ML20064G275
| ML20064G275 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 03/09/1994 |
| From: | Shelton D CENTERIOR ENERGY |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 2205, NUDOCS 9403160095 | |
| Download: ML20064G275 (5) | |
Text
-
)
l l
CENTERIOR ENERGY 6200 Ook Tree Boulevard Mal Address.
Donald C. Shelton independerite OH P.0 Box 94601 Senior Vice President 216 447 3153 Clevetand, OH 4410L4061 Nuclear fcx 216-447 3123 Docket Number 50-346 License Number NPF-3 Serial Number 2205 P
March 9, 1994 United States Nuclear Regulatory Commission Document Control Desk Vashington, D. C.
20555
Subject:
Second 10-Year Interval Inservice Testing Program for Davis-Besse Nuclear Power Station Gentlemen:
By letters dated December 2, 1991 (Log Number 3643), April 23, 1993 (Log Number 3981) and August 18, 1993 (Log Number 4055) the NRC reviewed and approved the "Second Interval Pump and Valve Inservice Testing Program" for Davis-Besse Nuclear Power Station (DBNPS) and associated requests for relief.
Subsequent to the review and approval, an additional need for relief from the requirements of the American Society of Mechanical Engineers (ASME) Code has been identified.
Relief Request RP-7 is enclosed as Attachment 1.
This relief request vill allow the use of pump performance curves to determine acceptable pump performance in lieu of using fixed reference flovrates and differential pressures for the Component Cooling Vater (CCV) pumps at DBNPS. Relief Request RP-7 is similar to a relief request for the Service Vater Pumps at DBNPS (RP-4) which was previously approved in the December 2, 1991 letter.
In addition, during an inspection of the Inservice Testing Program in April 1993, the inspectors made an observation regarding pump testing methods used at DBNPS. The observation is documented in Inspection Report 50-346/93009 dated May 17, 1993 (Log Number 1-2851; and is summarized below.
It is the present practice at DBNPS to obtain multiple data points during a single pump test. This data is typically taken at five minute intervals with no adjustment of the equipment.
Reference values for pump performance characteristics are determined.in a like manner. The data points are then averaged to determine the l
l
[
1d0042
\\,
[c',
opeating corrpomes:
i creveland Dectoc mummatra 9403160095 940309 D'd U'S "
PDR ADOCK 05000346 P
Docket Number 50-346-
- License Number NPF-3 Serial Number 2205 Page 2-pump performance characteristics for that test.
Section XI of the ASME Code,-IUP-3500 permits obtaining multiple data points during a single pump test, however, the ASME Code does not address averaging-of the data.
This was acknowledged in an the ASME Code Committee response to ASME Code Inquiry submitted by Toledo Edison following the inspection.
Since the ASME Code does not prohibit averaging of multiple data points, relief is not being requested, nor is TE proposing an alternative test method to a Code-required method.
Accordingly, TE intends to continue this present. practice as it provides a more consistent and reliable indication of pump performance and degradation.
Should you have any questions regarding this submittal, please contact Mr. Villiam T. O'Connor, Manager - Regulatory Affairs, at (419) 249-2366.
Very truly yourr,
,,-~h
/hc/
4KP/ eld cc:
J. B. Martin, Regional Administrator, NRC Region III S. Stasek, DB-1 NRC Senior Resident Inspector R. J. Stransky, NRC Project Manager Utility Radiological Safety Board a
f p
f m
-..e..-.-- - ~.
e
..-.<r w
y y
- _. _ _. _. _ _ _=_-- _
Docket Number 50-346 License Number NPF-3 Serial Number 2205 Page 1 PUMP RELIEF REQUEST RP-7 SYSTEM:
Component Cooling Vater (CCV)
PUMPS:
P43-1, P43-2, P43-3 CLASS:
3 FUNCTION:
Provides cooling water to transfer heat from safety-related equipment to the Service Water System.
TEST REQUIREMENTS:
IVP-3100 requires that each measured test quantity in Table IVP-3100-1 be compared to the reference value of the same quantity.
Any deviations determined shall be compared to the limits given in Table IVP-3100-2, Allowable Ranges of Test Quantities.
BASIS FOR RELIEF:
The use of reference curves for pump evaluation in lieu of reference values for differential pressure, vibration, and flow are requested. The use of vibration velocity in lieu of vibration amplitude as a measured quantity is also requested. The difficulty of reproducing the same system flow resistances, the use of a more reliable measured quantity (i.e., vibration velocity) and man-rem exposure savings are the basis for this request.
During power operations, one component cooling train is aligned to service standby essential loads and is not in service.
One component cooling train serves non-essential loads and is in service.
During
~
shutdown conditions, both component cooling trains are in service. The system also contains a. third, spare train as a backup to either the essential or non-essential trains.
Flow through the pump aligned to service esser'ial loads remains essentially constant when the train is in service.
Fle through the pump aligned to service the non-essential loads cannot bu fixed because system resistances are continuously varying and flows to parallel loads are dependent on each.other.
Spent fuel.
cooling and boric acid evaporators have temperature control valves which vary demand on the CCW system according to heat load.. Component cooling water flow to the reactor coolant pump coolers varies, dependent on as left throttle positions on the supply lines for the four pumps.
Component cooling water flow to the control rod drive booster pumps passes through cleaning filters and flow will change dependent on filter loading. Thus, flow cannot be reliably throttled to a fixed reference value.
l l
sl
~.. ~, - -.
~..
Docket Number 50-346 License Number NPP-3 4
Serial Number 2205 Page 2 BASIS FOR RELIEF (Continued):
Presently, the quarterly pump test for each pump may be performed on either train, depending upon the plant conditions at the' time of the test.
Manual butterfly valves in each of the in-service trains are used to throttle flow during pump testing to achieve the same operating point for each test.
These valves do provide control, however, repeatability is poor, as butterfly valves are not designed to throttle flow.
In the essential train, these valves are located in an elevated radiation field and, dependent on plant conditions, a high radiation area.
Entering this radiation area for up to 12 tents per year to monitor and throttle flow causes unnecessary radiation exposure. Based upon present radiation levels, estimated exposure for performing 12 tests is approximately 0.4 man-rem per year.
ALTERNATE TESTING:
As discussed above in the Basis for Relief section, it is extremely difficult to return to a specific value of flow rate or differential pressure for testing of these pumps.
An alternative to using testing requirements of IVP-3100 is to base the acceptance criteria on pump reference curves.
Pump performance curves, giving reference values for vibration velocity and differential pressure as functions of flow between i
3000 gpm and 8000 gpm have been established. The flow ranges reflect normal and accident flow rate conditions. The differential pressure versus flow data was plotted and compared to established manufacturer's pump curves. The vibration velocity data vas obtained at the same flow data points and was also plotted to establish reference curves.
These curves vill serve as the basis for the alert and required action levels to ensure pump degradation is identified. The alert level and required action level parameters, as defined in the ASME code, 1989 Edition,Section XI, OMa-1988, Part 6, Ranges for Test Parameters, are superimposed as curves.
The methodology employed for establishing a reference curve is similar to that for performing a comprehensive test being proposed by the ASME Code Committee. To reduce the uncertainty associated with the pump curves and the adequacy of the acceptance critetia, special test gauges (2 5 % full scale accuracy or better) have been. installed to 0
obtain test data.
Flov indicator damping devices were also installed to limit flow gauge oscillations.
Measurements of vibration velocity and differential pressure at a minimum of six flow data points have been obtained for each pump, then plotted to compile the pump performance reference curves.
During pump testing, the flow vill be established within the-domain of the reference curve.
As plant conditions permit, approximately the same operating point vill be established. Vibration velocity and differential pressure vill be measured and recorded.
Pump performance vill be considered acceptable if parameter values fall within the regions bounded by the defined alert and action level curves, tather than a specific value.
I i
Docket Number 50-346 License Number NPP-3 Serial Number 2205 4
Page 3 ALTERNATE TESTING (Continued):
The rated speed of these centrifugal pumps is 1100 rpm. Table IVP-3100-2, Allowable Ranges of Test Quantities, requires one vibration data point to be taken and this point is a measurement of amplitude. The ASME Code, 1989 Edition,Section XI, OHa-1988, Part 6, Ranges for Test Parameters, requires five vibration data points to be taken and these points are measurements of velocity.
Use of vibration velocity measurements in lieu of vibration amplitude and additional data point evaluations vill ensure an earlier and more reliable prediction of pump degradation.
After any maintenance or repair that may affect the existing reference pump curves, and new reference pump curves vill be determined or the existing pump curves revalidated by inservice testing.
Using this alternative testing method, an acceptable level of quality and safety is provided.
l
}
e m
e v
~ _
'f**
CENTERDOR ENERGY 6200 Ock Tree Boulevo'd Mail Address Donald C. Shelton independence OH PO Box 94661 Semot Vice President 216-44L3153 C!emiand OH 441014661 Nuclear fou 216-44h3123 Docket Number 50-346 License Number NPF-3 3erial Number 2205 March 9, 1994 United States Nuclear Regulatory Commission Document Control Desk Vashington, D. C.
20555 Subj ec t :
Second 10-Year Interval Inservice Testing Program for Davis-Besse Nuclear Power Station Gentlement By letters dated December 2, 1991 (Log Number 3643), April 23, 1993 (Log Number 3981) and August 18, 1993 (Log Number 4055) the.NRC reviewed and approved the "Second Interval Pump and Valve Inservice Testing Program" for Davis-Besse Nuclear Power Station (DBNPS) and associated requests for relief.
Subsequent to the review and approval, an additional need for relief from the requirements of the American Society of Mechanical Engineers (ASME) Code has been identified.
Relief Request RP-7 is enclosed as Attachment 1.
This relief request vill allow the use of pump performance curves to determine acceptable pump performance in lieu of using fixed reference flovrates and differential pressures for the Component Cooling Fater (CCV) pumps at DBNPS.
Relief Request RP-7 is similar to a relief request for the Service Vater Pumps at DBNPS (RP-4) which was previously approved in the December 2, 1991 letter.
In addition, during an inspection of the Inservice Testing Program in April 1993, the inspectors made an observation regarding pump testing methods used at DBNPS. The observation is documented in i
Inspection Report 50-346/93009 dated May 17, 1993 (Log Number 1-2851) and is summarized below.
It is the present practice at DBNPS to obtain multiple data points during a single pump test. This data is typically taken at five minute intervals with no adjustment of the equipment. Reference i
values for pump performance characteristics are determined in a like manner. The data points are then averaged to determine the operm,ng ccmpanes C:evetand fiectoc 1%mmating hfL [dhk bedo Edson
Docket Number 50-346 License Number NPF-3 Serial Number 2205 Page 2 pump performance characteristics for that test.
Section XI of the ASME Code, IVP-3500 permits obtaining multiple data points during a single pump test, however, the ASME Code does not address averaging of the data. This was acknowledged in an the ASME Code Committee response to ASME Code Inquiry submitted by Toledo Edison following the inspection.
Since the ASHE Code does not prohibit averaging of multiple data points, relief is not being requested, nor is TE proposing an alternative test method to a Code-required method.
Accordingly, TE intends to continue this present practice as it provides a more consistent and reliable indication of pump performance and degradation.
Should you have any questions regarding this submittal, please contact Mr. Villiam T. O'Connor, Manager - Regulatory Affairs, at (419) 249-2366.
Very truly your
~_ -
h KP/ eld cc:
J. B. Hartin, Regional Administrator, NRC Region III S. Stasek, DB-1 NRC Senior Resident Inspector R. J. Stransky, NRC Project Manager Utility Radiological Safety Board
'r-*
'-W-4'=
m
.+
m--
q--
y-y-
.e; m
q y
n,...7 g
re.
'y,
~
~. -
Docket Number 50-346 License Number NPF-3 Serial Number 2205 Page 1 PUMP RELIEF REQUEST RP-7 SYSTEM:
Component Cooling Vater (CCV)
PUMPS:
P43-1, P43-2, P43-3 CLASS:
3 FUNCTION:
Provides cooling water to transfer heat from safety-related equipment to the Service Vater System.
TEST REQUIREMENTS:
IUP-3100 requires that each measured test quantity in Table IVP-3100-1 be compared to the reference value of the same quantity.
Any deviations determined shall be compared to the limits given in Table IVP-3100-2, Allovable Ranges of Test Quantities.
BASIS FOR RELIEF:
The use of reference curves for pump evaluation in lieu of reference values for differential pressure, vibration, and flow are requested. The use of vibration velocity in lieu of vibration amplitude as a measured quantity is also requested. The difficulty of reproducing the same system flow resistances, the use of a more reliable measured quantity (i.e., vibration velocity) and man-rem exposure savings are the basis for this request.
During power operations, one component cooling train is aligned to service standby essential loads and is not in service. 'One component cooling train serves non-essential loads and is in service.
During shutdown conditions, both component cooling trains are in service. The system also contains a third, spare train as a backup to either the l
essential or non-essential trains.
Flow through the pump aligned to service essential loads remains essentially constant when the train is in service.
Flow through the pump aligned to service the non-essential loads cannot be fixed because system resistances are continuously varying i
and flovs to parallel loads are dependent on each other.
Spent fuel cooling and boric acid evaporators have temperature control valves which vary demand on the CCV system according to heat load.
Component cooling water flow to the reactor coolant pump coolers varies, dependent on as left throttle positions on the supply lines for the four pumps.
Component cooling water flow to the control rod drive booster pumps passes through cleaning filters and flow vill change dependent on filter i
loading.
Thus, flow cannot be reliably throttled to a fixed reference
- value, i
-m-e-a p
r 3
e--
dw:
p-
Docket Number 50-346 License Number NPF-3 Serial Number 2205 Page 2 BASIS FOR RELIEF (Continued):
Presently, the quarterly pump test for each pump may be performed on either train, depending upon the plant conditions at the time of the test.
Manual butterfly valves in each of the in-service trains are used to throttle flow during pump testing to achieve the same operating point for each test. These valves do provide control, however, repeatability is poor, as butterfly valves are not designed to throttle flow.
In the essential train, these valves cre located in an elevated radiation field and, dependent on plant conditions, a high radiation area.
Fntering this radiation area for up to 12 tests per year to monitor and throttle flow causes unnecessary radiation exposure.
Based upon present radiation levels, estimated exposure for performing 12 tests is approximately 0.4 man-rem per year.
~
ALTERNATE TESTING:
As discussed above in the Basis for Relief section, it is extremely difficult to return to a specific value of flow rate or differential pressure for testing of these pumps. An alternative to using testing requirements of IVP-3100 is to base the acceptance criteria on pump reference curves.
Pump performance curves, giving reference values for vibration velocity and differential pressure as functions of flow between 3000 gpm and 8000 gpm have been established. The flow ranges reflect normal and accident flow rate conditions.
The differential pressure versus flow data was plotted and compared to established manufacturer's pump curvas. The vibration velocity data was obtained at the same flov data points and was also plotted to establish reference curves. These curves vill serve as the basis for the alert and required action levels to ensure pump degradation is identified. The alert level and required action level parameters, as defined in the ASME code, 1989 Edition,Section XI, OMa-1988, Part 6, Ranges for Test Parameters, are superimposed as curves.
The methodology employed for establishing a reference curve is similar to that for performing a comprehensive test being proposed by the ASME Code Committee. To reduce the uncertainty associated with the pump curves and the adequacy of the acceptance criteria, special test gauges (2 5 % full scale accuracy or better) have been installed to 0
obtain test data.
Flov indicator damping devices were also installed to 11mic flov gauge oscillations.
Measurements of vibration velocity and differential pressure at a minimum of six flow data points have been obtained for each pump, then plotted to compile the pump performance reference curves.
During pump testing, the flow will be established within the domain of the reference curve. As plant conditions permit, approximately the same operating point vill be established. Vibration velocity and differential pressure vill be measured and recorded.
Pump performance vill be considered acceptable if parameter values fall within the regions bounded by the defined alert and action level curves, rather than a specific value.
Docket Number 50-346 License Number NPF-3 Serial Number 2205
'P1ge 3 ALTERNATE TESTING (Continued):
The rated speed of these centrifugal pumps is 1100 rpm. Table IWP-3100-2, Allovable Ranges of Test Quantities, requires one vibration data point to be taken and this point is a measurement of amplitude.
The ASHE Code, 1989 Edition,Section XI, OMa-1988, Part 6, Ranges for Test Parameters, requires tive vibration data points to be taken and these points are measurements of velocity.
Use of vibration velocity measurements in lieu of vibration amplitude and additional data point evaluations vill ensure an earlier and more reliable prediction of pump degradation.
After any maintenance or repair that may affect the existing reference pump curves, and new reference pump curves vill be determined or the existing pump curves revalidated by inservice testing.
Using this alternative testing method, an acceptable level of quality and safety is provided.
_.