A few vibration questions...

[geesamand]

I am testing on a 4.2:1 belt drive reducer by an 1800rpm motor. (That's 425rpm output).

Spectrum analysis during an unloaded test shows vibration at 425rpm and 860rpm, mostly at 860. The shaft first critical speed is measured to be 860cpm vertical and 675 and 900cpm horizontal. We believe balancing is good, but are in the process of checking it. The fact that there are many even multiples here (no obvious smoking guns) is raising questions.

1) Could the motor excite the output shaft first critical speed? In general, can a 2X forcing frequency excite a response at 1X?

2) Could an imbalance/misalignment on the output shaft (textbook response = a peak at 1X and smaller peak at 2X) be amplified to a peak at 1x and bigger peak at 2x if the 2X excites shaft critical speed?

Thanks for your time. I've been studying vibration but the answers to these fundamental questions haven't fallen out from my sources.

Dave

 

[electricpete]

"The shaft first critical speed is measured to be 860cpm vertical and 675 and 900cpm horizontal"

Based on coastdown test?
Are we talking the high-speed shaft or the low-speed shaft?

1) Could the motor excite the output shaft first critical speed? In general, can a 2X forcing frequency excite a response at 1X?

Normally there is a forcing function present at 1x and perhaps harmonics. This forcing function can excite resonances at 1x and perhaps harmonics. I'm not sure how this relates to the original scenario.

2) Could an imbalance/misalignment on the output shaft (textbook response = a peak at 1X and smaller peak at 2X) be amplified to a peak at 1x and bigger peak at 2x if the 2X excites shaft critical speed?

Yes. Sounds like this might be what is going on.


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

 

[sms]

The shaft critical speed of which shaft and how do you know? What kind of measurement did you make? 865 vertical, and 675 and 900 cpm horizontal all sound way too low for real shaft critical speeds. If you are casing data with an accel or velocity sensor, then those are probably structural, not rotor related....

"Why don't you knock it off with them negative waves? Why don't you dig how beautiful it is out here? Why don't you say something righteous and hopeful for a change?" Oddball, "Kelly's Heros" 1970

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

 

[geesamand]

Thanks for the responses.

Yes, those are the output shaft critical speeds, and we have used the simple 'bump' test to measure them. I also did a quick FEA modal analysis to investigate sensitivity and this has confirmed the numbers. It is an overhung shaft, if that helps explain the magnitude.

So regarding question #1, it seems you're confirming that a 1780rpm motor should not in any way, ever excite another part of the structure with a natural frequency of 890rpm. It would seem that the 2X harmonic of something on the output shaft is what's driving it.

One last question: if the shaft resonates at say 860cpm in the horizontal direction, and 900cpm in the vertical direction, can I assume it will tend to resonate in some manner anywhere between 860 and 900cpm? I should therefore aim to get the shaft critical <800 or >950cpm?

Thanks,

David

 

[sms]

So the 425 rpm output shaft has a critical at 800cpm? That could be excited by twice the output speed due to misalignment or the belt drive etc...


I would be interested to know how the output shaft is supported. 800 cpm is still very low for a critical speed.

"Why don't you knock it off with them negative waves? Why don't you dig how beautiful it is out here? Why don't you say something righteous and hopeful for a change?" Oddball, "Kelly's Heros" 1970

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

 

[Strong]

David,

The high vibration at 860 is not 2x, since that would be 850 cpm. Increase spectrum resolution, as necessary, to confirm exact frequency. Since you have a belt drive, have you analyzed belt frequency and harmonics (belt speed) and belt natural frequencies (depends on tension)? Sheave misalignment may cause 2xSS force/vibration. Measure/correct sheave alignment and adjust belt tension to see if vibrations change.
Even though you impacted the rotor or shaft in both horizontal and vertical directions, you may have measured the resonate support structure, but as you say, you did verify crtical speeds by calculation. I would also measure the natural frequency of the motor, since it probably shares the same structure as the belt driven (fan?) shaft.

Walt

 

[Tmoose]

Does belt drive = V-belt? Single, multi?
Is the vibration highly directional, and maybe highest in-line with the belt pull?

I love V-belts, but every v-belt has a fat spot(or 2). Notched V-Belts and other belts with cut or raw sides (

http://mdmetric.com/prod/mitsuboshi/ourpro/images/e_12.jpg)

are usually much more consistent width than wrapped belts (

http://mdmetric.com/prod/mitsuboshi/ourpro/images/e_7.jpg.)

Each time the fat spot engages a sheave/pulley the belt tension increases, trying to draw the Drive and driven components closer together. The effect is not a sinusoidal force event. It is more like a pulse train at 2X belt rotating frequency, so it contains even AND odd harmonics, unlike a true square wave, which mostly has odd harmonics.

http://www.dspguide.com/ch13/4.htm

Stiff shafts mean the drive/driven components move closer together in time with each belt tug.
More typical flexible shafts pitch the shaft in an arc opposite the belt tug (inducing first bending mode). Even modest amounts of sheave offset/overhang give the variation in belt pull plenty of help arcing the shaft.

A vintage tool like a tunable or better yet triggered strobe light can be a mighty handy for helping ID stuff like this.