ML19220C681
| ML19220C681 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/15/1979 |
| From: | Mccreless T Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| Shared Package | |
| ML19220C682 | List: |
| References | |
| ACRS-SM-0079, ACRS-SM-79, NUDOCS 7905140026 | |
| Download: ML19220C681 (42) | |
Text
_
fl cc.s-s/q 99
- y 4/15/79 h
Ll'ff SITE VISIT 'ID THREE MILE ISIRJD N"JCLEAR PLANT APRIL 9-14,1979 T. G. McCreless During the subject visit I was joined by Mr. Etherington (April 9-12),
Mr. Ray (April 10-11), Mr. Michelson (April 10-12), Dr. Catton (April 12-15) and Mr. Fraley (April 12). Previously Mr. Sender, Dr. Lavroski and Mr. Wright visited the site (April 6-8).
! believe the recovery operations are proceeding in a very deliberate man-ner, and I was impressed with the qualifications of the people that I found available. Needed equi;r.ent was on hand. When additional needs are identified, there is true effert to obtain items in shortest possible As an exa ple, ten railroad cars to hold liquid wastes were obtained time.
within 43 hours4.976852e-4 days <br />0.0119 hours <br />7.109788e-5 weeks <br />1.63615e-5 months <br />; the Air Force has been flying radioactive samples for analyses.
CURRENT PL'd:T STA"'L'S (See Attachment A)
Degassing by cycling the primary system pressure frco &l000 psi to A350 psi and venting of accumulated gases has been completed.
Presently the reactor temperature is being decreased by increasing steam flow in Steam Generator A by increasing ficw in the second-It ary system by adjusting the turbine bypacs valve setting.
the reactor temperature would be lowered f rom was planned that 99 242
'2 9 051400.7 C k;;g f :
O
4<.
.. a 280 F to 230 at a rate of about 5 F/hr.; the primary pressure would be maintained at 1000 psi. This was started at 10 a.m. on April 13.
At 8 a.m. on April 14 the average core temperature was 250 F and was holding steady.
It appears that 250 F may be a physical limit for the current system, but it is not thought daat this will present a problem.
One of the three level sensors of the pressurizer has been lost, and another seems to be unrealiable.
A Heise gauge has been in-stalled and is being cslibrated to measure pressurizer level.
As reported on April 14, the calibration was delayed because of a con-tamination of the area in which the gauge is located by a leaking chemical sample.
Another Heise gauge is to be installed to measure containment water level. This gauge will be installed on some associated piping that is located in the Auxiliary Building.
. Pressurizer heaters are about 60% operational. This is expected to be adequate to uiaintain system pressure for all planned con-ditions.
It is thought that failure of a portion of the a7 99 26a
. pressurizer heater is due to moisture in the cable, and it is planned to increase the current in that cable in hope of drying it out.
The exchange of charcoal filters in the Auxiliary Building is currently underway. Work in the area where the filters are located requires airpaks and replacement is time consuming.
Forty-eight of the core diermoccuples are still operating. The high 5 readings at 8 a.m. on April 14 were:
350, 348, 320, 307, and 282.
(The cold and hot leg tenperatures were 250 F at this time.)
Ebur recirculating pumps (two in each loop) are available.
Previously one of these pumps was tripped by a signal that suf-ficient perp coolant was unavailable. The fault was determined to be in the detector and not in the pump.
Primary coolant samples indicate that boron concentration is greater than 3000 ppm.
B&W analyses indicate that 3000 ppm boron is a safe concentration for all core conditions.
t 99 244 s
r-
, The analysas of coolant samples also indicate extensi'.e fuel damage but little core melt. Core flow restriction has in-creased frcn a 2h9 of about 1-2 psi to about 18 psi.
It was believed that Steam Generator B had a sizeable leak.
Recent information indicates that it is not leaking or at any rate not leaking much.
It is believed that the pressurict: relief valve is still stuck open as thermocouple read!.ags near the block valve are higher than expected. "his could occur only if some fluid is still flowing; fles would not be expected if the relief valve had reseatei or if the block valve was fully closed.
The Recombiner is not operating. Shielding is being positioned so that instrument can be worked on.
Venting is not being done.
RECCVERY CFERATICNS The primary ef fort at TMI has been on recovery and very little has been done to establish what happened to cause the accident and who is responsible. This is true both for the NRC Staff now at the scene and for the licensee and his industrial supporters.
99 nar44J r
i f
_5_
Senior NRC Staf f at the site include Harold Centon, Roger Mattson, Victor 5tello, Tom S'ovak and Dick Volmer, all from NRR, and Boyce Grier fron
'5E.
I estimar that in all, NRC must have 100 members on t."e scene. Two are alwafs present in the control room to mon-itor cperations; about five health physicists are used to follow plant repairs and modificetions.
I spoke with a senior man from I&E and asked if they had encountered unexpected problems.
He explained that he had thought, prior to this accidnet, that ILE effort would be concentrated within site boundaries and that some other agency would be measuring offsite doses.
As it has turned out, EPA is monitoring doses outside of a three mile radius and NRC has had to monitor the closer distances.
This has forced the NRC to increase the number of instruments and personnel needed to cope with this accident.
The industrial task force established to conduct the recovery operations is shewn in Attachment B.
It is apparent from this attachment how broad the industrial participation is with representatives from GPU, Duke, Metropolitan Edison, Ccmmon-wealth Edison, Cembustion Engineering, B&W, Burns and Roe, and i
e
_s_
EPRI. Not specifically shown in the chart but also participating are representatives from Westinghouse, General Electric and Sechtel.
Joe Palladino is participating as a special representative frca the Governor of Pennsylvania.
Pen Rusche is coming to head up waste menagement efforts. The Industry Advisory Group under Milt Levenson and/or Sol Levy is called the "think tank."
Attachment C is a list of tasks currently underway.
RECOVERY PLAN (Attached is a cocplete description by Dr. Catton (Att. D)).
A summary of the recovery plant is included at Attachment E.
This plan is depicted in a FERT chart which is included as Attachment F.
In brief, the recovery plan calls for:
1.
Some repiping of the secondary side of Steam Generator B so that when that loop is used and should leakage in Steam Generator B occur radioactivity could be confined and not released from the plant. This repiping involves the use of the turbine coolant system. When this modification is completed (about April 20),
Steam Generator B will be taken water solid and A will be iso-la t ed.
2.
The secondary side of Steam Generator A will then be modified by rerouting piping to the condensor to improve its heat transfer o
- 7, L4
~
, characteristics. hhen modifications are completed (by then core temperature is expected to be about 100 F) Generator A will be taken solid.
3.
The preceeding actions are scheduled to be completed by April 29, and at that time the reacter circulation pumps are te be tripped and natural circulation established in the primary system.
Pri-mary pressure is then to be reduced to cbout 100 psi.
4.
A further modification of the secondary system is planned so that the secondary side can be pressurized to a level in excess of the primary even if EPIC pumps are called for, to assure leakage, if any, is from secondary to primary.
I have included as Attachment G an interesting study made by B&W on natural circulation alternatives for long-tern core cooling. B&W states that the preferred mode for natural circulation is for both steam gen-erators to be water solid, with seccndary flow of each generator greater than 3000 gmp. This mode will produce a maximum core flow rate of more than 800,000 lb/hr, a minimum core 21T of less than 30 F, and a minimum reactor coolant average temperature of Al20 F.
The use of a single steam generator is expected to increase the reactor coolant temperature by about 5 F.
nao l'?O F~
1 Additional work is being performed Dy Westinghouse in designing a new decay heat removal system and a waste processing system.
ACCIDE.NT SCE9ARIO As I mentioned previously little has been done to establish the exact scenario and to establish responsioility.
I discussed en April 13 the accident investigation with the Chief I&E investigator for this ac-cident, and he said that they were just starting to gather information and had not wanted to interfer with recovery operations.
The techni-cal members of the investigation team were not available to me as they were in Bethesda at that time.
Some preliminary work has been done by the Industrial Advisory Group in crcating an accident scenario so that they might better understand the present condition of the reactor core.
Mr. Michelson discussed with this Group the accident scenario as he saw it.
I believe that Sol Levy was in general agreement with Mr. Michelson's proposal.
Mr. Michelson discussed plant operations and responses with a plant operator, and he believes seme of his earlier questions have been answered by these discussions.
99 249 e-
, Mr. Kaufman, INEL, prepared the Industrial Advisory Group's core assess-ment.
I have attached a copy as Attachment H.
His analysis includes a scenario that starts at one hour af ter the initial event.
He concludes from a study of ion chamber signals that significant core voidire oc-curred between 6 and 7 a.m. to do significant core damage.
Dr. Catte. has prepared a summary of core damage (Attachment I).
COMMON QUESTICNS 1.
Initiating Event A condensate pump failed. This lead to loss of feedwater to the steam generator.
2.
Sump pump activation Sump pumps are tripped by ESFAS signal.
Reset of ESFAS signal permits the sump pumps to operate on sump level.
3.
Containment Isolation Containment isolation occurred at approximately 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> hen contain-ment pcessure reached 4 psi.
4.
Activation of HPI pumps and Containment Isolation The liPI pumps are used as charging pumps and do not initiate contairnent isolation.
99 250 c-
e
. 5.
Containnent Isolation Containnent isolation is only actuated by containment pressure.
6.
Feactor coolant Orain Tank Pumps (different from sump pumps)
These drain pumps are on tank and transfer water to radwaste.
oOo Attachments:
A - Current Plant Status B - TMI Recovery Organization C - Industry Advisory Group D - Recovery Plan - Dr. Catton E - Base Case Cu- :ary F - Base Plan CPM G - A Summary of Natural Circulation Alterna-tives for Long-Term Core Cocling at TMI-2, by B. A. Karrasch, B&W dtd. 4/10/79 H - TMI Core Assessment by N. Kaufman, dtd. 4/6/79 I - Status of the Core - Dr. Catton nri hh d3I r-s
0%
- t T i M E_.
4.st.-79 T'AT C O
g 2.6 I Rc. Pure D ATA I
N5M-T fp/ -b 3pt g er, nA
, - m ~ ~ y" m,
c.n a
m.s h# 85
gJ{
[25.7 LUU
_Z.t u
' " ' " ~ "'
h ~'
k,
' 33B II*C 'PE E i
Q h bod
,a e U
Y
@',3 7
x to W-12i.2
- p_
- (> -
%-.*r
~ ~- l 1
4 ?
o egess PRC k T
~
Mt D!Akdilllil*a
~
g yg>,
GP= 'E f, iML'i3 a eh7
_740 P
t W
(M Q
ymm 3do Ib 4.
m af8 21 5 M
,,3 m ^ s.
2.
-.3 9
g 6
ux p
m.,,
[i
/
H p 7, j
tNCOR6 TEMP 5
- E 12dilOE N
h{
g%
ff Gw.J I
Y l
14 4. f, Mu g,3
. y e 1 1 4 T
- 7 f-i
'O
I-
'I
' +
- 7Ui/~~ IK TANK
--- h
/
L.'
L m3
/
g W. (n,
- 7 Ib
%],.f-F
r$PtL P<
s
-T C P1 t' C.
D T'k[
G g@ L_
TOTAL iv IN STA Lt.E n M em nu status
---N g
(
- t. essgr ega_33.3
%N
~
~
_E,P
? eto Aposvio d
~
t-(, or {
UKr.K5 ctosco 3! _
g g
g
- 3. H % cAtc g,
et
_, _c E
e F C+J ERG 7,to,
- M^'g,3q 37, SC G
~
T-
~
e.oe4mcs, rtop e.
1392.
+ 4 d" PKW_ ' ' ' TMS o
u mm g
.]
t e
CPU II. DIECKAMP DEP. W.
LEE-DUKE hT h!ET-ED)
GPU OPNS. IMR ADMIN 6 LOGISTICS PUBLICfGOVT. AFFAIRS 3.
FABIAN R. art:0LD, 11. ROBIDOUX R. IlYDE B. LEE TASK MCT/SCllED F.
STERN-CE INJUSTRY ADVISORY TECil WORKING GROU?
GROUP OPNS TASK MGMT NRC LIAISON TECl! SPT IAG IIVENSON Br. R UASTE MGMT ESU
';R C 3@
CPU TECll SUPPORT MET-ED PLA!Tr Ol'NS WASTE MGMT.
PLANT MODIF ACRS F.
PAIJtER 1
B. COPEAN I
R. WILSON J. IIERllEIN R.
PAVLICK l
LIAISON DECAY SYSTEM I
MOD CONTROL L7 ELEC Pl!R L
DATA 0FF NOR'tAL PLANT !!0D TECll SPT I
GPNS ANALYSIS REQTS SilIFT OPNS
-INSTR UPGRADE C
E
_ RAD WASTE MOPS S
NO l-g S
9
- 11. P.
SilIFT SEC MAI!Tr d
I N
OPNS g
N b
t-i
/
(
NRL G
vs
~M
{
INTERFACE N*C I'!ERFACE.
S h7G
f ndus t ry Ativ i so ry Grpgi, Task Description Priority Status /11 ate Due Coo rd.
Deteraine method of finding 1+
ASAP
!!. I,awborsk i leak in vent header Provide recommend:st ion fo r 1
In progress (I. A. Group) alternative methods of P/V Ackerman control Evaluate fire in containment 1
In progress Thiesing Long-term heat removal 1
Complete Thicsing Present 4/11 Unit 2 Containment Bldg.
- a. Possibic causes of 1
4/11 Stuench change of state
- b. P/T suitabic for 30 1
4/11 Stu enc h days
- c. Cleanup options for 2
In progress Lavroski cont. atmosphere Current assessment of core status
- a. From thermal-hydraulics 1
In progress Solbrig instrument data
- b. Sequence of events and 1
In progress Dietrich (i.ead) core description from event understanding Provide documentation of 2
Ongoing completed items Surveillance of Waste Stat. Group 2 To be assigned Till Uni'. I start-up criteria 3
To be assigned or sceuring criteria Reflux Boiler
- a. Non-condensib1,s/ water 2
Initiarcd Stuench l ev e l/ItP'<
- b. Temperature S pressure 2
Initiated Koler stusly of low reactor pressure In crument diagnosties 1
Ongoing Ackerman (reaetor core ins t riunenta t ion)
(Cont i nua l)
Specifications for lieflux Boiler Initiated Fernande:
lixp f r iment s to be carried out hy luiN 99 n(po, Instrument for measurement of Initinred Ackernun water level i n !!!'V Sioac i for noron ana gas in in nroeress s.n ro, o
4/15/79 N1'rACINOJT D RECOVERY PERJ Dr. Tvan Catton, ACRS Consultant At present the reactor system is stable. The core-coolant temperatu*re is 247 F.
Cooling is achieved by steaming in Steam Generator A.
B Steam Generator is isolated due to leakage from the primary side. We primary concerns at this time have to do with instru entation failure. Be in-tent is to get to solid water in both steam generators with natural cir-culation in the primary side at a pressure in the range 20 to 50 psi.
A plan has been formulated to achieve the desired goal with a number of backup options.
he first step in the process is to modify Steam Generator B flow loop so that it will be a closed cycle. Existing heat exchangers and pumps in the turbine building will be used. Up to 12,000 GPM of dominerali::ed water is available. We work will be completed within the next eight days. On completion, the B loop will take the reactor heat load and the second step will be initiated. Sten Generator A will be cooled to water solid. With the A loop free, modifications to the A condenser will be made so that it can perform as a liquid-liquid shell arx! tube heat exchanger. toop A will be run using the three available conden-sate pumps.
O F
s
. The secondary side at this stage will be limited to 100 psi and is viewed as a short-term provision.
Burns and Roe will have this work accomplished by April 22.
Two diesels will be ready soon for emer-gency power.
Both A and B loops will be tied to a 650 psi system via flangers installed during the short-term.Todificat.cns.
With both stean generators solid water, the primary system will be allowed to go over to natural circulation v.hile at 1000 psi. The pressure will then be decreased to scmewhere between 20 to 50 psi.
Calculations based en INEL water sample analsyis indicate that a 350 ft.3 bubble (at STP) may form. This is not thought to be a problem.
Several cackup precedures are simultaneously being pursued. There are a number of reasons for concern. Tao out of the three level sensors on the pressuriner are lost.
If the other is lost, the only means of level sensing will be the RTD's at the 21/2 foot level, and they are not that reliable. An effort is being nede to use neise gauges and ab-solute pressure measurements.
It is aoped that this can be ac-complished in time to use the remaining pressuriser level sensor for calibratiori. The water level in the containment is thought to 99 or(o t-a
. be close to electrical gear that can become incperative if scaked.
There is at present no way to determine within 16 inches or so what the level is.
Steps are being taken to install a pressure on the sump line gauge that will yield the level. There is a high proba-bility that contaminatien of clean parts of the Auxiliary Building There is, isagreement on whether the risk is piping may occur.
worth the information gained. The possible loss of instrumentation is also bringing pressure on the TMI team to go to natural circula-tion before the A and 3 secondary loops are solid. This is being resisted by NRC.
A temporary RHR system using standard equi;r.ent frca Surry is being installed outside the fuel handling building by Westinghouse.
At-ta:hments will be made to the existing system.
The new RIIR system will be skid mounted and put in a Quonset hut with full air control.
The goal is to have a backup system available as soon as possible.
It is estimated that it will be available within a week.
The exist.ing :wo MIR systems are felt to be operable. Westinghouse is upgsling the system and installing trote lubrication capability.
The work should be ccmp]eted by the end of the week.
oC7 99 eu,
. An additional two train permanent system is heirx3 designed for long-term heat removal.
It will te located in its own building.
If the need to use the RHR arises, it is not clear what system will be used first.
9&W feels that the Westinghouse temporary system should be used then thrown away. Bis way they would not contaminate the system they feel will be a part of the ultimate cleanup.
B&W also indicated that they were not sure how lorg the existing RPR system would run because pump seals and valve packings may not ccafine present activity to coolant. RHR pump seals are mechanical and there is sc.r.e concern abcut their ability to withstand the grit in the water. Further, a strainct in the line for startup may still be in and a chance of clogging exists. No one seems to know how long the RHR system will run unattended.
Another backuo is the so-called reflux mode where water is boiled in the core and condensed in the steam generator. Bis mode may lead to difficulties during the early course of the accident.
It is believed by scoe that ncncoadensibles will inhibit condensation and the core will reheat.
Finally, teed and bleed frcta the pres-surizer can be resorted to.
Its rot clear **ta
_o do with all the contaminated water.
hh nro[_ 30
,a.s, TMI-2 Recovery 1200
- BASE CAS E %Tt.'. W Mevision :)
CPU Service Corporahen el S-N)JW ( Q <- pa n.,'i m C p,,.., gy n,..,'n m
.e.,o (.r. -r
, m 5 r-3 y
- r.,.. s.
- );Q y L,.4
_/..
7 -
x., 4...co C
S m a m '" '
la00 ~
b' b'
3A k
4*)
O
- n
'A E
V c
9 A'
O a
i o
RC TE'T (*f)
S (1)
Degas at A.
I.cwer pressure (A-+A') while degassing, thea return to A.
-(Co=pleted 4/12/79)
(
Continue design / installation of static & active syste=s for primary
,2) volume / pressure control and secondary ecoling systems for OTSG "B" and OTSG "A".
(3)
Reduce tecperature (A*3) by steaming on "A" OTSG.
Verify operability of secondary ecoling system for "3" OTSO.
isolate (4)
At 3, establish solid cooling to "B" 0TSG, and when stable, "A" 0TSC.
(5)
Reduce te perature (3-#C) with OTSG "3".
Complete modifications to g'
"A" OTSG secondary system and start solid operation with "A" OTSC.
(6)
Ar C, trip operating RC Pump and establish natural circulation in "A" and "3" loops.
(7)
Reduce primary pressure.
(C4D).
(3)
Take primary system solid - activate new primary volume / pressure control system.
5 E':D POINT Primary:
- autral circulatter soild water, long; term pressure / volume control Secondary:
Solid water, lor g tera heat dump system APPROVEl TOR ISSUE:
M LffMA d {,'.1[k.( f h
RCA:cib R. G. Aruptd CPU M vv t 'or por.a:or.,.i.ut*..t h.ay of G. c r.il h it.lic Uvm.. C. a t ur.do.
99 59
- h..\\ '-Y9b'> %s C >f --
S hy ' ' y U.-
,(
~
~.
h..g;n ~* I t% %f??
k D
'^
. b
.g MC ddM u
j a______________.
A I
i g
g
_g
,N.a a
,'l
'n h a Q
N i
\\
l x%
w 4
,Q tN ;'!
.a i
QJ Q
i r-r et'-
.x b e y)
N w *$ y..yeQ l
---.. - {
Q N
n
+
g i
~
i
.s g
g.
I I
i
.. g
.).
4 M
l t
i m<
q v
,s w
's us g+
._Q 3 %
____...._..I
) p,h WE I
a e
A g';-
n -
I N
--- T M
4
. y _ _.__ _ _g g %g r
,o G-w
--w i
_ p. _ m \\ J....__q%,>2
_ _. _.._..i..._
_u_
i s
I
+
gc.
.__. l.... _.
n \\,. _ _q,s43 n
3 y
i s
I g
.._ Q s.'\\q 3
(n '
h,
\\y s
y
_.J.
g w
i 99 1
O C
4 W,
t
\\,:
v1 m.__
i y
y Y d E
)
I tti E I
I i
.d i
s e
w Qi q
N.
$c
-I i
i y
3 g
Q k
k i
8 K,l y 3
(;
.g i
'o g.
l_.____.
wi,
.g;.
.- g u:3
-_2 i
s gA, -
- g s
4.-
tJ s 4
S Wi 7.9 G
N N+N y
gs
&e h.n.
1
.t w
>c h f It
,- E - ?;
@ -l' l 3.
q t
t-o gi g %q, -- -- $ c3 4-C-pT - - h; i-
\\eql ?
)9 k>
s)
's y
s i
b h "s i
v l C F '
P A
b.
M Q {,:,. __ Q
&._. W' N 'L_. _
s
.s
-Q yi
'Q h
~
wi 4
e s
wa
't
%q s
{4 C;
"D 4 q
+=
cl 5
D
.o%
h-5 6
g u __.% __ o n ___ t.
__ ~
4 h-t --_ fu a 6, t
a s
k.
.h.
h
.?,-
h.i bk h$w 9.9
^c 6 0 s Q.
e-
-. @,.__ q _ y
- _. 9....n, a.
nD N.
m s
t, 9
N en it a
~
i.'/
s '.
La
?
Y k
N
-- r w-O a$ gO
's;i.C 0.. a* a n g,, s 2 w
x n
3, y
n O
C) F- - -
.s uf~ s pY. i e
s i, 4
8
,3
- a. 't- ;r-s D., x.,
p'-
6==-
q s
3 ?1 i.
Qs I c,:g es.-
-