What's new
What's new

Pulley Runout, Vibration, & Belt Choice

catalytic

Cast Iron
Joined
Feb 4, 2010
Location
Boston, Los Angeles, and Cleveland
I've chased the vibration in a 20" upright bandsaw to the belt & pulleys. I have three questions:

1. How much radial & axial runout is acceptable on machine pulleys in order to get the machine to pass the nickel test? Belt design books suggest the following:

Radial Runout Guideline:
Up through 10" Outside Diameter = 0.010"
For each additional inch = + 0.0005"

Axial Runout Guideline:
Up through 5" Outside Diameter = 0.005"
For each additional inch = + 0.001"
Pulley-Runout.gif


However, it is unclear to me whether pulleys that fall within these guidelines will have imperceptible vibration, or whether this is still the main source of my vibration problem.

My 4" motor pulley measures 0.003 radial & 0.002 axial. Balance not checked.
My 8.5" lower wheel pulley measures 0.005 radial and 0.006 axial. This pulley is balanced.

2. If this runout is an issue, do you guys recommend having them trued up in a shop & balanced, or just get new pulleys?

3. What belt would you recommend to mitigate vibration? I would prefer to avoid link & powertwist belts. I have heard that a Gates cogged belt may be the ticket, but I haven't tried one.

Thanks,
Adam
 
I don't think there are any good rules of thumb. It depends on the specific setup and how fussy you are. The radial runout as shown shouldn't mean anything, since the belt never touches there. IMO, .005" for the axial runout is the maximum I'd ever want to see on the smaller pulley and what's more important is that the sides are symmetric; you don't want the belt moving up and down, though it would take a pretty crappy pulley to do that. Look at it! Assembled sheet metal pulleys could suffer from that. Very often the Fenner Powertwist belts will reduce vibration and other than cost I don't know why you want to avoid them. Look for the belt series with the slots on the ID; sometimes they run smoother. Are the parts balanced? Is the belt long enough to flap? If you want to get more sophisticated, you could get somebody to put a scope and vibration sensor on it. That would tell you if the frequency of vibration matches anything in particular. Sometimes you can tell just by listening and feeling.
 
If you are running a single phase motor, you will have uneven pulses of power compared to a 3 phase... Add in just the right pulley diameters, and you could get some harmonics going...

The motor might be out of balance, far more rotating weight there, than the pulleys.

Belt is easiest to replace, I have seen many, very lumpy/poor quality belts out there...
 
1. How much radial & axial runout is acceptable on machine pulleys in order to get the machine to pass the nickel test?

Nobody will be able to provide you with a simple answer to this question without knowing the thousands of variables that define the machine in question. Vibrations are strange things; the mere word sends many engineers running for the hills.

Analysis on a simple shaft is fairly straightforward, but the moment that those vibrations begin to propagate through a machine the problem grows vastly.

2. If this runout is an issue, do you guys recommend having them trued up in a shop & balanced, or just get new pulleys?

This is certainly one way of improving an imbalanced shaft. It may or may not be right for you though. What speed is this shaft running? What kinds of loads are present? What kind of bearings constrain it?

3. What belt would you recommend to mitigate vibration? I would prefer to avoid link & powertwist belts. I have heard that a Gates cogged belt may be the ticket, but I haven't tried one.

A cogged belt would involve all new pulleys. Bandsaws have been running fine with V-belts for decades. If it came to a choice of all-new hardware I'd go with new V-belts and corresponding pulleys before I went to a cogged belt in that application; you'd probably come out money-ahead as well.

Henry
 
Thanks for the replies. I should have mentioned that the motor is 3-phase and the machine does not vibrate when I remove the belt and run it -- I'm 90% sure that the vibration is either in my belt or pulleys. I'm planing to change out the belt -- it was a standard v-belt which had taken set due to previously misaligned pulleys. While I have things apart, I thought it would be a good time to work on the pulleys as well.

I'm running it with a VFD, and while vibration is certainly more noticeable at certain frequencies, I haven't been able to connect this to the RPM's of any specific part. Is there a better way to do this?

What's the best way to measure the symetricality of pulley sides? Besides checking runout on both sides and seeing if they match in the same places around the perimeter, I'm not sure how to do this. If this is the problem, is this something that could be corrected without too much fuss? The pulleys are both cast iron discs, and the smaller one is an H-Bushing interchangeable bore type.
 
I'm planing to change out the belt -- it was a standard v-belt which had taken set due to previously misaligned pulleys. While I have things apart, I thought it would be a good time to work on the pulleys as well.

If you had a bad belt, that is probably the whole problem. In any event, a good diagnostic rule things is change one thing at a time. If new belt solves problem, every minute you spend futzing with pulleys is a complete waste, and risks misleading you.
 
I have had great sucess with belts that are cut, not molded and have the notched inner portion, Gates and others.

95% of the time using cast iron/steel sheeves with this setup.
 








 
Back
Top