Dear all,

We're facing a problem with the CWP in our thermal power plant. We have 04 pumps 1A, 1B, 2A and 2B. All of them will increase vibration when low tide event (as manual shows). But one of them, pump 1B has a higher vibration than others, it reaches above 7mm/s at point NDE (as attached file), you can find more detail in the attached word file.

We are afraid of structure vibration (something gets weaker) of resonance. But bump test didn't come with a good result, can't find the natural frequency. I think this pump is too big to bump by a hammer?

I post here the PI spectrum between vibration and tide level, and the spectrums from CSI 2600.

Please advise me with some ideas. Thank you so much!

CWP1B-after 20180401(Vibration and tide level after 01-04-2018)

CWP1B-Apr 2016(Vibration and tide level Apr 2016)CWP1B-Jan 2016(Vibration and tide Jan 2016)CWP1B-low water level[6.4m)-axial vibration at thrush bearing-waveform and FFT(CWP1B-low water level(6.4m)-axial vibration at thrush bearing-waveform and FFT)CWP1B-low water level[6.4m)-DE motor-phase(CWP1B-low water level(6.4m)-DE motor-phase)CWP1B-low water level[6.4m)-DE motor-waveform and FFT(CWP1B-low water level(6.4m)-DE motor-waveform and FFT)CWP1B-low water level[6.4m)-NDE motor-phase(CWP1B-low water level(6.4m)-NDE motor-phase)CWP1B-low water level[6.4m)-NDE motor-waveform and FFT(CWP1B-low water level(6.4m)-NDE motor-waveform and FFT)CWP1B-low water level[6.4m)-shaft-raw orbit and full spectrum(CWP1B-low water level(6.4m)-shaft-raw orbit and full spectrum)CWP1B-low water level[6.4m)-shaft-waveform and FFT(CWP1B-low water level(6.4m)-shaft-waveform and FFT)CWP1B-normal water level[7m)-axial vibration at thrush bearing-waveform and FFT(CWP1B-normal water level(7m)-axial vibration at thrush bearing-waveform and FFT)CWP1B-normal water level[7m)-DE motor-phase(CWP1B-normal water level(7m)-DE motor-phase)CWP1B-normal water level[7m)-DE motor-waveform and FFT(CWP1B-normal water level(7m)-DE motor-waveform and FFT)CWP1B-normal water level[7m)-NDE motor-phase(CWP1B-normal water level(7m)-NDE motor-phase)CWP1B-normal water level[7m)-NDE motor-waveform and FFT(CWP1B-normal water level(7m)-NDE motor-waveform and FFT)CWP1B-normal water level[7m)-shaft-raw orbit and full spectrum(CWP1B-normal water level(7m)-shaft-raw orbit and full spectrum)CWP1B-normal water level[7m)-shaft-waveform and FFT(CWP1B-normal water level(7m)-shaft-waveform and FFT))


Original Post

What is mean/normal low tide?

What is maximum low tide?

What is tide range (high to low)?

I have used Tide Charts, but I guess you have a water level sensor (LVL onstream TBS), correct?

It looks like high vibration that has been increasing is at motor NDE-X with dominant frequency at 2xSS, correct? Is the X-direction perpendicular to pump discharge pipe axis? 7-mm/sec on "problem" pump, but what are the other three pump's vibration levels for comparison?

The dominant frequency in thrust direction is at 5xSS or vane pulsation, and this changes with suction pressure; so this is normal.

No details for the hammer test. It is difficult to excite a large structure at 25-Hz. Did you try to excite pump column from dewatered pump bay? I have done this, but you have to correct data for water column mass effect and possibly for impeller mass/inertia effect.

Some possible causes for change over time:

Foundation/grout deterioration

Poor suction from sand, silt, or shells in suction bay; or altered flow from adjacent bays.

Accelerated wear ring or bearing wear from sand or corrosion or rubbing

Incorrect height setting for shaft axial position at thrust bearing can affect impeller-case rubbing


Accurately measure pump performance and efficiency to compare to other pumps -- to reveal excessive internal clearances

Inspect suction bay for suction bell damage and compare to one or more other bays for unusual flow patterns and fouling

Conduct Operating Deflection Shape (ODS) to show structural motion at 1xSS and 2xSS -- especially at machine base and foundation interface

Consider installing two underwater accelerometers on pump case at impeller elevation -- I have done this and it is not easy or reliable -- would help with ODS

Conduct advanced motor current spectrum analysis for pump cavitation and pulsation (compare pumps) and to correlate with performance test

Final Thought

Probably no one is going to have too much concern about the vibration until or unless pump performance falls off. Good luck getting support for expensive testing.


CWPs for thermal power plants can be very tall and difficult to excite by a standard instrumented hummer. You may need a much bigger object like 4"x4" plywood with the help of a crane.

Sometimes, the background excitation could generate enough vibration to guess the natural frequency.

I recall that the tide level had a decent effect on the vibration on most of the CWPs I worked with. But there are other operational factors that one needs to consider including the resistance on the line downstream the pump. Backpressure coming from the condensers or other heat exchangers pushes the pump to run away from its design BEP.  In such cases, the current of the motor decreases also as a consequence.

Regards- Ali M. Al-Shurafa

Have you checked the history of these pumps? The problem is from erection to now or just in recent time?

How the vibration in comparison with three rest pumps? how arrangement of these pump? which location of the pump with highest vibration compared with the other pump?




Still looking at Tide Tables?

I have used the PCB large impulse hammer (12-lb. head) with a custom 20-lb. extension mass on large circulating water pumps. Impact locations were at top of motor and on pump case at impeller (dewatered sump) with impact-response in both X and Y directions.


Walt Strong posted:


Still looking at Tide Tables?

I have used the PCB large impulse hammer (12-lb. head) with a custom 20-lb. extension mass on large circulating water pumps. Impact locations were at top of motor and on pump case at impeller (dewatered sump) with impact-response in both X and Y directions.


First of all, I want to thank all your recommendations. I'll still looking for the test to determine the main reason in this case.

Dear Mr. Walt,

The low level is 6.4m and normal level is 7m.

The highest vibration at NDE and it's 2X peak, yes. The X direction is 45 degree from the discharge direction. Please focus on orbit at pump bushing. Do you think it's the sign of impact because the plots are not circular. Look like the travel of shaft is limitted by something?


Photos (2)

"The low level is 6.4m and normal level is 7m."

Is Normal level the typical average? Usually the concern is the difference between High tide and Low tide. For example in the Boston, MA area the tide range is about 9-feet or 3-meters. This is a lot more than your 0.6 meter tide range! Moon phase and orbit distance will typically change tide levels.

The shaft orbit is at the coupling area, and may indicate a rub. Perhaps the shaft is not centered in seal housing. It is not obvious that there is any great connection between shaft vibration and motor housing vibration and significance of the tide. Are you mixing apples and oranges with your data?


Add Reply