ML13333A925
| ML13333A925 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 02/23/1981 |
| From: | Haynes J SOUTHERN CALIFORNIA EDISON CO. |
| To: | Engelken R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| References | |
| TAC-65149, NUDOCS 8103100528 | |
| Download: ML13333A925 (5) | |
Text
.
REGULATORY *-ORMATION DISTRIBUTION SY M (RIDS)
ACCESSION NBR:8103100528 DOC.DATE':81/02/23 NOTARIZED: NO DOCKET #
FACIL50-206 San Onofre Nuclear Station, Unit 1, Southern Californ 05000206 AUTH.NAME AUTHOR AFFILIATION HAYNES,J.G.
Southern California Edison Co.
RECIP.NAME, RECIPIENT AFFILIATION ENGELKENR.H, Region 5, San Francisco, Office of the Director
SUBJECT:
Provides status rept on corrective actions taken to ensure continued operability of safety-related air-using equipment Blowdown of header performed to remove dessicant.Prevehtive maint program is underidevelopment, DISTRIBUTION CODE: A0028 COPIES RECEIVED:LTR L ENCL SIZE:VYIL....2 TITLE: IncidentReports NOTES:1 cy each:SEP Section Leader & J Hanchett (Region V) 05000206 RECIPIENT COPIES RECIPIENT COPIES ID CODE/NAME LTTR ENCL7 ID CODE/NAME LTTR ENCL ACTION:
CRUTCHFIELD 04 3
3 INTERNAL: A/D COMP&STRU06 1
1 A/D ENV TECH 07 1
1 A/D MATL & QU08 1
1 A/D OP.REACT009 1
1 A/D PLANT SYS1U.
1 1
A/D RAD PROT 11 1
1 A/D SFTY ASSE12 1
1 ACC EVAL BR 14 1
1 AEOD 3
3 ASLBP/J.HARD 1
1 AUX SYS BR 15 1
1 CHEM ENG BR 16 1
1 CONT.SYS BR 17 1
1 CORE PERF BR 18 1
1 DIRENGINEERI20 1
1 DIRHUM FAC 821 1
1 DIRSYS INTEG22 1
1 EFF TR SYS BR23 1
1 EQUIP QUAL 8R25 1
1 GEOSCIENCES 26 1
1 I&C SYS BR 29 1
1 I&E 05 2
2 JORDANE./IE 1
1 LIC GUID BR 30 1
1 MATL ENG BR 32 1
1 MECH ENG BR 33 1
1 MPA 3
3 NRC:PDR 02 1
1 OP EX EVAL BR34 3
3 OR ASSESS BR 35 1
1 POWER SYS BR 36 1
1 RAD ESS BR39 1
1 REACT SYS BR 40:
1 1
01 1
1 REL & RISK A 41.
1 1
SF PROG EVA42 1
1 STRUCT ENG,BR44 1
1 SYS INTERAC B45 1
1 EXTERNAL: ACRS 46 16 16 LPDR 03 1:
1 NSIC 05 I
1 TERA:DOUG MAY 1
1 TOTAL NUMBER OF COPIES REQUIRED:
LTTR
/0 ENCLV-
Southern California Edison Company P. 0. BOX 800 2244 WALNUT GROVE AVENUE ROSEMEAD. CALIFORNIA 91770 J. G. HAYNES 6
TELEPHONE MANAGER OF NUCLEAR OPERATIONS (213)572-742 February 23, 1981 U. S. Nuclear Regulatory Commission Office of Inspection and Enforcement Region V 1990 North California Boulevard Suite 202, Walnut Creek Plaza Walnut Creek, California 94596 Attention:
Mr. R. H. Engelken, Director DOCKET No. 50-206 U
SAN ONOFRE -
UNIT 1 Iasses r
Subject:
Instrument Air System Malfunction San Onofre -
Unit 1
References:
- 1)
K.P. Baskin (SCE) to D.M. Crutchfield (NRC) letter dated October 8, 1980.
Subject:
Failure of the Salt Water Cooling System.
- 2)
Licensee Event Reports80-003 and 80-032
Dear Sir:
As previously reported to you in references 1 and 2 above, the breakdown of silica gel desiccant in our instrument air dryers and subsequent by-passing of the instrument air filters, resulted in the contamination of our instrument air system. The purpose of this letter is to provide a status report on the corrective actions taken by us to date to ensure the continued operability of all safety related air using equipment at San Onofre, Unit 1.
The presence of gritty material in our air system was first observed during the October, 1978 refueling outage. This condition was not reported at that time as the affected equipment (feedwater regulator) was not safety related and the extent of the problem was believed to be limited to that equipment.
S 8103 1o 0r
U.S. NuclearlRegulatory Commission Page # 2 An investigation was conducted to determine the identity of the gritty material.
A visual inspection of the material indicated that it resembled a fine brown colored sand.
Samples of the gritty material, beach sand, sand blasting sand, and desiccant from the instrument air filter bags were sent to an independent lab to determine if the chemical composition of any of these materials were similar. The results of the sample analysis were not conclusive as all samples were similar in that they contained large percentages of silicon (Si) but differed in the composition of the remaining elements. It should be noted that silica gel desiccant resembles coarse rock salt and was not sent out for analysis as it was not similar to the gritty material.
On January 9, 1980 containment isolation valve CV-537, service water to containment, failed to close during routine testing. An investigation revealed that the gritty material had entered the solenoid valve and prevented if from operating. At this time it was discovered that the desiccant in the instrumentiair dryers was breaking down into small particles and by-passing the instrument air filters.
Discussions with the air dryer manufacturer indicated that silica gel will turn brown and break down into small particles if left in service too long.
The silica gel desiccant in the drying towers was subsequently replaced and temporary filters were installed in the instrument air filter housings to prevent air by-passing the filters. Plans were then made to blowdown air lines supplying safety related valves during the current refueling outage.
The instrument air filters are manufactured by the Beach Sta-Dry Filter Company and are designed to absorb water vapor, oil mist, and particulate solids as small as.5 microns.
Each of the three filter housings are cylindrical and contain two bags mounted in series within the housing.
Each bag contains an alumina oxide type desiccant.
The manufacturer states that the filter bags will conform to.the filter housing walls so that no air can by-pass the filter.
Our experience indicates that the filter bags will allow some air (and any entrained solids) to by-pass between the bags and filter housing wall.
Each of the two instrument air drying towers is charged with sufficient desiccant to remove moisture at the rated air flow capacity. The operation of each tower is cyclic iin that reactivation heating is required after each absorption cycle.
Discussions with the drying tower manufacturer indicate that desiccant can breakdown due to excessive service life, overheating during reactivation, or improper tower cycling. The breakdown of desiccant in.our air drying towers has been attributed to the excessive service life between replacement intervals.
U. S. Nuclear Regulatory Commission Page # 3 Actions taken to remove desiccant from the instrument air header included a blowdown of the header at numerous points selected to ensure complete removal of the desiccant from the main air header and secondary supply lines. These points include all accessible locations on the main header and all secondary lines supplying air to safety related devices. Individual air lines supplying air to safety related valves will be tested in accordance with approved procedures to ensure that the supplied air is free of particulate matter.
To prevent additional desiccant from entering the instrument air header, temporary filter pads were installed in the air filter housings to prevent desiccant from by-passing the main filters. A clear plastic see-through filter housing was added to provide a means to determine visually if any desiccant was entering the filter elements. As an interim measure the filter pads are changed at frequent intervals to ensure that any desiccant entering the air filters is promptly removed. A permanent positive seal filter system consisting of two full flow filter housings is tentatively sqheduled to be installed by March 14, 1981. These filters will be installed downstream of the desiccant towers and contain cartridge type filter elements desilgned to eliminate air by-passing of the filter media.
A preventive maintenance program for the instrument air dryers and filters is currently under development and will be implemented prior to the completion of the current outage. The program will provide for periodic sampling of desiccant and frequency of replacement. Operating procedures will also be revised to require routine observation of the drying towers and air filters. These procedures will provide for monitoring of desiccant temperature, proper drying cycle operation, and differential pressure across the air filters. A preventive maintenance program for safety related solenoid and control valves is also being developed and will be implemented by August, 1981.
As reported in Reference 1, our experience concerning the effects of desiccant on equipment performance indicate that there are two predominant mechanisms in which the presence of desiccant may lead to equipment failures.
The first mechanism identified is one where the desiccant enters a solenoid core in sufficient quantities to prevent proper operation. The second mechanism involves the gradual wear of solenoid components due to the abrasive action of the desiccant on moving parts.
U., S. Nuclear Regulatory Co(bsion Page # 4 A station procedure has been developed to ensure that all safety related equipment served by the instrument air system is included in a testing and inspection program. This program will ensure that each safety related solenoid, pilot valve, control valve and pneumatic instrument continues to function as designed. The air pressure to each solenoid will be verified to be within the design operating pressure of the solenoid and inlets to solenoid valves will be inspected for any indications of foreign material.
Should any contamination at the inlet be observed, the solenoid will be thoroughly cleaned and inspected for wear and parts replaced as necessary.
Samples of desiccant removed during the air header blowdown were sent to an independent lab for particle size analysis. The particle size distribution was determined to ensure that a filter media of the correct micron rating is installed in the new air filter housings. Since an earlier analysis indicated that iron was present in the desiccant removed from the air system, the lab was also requested to determine the iron content of the sample. The analysis indicated that the sample contained a mixture of red.
iron oxide and black iron oxide. Since there is an insignificant amount of iron present in silica gel desiccant,.it appears that the iron originated from the carbon steel piping installed at various locations in the instrument air system. The iron would be present as a result of original mill scale not removed during installation (black iron oxide) or from oxidation of the tarbon steel (red iron oxide).
The blowdown of the air header ensured that loose iron deposits have been removed from the instrument air system. The generation rate of iron oxide is a function of both the quantity of carbon steel in the system and the concentration of water vapor and oxygen. The generation rate of iron oxide particles within the carbon steel pipe will not be significant as instrument air is maintained at a very low dew point, minimizing the concentration of water vapor. In addition, the use of carbon steel pipe in the air system downstream of the instrument air filters is confined mainly to secondary supply lines of small diameter, i.e one inch or less. A refueling interval blowdown of the air header will be conducted however, to prevent any significant iron particle accumulation over time.
The above actions will ensure that the instrument air remains free of particulate matter and that all safety related equipment served by the instrument air system will function as designed.
If you have any questions or desire additional information concerning this subject, please contact me.
Sincerely, J. G. Haynes Manager of Nuclear Operations SS:dh:17U cc:
L. F. Miller (NRC Resident Inspector - San Onofre Unit 1)
Director, Office of Management Information & Program Control Director, Nuclear Safety Analysis Center