ML19257A388

SEP Topic V-12.A, Water Purity of BWR Primary Coolant, Revised Safety Assessment

ML19257A388
Person / Time
Site:	Millstone
Issue date:	12/17/1979
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML19257A386	List: ML19257A385 ML19257A388
References
TASK-05-12.A, TASK-5-12.A, TASK-RR NUDOCS 8001030919
Download: ML19257A388 (11)

Text

MILLSTONE 1 TOPIC V-12.A WATER PURITY OF B0!L'"G WATER REACTOR PRIMARY COOLANT Topic V-12.A was included in the Systematic ' Evaluation Program in order that BWR compliance with Regulatory Guide 1.56 be determined. As part of this review, the guidance of Regulatory Juide 1.56, Revision 1, was compared against the facility design. This Regulatory Guide had been specifically identified by the NRC's Regulatory Requirements Review Comittee' as needing consideration for backfit on operating reactors. The review documents whether the facility design complies with the guidance of the Regulatory Guide or has some equivalent alternative acceptable to the staff. The acceptability or non-acceptability of the deviations listed below and the need for further action shall be judged during the integrated assessment for this facility.

Although the topic definition mentions that there are proposed revisions to Regulatory Guide 1.56, and although NUREG-0531 (" Investigation and Evaluation o'f Stress Corrosion Cracking in Piping of Light Water Reactor Plants,"

February 1979) recomends oxygen control in BWRs, we have detemined that Regulatory Guide 1.56, Revision 1, July 1978, in its "For Coment" form, is the latest approved staff guidance and have therefore based our review upon this guidance. The specific points of the regulatory position are quoted below and an explanation of the licensee's degree of conformance follows.

Information for this review was obtained from the licensee's Technical Specifications, the Millstone Unit No.1 Final Safety Analysis Report (FSAR),

applicable plant drawings, April 25, 1979 telephone conversations with his staff, and Appendix H to the Special Report, Chloride Intrusion Incident, February 9,1973. This report was written to document the circumstances of the September 1,1972 salt water intrusion and the inspections and modifications 8001039 j

December 17, 1979

which followed. The salt water leakage resulted in a peak reactor water chloride level of 15 ppm and a minimum rtactor water pH of 3.

I.

Regulatory Position 1.

The licensee should establish appropriate limits for the electrical conductivity of purified condensate to the reactor vessel (the electrical conductivity of the BWR feedwater cycle and that of the reactor water cleanup cycle). Separate limits may be required for such operating conditions as startup, hot standby, low power, high power, and at temperatures 5,2120F (1000C).

Chemical analyses for dissolved and suspended impurities should be perforr.ed as called for in the plant technical specifications.

A conductivity meter should be provided at each condenser hotwell or in the line between the hotwell and the condensate demineralizer with sufficient range to measure at least all levels of conductivity up to and including the limiting conditions of the technical specifications that require imediate shutdown of the reactor. The recording conductivity meters recomended in regulatory position 4.a may be used for this purpose.

Millstone 1 Technical Specification 3/4 6.C includes limits and requirements '

for the sampling of reactor coolant conductivity and chloride ions. Separate limits have been established for startup, steaming at less than 100,000 pounds per hour and steaming at or greater than 100,000 pounds per hour. Specific reactor coolant conductivity and chloride sampling intervals have been established at once per four hours during startups and at steaming rates below 100,000 pounds / hour, and at least every 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br /> (or when continuous conductivity monitors indicate abnormal conductivity (other than short-term spikes of one minute or less)] when steaming at rates equal % or greater than 100,000 pounds / hour.

Although there is no specification on the conductivity of the Reactor Water Cleanup (RWCU) System, there is a control room alarm on the RWCU demineralyzer discharge to alert the operators to the need to change the system lineup to another demineralyzer (although the system could be run at one and one-half times the 1668 196

normel throughput using all three demineralizers, Millstone 1 utilizes it as a 2/3 capacity system with one demineralizer as a standby unit.) Plant procedures do require that reactor coolant silica levels be checked once per week.

If there is indication that the silica level is increasing, then the discharge of the RWCU demineralyzers on service is sampled and, if warranted, flow is shifted to the remaining RWCU demineralizer. The RWCU resins are not regenerated, but are replaced upon silica breakthrough as determined above.

There is no Technical Specification or plant procedure which calls for the chemical analyses for dissolved and suspended imourities. How:ver, given the inline monitoring capability. sampling, and determination of resin capacities as explained below, we have determined that s'uch analyses would add little it. formation and are therefore not necessary.

iI. Regulatory Position 2.

The licensee should' establish the sequential resin regeneration frequency or resin replacement frequency required to maintain adequate capacity margin in the condensate treatment system for postulated condenser cooling water inleakage. The capacity required and corresponding resin regeneration or replacement frequency will depend on several parameters, including condenser cooling water composition, chloride concentration, flow rate in each demineralizer unit, type and quantity of resin, cation / anion resin ratio, postulated condenser leakage, and time for orderly reactor shutdown.

As noted in the February 9,1973 Appendix H to the Special Report, " Chloride Intrusion Incident", a resin regeneration scheme has been placed in force at Millstone 1.

Technical Specification 3/4 6.J statas the Limiting Condition 1668 197

for Operation and the Surveillance Requirements. Although the Millstone 1 method does not call for regeneration at any particular frequency, it does establish means by which the remaining ion exchange capacity is monitored, does specify when regeneration shall take. place, and also provides a criterion for the total replacement of the resin bed (s). First, it is important to note that the Millstone 1 condensate demineralizer system consists of seven demineralizer units, only six of which are needed to accocTnodate full flow.

Thus the seventh (on an alternating basis) remains regenerated and acts as a spare.

Plant procedures, which form part of the basis for this review, require that flow be measured every four hours through each condensate demineralIzer on service and that once per day the unused capacity of each bed be calculated by combining flow with the integrated conductivity readings for the day. This represents a conservative and beneficial tightening of the requirements of Technical

~

Specification 3/4 6.J. which would require weekly detemination of anionic exchange capacity when influent conductivity is between 0.055 and 0.3 umho/cm and daily determination when influent conductivity is equal to or greater than 0.3 umho/cm. It should also be noted that, although Specification 3/4 6.J would require regeneration of a condensate demineralizer resin charge before the unused capacity reaches a minimum value of 30 pounds of chloride ion, plant procedures establish the lower limit at 34 pounds, again providing additional ma rgi n.

To quote from the Technical Specification bases:

1668 198

.a.

5-

"The criteria of the resin monitor,ing program and the resin replacement program have been established to protect the reactor from high chloride levels should a seawater leak occur in the main condenser. These criteria will provide for a minimum unused capacity of 30 pounds of chloride ion (50 percent depletion in a resin which is approaching the limit of 0.75 meq/ml) before a planned regeneration of a resin. Shou ~ i a seawater leak occur when a resin has 30 pounds of capacity remaining, this criteria will allow a sufficient buffer for an orderly plant shutdown.

The resin depletion can be calculated using the measured salt-splitting capacity, the flow through the bed, and the average influent conductivity. Based on this result, a depletion can be calculated which will assure a 30 pound chloride ion exchange rese rve. Regeneration prior to this level of depletion will assure a sufficient ion exchange reserve for removal of chloride from the condensate system."

We find this program, accepted previously, which included the incorporation of revised Technical Specification, to be acceptable.

III. Regulatory Position 3.

The initial total capacity of the new anion and cation deminerali:er resins should be measured. Anion exchange capacity may be detemined by a procedure recommended by the resin manufacturer.

The total exchange capacity of the cation resin may be measured by a procedure recomended by the resin manufacturer or by paragraphs 41 through 49 of ASTM D2187-71, " Standard Methods of Test of Physical and Chemical Properties of Ion-Exchange Resins." For resins that are to be regenerated, these deteminations should be repeated at least semiannually. The resins should be discarded and replaced when their capacity following regeneration falls below 60 percent of the initial value. More frequent determinations should be made at plants using seawater or other water containing large amounts of dissolved or suspended matter as coolant in their heat exchangers.

For resins that are not regenerated but are instead replaced periodically with material of the same type, measurements of initial capacity should be made on a samole of new material at least once a year (when the time between replacements is less than 1 year) or at each replacement (when the time between replacements exceeds 1 year). When the type of anion or cation resin is changed, a measurement of total capacity of the replacement resin should be made prior to use in the demineralizer.

1668 199

6-Millstone 1 does not detemine the initial total capacity of new anion and cation demineralizer resins. Instead, the ' resin capacity, in the fom of remaining salt-splitting capacity, is detemined quarterly in accordance with Technical Specification 3/4 6.J.

To quote again from the above-referenced Appendix H:

"The salt-splitting capacity of ' anion resins is the most limiting factor of the condensate demineralizing system. Based on the lowest allowable anion capacity of 0.75 meq/ml, these resins can remove 61.2 lbs of chloride fn:m the condensate per demineralizer charge.

Through September 1,1972, the anion resins at Millstone have been subjected to very harsh conditions. Since startup of the unit, however, the salt-splitting capacity of the resins has only degraded 15% from its initial value of 1.2 milliequivalents.

A, quarterly sampling frequency therefore will be sufficient to detect the slow, long-tem degradation of the resin. The salt-splitting capacity is obtained in the laboratory using station resin analysis procedures.

We concur that the quarterly detemination of salt-splitting capacity is satisfactory and agree with the licensee that initial resin capacity detemination is unnecessary, especially when the detemination is made quarterly. Note that, were the new capacity to be 1.2 meq/ml as stated in the licensee's report, the lower limit of.75 meq/ml represents 62.5% of the original, which cc.mpares favorably with the SO: limit in the regulatory position.

IV. Regulatory Position 4 The licensee should verify the minimum residual demineralizer capacity in the most depleted demineralizer unit established in accordance with the recommendations of regulatory position 2 is maintained. The following is an example of an acceptable method 1668 200

l.

p 7

for detemining the condition of the demineralizer units so that the ion exchange resin can be replaced or regenerated before an unacceptable level of depletion is reached, a.

Recording conductivity meters should be installed at the inlet and outlet of both the condensate treatment system and reactor water cleanup system. The range of these instruments should be sufficient to measure all levels of potential water conductivity specified in the plant technical specifications. For the condensate treatment system, the conductivity meter readings should be calibrated so that estimates of condenser leakage can be made based on cooling-water conductivity, condensate conductivity, and flow rate. The chemical composition of the cooling water and its relation to specific conductance should be established and periodically confimed so that estimates of residual demineralizer capacity i

can be made.

i b.

A recording flowmeter should be used to measure the rate of flow through each demineralizer.

c.

The quantity of the principal ion (s) likely to cause demineralizer breakthrough should be calculated by:

(1) Converting the conductivity readings of the water entering the demineralizer to weight fraction (e.g., ppm or ppb) of the principal ion (s) and (2)

Integrating over time the product of concentration of this ion (s) and demineralizer flow.

The input quantity of ion (s) to the demineralizers should be determined at a frequency adequate to ensure sufficient residual ion exchange capacity in the event of a major condenser leakage to prevent exceeding reactor coolant limits.

d.

Each demineralizer unit should be replaced or regenerated when the remaining capacity (calculated by subtracting the total utilization detemined from conductivity and flow measurements in accordance with regulatory position 4.c from the initial capacity determined in accordance with regulatory position 3) approaches the minimum residual demineralizer capacity detemined in accordance with regulatory position 2.

The accuracy of the above calculation

~ should be checked by measurements made on resin samples taken when demineralizer units are removed from service for regeneration or resin cleaning. Measurements on samples from each unit shculd be made at each of the first two such removals from service and at every fifth such removal from service thereafter.

If appropriate, the actual measurements may be used to adjust the calculated value cf residual demineralizer capacity. Such adjustment and its justification should be reported to the NRC in the annual operating report. -

1668 201

~

The preceding sections have demonstrated that Millstone I has an acceptable alternative procedure to the one detailed above. The instruments of the condensate demineralizer and RCW systems are listed below.

A.

Conductivity instruments:

(1) condensate system -

(a) discharge of hotwell - four, inline, dual range (0-1, 0-10 umho/cm, all read in control room, alam setpoint.5 tsnho/cm (b) condensate pump discharge header - two, inline, one 0-2 umho/cm local readout one 0-25 umho/cm, read in control room, alarm setpoints 1.0 umho/cm and 15 umho/cm on the 0-25 unit and 1 umho/cm on 0-2 unit (c) condensate demineralizer effluent monitors - one inline for each of the seven units, 0.2 umho/cm, local readout, alarm setpoint.1 umho/cm (d) condensate booster pump discharge - two, inline. 0-1 umho/cm, read in control room, alarm setpoints.2 umho/cm and.5 umho/cm for both units (e) feedwater system - one, offline. 0-1.- umho/cm, read in control room, alarm setpoint.2 umho/cm (2) Reactor Water Cleanup System (a) influent - one offline. 0-5.0 umho/cm, read in control room, alarm setpoints.5 umho/cm and 2 umho/cm

• (b) effluent of demineralizers - three (one each demineralizer), 0-1.0 umho/cm, read in control room, alarm setpoint.1 umho/cm B.

Flow measuring instruments:

(1) Condensate system - one flow element and transmitter for each of the seven demineralizers 1668 202

9 (2} Reactor Water Cleanup System -one flow element, transmitter, indicator and low alarm switch on influent l'ine to the three demint.ralizers. This is satisfactory because, as noted above, only two of the three demineralizers are in service at any time.

V.

Regulatory Position 5 The conductivity meter (s) located at the inlet and outlet of the demineralizer(s) of the condensate treatment system and the reactor water cleanup system should be set to trigger alarms in the control room when, as a minimum, either of the following conductivity levels is reached (values of which should be detemined by the licensee);

a.

The level that indicates marginal perfomance of the demineralizer systems.

b.

The level that indicates noticeable breakthrough of one or more demineralizers.

The alam setpoints are noted above in Regulatory Position 4.

There-is sufficient instrumentation to cover the objectives of Sa. and 5b. above.

VI. Regulatory Position 6.

The chloride content in the reactor vessel water should be maintained as low as practical. The ionic equilibria of the reactor vessel water should be controlled to ensure a neutral pH. The licensee should establish limits for conductivity, pH, and chlorides in the reactor vessel water and should specify procedures to be used for their detemination. Acceptable reactor water chemistry limits are given in Table 1 of the appendix to this guide. If the limiting values of the conductivity, pH, or chloride content are exceeded, appropriate corrective actions as defined in the plant technical specifications should be taken.

A comparison of the guidance limits and the licensee's limits follows:

1668 203

Parameter R.G. 1.56 Limit Millstone 1 Limit (a) Conductivity (1) 2 umho/cm at steaming rates 2 umho/cm at steaming rates less than is of rated steam less than 100,000 pounds /

flow and during startup and hour, except that during hot standby. Time above 2 startup, conductivity shall umho/cm and.1 ppm chloride not exceed 10 umho/cm for not to exceed 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> for the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after the any single occurrence.

reactor is placed in the power operating condition (2) 1 umho/cm at steaming rates 2.0 umho/cm at steaming rates greater than 1% of rated equal to or greater than steam flow (maximum limit 100,000 pounds / hour, except established at 10 umho/cm, that maximum values of 10 which would require orderly umho/cm and.1 ppm chloride shutdown, but 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> above are allowed for reactor 1 umho/cm and.2 ppm chloride startups and for the first allowed per incident not to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after placing the exceed 2 weeks per year) reactor in the power operating condition (b) Chloride (1).1 ppm for startup/ hot stand-

.1 ppm for steaming rates less by and power operation with than 100,000 pounds / hour and steaming rates less than 1%

for startups of rated steam flow (2).2 ppm at steaming rates

.5 ppm at steaming rates equal greater than 1% of rated to or greater than 100,000 steam flow (maximum pounds / hour, except limited established at 0.5 ppm, to.1 ppm during startup and which would require orderly the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after shutdown, but 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> above placing the reactor in the

.2 ppm and 1 umho/cm allowed power operating condition s

per incident (c) pH limit: between 5.3 and 8.6 5.6 - 8.3 (as noted in the Appendix H report a meter is in operation at the Condensate Booster pump discharge header)

(d) Conductivity of the

.5 umho/cm at the inlet to 1.0 umho/cm (alarm)

Feedwater System the demineralizers (maxi-mum of 10, requiring orderly shutdown

.2 umho/cm at the individual

.1 umho/:m (alarm) demineralyzer outlet 1668 204

It is our belief that BWR pipe cracking has been proven to be such an extensive problem that the Position 6 limits of Regulatory Guide 1.56 should be conformed to where the licensee does not already have equivalent or lower limits.

Conclusion With the modification, of Position 6 above, the licensee will conform to.the intent of the regulatory guide. However, since the need for modification to the plant (including technical specification and procedural rodifications) will be judged only during the integrated assessment, the licensee need not take any action at this time.

It should also be noted that the staff considers that certain plant pro-cedures taken credit for in this analysis will require tighter control over their bcing changed than that allowed in 10 CFR 50.59. These pro-cedures are:

(1) That governing the sampling of the RWCU demineralyzer on service and subsequent shifting of flow if warranted (page 3); and (2) That governing the measurement of flow every four hours through each condensate demineralyzer on service and the daily calculation of unused capacity of each bed (page 4).

The staff is considering the means by which this tighter control should be implemented (FSAR changes, technical specification changes, etc.) and will include its determination in the integrated assessment.

1668 205 m