belt bump test - harmonics of fundamental sort of

[electricpete]

On another site, the results of a bump test of a vee-belt (not running) were posted as shown here:

http://home.comcast.net/~electricpete/eng-tips/Belt_bump_test.doc

From simple theory of strings under tension, we expect that the resonant frequencies will be a series of harmonics F1, 2*F1, 3*F1, 4*F1 etc (similar to the overtone series of a guitar string).

The first two peaks (F1, 2*F1) match that pattern pretty well. Then the third and four peaks appear to drop farther and farther below the harmonic frequencies we expect from simple theory.

Does anyone have any explanation for this?

I suspect there is some other effect that comes into play as we go to higher mode frequencies and/or higher order mode shapes. Perhaps the viscoelectric damping acts to decrease the frequency of the higher modes (although I would expect the magnitudes would decrease much more before we saw this noticeable decrease in frequency).









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

 

[BobM3]

More damping at the higher frequencies?

 

[electricpete]

It just occurred to me that the "harmonic cursor" labels may be misleading. This could be close to an exact harmonic series if the frequencies were:
640
1280
1920
2560
I will try to get the person who generated this plot to give me the exact frequencies.

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

 

[btrueblood]

Dunno about viscoelectric damping...you say the unit wasn't running?

Damping can increase at higher frequencies in assemblies due to greater losses across contact faces (coulomb damping or friction damping). There's a lot of frictional damping that will occur in a V-belt drive, I'd think.

 

[electricpete]

It just occurred to me that the "harmonic cursor" labels may be misleading. This could be close to an exact harmonic series if the frequencies were:
640
1280
1920
2560
I will try to get the person who generated this plot to give me the exact frequencies.

Nope, that's wrong. The second frequency is definitely more than 1250 and the fourth is definitely less than 2400, so it definitely deviates from an exact harmonic series.

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

 

[electricpete]

Correction to the correction (where's that darned edit button?)

2400 should have been 2500 (obviously?).
Again the conclusion is that it is not a harmonic series.

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

 

[MikeyP]

Perhaps the belt has some rotary inertia which becomes apparent at higher frequencies (more wavelengths along the belt = greater gradient along the belt = more rotation of a differential element along the belt)? Damping does look like it is increasing (though it's difficult to tell on a linear frequency axis) which may be enough to shift the frequency.

I presume that your model assumes that boundary the boundary condition is that the belt is fixed at the point where it meets the pully. This may also become less valid as frequency increases.


M

--
Dr Michael F Platten