Whirl mode harmonics

[UCengno1]

Is there a such thing as a half harmonic of a whirl mode?

I am chasing a vibration issue that is dominant at a specific speed. It is not a fixed event that I can run through as it appears at a lower magnitude at lower speeds and increases in magnitude up to this specific speed. I created a Campbell diagram and cannot find any driver that is a multiple? It is a axial turbine device with a gearbox output so there are lots of possible drivers.

Comments welcome.

 

[sms]

Depends on your nomenclature. How do you define a whirl mode...

There are a number of things that can cause 1/2 running speed vibration. Is that what you are refering to? Is the vibration frequenyc 1/2 of the turning speed of the shaft, if so it could be instability, or a rub...

What kind of bearings on the machine? What is the first critical speed? What is the actual vibration frequency that is bothering you, and what is running speed?

Finally what are the amplitudes? Have you looked at an orbit? If so is it big and round, or is it scrunched up and squashed?



-The future's so bright I gotta wear shades! B-)

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[GregLocock]

1/2 speed whirl is a classic case in the literature. Can be caused by assymetrically flexible shafts assymetric rotors, or the dreaded oil film whirl.

The first two should not occur in the absence of gravity.

http://www.dliengineering.com/vibman/oilwhirl.htm

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[rob768]

check the frequency. As greg says, oil whirl or whip is the most expected cause. whirl occurs a frequency just below 0.5th order (.44 -0.48 i believe). A whirl is not too dangerous, a whip may cause bearing damage.

 

[UCengno1]

It is a 7" long turbine shaft with a sun gear at one end a .4#, 35 bucket turbine rotor at the other. It has a traditional, preloaded 2 bearing support with a calculated 1st critical speed > 70K rpm. The rotor is fed through a 21 passage nozzle.

I have a frequency spike being picked up on my case mounted accelerometer that tracks with speed but at a different slope. It hits some sort of resonance at 35,000 rpm turbine speed with a vibration spike at 7k hz. I also picked up a distinct harmonic at 14k hz also. The vibration spike increases in magnitude by almost 10X between 32k rpm and 35k rpm. I can not find some multiple of my 1/rev or anything else that comes close to the 7k hz that I am measuring. I am a little stumped.

 

[sms]

So it is 12 times running speed? Sorry the way you are describing the problem is not clear to me.

At a turbine speed of 35000 rpm, you have a vibration spike at 7 KHz and one at 14 KHz. Is this correct?

As you vary rpm the amplitude increases between 32K and 35K rpm, so it sounds very much like a resonance being excited.

Are the 7 KHz and 14 KHz frequencies fixed vs speed, or do they track running speed at all, or do they just appear? In other words at 32000 rpm do you see peaks at 6400 Hz and 12800 Hz (12 and 24 times running speed)?

If so you need to find something that would cause 12 time running speed correct? So what more can you say about the machine configuration?

But I get the feeling that is not what you are trying to tell us. If the frequencies are not directly related to running speed, they may be an acoustic excitation due to the fluid flow through the turibine or turbine piping. Flow induced vibration is not uncommon and is usually a resonant phenomonon.


-The future's so bright I gotta wear shades! B-)

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[UCengno1]

They do track running speed at a multiple that doesn't correspond to any driver I can identify. They are much lower in magnitude at the lower speeds but I can still pick them up. Your assessment of a 12:1 driver is correct but it is currently unidentifiable. This is a cold air turbine device that runs off 100-150 psi drive air. We are just spinning up the unit as we have not seen any relationship between the existance of the frequency spikes and power output.

 

[sms]

How many teeth on each of the gears in the gearbox? It could be a gear tooth combination.

What is the gear box driving?



-The future's so bright I gotta wear shades! B-)

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[GregLocock]

Well, having misled us once, can you post a graph, preferably a waterfall plot or campbells diagram, so we can see what is going on.

Do I gather you have 35 blades on the turbine?



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[btrueblood]

Ball/roller element bearings or oil film (you said preloaded, so let's assume rollers)? If roller elements, what precision level (ABEC #), how many rollers? Maybe 12 or 24?

 

[electricpete]

If you can use high resolution spectra, you can nail down whether it is an exact multiple of running speed or not, which may help focus the investigation.

As sms said, gear tooth combination would be helpful. Maybe simple gear mesh frequency with 12 teeth on the pinion?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.