ML061980186: Difference between revisions
StriderTol (talk | contribs) (Created page by program invented by StriderTol) |
StriderTol (talk | contribs) (StriderTol Bot change) |
||
(2 intermediate revisions by the same user not shown) | |||
Line 17: | Line 17: | ||
=Text= | =Text= | ||
{{#Wiki_filter:p.-~,y~O '71 Ril~-5A4 3q | {{#Wiki_filter:p.-~,y~O '71 I i Ril~-5A4 3q Jolrnson"PUiMP Company Engineering Evaluation Regarding Short Term Dty-run of-Rubber Column Bearings Cooper Nuclear Station, Nebraska Public Power District fByron Jackson 28KXL I-STG! Chattanooga Sei'ice Center Report Serial No.: 3P04-20)Rev. 02 | ||
°o....%°..v;-. | ......- o..°.;...;:..o.... °o....%°..v;-. | ||
bigireei | bigireei Dept.,B1rooks ire, D | ||
RciHO Cf LnE.o.:O2. | |||
Prpae 3W\Rhd1CugLP | Prpae 3W\Rhd1CugLP | ||
~itai26~4 Johnston Pump Company Engineering Evaluation Regarding Short Term Dry-run of Rubber Column Bearings Cooper Nuclear Station, Nebraska Public Power District Byron Jackson 28KXL I-STG, Chattanooga Service.Center Report Serial No.: JP04-20 Rev. 02 I Originating Department: | ~itai26~4 | ||
Johnston Pump Company Engineering Evaluation Regarding Short Term Dry-run of Rubber Column Bearings Cooper Nuclear Station, Nebraska Public Power District Byron Jackson 28KXL I-STG, Chattanooga Service.Center Report Serial No.: JP04-20 Rev. 02 I | |||
Originating Department: | |||
Engineering Dept., Brookshire. | Engineering Dept., Brookshire. | ||
Revision No.: 02 Prepared By: Michael Cuga], PE Initial: Date: 07/] 6/04 Checked By: Lanka Pannila, P.E.Initial: | Revision No.: 02 Prepared By: Michael Cuga], PE Initial: | ||
{ ~ I i Table of Contents 1. Introduction | Date: 07/] 6/04 Checked By: Lanka Pannila, P.E. | ||
Initial: | |||
{~ I i | |||
Table of Contents | |||
: 1. Introduction 1 | |||
: 2. Discussion | |||
: 4. Summary/ Conclusions | |||
JOHNSTON PUMPS 800 Koomey Road Tel: (800) 926-6688 Brookshire, IX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number Customer NPPD - Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-STG Project Manager Jerry Harrelson | |||
: 1. Introduction T"he.purpose of this report is to provide an engineering evaluation of the following question: | |||
Given conditions that existed at Cooper Nuclear Station(l), will the Service Wat6r pumps(2) continue to function 48 hours after Gland Water low is reduced to zero flow for 90 minutes and then restored to normal? | |||
(1) During Janiary 21, 2004 - February 11,.2004: | |||
River water level 875.5 MSL - 877.5 MSL. | |||
Service water temperature 545°F. | |||
Pump running at capacity (5500 gpm). | |||
Average discharge pressure 50 psig average. | |||
Gland water flow -678 gpm to the enclosing tube (16-24 psi). | |||
Vibration (1ST) normal.. ....... | |||
-(2) Byron Jackson 28KXL 1-Stage VCT pumps with 180 rpm, 300 lip Motors.. | |||
The following material conditions existed as of February 2003: 1). New pump assembly, packing, coupling, shafts,.cutlass bearings, and impeller. 2). Rebuilt outer column, discharge nozzle, registers, and spider bushing supports. | |||
: 2. Discussion | : 2. Discussion | ||
===Background=== | ===Background=== | ||
The function of the Gland Water flow is to provide clean water lubrication lo the pump's stuffing box area, and column bearings during the pump operation. | The function of the Gland Water flow is to provide clean water lubrication lo the pump's stuffing box area, and column bearings during the pump operation. | ||
In vertical pumps, the column bearings provide 'bumper" bearing support. Due to negligible pressure differences-across the bearings, they do not behave as typical hydrostatic bearings. | In vertical pumps, the column bearings provide 'bumper" bearing support. Due to negligible pressure differences | ||
However, a thin" fluid film between the bearing and the shaft results in some bearing damping and fluid stiffness..... | -across the bearings, they do not behave as typical hydrostatic bearings. However, a thin"fluid film between the bearing and the shaft results in some bearing damping and fluid stiffness..... | ||
For the'subject pump, at the given river level, if the gland water flow is stopped, the top 7 bearings (including the stuffing box bearing) would not receive the lubrication. | For the'subject pump, at the given river level, if the gland water flow is stopped, the top 7 bearings (including the stuffing box bearing) would not receive the lubrication. The bottom 5 bearings (3 bottom column bearings and 2 bowl bearings) 'wouldbe submerged in the *umping fluid. | ||
The bottom 5 bearings (3 bottom column bearings and 2 bowl bearings) | In order to provide the answer to the posed question, we underook the following steps: | ||
' | Prepared By Checked By , Report No. Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16)04* 02 08105104 MC J | ||
02 08105104 MC J JOHNSTON PUMPS 800 Koomey Road Tel: (800) 926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number Customer NPPD -Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-STG I Project Manager Jerry Harrelson] | |||
The packing box bearing will heat up due to contact from the shaft. The effect of this could depend on the amount and force of the contact with the rotating shaft. If the shaft rotates without much wobble, it is conceivable that damage to the bearing or shaft can be minimal, Ifthe shaft contacts the stuffing box bearing constantly, the packing gland will get over-heated and dry, and the bearing clearance will increase.A typical vertical shaft is in tension and, in theory, it will rotale freely without any contact if the alignment and register fits are true and no manufacturing and machinin2 tolerances exist. In real operation, hov%,,.ver,.ThejriduaL unbalance of theorotor, Some hydraulic unbalance, allowable manufacturing tolerances, and the natural frequency of the rotor could cause the rotor to wobble and contact the column bearings.,If lubrication to the bearings is not present, we can foresee local heating of the rubber bearings and wveakening of the rubber material. | JOHNSTON PUMPS 800 Koomey Road Tel: (800) 926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number Customer NPPD - Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-STG I Project Manager Jerry Harrelson | ||
The constant contact with the shaft could result in loss of the bearing material. | ] Technical assessment by the Johnston Pump engineering staff. | ||
The severity of the mhaterial loss would highly depend on the nature and amount of contact. However, wye do believe that in general the bearing clearances will increase significantly due to the contact of the shaft..... hbearing clearances increase, the pump can experience higher vibration of the pum shaft and possibly the ump stfuctures m general. The severity and effect o e vi ration are difficult to predict, but a catastrop ic pump failure within 9U minutes of operation without the gland water supply and subsequent 48 hours of operation with the gland water supply is not likely to happen due to the vibration from increased bearing clearances. | : 2. General review and study of the field repair history of similar cases. | ||
There is a remote chance of a pump shaft seizure when it is operated without lubrication against rubber bearings.However, the probability of a seizure is very low if the pump does not have to stop and restart.2.2 Field repair history We were not able to locate any written repair rep6rt that dealt with the exact nature of this case. No record was'.found w',here the exact duration of the dry-running was documented. | : 3. Rotor dynamics analvsis. | ||
Our assessment was based on verbal communications with various Johnston Pump service centers. In most cases, operating pumps withoutlubrication to the rubber | 2.1 Technical assessment After gland flow is stopped, the packing ]hebox area will not receive proper lubrication and cooling flow, and the top 6 column bearings will not receive lubrication. The packing box bearing will heat up due to contact from the shaft. The effect of this could depend on the amount and force of the contact with the rotating shaft. If the shaft rotates without much wobble, it is conceivable that damage to the bearing or shaft can be minimal, Ifthe shaft contacts the stuffing box bearing constantly, the packing gland will get over-heated and dry, and the bearing clearance will increase. | ||
In their experience, when the gland water supply is stopped, the rubber bearings get over-heated, and in some cases they have seen smoke coming out of the column. In most cases, the pump will begin to vibrate, and eventually it will be pulled for a repair.. There were some cases where one or two bearings w,.ere completely damaged and pumps operated for over 48 hours. The general opinion of the service technicians is that the pump could have survive the 90 minutes of dry-operation though the bearings may have gotten severely damaged, and operation of the pump for the subs.equent 48 hours with re-introduction of the gland water supply would have been possible.2.3 Rotor dynamics analysis We believe that the loss of gland water supply will cause loss of bearing damping effect (by losing the fluid film between the bearing and shaft) and could open up the running clearances (which will change the bearing stiffness). | A typical vertical shaft is in tension and, in theory, it will rotale freely without any contact if the alignment and register fits are true and no manufacturing and machinin2 tolerances exist. In real operation, hov%,,.ver,.ThejriduaL unbalance of theorotor, Some hydraulic unbalance, allowable manufacturing tolerances, and the natural frequency of the rotor could cause the rotor to wobble and contact the column bearings.,If lubrication to the bearings is not present, we can foresee local heating of the rubber bearings and wveakening of the rubber material. The constant contact with the shaft could result in loss of the bearing material. The severity of the mhaterial loss would highly depend on the nature and amount of contact. However, wye do believe that in general the bearing clearances will increase significantly due to the contact of the shaft. | ||
In order to study the rotor dynamics behavior in more detail, we conducted a lateral rotor dynamics analysis (report JP04-18).Due to time constraints and the unavailability of some pertinenilgeometrv information (i.e. mnotor rotor details),. | .... hbearing clearances increase, the pump can experience higher vibration of the pum shaft and possibly the ump stfuctures m general. The severity and effect o e vi ration are difficult to predict, but a catastrop ic pump failure within 9U minutes of operation without the gland water supply and subsequent 48 hours of operation with the gland water supply is not likely to happen due to the vibration from increased bearing clearances. | ||
we had to use this study diligently. | There is a remote chance of a pump shaft seizure when it is operated without lubrication against rubber bearings. | ||
Our conclusions from the analysis are as follows: 0 It is quite typical for. a vertical, pump rotor system to have a number of critical speeds that are close to the excitation frequency. | However, the probability of a seizure is very low if the pump does not have to stop and restart. | ||
Such a case was proven to be true in this analysis. | 2.2 Field repair history We were not able to locate any written repair rep6rt that dealt with the exact nature of this case. No record was'. | ||
In this study, we focused on the modal shapes of the rotor in order to study the effect of the shaft and bearing contact.0 The baseline analysis (pump in normnal operating condition with gland water supply) indicated that there is indeed a natural frequency mode within 4.5% of the running frequency. | found w',here the exact duration of the dry-running was documented. Our assessment was based on verbal communications with various Johnston Pump service centers. In most cases, operating pumps withoutlubrication to the rubber fa-lure, bearings the shaft was seized, butinthe resulted severe damage duration to ihe of the operation bearings columnseemed and shalts. | ||
The mode shape, however, indicated that the shaft and the inner column deflect in the same phase, providing a non-contacting mode (refer to report JP04-] 8)." The worst case analysis (pump without gland w, ater supply and assuming all 7 top'bearings are lost and do not provide any bearing support) indicated that a higher mode natural frequency is near the running speed (within 6 %), and that the deflection mode of the shaft and inner column are not in the same phase. This would indicate a contact between the shaft and bearing, and that severe damages to the bearings can be expected." Based on the mode shape analysis, we believe that the worst case scenario would not happen during the pump operation. | to have In somecases been longer ofpump than a day or so. In ih-es'e cases, tme gland water supply was not restored. | ||
The mode shape indicated ihat two bearings may have come in severe contact with the shaft, but not all dry-running bearings would have been damaged.Our conclusion based on the rotor analysis is that when the pump is run dry some shaft contact with the bearings will occur. Based on the modal shape in the worse case scenario, we believe that contact will be made on some of the bearings, but loss of all column bearing support is not likely to happen.Prepared By Checked By Report No. Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16/04 .02 08/05/04 ,- MC 3 JOHNSTON PUMPS 800 Koorney Road Tel: (800) 926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number I Customer NPPD -Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-SIG Project Manager Jerry Harrelson Based 0i the modal shape analysis, the inner column mode shape was detected near the running speed. If the inner column fails, the result could be a re-supply of the lubrication to the column bearings as the discharge pressure pushes the fluid into the column. This would become somewhat beneficial in this case (depending on the nature of the failure), as it would introduce the lubrication and damping back to the column bearings.3.0 Summary and conclusions It is general knowledge and recommended practice never to operate a vertical pump withfrubber bearings without proper lubrication*. | Prepared By Checked By Report No. " Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16/04 02 08/05/04 MC 2 | ||
We wish to clearly state that the evaluations and statements of this report regarding this issue does not change such views. We also wish to state that the statements and summary of this report are intended only to share our views on this particular operation case, and we do not recommend or accept any future operation of the pump without the gland water supply for any length of time without a change to the existing pump bearing design.Our database search and study of-past field service records (.written and verbal) did s-ow that some damage to the.bearings and/or shaft can occur due to the lack of pump.gland water supply. HQ'wev-e,_-e.do.-noLhave-conclusive data which indicates a catastrophic pump failure would have resulted within 90 minutes of operation without the gland water supply and subsequent 48 hours with re-supply of gland water. Our rotor dynamics analysis indicated that it is highly unlikely to damage all top rubber bearings subjected to the dry-running condition. | |||
We believe that given the ample motor horsepower and the good existing maintenance practice in this case (ali mament, good fits and registers, etc.), the pump could have survived the 90 minutes of "dry" operation. | JOHNSTON PUMPS 800 Koomey Road Tel: (800).926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 | ||
When the gland water supply was re-introduced, the pump could have operated (possibly with higher vibration and some damaged bearings) for an additional 48 hours.Prepared By Checked By Report No. Date Revision Rev. Datel Rev. By Mike.Cugal Lanka Pannila JP04-20 07/16/04 02 08/05104 MC 4}} | _ _ _Engineering Evaluation on Short-term Dry-run i Job Number Customer NPPD - Cooper Nuclear Station' I Pump Size and Model BJ 28KXL 1-STG Project Manager Jerry Harrelson Our interviews with senior field service technicians resulted in similar conclusions. In their experience, when the gland water supply is stopped, the rubber bearings get over-heated, and in some cases they have seen smoke coming out of the column. In most cases, the pump will begin to vibrate, and eventually it will be pulled for a repair.. There were some cases where one or two bearings w,.ere completely damaged and pumps operated for over 48 hours. The general opinion of the service technicians is that the pump could have survive the 90 minutes of dry-operation though the bearings may have gotten severely damaged, and operation of the pump for the subs.equent 48 hours with re-introduction of the gland water supply would have been possible. | ||
2.3 Rotor dynamics analysis We believe that the loss of gland water supply will cause loss of bearing damping effect (by losing the fluid film between the bearing and shaft) and could open up the running clearances (which will change the bearing stiffness). In order to study the rotor dynamics behavior in more detail, we conducted a lateral rotor dynamics analysis (report JP04-18). | |||
Due to time constraints and the unavailability of some pertinenilgeometrv information (i.e. mnotor rotor details),. | |||
we had to use this study diligently. Our conclusions from the analysis are as follows: | |||
0 It is quite typical for. a vertical, pump rotor system to have a number of critical speeds that are close to the excitation frequency. Such a case was proven to be true in this analysis. In this study, we focused on the modal shapes of the rotor in order to study the effect of the shaft and bearing contact. | |||
0 The baseline analysis (pump in normnal operating condition with gland water supply) indicated that there is indeed a natural frequency mode within 4.5% of the running frequency. The mode shape, however, indicated that the shaft and the inner column deflect in the same phase, providing a non-contacting mode (refer to report JP04-] 8). | |||
" The worst case analysis (pump without gland w,ater supply and assuming all 7 top'bearings are lost and do not provide any bearing support) indicated that a higher mode natural frequency is near the running speed (within 6 %), and that the deflection mode of the shaft and inner column are not in the same phase. This would indicate a contact between the shaft and bearing, and that severe damages to the bearings can be expected. | |||
" Based on the mode shape analysis, we believe that the worst case scenario would not happen during the pump operation. The mode shape indicated ihat two bearings may have come in severe contact with the shaft, but not all dry-running bearings would have been damaged. | |||
Our conclusion based on the rotor analysis is that when the pump is run dry some shaft contact with the bearings will occur. Based on the modal shape in the worse case scenario, we believe that contact will be made on some of the bearings, but loss of all column bearing support is not likely to happen. | |||
Prepared By Checked By Report No. Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16/04 .02 08/05/04 ,- MC 3 | |||
JOHNSTON PUMPS 800 Koorney Road Tel: (800) 926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number I Customer NPPD - Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-SIG Project Manager Jerry Harrelson Based 0i the modal shape analysis, the inner column mode shape was detected near the running speed. If the inner column fails, the result could be a re-supply of the lubrication to the column bearings as the discharge pressure pushes the fluid into the column. This would become somewhat beneficial in this case (depending on the nature of the failure), as it would introduce the lubrication and damping back to the column bearings. | |||
3.0 Summary and conclusions It is general knowledge and recommended practice never to operate a vertical pump withfrubber bearings without proper lubrication*. We wish to clearly state that the evaluations and statements of this report regarding this issue does not change such views. We also wish to state that the statements and summary of this report are intended only to share our views on this particular operation case, and we do not recommend or accept any future operation of the pump without the gland water supply for any length of time without a change to the existing pump bearing design. | |||
Our database search and study of-past field service records (.written and verbal) did s-ow that some damage to the | |||
. bearings and/or shaft can occur due to the lack of pump.gland water supply. HQ'wev-e,_-e.do.-noLhave-conclusive data which indicates a catastrophic pump failure would have resulted within 90 minutes of operation without the gland water supply and subsequent 48 hours with re-supply of gland water. Our rotor dynamics analysis indicated that it is highly unlikely to damage all top rubber bearings subjected to the dry-running condition. | |||
We believe that given the ample motor horsepower and the good existing maintenance practice in this case (ali mament, good fits and registers, etc.), the pump could have survived the 90 minutes of "dry" operation. When the gland water supply was re-introduced, the pump could have operated (possibly with higher vibration and some damaged bearings) for an additional 48 hours. | |||
Prepared By Checked By Report No. Date Revision Rev. Datel Rev. By Mike.Cugal Lanka Pannila JP04-20 07/16/04 02 08/05104 MC 4}} |
Latest revision as of 03:09, 23 March 2020
ML061980186 | |
Person / Time | |
---|---|
Site: | Cooper |
Issue date: | 08/05/2004 |
From: | Cugal M Johnston Pump Co |
To: | Nebraska Public Power District (NPPD), Office of Nuclear Reactor Regulation |
References | |
FOIA/PA-2006-0007, R4-5A39 JP04-20, Rev 2 | |
Download: ML061980186 (7) | |
Text
p.-~,y~O '71 I i Ril~-5A4 3q Jolrnson"PUiMP Company Engineering Evaluation Regarding Short Term Dty-run of-Rubber Column Bearings Cooper Nuclear Station, Nebraska Public Power District fByron Jackson 28KXL I-STG! Chattanooga Sei'ice Center Report Serial No.: 3P04-20)Rev. 02
......- o..°.;...;:..o.... °o....%°..v;-.
bigireei Dept.,B1rooks ire, D
RciHO Cf LnE.o.:O2.
Prpae 3W\Rhd1CugLP
~itai26~4
Johnston Pump Company Engineering Evaluation Regarding Short Term Dry-run of Rubber Column Bearings Cooper Nuclear Station, Nebraska Public Power District Byron Jackson 28KXL I-STG, Chattanooga Service.Center Report Serial No.: JP04-20 Rev. 02 I
Originating Department:
Engineering Dept., Brookshire.
Revision No.: 02 Prepared By: Michael Cuga], PE Initial:
Date: 07/] 6/04 Checked By: Lanka Pannila, P.E.
Initial:
{~ I i
Table of Contents
- 1. Introduction 1
- 2. Discussion
- 4. Summary/ Conclusions
JOHNSTON PUMPS 800 Koomey Road Tel: (800) 926-6688 Brookshire, IX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number Customer NPPD - Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-STG Project Manager Jerry Harrelson
- 1. Introduction T"he.purpose of this report is to provide an engineering evaluation of the following question:
Given conditions that existed at Cooper Nuclear Station(l), will the Service Wat6r pumps(2) continue to function 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> after Gland Water low is reduced to zero flow for 90 minutes and then restored to normal?
(1) During Janiary 21, 2004 - February 11,.2004:
River water level 875.5 MSL - 877.5 MSL.
Service water temperature 545°F.
Pump running at capacity (5500 gpm).
Average discharge pressure 50 psig average.
Gland water flow -678 gpm to the enclosing tube (16-24 psi).
Vibration (1ST) normal.. .......
-(2) Byron Jackson 28KXL 1-Stage VCT pumps with 180 rpm, 300 lip Motors..
The following material conditions existed as of February 2003: 1). New pump assembly, packing, coupling, shafts,.cutlass bearings, and impeller. 2). Rebuilt outer column, discharge nozzle, registers, and spider bushing supports.
- 2. Discussion
Background
The function of the Gland Water flow is to provide clean water lubrication lo the pump's stuffing box area, and column bearings during the pump operation.
In vertical pumps, the column bearings provide 'bumper" bearing support. Due to negligible pressure differences
-across the bearings, they do not behave as typical hydrostatic bearings. However, a thin"fluid film between the bearing and the shaft results in some bearing damping and fluid stiffness.....
For the'subject pump, at the given river level, if the gland water flow is stopped, the top 7 bearings (including the stuffing box bearing) would not receive the lubrication. The bottom 5 bearings (3 bottom column bearings and 2 bowl bearings) 'wouldbe submerged in the *umping fluid.
In order to provide the answer to the posed question, we underook the following steps:
Prepared By Checked By , Report No. Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16)04* 02 08105104 MC J
JOHNSTON PUMPS 800 Koomey Road Tel: (800) 926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number Customer NPPD - Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-STG I Project Manager Jerry Harrelson
] Technical assessment by the Johnston Pump engineering staff.
- 2. General review and study of the field repair history of similar cases.
- 3. Rotor dynamics analvsis.
2.1 Technical assessment After gland flow is stopped, the packing ]hebox area will not receive proper lubrication and cooling flow, and the top 6 column bearings will not receive lubrication. The packing box bearing will heat up due to contact from the shaft. The effect of this could depend on the amount and force of the contact with the rotating shaft. If the shaft rotates without much wobble, it is conceivable that damage to the bearing or shaft can be minimal, Ifthe shaft contacts the stuffing box bearing constantly, the packing gland will get over-heated and dry, and the bearing clearance will increase.
A typical vertical shaft is in tension and, in theory, it will rotale freely without any contact if the alignment and register fits are true and no manufacturing and machinin2 tolerances exist. In real operation, hov%,,.ver,.ThejriduaL unbalance of theorotor, Some hydraulic unbalance, allowable manufacturing tolerances, and the natural frequency of the rotor could cause the rotor to wobble and contact the column bearings.,If lubrication to the bearings is not present, we can foresee local heating of the rubber bearings and wveakening of the rubber material. The constant contact with the shaft could result in loss of the bearing material. The severity of the mhaterial loss would highly depend on the nature and amount of contact. However, wye do believe that in general the bearing clearances will increase significantly due to the contact of the shaft.
.... hbearing clearances increase, the pump can experience higher vibration of the pum shaft and possibly the ump stfuctures m general. The severity and effect o e vi ration are difficult to predict, but a catastrop ic pump failure within 9U minutes of operation without the gland water supply and subsequent 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> of operation with the gland water supply is not likely to happen due to the vibration from increased bearing clearances.
There is a remote chance of a pump shaft seizure when it is operated without lubrication against rubber bearings.
However, the probability of a seizure is very low if the pump does not have to stop and restart.
2.2 Field repair history We were not able to locate any written repair rep6rt that dealt with the exact nature of this case. No record was'.
found w',here the exact duration of the dry-running was documented. Our assessment was based on verbal communications with various Johnston Pump service centers. In most cases, operating pumps withoutlubrication to the rubber fa-lure, bearings the shaft was seized, butinthe resulted severe damage duration to ihe of the operation bearings columnseemed and shalts.
to have In somecases been longer ofpump than a day or so. In ih-es'e cases, tme gland water supply was not restored.
Prepared By Checked By Report No. " Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16/04 02 08/05/04 MC 2
JOHNSTON PUMPS 800 Koomey Road Tel: (800).926-6688 Brookshire, TX 77423 Fax: (281) 934-6056
_ _ _Engineering Evaluation on Short-term Dry-run i Job Number Customer NPPD - Cooper Nuclear Station' I Pump Size and Model BJ 28KXL 1-STG Project Manager Jerry Harrelson Our interviews with senior field service technicians resulted in similar conclusions. In their experience, when the gland water supply is stopped, the rubber bearings get over-heated, and in some cases they have seen smoke coming out of the column. In most cases, the pump will begin to vibrate, and eventually it will be pulled for a repair.. There were some cases where one or two bearings w,.ere completely damaged and pumps operated for over 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. The general opinion of the service technicians is that the pump could have survive the 90 minutes of dry-operation though the bearings may have gotten severely damaged, and operation of the pump for the subs.equent 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> with re-introduction of the gland water supply would have been possible.
2.3 Rotor dynamics analysis We believe that the loss of gland water supply will cause loss of bearing damping effect (by losing the fluid film between the bearing and shaft) and could open up the running clearances (which will change the bearing stiffness). In order to study the rotor dynamics behavior in more detail, we conducted a lateral rotor dynamics analysis (report JP04-18).
Due to time constraints and the unavailability of some pertinenilgeometrv information (i.e. mnotor rotor details),.
we had to use this study diligently. Our conclusions from the analysis are as follows:
0 It is quite typical for. a vertical, pump rotor system to have a number of critical speeds that are close to the excitation frequency. Such a case was proven to be true in this analysis. In this study, we focused on the modal shapes of the rotor in order to study the effect of the shaft and bearing contact.
0 The baseline analysis (pump in normnal operating condition with gland water supply) indicated that there is indeed a natural frequency mode within 4.5% of the running frequency. The mode shape, however, indicated that the shaft and the inner column deflect in the same phase, providing a non-contacting mode (refer to report JP04-] 8).
" The worst case analysis (pump without gland w,ater supply and assuming all 7 top'bearings are lost and do not provide any bearing support) indicated that a higher mode natural frequency is near the running speed (within 6 %), and that the deflection mode of the shaft and inner column are not in the same phase. This would indicate a contact between the shaft and bearing, and that severe damages to the bearings can be expected.
" Based on the mode shape analysis, we believe that the worst case scenario would not happen during the pump operation. The mode shape indicated ihat two bearings may have come in severe contact with the shaft, but not all dry-running bearings would have been damaged.
Our conclusion based on the rotor analysis is that when the pump is run dry some shaft contact with the bearings will occur. Based on the modal shape in the worse case scenario, we believe that contact will be made on some of the bearings, but loss of all column bearing support is not likely to happen.
Prepared By Checked By Report No. Date Revision Rev. Date Rev. By Mike Cugal Lanka Pannila JP04-20 07/16/04 .02 08/05/04 ,- MC 3
JOHNSTON PUMPS 800 Koorney Road Tel: (800) 926-6688 Brookshire, TX 77423 Fax: (281) 934-6056 Engineering Evaluation on Short-term Dry-run Job Number I Customer NPPD - Cooper Nuclear Station Pump Size and Model BJ 28KXL 1-SIG Project Manager Jerry Harrelson Based 0i the modal shape analysis, the inner column mode shape was detected near the running speed. If the inner column fails, the result could be a re-supply of the lubrication to the column bearings as the discharge pressure pushes the fluid into the column. This would become somewhat beneficial in this case (depending on the nature of the failure), as it would introduce the lubrication and damping back to the column bearings.
3.0 Summary and conclusions It is general knowledge and recommended practice never to operate a vertical pump withfrubber bearings without proper lubrication*. We wish to clearly state that the evaluations and statements of this report regarding this issue does not change such views. We also wish to state that the statements and summary of this report are intended only to share our views on this particular operation case, and we do not recommend or accept any future operation of the pump without the gland water supply for any length of time without a change to the existing pump bearing design.
Our database search and study of-past field service records (.written and verbal) did s-ow that some damage to the
. bearings and/or shaft can occur due to the lack of pump.gland water supply. HQ'wev-e,_-e.do.-noLhave-conclusive data which indicates a catastrophic pump failure would have resulted within 90 minutes of operation without the gland water supply and subsequent 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> with re-supply of gland water. Our rotor dynamics analysis indicated that it is highly unlikely to damage all top rubber bearings subjected to the dry-running condition.
We believe that given the ample motor horsepower and the good existing maintenance practice in this case (ali mament, good fits and registers, etc.), the pump could have survived the 90 minutes of "dry" operation. When the gland water supply was re-introduced, the pump could have operated (possibly with higher vibration and some damaged bearings) for an additional 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.
Prepared By Checked By Report No. Date Revision Rev. Datel Rev. By Mike.Cugal Lanka Pannila JP04-20 07/16/04 02 08/05104 MC 4