Dear All,

We have AHU vibration problem at our customer site. This AHU is running in paint shop.

All the details & vibration data is in the document.

Kindly post your views on this to solve the problem.

Advance Thanks for your suggessions

Thanks & regards
Mahesh

AHU_Vibration_Data_-_Posting.doc (168 Kb, 84 downloads)

I haven't see one like that in terms of base construction (separate) or vibration pattern. Lack of fan 1x vib showing on motor suggests it's not sheave misalignment I think.

Highest vertical in-phase at both fan bearings. Second highest axial, 90 degrees from vertical. Would almost suggest movement rotating in the V/A plane (vs normal rotating in the V/H plane). Don't know what that would mean.

Since the base seems to have gotten a lot of attention in the past, some investigatory checks would be:

1- Check for vibration pattern and level on the base.
2- look for abrupt change in vibration and phase on the base. Check for any obvious cracks.
3 - Bump test to try to excite resonance in vertical, axial, and angular in the vert/axial plane (sort of the way you have your DE and NDE arrows labeled).
4 - Careful inspection of spring mounts. Check the effect of wedging something underneath the base to change stiffness.
5-Strobe the shaft.

Was the fan in a normal configuration during balance attempts dampers open and no access ports open?

Just some wild guesses.

This message has been edited. Last edited by: electricpete,

Electricpete,

You are correct. Blower being balanced under normal configuration. No access port open. But spring dampers were present under the blower.

I have seen very similar data in the paint shop fans I monitor; excessive vibration on the fan verticals and axial measurement may indicate poor mountings
Follow Epete's advice he is right on target
Inspect the vibration isolators visually if cap type check that similar gaps between top and bottom. If they are rubber are they squashed or warped.
When and when did they weld the l angles?
Mike.

MEAS_REP.xls (26 Kb, 34 downloads)

Run a coast down test and acquire amplitude/phase vs. RPM data to construct a Bode plot. It will give you an idea where you are in reference to resonance. Monitoring phase stability at running speed will also help.

In your attached data sheet, you said this started all of a sudden. What is "all of a sudden", over night, a few days, etc.?
I would check first for something obvious on, like the guys have siad, mounting, etc. What about the fan blades themselves, nothing slung off has it, like an old weight?

Why are all the angles showing a negative number?

I had a configuration like this, but in my case the fan was hard mounted and the motor was on rubber isolaters.
There turned out to be a resonance at belt frequency.
Spring Isolators can change over time, and your fan speed is fairly low.. (534CPM) Which can be in the range that isolators operate. It wouldn't suprize me if your spring isolators are resonant at fan turning speed.
The only problem with the resonance theory is that you have 2 dimensions of high amplitude which is not chacteristic of resonance.. but since springs are generally more flexible than rigid structures in all dimensions, I wouldn't discount it either.. it is easy enough to test.

Assuming pulley alignment is good, so other than a "two Plane " resonance, what could acount for the Axial vibration?
I am assuming the suction is single sided here..

Has the flow changed to this fan recently.. clogged filters, flow dampers adjusted ect?
Are the motor loads monitored, and if so has the motor load escalated with this problem?

Just some ideas.
instinct says resonance.. do a bump test asap.

As per my knowledge, Vibration levels changed over night.

I have one more idea, Is this give some result pl. comment.....

Speed of this AHU can be changes by 1 to 2 Hz. Shell I try once again balancing by changing the speed? Does it give some response? If it gives some response to balancing, I will do fine balancing. Then will change to original speed. Is this gives us some results? Just I thought. Comments pl.

Hi Mahesh,

I was wondering, if the trial weight had no effect on the amplitude and phase, how did the program come up with the correction weight and angle it gave you? Most programs will not give an accurate answer with no change from the reference numbers and some will not even attempt to give an answer at all.

Which direction were you using for your transducer and tach light?

You say you can change the speed by 60 to 120 rpms, what are the readings when you do this, lower or higher?

Have you done a coast down for amplitude vs phase and amplitude vs speed in the worse direction, vertical?

From your drawing this looks like an overhung fan, but is it a squirrel cage type or large blades? How many blades? Starting over night, I would check the blades carefully. Did the amplitude and phase return back to its original numbers after the weights were removed? Can you take the belts off and see if there is a huge static"roll" in the free state?

Can you "block" off the fan spring mounts to see if it changes anything? The phases are close to being in phase in each individual direction, and is vibrating in the direction of least resistance (vertical and axial),so the horizontal stiffness is enough to restrain the cause of the vibration. The spring or springs may have collapsed suddenly. Check this close.

After checking all the springs and blades, I would do a bump test in the vertical and axial directions, checking the vertical position while bumping in the axial direction as well as while bumping in the vertical direction and also checking the axial position while bumping in the vertical and axial directions.

Mahesh
Ralph has offered some good advice the fans I monitored were large blades, at the time I couldn't carry out coast down or phase measurements
I did however block off the fan spring mounts and it the vibration levels altered significantly

As Ralph says the springs may have collapsed, check out the attachment, I was trying to explain to a client that there were poor vibration isolators on a fan in this instance the motor had the worst levels of vibration.
It was very long winded to collect the data and show it if I could have taken phase and coast down measurements I could have proved that there was a mounting problem a lot easier.

The isolators were Cap type and visually I could see that there was a different gap between virtually all of them "worth a look it may give you some help".
Eventually I convinced my client to fit new vibration isolators and balance where necessary but as there were many fan units showing these symptoms it took time to put them all right.

Initially this seemed to solve the problem but after a while levels started creeping up again. More importantly further investigation "two three years later" revealed that the fans were splitting and had to be replaced.
Apparently the impellor material was not correct and the paint shop chemicals that were present were gradually rotting the blades.

Just some info hope it helps
Mike.

ANOTHER_LAST_MEAS_REP.xls (32 Kb, 14 downloads)

If you have not already done this, I would check all the blades for cracking. That seems to be the most logical since this happened suddenly and the fan does not respond to balancing.

mcdm,

quote:

if I could have taken phase and coast down measurements I could have proved that there was a mounting problem a lot easier.

This is interesting for me...I'm new to the subject of balancing. Could you please explain why, or explain your statement. Thank you.

If I could have taken phase I would have looked at the phase relationships between the concrete floor and base plate, the skid and then the motor.
90˚ to 180˚ difference on Vertical measurements a sure sign of looseness also soft foot may have shown different phase measurement from the other measured points.

As for coast down I still cannot take this as the machine cannot be turned off, but I would have tried to see if I could identify and confirm the presence of resonance in a machine or structure.

I would have been happy enough to report that there was looseness present shown some simple phase figures rather than having to collect all the data and manually type up a report and drawing

Well, the fact that the vibration changed overnight makes the "isolators are resonant" theory suspect.

Others have suggested a cracked or loose blade.. I have seen this happen also.. It surely can cause this kind of vibration on a spring isolated system.
The problem may be hard to find though, check carefully!

mcdm,

Thanks for the explanation.