Dear all!
We have a special problem in one of our pumps. This is an ordinary centrifugal overhung pump. It has high vibration only in Horizontal direction. (about 38 mm/s rms at almost one peak at 62.5Hz). The speed is 49.4 Hz. I first think about resonance, but I haven't found any exciting force in this freq. The vibration is highest on the casing of the pump. Any idea?
Regards,
Farahani

My guess: is operating point way off best-efficiency-point and/or gases in fluid? This can introduce flow-induced resonance at 63 hz in the flow. Try to change operating point and observe what happens.
Good luck, Arie Mol, NL

could u pls show the detail vibration data of this pump?

Dear arie
Unfortunately I have not access to performance curve of this pump, maybe it is away from BEP. But in the normal operation, because of handling hot water in low pressures, formation of gases is possible. I suspect to cavitation or any type of turbolance. But what makes it difficult: in one situation the pump has been started in low temperature of water, but the vibration was also present!!
I tried to make some changes in operational conditions (openning and closing valves, . . . ) but I didn't find any result. Do you have any idea to test more things?

Dear MD M ISLAM
The vibration is almost low on the motor. On the pump only in the H direction, vibration is high. below is a spectrum from pump outboard bearing in H direction.
Regards,
Farahani

pls let me know these following data.
is there any flactuation in outlet pressure?
and is the bearing is hot or any noise?
How many vanes of the pump?
has the vibration increased abruptly or in slow pace?

What is the dominating frequency of Driving side bearing?

are the axials high in magnitude? and there frequency?

Arie
I'm curious.
You stated "This can introduce flow-induced resonance at 63 hz in the flow."
How were you able to determine this without additional knowledge of the system?
Have you seen flow-induced resoanance at specific frequency? or were you just indicating that flow noise/cavitation impacts could excite a structural resonance (whatever frequency that happens to be)??

Ivibreight:
Simple: determined by means of FFT analysis of electronic 4 .. 20 mA pressure transmitter!
Flow-induced resonance have been found in 50 .. 100 hz range in sewage pump applications.
Farahani:
Perhaps 63 hz is impeller resonance frequency (or 'critical speed'of impeller-shaft-bearing arrangement). Try a bump test on coupling, in vertical direction, at standstill, possibly you find a 63 hz resonance.
Now this resonance can be excited by flow induced vibration as a result of operating point way off best-efficiency-point or turbulence in flow system.
It is still nothing more than my guess, sure I could be wrong.
Regards, Arie Mol, NL

Dear arie,

how the natural frequency of the impeller has changed? is there any crack in the impeller?

The problem is recirculation from worn wear rings. This is very typical for condensate and boiler feed applications. The horizontal vibration is due to the cut water location where the highest pressure is generated in the volute. The liquid will flow back through the wear ring to the suction side. I typically see a large peak with a wide skirt and is between 1x and 2x. If you check previous readings the same frequency will exist but very small.

Interesting point Erik, do you also have an idea as to the reason why a specific frequency is introduced?
Greetings from srilankan shores, wow,
Arie Mol, NL

Erik,

what do u suggest for this case?
do they need to check the wear rings?

Dear MD M ISLAM
Outlet Pressure = 18 bar with about 2~3 bar fluctuation.
The bearings were not noisy nor hot and Envelope didn,t show any bearing damage.
No of vanes = 5
After start of the pump, vibration is high. (the rolling is very fast).
The motor bearing are in healthy condition and their freq. are not about 62 Hz.

Dear arie
I am familiar with pressure analysis with special sensors (for analysing pulsations or ...), but unfortunatly I have not access to an approperiate pressure sensor to do a test on this pump.
I did the bump test but I didn,t find anythings. I think the casing of pump and rotor system are stiff and the natural frequencies are high. (not near 63Hz.)
But I think maybe the exciting force at 63Hz (as you said) are high enough to cause vibration. Now I need any way to confirm this (flow induced excitation at 50~100 Hz). Do you have any idea for this (changing the arrangement of piping? Installing orifices for changing working point? or . . . )

Dear Erik
In this problem we completely change the pump with another one that has worked witout problem before, but the vibration was still present. Anyway I will check the wear rings with more attention in the first opportunity for maintenance.

Regards,
Farahani

When you replaced the pump did you also replace the volute. The problem is a flow induced vibration caused by recirculation or some other flow problem. Check the case, suction and discharge lines, one of them will vibrate at 62Hz. This will confirm somekind of flow induced vibration.

Dear Erik
Yes, the volute also replaced. Also the casing of pump, inlet and outlet piping, all of them have vibration at 62.5Hz. But because vibration is a little more in casing, I think the source is something in the pump (not piping).
Is there any pilot modification to confirm flow problem in this case? below is a photo of the pump. It has a redundant that has not any vibration problem!

Regards,
Farahani

Farahani, the problem is in the pump as you mentioned. You can measure the pulsation with a dynamic pressure tranducer.
I noticed you have an external seal flush system going to the seal and case. This is a long shot, but try slowly closing the amount of flush. Also check for missing orifice or change from the good running pump.

Farahani,

Problems is simple. What Erik has said is correct. But the reasion to happen this is due to nonlinearity and squareness of the stuffing box to the casing. Just check phase across stuffing box plate you will find the pump is rocking. Also this type of problems can be solved with varying stiffness of the suction and discharge pipe support. Just provide all readings and spectrums.

All the best

Krishna nangare

Hi Farahani,

The motor is not very big, so you could do us all a "learning favour" by temporarily installing a VF drive and this will allow you to see how changing speed and the resulting flow rate changes the characteristics.

For instance a waterfall diagram may show it proportionally increasing with speed, or it may appear suddenly, which can indicate a non linear mechanism.

What were the horizontal and vertical and axial natural frequencies of the pump casing that you measured ?

Have you measured the pump shaft/impellor natural frequency by tapping the coupling hub ?

Have you tried adjusting the seal fluid flow ?

The few hundred dollars for a pressure xdcr could save you a lot of money when the labour cost of messing around with this pump is added up.

Cheers

John