I am somewhat stumped. Recently evaluated a roll-turning lathe that is producing an unacceptable finish on the rolls. Apparently it ran fine for a number of weeks before unacceptable vibration developed. Measuring in line with the lathe tool, there is a very clear vibration at 2200 cpm which varies in amplitude -- but not frequency -- with changes in roll speed or changes in travel speed. When first entering the cut, the vibration is minimal, but the amplitude slowly increases with time. If you back out, and then restart the cut, vibration is again minimal but increases with time.

Impact tests clearly reveal a resonance near 2200 cpm which varies somewhat with a change in turrett gib tightness.

We cannot find anything that operates at 2200 cpm (or a sub-harmonic).

Clearly the entire turret assembly is rocking, with minimal vibration at the base and maximum vibration at the top.

I have two questions: 1) Is it possible for the turret assembly to "self excite"? 2) Has anyone seen such a problem before, and if so, how did you solve it?

When I was an apprentice fitter & machinest we had to turn a lot of conveyor rolls.It was common that these would generate a vibration throughout. it started out light but as the cut continued so did the amplitude. We had a couple of solutions back then (pre my days of vibration testing). The first was to put a piece of wood in the other end of the toolstock and have it press hard against the roll. This acted like putting your finger on the wineglass once it started to resonate. Others used big "O"rings placed over the roll.
My thoughts are the noise travels up the roll hitting on the end shell which then bounces back and starts all over again when it reaches the cutting tool.You can almost whatch this happen if you use a continous coolant on the tip. Try the wood to see if it goes away.
Hooch

If mass damping through an absorber as described by Hooch is not sufficent, the machinist has to go back to basics and address all sources of tool deflection (tool post, saddle), tool material and profile, tool holder dimensions, speeds and feeds, stock material, etc. It may be an "operation problem" rather than a "vibration problem"

Another avenue is to try posting to http://www.eng-tips.com/threadminder.cfm?pid=281

Hey Rusty, I'm sure the machinist is quite experienced and knows these things, but from a machining perspective, when the tool "chatters" it's common practice to reduce spindle rpm and increase cutting feed rate. Also reduce depth of cut. Also make sure that the cutting insert is at or a few thousandths below spindle centerline. Lenght of tool overhang should be as little as possible as well. Cutting insert geometry, chip breaker and edge prep is a major factor as well. If you put a 10th's indicator on the spindle or chuck and push and pull, there should be around .0005 deflection. You could also check to make sure the x-axis and compound gibs are snug. There may also be gib locks as well that would help.

What size lathe, approx dimensions of the roll?

Again, I may be stating the obvious, but I've seen lots of instances where inexperienced guys are just thrown under the bus and told to run the lathe.

Andy

This message has been edited. Last edited by: Andy Paul,

Thanks for the input guys. I will post some spectra later, but this really does NOT look like a tool chatter problem. The operator has been there 16 years and they never have this problem with the 2 older (and more massive) lathes, just this new, smaller one that is supposed to work just as well (according to the mfg). They've had a tech in to check everthing (including the gibs), and even a guy from Germany and they can't find anything "wrong." This is a large machine, but smaller than the older ones (sorry I don't have more details - this was a quick trip before I left town for a week).

The entire turret assembly - not just the tool holder - is "rocking". They've tried every combination imaginable with cuts as little as 0.005" and as great as 0.050" (typical).

I am concerned the lathe is too small, light, and flexible to turn such large rolls (it's about half the size of the older machines, and cost a lot less) but have no way to objectively determine that.

I'd rather characterize it as "self-feeding" excitation process which always takes place in machining. In some cases it won't go wild whereas in others - it will. Self-feeding is a process when low magnitude white noise during machining excites turret's resonance and then impacting due to turret movement which excited turret resonance gets progressively worse. The process/equipment by its nature and built-in positive feedback is pron to resonance, in this case it happened at 2200 cpm.

I believe that the only way to alleviate the problem is to add damping to the turret.

An afterthought... Is it possible that the roll also resonates at 2200 cpm?

David

This message has been edited. Last edited by: David_G,

Hi Rusty,

I long time ago I solved a "chatter" problem at a steel mill on a roll grinding machine. When the machine was run with both varying chuck and grinder speed, it would produce a barely acceptable roll. The turret was mounted on a separate inertia block from the one the grinder was mounted on. If I remember correctly the starting amplitude was around 0.0002 IPS pk. It would rapidly climb to unacceptable levels within a couple of passes. My frequency was 3457 rpm. It was the cooling fan motor on the DC drive motor for the grinder. One of those little 3/4 HP jobs the mounts over the commutator. Balanced it down and separated from the motor with some air hose and duct tape. Viola! But there was a lot of playing with the roll setup prior to anyone believing me. Tail stock pressure was something that I remember being able to produce an adverse effect on the vibration. Hope something in this story helps. Sean

Rusty,
Use a tool like a single point screwcutting tool, with the lowest possible tip radius, and a sharper angle than 60 deg.
If the lathe has movement on the turret, there may a manufacturing fault, however, there may be a facility to lock all the possible movements, which should help to reduce chatter.
On the theme of using wood to dampen resonance, I have positioned a brush handle or any handy light piece of wood, overhanging the lathe saddle, allowing the wood to rub on the article being machined. It may be enough to prevent resonance.
Best of luck
Joe Mc Cormack

Hi Rusty,

We made vibration measurements related to chatter problem way back in 2007. please find the attachment. hope it helps.

Madhu

Chatter_report.pdf (649 KB, 24 downloads)