Hi guys, wondering if someone can help me out here. One of our cooling tower fans has had a large peak develop at around 72Hz, which works out to be 4x blade pass. Any ideas on what could have changed to cause this?

Shane,

IMO resonance is a possibility.

Most of the cooling tower fans I survey run at two speeds or are VFD. If that's the case, and you have the chance, change the fan speed to discard the possibily of a resonance.

hope this helps

Would you be more specific, how many blades the fan has? Is it a variable or 2speed? Is 4x component is dominant at other operating speed? Did you consider the phase of this ampliudte?

If this 4x of blade passing as you say then this value is devided by 4 then by # of blades to get the rotating speed. may be 4.5HZ

Is this fan slow to this dgree??

At first glance, I guess your fan is supported on a 2 crossed concrete members and you have a unbalnce problem in a certain blade.

Shane,
It is possible the hardware holding blades became loose or is broken and blades are vibrating loose. I would do blade inspection ASAP.

Guys, thanks for the responses so far. Unfortanatly I dont have the equipent to check the phase as Im currently using pretty basic equipment. Fan output speed is 2.2Hz, 8 blades. Alex J, that thought has crossed my mind as we have had a loose blade on a previous fan.

quote:

Originally posted by k.kahtib - Gaza:
Would you be more specific, how many blades the fan has? Is it a variable or 2speed? Is 4x component is dominant at other operating speed? Did you consider the phase of this ampliudte?

If this 4x of blade passing as you say then this value is devided by 4 then by # of blades to get the rotating speed. may be 4.5HZ

Is this fan slow to this dgree??

At first glance, I guess your fan is supported on a 2 crossed concrete members and you have a unbalnce problem in a certain blade.

Fan is a fixed speed, and its supported by the wooden cooling tower structure (obviously with a steel sub structure)

Shane is this fan belt drive or through a gearbox? 2.2 Hz sounds pretty slow.

quote:

Originally posted by Jonesy:
Shane is this fan belt drive or through a gearbox? 2.2 Hz sounds pretty slow.

gearbox driven. The fan blade diameter is almost 7 meters, so thats why its so slow. Its setup with a motor driving through a 3 1/2 meter driveshaft, gearbox is a triple reduction 11.18:1 ratio

Blade pass is 8*2.2 = 17.6 hz.
4* bladepass is 70.4hz ~ your large peak.

But there is a spacing of approx 8hz repeated through the pattern ~ 4x running speed. For example those large peaks on each side of 72hz.

Is motor speed 25hz? And As asked above - Belt or gearbox and number of teeth?
What type couplings?
What type bearings (rolling or sleeve and if rolling what part number)
Also if you can use a peak label feature to better define the peaks (perhaps with higher resolution spectrum also), it would help to define the pattern rather than using eyeball guesswork.

2.2 X 11.18 = 24.59 Hz motor shaft speed. The 72 Hz peak is near 3X motor speed. This could be a coupling or alignment issue. Where did you take this measurement, on the motor or gearbox or? A high resolution spectrum would help a lot.

ShaneWa,

If the peak is at 4x blade pass and there are no loosen or eroded blades you might check the angle of the blades. Very often these fans come with adjustable blade angle, and it may be causing some turbulence in the air flow, and the correspondent peak vibration. Also if the fan has shutters above it check if they are not closed or nearly closed.

This message has been edited. Last edited by: mechanical pawn,

I had something simular this last summer. We are running Fiberglass blades on Two of our towers. Took me a while to figure that out sad to say. At the tips of two of my blades it was delaminating which caused an imbalance. Sanded it down and epoxied the tips of the blade rebalance good to go now.

Dear sirs, If we think deeply in one blade unbalnce then 2.2HZ and multiples are revealed- ignore that.4X of blade passing frequency points out at the whole blades share in this component.I still doupt that the 4x of passing f is resulting from the X structure of the fan as the blades passing the 4 links consquently.

this means a structural problem, may be loosnes in gear supporting. Please re-check the regidity of your system.
I our power plant, we re tight the blades and gear periodicaly and find loosness. It is a howden- holland 7 meters 7 blades ones.

ShaneWA,

IMO there is rather a structural resonance close to 72 HZ then a blade issue. Also what is spectral resolution and sidebands spacing?

Dears David and ShaneWA,
The spectrum advice to check out the fan as a hole. After structure you have to check the upper bearing of the G.box, any loosnes is critical. You have to check out, May you do it by a dial gauge to measure the misalignment due to bearing loosnes or structure destortion.

Any way these fans create a pressrue field above the blades, this pressure field tends to bend a side the whole fan from the vertical.

Well, just a quick follow up. Fan tripped out last night on high motor amps. Had a quick run up this morning and we could physically see the fan wobbling from side to side, so at this stage it looks like the output bearings are gone.