ML20064E471

Responds to 780627 NRC Request for Review & Eval of Actions Recommended by Dupont Re Waste Tank Safety.Plans to Augment Procedure for Transfer of Waste from Insvc Tank 8D-2 to Spare Tank 8D-1 in the Interest of Greater Conservatism

ML20064E471
Person / Time
Site:	West Valley Demonstration Project
Issue date:	10/12/1978
From:	Clark J NUCLEAR FUEL SERVICES, INC.
To:	Starostecki R NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
References
NUDOCS 7811170233
Download: ML20064E471 (9)

Text

~ '

,... - -,r-cypp<".4 d' 0 0 Q W, I

w -

V Nuclear Fuel Services, Inc. 6000 Executive Boulevard, Suite 600, Rockville. Maryland 2085

][h A Subsidiary of Getty Oi! Cornpany I O *'2 M

(301)770-5510

' 3 CCI i6 l'.'.Cddober 12,1978 Docket No. 50-201 u,( , G.

S:t '

' ~

7.; N Mr. Richard W. Starostecki, Chief /

DOCKETED Fuel Reprocessing and Recycle Branch USNRC $

Division of Fuel Cycle and Material Safety 6- '

U. S. Nuclear Regulatory Commission "

OCT 171978 > b Washington, D. C. 20555 NM59 f

/-

c3 DC SECT!oN /

Reference:

NRC Letter to NFS dated June 57,1978 D.CCKET CLERK /

Dear Mr. Starostecki- 'Y j &,

M"/ ..:

Your letter of June 27th requested NFS to review and evaluate some specific actions recommended by DuPont in their report entitled " Safety Related Information Available on NA Waste Tanks." Per your request, NFS has again carefully considered its procedure for the transfer of waste from the 5-service tank 8D-2 to the spare tank 8D-1 and has again concluded that the present procedures allow for such a potential transfer to proceed in a timely, safe and efficient manner; however, in the interest of additional conservatism, we are planning, as described below, to augment that p ocedure.

1. EVALUATION OF RESPONSE TIME *

a. Estimate of Leakage Rate

,' Because the design and the operation of the waste tank 8D-2 have been directed at woiding the deficiencies that had arisen in some high level waste tanks at U. S. government. sites, there has.not been . leakage .fr.om a ..

similar waste tank;' therefore, there is no historical base to ~ estimate potentialleak rates. Any estimate of potentialleakage would be conjectural and is irrelevant because the permanently installed 8D-2 pan pump would return leakage of up to about 20 gpm back to the 3D-2 tank, thus eliminating

~

- ~

"or at least significantly redu~cing any post'ula'ted leakage.

b. Time Required to Detect Leak The liquid levels in the pan and in the vault of the 8D-2 waste tank complex are continuously monitored by instrumentation and annunciated, via a trouble alarm, in the plant's Control Room as well as in the Waste Tank Farm Shelter. The liquid level detector in the 3D-2 pan is set to annunciate ct six inches, which is approximately 1,200 gallons. An operator, licensed by the U.S. NRC would acknowledge the alarm, investigate the cause, and report to a NRC-licensed Shift Supervisor, one of whom is always on site.

7///D&M2 A % e.15 - Z o ~L o l ' 0

• Identification corresponds to that in referenced NRC letter. 11C($

~ -- v = 7 . yf b

Nuclear Fuel Services, Inc.

Mr. Richard W. Starostecki October 12,1978 Page Two The Shift Supervisor is authorized by NFS to take any or all of the following actions should he suspect that waste tank 3D-2 is leaking:

• withdraw a sample of the liquid from the 8D-2 pan for radiation / radioactivity measurement;

^ - ~

-~ ~

sarnple the airborne' radioactivity in the' vault;

~

~

'* 'idittatehuinpdown'of the bhllast ligdid in'8D-1' waste tank; and/oi, =

J_ -

- * . pump the liquid accumulating in the 8D-2 pan to the 8D-2 tank via - . .

the 8D-2 pan pump.

Meanwh'ile/ the Shift Supervisor,- in- accordance- with the established- -- -

Emergency Plan, would notify. the members of the Plan Safety Committee.

- - - c. Instrument Reliability The 3D-2 waste tank' pan level instrument, including the associated-high level alarm, is:

'.(1) . monitored at_least once each shift by an NRC-licensed operat'or who.

records the pan level instrument reading on a daily record sheet pertinent only to the Waste Tank Farm;

~ ~

(2) func~tionally checked'tnonthly by a simulated signal to the transmitter; -

4 this check includes assuring .that the "WTF Trouble" alarm annunciates

.- . -- . -i.n the plant's Co.n. trol-Room; . . .

.(3) calibrated every six months with simulated signals to the transmitter.

.To reconfirm the pan level detection system, NFS willin the next month add - -

water to the 3D-2 pan and verify the instrument response.

^ ~~' ~ ~~ ~ ~~'

'd. Pa'n Pump ' Reliability

~

- Because of both the high integrity.of the .8D-2 vault walls and the high --

ambient temperature.in the 3D-2 vault, there has been no accumulation of . '~

moisture ir) th.e 3D-2 pan; therefore, the l i- an pump has had little servic6.

~

Whilele' have no~rea' son to- believe that the- 8D:-2 pan pump would not --- ~~

operate properly if called upon, we will test the 3D-2 pan pump during 1978' and annually-thereafter by adding water to the pan and pumping it to Lagoon No. I via the. Installed line. The 8D-1 pan pump is identical to the 3D-2 pan pump'a'nd ~the'reby' serves as a readily available spare. ,

i Nuclear Fuel Services, Inc.

l .

Mr. Richard W. Starostecki l October 12,1978 l Page Three

e. 8D-1 Condensate Pumpdown j

A submersible pump in 3D-1 is used to remove co:.densate from 8D-1. Upon i removal, the condensate is either treated by ion exchange and routed to the l lagoon system or returned to the 8D-2 storage tank. This presently installed  !

pump has a capacity of about 12-15 gpm. Because the minimum 8D-1 condenstate level will be maintained at about 120,000 gallons as an anti-bouyancy measure, it would presently take up to seven days (depending upon the amount of condenstate that the Safety Committee decided to remove) to -

/ remove the condensate from 3D-1.

In order to significantly reduce the potential time requirements of this step in the waste transfer procedure, NFS has decided to replace the present pump with -the special waste transfer pump.(8G-2)-which has a capacity of about 100 gpm, thus reducing the maximum time requirements to about twenty hours. As indicated above, the 8G-2 pump will be used for the routine removal of condensate collected in SD-1; therefore, its reliability will be monitored.

f. 8D-2 Pumpout Procedure On August 23rd, NFS delivered to your Staff copies of "8D-2 Pumpout Procedure, Revision 1," which is the presently effective procedure for responding to a potential leak in waste tank 8D-2. As noted above, NFS intends to install waste transfer pump (8G-2) in the spare waste tank (8D-1) as a means of routinely testing the availability of the transfer pump. When

( the installation is complete, the pumpout procedure will have been revised to reflect the status of the transfer pump.

Attached is our time and motion summary of the procedure which starts with 8G-2 in the 8D-1 tank.

2. OPERATIONAL CAPABILITY OF 3G-2 ii

a. Motor Checks NFS will begin work on the installation of the special transfer pump (8G-2) in waste tank (8D-1) this month. After installation, the pump will be functionally tested at least quarterly. In addition, the motor will be subject to NFS' routine preventative maintenance program. - -

b. Pumping Action The SG-2 pump, when installed in waste tank 8D-1, will be used to transfer

, the approximately 100,000 gallons per year of condensate that collects in 3D-1 to either waste tank 8D-2 or to the lagoon system. This will provide

[ confirmation of proper pump performance.

P

~

l

)

Nuclear Fuel Services, Inc.

Mr. Richard W. Starostecki October 12,1978 Page Four

c. Sludge Pumping The waste transfer pump, 8G-2, is a deep-well pump that was especially selected for the waste transfer operation. It is a two stage, centrifigal pump rated to deliver the 100 gpm when pumping a sattgrated NANO 3 solution at a specific gravity of 1.4 and a temperature of 210 F. The pump design includes a sparge ring at the pump inlet.

/ A synthetic 3D-2 waste prepared by NFS to match the waste tank input records indicates that the sludge portion is relatively free flowing such that the bulk of the tank contents should be pumpable.

d. Back-up Transfer Equipment Although NFS believes that the rugged 8G-2 pump would perform adequately for periods much longer than the relatively short time (less than 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />) necessary to transfer waste from 8D-2 to 8D-1, NFS is committing to additional back-up transfer equipment as a " defense-in-depth."

NFS has initiated the purchase of a positive displacement pump capable of at least 100 ppm transfers. A formal procedure will also be developed for use of this pump should 8G-2 become inoperative.

3. RADIOLOGICAL CONTROLS

( a. Personnel Exposure Although a highly unlikely event, the transfer of waste from 8D-2 to 8D-1 would be a preplanned event involving rather straightforward techniques which would be conducted under the close scrutiny of the Plant Safety Committee and other management. With the operations preplanned, bulk shielding in place and personnel at a distance during the actual transfer operations, personnel exposure should be insignificant.

The procedure calls for three feet of soil or equivalent shielding over the pipe during the transfer. Calculations (attached) show that this would reduce the radiation dose rate to less than 1 mr/hr at the surface of the dirt.

b. Shielding Effectiveness The three foot thick dirt shield called for in the proceoure is only a target minimum. It could easily be augmented if dose rates were found to be higher than desirable during the actual transfer.

c. Joint Leakage The piping system is predominately welded, but there are some mechanical joints. The 8G-2 pump specifications called for zero leakage past the

c>

• 1 1

Nuclear Fuel Services, Inc.

Mr. Richard W. Starostecki October 12,1978 Page Five mounting flange. All joints would be assembled by maintenance personnel skilled in assembling equipment for radiochemical service.

We would be pleased to meet with your Staff and discuss these actions in additional detail.

Very truly yours,

[

c-James R. ark Manager, Quality Assurance and Licensing 3RC:jnw Enclosure

~

U

~

Rev. 2

- 10/13/78 Appendix A 8D-2 PUMPOUT TIME ESTIMATE Sub-Total Total Man- Elap. sed Time Per Elapsed Job Hrs Time Job Time

1. a. Assemble crane and move to

. WTF:

Crane operator 4 hrs Oiler 4_, hrs 8 4

b. Prepare 80-1 area, make plastic bags, etc.:

2 Operators 4 hrs 12 6

c. Connect 8D-1 pumpout line 2 Mechanics 2 hrs 4 2 2 2

2. a. Pumpout 8D-1 to Lagoon 1.

Use 8G-2 0 100 gpm.

Assume 100,000 gal in 8D-1 17 17 19

(

3. a. Prepare 80-2 area for plug pulling and pump install-ation.

2 Operators 6 hrs 1 H&S 6_ hrs 18 6 (6)

b. Remove port plug from 8D-2.

~

Crane Operator 4 hrs Oiler 4 hrs .

2 Mechanics 4 hrs 1 H&S 2, hrs 18 4 (10)

g .

Appendix A Sub-Total Total Man- Elapsed Time Per Elapsed Job Hrs Time Job Time

4. a. _ Remove 8G-2 pump from 80-1 and install in 8D-2 Crane Operator 6 hrs Oiler 6 hrs 3 2 Operators 6 hrs H&S 6, hrs 24 6 6 25

5. a. Connect discharge line from 8G-2 to 8D-1.

Crane Operator 1 hr 2 Mechanics 6 hrs Oiler 1,hr 8 6 6 31

6. a. Run water line to sparge ring on pump. ,

2 Mec'anics n 3 hrs 6 3

b. Cover line from 8D-2 to 80-1 with 3' of dirt.

i 4 hrs Bulldozer Op.

Oiler 4 hrs 8 4

c. Shield around pump with lead and steel.

2 Operators 4 hrs 6 4

d. Wire pump.

2 Electricians 8 hrs 16 8 8 39 TOTALS 130 M-H 39 Hrs

g

/

we 8

7 2/

3

.1 v/

e0 R1 0

,4 n

6 3

. 2 3

N O l l I

T A

M I 8 T

S

.2 E )

s E r M u I l o

T ia 4 H

. (

2 T e U

0 i m

P M T U

P 0 d

. 2 e

+ 2 s

- l p

3 in a

8 l E

.1 6 i3 2

l iI .1

.8 l l

.4 j

j j a b c a a b 1 2 3 4 5 6 l2$ =g

Appendix B BASIS FOR THREE FEET OF EARTH SHIELDING OVER TRANSFER LINE The 80-2 solution has Cesium-137 and -134 isotopes as the principal gamma emitters. A concentration of 5.5 mci /ml Cs-137 was used.

The model for calculation was the series of point sources. The unshielded dose rate at one meter was 22 Roentgens per hour.

The mass attenuation coefficients listed in the Radiological Health Hand-book were used to determine half value layers and with the application of an estimated dose buildup factor, a shielded dose rate was determined to be 0.6 milliroentgens per hour.

The mathematical calculations were compared with standard shielding graphs from the Rad Health Handbook which generally confirmed attenuations of ten thousand fold by about 40 inches of earth, k

i

-- _ _ _ _ _ , - . ,