Coupling failure Risk in VSD motor driven Lobe type blower 1

[svkd]

Guys,

We have VSD motor driven Lobe type blower.Due to nature of process, this blower can see large and quick swings in the speed output. Its also a fairly large motor(1000 HP) and being a lobe type there’s quite some mass on the coupling end.Does these quick speed changes creates any risk for coupling failure due torsional vibrations

 

[EdStainless]

Do you know all of the critical frequencies? You program the drive to ramp through these quickly and never allow operation near one.
As for torque, you need to size the coupling for the expected loads, including accelerations.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[JJPellin]

I would not expect to see coupling problems associated with rapid speed changes. The coupling should have a service factor of at least 1.5. I would be more concerned about torsional resonance. We converted an induced draft fan to a VSD and experienced a catastrophic coupling spacer failure. I have attached a picture of the failed spacer. Analysis after the fact found that we were exciting a torsional resonance in the fan train. Our motor driver was only 350 HP. I would strongly recommend that you perform a torsional analysis before making this change.

Johnny Pellin

 

[1gibson]

What if a fast speed change (reduction) reverses the load on the coupling? Commanding a fast reduction to driven equipment with a high inertia will load the other side of the key, so any slop in the fit could be a problem eventually. More of a basic mechanical consideration than torsional vibrations or natural frequencies.

 

[svkd]

Johnny ,

Please help me with questions that need to be asked to OEM to make sure that we will be in safe position as far as coupling failure due torsional resonance is concerned. Apart from torsional analysis ,what are data /documents that can be asked to OEM that confirm that there will be no torsional resonance.

 

[JJPellin]

The torsional analysis is really the key to avoiding a torsional resonance. You could ask some questions to gauge how much risk there is of stumbling upon a torsional. Have they ever built this same machine in this same size and speed for another customer? How many identical machines are in service and how long have they been running? You can ask for references of customers that you can contact that have the same machine train to find if they have had any problems. In our machine, we had problems with the cooling fan on the motor and unexplained radial vibration before the coupling failed. In hind-sight, these were indications of the torsional resonance problem before the coupling spacer failed.

Johnny Pellin

 

[electricpete]

I don't work with vfd's and haven't had any experience with torsional resonance so I definitely defer to Johnny and also suspect torsional resonance is probably a more detailed excercize than can be accomplished here.

Nevertheless I'll try to talk through the way I look at it.
There are two distinct aspects of a torsional resonance:
1 - excitation of resonance with a discrete forcing function at the resonant frequency
2 - exciting the resonance by a broadband / impulse-like forcing function (analogous to ringing a bell or impacting a structure for vibration bump test).

Scenario 1 requires a speed-dependent torsional excitation to be relevant to this discussion. For a given set of endpoints, rapid change in speed is better than slow change in speed with respect to exciting a torsional resonance (it's better to pass thru a resonance fast than slow… if endpoints are resonant than either case is equally bad). So I guess scenario 1 is not really relevant to this discussion.

Scenario 2 requires something like an impulse in torque. That could maybe be a rapid speed change. In that case the oscillation does not persist long after the transient. For a give speed change what determines how large the oscilation is and how long it lasts is heavily dependent on torsional damping. What type coupling is it?


=====================================
(2B)+(2B)' ?

 

[svkd]

Electricpete It is diaphragm coupling Flender N EUPEX .

 

[JJPellin]

The concern with a VSD is that they generate a simulated sine wave to change the speed of the motor. On older systems, this simulated wave was a stepped wave that produces other frequencies that could excite torsional resonance. So, in addition to the fact that the running speed is changing, these other frequencies give you more opportunities to accidentally have a forcing function that perfectly lines up with a torsional resonance. I am not up on the latest technology. It is very possible that newer VSD’s are better able to produce a smooth sine wave without the extra frequencies present. But, even if this is the case, I would want to check to be sure that no torsional resonance frequencies are within the range of speeds where the VSD would be running.

You describe the coupling as a diaphragm design. But, the drawing provided looks more like an elastomeric coupling. A diaphragm or disk pack coupling would not have much damping for torsional vibration. An elastomeric coupling would tend to have more torsional damping. The coupling design would play a major role in the analysis for torsional resonance response.


Johnny Pellin

 

[electricpete]

From word description and google search seems like diaphragm coupling.

I hijacked the discussion toward trying to figure what implications of torsional resonance are with respect to rapid speed change. Now I rereadJPellin's comment and see it was just about torsional resonance in general not associated with any rapid speed change.

=====================================
(2B)+(2B)' ?

 

[svkd]

Johnny Pellin , The coupling is diaphragm or disk pack coupling Flender (N EUPEX).

 

[svkd]

Electricpete,
Quote: For a given set of endpoints, rapid change in speed is better than slow change in speed with respect to exciting a torsional resonance (it's better to pass thru a resonance fast than slow… if endpoints are resonant than either case is equally bad)

Mean by endpoints is speed setpoints here??

 

[electricpete]

Electricpete,
Quote: For a given set of endpoints, rapid change in speed is better than slow change in speed with respect to exciting a torsional resonance (it's better to pass thru a resonance fast than slow… if endpoints are resonant than either case is equally bad)
Mean by endpoints is speed setpoints here??

Yes. I am picturing transition between two steady state speeds. I was pointing out that we have a lot more to worry about than transition rate if the the steady state speeds are resonant.

diagphram coupling will have relatively low torsional damping.

=====================================
(2B)+(2B)' ?

 

[CouplingGuru]

Svkd,

There are two main types of coupling vibrations and resonances that we concern ourselves with. 1st is when we have a lateral resonance, this is typically associated with longer span couplings operating at or near their critical speed. This is where rapid changes in speed can help you get through the critical speed limitations. I don't think that is what you are dealing with in your application. The other main cause of vibration is due to Vibratory torques, these vibratory torques occur when when the load and driving equipment are essentially fighting each other. Many times this has to do with various operating speed points in an application. We typically see this in reciprocating compressor applications where speed is changing and the load is vibrating due to the nature of the compressor. In your case with a blower, there are natural vibrations that will occur due to the physical design of the blower and how it moves air at various speeds. Most often times Blowers are less troublesome then compressors. However, your selection of a Neupex coupling does concern me slightly, the rubber elements that are in that style coupling are designed to eat energy as well as flex for misalignment and in your application they could wear out quite rapidly. This could cause a lot of maintenance and down time. We typically see Gear couplings for type of application, and if you are looking for a long life low maintenance solution go with a disc coupling with flexible metallic elements.

When it comes to couplings we are always here to help.

http://WWW.PSCCOUPLINGS.COM