IR 05000440/1986020

Insp Rept 50-440/86-20 on 860623-0821.Violations Noted: Inappropriate Procedures Used to Control Testing on Augmented Quality Sys & Failure to Have Administrative Procedures to Control Fire Hazards

ML20215C661
Person / Time
Site:	Perry
Issue date:	10/03/1986
From:	Gill C, Grobe J, Holmes J, Odwyer G, Williams C NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20215C617	List: IR 05000440/1986020 ML20215C613 ML20215C640
References
50-440-86-20, NUDOCS 8610100390
Download: ML20215C661 (18)

Text

-

,

-' .

..

-

U.S. NUCLEAR REGULATORY COMISSION --

-

'

REGION III

Report No: ,50-440/86020 Docket No: :50-440 License No: NPF-45

' Licensee: Cleveland Electric Illuminating Company Post Office Box 5000 Cleveland, Ohio 44101 Facility Name: Perry Nuclear Power Flant, Unit 1 Inspection At: Perry Site, Perry, Ohio

. Inspection Conducted: June 23 through August 21, 1986 C'. W A AA.> W' y ;

Inspectors: G. F. O'Dwyer lo-3-56 Date- l

. A. Grobe /0-Z- %

l Date

, . Holmes m. z- 8 4

!

Date C. ill % - #-3-4 Date

& pDJC.97"L(k w Approved By: C. C. Williams / c) -3 -16 Assistant to the Director Date Division of Reactor Safety Inspection Summary Inspection on June 23 through August 21, 1986 (Report No. 50-440/86020(DRS))

Areas Inspected: Announced, special safety inspection by region based and resident inspectors of the charcoal adsorber bed combustion events of June 20, 1986.through June 23, 1986 and the re-ignition of the charcoal beds from July 6'

to July 8, 1986. The inspection was conducted consistent with NRC Inspection module 9370 Results: Two violations were identified (inappropriate procedures to control testing on Augmented Quality Systems - Paragraph 4; and failure to have administrative procedures to control fire hazards - Paragraph 5).

$k0 00 go G

,

,

, ._ __ _ _ _ _ _

- _ . _

,

. .

.

r

-

-

DETAILS j:

1 . Persons Contacted

-

i'

• D. Lyster, Manager, Perry Plant Operations Department

• C. M. Shuster, Manager, Nuclear Quality Assurance Department

.

cB. D. Walrath, General Supervising Engineer, Operation Quality Assurance

'

.

i~

R. A. Stratman,-General Supervising Engineer, Operations

• K. R. Pech, General Supervising Engineer, Nuclear Engineering Department

.

• V. K. Higaki, General Supervising Engineer, Maintenance Quality Sectin

• R. J.' Tadych, Manager Alternate, Perry Technical Department l *M. D. Makar, Fire Protection Engineer, Nuclear Engineering Department

.

! *S. J. Wojton, General Supervising Engineer,-Perry Technical Departmen .

• J. S. Grimm, Chemistry Staff, Perry Technical Department '

• C. C. Enos, Operations Engineer, Perry Technical Department

, *D. H.:Reyes,; Plant Chemist, Perry Technical Department 4 *T. Heatherly, Operations Engineer, Perry Technical Department

, *R. Awan, System Engineer, Perry Technical Department

, * Denotes persons attending the' exit meeting of August 21, 1986.

l The inspectors also interviewed other licensee employees including members

'

of the quality assurance,< technical, operating and testing' staff . Synoptic Summary of Off-Gas System Charcoal Combustion Event On June 20, 1986, at.1145 hours0.0133 days <br />0.318 hours <br />0.00189 weeks <br />4.356725e-4 months <br />, an Unusual Event was declared'by the licensee based on their determination that the activated charcoal in

} at least two of the'off gas charcoal adsorber vessels was burning. The l fire was apparently initiated by radiant energy type space' heaters placed in close proximity to the' charcoal adsorber bed vessels. Prior to the

~

i

ignition of the charcoal, the space heaters were being utilized to elevate F system temperatures in preparation for a test which required that the vault

! and equipment have an initial temperature of at least 150*F for testing'of j the temperature pulldown capability of the Off-Gas Vault Refrigeration ,

f Syste After identification of the fire, a nitrogen gas purge of the l Off-Gas System was initiated to control the fire at 1219 hours0.0141 days <br />0.339 hours <br />0.00202 weeks <br />4.638295e-4 months <br /> on June 20,

198 NRC Headquarters and Region III Response Centers were activated on

, June 20, 1986 to monitor and evaluate progress of the licensee's effort

,

to extinguish the fir On June 23, 1986 a Confirmatory Action Letter (CAL) was issued by Region III in which the licensee committed to preserve information, equipment, and j specific documentation regarding this fire. The licensee also committed to develop a recovery plan.

I On June 23, 1986, at 1151, the Unusual Event was terminated. The nitrogen

' purge activities continued until adsorber bed vessel entry on June 27,

1986. After charcoal sampling activities were completed, a nitrogen

. blanket was established and maintained on both A and B trains of the I adsorber beds.

l

.

I

- .

. - - - - - - - . _ _ _ _ _ _ - _ _ _ _-

. _ _ _ . _ _ . . . - _ _ _ _

,

. .

- On July 6, 1986 the nitrogen purge of the off gas sytem was stopped.

-

Subsequently, a dry instrument air purge at 60 scfm was initiated through trains A and B. Approximately one hour after the initiation of this

' '

instrument air purge, the center therraocouples on adsorber beds 14A and 148 indicated: temperatures greater than 250'F. On July 6, 1986 at 2037 t

.

. an Unusual Event was declared based on the determination that the '

adsorber bed charcoal had re-ignitied. A nitrogen purge was. initiated-through the beds!. The second Unusual Event was terminated on July 8,

- 1986 after it was determined that an acceptable rate of decrease in-adsorber bed temperature and negligible concentrations of combustion

'

products were present in_the nitrogen purge stream. During July 15 thru.

17,.1986 the existing adsorber bed charcoal was removed from the vessel L Systems Description As stated in the Perry FSAR, Section 11.3.1.1, the objective of the gaseous waste management system is to process and control the release of gaseous radioactive effluents to the site environs so as to maintain the exposure j of persons in unrestricted areas to radioactive gaseous' effluents (Appendix I~to 10 CFR 50) as low as reasonably achievable.

i The Perry Off-Gas System (a non-safety, augmented quality system)-is made

'

up of three segments. The first. segment consists of preheaters, catalytic i- recombiners, an off gas condenser,_and a water separator. These components

, are'used to recombine the hydrogen and oxygen, and to remove moisture from the off gas process stream. The second segment consists of a hold-up line which allows any short-lived activity in the off gas process stream to deca The third segment consists of cooler condensers, moisture' separators,

,

prefilters, desiccant-type gas dryers, gas coolers, charcoal adsorber trains, i and after filters which remove trace moisture, allow adsorption of iodine, and delay the passage of Krypton and Xenon.

i The charcoal adsorber trains are comprised of eight charcoal adsorber

,

vessels (each 25 ft. high by 4 ft. in diameter) arranged in two trains of four vessels each, numbers 1N640012A and B, 13A & B, 14A & B and 15A &

I Each adsorber vessel contains approximately three tons of activated coconut shell charcoal. The adsorbers are located in the Off-Gas System i charcoal vaults in the Off-Gas Building. The first adsorber in each train 4 is located in a separate vault. The remaining three adsorber vessels in each train are located in another vault along with two process gas cooler .The Off-Gas Vault Refrigeration System cools the four vaults in which the charcoal adsorbers are located. It is designed to maintain the temperature in the vaults at approximately 0*F for maximum efficiency of the charcoal adsorption proces Each adsorber vessel has three thermocouples approxi-mately equally spaced vertically through the charcoal vessel . System Test Review During the inspector's review of testing related to the Off-Gas System

and the fires which occurred in that system, concerns developed in the

areas of System Quality Designation, Procedure Inadequacies, Charcoal l Ignition Temperature, a Previous Acceptance Test, and Test Review. These areas are discussed in the following paragraphs 4.a. through 4.e.

l

! 3

. . - . . - , - + . - ,.-~,..---,,,---,_,,_.-n_-__n.,,. - _ - - , - , , , , . - , _ - - . - - - _ , - , . . , , , - , .

.

_ __ _.

. . System Quality Designation: Criterion V, " Instructions,' Procedures and Drawings" of Appendix 8 of 10 CFR 50 applies to the subject portions of the Off-Gas System. The Perry Nuclear Power Plant, Operations Quality Assurance (0QA) Plan, Section 5, " Instructions, Procedures and Drawings", Revision 3, effective November 14, 1986, implements 10 CFR Part 50, Appendix B, Criterion V. Appendix M of the 0QA plan states that Criterion V is applicable to those components of the Off-Gas System which are designated Augmented Quality on the Perry Q-List. Augmented Quality is defined in Perry Power Plant-Huclear Design Procedure 91-0355, Revision 3, effective May 5, 1986 as Safety Code 4. The Q-List (defined in the Glossary of the Quality Assurance Plan) is the controlled list of all permanent plant equip-ment and items which provides the safety and seismic classifications assigned to each piece of equipment or item. The Q-List designation of the N64 (Off-Gas) System lists several components as safety code 4 including the charcoal adsorber vessels and the thermocouples within the vessels, Procedure Inadequacies: (1) Attachment 1 to the Generic Procedure, GEN-M-021, Data Sheet actually details the testing that caused the fire in the charcoal. Attachment 1 directed the heating of the eight charcoal adsorber beds utilizing external heaters and an internal warm instrument air supply through the N64C (Off-Gas System) process path (i.e. through the vessels). Attachment 1 was inappropriate for these activities affecting quality in that neither Attachment 1 nor any of the procedures used for this test described the heating-process or what kind of heaters to use or where to place the Consequently, some of the heaters were placed approximately 2 feet, and possibly within 1 foot, from the charcoal adsorber bed vessels which ultimately resulted in temperatures of approximately 1000 F in the metal of the vessel immediately adjacent to the charcoal. The licensee made these calculations in Attachment 1, " Radiant Heater Temperature Analysis Calculations," of Section 5 of " Analysis of the Mode of Ignition and Combustion Propagation for both Charcoal Adsorber Bed Combustion Events," contained in the " Charcoal Adsorber Combustion Events Final Report."

The proximity of the heaters to the charcoal adsorbers was not in agreement with Step 2 of the installation instructions from the l radiant heater manufacturer, Aitken, which stated that combustible  !

material should not be placed in the radiation path within 5 feet of a heater. The Aitken installation instructions for the heaters were  !

not used during placement of the heaters for the tes ;

The testing procedure did not specify any method for controlling or i monitoring the temperature of the metal of the vessels induced by the '

heaters. For example, it did not direct the attachment of temperature measuring test equipment to the surface of the vessels near the radiant heaters during the heating of the vessel Test personnel did not understand that at close proximity the radiant heaters would deposit most of their energy into the metal of the vessels. This lack of understanding resulted in the temperature of the air of the vault, being monitored instead of the metal of the

. . .- - -- - . _ - - . - - - - - -- - - - _ _ - - -

,

,

. . . _ ._ _ . _ - _ ._ . . .- _- ._

. ~.

,

!

~

vessels closest to the heaters. Licensee personnel were cognizant o the placement of the radiant heaters because it is stated in the official chronology that on' June 17, 1986, at 1200 hours0.0139 days <br />0.333 hours <br />0.00198 weeks <br />4.566e-4 months <br /> PPTD Technical Section engineers performed vault inspections for-fire safety concerns

(i.e. verify no combustion hazards) and acceptable heater placemen This official chronology.was a part of.the-Investigation Summary for;

.

-

'

Condition Reports;86-595,86-596 and 86-656. These Condition ReportsL

. officially ~ document the licensee's investigations of.both Charcoal

, Adsorber Bed Combustion Events and the entire submittal will be-hereinafter referred to as.the Investigation Summar Failure to

~ develop an adequate procedure as detailed in this subparagraph 4.b.(1) is considered to be an example of a violation of 10 CFR 50, Appendix'B, Criterion V (440/86020-01a(DRS)).

'

(2) In paragraph 4.3.(2) of this report a discussion is provided-regarding the lack of appropriate review for the testing of the

Off-Gas System due to the use of the Data Sheet Attachment of f GEN-M-021 as a test procedure in lieu of a specific procedure. In addition to this problem, however, the Data Sheet for GEN-M-021'did

not contain all the necessary information such that, even if the required reviews had been performed, the reviewers may not have been.

able to fully appreciate the fire hazard involved. The information

-

which was missing concerns the Aitken radiant heaters and their size, ,

i capacity and placemen The data sheet of GEN-M-021 did not list the i

Aitken radiant heaters as test equipment in the test equipment block

._ of the Data Sheet as required by Step 4.4 of GEN-M-021, " General

Mechanical Equipment Initial Checkout and Review Test Procedures,"

Revision 2. None of the procedures used for this test described the heating process nor did they indicate what kind of heaters to use or

{ where to place them. Had this test procedure been prepared properly, i as a specific procedure in accordance with Test Program Instruction j ~TPI-6, this omission of significant equipment should not have occurre '

TPI-6 requires the listing of any test equipment needed to perform the procedure (which is not permanently installed plant equipment) and provides examples such as temporary pumps, compressors and blower TPI-6 further requires that attributes of equipment such as model number, size and capacity be specified. The radiant heaters, which

. were' rated.at 6 and 13.5 kilowatts, were significant pieces of test'

! equipment and should have been described in the test. This omission

! of significant equipment from the test further supports the violation I previously described in 4.b(1) that the test procedure was inadequate.

< Charcoal Ignition Temperature: Even if the temperature of the charcoal

,

had been closely controlled during the testing, the actual ignition

temperature of the charcoal would probably have been exceeded due to the licensee's lack of knowledge regarding charcoal ignition temperatur ,

Licensee personnel informed Region III personnel as late as June 23,

! 1986, that the minimum ignition temperature of the charcoal was about

1000*F. Upon investigation the licensee found that the supplier of f the charcoal, the Barneby-Cheney Company, would ' certify the minimum L ignition temperature no higher than 250*C (or 482*F) as specified in

'

GE Purchase Specification for the charcoal (Specification 21A9375,

, Revision 2, Step 4.1.1.6, Sheet 4) which states, "that the ignition i

i

5 I

a -__ _ - .. ._ - _ _ - . -.:

>

s . .

l temperature shall-not be less than 250 C (or 482 F) measured in flowing air, in accordance with the ASTM standard methods for testing ignition temperature."

Furthermore, in testing conducted June 30 through July 6, 1986, the Barneby-Cheney Company found that the ignition temperature of 250 C (482*F) as determined by the ASTM standard methods was not appropriate to the subject testing because the. face velocity of the air flow for the ASTM method was 100 feet per minute (fpm) whereas the face velocity of the air flow induced by the subject testing has been determined to be 4 fpm. A face velocity of 4 fpm has been subsequently found to reduce the ignition temperature of the original charcoal to as low as 153 C (307.4 F). Therefore, even if the licensee had carefully moni-tored the highest temperature of any part of the metal of the vessels, closely controlled the heat input, and used 482 F (250 C) instead of 1000*F as their upper limit, the actual minimum ignition temperature of 307.4*F would probably have been exceeded causing ignition of the charcoa The documented instructions and procedures regarding the subject testing made no reference to any consideration of the ignition potential of the charcoa Likewise there were no precautions in the procedure to terminate heating upon indication that charcoal ignition had occurred nor did the procedure prescribe actions to be taken if a fire did occu The inspectors remain concerned that operational modes of the Off-Gas System are properly evaluated for charcoal ignition. This concern will be tracked as an unresolved item (440/86020-03(ORS))

pending demonstration that the current understanding of the adsorber bed charcoal ignition temperature (307.4 F) does not compromise any mode of normal or abnormal off gas system operation or testin d. Previous Acceptance Test: Based on the inspectors review, "The Off-Gas Vault Refrigeration System (0GVRS) Acceptance Test", 1N64-A001, Revision 0, is another example of an inadequate procedure controlling

-

activities affecting quality. This test which had been performed in May 1985 (prior to the fire) also involved heating of the vault with space heaters. No combustion occarred in the adsorber beds during the heating process directed by this acceptance test, because there was a nitrogen blanket on the adsorber vessels and smaller radiant heaters were used for a shorter time period and only the air in the vaults was heated to 150 F. However this test was also inadequate for the following reasons:

• The acceptance test did not list test equipment as required by TPI-7, "Preoperational, Acceptance and Special Test Procedure Preparation, Review and Approval, Revision 4, Step 4.1.2. There were no steps in the test instructions addressing the heating of the adsorber bed charcoal and vessel * There were no precautions to protect equipment as required by Step 4.1.2.2.d. of TPI-7.

i 6

, .

- - - - . - - - - . - -- ., . - _

_ _ ____ _ ____ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ - - - - - - - - - - - - - - _ - - -

.

,

. . Test Review: Attachment 1 to the GEN-M-021 Data Sheet did not receive the reviews that it should have received. These reviews (had they occurred) might have detected and corrected the inadequacies in the test methodology of Attachment 1 and previnted the fir (1) The first omission of a required review was caused by the incorrect safety classifications of Off-Gas System components on Work Order (WO) 860007034 used to conduct the testing. The WO incorrectly designated the testing as affecting mechanical and electrical components of safety Class 5. However, the testing used the thermocouples (safety class 4) in the charcoal s beds to take test data. The charcoal beds (safety code 4) were involved in the testing in that the charcoal was heated by powerful radiant heaters, the instrument air purge through the charcoal beds was clearly delineated in the work order package, and the entire point of the testing was to pull the temperature of the adsorber beds from 150*F to 0*F and maintain the temperature at 0*F for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The Perry Quality Items Classification List (Q-List) correctly identifies the thermocouples and charcoal beds as safety code 4. The planner incorrectly designatet the work order safety code 5. The work order was also written against the entire Off-Gas System which is uncommon and requires the planner to purposely decide the safety classification. Due to the incorrect safety classification of code 5, the WO was not given a Nuclear Quality Assurance Department (NQAD) review per NQAD-0541, which would have been required by step 6.3.1. Consequently, this misclassification denied NQAD the opportunity to review the documents as required. Their revfew possibly would have questioned the lack of adequate instructions regarding the external heaters. Incorrect safety classification of the WO is considered to be an example of an inappropriate procedure and a violation of 10 CFR 50, Appendix B, Criterion V and shall be tracked as item (440-8G020-01c(DRS)). The inspectors' are also concerned that other W0s may be incorrectly classified as to safety code. This subject will be followed as an open item (440/86020-04(DRS)) pending licensee demonstration that safety code classification of W0s is prope (2) The second omission of a required review occurred because GEN-M-021, " General Mechanical Equipment Initial Checkout and Review Test Procedures," Revision 2, was used to write the testing procedure (Attachment 1 to the Data Sheet of GEN-M-021)

instead of writing a new specific generic procedure in accordance with Test Program Instruction (TPI-6), "IC&R Test Procedure, Temporary Operating Instruction and Alarm Response Instruction Preparation, Review and Approval," Revision GEN-M-021 does not require a pre-implementation Test Program Review Committee (TPRC) Review and approval for testing proce-dures on Augmented Quality equipment such as the Off-Gas Syste A requirement for TPRC review includes an Operational Quality !

Section (0QS) review due to 0QS membership on the TPRC. Unlike GEN-M-021, TPI-6, Revision 6, Stcp 4.2.2 does require such a revie ___

m .

.

. .,

!

'

Lack of TPRC review for testing is not in'accordance with the Operations Quality Assurance (0QA) Plan, Section 5, Step 5. !

which states, " Temporary procedures and instructions shall be reviewed and approved by the organizational elements that review  ;

and approve normal procedures for the affected activity." Normal

'

generic procedures for testing activities for Augmented Quality components get review and approval by TPRC (with OQS membership)

as required by TPI-6, Step 4. Therefore,.in this case the temporary procedure, Attachment 1 to GEN-M-021, should have been reviewed and approved by the organizational element (TPRC) that reviews and approves normal procedures for the affected activity (testing). This lack of appropriate review and approval for the testing procedure is considered to be another example of a violation of 10 CFR 50, l Appendix B, Criterion V (440/86020-01c(DRS)). l

In the third paragraph of Section IV.A.1 of the Investigation ,

Summary the licensee seems to contradict this NRC finding when l it states that GEN-M-021 was used to conduct the test. This is ,

incorrect. Atta.hment 1 to the Data Sheet of GEN-M-021 actually detailed the testing and therefore conducted the tes GEN-M-021 does not detail any testing but rather directs that the detailed steps for testing be referred to by the data sheet. In this case  :

the System Test Engineer (STE) was to write the detailed steps *

'

and attach them to the GEN-M-021 Data Shee In the very next

'

sentence of the summary the' licensee states that, "This procedure was approved by the TPRC". This implies that the TP!iC approved  :

the procedure that detailed the testin This is incorrec ,

, When the TPRC approved GEN-M-021, they only approved a procedure that directed the System Test Engineer (STE) to fill out a data i sheet and the blank data sheet itself. The actual detailed test I steps (Attachment 1 to the GEN-M-021 data sheet) were not reviewed and approved by TPR ,

!

There is also the possibility that implementation of the GEN-M-021 ;

procedure may have generated other inadequately reviewed test i procedures. This will be tracked as open item (440/86020-05(DRS))  !

pending review of procedures generated by GEN-M-02 ! Sequence of Events - Fire Protection Review Licensee-Staff Actions Regarding Off-Gas System Charcoal Adsorber Bed Fires As previusly stated, on June 19, 1986, a fire was ignited in the Unit 1

,

off gas system charcoal adsorber beds. . Combustion was halted using a '

,

nitrogen purge process'on June 23, 1986. The fire rekindled on July 6, 1986, following reintroduction of oxygen into the adsorbed beds. Combustion was again halted using a nitrogen purge process on July 8,1986. The fires ,

were caused by a testing evolution in progress to evaluate the effectiveness-of the adsorber bed vault refrigeration system using radiant type space heaters as a heat source on the vault .

.

_ ., =____-__ _ . _ , . . . - . _ _ _ _ . _ _ _ _ _ _ _ - . _ . _ . ,

• .

a. Chronology of Events An abbreviated sequence of events for the first fire is provided belo June 18, 1986 11:00 Radiant type space heaters were energized to establish 150'F ambient temperature and equipment sensible heat for the refrigeration system testing initial condition June 18-19, 1986 Adsorber bed vault and equipment temperatures were gradually and satisfactorily increasin June 19, 1986

'

6:00 Clean dry air flow was established through the adsorber beds to simulate off gas process flow during refrigera-tion system testin :20 Adsorber bed vault and equipment temperatures were gradually and satisfactorily increasin :30 The heater in the vault containing adsorber bed 12B was deenergized due to smoke and pungent odor noted outside the vaul :00 The smoke and pungent odor cleare :15 Adsorber bed 14A bottom temperature element experienced an abnormally high temperature increase (10"F in the previous hour).

6:20 Heaters in the vault containing adsorber beds 13A, 14A and 15A were secure Heater in the vault containing adsorber bed 12B re-energized due to satisfactory environmental conditior.s and adsorber bed temperature :15 Adsorber bed 14A bottom temperature element decreased 14*F since securing the heaters in that vaul :41 Personnel noted that adsorbed bed 148 mid-height temperature element had very rapidly increasef from 112 F to offscale high (>250 F) and adsorber bed 14B bottom temperature element had increased from 157 F to 205 F in ninety minute :10 All heaters in all adsorber bed vaults were secure I

'

--

- - . . . -.

-- .-. _ _ - _ - - .

- - . . . - . . - ._ .

. .

,

9:20 Personnel'noted that adsorber bed 14A mid height

- temperature element had very rapidly increased from approximately 126'F to offscale high (>250'F) and ,

-

adsorber field 148 bottom temperature element had continued increasing to offscale high (>250*F).

'

Shift supervisor dispatched shift Instrumentation and Control (I&C) technician to obtain temperature element ,

readings at the local instrument rack for those devices indicating offscale high in the control roo ;

I 9:25' I&C technician reported back to the shift supervisor that adsorber bed 14A bottom temperature element was reading  !

575'F and the other elements were failed based on his

'

observations that their output voltage, converted to

'

temperature, indicated in excess of 1000*F, and they exhibited a measured resistance of 3 megches.

9

45 p.m, Operations personnel observed that no smoke was emanating from the adsorber bed process stream discharge poin Based on the apparent lack of combustion products j coming from the adsorber bed process stream and the apparently failed detectors, the shift supervisor concluded that the high vault temperatures must have

- damaged temperature element cabling in the vault causing the erroneous reading

45 Light smoke was observed outside the adsorber bed vault area.

11:15 Operations personnel concluded that the smoke was

,

due to smoldering temperature element cable insulation.

,

-June 20, 1986

, 12:15 Fire Protection Coordinator opened the adsorber bed l vault top hatch (vault temperature was 165*F) and '

observed no open flame, but the adsorber bed vessel coating was fuming.

!

'

Shift Supervisor initiated vault'cooldown to facilitate

,

personnel access for physical inspectio t

, 12:30 Personnel noted that adsorber bed 148 top temperature element rapidly increased from approximately 220*F to

,

offscale high (>250*F).

.

12:45 Test and Operations personnel began trying to determine which drawings described the temperature element cable 1 layout so they could establish the verac i ty of thei r hypothesis that the erroneous detector readings were caused by vault high temperature $

. .

,_w +, y-4,- , , - - - .- ,,m.,,_y.,-w-o-,,.,._ -c.,,.-r__.. ,,_ . _.-.___,.__.,m-,-. _- , , . ,,_.,,--_v-,-,-,, _ , - . -. . - r-_,.

. . . _ . - . .. - -_ _ _ .-- - - -

. .

'

!

2:30 Personnel noted that adsorber bed 14A top temperature element rapidly increased from approximately 200*F to offscale high (>250'F).

Adsorber bed vault temperature had been lowered from 165'F to 128'F in approximately two hours by operation  !

'

of the vault refrigeration system.

4 4:30 Personnel noted that adsorber bed 15B bottom temperature element increased high offscale (>250*F).

6:30-7:30 Test personnel located conduit layout drawings for the temperature element cablin :30 Test personnel completed review of the conduit layout drawings and refuted the hypothesis that a common mode

.; failure of the temperature elements was occurring due l to heater placement and high vault temperatur :30 Personnel noted that adsorber bed 15A bottom temperature

element increased offscale high (>250'F).

7 Test personnel discussed their findings from the conduit ',

l layout review with operations personne '

l Adsorber bed vault temperature had been lowered to 85*F i- by the refrigeration system and personnel-access was

possibl

00. Vault entry was made and adsorber bed 14B vessel top .

contact temperature reading was 254" Process air flow through the adsorber beds was secured and the vessels were isolate :00 Operations personnel began preparing for adsorber bed

. nitrogen purge operation '

11:45 An Unusual Event was declare :19 Nitrogen purge of the A and B adsorber beds was initiate June 23, 1986 11:25 The Unusual Event was terminated after all adsorber bed temperatures and effluent product of combustion measurements clearly indicated that internal combustion had cease l

i

.

+ = - - - ,- -,...--..-,.,,--.,-em _,w,---ny, %-m-p -.%, , ,-,+.-, , - - - ...-3,, ---.,eww.w,, -,.,,....w.-,,----t+- -=--pwe,-+y,*ew"*

. .

Sequence of Events -Rekindled Fire July 6, 1986 6:00 Clean dry air purge initiated through the adsorber bed :10 Adsorber bed 14A and 14B mid height temperature elements rapidly increased to offscale high (>250*F).

7:30 Shift supervisor dispatched:

I&C technician to take temperature element readings at the local instrument rack, Operators to secure the adsorber bed air purge evolution, Operators to prepare for adsorber bed nitrogen purge evolution, and Chemistry technicians to sample adsorber bed effluent for products of combustio :42 Air purge of the adsorber beds was secured and the beds were isolate :32 Adsorber bed 14A mid-height temperature element was reading 656*F. Adsorber bed 148 mid-height temperature element was reading 578 :37 An Unusual Event was declare :07 Nitrogen purge of the adsorber beds was initiate July 8, 1986 4:45 The unusual event was terminated after all adsorber bed temperatures and effluent product of combustion measurement clearly indicated that internal combustion had cease Implementation of Fire Protection Requirements Review (1) The Perry Nuclear Power Plant Facility Operating License N NPF-45, Section 2.c.(6) states that CEICo shall implement and maintain in effect all provisions of the approved fire protection program as described in the Final Safety Analysis Report (FSAR)

Section 9A.6 as approved in the Safety Evaluation Report (NUREG-0887) dated May 1982, and Supplement Nos. 1 thru 9 thereto, subject to the following provisions:

1

e

.. .

CEI may make changes to the approved fire protection program without prior approval of the Commission only if these changes would not adversely affect the ability to achieve'and maintain safe shutdown in the event of a fir *

Section 9A.6 entitled " Point-by-Point Comparison" of the FSAR to Appendix R,Section III.K, Administrative Controls," requires that administrative controls be established to minimize. fire hazards in areas containing structures, systems and components important to safet *

'The applicant's response-to the stipulation above stated,-"CEI will comply" as documented in CEI letter dated April 29, 198 (Letter from D. Davidson, Vice President, to A. Schwencer, NRR, dated April 29, 1982 Section III.K entitled " Administrative Controls".)

Supplement No. 1 of the Safety Evaluation Report (NUREG-0887 Supplement No.1) related to the operations of Perry Nuclear Power Plant Units 1 and 2, Section 9.5 entitled " Fire Protection Systems," states that the applicant documented his commitment to implement the fire protection program and administrative controls delineated in Appendix R to 10 CFR 50, Section II *

Section 1.B of the FSAR the applicants' retponse to Appendix A of the NRC Branch Technical Position APCSB 9.5-1 entitled

" Administrative Procedures, Controls and Fire Brigade", states that CEI has implemented the following Plant Administrative Procedures (PAP):

1910 Fire Protection Program 1911 Fire Emergency 1912 Burn Permit for Ignition Sources 1913 Control of Transient Combustibles-

'1914 Impairment 1915 Fire Report '

1916 Duties of Fire Watch 1917 Fire Protection Training Programs 1918 Fire Drills 1919 Fire Brigade Equipment Inspection, Inventory and Maintenance  ;

1920 Periodic Fire Inspections 1921 Fire Barrier Removal 1922 Pre Fire Plans (2) The use of space heaters during this testing resulted in the ignition of the charcoal. Controls for space heaters is not included in the plant procedures listed above. It appe..s that the licensee has not provided administrative controls for portable space heaters such as those used in the subject testing, and is therefore in violation of Section III.K of Appendix The violation described in the foregoing paragraphs, is identified as item number (440/86020-02(DRS)). ,

t I 13

-

+

, 3-

- ,- - , , . - , , , . - . , - . _ ~ . . , - . _ - . . . , , . -,,-4 --..m , , _ _ . -..,.-.,,,m , , _ , . , - . . , _ ,m_,_- n__----+-

- -

,-

• .

(3) The inspector also reviewed the related work order #860007034 dated June 17, 1986 which indicated no special permit (such as a burn permit, fire barrier removal permit, transient combustible permit etc.) or special precautions to be taken during this work activit The work order did not specify any precautions as specified in the radiant energy heater manufactures instructions to prevent the ignition of combustible material c. Fire Protection Organization Involvement Although the Fire Protection Coordinator was requested to assist in the investigation of the charcoal adsorber beds on Thursday June 19, 1986 at 2230 hours0.0258 days <br />0.619 hours <br />0.00369 weeks <br />8.48515e-4 months <br />. The fire protection engineer was apparently not immediately notified. It is our understanding that the fire protection engineer became aware of this event through the routine reading of the daily log. This does not appear to be appropriate or comprehensiv This issues is considered an open item pending further review and discussion with the licensee (50-440/86020-06(DRS)).

d. Analysis of Licensee Staff Actions From analysis of the adsorber bed internal temperature profiles, it appears that a charcoal fire ignited inside the 14A and 148 adsorber beds between the times of 6:15 pm and 9:30 pm on June 19, 1985. By 9:20 p.m., two temperature elements, 14A and 148 mid-height location, had essentially instantaneously increased off scale and one element, 148 bottom location, had trended offscale high. The shift supervisor initiated proper actions to assess actual bed temperature by dispatch-ing an I&C technician to obtain local readings. At 9:25 p.m. the technician reported the mid height detectors as failed which was corroborated by the manner in which the detectors responded as shown on the chart recorder in the control roo Post event testing of thermocouple response by the licensees I&C section has been unable to reproduce conditions as observed by the I&C technican; e.g., millivolt readings corresponding to 1000 F -

2000*F and 3 megohms resistance reading. Additionally, the licensee found those thermocouples undamaged and within calibration tolerances after the fir If one of the thermocouples had failed open or if the technician had attempted to measure resistance using improper terminal points, with no continuity, the measuring device used by the technician, a Fluke 77 Multimeter, on reading resistance, would not provide a numerical readout, but would indicate an open circuit. It appears that the technician's conclusion that the thermocouple had failed was erroneous. That information significantly misled operations personnel in their decision makin Under the belief that two thermocouples had failed; that the other temperature elements were either onscale (>250 F) or below what was thought by the shift supervisor to be the ignition point of charcoal; with apparent lack of combustion products in the off gas system effluent and the existence of smoke and/or fumes emanating from the adsorber bed vaults, the shift supervisor concluded at 9:45 .-. . . - -.. . .

r 1

. .

that this event was due to high temperatures external to the adsorber

< beds. During the evening and early morning hours, several other observations were made which could have alerted the operations staff ,.

that an internal bed fire was occurring:

- Four other temperature elements trended offscale,

- Visual observations indicated that the surface coatings on the adsorber beds with high indicated internal temperatures were fuming three hours after de-energization of the heaters, and

- The internal indicated temperatures on adsorber beds 15A and B continued trending up while vault temperature was decreasin Two actions could have been taken or expedited by. operations staff which would have facilitated earlier recognition of the internal combustico even Conduit layout drawings for locating the thermocouple cables could have been identified and obtained more expeditiously, and ,

'

- Local temperature readings could have been obtained on other temperature elements which were "failing" off scale high as indicated on the control room chart recorder Early completion of those actions may have resulted in earlier identification of the fire. Once the conduit layout diagrams were located, after approximately six hours, it was quickly concluded that it was unlikely that a fire in the vault external to the adsorber beds was the cause of the indicated high temperature e. NRC Concerns It appears that the licensee did not respond to the information indicating a fire in a timely manner as indicated by' the following:

(1) After it was identified at 0015 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> on June 20, 1986 that the fire protection coordinator did not observe any flames but the paint on the charcoal vessels was fuming, effective action was not taken to verify if the charcoal was burning or determine the source of this hea (2) At 0400 hours0.00463 days <br />0.111 hours <br />6.613757e-4 weeks <br />1.522e-4 months <br /> on June 20, 1986 approximately 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after the fire coordinator observed the vault, the vault temperature dropped from 165 F to 125*F but the charcoal adsorber temperature instru-ments trer.ded upward. This should have indicated the presence of a internal fir (3) Wnile this fire was not a significant safety evf.nt, there was a lack of timely action on the part of operations personnel to substantiate their theories which was further compromized by the inaccurate and unaggressive support from I&C and test personne This (itemt e.(1), (2) and (3) above) is considered an open item (440/86020-07(DRS)).

,

,

. .

- Related Issues - Adsorber Bed Charcoal

.- Off-Gas System - Storage, Maintenance and Installation The inspector reviewed and found no problems with the licensee's documentation regarding the purchase orders, certifications, storage, handling, maintenance, surveillances and installation of the original charcoal and adsorber vessels from their original purchase through installation up to but not including the performance of "The Off-Gas Vault Refrigeration System Acceptance Test." Activated Charcoal Analysis Review The inspectors reviewed the applicant's analysis of the physical and chemical properties of the original and replacement activated carbon and interviewed licensee representatives concerning sampling techniques, laboratory methodology, and test result acceptance criteria; no significant problems were note Witnessing of Adsorber Bed Charcoal Sampling and Unloading Activities (1) On July 2, 1986 the inspector witnessed the sampling of the 12B '

adsorbed bcd through the bottom thermocouple port. The inspector found the licensee personnel sufficiently knowledgeable of all activities associated with the samplin Area cleanliness was sufficient to prevent introduction of contamination into charcoal adsorber vessel 128. The thermocouple was maintained adequately while removed to prevent contamination of the adsorber bed when the thermocouple was reinstalle The sampler was driven radially into the adsorber be The sampling was done in accordance with approved procedure N OM19E:RT1-0800, Revision 1. The sampler was inserted approximately two hundred fifty time The inspector looked into the port about fifty times after the samples were removed from the adsorber bed. The charcoal always maintained the shape in the port that existed previous to the insertion of the sample The charcoal appeared to act suffic-

-iently like a fluid that the inspector has no concerns for voiding in the charcoal as a result of sampling. The inspector could not discern by visual observation any evidence of ash or contamination in the samples take On July 2, 1986 the inspector witnessed the sampling of vessel 14B through its bottom thermocouple port. This process was identical to the satisfactory sampling of vessel 128. During these sampling activities the inspector witnessed the satisfactory removal and replacement of the bottom thermocouples of vessels 128 and 14 On July 3, 1986 the inspector observed the satisfactory sampling of vessel 12A. The process was essentially identical to the

'

sampling of vessel 12 .

_ _ - , - .- -- .. ..

- -- - - -

_ _ _

_

,

- .

G (2)' NRC Witnessing of Charcoal Adsorber Bed Unloading Activities At 10:30 pm on July 14, 1985; the inspector attended a briefing '

~ before the wet charcoal unloading. The briefing was informal, brief and seemed to assure that personnel were adequately

. prepare The Mechanical Test Instruction (MIT) 0006, " Unloading, Drying and Loading of N65 Activated carbon Adsorber Vessels", Revision 0 was performed under Work Order (WO) 860010348 and'its attached job. traveler. The water spray nozzle was attached about 36 inches above the bottom of the vacuum hose..The vacuum truck was grounded to the metallic. vacuum hose sections which were carbon steel with galvanized coating. The hose sections were in turn grounded to a-building ground. Hose clamps on the four inch diameter vacuum hose were determined to be too large to allow the hose to be lowered to a sufficient depth into the vessel to accomplish complete removal.of the charcoal. Therefore, all base dimensions were deleted from the drawings attached-to MTI-0006, Revision 0 by TCN-001 and the Recovery Plan. Health Physics adequately

. sampled air and ensure safety. A charged fire hose was staged in the Off-Gas Buildin Photographs and videotaping were made of the inside of the first vessel opened, 15A, which w6s opened at approximately 3:15 July 15, 1986.

. Water was finally applied to the adsorber bed about 3:40 a After about 100 gallons of water was injected the vacuum was initiated and charcoal could be heard rattling up the hos Water at approximately 70*F trickled out of the 15A and 14A

drains and seeped out of the bottom thermocouple flange of 14A.

! This indicated that water sprayed into the 15A vessel was filling

.

'

the bottoms of the 14A and 15A vessles through the connecting piping.- The indicated completc. and thorough soaking, ensured

-

that the fire was out and that any hot spots were cooled by the soaking wate . Metallurgical Evaluation

!

In an' effort to evaluate potential degradation of vessel mechanical properties t'ue to the thermal excursions (charcoal fire), the licensee -

performed a metallurgical analysis of the vessels involved in the even !

The analysis' included a metallurgical comparison of those areas exposed '

i

, .

to heat with those areas not involved in the incident. This comparison

'

.

included visual examination, hardness testing, and examination of the .

microstructure, i.e., grain size and carbide morpholog )

I

'

'

The visual internal examination was performed to identify any structural

! damage or formation of high temperature oxide scales. The relative uniform appearance and the lack of oxide scales on the interior surface

- indicate that the temperature excursion was not in excess of 1500 F.

!

l

4 i

y w r*mrg- .= -yy--ow ,-w-y ge---s-eew- -

wy- c y y---w-e----- ---w-,--,w --i-trerr'-Dr--up-ew- * - - ++*---r~'w-er - " - 'r-**rw--eevi wae -car -w----e i we w h

r

, .

A comparison of metallurgical replicas taken from areas of the vessel where paint had blistered and areas where no heat damage to the paint had occurred, revealed no apparent change in gain size or carbide morpholog An increase in grain size or a change in microstructure would have been indicative of potential degradation of impact toughness and possibly tensile strengt Changes of this type would be expected to occur at temperatures in excess of approximately 1250* Hardness tests.were performed in the same areas that the replicas were taken from. The comparison of hardness values indicated that no significant change in hardness had taken place. A change in relative hardness would have indicated an attendant change in tensile propertie In an effort to further explore the potential of vessel material degradation, a steel plate of similar chemical corrposition and processing was heat treated to attempt to induce the phenomenon of temper embittlement. Temper embittlement typically occurs in certain alloys at a temperature of approximately 650 F to 1050 F. The occurrence of temper embrittlement causes a reduction of material fracture toughness. This sample of material was exposed to elevated temperature considered optimum to induce temper embrittlement. The results of Charpy impact testing of this sample both before and after the heat treatment indicates that no significant loss of fracture toughness would be expected from exposure to these temperature An analogous test was conducted on weld material similar to that used for the vessel fabrication. These results were consistent with the positive test results achieved by testing plate material in that no significant degradation of fracture toughness was apparen In summary, the testing, examination, and evaluations performed on the vessel materials indicate that no significant damage was caused by the thermal excursio . Unresolved Items Unresolved items are matter about which information is required in order to ascertain whether they are acceptable items, violations, or deviation The unresolved item disclosed during the inspection is discussed in Paragraph . Open Items Open items are matters which have been discussed with the licensee which will be reviewed further by the inspector, and which involve some action on the part of the NRC or licensee or both. The open items disclosed during the inspection are discussed in Paragraphs 4 and . Exit Interview j The inspector met with licensee representatives denoted in Paragraph 1 on l August 2.1, 198 The inspector summarized the scope and findings of the l inspection and discussed the likely content of this inspection report. The !

licensee did not indicate that any of the information disclosed during the j

inspection could be considered proprietary in natur j l

l

l

__ ___ -___