IR 05000346/1985028

Insp Rept 50-346/85-28 on 850903-1031.Violation Noted:No Procedures Provided for Fabrication,Installation & Insp of Boot Seals

ML20137C062
Person / Time
Site:	Davis Besse
Issue date:	11/21/1985
From:	Guldemond W, Holmes J, Maura F, Ulie J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20137C005	List: IR 05000903/2010031 ML20137B999 ML20137C062
References
50-346-85-28, NUDOCS 8511260421
Download: ML20137C062 (22)

Text

. .

U.S. NUCLEAR REGULATORY COMMISSION

REGION III

.

Report No. 50-346/85028(DRS)

Docket No. 50-346' License No. NPF-3 Licensee: Toledo Edison Company Edison Plaza 300 Madison Avenue Toledo, OH '43652 Facility Name: Davis-Besse 1 Inspection At: Oak Harbor, OH .

Inspection Conducted: September 3 thorugh October 31, 1985

-

SM pl.g/p-Inspectors: . olm '

~Date

. Ma a llvkl Date

. U11 M[

Approved By: . emond, Chie f/u/p/

Operations Program Section Date ~

Inspec_ti_on n Suninary Inspection on September 3 through Octobe_r 31,1985(R_eportNo. 50-346/85028 (DRS)) Areas _ Inspected: ~EutTn'e, t announced inspection to assess the tei:YnTEal status of outstanding fire protection issues including a review of the fire

_

protection program as it relates to the outstanding issues, and a review of allegations received by the NRC relative to penetration seals. In addition, leak testing of containment purge valves was reviewed. The inspection involved a total of 138 inspector-hours onsite and in-office review by three NRC inspectors including 17 inspector-hours onsite during off-shift Results: In the four areas inspected, no apparent violations were identified in three areas; one apparent violation was identified in the remaining area (no procedure (s) existed which provided for the fabrication, installation, and inspection of the plant flexible boot seals - Paragraph 4).

8511260421 851122 PDR ADOCK 05000346 9 PDR u

.

.

DETAILS 1. Persons Contacted Toledo Edison Company

• B. Beyer, Nuclear Projects Director

• Connor, Assistant Plant Manager Operations

• T. Hart, Fire Protection Engineer

.

"

• J. Helle, Nuclear Facility Engineer-Director

• J. Kapser, Mechanical Superintendent D. Missig, Plant Engineer

• D. Momivez, Quality Assurance Supervisor

• Murtla, Fire Protection Engineer

• T. Myers, Nuclear Safety and Licensing Director W. Nissia, Fire Protection Coordinator

• S. Quennoz, Group Director Nuclear Engineer

• L. Storz, Plant Manager

• A. VanDenabele, Project Manager

• S. Wideman, Senior Licensing Specialist

• J. Williams, Senior Vice-President

• J. Wood, General Superintendent-Facility Engineer Cygna

• J. Haverly, Licensing Support USNRC

• D. Kosloff, Resident Inspector

• Rogers, Senior Resident Inspector The inspectors also contacted other plant personne ,

• Denotes persons attending the exit meeting of September 6, 198 . Action On Previous Inspection Findings During the period July 11-29, 1983, the NRC conducted a special safety inspection of the licensee's implementation of the requirements of i

10 CFR 50,' Appendix R (Sections III.G, J, L, and 0) and the fire pro-tection program. Numerous deficiencies were identified during that inspcction as documented in Part I of Inspection Report No. 50-346/8301 Upon review of these deficiencies, the NRC concluded that certain issues had to be resolved prior to plant restar During September 1983, a followup inspection was conducted to establish the status of those issues identified as restart prereouisites. As identified in Part II of Inspection Report No. 50-346/83016, all restart issues were satisfactorily resolved; however, they were administrative 1y left open pending receipt of a formal response from the licensee on those issues.

,

,

. . -. . .. -- - - - . - -

_,

. .

During the period September 3 through October 31, 1985, a followup inspection was performed by Region III inspectors to revisit a number of issues identified in Inspection Report No. 50-346/83016: remaining open. The status of those issues is discussed belo (0 pen) Violation (346/83016-01A): The alternative shutdown capa-bility for the control room and cable spreading room did not meet the requirements of 10 CFR 50, Appendix R in that the initial conditions assumed were not consistent with the range of fires which could be postulated, interaction of circuits of concern was not addressed, it failed to uniquely address loss of offsite power, it failed to adequately address the use of pressurizer heaters, it was not totally independent of the control room and cable spreading room, it failed to provide for achieving cold shutdown in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, and the alternate shutdown panels lacked indications of neutron flux and plant temperature. The licensee has re-evaluated the station level of com-pliance with the requirements of 10 CFR 50, Appendix R, Sections III, G, J, L, and The results of this review were submitted to the NRC in September 1984 as the Appendix R Compliance Assessment Report (CAR). The CAR identifies the licensee's impression of their specific degree of compliance with Appendix R, identifies the procedural or physical modifications required to bring the plant into compliance with the regulation, and, in specific areas, requests exemptions from specific requirements of Sections III, G, J, L and While the NRC has not acted yet on the CAR, the licensee is proceeding with the implementation of the corrective actions described in the repor The inspection of this item involved a review of specific items as described in the CAR, station procedures, and different tracking systems being used by the licensee to ensure that procedural and physical corrective actions are carried out. During this review, it was noted that the documents did not generally agree with each othe The licensee stated that since the CAR was issued, additional analyses have been completed which are reflected in the summary of fire protec-tion activities report, but that the tracking systems only reflect Revision 0 of the CAR. The licensee agreed to, by March 6, 1986, submit to the NRC a revised CAR and Fire Hazards Analysis Report (FHAR) identifying all analysis work required to ensure compliance with 10 CFR 50, Appendix R and any other commitments in the area of safe shutdown and fire protection and to update all tracking mechan-isms to correctly reflect all modifications required to bring the plant into compliance as stated in the revised CAR and FHA With respect to the specific findings identified as part of violation 346/83016-01A, the following findings were made:

(1) As indicated in Part II of Inspection Report No. 50-346/83-16, the licensee developed procedure AB 1203.26, Serious Control Room Fire, to address the necessary actions to achieve hot standby and subsequent cold shutdown in the event of a fire

m

>

s .

s ( '

resulting in total loss of the control room or cable spreading room accompanied with loss of offsite power. The procedure is considered " interim" since it contains a number of repair acti-vtties required to achieve hot standby, pending satisfactory completion of the modifications required to meet Appendix This portion of this item remains ope (2) The CAR has identified that, in a majority of the fire areas, the pressurizer heaters remain available for pressure control; however, in five specific areas where both trains of pressurizer heaters could be lost as a result of a fire, the plant would proceed to cold shutdown as directed by procedure AB 1203.02, Serious Station Fire. As a result, the licensee does not con-sider the pressurizer heaters as a required safe shutdown component. The inspector reviewed procedure AB 1203.02 for two (A and D) of the five fire areas identified in the CAR where the heaters could be lost and noted that the procedure for fire area A (Attachment 16.35 to AB 1203.02) "RCS System", identifies the heaters as equipment affected, but does not require the operators to proceed to cold shutdown. The licensee had stated earlier, that all procedural changes required by the CAR had been completed. As a result, the licensee will re-review all the procedural changes required by the CAR to ensure they have

been performe This portion of the item remains ope (3) During the July 1983 inspection, it was determined that for a control room / cable spreading room fire with loss of offsite power, the licensee was not capable of achieving cold shutdown within the required 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. On September 30, 1983, the licensee requested an exemption from the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> requirement of Appendix R,Section III.L.1. The NRC granted the exemption on August 20, 1984; therefore, this issue is resolved and this portion of this item is close (4) With respect to the missing process variables necessary to perform and control reactor shutdown from outside the control room, the licensee plans to provide source range flux and reactor coolant temperature (Th and Tc for both loops) indica-tion outside the control room. Specific completion dates for these modifications have not been established. Inspection Report No. 50-346/83-16, Part II, describes the NRC accepted interim measures being used to compensate for the present lack of these process variables. This portion of the issue remains ope Violation 346/83016-01A remains open pending completion of all procedural and physical modifications required to bring the plant in compliance with Appendix R, NRR acceptance of the licensee's pro-posed resolutions including all requested exemptions in this area, and a satisfactory reinspection to ensure compliance with the regulatio .

b. (0 pen) Violation (346/83016-01B): Failure to perform a spurious signal analysis for valves in the service water discharge line, the pressurizer Power Operated Relief Valve (PORV) and block valve, and the letdown cooler isolation valves. Prior to restart after the July 1983 inspection, the inspectors verified that the licensee had taken procedural actions to ensure that at least one of the four service water discharge valves is normally open and de-energized to assure a service water flow path regardless of fire location, and that the PORV block valve (RC-11) and the letdown isolation valve (MU2B)

would be closed prior to control room evacuation. This was documented in Part II of Inspection Report No. 50-346/8301 In the case of the service water discharge valves, these actions provide a permanent acceptable fi The CAR addresses the concern of spurious signals activating any of these valves. In the case of the PORV and its block valve, RC-11, the analysis showed that a single fire in one area, Area U, could cause the PORV to open and simultaneously prevent closure of RC-11, thereby producing an unisolable leakage path from the reactor coolant system. The licensee has proposed wrapping cable tray 1CJM with a one hour fire barrier to ensure the operability of valve RC-1 The licensee had scheduled this activity for completion prior to startup following the next refueling outage; however, Region III has recommended to NRR that completion of this modification be made a condition for startup from the current outag In the case of the letdown coolers' inlet and outlet isolation valves, the analysis indicates that a fire in containment could disable both valves; however, isolation capability can be provided by a third isolation valve (MUO3) downstream of the other two valves for either cooler. Valve MUO3 is a normally open air operated valve which fails closed on loss of power. Its circuits are not within containmen The inspector verified that procedure AB 1203.02, Serious Plant Fires, Attachment No. 43 addresses the closing of MUO3 if the primary isola-tion valves for the letdown coolers are affected; however, it was noted that there is a lack of guidance available to the operator to make the determination that the isolation valves for either letdown cooler have remained ope This item will remain open pending a revision to procedure AB 1203.02 to adecuately reflect operation of valve MUO3, and satisfactory resolution of the concerns related to simultaneous spurious operation of the PORV and its block valv c. (Closed) Violation (346/83016-07): Reactor coolant pump cil collec-tion systems not capable of holding the entire lubricating oil system inventory. On September 30, 1983, the licensee requested an exemption from the requirements of 10 CFR 50, Appendix R, Section III.O. to have a reactor coolant pump oil collection system large enough to hold the entire lubricating oil system inventory. On August 20, 1984, the NRC granted the exemption, resolving this issu b

.

d. (Closed) Open Item (346/83016-08): Failure of procedures to address spurious actuation of decay heat isolation valves. Prior to restart following the July 1983 inspection, the licensee revised procedure SP 1104.04 to require deenergizing one of the two motor operated valves, DH-11 or DH-12, in the decay heat removal system. This is documented in Part II of Inspection Report No. 50-346/83016, and is considered an acceptable permanent resolution to this issu e. (0 pen) Open Item (346/83016-09): Failure to perform an analysis to determine the level of Gaitronics Communication system operability that would remain after a control room / cable spreading room fire. At the conclusion of the September 1983 inspection, prior to plant restart, (Inspection Report 50-346/83016, Part II), the inspectors determined that the licensee had developed procedures to ensure an adequate alternate means of communications using portable handheld VHF radios. The CAR addresses the communications problem, identi -

fies the need for a sound powered phone system, and the need to determine if additional equipment is required after the communication needs for shutting dcwn the plant are identified. The installation of the sound powered phone system was completed in May 1984. The licensee stated that an informal field verification test was performed to ensure that adequate communication systems were available for the implementation of safe shutdown procedures (AB 1203.02 and AB 1203.26). The licensee agreed to perform and document a formal verification tes Pending the completion of the formal test to verify that adequate communication systems are available to implement AB 1203.02 and AB 1203.26, this item remains ope f. (0 pen) Unresolved Item (346/83-16-14): Automatic sprinklers not installed in accordance with Chapter 4 of NFPA 13 in 19 areas containing equipment required for safe shutdown. Specifically, the systems do not appear to adequately accommodate ceiling congestio Applicable requirements include Section 4-2.5 of NFPA 13 (1978) whi requires that a minimum of 18 inches clearance be maintained between the top of storage and ceiling sprinkler deflectors. Section 4-3 of NFPA 13 (1978) requires that deflectors of sprinklers be located not less than 1 inch or more than 22 inches from ceilings. Section 4-4.13 of NFPA 13 (1978) permits the installation of sprinklers beneath ducts over four feet wide if ceiling sprinklers cannot be spaced in accordance with Table 4-2.4.6 and Section 4.3 of NFPA 1 A review of the NRC's correspondence to the licensee (J. F. Stolz-NRC to L. E. Roe-TED) dated October 3, 1978, identified that the NRC apparently endorsed the installation of sprinkler systems below obstructions in heavily congested ceilings in certain area It is not clear that this correspondence also endorsed not installing sprinklers above the obstructions as required by NFPA 13. The licensee is committed to NFPA 13 by Paragraph 6 of Section 2.1 of the

i

.

fire protection SER for Davis-Besse dated July 26, 1979, which states in part, " Automatic sprinkler systems...are designed to the require-ments of NFPA Standard 13." Additionally, Section 9.5.1.2.1 of the Updated Safety Analysis Report (U.S.A.R.) Revision 0, dated July 1982, does not identify any deviations from NFPA 13. Thus, prior to the inspection documented in Inspection Report No. 50-346/83016, the licensee had not formally identified to the NRC that sprinkler system design did not conform to NFPA 1 A May 18, 1979 correspondence to the licensee from their Architect-Engineering (AE) firm (M. Malcolm of Bechtel to J. Balfour, Jr. of TED) identified that the sprinkler installations deviated from the requirements of Chapter 4 of NFPA 13 because the ceiling spaces are heavily congested with large diameter steel piping, rigid steel conduits, and steel cable trays with solid bottoms. The justifications provided by the AE firm for the deviations were that:

(1) The Nuclear Regulatory Commission (NRC) has requested that the sprinkler heads be placed in such a manner that they will adequately control and extinguish any fire that could initiate on the floor and cause an exposure to equipment required for safe shutdow (2) The ceiling spaces of the various rooms which are to be sprinklered, are heavily congested with large diameter steel piping, rigid steel conduits, and steel cable trays with solid metal bottoms. The cabling in the cable trays has been fire tested and has proved to be a non propagating fire retardant type. The tops of the cable trays are enclosed with a 2 inch thick Kaowool thermal blanket which has a fire resistive rating of 1/2 hour. Hence, the equipment located within the ceiling spaces is considered to be non-combustibl (3) With the sprinkler heads located near the ceiling slab, inadequate sprinkler water distribution and accessibility to the heads for inspection and replacement would exis ,

(4) The heavy congestion of piping, cable trays, and conduits in the ceiling space c. eates a false ceiling effect, which in turn creates an area of heat collection below the lowest level of cable trays and conduits, thereby ensuring timely actuation of the sprinkler syste As noted above, this was not officially brought to the attention of the NRC prior to the inspection documented in report 50-346/8301 In response to the findings of inspection report 50-346/83016, the licensee provided justification for the deviations from the requirements of NFPA 13, in their September 13, 1983 submittal to the NRC (R. P. Crouse-TED to D. G. Eisenhut-NRC, Serial No. 986).

Additional information on this subject was provided in an August 28, 1984 letter from the licensee to the NRC. In both submittals, in addition to the justification provided by the AE, the licensee makes reference to a Union Carbide test that was conducted for the NR , .

The inspectors noted that this test was conducted to determine the feasibility of installing sprinklers three feet below the ceiling, not as much as approximately fifteen feet below the ceiling as is the case in some configurations at Davis-Besse. Although some of the sprinkler configurations tested resulted in acceptable actuation times for fast developing fires, the test summary points out that not all fires envisioned in the test occupancy would begin as rapidly as the test fire. Any slow buildup of heat due to smaller fires would result in a delay of sprinkler actuatio Figure II of the test report indicates that in some of the tested sprinkler configurations, delayed sprinkler actuation averaged more than seven minutes. More significant is the fact that the room or enclosure configuration used in the test had a three foot wide opening in the ceiling running the length of the enclosure which created a chimney effect. The inspec-tors did not observe this type of room or enclosure configuration at the Davis-Besse facility. The complexity of piping, fixtures, and structural interferences that initiated the Union Carbide test was not described in the test report; therefore, there is no clear correlation between the configuration at Davis-Besse and that of

-

the Union Carbide tes Because of the apparent inconsistencies discussed above, it is not clear that the NRC has formally accepted the licensee's justification for this deviation from NFPA 13. During this inspection, the licensee indicated that although their position is that NRR approved this deviation, an independent fire protection consultant has been employed to provide a documented technical evaluation of the sprinkler installations in question, which will include recommendations concerning adequac Purchase Order No. 035-F-87391-A-MB verified the licensee's procurement of the fire protection consulting firm's services for this purpos This item will remain open pending NRC review of the licensee's reevaluation of system adequac g. (0 pen) Unresolved Item (346/83-16-17): The post indicator valves and fire hydrants throughout the plant yard are accessible to mechanical damage from vehicular traffic. No permanently affixed physical barriers (concrete or steel post) are installed to protect these fire protection features from such damage as required by National Fire Protection Association (NFPA) Standard 2 The licensee provided additional information in letter Serial No. 986, dated September 13, 1983, and in the " Fire Protection Activities Summary," dated September 3, 1984, and concluded that hydrants identified during the 10 CFR 50 Appendix R audit are located within the protected area where vehicular traffic is held to a minimu In addition, all post indicating valves and hydrants are of the breakaway typ . .

In a telephone conversation between Regian III and NRR during the week of October 14, 1985, NRR indicated that permanent affixed physical barriers should be installed to protect the post indicator valves and fire hydrants as required by NFPA 2 Also during the week of October 14, 1985, the licensee was notified of the results of the telephone call between Region III and NRR regarding the physical barriers for the post indicating valves and fire hydrant In a telephone conversation on October 31, 1985, between the licensee and Region III, it was discussed that the licensee consnitted to installing physical barriers for the post indicating valves and fire hydrants and that the work is presently in progres This unresolved item will remain open pending verification of installed physical barriers for the post indicating valves and fire hydrant (0 pen) Unresolved Item (346/83-16-18): Approved pressure reducing devices on standpipe hose station outlets are required where the water pressure exceeds 100 psig according to Chapter 4, Section 4-7 of NFPA 14 The inspectors observed the static pressure on automatit:

sprinkler and standpipe hose stations to be at or above 150 psi Examination of hose stations indicated there were no pressure reducers installed on standpipe hose stations throughout the plant as required by NFPA 1 As discussed with the licensee, pressure reducing devices should be installed. As an alternative. solution, the licensee could train the fire brigade on the maximum pressure that will be encountered when utilizing fire hose stations. Additionally, the licensee would have to clearly indicate to all employees that the fire hose stations are-for use by the trained fire brigade only. Finally, signs indicating that the fire hose is for fire brigade use only, would have to be located at the fire hose statio ~

In a telephone conversation on October 31, 1985 between the licensee and Region III, the licensee conunitted to utilizing the alternative solutio This unresolved item will remain open pending review of licensee's actions by Region II (0 pen)OpenItem(346/83-16-19): Technical Specification 3.3. specifies, "With the number of operable fire detection instrument (s)

less than the minimum of operable requirement of Table 3.3.14...."

Throughout the licensee's FSAR and FHA, the coninitment was made to install fire detectors in accordance ..th NFPA 72D. NFPA 72D requires that fire detectors be installed in accordance with NFPA 72 NFPA 72E specifies minimum installation and spacing requirements for fire detector _-. _

.

.

I I

In the " Fire Protection Activities Summary" dated September 3,1985, the licensee indicated that the rationale as to why the minimum number l of detectors in Table 3.3-14 of the Technical Specifications is less

! than the total number of detectors required for some zones is the conservative design approach which was used in their layout. The design exceeded the minimum spacing requirements of NFPA 72E-1978, which created areas of overlapping protection. This overlapping pro-tection is why one or more detectors, as long as no two are adjacent, can be out of service and still provide protection to the zone being protecte In discussions with the iicensee, it appeared that several areas did not have adequate coverage with the minimum number of fire detectors operable as indicated in the Technical Specification 3.3.3.8, because l of the pocket beam criteria, it was also discussed that when two adjacent detectors are out of service, adequate detection may not be provided even though the minimum number of fire detectors as indi-cated in Technical Specification LC0 3.3.3.8 has been achieve In a telephone conversation on October 31, 1985 between the licensee and Region III, the licensee coomitted to review each fire l

area against NFPA 72E and revise the Technical Specification that

'

when the minimum number of detectors are out of service, adequate detection will be availabl This item will remain open pending Region III's review of the licensee's actions.

(0 pen)OpenItem(346/83-16-20)

The 1980 Fire Protection Audit by l Professional Loss Control Inc., Section 3.3.3, identified the possi-bility of cold soldering of sprinkler heads due to water curtain and sprinklers installed at elevation 657 feet in the heater bay are (Cold soldering occurs when water from one sprinkler cools down the heat sensitive element in another sprinkler which prevents operation of the sprinkler being wetted.)

The licensee's " Fire Protection Activities Summary", dated September 3,1985, stated that the suppression at the 657 foot elevation is comprised of two independent systems. The first system has its sprinkler heads located between the roof's beams near the roof underside surface. The second system has its sprinkler heads uniformly distributed under the roof's beams to provide coverage of the area of interest belo This type of design is described in NFPA 13-1978 under Section 4- which gives permissible locations of sprinkler heads either under beams or under a roof or ceiling. A combination of these two posi-tions is permissible provided the clearance and positions outlined in Section 4-2.4 and 4-3 are followed. In addition, the licensee's internal memorandum of August 20, 1981, Item 5, indicates that Grinnell Fire Protection Systems Company, Inc., performed a design l

<

l

-

. _ _ _ - . , _

. .

review of these systems and the design was modified to eliminate the possibility of cold soldering of sprinkler heads. The licensee also has a contract with Professional Loss Control to provide an evaluation on the effects of cold soldering and develop recommendations (Toledo Edison Purchase Order 035-F-87391A-MB). This item will remain open

.pending review of the documentation of an evaluation on the effects of cold solderin (0 pen) Violation (346/83-16-22): Failure to Develop Adequate

. Surveillance Test Procedures, Administrative Controls and Quality Assuranc (1) (0 pen) Noncompliance (346/83-16-22a): Fire pump testing procedures are inadequat (a) The electric and diesel annual fire pump flow test procedures for 1980, 1981, 1982, and 1983, did not specify a " shutoff" (no flow) test as required by NFPA 2 The licensee has incorporated the " shut-off" test as required by NFPA 20 in the Ciesel Fire Pump Annual Flow Test Surveillance Procedure ST 5016.12 (Davis-Besse Temporary Modification Request dated July 3, 1985) and in the Electric Fire Pump Annual Flow Test Procedure ST 5016.03 (Revision 5 dated April 2, 1984). This part of the item is considered closed based on the licensee's corrective action (b) The electric fire pump annual flow test procedure did not specify measurement of vibration at peak load conditions (150 percent of rated capacity) as required by NFPA 2 Electric Fire Pump Annual Flow Test (Surveillance Test 5016.08) Section 5.3 states, " Maintenance personnel are available to record vibration data on the pump at all levels of flow during the test (i.e., 1250, 2500, 3750 gpm)." This part of the item is considered closed based on the licensee's corrective actio (c) The diesel fire pump annual flow test procedure for 1980, 1981, 1982, and 1983, did not specify measurement of pump speed (rpm), flow to 150 percent of rated capacity at 65 percent of rated head, or measurement of vibration at peak load conditions as required by NFPA 2 Diesel Fire Pump Flow Capacity Test (Sheets 1, 2, 3 of Enclosure II), of the Diesel Fire Pump Annual Flow Test (Surveillance Test 5016.12.5) includes measurements for pump speed (rpm) and flow to 150 percent of rated capacity at 65 percent of rated head; however, there is no provi-sion or specific requirement in the surveillance test to

_ ._ _ . .-

. .

take measurements of vibration at peak load conditions as required by NFPA. As previously identified in Report 50-346/83016, the licensee should specify measurement of vibration at peak load conditions as required by NFP This portion of this item remains ope (d) The diesel and electric fire pump test results for 1980, 1981, and 1982 were not properly evaluated to determine their acceptability. No characteristic curve was plotted from the test result to compare with the original manufacturer's certified shop test curve and acceptance test curv The licensee has submitted characteristic curves developed from the 1980, 1981, and 1982 diesel fire pump test; however, it is unclear how the curves were developed since the data sheet did not include pump speed (rpm), pressure corrections, or suction hea The licensee provided the inspector with the August 22, 1984 diesel fire pump test results which included a characteristic curve; however, it is also unclear in this test how the characteristic curve was developed because it does not appear that the corrections for pump speed or suction head were included. This portion of this item will

'

remain open pending review of the licensee's corrective action (2) (0 pen) Noncompliance (346/83-16-22b): Contrary to the NRC Supplemental guidance, entitled " Nuclear Plant Fire Protection Functional Responsib111 ties Administrative Controls and Quality Assurance" (supplemental guidance), surveillance test procedure 5016.09 did not specify verifying fire protection system valve operability weekly, monthly, and annually as required by NFPA 2 The licensee's " Fire Protection Activities Summary" dated September 3, 1985 stated, "NFPA 26, 1976, Chapter 3, allows a monthly inspection if valves in the fire system flow path are locked or sealed open."

The inspectors reviewed NFPA 26, 1976, Chapter 3 which states

'

"A systematic weekly inspection (or monthly in the case of locked-open valves) of each valve should be made, and a report

! form used to record the condition of each valve." The licensee is requested to conduct weekly inspections if seals are to be utilized or monthly inspections if valves are to be locked open as required by NFPA. This part of the Violation (346/83-1G-22) ,

will remain open pending review and acceptance of the licensee's submitta I

-__

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

. .

,

,

,

,

(3) (Closed) Noncompliance (346/83-16-22c): Automatic sprinkler system test results for 1980,1981, and 1982, did not verify simulation of actuation of the most remote sprinkler head by opening the inspector's test valve on the system as required by the surveillance test procedure. According to the licensee's ' *

staff, this test is not performed as required b'y NFPA 13 Instead only a two inch drain test is perfornied at the-base of the sprinkler system rise During this inspection, the licensee indicated that the inspector's test for the wet sprinkler systems has been conducted since 1980 and that the previous information given to the inspector was incorrect. The licensee explair.ed that the individual that provided the previous inspector with the information was not familiar with the inspector's test. The inspector was provided with Surveillance. Test Procedure ST 5016.07 entitled " Fire Protection System Automatic Sprinkler '

Syrtem" which requires the inspector's test to be conducted on the wet sprinkler system. In addition, the inspector was provided with the 1981, 1982, and 1985 inspector's tests for '

the wet sprinkler system. The licensee indicated that the 1980 inspector's test was conducted; however, test ro ults could not be located. The upgraded procedures were reviewed and it was verified that the inspector's test is currently being conducted ,

and also that the inspector's test has been conducted since at

'

least 1981. This portion of the item is close'd.,

(4) (Closed) Noncompliance (346/83-16-22): Contrary to the NRC )"

supplemental guidance, surveillance test procedure ST 50.16.06, '

did not specify measurements of sensitivity, periodid clCaning, -

and adjustment of photo electric fire detectors as required by -

NFPA 72D and NFPA 72E. The test proceduro specified an unsafe practice in that it required use of an "open flame" to test fire ,

detectors rather than a " puff" test as specified by NFPA 72 (a) Fire Protection Activities Sumary dated September 3,1985, indicated that Surveillance Test Procedure ST 5016.06, Fire Detector Functional Test, allows the issuance of a Maintenance Work Order in conjunction with the conduct of this test to document and correct any deficiencies found while performing this test such as needed cleaning or adjustment. This is adequate to ensure that cleaning and, ~.,

adjustment of detectors occurs. This part of violation N (346/83-16-22d)isconsideredclose ,

(b) Fire Protection Activities Summary, dated September 3, 1985, indicates that the test procedures specified an unsafe practice in that they specified use of an "open flame" to test fire detectors rather than a " puff" test as specified by NFPA 72E. The reference to the use of an "open flame" m

13

-

. - - - ---__.___----_.-_ _ ______--._____ _-

to test fire detectors has been removed from ST 5016.0 In addition, the licensee further indicates that an "open flame" has never been used at Davis-Besse to test any detector . This part of the violation (346/83-16-22) is considered close (5) (Closed) Noncompliance (346/83-16-22e): Periodic Test procedure PT 5112.01 did not specify quarterly surveillance of emergency

'

lighting units, or an 8-hour discharge test and periodic replacement of batteries in accordance with the manufacturer's recommendations. The inspector reviewed Revision 8 of Periodic Test Procedure PT 5112.01 titled, " Emergency Lighting System Test," and determined that the surveillance checks noted above

..

have been incorporated into the current procedure. The current procedure includes a monthly, quarterly, and annual surveillance test of each individual eight hour emergency lighting unit, having the integral batteries tested on a staggered frequenc This is to prevent a total blackout should an emergency occu The licensee's emergency lighting surveillance test program, as verified by the inspector, also incorporates the lighting unit manufacturer's recommendations. The licensee provided documenta-tion showing that 23 new emergency lighting batteries were installed in February 1984, replacing previously installed batteries as part of the licensee's corrective action in main-taining the emergency lighting unit (6) (Closed) Noncompliance (346/83-16-22f): Contrary to the NRC supplemental guidance, Administrative Procedure AD 1828-20, " Fire Brigade Training," does not specify the minimum physical requirements for fire brigade members as required by NFPA 2 Interviews with the licensee's staff indicated that no separate physical examinations are required for fire brigade members at Davis-Besse Nuclear Power Station; however, Step 5.2.3 of procedure AD 1810.00 Revision 7 indicates that fire brigade members shall have completed the respiratory protection plan in accordance with procedure HP 1605.02. Step 6.4 of procedure HP 1605.02, Revisions 11 and 12, dated November 8, 1984, revised since the July 1983 inspection, requires that each individual being certified to wear respiratory protection equipment shall have an initial physical examination, the results of which will be reviewed by a physician who will, and annually thereafter, review each certified individual's medical status. The physician will make the recommendation in writing. The procedure identi-fled 13 clinical conditions which the physical examination cover These requirements are acceptabl (7) (Closed) Noncompliance (346/83-16-22g): Contrary to the NRC supplemental guidance noted above, Administrative Procedure AD 1810-01 did not specify fire watch training, or that the fire watch remain on the work site while the work is performed, and

m

. - - .- . . .

. . - . ,-

A s

. .

'

.

r -

remain in the area for at least 30 minutes after the work.is t completed. NFPA 51 and 518 require that a fire watch be trained

!

and equipped to prevent and combat fires, and that the fire watch--

remain on the. work site for at least 30 minutes after the work b is completed to check for smoldering fires. -The licensee's staff provided no documented evidence that contractor personnel per-

. forming fire watch duty.are required. to be-trained in the use of 3 ffire protection equipment and fire preventio The current Administrative Procedure' covering the Fire Watch Qualification Program is AD.'1828.19, Revision 2, dated December 19, 1984. Steps 4 and.5 now include the minimum instruction.an individual contractor and onsite personnel are k to receive to. qualify as a fire watch during open flame, welding, grinding, and cutting. operations. This instruction consists of

.both classroom training and practical training in the use of fire extinguisher Fire watch requalification is required on an annual basis. Based on the above, the inspector determined

the licensee to be meeting their commitment to NFPA SIB. In-addition, TED has' incorporated a second type of fire watch given the title of " Fire Barrier Watch," defined as an individual trained in the> formal process for notification of the security

, force'and Shift ~ Supervisor that.a fire' exists. This individual will be trained through General Orientation Training in the use and operation of plant communication systems. The sole purpose r of this individual:is to monitor potentially degraded fire doors, j -dampers, barriers, and other items which degrade the adequacy

{ of the fire protection program.

,

Step 4.4 of Administrative Procedure AD 1844.07, Revision 2,

'

dated April 22, 1983, includes the necessary responsibilities of personnel performing activities associated with open flame, 1- -welding, grinding, and cutting operations as described by NRC supplemental guidance documents and NFPA SIB, including having the fire watch remain at the work area for a minimum of thirty

, minutes after the work activity is completed to check for smoldering fire a

! (8) -'(Closed) Noncompliance (346/83-16-22h): . Backup fire fighting capability from offsite fire departments. The previous inspector verified in Report 83016 that by letter dated-August 31, 1985,

the Assistant Fire Chief of Carroll Township. Emergency Medical _ ,

and Fire' Service, Incorporated, had agreed to provide-fire pro-j tection in accordance with the Davis Besse Emergency Plan. The

'

licensee had prepared Revision 7 (not. Revision 5 as previously identified in Report 83016) to the Emergency Plan which indicated

~

.that when answering to a call for assistance in fighting fires, the offsite fire 3 department will operate under'the direction o the licensee's fire brigade. The licensee committed to complete

'

the. revision by November 1, 198 *

'

t i

. -

.

. .

In the letter dated November 16, 1983 from R. Crouse, TED to J. Keppler, R-III, the Approved Modification Request to the-Emergency Plan concerning the offsite fire department support was provided for commitment completion verification. This item is considered. closed based on the corrective actions documented in the previous lette (9) (0 pen) Noncompliance (346/83-16-221): Specific qualifications and responsibilities for individuals responsible for implementing

- the fire protection program are specified in Table 4-1 (Sheet 1)

-of'the licensee's'FSAR for Davis-Besse, and in Section 2.0 of NRC supplemental guidance documen The following discrepancies existed in the licensee's personnel qualification requirements for fire protection / prevention program implementation:

(a) Interviews with the licensee staff disclosed that only the fire protection coordinator has direct responsibility for the fire protection program at the site. Other licensee

' staff are periodically assigned various fire protection responsibilities a small percentage of.the time. Sufficient-documentation was not available at the time of the inspec-tion, including a TED Table of organization, showing the responsible fire protection personnel'and their interface to the fire protection program. This portion of the item

-

will remain open pending further review of the licensee's commitment and other related informatio (b) The fire protection coordinator and the-fire marshal are the same individual, whose demonstrated knowledge of fire protection-is vested in his training and experience as a volunteer firefighter. Attachment No. 1 of the NRC supple-mental guidance document describes the responsibilities of the fire protection organition including the onsite position commonly referred to as either the plant " fire protection coordinator" or plant " fire marshal." According to the guidance document, the onsite position (s) responsibilities include. implementing periodic fire protection inspections, training, assisting in the critique-of all fire drills, review of proposed work activities for transient fire loads,

.

implementation of a program for contractor personnel in

administrative procedures, and instruction of personnel on other related fire protection topic According to the licensee, three individuals now hold the

position of." fire protection coordinator". Subsequent to the inspection the licensee provided copies of these indi-L viduals resumes to the inspector for review. The following i information was derived from each individual's resume:

(1) Individual No. I currently has twelve years of combined

k i

.- .

nuclear fire protection experience attaining " reactor operator" status and-lead fire protection responsibility including emergency medical experience. Further, this individual has attended numerous fire protection seminars totaling in excess of six hundred and seventy five hours;

'(2)-Individual No. 2 currently has ten years of combined nuclear fire protection experience attaining " auxiliary operator" status, advancing through the operator qualifi-cation program and also lead fire protection responsibilit This individual has attended some. twenty-five fire protection related seminars since 1975; and (3) Individual No. 3 currently has in excess of eight years fire protection experience including nuclear experience and an extensive emergency medical and fire training background. This individual has also attended numerous fire safety seminars and is a certified Ohio firefighter satisfying the required two hundred hour training program. According to the resumes, each of these individuals utilize their talents towards nuclear fire safety as part of their responsibili-ties at Davis-Besse. Based on the above, the inspector concluded that these individuals' academic and work experi-ence satisfy NRC guidelines "equired of licensee fire protection personnel. This portion of the item is considered close (c) There is no qualified fire protection engineer who is a graduate of an engineering curriculum of accepted standing in fire protection engineering on the licensee's staff o onsite as a consultant. During an NRC Fire Protection inspection conducted during June 1984, the licensee's position was that a qualified fire protection engineer is on the TED staff at the Davis-Besse site. According to the NRC supplemental guidance document, the qualifications of a fire protection engineer should include one who is a graduate of an engineering curriculum of accepted standing, and who shall have completed not less than six years of engineering competency and achievement, three of which shall have been in responsible charge of fire protection

'

engineering work. These requirements are the eligibility requirements as a member in the Society of Fire Protection Engineer At the time of the Appendix R Fire Protection / Safe Shutdown inspection (July 1983), the individual on the licensee's staff in the position of fire protection engineer was not a member of the Society of Fire Protection Engineers (SFPE)

and therefore had not had his academic and work history evaluated to be qualified as a " member" grade in the SFP Additionally, this individual was'not devoted full time to the area of fire protectio .. .

Since the Appendix R, Fire Protection / Safe shutdown inspection, this individual has had his academic and work history evaluated by SFPE's and was qualified as a " member" grade in the SFPE in October 1983. The inspector also met with the General Supervsior-Facility Engineer to discuss the present scope of the fire protection engineer's job responsibilities. The inspector was assured that since July 1983 the fire protection engineer's sole responsibility was fire protection and that all other non-related responsi-bilities have been reassigned to other personne This individual is a member of the Fire Task Group Team and has been since its inceptio Further, since the June 1984 inspection, an additional fire protection engineer as described in NRC criteria has been added to the TED staff in the Nuclear Facility Engineering Department. This portion of the item in considered close (10) (0 pen) Noncompliance (346/83-16-13 and 346/83-16-22j): Failure to establish a fire watch on inoperable fire dampers and an inadequate surveillance test procedure. In the Report 346/83-16, section entitled " Details - Part II" (pg 43), it states, "The inspector verified that procedure ST 5016-11 had been revised and its implementation was observed by the inspector on September 8, 1983 and found acceptable". This portion of the Open Item is close The open item previously discussed also states "Due to the high failure rate that occurred during the fire damper surveillance retests, the licensee has committed to an on going testing program of all dampers until the confidence level increase This program, plus the schedule completion date for the remaining fire dampers yet to be tested, will be submitted to Region III by November 30, 1983 after an engineering analysis is performed on locating and evaluating any necessary work needed to perform the surveillance test. As discussed on October 31, 1985, the licensee committed to provide Region III a summary or history of events and corrective actions regarding this portion of the Item. This item will remain open pending Region III review and acceptance of the licensee's submitta . Containment Purge Valves The containment inlet and outlet purge valves (4) were inspected, and the general local leak rate testing method was discussed with the licensee to ensure that no possible leakage paths were being bypassed by the testing methods in us No violations or deviations were identifie J

. . Review of Allegations Concerning BISCO Penetration Seals The NRC received three generic (not specifically mentioning the Davis-Besse Plant) allegations regarding Bisco penetration seals indicating that:

(1) test data to support the penetration seals' fire resistance capability did not exist including test data to show that the 6 inch silicone foam fire barrier penetration seals are rated for three hours, (2) penetration pressure seals were installed with a 2 inch overlap on the seal boot instead of the required 3 inch overlap, and (3) Bisco procedures SP504, SP505, SP505-1, SP505-2, and SP505-3 were deficien The licensee provided a comprehensive package of information pertaining to Bisco penetration seals including NRC correspondence on this subject, penetration seal specifications, Bisco procedures covering flexible boot seals, TED procedures covering the Dow Corning silicone foam penetration fire seals, and test reports of penetration seal Resolution of the allegations was pursued through inspector review of the documentation provided by the licensee; a visual inspection and penetration seal walkdown of several flexible boots and silicone penetration fire seals; and discussions between the licensee's staff, resident inspectors and the inspector to determine whether the allegations could be substantiated. As a result of these activities the following findings and conclusions were made, Allegation No. RIII-83-A-0029-01: Test data to support the penetration seals' fire resistance capability did not exist including test data to show that the six inch silicone foam fire barrier penetration seals are rated for three hour In response to this allegation the inspector reviewed two Bisco Fire Test Reports numbered 748-57, dated November 6, 1981, and 748-42, dated July 29, 1981, noting the test standard criteria being utilized, penetration seal fabrication, type, and design configuration including any penetrating items, test observations and the conclusions of the fire tes Both tests showed that the material used as the fire seal was Bisco SF-150NH (High Density Silicone Foam) installed to a nominal depth of 12 inches and utilizing cable tray and conduit design config-urations. The inspector determined that the fire seal systems tested in the two tests met NRC guidelines and therefore are acceptabl Section 7.2.4.e of Specification No. 7749-M-25500, dated August 11, 1983, indicates for fire barriers using low density silicone foam that a minimum final fill depth of 11 inches is required and for high density silicone sealant material 12 inches is require The test reports reviewed by the inspector regarded only the high density silicone sealant material; however, the licensee did provide a letter (Serial No. 284) from L. Roe, TED, to J. Stolz, NRR, dated May 25, 1977, regarding a fire equivalency test for low density silicone foam (SF-20 verst.s Dow Corning 3-6548). The purpose of the letter was to inform NRR that TED planned on substituting the SF-20 product with the Dow Corning 3-6548 product due to the discontinued

'

L_ - _ _ _ _ _ _ 1

,

, .

manufacture of the SF-20. The TED letter also indicated that a three-hour fire exposure test had been performed at the Fire Research Laboratory of the Portland Cement Association in Skokie, Illinois on April 12, 1977. A 12 inch thick concrete slab was prepared containing four one-foot square blockouts. Two of the blockouts were filled with SF-20 and two with Dow Corning 3-6548 silicone foam. The report further stated that, "Each blockout was filled to the full depth of i 12 inches, which is the minimum depth used as a fire barrier at Davis-Besse Unit 1."

Documentation was not available at the time of the inspection to support the specification statement of having low density silicone foam configurations (previously installed SF-20 or more recently installed Dow Corning 3-6548) using a fill depth of 11 inches. This is considered an open item (346/85028-01) pending inspector review of test documentation demonstrating the adequacy of installed low density silicone foam with a fill depth of 11 inches. As part of

! this open item the inspector requested the licensee to provide l adequate assurance that no penetration fire seal located in a fire area boundary barrier was less than 11 inches in depth. As this

'

[ assurance could not be provided during the inspection, this allegation remains open. On October 18, 1985, in a telephone call between Region III management and TED management, it was discussed that this assurance would be required as one of the items to be closed prior to plant restart from the current outag Three maintenance procedures numbered MP 1405.03.1 (Step 8.3.1),

! MP 1405.04.3 (Step 6.3.1), and MP 1405.08.0 (Step 8.3.a) each l Indicate that the minimum depth of foam to be installed is 11 inches

'

for low density silicone foam or 12 inches for high density silicone l foam in any penetration conduit through a wall or floor, except where the wall or floor thickness is less, then the depth is to be equal to the thickness of the wall or floor.

!

l

'

The three maintenance procedures do not differentiate between fire barrier walls and floors and non-fire barrier walls and floors when mentioning the amount of sealant material to install as does Section 7.2.4.e of Specification No. 7749-M-255 This is considered an open item (346/85028-02) pending inspector review of the procedural revisions to the three maintenance proce- ,

dures to reflect Section 7.2.4.e of Specification No. 7749-M-255Q.

, b. Allegation No. RIII-83-A-0029-02 and 03: (02) Penetration seals I

were installed with a 2 inch overlap on the seal boot instead of the required 3 inches. (03) Bisco Company procedures SP504, SP505,

, SP505-1, SP505-2, and SP505-3 were deficient. The inspector determined that these two allegations were interrelated and as such they are discussed together.

l s

,

! a 20

_ .

. .

The flexible Boot Seal is designed for use as a pressure barrie According to the licensee, procedures SP505-1, SP505-2, and SP505-3 were not being used at the Davis-Besse site. As such, that portion of the allegation is not applicable to Davis-Bess Bisco's flexible boot seal fabrication procedure (SP504), step 5.3.3, and Bisco's Installation procedure (SP5051), step 7.3, both state

-

that, "A bead of the Bisco RTV Sealant is applied to the inside of the overlap seam and spread to a minimum width of 3 inches." The Davis-Besse site installer of flexible boot seals (installation of the flexible boot seals is performed on an infrequer.t basis)

interpreted the two steps noted above to mean that the minimum overlap seam is to be 3 inches.

I Subsequently, the licensee, through Bechtel, contacted Bisco personnel who indicated that the overlap axial seam criteria was to be based on the diameter of the pipe penetration being sealed. The inspector requested the licensee to provide documentation to support the vendor's position regarding the overlap seam criteria. This is considered on open item (346/85028-03) pending inspector review of this documentatio The inspector performed an inspection of seven flexible boot seals including measurement of overlap axial seams with assistance from the licensee's staff. Of the seven boot seals inspected, two failed to meet the acceptance criteria as given by the Bisco personnel and five boot seals failed to meet the acceptance criteria as given by the plant installer. As such, this portion of allegation (02) remains ope The Bisco Fabrication and Installation procedures also indicate that the RTV sealant material should be allowed to cure for a minimum of sixteen hours before any further work is performed. No TED procedure existed for the responsible Quality Assurance inspector on when to inspect the boot seals' final installation so as to verify that no slippage of the overlap seam had occurre The TED procedure (s) established for the installation, and fabrication of the flexible boot seals were inadequate in that these procedures failed to provide clear and concise installation, fabrication and inspection instructions. The inadequacies in the flexible boot seal procedures is considered a violation of 10 CFR 50, Appendix B, Criterion V (346/85028-04).

5. 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. Open items disclosed during the inspection are discussed in Paragraph 4.

I

l l

- -- - - - _ . . .__ . - __ _ _ - -

.

. ..

6. Exit Interview The inspectors met with licensee representatives (denoted in Paragraph 1)

at the conclusion of the onsite inspection on September 6, 1985, and summarized

the scope and findings of the inspection. The licensee acknowledged the

_ statements made by the inspector. The inspector also discussed the likely informational content of the inspection report with regard to documents reviewed by the inspector during the inspection. The licensee f

did not identify any of the documents as proprietar In a telephone conversation on October 18, 1985 between Region III and the licensee, it was discussed that (0 pen) Violation (346/83016-018)

regarding the operability of the PORV and its block valve (Paragraph 2.b.)

and Open Item (346/85028-01) regarding the adequacy of the low density foam (Paragraph 4.a.) would be resolved prior to restar Also, in a telephone conversation conducted between Region III and the licensee on October 31, 1985, items 83-16-17, 83-16-18, 83-16-20 and 83-16-22 (a, b, c, d, h, j) were discussed and are documented in this repor l r

!

22