I'm doing a bunch of Robinson Tubeaxial fans and am wondering if anyone has a better method of getting vibration data on the fan bearings. I am trying to locate the bearing pedestal in the duct and taking readings off of it. I have had pretty good luck with smaller Chicago vane axial fans, but the construction of the bearing mounts allows for a nice transfer and I'm not as comfortable with these bigger ones.

Any suggestions?

Thanks,

Danny

btw, Anybody know what happened to Jim Shinn that used to post on the old site and works (or worked) for Robinson?

On a number of mine I have permanetly mounted transducer; some I use an extension pole on. I do monitor several hundred fans - what are you seeing?

The last info I have on him is he is working for HowdenBuffalo Fans. This is about 2 years ago. I have a number for him also. If you would like it email me at mike-collier at txgenco.com

Sam,

How do you use the extension pole? Can yo reach past the v-belt drive to get to the pillow block bearings? I don't think I can do that here because I would have to get on top of the duct and some of it might collapse.

Mostly I'm seeing running speed harmonics with some impacting at running speed. I figure that there is at least some imbalance and possibly some broken steel in some of the pedestals or vanes. These thing are probably 20 or 30 years old and have not had the attention they should have.

Thanks,

Danny

Danny,

I think you can get reasonable measurements at low frequencies by measuring on the outside, but I doubt you get very good transfer of high frequency noise, such as from deteriorated bearings. I've often taken measurements on the flanges at each end when I couldn't tell what the mounting arrangement was or the fan was insulated.

We usually recommend installing a radial accelerometer on each bearing, cabled to a termination box mounted on the outside. This will detect balance and bearing problems, which pretty much covers what can go wrong on these fans. This is one of the few aplications where I'd recommend an integral cable accelerometer - since there's no way to reach it when it's in service, the connector is just an added expense.

The cost to install accelerometers need not be extreme:

2 each AC-140-2A/010Z low cost accelerometers: $188
2 each MH101-1B integral stud adhesive base: $15
1 Spintelligent Labs 2 channel Economy Term Box: $69

TOTAL: $272

This has to be weighed against potential repair costs, shutdowns required because of failure and potential damage that could result from a failure. A few recent installations for us were in supply fans to operating rooms...clearly not a place for flying parts to fall from the ceiling.

Jon
http://www.spintelligentlabs.com

Jon,

These guys have new management that is much more inclined to do things like that, but we have just brought them into the 20th century, I'm not sure they are ready to go all the way yet.

I'm going to suggest it tomorrow in a meeting. We'll see how it works out.

Thanks,

Danny

We have over a hundred, similar to Robinson. Mounted inexpensive accels on them, drilled a hole in the barrel, put in a rubber grommet with a strain relief, and ran the cable down to an access point. Saved us a fortune, made testing them a lot easier and more accurate, and cost us less than $100 each.

On those configured so that it is trouble to get to we have permanently mounted transducers. Seems we have various configurations; some too far off the ground to reach safely.

I got a good look at one of these today and my prior assessment that it was about to fall apart was correct. It threw a blade last night. Every blade was about to fall off with far more broken weld than good. They are going to save the bearings for me so that I can guage the condition of the others as compared to a known.

This could prove very interesting if things keep flying about.