Dear All,

We are having problem with our ID Fan, it is centerhang fan supported by sleeve bearing, variable speed with maximum speed 700 rpm, the problem was found after outages, vibration at Fan drive end bearing(fixed bearing with thrust) at axial direction was increase and triggered alarm at control room. The vibration dominated at 1x rpm and TWF show sinusoidal pattern and no impacting. The amplitude is highest at highest fan speed. Our mechanics complain that alignment was good an still within tolerance (we can’t shut down the fan to check alignment). Our first action is attach hydraulic jack at bearing support to increase the stiffness of the structure and it did decrease the amplitude. But there is still a question what is the real cause of this problem, is it structural weakness/looseness, bad alignment or flow induced problem .
Can anybody give some clue about this?

Vibration_spectrum_before_hydraulic_jack_attachment.doc (42 Kb, 61 downloads)

A jack inserted is more like shock absorber or hydraulic damper in action, not actually stiffening that much as the hydraulic system is not perfectly stiff (at least in the jacks here). My father once put a thingy he had in his car on a turbine with similar result but the problem was that the jack absorb energy so it was burnt out in a week, so check if yours getting warm. In my view this indicate some kind of resonance, if it´s new something must have changed either in the pedestal mount/structure eg. check for cracks etc. you could also add some mass and see if it makes the same change or the impulse activating the resonance eg. 1xRPM axial in a center hung fan may be what changed. Normal things to check is then: wobbling of the wheel axially or my best guess, bent shaft. Has it stopped while very hot or something like that? Only my 2 cent. Olov

Thank you OLI, but somehow i don't think this is unbalance or rotor bow , since radial vibration are low (0.3 mm/s and 0.2 mm/s). we have similiar fan that exhibit unbalance and vibration at radial are higher than axial direction. We also did not found any crack, but i'll check it again. For fan wheel wobbling, how can i verify that?
Picture of our fan may be can help.

Fan_pic..doc (1,756 Kb, 64 downloads)

A bit more information....
You stated,

quote:

the problem was found after outages

Was there any maintenance done on this machine during outage?
Also, does the free end bearing show similar increase?

Jim P

Again, one point, one spectrum/waveform makes it difficult to help. If there was an alignment performed during the shutdown, I would worry about coupling/shaft axial spacing and assembly of the coupling (if it was disassembled). If it was not disassembled, is the coupling locked up? Some phase readings around the shaft in the axial direction (0/90/180/270 pos.) as well as comparison to axial phase on non-coupled end bearing may show this.

You stated there is no impacting. The time waveform you posted is in velocity, so it would be difficult to see impacting. Acceleration TW and peakvue reading may help.

An excited natural frequency suddenly appearing is not unheard of, but I would look for forcing functions first and eliminate them one by one.

Good luck and let us know what you find.

Pretty large thing, does it have plated blades? Had a case last year where the plating suddenly started to fall off due to some plating problems, just a while after starting operation. It gave very strange behavior and some 1xRPM. Looking at where you put the jack, there are no vertical stiffeners on that side as on the opposite side of the vertical plate? I would scrape some paint with my screwdriver and look for some cracks in the welding, especially on the other side from the jack. Only way to find a wobbly fan as far as I know is to look at it while turning slowly...... Olov

If this was not going on before the outage, what was done during the outage? Ask some questions.

Maybe something was moved inside the bearing which is causing the thrust surface to "push" the bearing pedestal in the axial direction. The thrust disc may be running in a wobbling direction and not impacting but mearly pushing and releasing the thrust surface mount to the fan shaft. Too much runout.

Also have you checked to see if there is any pedestal looseness where it is mounted to the foundation. Was the fan slow rolled to remove the bow caused from being down an extended time or was it started up at max speed from the get-go?

What type coupling does it have, greased? May be, if greased, been turned into a hydraulic jack itself.

jasmine,

1. Was motor IB and fan OB bearing vibration also affected in axial direction and how much?

2. Did you measure phase difference in axial direction between IB and OB bearings?

3. Did you conduct a bump test with equipment running?

4. Has the process changed causing higher
thrust?

5. Has bearing temperature changed after outage?

David

Jasmine,

Is the supply on each end of the fan? If so are there control dampers? Maybe one side is hanging up to give you thrust problems.

As I started reading this, my first thought was the same as Ron's.... imbalanced flow. Is this a double-suction fan with axial flow from both ends, and a common radial discharge? If so, check the position of both inlet dampers, and also be sure all the individual blades are attached. Had a fan like this once that had one drive for both dampers, and one of the arm pins sheared and the dampers on one end were shut. Gave axial vibration. Just something to check.

Thank's guys

ivibr8
Still looking for the outages activities, not sure ,but may be there were bearing replacement. free end bearing does'nt show any increase.

Bill Kilbey
This is attachment for acceleration TW bill. Is there another way to check for coupling lock up beside phase measurement, cause i find it difficult to do this since there is not much space to do it.

OLI
no, it is not plated blades fan. Actually there are vertical stiffener but only two.

Ralph
All bolts are tight and there is no abnormal report during shutdown (we are using turning gear during shutdown to roll the fan)and if it's caused by something that pushing the bearing pedestal, will it increase the bearing temperature?.

David
OB bearing is not affected, i can't measure phase different between IB and OB (distance is too far). At which point should i do bump test? bearing housing?. No temperature change. No process change.

Ron and rusty
i've check damper opening at Control room and it's indicate the same opening 81%, but i am going to ask instrument guys to make some calibration.

So, do you think this fan stand for another year, cause there is no activity we can do until next outage?

thanks

Doc2.doc (36 Kb, 22 downloads)

I have seen this before in other fans that have the same bearing arrangement and normally if it is mainly in the held bearing it is like Ralph suggested there is probably problems with the thrust collar. If you have the oppertunity check the collar for run-out and also make sure the fit is tight were it mounts to the shaft.

After looking at your data, I agree with David and others that this is probably a thrust collar problem. For a direct drive fan running this slow, there is really no other problem that will give you such high axial vibration with so little radial vibration. It is highly unlikely the coupling is locked up, and I don't think it would give this much axial vibration even if it was.

That said, you need to go out and physically verify that the damper position is what is "indicated" in the control room. When you have a vibration problem, you should NEVER trust any "indication" that can be physically verified.

Worked in a powerplant where lots of coal pulverizers were blown up because the air flow "indication" said 100% (venturi section in duct) when the flow was in fact 0%. This occurred over a period of several years and no one ever thought to actually measure the pressures to verify the flow indication.

Yeah never trust the indication. I know I have seen this on ID fans when the inlet damper has become stuck or broken off from the control arm. The indication was off the arm and it read great but, the damper (or part of it) was not moving.

MY EXPERIENCE: I used to be called in to balance ID fans at (NO NAME) company and it was always after they had a long shut down or after the Unit was down for more than a week. After balancing the ID fan, I'd be called back in a month later to re-balance the ID fan...It made me wonder if I was balancing out a bowed shaft from the unit just sitting there idle so long. One way to find out is to balance it and come back to it in a month to see where it is. I think you will find a 180 degree phase shift if this too is happening to you. Check for a bent/bowed shaft or attemtp to balance out the bow - eventually with heat it will straignten. Close monitoring is the key after the first balance attempt.

Regards,

Dan

thanks guys
i will check it at the first opportunity -don't know when and i'll let you know the result.

good day

High axial and low radial 1xSS vibrations I have encountered:

1) Thrust bearing/collar damage
2) Dynamic unbalance (couple)
3) Loose bearing support pedestal

Walt

from da photo it looks like over hung fan....in this case high 1XTs in da axial direction with such a sinusoidal waveform leads to dynamic (couple + static) imbalance problem...although phase difference would be very helpfull to asure its imbalance
if u were talk to an ID fan in a cement factory it simply could be a cout on the impler
although 1XTS peak have skirt which could be from axial resonance...which also explain how does it acted whith the hydraulic jack applied
have u tried to change the speed by increasing or decreasing by 15% of this speed it and see the readings?
also a pump test is very helpfull also u have to turn ur fan off for better results
regards

Hi,

Just observing the time waveform plot pattern posted earlier. see the attachment. why is this so... just curious to know... does this tell us something. normally every cycle should look simillar or may be its a velocity plot is that the reason...???

Madhu

PPT.pps (48 Kb, 12 downloads)

Madhu,

Amplitude modulation is present. A longer time record would help to identify the period of the amplitude variation and show if it is periodic. A possible cause is low frequency pulsation. Another cause of amplitude variation is Beating with another vibration source such a nearby ID or FD fan.

Walt