Drill / Mill Spindle-to-Pulley Spline Fit

[extropic]

well, there are indeed two positions of the pulley relative to the spindle that all but eliminate the chatter....and they are clocked 180 degrees from each other.

I *think* this confirms the hypothesis that the tolerance between splines is the culprit. happy to take objections if you have them

Did this solution turn out to be a mirage?

 

[jdeng]

good question. It was definitely the best of the permutations so far and maybe I should stop striving for better than that.

 

[extropic]

It's difficult to accept, without accurate dimensional inspection, that the spindle splines and/or the pulley splines are irregularly spaced around their circumferences.
However, I imagine such irregularity would make plenty of noise.

One other test, that I don't remember reading about, is to run the pulley with the spindle removed. If that test proves satisfactorily quiet (each belt in turn) it eliminates significant elements from the issue. If the test produces unsatisfactory noise (which seem unlikely to me) it helps focus your search.

 

[Ischgl99]

You definitely don’t want to run it crowbar tight, but since it is the same at all tensions, we can eliminate too tight causing it to flex. Have you checked the spline section on the spindle to see if they are bent?

Extropic posted a good suggestion, try running without the spindle in and see what happens. Also his comment about the splines not being regularly spaced. That would be good to check both the spindle and pulley splines, it could be that they are off enough that it causes binding during rotation.

 

[jdeng]

great suggestions.

I ran the pulley without the spindle installed. No chatter but the pulley still gets hot (140F) after 10 minutes at full speed. This is the new pulley, new bearing, and new belt.

Included photos below with some dimensions of the splines. Note that I truncated a sig fig from the measurements because the teeth have draft and the broaches are not cut concentric with the bores. ie, it's difficult to pull precision measurements from my calipers. Accurate but not precise. I also can't get my calipers into the pulley to measure the width of the spline-tooth.

Last photo is of the splines on the spindle shaft for reference.

edit: to add that the dimensions are consistent (within my precision) around the circumference.

 

[Ischgl99]

I’m not seeing anything wrong with the splines.

140f on a bearing is fine. You wouldn’t want it that hot with spindle bearings with preload, but a deep groove ball bearing in that application can handle that no problem.

Mark brought up something a few days ago that I missed, how is the fit of the quill to the housing? If the pulley bearing housing has 0.25mm clearance, maybe there is excessive clearance there too. I wouldn’t think that would be the issue with it happening with one speed belt and not the other, but you never know.

Since you can’t reach full speed, I’m thinking this might be a drive problem. If the drive is inducing pulsations due to a bad phase or some other electrical problem like that, it could explain why you can’t reach full speed. Does this have a DC or AC motor?

 

[extropic]

Lots of information in those pictures.

Let's see if I can communicate my next suggestion. I'm interested if you can perceive any nonuniformity in the spacing of the splines around the circumference of the spindle shaft. For the purposes of description, in your picture above, lets agree that the uppermost spline tooth #1. The next tooth, clockwise, is #2 and so on. Please measure the dimensions of the outside point-to-point spanning teeth 1 and 2, 2 and 3, 3 and 4, 4 and 5, 5 and 6, 6 and 1.

25mm ID mates with 24.4 mm OD. Diametral clearance is .6 mm = .0236".
I think you need to get some shim stock and do your very best to insure that the pully and spindle are concentric at assembly.

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

 

[jdeng]

I very meticulously slipped some custom-cut, 5mil shim strips down each of the splines and managed to get the spindle assembled to the pulley. It has indeed eliminated the rattle.

But, of course, my downfeed won't work. Pulling the strips even after the pulley-assembly is fastened down results in the rattle re-emerging. Still, this is progress, methinks.

I keep coming back to the question of how they assembled this at the factory? trial and error? Seems crazy but what do I know. There's still always the possibility that I messed something up during the refurb too.

 

[jdeng]

I really appreciate the amount of thought y'all have put into troubleshooting this with me. I need to take a break for a few days to let my mind process everything and to cool my temper a bit. Will report back.

 

[extropic]

I very meticulously slipped some custom-cut, 5mil shim strips down each of the splines and managed to get the spindle assembled to the pulley. It has indeed eliminated the rattle.

But, of course, my downfeed won't work. Pulling the strips even after the pulley-assembly is fastened down results in the rattle re-emerging. Still, this is progress, methinks.

I keep coming back to the question of how they assembled this at the factory? trial and error? Seems crazy but what do I know. There's still always the possibility that I messed something up during the refurb too.

So, can we consider the source of the noise definitively isolated to the splined interface between the pulley and the spindle?
Maybe we are not certain of the "why", but it sounds like you've zeroed into the "where".

It's seems like too much work for me to pursue implementing shims that will prevent the rattle, but still allow the down feed to function.
It's your machine and what else would you have to work on if the problem was solved?