IR 05000321/1988013

Insp Repts 50-321/88-13 & 50-366/88-13 on 880418-20.No Violations Noted.Major Areas Inspected:Hydrogen Water Chemistry & Plant Chemistry Control

ML20154Q087
Person / Time
Site:	Hatch
Issue date:	05/13/1988
From:	Kahle J, Ross W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML20154Q084	List: IR 05000321/1988013 ML20154Q080
References
50-321-88-13, 50-366-88-13, NUDOCS 8806060182
Download: ML20154Q087 (9)

Text

-__

.

.

UNITED STATES

1. g 0 7800g D

<f o NUCLEAR REGULATORY COMMISSION 8 'i *Y REGION 11 h.)

• i 'c 101 MARIETTA STREET, AT L ANTA, GEORGI A 30323

'.....* HAY 171988 Report Nos.: 50-321/88-13 and 50-366/88-13 l

Licensee: Georgia Power Company P. O. Box 4545 Atlanta, GA 30302 Docket Nos.: 50-321 and 50-366 License Nos.: OPR-57 and NPF-5 Facility Name: Hatch 1 and 2 Inspection Conducted: April 18-20, 1988 Inspector: /

MM(dli 5 3//7

@ W. J. Ross Date Signed Appro ed by: L/(~c s 3/F)

J fBf Kahle, Sectibn Chief Dat'e Signed Old sion of Reactor Safety SUMMARY Scope: This routine, unannounced inspection was conducted in the areas of hydrogen water chemistry and plant chemistry contro Results: No violations or deviations were identifie PDR ADOCK 05000321 O DCD _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

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

- '

.

,

.

REPORT DETAILS Persons Contacted Licensee Employees B. Arnold, Chemistry Supervisor R. Bryant, Staff Chemist R. Dedrierson, Assistant to the Vice-President / Hatch B. Feimster, Chemistry Foreman

• Kirkley, HP/ Chemical Engineering

• V. McGowan, Chemistry Supervisor

,

W. Rogers,- Chemistry Superintendent S. Shipman, Systems Engineer R. Tracey, Systems Engineer

• P. Read, Plant Support Manager

• Self, Nuclear Superintendent Other Organizations M. Terrell, General Electric Company NRC Resident Inspectors P. Holmes-Ray, SRI

• J. Menning, RI

• R. Musser, RI

• Attended exit interview Exit Interview

>

The inspection scope and findings were summarized on April 20, 1988, with those persons indicated in Paragraph 1 above. The inspector described the areas inspected and discussed the inspsp ton results. No dissenting comments were received from the license In response to a request by the inspector for clarification of actions

] taken in response to findings by the Institute of Nuclear Power Operations (INPO), the licensee provided information that was considered proprietary to INP This subject is not included in this repor . Licensee Action on Previous Enforcement Matters This subject was not addressed in the inspection.

(

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

.

.

'

.

2 Hydrogen Water Chemistry This phase of the inspection was a review of actions taken by the licensee since the last inspection in this area (see Inspection Report Nos. 50-331; 366/87-03 dated February.11, 1987) to implement a hydrogen water chemistry (HWC) program to control initiation and/or growth of intergranular stress corrosion cracking (IGSCC) in the reactor coolant recirculating pipin This review consisted of discussions with personnel from the licensee's Chemistry Department and Systems Engineering Department as well as with the Gereral Electric contractor who was operating the current hydrogen and oxygen injection system in Unit In addition, the inspector walked down the current system and the proposed pipe runs for permanent injection systems for both units, Status of HWC for Unit 1 Beginning in February 1987 the licensee had performed a second HWC test for approximately ten weeks to establish optimum conditions for preventing IGSCC while minimizing radiation levels caused by carry over of volatile nitrogen-16 spec.ies with the stea Li kewi se ,

beginning September 1987 Unit I had been operating in the current fuel cycle using HWC control. As occurred in the initial feasibility tests, the results of the second test were inconclusive in that a stable corrosion potential between stainless steel and the reactor coolant could not be established through the reduction of dissolved oxygen by a hydrogen flow rate as high as 44 SCF At a rate of 44 SCFM of hydrogen the radiation level at critical parts of the plant was as much as four times the level under normal (non-HWC) condition Consequently, during most of the test and during the current fuel cycle the injection rate of hydrogen had been reduced for 22 SCFM. At this rate the radiation levels were as much as twice normal background. Also, at this injection rate the concentrations of dissolved oxygen in the feedwater and recirculatory water were 220 ppb and 6 ppb respectively. The electrochemical potential of stainless steel stabilized within the range of 10 millivolts versus a standard hydrogen electrode (SHE). This electrochemical potential is considerably more positive than the 230 my range recommended by the BWR owners group for prevention of IGSC The inspector was informed that a permanent HWC procedure was being developed based on an injection rate of 22 SCFM with the expectation that crack initiation or crack growth will be significantly hindered even though optimum conditions are not attainable. In the interim, Special purpose Procedure "Hydrogen Injection and Control" was being followed to govern the addition of hydrogen to reactor feedwate This procedure also covered the addition of oxygen into the common offgas line downstream of the third stage Steam Jet Air Ejector to assure an oxygen-rich mixture in the recombine This interim procedure also provided for personnel requirements, coordination with

~ , '

.

,

'

.

.

,,

'

the Operations Department, as well as for operating and safety precautions and authorizatio The injection system was being monitored and operated by General Electric contract personne The inspector verified that Technical Specifications (Tables 3.2-1

- and 3.2-8) had been revised to allow the trip setting of the Main Steam Line Radiation Monitor to be maintained at $3 times normal full background under the following conditions:

"Within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> prior to the planned start of the hydrogen injection test with the reactor power at greater than 20*4 rated power, the normal full power radiation background level and /

associated trip setpoints may be changed based on a calculated value of the radiaticn level expected during the tes The background radiation level and associated trip setpoints may be adjusted during the test based on either calculations or measurements of actual radiation levels resulting from hydrogen injection. The background radiation level shall be determined and associated trip setpoints shall be set within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of re-establishing normal radiation levels after completion of hydrogen injection and prior to establishing reactor power levels below 20*4 rated power."

An operations standing order had been implemented to assure that these setpoints were properly coordinated with power leve ,

The inspector walked down the lines of the currentJWC injection system from the skids for hydrogen and oxygen cylinders to the penetrations for the hydrogen lines at the suction of the three

/ condensate booster pump Injection flow was being controlled at a panel in the Unit 1 Turbine, and hydrog'en was being injected at a rate of 24 SCFM and oxygen of 12 SCF Hydrogen could be manually isolated if the reactor tripped or was isolated, or if'the injection rate exceeded 35 SCFM. The isolation valve panel was being monitored by hydrogen analyzers that would also isolate hydrogen flow if the atmospheric hydrogen concentration exceeded two percen Finally, hydrogen flov would also be isolated if the concentration of oxygen in the discharge of the hydrogen-oxygen recombiner in the condenser air ejection system is <5' Permanent HWC System The inspector reviewed the licensee's plans for installing a permanent HWC injection system and established the following facts:

• Plans for initiating HWC on Unit 2 were being postponed until the effectiveness of IGSCC control in Unit I could be establishe E , .

" s e

e, ; s

. .

.

,

i h

t 4

I

'

Designs for a permanent injection system were being developed by Southern Company Services and construction was to begin in June 1988. These designs are based on recommendations published by the Eltetric Power Research Institute (EPRI) and appropriate fire protection._ code * Use of the permanent system is scheduled to begin after the next Viitt i refueling outage in the fourth quarter of 198 t

• Hydrogen will be stored as a liquid in a 20,000 gallon tank and oxygen will be stored in a 9000 gallon tank. These tanks are to be located outside the Protective Area fence at the southeast corner of the plant site, approximately 1000 yards from the power bloc * Both hydrogen and oxygen will be pumped as gases through i stainless steel lines (1h inch for oxygen and 1 inch *or hydrogen) buried below the frost line (i.e. ~18 inches deep) in a single tranch from the cryogenic tanks to the west cableway at the south end; af the plant's . turbine buildin The hydrogen line will be b ciosed in a carbon steel guard pipe whenever the

'line passes uWer a road or rail road. Within the west cableway the lines will be reduced to 3/4 inch diameter and routed to the existing isolation panels and then to existing penetrations at the suction of condensate booster pumps and at the air ejector lines. Hydrogen monitors are to be installed in the guard pipes ,

that will enclose the hydrogen lines throughout the unventilated cableway. All joints will be welde * Purge valves will be installed at the condensate booster pumps so that hydrogen lines can be purged in a backwards direction if require ,

'

-

The licensee's plans were considered to be consistent with guidance provided by the NRC (see letter dated July 13, 1987 from James E. Richardson, Office of Nuclear Reactor Regulation to Mr. G. H. Neils, Chairman Regulatory Advisory Committee, BWR Owners 2 Group II for IGSCC). However, the inspector reemphasized the need to take maximum precautions to design and operate the hydrogen and '

oxygen injection systems so as to minimize leaks and to maximize surveillance for leaks, especially in regions of the plant with poor i ventilatio l As part of the permanent HWC installation the licensee had also t acquired i Crack Arrest Varification (CAV) system which will be used -

l l

to monitor the effect of HVC on three representative stainless steel coupons, and, thereby, monitor the effectiveness of HWC contro In l

addition, two new inline monitors for dissolved oxygen had been installed to increase the licensse's capability to determine and to !

'

trend the oxygen concentreion in the condensate and feedwate Finally, in a effort to establish the effect of metals (dissolved and

.

, -.

_

/ . . _ ~ w)

'/

• * >

-

/

,

wq \<

<

,

g

, I_a , i/

'

1, r# g

"

1 5 i r' ,

f

, ,f ' ,

(

,' ,

soluble) o l }iWC, three inline corrosion product samples had been y ,. installed to monitortwater in the hotwell (especially for corrosion u- products from the brhss condenser tubes), in the condensate polisiier '

,

'

effluent, and in the feedwate ( -

'['

As will be discussed later, the instability of the electrochemical potentials observed in thei l

p l; {, '< recirculatingwattjchenhydrogenwasinjectedintothefeedwaterha6 been attributed, in,part, to the presence of metallic species in the

-

'

reactor water and rpcirculating water, ,

f a Plant Chemistry' . e

'

_

Through discussions w\it/co~gnizantplantpersonnelandthroughanauditof ( chemistry control dats jhe insdector reassessed the licensee's hbility to i prevent degradation of the reactor coolant pressure b6ndary. This part

./; of the inspection consisted of a review of the design and operation of key components of thej reactor water system and the implementation of the licensee's water ahemistry progra , ,

i- , Plant Design and Operation T

>'

'

'

9uring the fifteen month interYal since the last inspection in this area, both uniti had operated in stable modes exce outages (April - June 1987 for Unit 1 and January}t- during April 1988 refueling for l

Unit 2). Except for four brief (1-2 days) shutdowns, Unit I had operated at 100% power and, most of the time, with hydrogen water chemistry control . Although Unit 2 only had three short shutdowns, the licensee had decreased the power level of this unit to 90s In May 1987, and to lower levels since Octuber 1937, because of fuil f

'

fa0 lure problems.

'

Although air inle.kage had remained higher than desired (~15-40 SCFM in Unit 1 and ~15-20 SCFM in Unit 2) the main condensers had provided effective barriers against ingress of condenser cooling water. Only one tube leak had occurred, and this leak had been caused by mechanical damage rather than by corrosion mechanis The licensee

' attributed the 1stegrity of the main condensers, in part, to a i f ',

program of visual and eddy current tests performed during refuelin'g b {

/. outage <

s r The principal concern with thel inain condensers was still the ,

t,'

continual loss of ntal from the condenser tubes as sol ie and/or [{

insoluble species of clpper and zinc. In an effort to prevent these m is ', metals, especially ccpper, from being tyhsported to the reactor, the

,li,censee had attempted @ enhance ths 4fficiency of the condensate

. polishers in Unit 2 by increasing "the length of the

/ ', / filter-demineralizer tubes from 70 indes to 80 inches. Also, ' body-

,, < feed' technique of cor.tinually adding thin layers of ion-exchange resins to the filter-demineralizer,el',aents was still in use. This technique had extended the intervals between proecoating (complete removal and replacement of resin fmm 'the tubes) to approximately threk weeks; thereby, conserving water, resin, and manpowe '

,

i J

i

'

c 1

r )

- '

.

,

'

.

l

However, an audit of analyses of effluents from the polishers, as well as analyses of- feedwater samples, showed that measureable amounts of cooper (~0.3 ppb) were still passing through the demineralizers and into the reactor, where the copper is a concentrated by a fa.ctor of ~10 Both Hatch units have experienced fuel rod failure as the result of corrosion and embrittlement of zircaloy cladding attributed to

' copper crud' (see Inspection Report 50-321:366/84-06 dated March 23, 1984). The presence of easily reduced species of copper has been considered by the licensee to also contribute to the instability of the electrochemical potential of the reactor water and the recirculating water when hydrogen is injected to reduce dissolved oxygen (see Paragraph 4a of this report). Consequently, the licensee has an ongoing program to reduce the concentrations of both soluble and insoluble copper in the feedwate From the audit of control and diagnostic chenistry variables the inspector observed that trace (0.1-1.0 ppb) amounts of soluble iron, zinc, and nickel were also being . transported into the feedwate This magnitude of iron and nickel is not considered to be a significant contributor to the loading of the reactor with oxide sludg Recent investigations by EPRI have shown that the presence of trace amounts of zinc in the reactor water acutally is beneficial to the maintenance of low ex-core radiation levels throughout the reactor coolant system and does not pose a corrosion hazar The Chemistry Groups's capability for monitoring general corrosion of the brass condenser tubes as well as carbon steel and stainless steel pipe throughout the reactor coolant system had been enhanced by the installation of corrosion product samplers for hotwell water, polisher effluents, and feedwate The ir.spector also established that although the licensee had not encountered micro or macro biological problems in the Service Water systems, two intrusions of raw river water into closed cycle cooling systems had occurred through heat exchangers cooled by Service Wate One of these leaks contaminated the Unit 1 Recombiner Closed Cycle Cooling Water System.

l The other leak contaminated the water in the Suppression Pool of

, Unit I with approximately 0.5 ppm of chloride and 300-400 ppb of

'

silica; thereby, exceeding administrative limits of <50 ppb chloride and <100 ppb silica. At the time of this inspection the licensee had not begun to reduce the levels of these contaminants. The inspector was informed that the reason for the delay was because an efficent

! cleanup method had not been settled on. The radwaste system had a l capacity of about 2000 gallons per week while the Torus water volume I was over 600,000 gallons. The licensee was considering the use of a 1 demineralizer system that could be used to cycle and cleanup this i

volume of water. Although the walls of the carbon steel Torus were l

l 1 -

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

_

' '

.

,

'

.

supposed to be coated with paint, the inspector emphasized the potential for corrosion when carbon steel is exposed to water with dissolved oxygen, detectable chloride concentrations, neutral pH, and, especially, in a stagnant conditio Finally, the inspector was informed that the discrepancies in the concentrations of corrosion inhibitor (sodium nitrite) in the Reactor Building Closed Cycle Water (RBCCW)(as shown on the HP/ CHEM Daily Report) was caused by leaks in pump seals that depleted the treated water. Because of difficulties involved with the cleanup or disposal of nitrite-containing water, the licensee had chosen to discontinue addition of the corrosion inhibitor until the pump problem was corrected. The licensee was aware of the potential fnr degradation of the carbon steel pipe in the RBCCW unless all dissolved oxygen in the cooling water was eliminte Water Chemistry Program The inspector reviewed the following elements of the licensee's water chemistry program Staffing Iraining Physical Facilities

Quality Control Program (1) Staffing - Since the last inspection in this area the organization reporting to the Manager-Health Physics / Chemistry had been modified so that three Superintendents (Health Physics, Chemistry and HP/ CHEM Support) provide a secondary line of supervisio The Chemistry Group was divided under two Supervisors, one for instrumental analysis and the second responsible for all other che.nistry activities. The analytical staf f, eight foremen and thirty-four technicians, was divided into five rotating shifts and one day shif t. The inspector was informed 1 hat the rotating shifts would soon be of twelve hours duration rather than the current eight hour The professional staff under the Superintendent / Support included chemists and chemical engineers who not only provided daily support but ;^o also had responsibilities for upgrading capabilities um initiating HW (2) Training - The stability of the Chemistry staff continued to improve dering the interval since the last inspection in this area. Consequently, the licensee had been able to continue with its formal training (classroom) orograms as well at with the on-the-job qualification program. An average of two weeks out of every ten was being dedicated to trainia . . __ _ __ _ . -_ __ __ _

, - . .

,

(3) Physical Facilities - By means of a walkdown of sampling rooms and laboratories the inspector reassessed the facilities available to perform the responsibilities of the water chemistry program. During the last year the licensee had developed an instrument laboratory separate from the existing hot laborator This additional space permitted more effective use of. the ion chromatography systems, but, because of inefficient venting, could not be used for atomic absorption ' spectrometric analysi As mentioned earlier, the sampling rooms for both Units had been upgraied with inline monitors for oxygen, conductivity, and corro.: ion product (4) Quality Control - This element of the water chemistry program was reviewed through observations and discussions with the Instrument Supervisor, Instrument Foreman, and technicians in the Instrument Sub Group while these personnel were involved in the analysis of a series of unknown standards that had been provided by the inspector. These standards consisted of aqueous solutions of the key chemistry ionic variables that have been identified for the control of. BWR chemistry by the BWR Owners Group (i.e., chloride, sulfate, silica, iron, copper, nickel, chromium, and sodium). These solutions were prepared in ppm concentrations for the NRC by the Brookhaven National Laboratory and were to be analyzed by the licensee's usual procedures after dilution to concentrations similar to routine control and diagnostic value Although some analyses had been completed and evaluated by the inspector before the end of the inspection, the results of all analyses will be presented in a supplemented Inspection Repor IFI 50-321/88-13-01; 50-366/88-13-01 (5) Conclusions Violations or deviations were not identified during any phase of this inspectio The inspector verified that the Technical Specifications pertinent to chemistry control had been met. The steps beir.g taken to implement HWC, to upgrade the design and operation of key plant components related to chemistry controi, and to further improve control and diagnostic capabilities were considered to be acceptable and to i ndicate a commendable understanding of corrosion mechanisms and industry approaches to this preventio ,

-w - - n , ..-,r .

, , , , ,,--...,.-.w- - , - . , . , - , , , , , - , , - e-- . ~ , - - - -,,,,m- _aaw<,w-, , ,