ML20217K394

From kanterella
Jump to navigation Jump to search
SER Approving Relief Request for Pump Vibration Monitoring, Brunswick Steam Electric Plant,Units 1 & 2
ML20217K394
Person / Time
Site: Brunswick  
Issue date: 04/24/1998
From:
NRC (Affiliation Not Assigned)
To:
Shared Package
ML20217K387 List:
References
NUDOCS 9805010193
Download: ML20217K394 (7)


Text

_

a uru p

k UNITED STATES g

j NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. was -1 l

SAFETY EVALUATION BY THE OFFICE OF NUCI FAR REACTOR REGULATION REllEF REQUEST FOR PUMP VlBRATION MONITORING BRUNSWICK STEAM FI FCTRIC PLANT. UNIT NOS.1 AND 2 CAROLINA POWER & LIGHT COMPANY DOCKET NOS. 50-325 AND 50-324

1.0 INTRODUCTION

The Code of Federal Regulations,10 CFR 50.55a, requires that inservice testing (IST) of certain American Society of Mechanical Engineers (ASME) Code Class 1,2, and 3 pumps and valves be performed in accordance with Section XI of the ASME Boiler and Pressure Vessel Code (the Code) and applicable addenda, except where attematives have been authorized or relief has been requested by the licensee and granted by the Commission pursuant to Sections (a)(3)(i), (a)(3)(ii), or (f)(6)(i) of 10 CFR 50.55a. In proposing attematives or requesting relief, the licensee must demonstrate that: (1) the proposed attematives provide an acceptable level of quality and safety, (2) compliance would result in hardship or unusual difficulty without a compensating increase in the level of quality and safety, or (3) conformance is impractical for its facility. Section 50.55a authorizes the Commission to approve attematives and to grant relief from ASME Code requirements upon making the necessary findings. Generic Letter 89-04

" Guidance on Developing Acceptable Inservice Testing Programs," issued April 3,1989, and its April 4,1995, Supplement 1 give guidance related to the development and implementation of IST programs. Also see NUREG-1482, " Guidelines for Inservice Testing at Nuclear Power Plants," and NUREG/CR-6396, " Examples, Clarifications, and Guidance on Preparing Requests for Relief from Pump and Valve Inservice Testing Requirements."

The 1989 Edition of the ASME Code is the latest edition incorporated by reference in Paragraph (b) of Section 50.55a. Subsection lWP of the 1989 Edition, which gives the i

requirements for IST of pumps, references Part 6 of the American National Standards institute /ASME Operations and Maintenance Standards (OM-6) as the rules for IST of pumps.

Standard OM-6 replaces specific requirements in previous editions of ASME Code Section XI, Subset, tion IWP.

2.0 BACKGROUND

in a [[letter::BSEP-97-0504, Requests Approval for Bsep,Units 1 & 2,to Use Alternative to Vibration Testing Requirements of ASME BPV Code,Section XI & Asme/Ansi Operations & Maint Stds,Part 6, IST of Pumps in LWR Power Plants, IAW 10CFR50.55a(a)(3)(ii)|letter dated December 8,1997]], Carolina Power & Light Company (CP&L), licensee for the Brunswick Steam Electric Plant (BSEP), Units 1 and 2, submitted a relief request concoming the vibration monitoring of standby liquid control sysicm (SLC) and service water system (SWS) pumps. On February 27,1998, the NRC issued a request for additional information (RAI) for this relief request. The licensee responded to this request in a letter dated April 2,1998. The NRC staff's findings with respect to authorizing attematives as part of the licensee's IST 9805010193 990424 DR ADOCK 0500 3 4 i

.__._-.s...,,-.

' I

-2 program are contained in this safety evaluation (SE). The licensee's IST program was developed in accordance with the requirements of the 1980 Edition through the Winter 1981 Addenda,Section XI, of the ASME Code, in the NRC SE dated January 4,1990, the licensee's proposal to implement OM-6 for pump vibration monitoring was approved for the second 10-year interval IST program, which will end on May 10,1998.

3.0 RELIEF REQUEST: PUMP VIBRATION MONITORING Relief is requested from OM-6, paragraph 4.6.1.6, which requires that the frequency response range of the vibration measuring transducers and their readout system shall be from one-third minimum pump shaft rotational speed to at least 1000 Hz. This regt,est pertains to the following pump:,: SLC pumps 1(2)-SLC-P-1 A and -1B; and SWS pumps 1(2)-SW-C-P-1 A, 1(2)-SW-C-P-18,1(2)-SW-C-P-1C,1(2)-SW-N-P-1 A, and 1(2)-SW-N-P-1 B.

3.1 Basis for Relief The licensee provided the following basis for the request:

Standbv Liouid Control System Pumos The nominal shaft rotational speed of the Standby Liquid Control system pumps is 385 RPM, which is equivalent to approximately 6.4 Hertz. Based on this frequency and the methodology delineated in OM Part 6, Paragraph 4.6.1.6, the required frequency response range of instruments used for measuring pump vibration is 2.14 to 1000 Hertz.

The instruments currently in use at BSEP (lRD Model 890 with #970 transducer) have a frequency response ranging from 5 to 1000 Hertz.

These pumps are of a simplified reciprocating (piston) positive displacement design with rolling element bearings (Model Number TD-60, manufactured by Union Pump Corporation). The requirement to measure vibration using instruments with a response range to % shaft speed stems from the need to detect oil whip or oil whirl associated with pump joumal bearings. In the case of these pumps, there are no joumal bearings to create these phenomena.

Compliance with the Code requirement would require procurement and calibration of instruments to cover this range to the lower extreme (i.e.,2.14 Hertz). Procurement and calibration of this equipment will be unusually difficult without a compensating increase in pump performance or plant safety. As noted in NUREG-1482, " Guidelines for Inservice Testing at Nuclear Power Plents," Appendix G, Comment 5.4-3, instruments that can read 2 Herti may be available, but th!s frequency is less than what is traceable to the National Bureau of Standards for calibration. Even though the current equipment (i.e., IRD Model 890, with an accelerometer) is not calibrated to monitor the low frequency required by OM Part 6, indication of pump degradation will not be masked by the existing instrumentation not being calibrated to collect data at sub-synchronous frequencies (i.e., less than 5 Hertz).

l l

I 1 1

Satisfying the frequency response range criteria will not provide meaningful information to assess the condition of these pumps. The significant modes of vibration with respect to monitoring these pumps are as follows:

1-Times Crankshaft Soeed - An increase in vibration at this frequency may be an indication of rubbing between a single crankshaft cheek and rod end, or cavitation at a single valve, hydraulic instability, or loose machine foot.

2-Times Crankshaft Soeed - An increase in vibration at this frequency may be an indication of looseness at a single rod bearing or crosshead pin, a loose valve seat in the fluid cylinder, a loose plunger /crosshead stub connection, or coupling -

misalignment.

Other Multioles of Shaft Soeed - An increase in vibration at other frequencies may be indications of cavitation at several valves, looseness at multiple locations, bea:ing degradation, rubbing, impacting, or binding.

Based on the foregoing discussion, using current vibration measuring instruments with at least a frequency response range of 5 to 1000 hertz will provide adequate information to evaluate pump condition and ensure continued reliability with respect to the pumps' function.

Service Water System Pumos The nominal shaft rotational speed of these pumps is 885 RPM which is equivalent to approximately 14.75 Hz. Based on this frequency and the methodology delineated in OM Part 6, Paragraph 4.6.1.6, the required frequency response range of instruments used for measuring pump vibration is 4.91 to 1000 Hertz. The instruments currently in use at BSEP (IRD Model 890 with #970 transducer) have a frequency response range of 5 to 1000 Hertz.

The requirement to measure vibration with instruments with response to % shaft speed stems from the need to detect oil whip or oil whirl associated with oil-lubricated joumal bearings. Specifically, vibration peaks for an oil whip typically occur at 40 percent to 48 percent of shaft speed. Since the existing instruments can measure vibration below 40 percent to 48 percent of shaft speed for these pumps (i.e., 5.9 to 7.1 Hertz), their use is consistent with the intent of the Code. Thus, using current vibration measuring instruments with at least a frequency response range of 5 to 1000 Hertz will provide adequate information to evaluate pump condition and ensure continued reliability with respect to the pumps' function.

The NRC 's February 27,1998, RAI, and the licensee's April 2,1998, responses are as follows:

NRC Reauest 1:

The Code requires the frequency response range of vibration measurement instruments

)

to extend as low as one-third minimum pump shaft rotational speed. For the positive i

g 4

displacement standby liquid control system pumps in question, one-third of the minimum pump rotational speed is approximately 128 rpm, which corresponds to 2.1 Heitz.

However, the frequency response range of instrumentation used for these pumps extends only as low as 5 Hertz, which falls short of the requirements by approximately 2.9 Hertz. In order to support the statement in the basis that this instrumentation will.

f provide adequate information to evaluate pump condition, please provide j

documentation, including manufacturer's recommendation, that demonstrates that these pumps are not susceptible to rolling element bearings degradation mechanisms that result in increased vibration levels seen at frequencies below the pump rotational speed frequency, such as impacts from rolling element bearing cage defects.

Also, instruments with a response range that extends as low as 3 Hertz a a commonly used and available. Address the practicability of procuring such an instrument for use on these pumps.

CP&L Response i

The Standby Liquid Control system pumps have rolling element bearings. These j

bearings are susceptible to degradation that can result in increased vibration levels.

These degradation mechanisms are detectable through: cage frequencies (Fundamental Train Frequencies (FTF)), Ball Spin Frequencies (BSF), Outer Race Defect Frequencies (i.e., Ball Pass Frequency Outer (BPFO)), and Inner Race Defect Frequencies (i.e., Ball Pass Frequency Inner (BPF1)). Although rolling element bearing problems are typically identified in the high frequency area of a spectral plot, the FTF is sometimes found in subsynchronous area ranging from.33 x RPM up to.45 x RPM. However, the cage -

frequency will not normally appear at its fundamental frequency. Instead, it will most often show up as a side band of either ball spin (i.e., BSF) or one of the other bearing race frequencies with a side band difference frequency equal to the FTF. In some cases, vibration data has indicated ball spin frequency on a rolling element bearing, s.nd the result was a broken cage. The bearings used for the inboard and outboard crank shaft on the Standby Liquid Control system pumps are Timken 772. They have specific fault frequencies at the following:

FTF: 165 RPM (2.7 Hertz)

BSF: 1,509 RPM (25.1 Hertz)

BPFO: 3,525 RPM (58.7 Hertz)

BPFl: 4,531 RPM (75.5 Hertz)

The FTF side bands would show at i 165 RPM around the other three fault frequencies.

Therefore, the FTF of a rolling element bearing can be detected by vibration equipment calibrated to 5 Hertz.

NRC Reauest 2:

OM-6 allows either displacement or velocity in measuring vibration. NUREG/CP-0111,

" Proceedings of the Symposium on Inservice Testing of Pumps and Valves," contains a paper presented by J. Howard Maxwell, " Measurements of Vibrational Parameters for

. Pump Testing," which states that displacement is the better parameter in determining vibration severity of machines which operate below 600 rpm. Please address the applicability of measuring displacement for the standby liquid control system pumps in this relief request.

CP&L Response-OM-6 allows either displacement or velocity in measuring vibration. The current instrumentation used at BSEP is not capable of providing accurate data when l

measuring displacement versus velocity. Accelerometers are being used to record data.

When the collector records a displacement reading, it is performing a double integration.

When a velocity reading is taken, it is performing a single integration. Converting accelerometer output to velocity or displacement requires integration. Integration enhances the low frequencies in the signal. Tribo effect and circuit component noise are greatly magnified by double integration when converting to displacerr.ent.

As noted above, CP&L is in the process of purchasing vibration data collection equipment capable of measuring the required frequency response range for the Standby Liquid Control system pumps. The purchase of the new vibration equipment will allow the use of Slow Speed Technology (SST) developed by Dr. Jim Robertson. The SST function allows for accurate vibration readings at extremely low machine speeds. Due to the effects of digital integration, it is difficult to get accurate velocity or displacement readings below 2 Hertz using an accelerometer. The SST function is an algorithm that corrects for the attenuation of FTF amplitudes that occur below 2 Hertz. Due to the single integration of acceleration to velocity, the velocity reading obtained using an accelerometer is more accurate than a displacement reading.

3.2 Prooosed Alternative Testina The licensee proposes:

Vibration levels of the Standby Liquid Control and Service Water system pumps will be measured in accordance with the applicable portions of OM Part 6, Paragraph 4.6, with the exception of the lower frequency response limit for the vibration measuring equipment (Paragraph 4.6.1.6). In this case, the lower response limit of the vibration measuring equipment will be 5 Hertz or less, based on the capability of the existing plant vibration measuring equipment, and the upper frequency response limit will be a minimum of 1000 Hertz.

In response to NRC 's February 27,1998, RAI, the licensee subsequently proposed the following in an April 2,1998, letter.

Subsequent to submittal of Carolina Power & Light (CP&L) Company's request dated December 8,1997, and as part of an enhancement of the Brunswick Steam Electric Plant's (BSIP) predictive maintenance program, CP&L initiated procurement of pump vibration data collection equipment. The new equipment will be capable of frequency measurements as low as 1.5 Hertz. The new equipment for monitoring the Standby l

. Liquid Control system pumps is scheduled to be installed, with test procedures revised

)

and personnel trained, by October 1,1998. Therefore, approval of attemative vibration monitoring for the Standby Liquid Control system pumps is only requested until October 1,1998.

i l

3.3 Evaluation The Code requires that the vibration instrumentation frequency response range used in i

quarterly testing be from one-third pump rotational speed to 1000 Hz. One-third pump rotational speeds are 128 rpm (2.1Hz) for the SLC system pumps and 295 rpm (4.9 Hz) for the SWS pumps. The plant has an instrument with a range of 5 to 1,000 Hz. This instrument does not satisfy the Code lower limit of the frequency response range for either the SLC system pumps or the SWS pumps at Brunswick Steam Electric Plant, Units 1 and 2.

Standbv Liould Control Svstem Pumos:

These are positive displacement pumps with rolling element bearings. Degradation i

mechanisms associated with these bearings include impacts from cage defects that may result in increased vibration levels seen at frequencies below the pump rotational speed. The licensee states that rolling element bearing problems, however, are typically identified in the high frequency area of a spectral plot and will most often show up as a side band of either ball spin or one of the other bearing race frequencies with a side band difference frequency equal to the FTF.

The licensee has proposed to use the existing instruments until October 1,1998. This interim period will allow the licensee time to incorporate a new instrument that meets the ASME Code requirements into the IST program. Immediate compliance would be an undue burden since the instruments are difficult to procure and calibrate. The proposed alternative testing should i

provide a reascnable assurance of operational readiness during the interim.

I Service Water System Pumos:

For the service water pumps, the required frequency response range of instruments used for measuring vibration is 4.91 to 1000 Hertz. The instruments currently in use at the plant have a frequency response range of 5 to 1000 Hertz. Therefore, the deviation from the Code-required lower limit is less than 2 percent.

The licensee has proposed to use the instruments currently available at Brunswick. Requiring the licensee to procure new instrumentation to meet the Code requirements for the service water pumps would be a hardship without a compensating increase in the level of quality and safety because the deviation from the Code requirement is small (less than 2 percent) and falls within the i 5 percent accuracy allowed by OM-6 for vibration instruments. The proposed afternative testing should provide a reasonable assurance of operational readiness for the i

pumps in question.

.. 3.4 Conclusion Based on the determination that compliance with the specified requirements results in a hardship without a compensating increase in the level of quality and safety, the proposed alternatives to the Code vibration instrument frequency response range requirements for the SLC pumps are authorized pursuant to 10 CFR 50.55a(a)(3)(ii) until October 1,1998, to allow time for the licensee to incorporate a new instrument that meets the OM-6 requirements into the l

IST program. The proposed alternative to the Code vibration instrument frequency response range requirements for the SWS pumps is authorized pursuant to 10 CFR 50.55a(a)(3)(ii),

based on the determination that compliance with the specified requirements results in a hardship without a compensating increase in the level of quality and safety.

Principal Contributor K. Dempsey, DE/EMEB Date: April 24,,1998 l

{

I l

1 h

..