vibration/alignment/balance

I recently carried out high speed spindle balancing (12 no's). Mass provision was provided on the coupling assembly only. Not on motor and not on spindle.

Max operating speed was 15000 rpm. 1st spindle critical at 8000-8500. Coupling natural frequency 14500-16500 (Wide band frequency) I got this during bump test.

It's a vertical spindle assembly connected to motor (Also in vertical direction) through a coupling. All rigidly mounted.

I was unable to attempt any balancing at 14000 to 15000 as the response was poor and coupling was generating high noise which was unbearable.

I selected 12000 rpm as the balancing speed and finally I was able to bring down the vib levels within limits. But I was surprised after seeing the correction mass phenomena in all these 12 spindles.

I have attached a file for reference. This is just a sample data. The correction mass was first requested at 90 deg (Not exactly at 90, it was like 81 or 103 deg but there was no option so the nearer angle was 90 deg)

I added the first correction mass at 90 deg say 10 gm.

then the mass was requested at 180 deg (Less mass say 5-6 gm) and then in the opposite direction at 270 deg say 3 to 4 gm. Vibs reduced.

since the correction mass was at 180 deg I removed 10 gm at 90 deg and added only 7 gm at 90 deg. The vibs went up.

Puzzled by this I retained 10 gm at 90 and added 3 gm at 180 deg. The vibs reduced further and became within limits.

What's happening here. adding 3 gm weight at 180 deg is same as removing 3 gm at 90 deg. But the spindle vibs went up....??? All 12 spindles are behaving like this....

Seeing this behavior for the first time... any guesses

Nagaraj

Could you post your data for each correction run?
How many planes are you measuring for "balance."
CAn you run the motors solo (uncoupled) ?

Could you describe the test you made for the "coupling nature frequency" ?

If the unbalance (s) are not in the same plane as the correction interesting things can happen.

One thought, if results are not repeatable, there might be be looseness.

quote:

adding 3 gm weight at 180 deg is same as removing 3 gm at 90 deg.

I didn’t follow your scenario, but of course they're not the same the way they're written above. Adding 3 gram at 180 would be roughly the same (*) as removing 3 gram at 0 degrees. I think it's probably a typographical error.... sorry if I misunderstood what you were saying.

[* even in that scenario, adding at 180 vs removing at 0, there is small differece in rotating inertia which could affect dynamics (gyroscopic stiffening, but that would be a very tiny effect.]

quote:

All rigidly mounted.

Is the spindle/motor assembly on a common base, with the base on a slide of some sort?

I'd understand a picture better than words usually.

Nagaraj,
You certainly have a curse of criticals in this machine.
I agree with the others that you should try to balance the spindle uncoupled (or out of the mahcine) first. If you have that data and it is confirmed that the spindles were balanced to an ultra precision level, then good.
Next, try balancing the entire assembly, adding your weight where you are adding it, but do NOT use phase information. Instead, use the four run balance procedure (I know all of you old timers, Arny would say you only need three!!!!).
It is attached here.
Basically, you take your original vibration level and record it. Then, put the same trial weight at each position, one at a time, and record the vibration level. I have used the method, although not this software and it works wonderfully for rotors operating near or at critical. Of course, your spindles are ideal since your stops and starts are not involved or limited, say like balancing a 10,000 hp motor.
The whole reason balance programs get confused at critical is due to the phase shift associated with the critical riding on top of the phase shift due to the trial weight.
Try it! You'll like it!
Ron

It didn't attach. Try it again.
BTW, Thanks to Spintelligent for the paper.

Try again.

Man, this server at this joint is terrible! One more try.

A real 'home brew' article! Same principle.
By the way, it is the author's attempt to call this 'For Dummies'. I personally do NOT consider anyone in my chosen field a dummy.
Regards,
Ron Brook

pete sorry for the typo error... it is adding 3 gm at 270 deg or removing 3 gm at 90 deg...

There is no base here. it's a vertical machining center. Spindle is held in the vertical direction. so vertical vibs are less. max vibs observed in horizontal direction (Least stiffness). We cannot run the spindle solo... it requires drive from the motor....

whatever the correction I had given it is for the total system. I will try to find a snap and send it. photography is prohibitted... will try to arrange soon.

Thanx brook for the valuable paper.

Nagaraj

For those like me who are technically challenged, here is Ron's paper in pdf format.

Google images will retrieve dozens of images of vertical machining centers. I'd get one that looks close and edit it to show measurement planes, etc.

I think running the motor solo would potentially provide some useful info.

Are you balancing with a dummy symmetrical tool in place? A non symetrical tool should be balanced by itself.
If there is a drawbar retaining the tooling there must be a tool in place when running the spindle

I'm still interested in details of the test you made for the "coupling nature frequency"

15,000 cpm resonance corresponds to static deflection of around 0.00015 inches. It takes a stiff structure/system indeed to provide such low deflection supporting much of a spindle and drive motor if as a test gravity was "pointed" in each of the 3 UCS directions.

The four run method may point out a balance solution, but it will take some work to "split" the correction weight among the available 90 degree locations.

Hi Dan,

Bump test data of coupling attached for reference. Yes the balancing was carried out with a symmetrical dummy tool.

Nagaraja

Nagaraj,
Excellent bump test! Yes, you have a critical near operating speed. Go ahead and perform the 4 run balance and you will be fine.
Ron