resonace at vane pass frequency

[Ssampson]

I have a couple of 500 hp vertical pumps. Wastewater application, they are series centrifugal pumps. The pumps vibrate excessively readings are above 1.0 in/sec at 3 times turn speed. They have 3 vane impellers. Bump test indicates a natural frequency of the pump bearing case at 3 times turn speed.
How do I go about stiffening the pump case to change the natural frequency, or what would be the alternatives to stiffening.

 

[machinerys]

Other Alternative is installing Dynamic Vibration absorbers on the bearing Housing.

 

[Hatch]

Yes Dynamic Vibration Absorbers would be the easiest way to go. If you need info try a web search. In most applications (but not all) a dynamic absorber works quite well. Are you sure the resonance is the bearing housing itself or is it the bearing pedestal? If it is the pedestal, you could stiffen the pedestal to try and get the resonant freqeuncy away (higher) from the 3X excitation of the vane pass frequency. I am surprised something has not broken yet because you have direct excitation of the bearing natural frequency. Do not add mass (fill with concrete etc) because that would lower the natural frequency and during startup the vane pass would pass through the natural frequency. C. Hugh (

http://www.Hatch.ca)

 

[smpedi]

Sometimes, it is possible to reduce vibration to acceptable levels by reducing VPF excitation. This can be accomplished by increasing the clearance between the vane and the cutwater (sometimes called the B Gap). Of course, this will affect performance somewhat, but is a lot cheaper fix than modifying the case. Alternatively, you can often use a different number of impeller vanes.

 

[vanstoja]

Another alternative is a variable speed motor drive so you can dial in a running speed to avoid coincidence of any pump or motor vibration source frequencies with known resonance frequencies. Vibration absorber will split the resonance peak at the tuned frequency but leave you with two resonance peaks each side of the "tuned" frequency. If pump load changes shift your motor speed then you could excite one of the residual resonance peaks. If there are outlet guide vanes, the cutwater radial clearance should be more than 15% of impeller radius to avoid impeller blade passing vibration problems though pump hydraulic efficiency may suffer some. Outlet guide vanes numbers should be separated from impeller blade numbers by three or more ideally to avoid "efficient radiation zone " amplification effects. vanstoja

 

[Sivayokan]

I have a similar problem with a vertical pump using a corkscrew type impeller. I get very high vibration levels (45mm/s) when water is pumped through the system. The main component of the vibration is at 1X the motor speed. Can someone explain how the 1X component is produced. I assume it is developed by the corkscrew impeller when water is pumped through, but I can not work out the mechanism. How can the pump be modified to reduce this force.

 

[electricpete]

Sivayokan - would you mind posting this as a separate thread. That makes everything easier to follows.
Thanks
electricpete

 

[smpedi]

As you probably know, a common cause of 1X Vibration is imbalance. Even if the rotor has been balanced perfectly, a hydraulic imbalance can occur that will appear only when fluid is being pumped. This behavior is a consequence of variations in cavity volume from one part of the impeller to another. In multistage centrifugals, it is common to rotate each impeller slightly so that the cavities don't all line up.

You can try to balance the system in place (if your situation allows it). Perform a balance with the pump operating normally. If vibration reduction is possible, hydraulic unbalance is probably the problem. You can further check this hypothesis by checking balance on the pump when dry (probably on a balance stand or other fixture) and comparing the result to the wet case. Assuming that the shaft was balanced to begin with, the dry balance should now be off by the amount of weight and angle that you added for the wet case.

If this is the case, you can elect to leave the balance weights in place, or see if the impeller can be remade to tighter tolerances.

 

[vanstoja]

Returning to Ssampson's problem, there's a recent journal paper that may tell you everything you ever wanted to know and more about optimizing a so-called "standard" Voight-type Dynamic Vibration Absorber (DVA). The reference is:
Asami, T. etal, Analytical Solutions to H_infinite and H_2 Optimization of Dynamic Vibration Absorbers Attached to Damped Linear Systems, Transactions ASME, Journal of Vibration and Acoustics, Vol. 124, April 2002, pages 284-295.
All the equations, though mind-bogglingly lengthy, are at least in comprehensible BS form rather than the usual PhD form. vanstoja

 

[Franko]

Remote possible cause is an impeller resonance similar to that for some fans; see article by Peter Baade: Baade, P. K., 1998, "Vibration Control of Propeller Fans", Sound and Vibration Magazine, July 1998, pp. 16-26.
Also current issue of magazine has an article in their website on changing mass:

http://www.sandvmag.com/

Changing the Effective Mass to Control Resonance Problems, Richard Smith, keywords: effective mass structural resonance, 0105smit.pdf, 4 pp, 53 KB.