ML20141P237

From kanterella
Revision as of 07:17, 12 December 2021 by StriderTol (talk | contribs) (StriderTol Bot change)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

SER Re Util 860312 Plans for Mods to Facility in Order to Reduce Moisture Ingress Into Reactor Vessel.Design Mods Should Reduce Likelihood of Moisture Ingress.Licensee Committed to Conduct Further Studies If Ingress Continues
ML20141P237
Person / Time
Site: Fort Saint Vrain Xcel Energy icon.png
Issue date: 03/14/1986
From:
NRC
To:
Shared Package
ML20141P232 List:
References
TAC-59936, NUDOCS 8603190175
Download: ML20141P237 (5)


Text

~

. /  %, UNITED STATES

[ g NUCLEAR REGULATORY COMMISSION 5 j WASHINGTON, D. C. 20555 4

.....p SAFETY EVALUATION REPORT FORT ST. VRAIN NUCLEAR GENERATING STATION AUXILIARY SYSTEMS BRANCH '

Moisture Ingress Into The Primary System The Fort St. Vrain Nuclear Generating Station has had a significant history of problems associated with moisture in the primary coolant system. This moisture has almost exclusively been the result of helium circulator bearing water flowing up the shaft of the circulators during transient or upset conditions.

Numerous modifications of the helium circulation auxiliary system have been performed in an attempt to preclude or minimize moisture upsets. These included modifications performed by the vendor, Gulf General Atomic, prior to acceptance of the plant by Public Service Company (PSC) in 1979 and subsequent technical evaluations of the system and modifications by PSC thereafter.

A major plant outage took place in 1981/1982 during which a modification was made to isolate the buffer systems of the circulators of loop 1 from loop 2.

Although the primary purpose of this modification was to prevent circulator upsets in one loop from reflecting into the second, a significant goal of the modification was to reduce moisture ingress by minimizing the number of circulators responding to a system upset. Comparison of the records of circulator trips before and after splitting of the buffer loops indicate a substantial reduction in circulator trips which should result in a lowering of moisture ingress events.

However, moisture ingress events continued and the licensee has taken further actions to prevent moisture ingress. The actions of the licensee to eliminate moisture ingress events resulted in the evaluation of many possible modifica-tions to the helium circulator auxiliaries, evaluations of instrumentation hardware and software, evaluation of procedures, and evaluation of vario'us methods for improving transient analyses. Many activities were evaluated and dispositioned to include initiation of modifications which showed promise for 8603190175 860314 PDR ADOCK 05000267 P PDR

~

2 reduction of the frequency and severity of moisture ingress into th9 primary system. The following are system modifications that have been initiated to reduce moisture ingress into the primary system:

1. The licensee has determined that a high percentage of helium circulator transients, resulting in water ingress to the pre-stressed concrete reactor vessel (PCRV) are related to inability to drain water from the bearing cartridge. Several aspects of the system contributed to these water ingress problems. A loop seal existed between the circulator and the high pressure separator. During transients wherein water levels in the high pressure separator were changing rapidly, water from the loop seal was forced back into the circulator, resulting in an upset of the pressure relationships of buffer and bearing water within the circulator bearing water cartridge. During the recent outage, this piping has been modified-to eliminate the loop seals.
2. The high pressure separator previously drained into the main drain line.

During plant transients, this drain path had considerable resistance and in effect acted as a loop seal in the drain from the high pressure separator. Fluctuations in the bearing water surge tank levels often resulted in flooding of the high pressure separator. During this outage the high pressure separator drain was re-routed directly to the gas side of the bearing water surge tank. This change should provide proper high pressure separator draining during circulator transient conditions.

3. Another problem relating to flooding of the high pressure separator was the insufficient piping flow capacity of the system. The old system had a capacity of approximately 5 gallons per minute. This capacity was increased to 20 gallons per minute, which should help to eliminate high pressure separator flooding during transients. ,

3

4. The original main drain control system was pneumatic with a pneumatic operated valve. There is an inherent system lag in this type of system due to the characteristics of an air control system. It has been deter-mined from computer model studies that if the main drain valve responds fast enough, the circulator can go through severe plant transients without water ingress. In order to speed up the response of the main drain valve, pneumatic boosters were added. Plant transients that were experienced after installation of these boosters did not result in water ingress events.

While the booster improved the main drain performance, they also intro-duced control instability problems. When the licensee attempted to slow down the system to eliminate this instability, the effectiveness of the boosters was reduced. An investigation into better main drain control showed that converting the pneumatic system to electronic controls should make significant improvements. During the recent outage, the licensee did convert the main drain control for all four circulators to an elec-tronic system.

The electronic main drain controls for the circulators may enhance the performance of the existing valves so that they provide satisfactory response. With the installation of electronic control for the main drain system, the licensee is giving the operator more control options. The high pressure separator level control now provides a feed forward signal to the main drain setpoint. With the higher response speed of the electronic control system, this may not be necessary. The licensee is providing an option where the main drain is controlled strictly from main drain to buffer differential pressure. Computer model studies performed by GA indicate that this is a more effective mode of control. The licensee is also providing a manual mode of operation where the operator can take direct control of the main drain valve position from the contro1 room.

r

  • 4
5. In the past there have been circulator trips in which analysis of the transient could not satisfactorily identify the cause of the trip or the cause of the resulting moisture ingress. In order to pursue a more intense effort to resolve these moisture ingress events, the licensee performed an in-depth analysis of all plant transients such as circulator trips, loop shutdowns, and reactor scrams. One of the items identified by this transient review was the need for additional data points at a higher sampling frequency than the plant data logger could provide. A dedicated system has been installed to provide this information and additional inputs have been added as the need has been identified.
6. Another item identified by the licensee as a possible source of circulator trips was the possibility of spurious trips induced into the system. The licensee identified that the signal cables for the bearing water, buffer and mid-buffer pressure differential switches were not shielded. The licensee installed shielded cables during the recent outage. The new shielded cables were connected to the buffer, mid-buffer switches and many System 21 computer data signals during the last shutdown. It is the licensee's intent to replace the bearing water pressure differential switches in the future and the new shielded cables will be connected to these switches at that time.
7. During normal circulator operation, bearing water for the circulators is provided from the bearing water pumps. If a pump is lost, the back up bearing water automatically comes in to provide adequate water to the circulator bearings. If the pump was restarted, a surge of bearing water to the circulator bearing cartridge occurred until the automatic control on the back up bearing water could act to reduce this flow.

After the helium circulators are operated on back-up bearing water. it is necessary for the operator to return to the normal bearing water supply from bearing water pumps. A severe transient was sometimes ind'uced if the bearing water pumps were brought on line in addition to the backup bearing water source. The operator had no control over the rate at

5 which the normal bearing water was pumped into the system. The, licensee installed controls to allow the operator to bring normal bearing water supply in gradually, and this supply will back out the backup bearing water system so that this transition can be made in a more controlled manner.

8. During overhaul of the control rod drives, the licensee observed indica-tion that moisture may be coming into the drive mechanism through the purified helium header. A bearing water leak from the A circulator into the penetration interspace has been identified, and this is the probable source of moisture into the purified helium header. In order to detect and eliminate moisture from any source in the purified helium header, the licensee plans to install moisture removal devices and moisture monitors in the purified helium header. The moisture removal devices and instrumentation are not available for installation during this outage, however, piping revisions will be made to allow installation of these devices when they become available.

The licensee has identified several potential paths for moisture ingress into the reactor vessel and modified the plant design to reduce the likelihood of moisture ingress through these paths. Furthermore the licensee intends to continue to analyze moisture ingress and evaluate the effects of the plant modifications on reducing this problem.

Based on our review of the licensee's submittal we conclude that the design modifications to the Fort St. Vrain Plant should reduce the likelihood of moisture ingress into the reactor vessel; however, due to the minute moisture content limit (500 ppm) and the close proximity of helium circulator bearing cooling water to the reactor vessel penetration, moisture ingress may still cause the reactor moisture content limit to be exceeded. If moisture ingress to the reactor continues to exceed the moisture content limit the licensee has committed to conduct further studies and propose further design modifications to the plant.

_ _ _ - - _ _ _ _ _ - - . _ . _ _ _ _ _ _ . , _ __ _ _ ____ _ _ _ _ _. _ __ _ __