ML20207C773

Submits First Quarterly Rept on Status of Reactor Bldg Sump Beta Radiation Monitoring Sys.Monitor Installed on 860922. First Sample Release Caused Silicate Cell to Plug.Attempts to Unplug Cell Unsuccessful.No Addl Releases Monitored

ML20207C773
Person / Time
Site:	Fort Saint Vrain
Issue date:	12/22/1986
From:	Gahm J PUBLIC SERVICE CO. OF COLORADO
To:	Berkow H Office of Nuclear Reactor Regulation
References
P-86655, TAC-61580, NUDOCS 8612300233
Download: ML20207C773 (6)

Text

.

h Public Service- =: %

16805 WCR 19 1/2, Platteville, Colorado 80651 December 22, 1986 Fort St. Vrain Unit No. 1 P-86655 Director of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Mr. H. N. Berkow, Director Standardization and Special Projects Directorate Docket No. 50-267

SUBJECT:

Status of Reactor Building Sump In-Line Beta Monitor

REFERENCE:

PSC Letter, Gahm to Berkow, dated 9/18/86 (P-86559)

Dear Mr. Berkow:

In accordance with Public Service Company of Colorado's agreement with you this is the first quarterly report on the status of the Reactor Building Sump Beta radiation monitoring system.

The Reactor Building Sump Beta Monitoring System is comprised of the following components:

A positive displacement pump, dual piston (for non pulsating output), variable flow 0-10ml/ min., operating pressure from 0-6000 psig. The pump has two removable check valves containing a spring loaded ruby ball and sapphire seat. It is programmable for constant flow, constant pressure and shut off on minimum or maximum pressure. It was manufactured in West Germany.

8612300233 861222 ggl PDR ADOCK 05000267 R PDR 6{l

• i P-86655' December 22, 1986 .

e a Ramona-D Scintillation detection unit comprised of two-photo multiplier tubes 180lmdeglx-apart facing a lead lined light tight chamber for holding the scintillation cells. _ The unit incorporates coincidence circuitry, and a 1000 channel analyzer.

Key pad programming is used to display any number of channels in each of 2 windows as well_ as background subtract, and a correction for spill-over from one window to the other. It was also manufactured in West Germany.

Molytek 32 Channel recorder data logger, fully programmable.

Low volume flow cell from Fluid Components for determining a flow or no flow condition.

Scintillation cells -

2 different cells composed of a spiral wound teflon tube, encapsulated in a epoxy matrix, filled with either yttrium silicate or calcium fluoride scintillating

' material with a particle size of 40 micron. These Cells are capable of withstanding pressures to 900 psig.

Cabinet - Holds all of the above equipment.

Filtration System - A dual train filtration system with 3 filter-banks per train utilizing a 25 micron, 10 micron and 1 micron filter for each train.

The following is a chronology of Beta Monitor in-service operation.

The unit was installed on schedule 9-22-86 along with the associated filtration system.

Prior to the first Reactor _ Building Sump Release being monitored, a preliminary calibration was performed. A known tritium standard was pumped through the yttrium silicate cell. A window width and efficiency was determined based on counts per second/ microcurie per milliliter. A second standard, sulphur-35, was pumped through the yttrium. silicate cell and an efficiency and window width was determined based on counts per second/ microcurie per milliliter.

Based on the above results a correction for contribution of one channel to the other was calculated and programmed into the Ramona detection unit.

The first Reactor Building Sump sample released with the Beta monitor on line caused the yttrium silicate cell to plug and the pump to shut off on high pressure after approximately 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> of operation.

Attempts to unplug the cell using soap solutions and solvents were unsuccessful. Back flush of the cell was impractical due to the cell design.

P-86655 December 22, 1986

.A second Reactor Building Sump release was monitored using the calcium fluoride Cell. No calibration was performed, due to skepticism as to whether the cell would become plugged. The cell operated during the entire release, however the calcium fluoride became covered with a brown residue and pressure required to maintain the initial flow rate had nearly doubled. Subsequent cleaning of the calcium fluoride was unsuccessful. No further releases have been monitored. Until the resolution of some of the more critical problems, no further releases will be monitored.

Beta Monitor problems encountered during this reporting period:

Pump The pump check valves during the test period became inoperative, due to foreign particles on the valve seats or sticking of the ruby ball caused by the spring within the valve. The valves had to be removed, ultrasoaically cleaned and blown dry.

The piston seals developed a constant leak.

Scintillation Cells The small granule size (40 micron) of the scintillator in the calcium fluoride and yttrium silicate caused a large back pressure, 250 psig for the yttrium silicate and 100 psig for the calcium fluoride. Due to the small particle size, the cell had a filtering effect on all Reactor Building Sump water that flowed through it.

Cleaning by solvent did not work and the cells had no provision for the scintillation material to be removed for cleaning. The yttrium silicate cell developed a leak at 600 psig and was returned to the manufacturer for evaluation.

Filtration System Initial testing of the water from the outlet of the filtration system indicated particle sizes of greater than 1 micron.

Subsequent testing indicated the larger particle size had diminished but particles greater than 1 micron were still leaving the 1 micron filter bank. Traces of oil were detected in some of the filtered water samples.

t I

i .

P-86655 December 22, 1986

' Problem Resolution:

Pump New piston seals were ordered from the manufacturer to alleviate the piston leakage problem. New check valves were also ordered so that a supply could be on hand for rapid change out. As of late October, pump rebuild parts had not arrived and in early November the West German manufacturer of the pump agreed to take the pump back free of charge. On November 18th a new pump design was agreed upon. This pump is a magnetic drive variable speed electronic feed back, gear pump. Classified as a metering pump, it is positive displacement to 100 psig and has interchangeable heads for low flew capability. No check valves are utilized in this pump. The pump is available from Cole-Parmer of Chicago and manufactured by ISMATEC. Rebuild kits are readily available.

The first pump arrived on November 25, 1986. The speed control feed back loop to the motor appeared to have failed and was returned to Cole-Parmer the same day. At this time, we are awaiting its replacement.

Scintillation Cells The granule size of the scintillator causing a filtering media and an enormous back pressure was a major concern. A new design was conceived at the FSV Radiochemistry lab utilizing a larger diameter teflon tubing in the cell (approxirately twice the original size) and 80 micron particles. The West German manufacturer of the cells agreed to construct same using yttrium silicate scintillator.

FSV Radiochemistry laboratory personnel investigated our own cell design. Very little information was uncovered on the yttrium silicate scintillator other than it needed cerium as dopant. It was extremely expensive, and not available in the United States to our knowledge. The yttrium silicate was our first choice due to it's desirable inert characteristics and greater light output compared to the calcium fluoride cell.

Calcium fluoride was readily available as random cutting from Harshaw Filtrol Chemcial Company and Bicron Corporation. A 1 1/2" long x 1 1/2" diameter cylinder was purchased from Harshaw.

o P-86655 December 22, 1986 The Fort St. Vrain machine shop built a prototype cell of plexiglas with removable face plate, similar in a appearance to the West German cell. Off the shelf teflon tubing I.D. 4mm was used and sand blasting sand was ground and sieved to a 250-335 micron particle size. The tubing was packed with the sand particles and assembled in the cell. Testing of the cell showed that back pressure at full flow was 10-15 psig. Water was collected from the filtration system during a Reactor Building Sump release and processed through the prototype cell to test the pressure build up and ease of cleaning the cell. The pressure increased to approximately 20-25 psig and the cell was easily cleaned by back flush with a soap solution. The pressure reduced to near its original valve. Presently, the FSV machine shop is constructing two additional cells with the same dimensions as the prototype.

The West German cell arrived after 4-5 weeks and was tested for back pressure. The back pressure at low flow was extremely high (150-200 psig). A decision was made not to use the cell since there appeared to be no improvement over the previous 2 cells made in West Germany.

On November 17, the calcium fluoride arrived from Harshaw. The following day part of the crystal was ground sieved to obtain the 250-335 micron particles. The original prototype cell was disassembled, the sand was removed, and it was repacked with calcium fluoride. Back pressure testing showed a slightly higher back pressure (18 psig) than the sand filled prototype which we attribute to the kinks in the teflon tubing caused by its reuse.

Efficiency for tritium was calculated as being more than double the West German calcium fluoride cell and slightly higher than the yttrium silicate cell. Enough calcium fluoride is still available for an estimated 5-8 cells.

Filtration System At present, the cooperation of all site personnel is being sought to ensure water going to the Reactor Building Sump is as oil free as possible. This issue has been brought to the attention of management and routinely re-emphasized.

As of late November, some hydro-lasing (high pressure water cleaning) of the Reactor Building Sump has been completed in an effort to keep the sump clean. Additional measures are being investigated during change out of filters to ensure that water will in no way bypass any of the filters, and that each filter bank removes it's designed particle size.

l i

.e j i

P-86655 December 22, 1986 Should' you have any questions, please contact Mr. M. H. Holmes, (303) 480-6960.

Sincerely, 30)Qdw bf 66 09WE J. W. Gahm Manager, Nuclear Production Fort St. Vrain Nuclear Generating Station VJM:JWG/klw 9

r t-t e

v'*N-*y yM+<-4vv*,w-vyyp w'w e wev vygg--- ww-.--e--------rde*.- e -ae, em-w+vv.r-vnw.~-.-v ms--v----*