An Unusual (to me) Chatter Problem

At 3-1/2" are you using a tailstock centre ?

-- Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

John

Hi, I have about 2-1/2" sticking out of the chuck but must admit for the 1" bar I haven't been using any tailstock support. For turning smaller diameter 1/2" down I do use a centre, while the problem is better with support I still get a "pattern" about 1-1/2" from the chuck over about 3/8" or so. I haven't tried with the larger bar and support; I'll go out now and have a go and "report back".

Thanks for your time (again).

Best regards

Keith

I haven't tried with the larger bar and support;

John

Hi, whatever you are thinking you are on the right track. If I support with a live centre I can cut at 2000rpm and .030" without a problem over the full length. Remove the support and the problem is back at anything over 900rpm. Could it be that I have the spindle set too far forward so that the cone bearing is too loose? I have recently replaced the rear ball bearings and belt and have tried not too set the front cone too tight as it is an expensive replacement. At the moment it is set slightly tighter than Myford suggest as a starting point. It does warm up slightly at speeds over 1050rpm but I suppose that might be fairly normal.

The other thing that slightly bothers me is that the chuck although new is mounted on one of these pre-machined steel backplates. The register diameter although the tightest I could find at the Harrogate show is still a couple of thou bigger than I think it should be. The range I tested was between .002" and .009 over the nominal 1.5". Having said that the chuck being brand new appears to hold within .001 - .002 and is repeatable at the moment.

Thanks again

Keith

Are you using the leadscrew and change gears to advance the carriage or feeding manually with the handwheel on the saddle? Whichever one you're using, try the other as this will eliminate the rack or the leadscrew. Also try advancing the carriage using the leadscrew, but not under power (i.e. the handwheel at the tailstock end).

If you are power-feeding, make sure the leadscrew does not have any swarf in it. Clean with a brass brush if necessary.

I do not know if this is even possible, but if it points to the rack, you may be able to take it off and switch it end-for-end.

How repeatable is the chatter/banding length- and distance-wise relative to the face of the mandrel? If you change the distance between the (centre of the) half-nut and the point of the tool [e.g. with a Dixon toolpost, mount the tool at the headstock or the tailstock end of the topslide], does this change the position of the chatter?

Do you have a gearbox model? If not, is your leadscrew in two pieces with a roll pin connector between the two halves?

John has given very good advice concerning supporting the free end of the workpiece. You could also try turning without gripping the workpiece in a chuck at all -- between centres (using a soft centre in the headstock and truing it up if necessary). If you don't have a driver plate and dog, it's possible to use the chuck jaw onto a Jubilee clip to push the stock round.

John Hi, thanks for all your ideas it has taken me a little while to work through them all and I have certainly learned a lot more about my lathe in doing so (not all good unfortunately). Anyway to work through your suggestions:-

I mostly use the power feed via the leadscrew but have now tried all of these different methods. The original problem remained with all of them.

I did this and also adjusted the half-nuts and subsequently I can detect a little more backlash in the leadscrew towards the headstock end of the lathe. I adjusted them for the range I was using but no improvement in the problem.

Interesting one this, very repeatable in position of chatter band but moving the tool point towards the tailstock certainly made the depth of chatter worse. This has confused me as I would have thought it would have improved things? So I can vary the degree of chatter but not remove it.

It is a gearbox model

No problem whatsoever when turning between centres - I can take .025"+ at 2000rpm and get nice blue swarf with no chatter at all. This is also the case when using the chuck with tailstock support. Although this gets the "job done" I can't believe that an average Myford can't turn successfully 1" - 11/2" from the chuck if it is working correctly so I would still like to crack the problem.

Some of the results when going through your ideas has shown that the adjustment of saddle and spindle bearings was not as perfect as I had thought. I had replaced the two ball bearings on the main spindle with some I had (still in box) but I have now seen that when they are pre-loaded (lightly) they develop some roughness. I have ordered some new ones and will replace/readjust before I continue the "quest".

I have also discoverd that all inserts are not equal even when new, the problem is most pronounced when using some of the "gold" coated type. They also seem extremely critical on centre height, on centre or slightly below seems the best anything even minutely high produces an in/out cut pattern. This leads me to believe that it might be anything that increases tool pressure that is finding the "weak spot" in the wear or adjustment of the lathe.

I would like to thank everyone who has suggested things for me to check including those that e-mailed me with their own experiences. Although I have not yet cracked it completly, I have now got it to a workable situation. As long that is as I use a particular type of tip, cuts below .010" and keep the speed down slightly. Please keep the ideas coming if you have time, you have all improved my knowledge of how the lathe works (or doesn't) greatly. I very much appreciate that and am greatful for the time and trouble that you have all taken.

Best regards

Keith

Hi Keith, Reading your latest post reminded me of something that jarred with me in one of your earlier posts and I did not remember to reply then. This was your comment regarding the clearance of the chuck backplate. I'd expect minimal clearance on the register. I feel it is very important that the chuck is completely aligned both axially and radially by the two surfaces of the mandrel register and in no way by the thread which crudely put is just there to stop the chuck falling off. You found that the chatter stopped when turning between centres. This suggests the mandrel bearings are fine ( one of your earlier suspects) and that your saddle/topslide/leadscrew are without blame too. It does however cast doubts on the chuck or its mountings.

hth

Bob

Bob Hi,

Thanks for your reply it has prompted me to do a few more "tests" with interesting (to me anyway) and very useful results. The lathe is in its' re-adjusted condition and is still awaiting the new ball bearings as they haven't yet arrived.

I have several chucks available of various quality/condition and I have tried them all at the same speed/feed/depth/dia settings and the results do vary. I have measured the various internal register diameters and there would appear to be a possible relationship between the performance and clearance. The other factor that could be affecting it is the "overhang" as the chuck assemblies all vary in their length from the spindle register vertical face to front face of the jaws. I tried to eliminate this by using a longer piece in the shorter chucks so they were cutting at the same position but this leaves too much sticking out to give a fair comparison.

I have 3 Pratt Burnerd "threaded body" chucks (3 and 4 jaw) ranging from brand new to fairly worn, all of these have a register clearance of .0005" and a length of approx. 2.75". Now the lathe has been re-adjusted none of these produce chatter at .015" cut although the

6" 4 jaw (good condition) does produce a slight "ringing" sound. Up the speed to 1400rpm and they are still OK with the ringing a bit more pronounced but finish still OK.

A very old Crown 3 jaw backplate mounted with a length of 3.5" and a register clearance of .0008" (a little bit of a guess really but it is certainly over .0005" but less than .001") produces a definite ringing sound but the finish looks OK. However raise the speed a notch to 1400 and the chatter is back.

The original "problem" chuck is a brand new Universal that came with an Emco lathe and it is mounted on a brand new backplate bought at Harrogate. This is the longest at 4" and also one of the heaviest with a register clearance of .0035". This produces extensive chatter over a range of about .75" as before. I can eliminate it by reducing depth of cut to .005 and the speed to 420rpm.

Another puzzling thing I have noticed is that the ringing/chatter is the same distance from the chuck jaws in all cases irrespective of the length of the chuck and is therefore not in the same position relative to the spindle or the bed??

Anyway Bob thanks very much, prompted by your message this has been a very useful exercise which has resulted in me having 3 chucks which I can now use with confidence. I will buy a new backplate for the "problem child" if I can locate one with a proper register diameter. If not, I will transfer it to my Boxford and see if it is useable there. Another example of "new" not always being best and of course as always there is no substitute for "quality". Pity as quality can be difficult for us beginners to identify particularly as nowadays there appears to be little direct relationship between price and quality.

Thanks again Bob for your time

Best regards

Keith

Hi Keith, I only too glad to have been the stimulus for your experimetation but sorry I could not offer an expanation. I'm really curious abut this now and the problem keeps coming back into mind but not getting resloved.

As you say it seems curious the the 'chatter zone' remains fixed with different chucks having different masses. If it were due to some sort of resonance in the mandrel, I'd have expected the chuck mass to have made a difference. I don't suppose there coud be a stress crack in the mandrel? it's a long shot but we are not dealing with an everyday type problem. In 30 years of amateur truning with a variety of chucks, some new some bell mouthed etc I've not come across anything like this.

Please keep us up to date with progress.

Regards

Bob

Hi Bob, good call, when i have the spindle out to change the rear bearings I will pop it into my ex-employer and get the lads to give it the once over with the ultrasonic equipment. I will also get them to check it for straightness.

One of the possible explanations I have in mind is indeed to do with the mass and distance from the front bearing. If it is infact the rear ball bearings that are the trouble then I have been trying to get my mind around the affect of larger/smaller mass with different gyroscopic affect and greater/shorter leverage (assuming the front bearing is good) between tool/front bearing and front bearing/rear bearing. While it appears on the surface to be a simple mechanics issue the affect of the larger/smaller mass rotating at 1000rpm must change the dynamics somewhat. Unfortunately, I'm not a sufficiently knowledgeable theoretical engineer to work it all out but from my helicopter days I do know that the gyroscopic effects of differing masses and the stiffness of shafts can have some strange affects on shaft displacement. Of course, my "model" might well be me just "clutching at straws" to justify changing the rear bearings.

If I have time I will repeat the tests and confirm an accurate position of the chatter relative to the spindle register; my initial observation was based on a quick check with my 6" rule. I might also increase the pre-load on the rear bearings to see if that has any effect. I will of course let you know the outcome, good or bad.

Thanks again for taking the time to share your thoughts with me.

Best regards

Keith

experimetation

As chatter is a vibration problem you can play tricks by changing the frequency. Try setting the tool out from the holder as far as practical, and then in as close as it will go, these should produce different chatter patterns.

with the tool extended, use a small toolmakers clamp to clamp a couple of bits of rubber or cloth to the tool, this should damp the tool vibration...

Hi all, many thanks to everyone who posted or sent helpful hints, with all of this help I have finally managed to crack the problem.

Bob, sorry to say that it turned out not to be an "unusual" problem at all, rather a combination of simple errors. It did however involve the chuck mounting which you had already identified as a possible culprit. I had the spindle checked and it was found to be OK (rather good in fact, the aerospace guy who usually works with small numbers of microns was impressed at Myfords' ability to produce a straight and true spindle).

I replaced the rear bearings and increased the pre-load but the problem remained. When I was checking the actual position of the chatter I realised that the two problem chucks were in fact the same length, the different overall length is due to different styles of backplate. So I looked at the chuck/backplate mounting more closely. On my "problem child" I noticed a little bit of distortion on the very end of the bolts holding the chuck to backplate. There was no obvious movement in the joint but could they be too long? Yes is the answer, removed a couple of threads from the bolt length and instant success.

So I had a look at the second chuck which had been performing a little better, no problem with the bolt length here. A closer look revealed that I had not put sufficient chamfer on the end of the large backplate register and although I couldn't see any gap it was not fully entering the parallel portion of the chuck register. As this backplate is a little "thin" I had also been too tight with the length of this register. Re-machined the large register properly and even with a (to me) excessively thin backplate instant success again. I'm beginning to enjoy this "success" thing. As an additional bonus the

6" Pratt that had been "ringing" slightly is also now OK so I guess something in the new bearings or adjustment has been improved slightly as well. So thanks Bob, I know I'm a bit slow but with your prompting I got there in the end.

Jonathan, thanks for your pointer, in this particular case I had only .5" of the tool forward of the toolpost to try and eliminate tool flex (thanks to Thomas O on this one as well). I was not aware of the "tricks" you identified but I'm sure they will come in handy particularly with boring bars that seem to "sing" constantly. As a trial I wrapped some thin lead sheet round my favorite boring bar (too long, too thin!!) with some soft wire, thankfully it certainly leads to quieter turning although that "nice" pattern disappears as well!!.

So thanks to everyone who helped, I'm off now to make plenty of thick pretty blue swarf without the patterns.

Best regards

Keith

Glad to hear you have been able to put this one to bed now.

I'm quite envious of you being able to make swarf, I'm still nursing tendon and skin grafts to my right hand after my wood planer decided to show me it was boss! I guess it will be about another month before I can consider getting near the workshop. In the meantime I'm just adding jobs to my mental do-list without knocking any off.

Regards

Bob

Thanks Bob, sorry to hear about your disagreement with your wood planer and I hope the hand recovers well. Feed it a bit of hard oak next time that will teach it who's boss. At least you are in the right part of the country to enjoy this weather, I worked for a year or so in Gosport and while I enjoyed the sea and sunshine I seem to remember it also brought all the other "people" out as well. Ah well can't have everything.

Just think of all those jobs you can plan to the "nth degree" and then all those plans you can change when you actually get started - what fun. I don't have a do-list myself at least not one I can see it is buried under the wifes'.

Best regards

Keith