Well Another Fan problem for ya’ll. It’s been a while for me so you might have to take me to school.
We have a Clinker Cooler 1st Compartment, Belt driven, over hung rotor. We have been seeing an increase in the axial vibration in the trend. It has recently jumped up to .5 IPS with a 1x,2x motor speed dominant freq. in the fan bearings

The FOV spectrum looks very similar to the Axial but only a fraction of the amplitude. The FOH (in line with Belt) looks very different, much more data with Belt frequency present but not high. The FOH amplitude overall is only a fraction of the axial as well.

In all of the spectrums the blade pass in present again at a low amplitude.

With the 1x,2x Axial we thought that we should see some indication in the phase of an alignment issue, belt sheaves, bent shaft, or a cocked bearing. Today we ran a Cross Channel Phase Analysis and found that the phase was as normal machine should be. This is perplexing because we expected something very different.

We can not shut this fan down at the moment we are just trying to get an idea of where to start when we do come down.

Thanks For Your Time

Spectrums.doc (94 Kb, 55 downloads)

One of the first things I do when I see the motor dominant on the fan, and the fan dominant on the motor is to check the alignment. I usually use a tach though instead of cross channel. SHOULD be the same I guess.
It appears you have already went through this exercise.
Take a strobe and check the belt condition. Look to see if the motor base is loose, lockwashers broken, welds broken, etc.
If there is no misalignment, I would tend to think about looseness (1X, 2X) of the motor somewhere.

Out of curiosity, what does the outboard end of the motor look like?

Dave

I have attached Spectrums and trends from MOH and MIH. MIH shows a dominant peak at 1832 Cpm which does not correlate with what we believe are any operating speeds.

We have check the motor base, feet foundation, and the fan bearing, base, foundation and no looseness was found. I still believe that we have some miss alignment in the sheaves, I am just puzzled as to why we did not see it in the phase.

Thanks for the help

MOTOR_OUTBOARD_HORIZONTAL.doc (66 Kb, 23 downloads)

I'm w/ Dave. Fan vib on motor, and motor vib on fan, have usually led to sheave misalignment. At least for me...

For overhung rotors, unbalance results in axial high vibration. Can in any way, deposition of foreign material on fan cause unbalance?

Is the fan outboard bearing the only thing moving in the axial direction of the fan shaft or is the entire pedestal moving? Can you see the shaft or bearing moving on the fan inboard bearing. Looks like something else should be moving besides just the fan outboard bearing. Is the motor mounted on the side of the same pedestal as the fan or on the floor?

Kelly,
High 1xss of motor on fan OB axial indicates Natural frequency. You can map it out during operation by ODS test or shutdown and do impact test.
Walt

Kelly, it appears you have CSI AMS software, please use multiple point spectrum, tag HVA on motor and fan bearings, set the value of "N" to 18.

Post the waterfall once you have it please.

This will help in understanding what is moving and it which direction without phase data.

Gradual increase typically is a long term issue, like sheave/pulley wear, restriction of air flow due to dirty filter/duct work.

If you can show a history waterfall of the axial direction this too will help.
Dave

@1800 rpm .5 ips pk is about 0.005 " pk-pk or TIR. I'd never expect mechanical accuracy better than that on V-belt sheaves, and a typical flexible fan support structure would be willing to respond to belt tugging at 1X motor rpm.

Is the trend at the bottom of the page the fan axial location?
It looks like it was almost 0.55 ips last June.
How confident are you that the data is being taken in the same location?

Interesting point about the effect of runout.

Another way to look at it, if we ASSUME the systems are far below resonance. Ignore the masses, look at the springs. Then the movement occurs according to the springs. The belt spring is much more flexible than the bearing support spring (I imagine). So most of the movement will be accomodated by the belt spring rather than the bearing support spring. And therefore bearing housing displacement resulting from sheave runout would be much less than the sheave runout.

But if we drop the assumption of far below resonance, life gets more complicated and could go either way. But I haven't routinely seen belt driven fan bearing housings at 0.5 ips

A motor sitting on a separate pedestal can not effectively transmit its vibration through the belt in particular in axial direction since belt is too soft.

Ralph's question about motor's location unfortunately was ignored. If located on the same pedestal, then a lot of things could be explained. Also motor's axial vibration data will be helpful.

If high vibration is at motor frequency then resonance should be suspected as has been mentioned above.

P.S.
I never could buy sheave misalignment explanation.

Dave

Daveman - Just to be clear on my position fwiw, I was not ruling out Danny's comments, I was commenting on the magnitude. I certainly do believe that sheave runout and alignment type issues can cause frequencies from one machine (fan/motor) being transmitted to the other machine (motor/fan) due to the action of the belt (not thru the support).