ML18283B407

LER 1977-012-00 for Browns Ferry Nuclear Plant, Unit 3, Temperature Transients Were Experienced with Six Charcoal Adsorber Beds in Offgas System

ML18283B407
Person / Time
Site:	Browns Ferry
Issue date:	10/05/1977
From:	Tennessee Valley Authority
To:	NRC/RGN-II
References
LER 1977-012-00
Download: ML18283B407 (10)

Text

' *

.'DLPage 3

~

~

>> BPA85 10/5/77

~ e NRC FORth 366 U,S. NUCLEAR REGULATORY CoslthlSSION P 77)

',LICENSEE EVENT REPORT EXICI8 IT A 6

O )PLEASE PRINT OR TYPE ALL REQUIRED INFORMATION)

~~o.") ~>o

~

~el 7 8 9 A L B R LICENSEE CODf.

F 3 14 Q 15 0 0 - 010 0 0 LICEkSE kl>>II>>IDEA 0 - 0 0 Q) 26 4 I a LICENSE 'TYpE II a JP Q~Q>> 51 CAT 56 CON'T Bee:l:!*'. )" I>>

7 8 ep 61 'OCKET NUMBER 69 Evf NT DATE 14 15 AEPOAT DATE EVENT DESCRIPTION Atlo PROBABLE CONSEQUENCES OIO

~

~O2 13urin normal o eratio t:em erature trans'ent:s mere ex erienced with t he six charcoal adsorber beds in the of f as s stem, The stack radia ti'on

~n4 monitor showed no increases durin this period There mere no effects

~OS

[oOGJ

~ were n n ublic health re iu c and r

safet c

redundanc was not a licable'nd there O B 7 8 9 SYSTEM CAUSE CAVSE COMP, VALVE CODE CODE SUSCODE CDMPDNEIITCODE SUOCODE SUBCODE I .

[oasJ 7 8

~5f 9

B 10 Q ~xO) 11

~xQ)>>

17

~F 12 1 L T E R 18 O)>> ~zO)>>

19

~z 2D Q)>>

SEOVE NTIAL OCCURRENCE REPORT REVISION ~

EVENTYEAA

~

REPORT NO. CODE TYI' ND.

0>>>>>>>>>> ~II 3 ~OX 3 ~~ ~OX

~

~ . ' ~

o ACTION FUTURE TAKEN ACTICN Ie)S)t)S teS 21 22 EFFECT ON PLANT 2$

CAUSE OESCRIPTION ANO CORRECTIVE ACTIONS A drainline on 22 the

>>HU.DDVIN METHOD t>>O 25 off 24 as

~)'

DI O27 26 HOURS O22 27

'>>S 40 reheatex was found- lu ATFACHLIE'CT SVDMITTfD 41 e9 NSADP Fcr LI YJ6,

'MS Ieo 42 20 ed.

PRIME COMP.

SUI ~ LIER Dl Without ade uate

~O 44 22 COMPONENT 4ANUFACTUAER 47 reheatint diluti n te m'c binat:ion f vd

~13 the of f as s stem downstream o f the recombiner a oarent 1 i nited the char 4 coal drainline I 1 bed were reteree~d 7 8 9 ( See attached sheee )

FACILITY

'4 pcavEA O p

'EK 7 8 STAIUs 9

E Q>>: ~oa 10 IQ>>1 12 OTHER STA TUS 44 DISCOVEAY

~A 4$

O31 DISCOVERY DESCRIPTION O32 Increased heat in adsorbe beds SD ACTIVITY CONTf NT AELEASED OF RELEASE AMOUNTOF ACTIVI'tYQ3S LOCATION OF RELEASE O

~

fits f Z l 0" 7 8 9 LZJQ34 10 11 A

44 4$

N A

~ EASDNNEL ExposunEs NURSER TYPE DESCAII'TION O3B

~00 0 Q) Z Of>> N/A 7 8 9 ll 12 PERSONNEL INJUnlf5 12

~ 7 8 9 11 12 Loss of on DAMAGE To FACILITY Q43 T VPE'ESCRIPTION

~ZO42 N/A e 9 Ip PVOLICITY NRC USE ONLY ISSUED DESCRIPTION ~~

KH LJS 7 8 9 10 r s Release dated .Tul 18 1977.

68 69 NAME OF PREPARER PkONE:

~ I

<Revision

~ ~

~ ~

Cause Descry tion and Corrective Actions (Continued) service and satisfactorily passed operational testing.

JGD: SGS 3/27/78

~ ~

EVENT DESCRIPTION At 2215 hours0.0256 days <br />0.615 hours <br />0.00366 weeks <br />8.428075e-4 months <br /> on 7/15/77 the unit was returned to service after being shutdown since 7/7/77. During startup, problems were encountered with high offgas flow rates, low recombiner preheat temperatures, and low recombiner temperatures.

At 0900 hours0.0104 days <br />0.25 hours <br />0.00149 weeks <br />3.4245e-4 months <br /> on 7/17/77 operations was notified by a chemical laboratory analyst that the concentration of hydrogen in the offgas system was greater than 4%. One of the redundant hydrogen analyzers was in service during this period, but did not indicate an abnormal concentration of hydrogen in the offgas'system. Subsequent analysis at 1345 indicated hy'drogen concentration was less than 4%. About 1200 hours0.0139 days <br />0.333 hours <br />0.00198 weeks <br />4.566e-4 months <br /> on 7/17/77, an attempt was made to place the standby steam ]et air e)ector (SJAE) "B" in service. Problems were encountered with the pressure controller regulating dilution steam to the third stage of SJAE. During the 1-3/4 hour period associated with the unsuc-cessful attempt to transfer SJAE's, several significant but brief transients were noticed in system pressure, apparently due to excessive dilution steam.

Increases were noticed in system flow rate, dew point, and reheater inlet temperature. Efforts to transfer the SJAE were terminated at 1400 hours0.0162 days <br />0.389 hours <br />0.00231 weeks <br />5.327e-4 months <br /> on 7/17/77, and system operation using, the original "A" SJAE resumed. The stack radiation monitor showed no increase during this period. The charcoal adsorber beds were in service m.th parallel flow through two trains of three vessels each.

An effort to switch to the standby recombiner was unsuccessful because accept-able preheater temperatures could not be obtained.

DESCRIPTION OF THE OCCURRENCE At about 1200 hours0.0139 days <br />0.333 hours <br />0.00198 weeks <br />4.566e-4 months <br /> on 7/17/77, the temperature of adsorber beds "B," "C,"

"D," and "F" began to increase from their steady state average temperature of 67 F. The temperature of "C" bed peaked at 146 F at 1645 hours0.019 days <br />0.457 hours <br />0.00272 weeks <br />6.259225e-4 months <br />, and "D" bed temperature peaked at 146 F at 1720 hours0.0199 days <br />0.478 hours <br />0.00284 weeks <br />6.5446e-4 months <br />. The indicator temperature of "B" bed went offscale (150 F) at 1610, and "F" bed temperature peaked at 138 F at 1815 hours0.021 days <br />0.504 hours <br />0.003 weeks <br />6.906075e-4 months <br />. The temperature of the vault housing the charcoal vessels increased steadily beginning after 1200 hours0.0139 days <br />0.333 hours <br />0.00198 weeks <br />4.566e-4 months <br /> on 7/17/77, from its normal 75 F to 100 F over a 7-hour period.

The charcoal vessels were isolated at 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> and the offgas flow was routed through the bypass line. On 7/18/77 at 0420 hours0.00486 days <br />0.117 hours <br />6.944444e-4 weeks <br />1.5981e-4 months <br /> a nitrogen purge was began through "B" and "C" vessels.

On 7/18/77, at 1200 hours0.0139 days <br />0.333 hours <br />0.00198 weeks <br />4.566e-4 months <br /> the adsolber vault room was entered. The bottom of "D" vessel was found to be hot with its blue-gray paint discolored but not blistered or cracked. A slight discoloration of the paint covering the center portion of "B" vessel was evident. Thermocouples were attached to the exterior surfaces of vessels "B" and "D" for supplemental monitoring during the nitro'gen purge. The maximum skin temperature recorded was 142 F at midpoint of "B" vessel.

All vessel temperatures were within the normal range on 7/20/77,

CORRECTIVE ACTION A task force composed of TVA operating and design employees and GE employees met at BFNP on 7/20/77 and 7/21/77 to investigate the occurrence. On 7/22/77 the unit was removed from service to facilitate repair work recommended by the task force. During the outage, work on the preheater drainlines revealed that two common sections of P-1/2" line were restricted. Since external means were used to dislodge the blockage, the cause for the blockag'e is not known. It was these blocked drainlines which prevented normal steam flow through either of the recombiner's preheaters. Without adequate preheat, condensation formed in the preheater. This condensate carried over to the recombiner vessel, covered the catalyst, and prevented recombination. When the recombiner drain was opened, moisture drained off the catalyst reinitiating recombination.

The existing bottom drains on the charcoal beds were found to be inadequate to drain the entire system. A Design Change Request to install a low point drain on the inlet to each vessel has been initiated.

Restricted sample lines preventing the hydrogen analyzer from functioning were blown free. The sensor measuring the moisture content of the influent to the charcoal beds was coated with dirt and had to be replaced. Methods to improve the reliability of these two instruments are being pursued. On 7/23/77 the unit was returned to service after demonstrating satisfactory operation of both preheaters, both recombiners, both hydiogen analyzers, and l I Bll Plant procedures have been reviewed and revised to include explicit actions to be taken in the event vital offgas system parameters deviate from their designed range.

Samples of'he charcoal in "B" vessel were taken and examined. No evidence of discoloration or degradation was seen. The moisture content of the samples was determined, to be 2/. Samples of the charcoal in the other beds will be taken to ensure it is dry before the beds are placed back in service.

An extensive investigation was conducted to determine-the cause of the temperature excursion. All evidence suggests a hydrogen ignition in the offgas system ignited carbon fishes in adsorber beds B 'and D which resulted in some sustained combustion. Tfiis heat was transported by the offgas flow to other beds as evidenced by the swings in bed temperature.

The system was operated for approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> (midnight until noon)

)ust before the incident with no recombination taking place. Since there was no flow on the Steam side of the offgas preheater because of the blocked preheater drain, the dilution steam, instead of being preheated, was at saturated temperature. This resulted in condensation on the catalyst which prevented recombination from taking place. Without recombination, the radiolytic production rate of hydrogen (about 50 cfm at that time) and the offgas flow (200 cfm at the time of the incident) gave a hydrogen concentra-tion from the inlet of the holdup pipe to the stack of 25/.

The SJAE logic is such that one get must be shut completely down before the other can be started. This results in about. a two-minute period in changing SJAE's when there is no dilution steamflow.'uite likely, when the attempt was being made to switch from "A" to "B" set of SJAE's misting over of the catalyst ceased when "A" SJAE was shutdown, and, during the two-minute interval before'B" was started up, hydrogen diffused back into either recombiner (the valve on the standby recombiner was cracked open),

contacted the now-active catalyst, and ignition took place. Velocity of a flame front of this sort is about 6,000 fps (reference 1) and accounts for the "blown apart" appearance of the pre- and after-filter elements.

The theory of spontaneous combustion by moisture adsorption does not seem.

possible for this type of system. Experimental data utilizing moist gas flow (references 2 and 3) gave temperature increases at least an order of magnitude below the ignition point of carbo'n; and, in fact after an initial temperature peak, further temperature rises were not possible due to the thermal desorption of the water. In other words, spontaneous combustion appears to be a self-limiting phenomenon given a set of airflow and humid-ity conditions. Experimental data with no airflow also gave a very low temperature rise (46 0 P) for the size of carbon bed in question (reference 4).

Ignition by hydrogen flame appears the most plausible; and, in fact, an almost identical circumstance occurred in 1958 (reference 5).

The charcoal beds were returned to service on November ll, 1977. Special Test 155 was performed to closely monitor differential pressure and isotopic delay times during the first two months of service. All system..parameters closely paralleled values obtained during -the initial startup testing program.

Thus, the July 17, 1977, temperature excursion did not affect the performance of the charcoal beds.

JGD: SGS 3/27/78

Attachment III References

1. Garison, Knight, Henrie, Flame and Detonation Initiation and Pro a ation in Various H dro en-Air Mixtures With and Without Water

~S r~a , Atomics International Division, issued May 11, 1973.

2. J. L. Kovach,-The Evaluation of the I nition Tem erature of Activated Charcoals in Air Steam Ox en and Oxides of Nitro en, Nuclear Consulting Services, Inc., NUCON 024, 1968.

3. Dederick, Martin, Porter, Reactions of H dro en/ en Mixtures with Charcoal and Thermal Effects Associated with Ph sical Adsor tion, General Electric Company, Report No. 72CRD182, June 1972.

4. Personal correspondence between Dan L. Paul, TVA Chemical Section, and Willard Clark, President, Barnebey - Chancy, July 28, 1977.

5. Browning, Removal of Fission Product Activit from Gases, Nuclear Safety 1 (4), page 49, June 1960.

JGD:SGS 3/27/78