ML20010F860
| ML20010F860 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 09/03/1981 |
| From: | Hukill H METROPOLITAN EDISON CO. |
| To: | Stalz J Office of Nuclear Reactor Regulation |
| References | |
| L1L-149, NUDOCS 8109110467 | |
| Download: ML20010F860 (5) | |
Text
___ _ _ _ _ _ _ _ _ _ _ _ _ _
Metropolitan Edison Company Post Office Box 480 II
- L Middletown, Pennsylvania 17057 Writer's % ct Dial Number September 3, 1981
},
L1L-149 Office of Nuclear Reactor Regulation
?;'
g
(((O Attn: John F. Stolz, Chief g
y Operating Reactors Branch No. 4 U. S. Nuclear Regulatory Commiseton L
.s11
($
Washington, D.C.
20555 L.-
QV N
\\;
e-
.S
Dear Sir:
[/p,
(
b
, g\\ /
s..
Three Mile Island Nuclear Station, 'qit 1 (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 Reactor Bellding Gpray System This letter is in response to your letter of March 7, 1980 and supplements our responses of April 2,1980 (TLL154), October 2,1980 (TLL 496), and January 29, 1981 (LlL 016), concerning the Reactor Building Spray System (RBSS). Your letter requested that either a series of tests be performed to demonstrate that our existing RBSS is capable of meeting its design function or determine that the RBSS operation with sodium hydroxide alone is acceptable.
As stated in our previous letters, we have chosen the second approach, that of justifying an RBSS using sodium hydroxide alone. This will entail draining and isolating the stdium thiosulfate tank in addition to implementing the controls described in the attachments.
Sincerely, s
11.
. Hukill Director, TMI-l HDH:CJS: mar Attachments cc:
L. Barrett D. Dilanni
/)ff00l B. 11. Grier H. Silver i
B. J. Snyder f
i
// l f[hCK05000289 467 810903 D
P PDR Metropohtan Edison Company is a Member of the General Pubhc Utihties System
ATTACHMENT Page 1 of 4 i
' RESPONSE TO NRC LETTER OF MARCH 7, 1980 NRC I'.e= a)
Assure adequate system configuration and concentration at the holdup tanks so that the pH of 8.5 to 13 exists at the spray nozzles and that the long-term solution (af ter mixing) will be above 8.0.
Also consider the draw-down capability of the system from the holdup tanks to the spray nozzles in the containment building and a single worst case failure in this system.
Respense to item a
==
Introduction:==
For the modified system, adherence to a pH ange of 8.5 - 11.0 is only necessary when'o single Reactor ",uilding Spray System (RBSS) train is in operation since only that single train is effective in iodine removal.
For the cases 1nere both RSSS trains are in operation, physical processes permit a broader range of pH.
Discussion:
The lower spray solution pH lim.t of 8.5 is based on the system iodine scrubbing ef ficiency with only ane (100% capacity) header in operation.
T52 system pH may be reduced to as low as 8.0 if both headers are in oper-ation and still maintain dose reduction f actor (DRF) equivalent to or better than single. train performance with a pH cf 8.5, due to increased flow from two train operation.
Two physical processes can be operant in reducing the iodine renoval effectiveness with twc loop operation. These are the potential surface area reduction of :Pc droplets and the pH dilution potential when droplets of differing pH coalesce.
When the RBSS is in coeratien the spray droplets can merge together causing a change from the initial droplet conditions during the fall fr e the spray headers to the point of impingment.
Two droplete have greater exterior surface area than does the single droplet produced after they combine.
This ccalescent effect causes a reduction of the iodine removal coefficient of up to 151. This reductions is more than adequately compensated by the incretsed number of droplets as a result of the increased flow of two loop operation.
The iodine removal coeff:cient can be further affected by different pH levels.
Specifically, a droplet originating from a loop eith low pH (dea to component failures) could dilute a droplet with proper (high) pH to below the pH limit of 8.5.
The scenerios which could cause the greatest difference in spray loop pH are; one Decay Heat Removal Pump (DHRP) failing to start, or one Sodium Hydroxide Storage Tank (SHST) valve failing to open.
If the DHRP f ails to start, calculations show that one loop will have a zero colarity and the other loop will have 0.098 Mole NaOH/ liter molarity.
Should one of the SHST valves icil to open, one loop will have zero colar-ity and the other loop will have 0.14 Mole haOH/ liter molarity.
The combined droplet's pH, for these cases, are 0.049 Mole NaOH/ liter (pH 8.4) and 0.07 Mole NaOH/ liter (pH 8.6) respectively.
Based on the above, in both cases (single and two train) of operation the Iodine Removal Capacity of the RBSS is adequate considering worst case single failures.
ATTACHMENT Page 2 of 4
(
The spray solution high pH limit of 11 fs based on the following:
1.
Biological effects of sodium hydroxide on the operating or maintenance crew.
2.
Corrosion effects of sodium hydroxide en safety related equipment.
An operating or maintenance crew could only be exposed to sodium hydroxide if the system were inadvertantly activated.
In this unlikely event, a spray solution. with pH close to 11 (as a result of component f ailurcs) could cause serious coughing, bronchial spasms, or skin and eye irritation and damage.
An expc;ure of less than a half hour could potentially be fatal, according to NUREC-CR-0650, "A Comparative Evaluation of Containment Spray Additives, Detrimental Impact of an Inadvertent Spray Actuation." The drawdown analysis show that the spray solution pH of 11 is reached no sooner than 5 min. after the system activation in the worst component fail-ure scenario, this giving the plant personnel adequate time to effect the isolation of the RBSS and/or exit fcom the building.
We, therefore, consider the potential biological effects of inadvertant actuation of RBSS with sodium hydroxide as the sole chemical additive to the borated water acceptable.
The coalescense affect described above was considered when the system iodine removal ability was considered since it is a more conservative approach.
However, the coalescense effect was disregarded when corrosion aggressiveness of sodium hydroxide was considered.
For corrosion considerations evaluation a map of the coverage of. separate no zles was made to determine how loop A and loop B cover the reactor build-ing operating floor.* The results are as fellows:
1.
The entire operating floor is covered by each spray header.
2.
55% of the operating floor is covered by each spray header twice.
3.
28% of the operating floor is covered by each spray header three times.
If both spray headers are in normal operation the spray overlap described above will assure mixing in the horizontal planes of one liter of spray solution of "A" molarity with three liters of spray solution of "B" molarity as the worst combination of the above possibilities (assuming A < B and A is molarity of the A loop and B is molarity of the B loop.)
If the molarity of the 4 liter total is averaged, the resultant molarity will be thac of the actual spray solution at the moment of droplet impingment.
It is the solarity upon impingtent rather than at the nozzles which must meet the
- Safety related compon:nts which could be exposed to RB spray are generally located at or below the operating floor.
ATTACHbENT Page 3 of4 ASS-56.5 and DRP 6.5.2 criteria provided that it does not impair the system's iodine removal capability.
l Since the pH of 11 is equivalent to a 0.22 Mole SaOH/ liter solution the above can be represented with the following expression:
1/4 (A - 3B) < 0.22 As shown in the following example, a pH of 12 f or a single spray header during two header operators can be tolerated.
However, if only one header is in operation, a pH of 11 must be maintained.
Example:
Calculations show that in the case of a sodium hydroxide valve f ailure uith two RBSS, and two HPI pumps in operation, one header will have solution of 0.29 Mole NaOH/ liter at 30 min. after the system actuation and the other will have a zero molarity.
The 0.29 colarity corresponds to pH 12.
1/4 (0 + 3 x 0.29) < 0.22 0.2175 _< 0.22 Draw-down analyses performed for the full transient duration demonstrate that above expression is satisfied for all potential modes of RSSS operation.
In the case of peripheral nozzles at higher elevations, there the spray cenes are not developed nd do not overlap, the coating of the containment liner will be exposed to solution as it comes out of the no::les with pH varying (for transitory periods) from 4.5 to 11.
According te the RB coating manu-facturer. this exposure of the coating to extrema conditions will not cause degradation of the liner since the exposures are of short d u' rat ion s.
Technical and Administrative controls which will be implemented are as follows:
a) A differential level measuring system for the Sorated Water Storage Tank (bWST) and the Sodium Hydroxide Storace Tank (SHST) will be provided. This local level instrumentation will enable the operator to determine SHST level relative to the S'..'ST level to ensure that initial tank levels are within the range required for proper function-ing of the system.
During system operation the EWST and SHST levels will draw-down with velccities linearly proportional to each other because the tanks discharge piping are cross-connected and both open to the atmosphere via vacuum breakers or rupture discs.
The liquid columns in thetanks will remain in balance. This means that the sodium hydroxide flow rate does not depend on the number of sodium hydroxide loops in operation but cill remain proportional to the BWST rate.
b)
The SHST anc BWST will have Technical 3pecificatica limits on concentration and relative tank levels to assure that pH limits are met.
These Technical Specifications will be submitted upon NRC acceptance of this conceptual proposal and will be based on maintaing:
(1) SHAT Level with i 6 inches (water) of BWST level (2) SHST Concentration 12.0 - 13.2 weight % SaOH (3) BWST Level per existing Technical Specification Limits.
I l
a
e ATTACID!EhT Page 4 of-4 NRC item b) Determine the iodine removal ef fectiveness of the spray using the evaluation methods described in NUREC-CR-0009.
Response to item b The iodine removal effectiveness of the Reactor Building Spray System using only sodium hydroxa de was determined using the methods and Iodine Removal
~
Coefficients of ASSI/ANS-56.5-1979.
NRC item c) Demonstrate that offsite doses for the design basis accident will be within the 10 CFR 100 guidelines based en the calculated iodine reduction.
Response to item c Our calculations deconstrate that the offsite doses for the design basis accident will be within the 10 CFR 100 guidelines.
Specifically, the results obtained for minitur safety features (one spray header pump and one air cooling unit fan operating) show that the two hour dose would be 189 rem thyroid 7.6 rem whole body at the exclusion boundarv.
If both spray header pumps and all three air cooling unit fans were operating, tne two hour dose would be '.56 ret thyroid and 7.5 whole body at the exclusicn boundary.
These c.e conservative values, and are ::e:1 below the limits of 10 CFR 100.
The as7unptions used in the dose calculation are the spray removal coefficients as per iter b above, an exclusion boundarv distance of 610 meters, and an average Atmospheric Diffusion Factor (F/Q) of S.3 x 10 ~ sec/m# (X/Q derived
~
from meterological data collected since TMI-1 was licensed).
a
_