/30/1984 |
| From: | Lowe W AFFILIATION NOT ASSIGNED |
| To: | |
| References | |
| SP-I-TMIA-018, SP-I-TMIA-18, NUDOCS 8508200364 | |
| Download: ML20136H500 (32) | |
Text
--
- - s j s
'//MY TMI-2 -- 191
)
'F $y* ""fg"*"1 9
f m '
/ 4' 0730 Sunday, l WL's Activities re TMI-2 Accident -- 0830 March 28 April b .
Date/ Time Activity 3/28 0830 MHS notifies WL of TMI request for weather data and says si he understands from Dave Carl there has been a release from Unit 2.
fLG L3hw;'!M 0930 TEP notifies WL of same and that compenTA inquiry b_eing d
made to tower for &ata.L . ...&,
1025 JMV reports that Reppert says there has been accident at TMI-2 and that it is serious. 9 1*.w - c- - .'
. ) W / s . i.: t w a r. ; c' : . /.
1140 Jacle Thorpe calls WL and alerts him to standby for accident control. M '
7 W 1150 WL telecon to IG andMAbealertandrequeststhem to return to office (MA is in Iowa and KW in Houston) . l 1400 WL orders MA to come home.
jus41 i. . J 1631 WL calls Thorpe requesting status and learns that I
l yeremaybesteambindingofthereactorcoolant
)
i l
pumps.,
2200 KW. telecon to WL re status of weather data gsf2f' reduction p -
X/Q estimates. )
i 8508200364 841130 l PDR ADOCK 05000289 1 g PDR l
~ _
1
NUOLEAR BEGULATORY COMM!!3 ION Dettet F 0-
- 5 /2 ,
la the mittar of id " " cfscist E</ I' ExA No--[At h3 /[
Staff .
M ' " Y' USNRC att acoi EENTWO
- ranm_/ --
i.:m,,,r g a s,
- enig Cfft-- /d n r.ur:Ta- '85 AUS -7 P2:30
- a. ,rae:or_- f
,c,,,_ r .TE h l l u/ f '/ . Off;([ Or diirit iM.
__ va:.mt/<*uanef, 00CKETING s 5EPv10 SRANCH 9eserter[_b m#A%-- -
t 9
4
. - pago 2. ,
Date/ Time Activity WWL asks KW and TEP to standby to back up WWL for 3/29 0815 '
I accident control.
0830 Bob Arnold calls WWL re formation of [ vent h alysis and ,
4 boveryh1anning,teamandasksWWLtobeamemberand ,
to come to Observation Center by early afternoon.
0900 TRR reports to WWL re his conversations with about TMI-2 plant status (TRR out of town). p *qf I *
&%- & $b4yu[pll$,i, ^(j ' 3, hW
"----~~==*4=
0915 JMV reports to WWL 'as islant status,8= =
l m ,)$fl> ear O' WNW,
.< e u y; _
s I
i 0930 WWL calls Keaten re meeting at TMI-2 and suggestg
! ~
y% c;y gamma spect analysis of water in 1mateinment including M
silver 110 Ma M" /N
- gnd o' 0935 WWL calls Ron Williams jer* meeting at site.
1050 WWL leaves for TMI-2 via home to get clothes.
1400 WWL arrives at TMI-2 Observation Center and enters ongoing press conference with Senators Hart, Simmons, j
and Schweiker (?) along with Dieckamp, Arnold, Herbein, Williams,Crimn$ns,etal.
, ** s t e , or-
- -y- --m- -, y m -
g- p--+y g---- e w --,--+ew u-wro- --
. , pega 3 !
1 l
- j Date/ Time Activity i
3/29 1538 WWL arrives at TMI-l office building conference room ,
l h& Team is divided
\
for first meeting of recovery team.
into two parts: -"-a N =- ' ;;i: _..'. . _ m .2 Event analysis is* assigned to Wilson, Chairman Ed Wallace g
h 'y P T. Reppert k- ?" e, G. Broughton
.p /i / [ [ ' .k , G. Kunder f' J. Moore
'!g I - ,
Lee Rogers (B&W)
,s b.kf' j.f
.c Recovery Planning is assigned to ju .*,
f .-
Ron Williams, Chairman ghy #f*
T. Crimmins f6/.c *1,# %
W. Lowe Pg'/
/. ' D. Klingeman (not present) s
\ Bob Long.
i
. /)
N.Q4ygdjy.3"';fd[&M0*Al' % ' """l.
1730(?) , T. sj and . Kunder visit control room
/* ., 1 I
- ta u)uiL NML "
At this point iIbecomes obvious tilat the problem is b
not historical analysis or recovery planning but plant g/
y yl p j.in s w ,a -
stability. Operators having trouble with A level control.
b/ftj
- /f WWL and T -r t, rn toy <nferen .e::w.
room.k @." * /~ ' -t 7,
!s..M, &+.mCrimmins ou s ~,. L w ret ja/ ,s ~ .
... ... ..w :t , *
(?) %mpa, 3
,u i r:ressage received Governor Thornberg h e intends 4 u
Jche to evacuate on advice from NRC. Herbein leaves to call f.=.g ~C
/**
sefy% ..
I 1
]
peg 3 4.
Date/ Time Activity Governor;s office whseh Na does not succeed in /
us, 4 " 4-6-ly. WWL follows Herbein out of conference /.).w 4 5 0 0 % T .;; .u s.I D c.
h p , A ,..; n Y o k G R e o 7 F o d r M u n and te_le V m we mu.L-devalar. support for. stability-.in.
M s't " . W J *'* N #
C M te'M >
the control room. Herbein reenters conference room and strongly states this point and requests volunteers.
.7030 WWL and T..Crimmins volunteer and go back to control 1930 room ::: ;:n.
W ~2
+ and Crimmins and-Meeee briefed by Kunder and a
WWL and Crimmins M Ron Tool y tablish termination points for temporary lines to vent waste gas tankr to containment.
AS Tanks are nearing / pressure set point.
v WWL and Crimmins start to get briefing as to plant 0
g ,. % . i...,,.
I
, status from /elinger (sp?), Kunder and others. y:y j? VPy ' g-94 6 tr.;y P. .. . .Le . L LL-le-in pe -primaQ 2300 Operators lose control of pressurizer level. Logan, who has appeared in the meatime, takes charge of the
- b. operat crew. the midst of this prob 1M fire 'N y O
/ .
/ -in Unit i vent system is announced. Fire in Unit 1
~ #
is announced secure about five minutes later. WWL an
. paga 5.
pate/ Time y Activity p Cr.= =
Crimmins ask Logab if he wants 3bem"to leave until 4 ft the problem is under control and he says yes. WL
%9 and Crimmins leara for the office building for about
/
g ! .'
k, one hour. .
7.-
, ' N
\~ 11/ lf(P.&
If I WWL and Crimmins return to the control room and r$
pj toqrldl_-l
- y ~ - - .
/ ,O y (start discussing the problem. '
ftabou th is
.x k . . . .
3 ) time a young engineer approache WWL and pint; euL-
. ; s
> fy,nts A a.
f reactor building pressure e f 2 8 P S at 1400
/ .I gns k /s pys wa q l (. ' 3/28/79. WWL raqueses confirmation via other pressurej tj y,
I m 2 :;li.3 n*EB1r and v4egcontainment u
tr.r. b b temperature ,,,
/
s i \
e
. . p is l .cf j & l '' " .
These all indicate I M b
.g ,
s juJr e . N 5 " * /.ste. ) .
kk S
! hydrogen ignition.y o c ' ""*'
At this point WWL
,u . . wp..
,=";;; Ls that
\ ,
Ys .
bubble g o steam . e to find a F A A gt as4 .
Jm,
, (*gn ais (f*$p k.w icsay a vrLs.. .
~' b4 7 4+ 4' . Jim Moore gets b:.y o 'e fs r,.
(
{ w :. g z .ry;- G .a *ny th t " ! " N U 0
s ,. ,
h/JCn+M s * /*sNr,s W l f '
res re and-rr;; :t:gthat if we can take readings of f;p'*
e e we /m
=t pressurdlevel(eithmakeupandletdownsec untsyn 5 wp+
)# 5 . .; , :
4MMeld solve for the Volume of,,t.be' bubble.% o gan - ee s ****' ~
l he &hacY
}{MM data on IQtt.._}_
t :: i thb; =A = pe="nea. Moore
pagn 6.
Date/ Time Activity and WWL calculate the volume which appears to be between eb Sa 1000 and 2000 cubic feet. WWLopens[ontactwith B&W, Lynchberg and requests they perform check calculations onbubkesizewiththesameinputdata,takinginto '
account temperature and gas solubility.if th:y :: ,
j GaledWW
d- cent
&calculat L ns on.a simpler /l-d8),
~
and r uptde-7 an%9 9(# .
basis L. Lu D ' _..asl . _ . /A A, W' '
{yL. /:Ise a J Ai r /W h eApV wwse M &H $ $ b*
3/30 0245 WWL completes check of calculatio(n. ur bubble size; A
1471 cu f t at 280 o F A 970 PSIA it i_ .aw w a . I;_ .
(see attachment 1) .
. s and Jim Taylor of B&W in response to 0400 Don . . .
k3M.mt*ed.
WWL's previous request report back that the free volume under the head.down to the presure vessel 1129 cubic feet. This means that if the bubblercalculation isu .
M ' e .; < " ,; .
.? 4. **
right there are fgssh gases huMO b M N '
.w h f/AJ/A/ T[4. AS4.M, @ (*C-
~. ,[ circulating in tha ( d d e 8 , f,.. Ib,,
3 )* IL- . ..- .s
/
. or es candy-canes
- hypp
,/. ,
gpe NW
ptgo 7.
e Day / Time Acitivity cont. Another measurement was made at about 0330 but the pressure change was only 15 PSI. The resulting volume corrected to 875 PSIG which was adopted as '
i
. g -
the standard reference pressure was 1026 f m J ed.
Whilte this indicated a reduction in bubble volume, we strongly believed the error in the simple calculations and measurements were aufficient as it was only .
particularly since AP in this instance was whereas it was about 100 PSI . _ . . .
in the first measurement. Subsequent measurements throughout the night were made. The recewnmandation was made that the measurements be conducted using a AP of about 100 PSI. Corrections for change in volume in the makeup tank were made to the measurement of the buble size.
l Early in the morning of 3/30/79 WWL made calculations to try to confirm that the bubble was hydrogen and it could be removed by the recomended mechanism of' f
pzga 8.
Date/ Time Acitivty '
(attachement 2 & 3). This indicated that if hydrogen concentration in the containment had 4
been homogenous and at the lower flamability level when the apparent hydrogen ignition occurred at 1400 on 3/28 then the hydrogen in the containment at that time must have been at least 100,000 SCF (or 54 its volume) . The equivalent amount left in 3
the reactor of 1446 ft atfl0PSIGwouldbeabout 93500 SCF, so total hydrogen production had to be about 200,000 SCF or more, neglecting the hydrogen I
which had accumulated in the containment subsequent to the hydrogen ignition. .
l This is a substantial fraction of zirconium i reacted with water (later estimated to be about 30 %). Since this seemed to be possible, it did not negate the initial conclusions that the l
bubble was probably hydrogen.
f I
I
. - pago 9. j l
Date/ Time Activity WWL then did a rough calculation of how fast i
the hydrogen could be removed assuming that the
, spray rate was 100 gallon per minute that the spray was saturated with hydrogen (then s
l assumed to be equivalent to about .7cc of hydrogen l STP per fram of water) and assuming that the primary water had a homogenaous concentration of hydrogen as long a 1 main coolant pump was in operation and .
assuming that the transfarlacross thej;ef %
to g liquid interface of the bubble boundary was not limit . This indicated ~that,the hydrogen transfa rate could be about .7cc/ gram x 100 gal / min x 785 grams /
gal + 28300 cc per standard cubic foot = about 10 SCF/ minute.
Thsu, under these circumstances it would take about 10,000 minutes or about 7 days to bring the bubble volume down to 0. However, this was on the basis that the simple calculations of bubble size were right-and that the solubility was .7g/cc and the flow rate about 100gpm. It was judged'the flow rate could be' 1
.y -_ y -%-..g9.egymy.e.mi-y-_v. --,m.ey9 _ . _ , - . _ _-y--- - - . 9- -, e.- -_w.w ., m g.. - --
ptga 10 Date/ Time Activity higher and that.the bubble could be smaller._? # # .
- d. y !" , ."& ~l 4'~'J .7* ALfen Si' ' ** * ' ' "
- V " '
At t!!is time WWL requested a thorough examination
_.,- o &
of all reactor vessle head sna : _. q uou drawings a
to see if there were any possible venting pathways.
/ r;wel amb et was negative.
No better mechanism for getting rid of the bubble appeared to exist other than '
/lf yl
.t, venting through the pressurizer even though this / 'F N Y)(3 meant an increase in the radiactive releases.
from _
Therefore/about 0500 the recommendation was reiterated l
that venting continue and that bubble size measurement continue and this was done. The venting continued, the buble measurements continued .
WHL also called Bsw again to reiterate the urgency of getting the check calculations on bubble size.
The urgency felt to get rid of the buble derived from the assumption'that if the buble was still growing (which I doubted) orparitcularlyifthe system had to be depressurized, the fuel would be
-- - , , - - - . , , , , . , . - ~ , , ~ . - - - - - .... - - , , - - - - , . - - - -
paga 11
- l l
- 1 Date/ Time Acitvity uncovered again and subject to further damage and be subject to a possible " meltdown" (B&W Don Roy called l
WWL about 2000 on 4/1/79 when the night shift head-3 quarters had been moved to G. Miller's trailer and reported " good News" that their yalculations showed s
[
the buble had disappeared about 1540 hours0.0178 days <br />0.428 hours <br />0.00255 weeks <br />5.8597e-4 months <br /> on
\
i April 1) . He said that he would send me a copy of I
the draft immediately (which was received at 0044 I on 4/2/79) and that copies of the calculations would ,
i follow later in the early hours of the morning on 4/2. The calculations sent to WWL were done by j Don Nitti.
l The second most urgent concern after the buble i
problem was to find out what the hydrogen concentration
. ) V'O was in the containment. Night technical opeseterT requested that measurements be made as soon as I
possible. The first successful measurement was made from a sample taken at about-.0400 on 3/31/79 and
- indicated 1.74. Subsequent readings peaked at i
4 pngo 12 Date/ Time Actiivity 4
about 2.6% at 0700 on 4/1/79. The slope of this
' curve indicated that if the hydrogen concentration '
l l
kept climbing at the initial rate of about 14 every '
l l
27 hours3.125e-4 days <br />0.0075 hours <br />4.464286e-5 weeks <br />1.02735e-5 months <br />, the lower flamable limit (a 4) would i be reached again about midnite of 4/3/79.
4 Therefore an urgent request was placed on operators to put the recombiners into operation as soon as possible. Jim Henr was called in from California to be sure they were put in right and operated properly. It was not possible to discern at this point how much of the hydrogen,was coming from i
taking the bubble out of the reactor and how much was coming from radiation of sump water.
On the morning of 4/1 at about Oh0 (?) WWL, G.
Broughton, Jack De Vine and T. Crimmins tried to Plot hydrogen build up to distinguish the strength 4
of the sources but the inadequacy of the informatin 4
masked the answer.
l l
f #
l
~ ., _ _.___ .. . - . . . . - . . . _ . . . . - . . _ . , , - . ,_, . . , , . . , . - _ _ , _ _ , ,,. _ m.. ._ _
i Dato/ Time
! Activity i i
3 l
5 In the very early morning of 3/
~
being able to tell where the hydroga was heading called Don Kroneberg (sp Lakes and asked for an avaluation of
- reactor building could stand an igniG
,;...4
.p at the upperiexplosion
( -
limit. About-3, Kronberg called back and said he was e the containment building could take a-1 explosion with a hydrogen concentrati
~ ~
14%. ~ (Note that Don Kronberger was th
! designer of the Unit I containment bui a
which Unit 2 building was patterned.)
The third most urgen problem on w%
started i 0. ' .c.
/tr I ventwastegastankbaq 4
t i
containment since these tanks were n ,
1 limits.
If these tanks / */. 4- e* '
,3 I
doses could have been significantly high f^t m _ _ _ _ . _ . . . .
l i .
ptgo 14.
Date/ Time Activity coordinated this effort and a bypass line was built. .
WWL was insistant that a hydrogen flame arrester be placed in the line if one could be obtained. Two 1
of them were obtained and one was installed. WWL also strongly advised that the valve lineup be 4
completely described in a procedure and be carefully i.
checked before venting commenced. Joe Henrie
's #
helped check the valve lineup and ,/We**
that the waste tank gas pressure not be allowed to drop below 15 PSIG so that flows in the ;
pinch point would not [0 This would prevent any flame front from moving l
l back into the containment. WWL also requested l 1
i a sample of waste tank gas atmosphere as soon ,
i l
as it could be obtained to insure it was not an
.i explosive mixture. WHL also requested Mountain f
Lakes to make a careful check around the exit of the waste gas decay tank into the containment to i
besuretherewerenotsparks[ within about 1
1
. l
. l
prgo 15.
Date/ Time Activity a 12 foot radius. Joe Henrie also reconunended that the tanks not be bled down at more than 3 PSI l
i per hour, to promote mixing within the containment. ,
l on the 1st -attempt to sample the waste gas l
tank atmosphere the radiation monitor indicated 1
there was a leak .
, The leak was found and fixed. On the second attempt to sample WWL was observing the helicopter hovering ,
over the auxilliary building when it suddenly
- landed. Urgent inquiry disclosed that it had run out of gas. WWL strongly recomunended not refueling by another helicopter but by bringing in a truck.
l l
l The third attemt revealed a concentration of 584 hydrogen and the rest nitrogen and negligible oxygen. After the first attempt to vent the waste gas tank valve 341B which had been previously opened to take a sample would not open. WWL recommended l
l e<- e e ,
,, , _ , . , , - - , , - , . . , , . , . - - - , . , - - , , - , , - - - - ~ , . . . , ,
~ '
p:ga 16.
Dnt0/ Time Activity they try again, turnging the valve both ways to be sure it was not already open or stuck closed.
Several attempts were made.
, e .s s'.s yo.* I Y -
Then /~l /'I * *' '
V 143A. p t /W
.4
- As the venting commenced, the wind speed increased to i
l 50 to 60 mph and the noise of a bus rolling over a plywood cable protector (or of thunder) occured at the same time. The winds prevented helicopter i
1 measurements from being taken. The venting at about 15 PSIG. They were vented faster hwoever i
j
- than 3 PSIG/ hour.
Another urgent problem was to try to get a i
backup way to measure pressure levels since one I
instrument had failed, another had been erratic 4 ,
1 l and only one appeared to be reliable. We were aware i
that the pre-amperage for .
1 of the basement floor and that the water level was about 3 - , +3 . Nonetheless, we had to continue venting to get rid of the bubble.
f
. , . - . . --,,,-,,,,n -..n. - , - , - , , - - .
, _ , , , , - - . , - , - - . ~ _ , ,
pego 17. J. --
Date/ Time Activity or we would have to get into a solid mode in the primary system. Three basic approaches were taken j
i to l
, 'l
- The first was to put the gauge (03000 PSI) l l
on the primary system sample line. The gauge could reduce the pressure in the lower water j
space in the pressurizer with one valve setting. With anouther valve J
However, a high pressure pump-had to be rigged to fill the vapor space leg with water, so that there 1
was a water-filled reference leg of known density.
Apparently the water leg could not be filled because of the apparent leakage rate through the 4
which exerted on the water leg and sucked the water out of it.
Attempts were then made to -
i to pump water at a higher rate to overcome the
)
Bernewly (sp?) effect. Dick Wilson suggested an l 1 .
pago 18 Date/ Time ' Activity alternate method. That was to measure the water side pressurizer with a gauge and to measure the gas space pressure using the gas space s temperature sensor and derive the pressure from this i
I J assuming the gas space was saturated steam. This method gave a good correlation but it is only valid
\
if it is known that the gas sensor is seeing gas not water. Two other methods were tried to measure the pressure level by alternate means. One was to keep a running mass balance in the primary system assuming that the measured 2% gym leak rate stayed constant and by correcting the levelfor thermodynamic coupling condition. A J. !
1 test of this method over several ,
i for short time periods (e.g., a hay) one could 1 probably keep track of the level within a foot or so. But any l would lead to errors and one would have to go solid 2
4 s-
P2go 19 i
Date/ Time Activity l IN THE PRIMARY system. The third method was to
(
utilize the different heat transfer characteristics ,
of the gas or water in the pressurizer as they
. J.
affected the time response of the temperature sensor. I never saw a valid test of this method-although there was said to have been one.
Another urgent probeim was to keep enough pressure heaters availabl so as to control with pressurizar steam bubble. I asked Bill .
I Smo~ss (sp?) to bird dog that problem. His first i
- report indicated that of 1100 kw of poter (?)
capacity only only 1600 ,
of the could be received. He believed l .
this was because you could turn the heaters on but they would trip out. Therefore he recommended putting 1
. This was done but
didn't work. Another idea was to charge for 60 - 80 adlerage, was done but didnt' work.
j
- - . , - - - - - - , - - ---,nw,- - , ,, -,--e- n-- --,--
paga 20 Date/ Time Activity As far as I know we received very few of the heaters which went out.
Antoher urgent problem was to prepara a backup
>\
i procedure for backup circulation in case we lost the coolant pumps. We were concerned about thermo-
~
4 dynamic couplings. B&W replied that the surge
~
condensors on the meter starters had a tolerance .
8 They predicted that the total would l to 10 . rads.
l
- be exceeded on the morning of 4/1 (?) . This calculation,was made, I.believe, on the basis of assumptions concerning containment radiation levels .
i In case of the loss of pump 1A the problem was first
- to start pump 1B and if it didn't start to start 2B, and if it didn't start, then to start 13. If
' either 1 or 2 B had to be used the B steam generator
' would have to be steamed and it had activity in it on the second side. Despite urgent efforts a satisfactory e
solution took about 7 - 10 days to i
, develop. The first one was given to the control
pago 21.
1 l
Date/ Time Activivity room verbally by me following the outlines of Bsw -
repeated calculation and in effect said to keep the steam general water levels high, do not abort
,e attempt.at natural circulation unless at least 2 full.termperature thermocouples exceed F or more than 10 exceed the saturation temperature.
[
The c'aution was that the AT (T hot minus T cold) would probably go up very slowly and would
. 30 - 50 F and stabilize. So the operators would have to hang tight and wait and not take precipitous l action out of apprehension during this waiting period.
Later the limits were raised to 1000 F or,two thermocouples and thermocouples above saturation.
WWL advised that pressure must be kept high in order
> to prevent formation of gas bubbles (see WWL's final notes on this subject dated ).
Still another problem was that the iodine levels n the auxilliary building were reported to be about 1000 time MBC.
At about the same' time measurements J
___-___.m___- _
paga 22. J Dato/ Time Activity were reported to technical control indicating a break through on the auxilliary building charcoal filters. .
1 WWL was requested by Ron Williams to .
,l 4
I called Bob Arnold and told him that I would need j someone to handle this. . He sent over Jim Ramshaw (sp)
I told Jim to get hold of B&W and Keith Woodard and find out what computation to use for a sodium-hydroxide sulfate to put in the auxilliary building sumps and to spray on the auxilliary building floors and walls where there had been water potentially containing iodine. The requests were placed and within about 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> we were ready to dope the using back-rack, garden-type sprayers. But a hold order was indhed pending.further checks about the _
effect of the spray solution on the equipment. My view was that such a spray could not damage the equipment especially.if it were confined to floors, As a second action I called
> walls and sumps.
pego 23.
Date/ Time Activitiy Mountain Lakes and asked for DEC on an urgent basis i
i of a parallel charcoal filter system. The request was placed with Don Kronberger. About four hours s.
later Don reported that it was not possible to do such a system. I told him that that answer was unacceptable and that he had to get back to the design i and get a solution, which he did. It turns out that 4
they had to put the charcoal filter in a series.
l The decision was also made, but not in my group, that the existing charcoal filters should be changed out. At about 0600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br />. on Palm Sunday Morning, I got a call from NRC representative stating an IC5A .
airplane had just arrived from Richland, Washington
! with a consignment for me and that another was to e
4 arrive about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> later and another about 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after that. He didn't know what the cargo was.
~
I assumed the cargo was charcoal filter elements and thereofre turned the problem over to Jim Rimshaw i
e
, , - , , - _ _ . , . _ - , ,w--w, ,-.v.,wm-- -- - - -w g- , ---,,-,w ..-e,-++-,-g, y - g y v
, DatO/ Time Activity who turned it over to the next shift.
Another important problem which arose around then on 4/1 was an approximate 40 F increase on the l average in the fuel thermocouple temperatures. I ;
~
was afraid of fuel shifting and I called n Roy
./
at B&W with an urgent request. to analyze
\
and to report on its implications. B&W started taking thermocoupling readings every hour. Prior to that i
I requested we receive thermocouple temperatures every four hours. Don reported back that it looked ,
like the termocouple tempratures simply followed the l
water temperature and he didn't think lodged any peo&lem.
As a double check I asked John De Loma (sp) t to keep a plot on selected thermocouple temperatures (attachment ) which he did until he left.
About the 2nd of April a tsasic operation plan was developed -in coordination i:iith the Industrial i
Advisory Group which had the fc11owing steps aimed e
- , -. , - y
~
p2go 25 Date/ Time Activity at decastication of the primary coolant and cooling it off. The basic process was to start at a Primary pressure of about 1000 PSIG, hold the temperature at about 2350Ftand reduce pressure slowly to about 300 PSIG, keeping the temperature constant and then raising the pressure back up to 1000 PSIG.
While this was being done, maximum letdown flow would be maintained to try to get dissolved gas out of the system.
Precautions included stationing a sound monitor temperature to listen for degassification and to monitor main coolant pump amperage for any indication of cavitation (e.g. erratic amperage readage). The second step was to cool the primary system down to about 230 F while on a generator through out stabilization if possible to bring the temperature i
1 below the boiling point. The first maneuver was carried out successfully and pressure sample of indicated that the gas O
I e
^
p'ago'8@~
. Dato/ Time Activity concentration had been brought to about'60 cc STP per kg of water. But in the second leg of the maneuver we got stuck at F and couldn't get any lower.
Stan - and I discussed this matter thoroughly and concluded . water in the steam line. At the same time inquiries were made of B&W as to why the. temperature couldn't get lower.
and reported there were several_likely candidates. Bsw reported that 1
! the pressure drops with the paths we were using 1
and that we couldn't . That 1
pressure drop was about 15 PSIG to the condensor and .
a 15 PSIG breakdown pressure into the condensor.
Then condensor was then operating at about 1 PSIG.
f one thing that misled us was that the steam pressure gen,erators were reading about -1 PSIG and this quite apparently was inaccurate. ,
Later other pressure readings were taken and the temperature was brought down below the boiling point. Later I understand
, - _ . , _ . . _ . , . . , . _ . - .,_-m.. .
,-, -,- y r
prga 27 Date/ Time Activity that as some of us expected, water built up in the ~
blocking some of it and the temperature rose again.
i s While this was going on there were three principle plant models underway initially under supervision of our group. The first one'was to construct a heat sink.,(e.g. piles, heat valves, heat exchanger and -
pumps) between the feedwater-lines and steam lines of B steam generator so that it could be taken solid with this auxilliary cooling loop and used as a water per water . A second model on a slightly lower schedule to do the same thing with generator A after generator B had picked up the heat load in the solid mode, but to use more permanent equipment which had more delivery lines on condensor A.
Another model was to make a short circuit heat exchange ;
i l
system connected to the decay heat outlet and inlet systems. Another model was to place charcoal filters in the condensor exhaust.
4 i
..--,----.c,.w-me...-.....---
- pago 28 Acitivity j Date/ Time
\
One of the problems adressed early on was to ,
I identify'all equipment and valves within the containment building which might b'e subjected to failure of operating status' due to radiation, humidity or water and which were needed to be in certain operating modes in anticipation of futura .
We asked ML to provide such a list. We checked it, operators checked it and the equipment was put into .
the desired mode, i
Another problem was to provide emergency bus power to the pressurizer eletromagnetic relief and block valves. Cables were laid out for this power supply but final connections were not made
- l
! . Another major effort was to provide emergency l power for all equipment needed in case we lost off- 1 site power (there was particular concern about a P
helicopter crashing or sabotage). Auxilliary .
sets were moved in and' connected. This operation was not under the control of Technical support but
'e
pcga 29 Date/ Time Acitivity others. Another operation accomplished by others was the delivery of very large tanks. There may have been tanks 100 fett in diameter and 50 feet long. Some of these tanks were to be linked up to take excess liquid or gas should such excess develop.
Another urgen problem was to to include
! letdown flow so that the concentration of disolved gas in the primary cooling system and so that the baron concentration could be contained.
The letdown when the designated rate was 150.GPM. Some people suspected boron crystalization S
but I did not believe this was the cause of the reduced flow. Finally reasonably acceptable measurements i t
of letdown flow were obtained using heat balances and l it appeared to be about 20 gym. I suspect fuel or fuel cladding or other debris to be partially blocking
- 1et down in the line. We looked at back-washing but l
I e could find no way to do it. We also shut off the i
' ptgo 30
- Date/ Time Activity secondary side coolant in the let down flow coolers meanwhile having by-passed the letdown flow lon exchangers and isolated them. We got the temperature a
up to about 190"F but this didn't seem to increase the flow very much if at all. Apparently the coolers are under water.
Anotherproblemwhich%dnot feel strongly about was to vent the makeup -
b .
4 tank to the containment. Apparently there was a belief that the makeup tank or something down-stream was We did note that every time we built up pressure by shutting connections l
and then when we vented it, radiation levels would i
increase in the auxilliary building. However I was against venting into the containment until.
l .
so that we were sure we wouldn't i
blow it up. This was because the makeup tank was to our operators and I felt the tr 's off of I
[ .
,-,-w.g,m. --- -- -~ -.-,-,,__--,.,,-g--,. - - - ,n--n-m-,,-m m--,,m , , - - - . g a--, ,ym -p, ,,y w,,-.,,,,--wep-., gw-, ,.,,-y.,,,,w ,, -
7 i
Dat3/ Time Activity slightly increased releases of radiation was preferable to .
When we finally got a sample of the makeup tank atmosphere, it looked like it was in a non-explosive Tange which we would expect. A
\
very important but longer range problem was to degas '
.)
' the primary system fluid. ~The basic operation for l
degassing this was to maximize downward flow to the makeup tank spray and to keep the make up tank .
4 even putting vacuum pumps on if necessary and to de-aerate the makeup 4
i tarik water. It was my view that the system not be depressurized until the primary system concentration was below 15-20cc/kg, and that even so as B&W had pointed out there is a danger of non-homogene ous causing system binding in circulation. Therefore my only concern was that one should not depressurize l
until one had the short circuit decay heat system completed so that one could without depending I
on ,
and natural circulation. i i
l
! 1 I
, - , - - - + - - .- - - - - - - , - - , -- , . - - - - - - - - - - , ,,,w ,-