We have been facing high vibration problem in a vertical 5 stages pump since commissioning. Q is 15 CuM/Hr. Head is 8.5 Kg/Cm2.n=2920 RPM. The vibration was higher on motor NDE bearing(around 10mm/sec). Durinh a later test, it was 150mm/sec(We did not have a vibration analyser at that time). Alarmed, we opened up the pump. Its bottom guide bearing was found badly worn out. we replced the complete rotating assembly after balancing to 2.5G. The pump was intalled back. Vibration reduced but it again rose to around 25mm/sec. We checked the pump base and suspecting voids beneth, we completely dug out the old grout and regrouted the base plate. Pump was installed back but no change in vibration level was found. We replaced the motor bearings. Now we have an anlyser. We took spectrum of NDE bearing of motor (Attachment no.1). Peak is at 1X. No load of motor was carried out. It shows vibration less than 1.2mm/sec in all diretions and all locations. There is clearance around 0.1 mm in motor NDE bearing and housing.

My question is? Should we go for balancing of motor rotor. Will unbalance producing only 1.2mm/s vibration under no load accompanied with looseness in bearing housing multiply it to 25mm/sec on load.

This message has been edited. Last edited by: AKHTAR,

[QUOTE]Originally posted by AKHTAR:
The motor spectrum is also attched herewith.

What kind of product are you pumping? Does it have particulates in it? What material is used for pump bushings? Maybe the wear is very rapid. How long did you run the pump first time after you discovered worn bushings? Did you do an impact tests to see if this pump has resonance close to operating speed? The wave form shows a beat frequency. Put your analyzer on life mode and see if vibration fluctuates. Possibly vibration is transmitted from adjacent pumps?

Akthar,
Can you post the Spectrum and TWF collected on other direction of the motor. As Alex quoted, Resonance problems are most common in such type of pumps.
Regards,
Syed

It appears that you have beating with beat frequency of about 350 cpm. Take a higher res spectrum to confirm.

Akhtar,

Sounds like your pump vibration may be due cantilever effect excited by lack of system rigidity. Several factors can contribute to this phenomenon depending on how the pump train is installed ie. base plate rigidity, motor bracket rigidity (cylindrical support between motor and base) and pipe strain/stress. Did you try coupled and uncoupled runs to check vibration response? Is this pump operated under various speeds and if so, is the vibration higher at one speed then most?
It is very unlikely you'll resolve this vibration by balancing, Motor/bracket bracing may be in order.
Hope this helps...vibration profiles were not available on your thread.


PS. OK I see it now!
Cheers...Rajan Muthukrishnan

quote:

Originally posted by Alex J:
What kind of product are you pumping? Does it have particulates in it? What material is used for pump bushings? Maybe the wear is very rapid. How long did you run the pump first time after you discovered worn bushings? Did you do an impact tests to see if this pump has resonance close to operating speed? The wave form shows a beat frequency. Put your analyzer on life mode and see if vibration fluctuates. Possibly vibration is transmitted from adjacent pumps?

Alex,
We are pumping lake water with minute particulates. Pump bushings are made of finocot material. Composition not known. The pump has been running on and off since the last one year with approx. 2 to 3 hours a day. After discovering worn bushings, in the trial run itself , the vibratrion was found beyond limit.
Bump test was not performed. It will require to remove the pump out completely. Time waveform shows amplitude modulation but as is clearly seen ,the carrier frequency is 1X. Vibration is fluctuating rapidly and by large amounts. It can not be beats since all nearby pumps were stationary at the time of recording. Alex, what is your opinion regarding cruciform provided below the bellmouth assembly to prevent vortices.. Is it a must for all vertical pumps?
If not, why?

This message has been edited. Last edited by: AKHTAR,

quote:

Originally posted by RajSha:
Akhtar,

Sounds like your pump vibration may be due cantilever effect excited by lack of system rigidity. Several factors can contribute to this phenomenon depending on how the pump train is installed ie. base plate rigidity, motor bracket rigidity (cylindrical support between motor and base) and pipe strain/stress. Did you try coupled and uncoupled runs to check vibration response? Is this pump operated under various speeds and if so, is the vibration higher at one speed then most?
It is very unlikely you'll resolve this vibration by balancing, Motor/bracket bracing may be in order.
Hope this helps...vibration profiles were not available on your thread.
PS. OK I see it now!
Cheers...Rajan Muthukrishnan

Rajsha, It can be cantilever effect since there is no support for the hanging pump column . But I have seen so many pumps of much much bigger capcity and HP running quite smoothly in cantilever form. There is nothing spe ial with it. Motor bracket is as usual with windows for passing of discharge line. Rigidity will be lesss in this direction resulting in higher vibration. but this is true for all pumps and always vibration parallel to discharge line is higher due to this reason.

quote:

Originally posted by Syed Ibrahim:
Akthar,
Can you post the Spectrum and TWF collected on other direction of the motor. As Alex quoted, Resonance problems are most common in such type of pumps.
Regards,
Syed

Spectrum and waveform are very much similar on the other side.

quote:

Originally posted by AKHTAR:
...Time waveform shows amplitude modulation but as is clearly seen ,the carrier frequency is 1X....

One can never confidently conlude the above without verifying frequency content via a high resolution spectrum.

It is not unusual to have a certain amount of shaft wip in a vertical pump. The vibration can usually be reduced by trim balancing the motor and pump together. There should be a place on the top of the motor to add weight. If you have to add a large amount of weight to reduce the vibration you probably have problems in your pump.

Answer to your question: I feel there is no problem with the motor.

I still suspect some assembly problem in the pump side only.

Too late in responding. The problem was solved long back. Foundation grout was giving in. Also, there was resonance of motor stool. Stiffened the motor stool. Changed the foundation and regrouted. Pump is running fine below 5mm/sec.
Regards
Irshad

Akhtar,

Physically looking at the foundation of the pump, was it really looking bad?

One of our pump has a similar problem, the foundation is in a bad shape. Its so bad that one hold down bolt cannot be used. All the other bolts were tightened.

The pump started and vibrations reached upto 41 mm/sec peak. It was immediately shut down and I asked the Maintenance people to regrout the foundation, but they insisted that its not the case, because all other bolts were tight.

How to convince them that this can be a foundation problem.

I'll see if I can send you a snap of it.

ZEE-VIBE,
Hit with a hammer on the foundation all around. Where ever it is hollow, you will hear a different sound. Also put some water from the side . If this water goes inside the foundation rapidly, you can suspect hollow inside the foundation or grout. This is what we had done to confirm foundation problem. After confirming, we decided to inject high strength resin putty to strengthen it but as soon as drilling started the foundation crumbled and its crust came out. So we rebuilt the foundation.
REGARDS
IRSHAD

Akhtar,

Thanks for your suggestions. I've tried hammering the foundation, but didn't hear anything suspicious.

However the pump had been overhauled. Shaft was replaced with a new one and all the bearings replaced.

Maintenance staff told me that because this is water lubricated bearings and the water here has silt in it, therefore over a period of time, it deteriorates the bearings immensely.

The new pump started and the overall levels have decreased drastically. Attached is the data.

A newly designed pump will replace this one in Feb, 09. I hope this pump performs well till Feb.

Foundation_in_a_very_bad_shape.pdf (283 Kb, 38 downloads) Raw Water Pump A

.

Zee-Vibe,

Was 36.875 HZ peak the resonance or close to resonance frequency? If so, what is the excitation frequency? What are the apparent sidebands ? The same hump, although much smaller, is still there.

Not that it matters now, but just out of curiosity...

Dave

Akhtar,

quote:

I've tried hammering the foundation, but didn't hear anything suspicious.

I am not sure what you were trying to achieve by this test or how you did it. If you were trying to measure the natuaral frequency of the motor-pump on its foundation, then you would apply impacts and measure force-vibrations at the top of the motor. The missing bolts and possible other loose bolts and grout could reduce foudation stiffness and shift the natual frequencies downward compared to the original installation. This effect is quite common in both steel and concrete foundations for vertical pumps.

Walt

quote:

Originally posted by Walt Strong:
Akhtar,

quote:

I've tried hammering the foundation, but didn't hear anything suspicious.

I am not sure what you were trying to achieve by this test or how you did it. If you were trying to measure the natuaral frequency of the motor-pump on its foundation, then you would apply impacts and measure force-vibrations at the top of the motor. The missing bolts and possible other loose bolts and grout could reduce foudation stiffness and shift the natual frequencies downward compared to the original installation. This effect is quite common in both steel and concrete foundations for vertical pumps.

Walt

Walt,
I was not trying to get natural frequency by hammering the base. I was trying to find the hollowness below the foundation by hammering on it. If the grout below foundation is 100% ok , hammering on metallic foundation produces solid metallic sound otherwise, if it is hollow inside, the sound produced is different and indicates base problem.
Regards
Irshad