What's new
What's new

Strange lathe vibration at single point on part

cornerRad

Plastic
Joined
May 7, 2021
I've been loosing a lot of sleep over a new issue I've never seen before and have run out of ideas to try.

I have a 4in dia x 12in piece of 17-4 held in the sub spindle of the lathe with 4in tall steel soft jaws. 90% of the machining is done on the main side with 0 issues. I can face, drill, and bore on the sub with 0 issues.

However turning the last little bit of the OD has been an absolute nightmare. I only need to cleanup the last 0.6in on the end. The cut starts fine but once it's halfway through the cut an incredibly loud, low pitch vibration starts what pulls the part out of the chuck, ruins the finish and the part.

Chatter is typically a high pitch whine/ringing which this is not so I don't think it's that.
It only happens in this small band on the OD, the tool faces fine and starts fine so I don't think it's the tool or lack of rigidity.
I've tried half a dozen different insert geometries and anywhere from .01-.09 DOC, .005-.012 IPR, and 400-800sfm with 0 differences. They all produce the same result.

I have exhausted all possibilities that I can think of and have been counting the new grey hairs by the hour.

Does anyone have some suggestions on what might be causing this and how to fix it? (I've got 1000 of these to make)
 

Attachments

  • PXL_20231027_055708734.jpg
    PXL_20231027_055708734.jpg
    1.8 MB · Views: 110
Longer jaws, and undercut them. Pretty sure that much overhang is the issue. You might also try undercutting the back 1/3 of the jaws a tiny bit. Could be the jaws are deflecting at the outer end and not clamping the part. That setup just does not look very solid!
yeah, this is where I'd start too.
As an experiment take one of your scrapped parts and wrap a 1/2" wide strip of copper or sand paper around it right at the end of the jaws to help the back or "heels" of the jaws contact your part. The take some test cuts. I find longer jaws like this deflect. Remember as surface area goes up psi goes down. You may have some big jaws there but your actual contact patch could be small due to deflection.

1698422620557.png
 
Can you get a piece of .001 shim in between your part and the jaws in the back? Like KingBob said longer jaws tend to deflect especially if you cut them at one pressure and clamp the workpiece at a higher pressure. I would check for that and if they are you can cut a taper once you know how much they are deflecting. Also make sure they are relieved at the bottom I could not tell from the picture if they was.
 
Chuck a part and check the runout. Tap it true. Tap it with a mallet and see check runout again. Do this a few times. The amount of wobble should somewhat repeat. Use the wobble TIR to make an educated guess how much taper to backbore in the jaws. Make it just a touch more than you think you need. Better tight at the front than the back.

FWIW i most always backbore a small taper even on shallow jaws. For a 1"deep bore maybe .002. The jaws ALWAYS tilt so it helps.
 
Chuck a part and check the runout. Tap it true. Tap it with a mallet and see check runout again. Do this a few times. The amount of wobble should somewhat repeat. Use the wobble TIR to make an educated guess how much taper to backbore in the jaws. Make it just a touch more than you think you need. Better tight at the front than the back.

FWIW i most always backbore a small taper even on shallow jaws. For a 1"deep bore maybe .002. The jaws ALWAYS tilt so it helps.

Backtapering jaws in principle (I know this is mainly academic) does not create a perfect contact surface, it only ensures the jaws are touching at the front, but if they have full diametric contact at the front they will only have point contact at the back...

I always preload the jaws at the front with a ring and bore parallel.

I used to do a lot of similar parts in very long soft jaws and preload at the front with parallel bore always worked for me.

That all said, since it cuts cleanly at the start and then starts to chatter it's probably just a harmonic building up and that specific point on the shaft is the natural position for it to resonate. That is where varying the spindle speed comes into play.
 
Backtapering jaws in principle (I know this is mainly academic) does not create a perfect contact surface, it only ensures the jaws are touching at the front, but if they have full diametric contact at the front they will only have point contact at the back...

I always preload the jaws at the front with a ring and bore parallel.

I used to do a lot of similar parts in very long soft jaws and preload at the front with parallel bore always worked for me.

That all said, since it cuts cleanly at the start and then starts to chatter it's probably just a harmonic building up and that specific point on the shaft is the natural position for it to resonate. That is where varying the spindle speed comes into play.
I agree if you preload at the front. My guess that is not the case here. Hence the trouble.
 
Thanks for the suggestions, I should have added that I did test the jaws for contact.

I cannot fit a .001 shim between the part and jaws in the front or along the sides with any of the jaws. I also sharpie'd the part and looks like I'm making good contact across the length of the jaws.

The jaws were cut with a boring ring so they were preloaded and measured while clamped to match the diameter of the part within .0005". There is definitely a back taper when you measure them unclamped.

Runout is ~.002. I've tried wailing on the part with a dead blow but I cannot get the indicator to move.

The L/D is only 3.2 so I'd think that's ok. (It's 6 on the main side and there's 0 issues there).


If it a work holding rigidity issue, why would the tip machine fine, but sudden appear 0.25 down the part?
 
That all said, since it cuts cleanly at the start and then starts to chatter it's probably just a harmonic building up and that specific point on the shaft is the natural position for it to resonate. That is where varying the spindle speed comes into play.
Sorry I missed this on my initial response. I'd agree that this is probably a resonate node on the shaft. I've tried all sorts of RPMs but haven't been able to change it. I don't have the option to oscillate rpm that I know some controls can do.

Any suggestions on how to fix it?
 
Sorry I missed this on my initial response. I'd agree that this is probably a resonate node on the shaft. I've tried all sorts of RPMs but haven't been able to change it. I don't have the option to oscillate rpm that I know some controls can do.

Any suggestions on how to fix it?

Are you able to prove/disprove it by standing at the control and twiddling the spindle speed override while it's cutting?

What control is on this machine?
 
I've tried all sorts of RPMs but haven't been able to change it.
That would have been my suggestion. By "all sorts" are we talking 1% changes or 50% changes? If you haven't already, try a 25% reduction and a 50% reduction. Only needed for the part of the cut that's causing problems.
 
I've gone+/-50% on rpm with little to no change. If I adjust it in the cut, it seems to help but the second the rpm stabilizes, it comes back.

I'll give rubber bands a shot, though, not a very elegant production solution if it works.
 
You could also try leaving a lot of stock for your finish cut. Like .100 or so. Sometimes the extra stock stabilizes the cut.
 
Since you don't have jaw problems we move on to next thing..
1) Do you have ANY chatter or harmonics when roughing? *you say you are only cleaning up that last little bit
2) Trying a VNMG first with a .016 insert, then .008 at decent feed at slower SFM.
3) still doesn't work look if you have a neutral rake tool around with a small nose radius insert. Then last trick is to throw in a positive rake insert.

Report back. Quite impressed though with what you have accomplished. Next, bigger lathe.
 
The repeatability of the chatter can be used to your advantage. It may, for example, take 0.5 seconds to turn the first half of the section before the vibration of the combined tool, part, chuck, and lathe spindle structure becomes excessive. The solution may be to limit the turning cycle to 0.2 seconds, withdraw the tool, wait 1.0 second for the vibration to damp out, and then resume turning for another 0.2 seconds. The risk of restarting the chatter ahead of schedule can be reduced by taken the second cut from the right side of the section to avoid any existing surface defects and then complete the turning operation at the center of the section.
 








 
Back
Top