What's new
What's new

Machining large cylinders

jonwinchester

Plastic
Joined
Jul 22, 2016
I should note, I've worked exclusively as a repair machinist using only conventional machines. One of the jobs includes lathe turning large, 112" long 13"+ diameter rollers. A reoccurring issue I see is runout in the middle of the rolls. I'm wondering if anyone has experience or any tips to improve tolerance. The customer requires. 0008 throughout
 
What brand of lathe are they trying to do this on? What kind of shape is the machine in?

How would either of those things affect runout? Diameter for sure, but I'm not seeing the runout even if the machine is a beater.

Aren't whip from too little support / too much speed, imbalance, or internal stress more likely causes of runout?
 
Going to guess it’s a tube, what wall, in so saying anything 13’ long is a sag between centres.
Mark
I'm not sure if sag would be the cause so much as thermal expansion from the cut causing it to bow.

It's a common problem on long shafts, over constrained by the tailstock. Chuck is tight so it can't slip back, and tailstock is rigid. So if the part expands it can't go anywhere and bows.

Supporting the outer end in a live chuck rather than a centre would maybe allow it to expand without bowing as it could move in the chuck a little.
 
It's a common problem on long shafts, over constrained by the tailstock.

Would you expect that on something less than 9 diameters long? Yes it's large, but it's not really very long.

I was also thinking "long" when I mentioned whip above, but that's probably wrong.
 
Last edited:
I've dealt with smaller diameter cylinders of the same length. Honed ID's, OD's not so critical as long as the ID honed ID would drift using a special drift bar. Straightness end to end on the honed ID within .003". Any tighter than that you paid the price for it!
 
Would a toolpost grinder help? I assume that this cylinder is made of some kind of steel, not necessarily all that hard, but one can still grind it.

If the grinder contact point is at the same height as the rotation axis, the sag will not cause a problem.
 
Last edited:
The customer requires. 0008 throughout

So what is the runout you are able to achieve?

If the roller is constrained in the length direction, it's not too hard to see heat expansion bow the roller during machining. If the wall is not too thick, I'll bet you can see movement just pushing in the middle of the roll. Or especially letting sunlight warm up only one side.

First question would be "Does the customer have any means of checking the dimensions"............the answer is most likely NO.

Not really sure this makes a difference. Either it's. 0008 TIR or it's not, even if the customer can't check it.
 
Would you expect that on something less than 9 diameters long? Yes it's large, but it's not really very long.

Yes, absolutely. The length/diameter ratio is largely irrelevant except for how much it's likely to move vs. the machine it's mounted in. Small parts in very rigid machines exhibit the same behaviour.

Remember that OP is talking (relatively) very small numbers...
 
Put it in a centreless if you can then you'll really see how bad the straightness is!
 
I’m trying to picture the loading and unloading and laughing a lot
I ran centerless grinders in production for over twelve years before changing careers. We had all the books from Cincinnati and you would be amazed at some of the rigs and fixtures they showed for infeed and throughfeed grinding. With a little imagination, this would be quite doable.
 








 
Back
Top