ML20055C269
| ML20055C269 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 02/20/1990 |
| From: | Cockfield D PORTLAND GENERAL ELECTRIC CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9003010034 | |
| Download: ML20055C269 (12) | |
Text
._
0 e
=s=.
P g_==g ortlandGeneralBedricCompany David W. CocKfield Vice President, Nuclear j
Febtvary 20, 1990 Trojan Nuclear Plant j
Docket 50-344
'i License NPF-1 1
U.S. Nuclear Regulatory Commission ATTW: Document Control Desk l
Washington DC 20555
Dear Sir:
j Reply to a Notice of Violation Your. letter of January 19, 1990. transmitted two Notices of Violation associated with Nuclear Regulatory Commission Inspection Report 50-344/89-30. Attachment.1 is our response to the Notices of Violation. Attachment 2 provides our response to the request for additional information pertaining to our-analysis of causative factors to i
Violation A of Appendix A and the final conclusion reached regarding the.
investigation involving monitoring of the hydrogen ventilation system. providas our response-to Appendix B, Notice of Deviation.
Sincerely, i
Attachment c:
Mr. John B. Martin Regional Administrator, Region V
-l U.S. Nuclear Regulatory Commission Mr. David'StewarteSmith l
State of Oregon Department of Energy Mr. R. C. Barr d
NRC Resident Inspector Trojan Nuclesr Plant
([#q fi 9003010034 900220 PDR ADOCK 05000344 I'
o PDC 121 S.W. Salmon Street. Portiand Oregon 97204,503/464-8897
\\
t Trojan Nuclear Plant Document Control Desk-I Docket 50-344
-February 20,'1990 License NPF-1 l
Page 1 of 5 e
Violation A__
Trojan Technical Specification (TTS)' 6.8, " Procedures and Programs", reads, 'in-O
.part::
6.8.1 Written procedures shall be established, implemented and maintained covering the activities referenced.below:
a.
The applicable procedures recommended in Appendix ".A" of Regulatory Guide (RG)'1.33, November 1972.
.RG.1.33 (Safety Guide 33),: Appendix A, reads, in part:
A.
Administrative Procedures.
4.
' Procedure Adherence and-Temporary Change Method.
I.
Procedures for Performing Maintenance.
- 5.. General Procedures for the control of maintenance, repair, replacement, and modification work should' be prepared before-reactor operation is begun.. These procedures should include infonmation on areas'such as the:following:
1 Method for obtaining permission and clearance' for L
a.
operations personnel to work and for logging.of such q
work...
Administrative Order (AO)-3-9, " Maintenance Requests", reads','in part:
4.'5.3.f.
Immediately-bring to the attention of the work group craft supervisor and the Shift Supervisor problens~that:
- 1) Cannot be corrected without additional work direction.
4.6.1 Before the end of his shift, the craftsman / technician shall take any MRs to the Shift-Supervisor or-designee forLreview of p
post-maintenance testing requirements. This includes MRs which are ready for clearance release or ready toibe turned over to operations for use..
4.6.3 The 'Shif t Supervisor or designee shall initial tho SRO l
verification block 36 of the MR to' document his-review.
5
~
Trojan Nuclear Plant Document Control Desk Docket 50-344 February 20, 1990 License NPF-1 Page 2 of 5 4.6.7 ' The individual (s) performing any designated Post-tfaintenance testing shall:
a.
Document results in block 33 of the MR and sign'in entry block 34.
Contrary to~the above, on September 24, 1989, a maintenance worker' installed a Temporary Modification to Train B of the hydrogen vent system
- but signed for installing it to Train A, failing to stop work and obtain further direction when the direction provided in the Maintenance Request (MR) was inadequate to the work situation.
The Operations representative reviewing the associated MR and the individus1 performing the post-maintenance test failed to adequately review the MR to identify that work had been requested on Train B but had been signed-off as completed on Train A.
The Operations representative also failed to sign Block 36 to document the review.
This is a Severity Level IV violation.
Response
Portland Ceneral Electric Company (PGE) acknowledges the violation.
1.-
Reason for the violation:
The reason for the violation was personnel error, failure to. follow procedure delineated in Administrative Order'A0 3-9, Maintenance Requests. Contributing to the personnel error were additional we;Nnesses identified in the work control program. These items were:
The failure of supervisory personnel to properly review work packages with the craftsman prior to the start of work; inadequate supervision of the work in progress; and the failure of the craftsman involved to stop work-and seek: further guidance from superevisors.
Inadequate labeling of components and inadequate work instructions also contributed to the violation.
2.
The coneective steps that have been -taken and the results achieved:
l Hydrogen Ventilation System Train A (CS-9A), was tagged out.to-prevent further use while the Process Effluent Radiation Monitor (PERM)-1 was connected to Hydrogen Ventilation System Train B (CS-9B). 'A system walkdown was-porformed per Plant Systems Engineering Procedure-(PSEP) 40-3 to verify that PERM-1 was connected to CS-9B.
CS-9B was tested following'.the walkdown confirmation to verify-it was connected to PERM-1 and was designated as the means-for pressure reduction.
Trains A and B of CS-9,were identified and
,.C marked where the pipes connect to the purge exhaust and refueling t
1 Trojan Nuclear Plant Document Control Desk j
Docket 50-344 February 20, 1990 i
License NPF-1 Page 3 of 5 j
l pool supply und exhaust system (CS-2).
Temporary Modificatzon
]
(TM)89-072, Revision 1 was issued to limit the TM to only Train B of.
CS-9..
The Mechanical Maintenance Supervisor discussed the need to identify specifically the train on which. work was to be performed prior to commencing the work with the personnel of the Mechanical Maintenance Branch. This discussion also addressed the need to. assure the work closure documentation is' consistent with the work performed.
3.
The corrective steps that will be taken to avoid further violations:
Administrative Order (AO) 5-8, " Temporary. Modifications" will be evaluated for the need to improve control over temporary modifications issued for both trains.of a system.- This evaluation will be completed by March-1, 1990.
i The process of using identical field and document copies of '
Naintenance Requesto will be evaluated for inclusion into the new Work Control program. This evaluation will be completed by August 1, 1990.
4.
The date when full compliance will be achieved.:
Full compliance was achieved on September 25,.1989 following the identification of the installation ceror and the implemontation of the immediate corrective actions.
)
f Trojan Nuclear Plant Document Control Desk Docket 50-344 February 20, 1990 License NPF-1 Page 4 of 5 Violation B 10 CFR 20.311 reads, in part:
Any generating licensee who transfers radioactive waste to a land disposal facility.
. shall:
(1) Prepare all wastes so that the waste is class 5fied according to Section 61.55.
l 10 CFR 61.55 reads, in part:
Determination of the classification of radioactive waste involves two considerations.
First, consideration must be given-to the concentration of long-lived radionuclides.
. Second, consideration must be given to the concentration of shorter-lived radionuclides.
. The concentration of a radionuclide may be determined by indirect methods.
if there'is reasonable assurance that the indirect methods can be correlatei with actual measurements. The concentration of a radiecsclide may be. averaged
-over the volume of the waste,' or weight of the waste if the units are expressed as nanocuries per gram.
Contrary to 20.311 above, radioactive waste Shipment 89-90 used the volume of the resin liners in the shipment for determination _of radionuclide concentrations rather than the volume of-the resin itself.
l This resulted in an underestimation of the specific radionuclide concentrations required by Section 61.55 for waste classification and L
could have resulted in a misclassification if activities in the resin had j
been higher.
l This is a Severity Levol IV violation.
Response
l PGE acknowledges the violation.
1.
Reason for the violation:
The reason for the Violation was procedural inadequacy.
The l
procedure used did not accurately implement the-requirement of 10 CFR 61.55 to calculate radionuclide concentrations by averaging over the volume of the waste.
Instead, the procedure averaged the L
concentration over the volume of the container.
2.
The corrective steps that have been taken and the results achieved:
Further shipments of spent resin were suspended upon identification of the discrepancy, s%
I Trojan Nuclear plant Document Control D.sk Docket 50-344 February 20, 1990' License NPF-1 Page 5 of 5 3.
The corrective steps that will be taken to avoid further violations:
PGE will perform a test.to determine the actual density'of dewatered resin. Radwaste procedures will be revised t.o compute radionuclide concentrations based on this density'and the measured resin weight.
Spent resin' shipments will continue to be suspended until these actions are complete.. In the event that a shipment must be made, procedures will be deviated to specify a conservative resin density of 1.1 grams per cubic centimeter.'
4.
Date when full compliance will be achieved:
Full compliance will be schieved by March 31, 1990, following the completion of the resin density. testing and implementation of the results into the procedure governing the calculational method used to determine radionuclide concentrations.
\\
3 I
i 9
~
i i
J WJW/4325W
'i m.
E z
i
L 5
L Trojan Nuclear Plant Document Control Desk Dockot-50-344 February 20, 1990 I
License WPF-1 Page 1 of 2 i
f Recuest for Additional Information
{
)-
1.
Causative Factor Review:
~
Your' letter of January 19, 1990 stated an interest in our analysis of the :ausative factors of violation A of Appendix A of your transmittal and Violation B.3 of Appendix A from your letter to PGE dated March _ 15, 1989. Wo.found the causative factors discussed in eur response to Violation A to essentially be the same as those 1
l Factors contributing to Violation B.3 of March 15, 1989.
Both of the j
violations received pertain to ineffective work control practices.
~
L l.
Your letter also requested an explanation of how the Nuclear Division Improvement Plan (NDIP) should have or will preclude recurrence.
The Nuclear Division Improvement Plant-(NDIP) was issued on September 29, 1989, and addressed key issues w;.ere improvement was recognized as being needed. Work control was one arva addressed, It l
was recognized that PCE had to investigate the causative factors of I
violations involving the Trojan work control process and develop an.
(:
in-depth plan to address the areas of weakness to improve performance in this key area.
Revision 1 to'the NDIP was issued on-January 31, 1990 and further addresses the improvements planned for.the work control system.
The improvements planned include a centralized work control center (WCC) l containing Operations personnel that.will have the responsibility to j
review maintenance requests and work packages for quality and completeneas. The WCC wi11' control and coordinate equipment l
clearances and surveillance testing for all Plant work groups.
i Survelliances will be coordinated with equipment outages to ensure j
1bota trains of safety-related equipment are not removed from service at the same time.
The WCC will assign the priority af maintenance requests, with the exception of urgent requests, which will continue -
l to be done=by the Shift Supervisor.
L The objective of the NDIP is to move the-behavior of.the organization j
in a direction toward achieving the organizational values described j
in the plan.
These values-are needed to prevent events like the l
-Hydrogen Ventilation System monitoring incident from recurring.
Attention to detail, quality teamwork, appropriate supervision are
{
values that,.had they been adhered to, most likely would have j
prevented this type of event. By affecting-these; types of value
'l changes along with enhancements in the work control process, the recurrence of events of this nature will be precluded.
]
1 A
u-1
Trojan Nuclear Plant Document Control Desk Docket 50-344 February 20, 1990 Licens9 #PF-1 Page 2 of 2 2.
Root Cause Analysis:
Your letter referenced above also expressed concern over the-nbjectivity of the root cause analysis of the event 3nvolving the; radiation monitoring of the Hydrogen Vantilation System (HVS), anG' whether'the wrong train of this system was being monitored during :the pericd mid-July to mid-September 1989..
During the investigation of the event, Chemistry was requested to review release data to determine if the train'the monitor was aligned to could be identified.
Chemistry performed a reanalysis of the last three Containment pressure reductions (September 5,-1989; September 9, 1989; and September 13, 1989) that were performed = prior
^
to the maintenance request ~to remove the sample probe from the MVS.
Train A.
This analysis compared the~ count rate data from Process and 2ffluent Radiological Monitor (PERM)-1C (low-level gas channel monitor) prior to (sample taken from Cuntainment) and immediately after commencement (sample taken from hydrogen ventilation duct) of each of the;e three ventings. When corrected.for sampling vacuum, the venting-countrate agreed within 95% of the pre-venting (in containment) countrate. As a result, it was concluded that PERM-IC
~
was monitoring ~ Train A of the HVS.
PERM-10 was used: for the comparison as it was most likely to provide.the best data since the particulate and iodine channels did not show any levels-above background in the Containment.
While this evaluation did not furnish conclusive proof, it did provide an objective basis for concluding PERM-1 was monitoring the Train A vent from the mid-July to mid-September time period. The Performance Monitoring Evcnts Analysis (PMEA) Branch of the Nuclear Quality. Assurance Department assigned the Human Performance Evaluation System (HPES) coordinator the task of evaluating the event to provide an objective root cause c
analysis. The HPES coordinator used-Chemistry's' information and the testimony of the Maintenance Craftsman as inputs to the event report investigation so that after careful review of all infermation,ta final conclusion could be reached.
The event report has now been completed and concludes thc_ sample line was connected to Train B of the HVS from the time pettod of July 13, 1989 to September 18, 1989.
It has been postulated that the actual location of the:cample probe in Train B would have provided~ representative data to reflect the true effluent release from Train A because of its physical location in front of the outlet port of Train B.
The Train B discharge into the= purge exhaust duct is slightly downstream of the train A HVS inlet port. Since correlation of the PERM-1C'"in-Containment"
~
readings with HVS readings indicated values within 95 percent of each other, it was concluded that the releases between July 13, 1989 and September 18, 1989, were-adequately monitored.
-WJW/4325W s
9 Trojan Nuclear Plant Docurent Control Desk Docket 50-344
' Fet: n ary 20, 1990 License NPF-1 Attaehment 3 Page 1 of 4 4
i Appendix B i
Notice of Deviation A.. Trojan Final Safety-Analysis Report, Parsgraph 11.5.2.2, continuous cas Monitoring Systems. provides that. sampling devices and procedures reflect the recommendations of American Society of Mechanicel Engineers (ANSI) N13.1-1969, Guide to Sampling-Airborne Radioactive Material in Nuclear Facilities.
ANSI N13.1-1969, Appendix A. Guides-for Sampling from Ducts and Stacks, reads, in part.
_j The point at which a duet or stack should be entered for sampling should be selected after taking into account several factors.
In the region immediately~ downstream from bends or severe transi-tions, flow patterns may:be markedly distorted.
These locations should be avoided... Generally, the distance from the transi-tion or elbow to the point of sampling should be a minimum of five and perferably ten or more-diameters downstream.
. It is recommended that the velocity distribution be measured at_-the anticipated section to determine that flow is fully developed-and mixing complete.
. For laminar follow l(Re 1 100) in circular 2
ducts the flow profile is'a parabolic distribution of~
velocities.
The average velocity. occurs at about 0.7 the radial distance from the axis of the duct to the wall. A sample drawn from this point-through the inlet of the probe atla velocity equal to the average velocity in the' duct will; furnish a' valid sample.
. At greater velocities and flows, turbulent
' flow will likely exist.
Contrary to the above, the physical configuration to Temporary Modification (TM)89-072, installed to the Process Radiation Monitor (PRM)-1 sampling system on-July 13, 1989, did not meet the criteria, as specified in ANSI N13.1-1969,- for isokinetic sampling of thel 1
Hydrogen Vent System (HVS) effluent during Containment pressure reductions in that the sample line was not positioned in the HVS duct but in front of the HVS nozzle penetration into.the purge duct,t providing a radical flow transition; the sample line was not rigidly
~
. fixed in position allowing it to bo moved about by the' air flow, and
~
the sample line was secured by stay-ties end tape which would be expected to further disturb the air flow.
This is a deviation.
f
,....w
. sm...
L l
6 i
ir Trojan Nuclear Plant Document Control Desk l'
Docket 50-344 February 20, 1990 l
l License'NPF-1 Page.2 of 4 l
Response
1.
Reason for the Deviation.
The reason for the deviation was.to implement a revised sampling method to ensure the effectiveness of Process and Effluent Radiation Monitor (PERM)-1 during Containment pressure reductions.
It was recognized that there was a need to modify the existing PERM-1 sample inlet location-due to the original location of the sample-probe providing poor monitoring of the effluent during operation of the HVS.- Request for Design Change-(RDC)88-023 was_ initiated.
In addition. TM 88-079 was initiated to modify the sample location prior, to implemcoting the permanent design change.
The TM associnted with the desil.n change was installed 'and testing was performed to determins sampling performance. The results were compared to the original design _ configuration.- The sample results from the TM provided a better response signal to PERM-1 than the original sample arrangement.
It was recognized that the TM sample-configuration's.
better response time at the o?'let of the HVS nozzle would' allow an increase in the response ti-associated with.a Containment' Isolation.
Signal.
In this regard, th. fM configuration was viewed to be an improvement over the original configuration.
This evaluation was also used for TM 89-072 (update to the original TM 88-079).
Your letter expressed concern that the physical configuration of TM 89-072 did not appear to meet the criteria.- as specified in ANSI N13.1-1969, for isokinetic sampling.of the HVS effluent during containment pressure reductions in that the sample-line was not positioned in the HVS duct but in front of the nozzle in an area'of radical flow transition.
The TM called for the sample tube to be inserted into the HVS piping past the 2-inch nozzle. This configuration placed the sangAe tube in close proximity to an elbow in the 4-inch pipe.
It was recognized that this configuration was a compromise from the ideal configuration at the time the recom-7 mendation was made; however, it was considered to be acceptable for.
t the following reasons:
l L
ANSI-N13.1-1969_ points out'that, "Even with complete awareness of l
the many conditicas which must.be realized to obtain a l
representative sample, the person responsible fer the design must l
almost always arrivelat a compromise." The guide continues, "When compromises from the ideal are required, the degree to which-a sample may be in: error should be estimated and a determination made to see if this is tolerable for the particular use of the data obtained."
L F
I i
il'
P Trojan Nuclear plant Document Control Desk Docket 50-344 February 20, 1990 License NPF-1 Page 3 of 4 Engineering reviewed the reasoning behind the requiremant of placing the sample probe five to ten major duct diameters downstream of a flow transition.
The concern was, in the region immediately downstream from bends or severe transitions, flow patterns may be markedly distorted. This can cause particle segregation according to size if particles are greater than about five microns due to inertial effects. Additionally, establishing the correct sampling rate for isokinetic flow (if required) at i
such positions may be difficult due to cross currents or other preferred motions of flow set up by the transition or bends.
It was recognized that our HEPA and charcoal filters reraoved particles greater than five microns diameter with a nigh degree of efficiency from our Hydrogen Vont System air stream. As such, particle segregation was not a concern..The concern for isokinetic sampling also became a moot point as the Nuclear Regulatory Commission provided a clarification to Trojan in a letter [ Nuclear Regulatory Commission (NRC) to PGE letter March 1, 1982, Clark to Withers re: NUREG 0737 clarification]
which states that " isokinetic sampling is not necessary for effluent streams which contain a predominance of particles smaller than 5 microns." Therefore, by determining that we had no sampling error from placement of the sample tube at or near a flow transition, we evaluated our compromiso from the ideal configuration and determined we met the requirements of ANSI N13.1-1969.
The inspection report also commented ca the ability of PERM 11 to detect noble gas activity in the purgo exhaust duct subsequent-to a Design Basis Accident (DBA), with the TM installed. When the TM was recommended, this concern was identified by Nuclear Plant Engineering and Chemistry and a contingency was provided. Chemistry procedures 1
CMP-55 and CMP-58 were revised to require monitoring of the purge exhaust duct as long as the TM was installed by means of grab samples j
using a portable sample pump.
l 2.
The corrective steps that have been taken to avoid-further deviations and the results achieved.
Plant Systems Engineering reviewed the concerns of the-NRC and the sequence of events involved in Inspection. Report 50-344/89-30. A new -
TM (TM 90-006) was written to specifically address the deviation from the Final Safety Analysis Report (FSAR) and ANSI N13.1-1969.
This TM will continue to be in use during the remainder of the.
1989-1990 Operating Cycle to monitor Containment internal pressure reduction through the HVS.
t a
A 5;
Trojan Nuclear Plant Document Control Desk
'i Docket 50-344 February 20, 1990 5
License NPF-1 -
-l Page'4 of 4-
+
3.
Final corrective action will-be'the completion'of the original".
RDC 88-023 during the 1990 Refueling Outage. This modification'will-conform to the standards as necessary to prevent further deviations..
i The use of the TM (TM 90-006) will then no longer be necessary.J i
+
WJW/4325W
.i 5
P P
.\\
?
-?
e P
5 rt i
.r
. f r.
'