Part out of round but only on OD while using ID pie jaws

Scratching my head on this one trying to figure out what's going on and hoping one of you might be able to provide some insight.

I've got lathe job with a thin walled aluminum part that I am taking a finish pass on the OD that I am noticing 0.005" out of round afterwards but only on the outside on the last inch and half of the part on the chuck side. ID is still round.


The part: 4.5" OD x 11" 6061 Aluminum 0.100" wall thickness

The setup: ID pie jaws grabbing 3.5" deep on an 8" 3 jaw chuck on a Mori NL2000.

The process: We rough turn the OD leaving 0.020" finish stock, then we OD grip the part with 4" pie jaws to drill and bore the ID to size. Last op is ID gripping the part with pie jaws to finish the OD.

More details on the problem: I check roundness after finishing the ID and everything looks good <0.0004" all around. I've got my clamp pressure dialed as low as it will go and if I clamp, roundness still looks good on the machine. I take the part out before machining, you got it, roundness still looks good. However, after I take a 0.005" skim pass on the part is when all hell breaks loose. Once I take the part out and check it on the CMM, it has the classic trilobal shape you'd expect from jaw distortion but here's the kicker, its only on the OD. ID is still perfectly round. The distortion is only present on the last inch or two of the part on the chuck side, everything else is round. I used dykem to verify the part is properly seated on the jaws and making contact across its entire length & width.


Questions I can't find an answer to:
1)If its jaw distortion, why can't I see it from just clamping and unclamping the part? (only shows up after I take a skim cut)
2)The ID pie jaws probably grip close to 90% of the ID with minimal pressure, how can they be distorting the part?
3)Why is it only affecting the OD and not ID?
4)If its something else, say tool deflection, why it is then that if I spin an indicator on the part after cutting while still clamped, I get only a few tenths of runout? This tells me the machine is cutting round.

Any help is greatly appreciated.

Indicate the part after final machining with the part still clamped, is the part round at that stage. if so then it is most likely distortion due to the ID gripping and could be related to the spindle speed and centrifugal force of the jaws trying to expand outward while the spindle is rotating, or the gripping force.The issue with only the last inch or two towards the chuck could be from jaw lift when clamping. I am assuming the jaws were turned while expanded out on a ring and the same chucking pressure,if so this would minimize any possible taper on the jaws gripping diameter. Possibly try under cutting the last inch or two of the jaws to a smaller diameter so this doesn't influence the end of the part.

cornerRad said:

Questions I can't find an answer to:
1)If its jaw distortion, why can't I see it from just clamping and unclamping the part? (only shows up after I take a skim cut)

Click to expand...

Because the part flexes back into shape.

cornerRad said:

Questions I can't find an answer to:
2)The ID pie jaws probably grip close to 90% of the ID with minimal pressure, how can they be distorting the part?

Click to expand...

Because the part has a .100" wall thickness.

cornerRad said:

Questions I can't find an answer to:
3)Why is it only affecting the OD and not ID?

Click to expand...

Because you're only cutting the OD while it's clamped and not the I.D. If you were to hold on to the O.D. and cut the I.D. you'd see that deformation also.

cornerRad said:

Questions I can't find an answer to:
4)why it is then that if I spin an indicator on the part after cutting while still clamped, I get only a few tenths of runout?

Click to expand...

Because it's still clamped in the jaws and holding it's shape.




My advice would be to put them on an arbor and use a cap with a bolt. This way there is zero deformation happening and the O.D. will remain round.

It doesn't take much to flex a part like that, your chuck pressure would have to be extremely low. Lower than what most people would call safe.

Tom said:

Indicate the part after final machining with the part still clamped, is the part round at that stage. if so then it is most likely distortion due to the ID gripping and could be related to the spindle speed and centrifugal force of the jaws trying to expand outward while the spindle is rotating, or the gripping force.The issue with only the last inch or two towards the chuck could be from jaw lift when clamping. I am assuming the jaws were turned while expanded out on a ring and the same chucking pressure,if so this would minimize any possible taper on the jaws gripping diameter. Possibly try under cutting the last inch or two of the jaws to a smaller diameter so this doesn't influence the end of the part.

Click to expand...

Indicating the part after machining while clamped shows its round. Didn't consider spindle speed a factor before but I will try another test cut at half the rpm to see. Right now it cuts at only 1300 rpm.

I did cut almost a .0003"/in back taper in the ID jaws to counter act jaw lift and verified with dykem that the jaws are making full contact with the part.

Mtndew said:

Because the part flexes back into shape.

My advice would be to put them on an arbor and use a cap with a bolt. This way there is zero deformation happening and the O.D. will remain round.

It doesn't take much to flex a part like that, your chuck pressure would have to be extremely low. Lower than what most people would call safe.

Click to expand...

Measuring the part before clamping shows it round. Measuring it before cutting, while clamped shows it round. If the jaws are distorting it, I wouldn't think 0.005" removal is enough to cause the jaws to start flexing the part. I'm also perplexed that pie jaws which grab ~90% of the diameter can cause such a trilobal shape. I could understand there being taper from the jaws deforming the part but I thought the whole purpose behind pie jaws was to prevent uneven distortion.

I wish mandrel and a cap was an option but the part does not have a thru hole.

P.S. clamp pressure right now is 0.3 MPa which is as low as it will go and still move the jaws.

Its in the Kitagawa literature, force lost with speed. A B208 looses 5kN gripping force from 0 to 1500 rpm with std top jaws. Heavier jaws, more lost.

Thats is how much force is pushing on the tube ID.
Keeping the pressure down doesnt help because as you spin it, it grips harder. Even if the pressure was zero. They make internally counterbalanced chucks for this (Powerhold) but I think it is more to retain OD grip force at speed. But either way.

Mtndew has the right idea.

GENERALDISARRAY said:

Its in the Kitagawa literature, force lost with speed. A B208 looses 5kN gripping force from 0 to 1500 rpm with std top jaws. Heavier jaws, more lost.

Thats is how much force is pushing on the tube ID.
Keeping the pressure down doesnt help because as you spin it, it grips harder. Even if the pressure was zero. They make internally counterbalanced chucks for this (Powerhold) but I think it is more to retain OD grip force at speed. But either way.

Mtndew has the right idea.

Click to expand...

Well I think we're onto something here. I just ran a cut at 500 rpm (was 1300) and my roundness improved from ~0.005 to 0.003. This would also explain why the part still looks round after clamping cause there's no way I'm spinning it that fast by hand.

Looks like I'm off to try turtle speed to see if I can make it even better.

You can also try cutting away any unneeded portion of your jaws to reduce the effects of centrifugal force.