Any idea why this lathe is chattering? Spindle bearings?

[loply]

Hi folks,

Video showing the problem - shot of the surface finish is at the end -

So I've just finished doing a full refurb on this lathe, scraped it all in, replaced the old gear drive with a direct belt drive  & VFD, serviced all the parts etc.

Put it back together and it's chattering like mad. Anything steel above about 30mm is more or less unturnable, and even on steel around 15mm it only turns nicely if I use a high rake tool (like what you'd use on aluminium). It seems like it just can't handle a heavy load.

I've double checked everything and eliminated all obvious variables - tried different bits of stock, steel, alu, all different tools etc. Checked nothing is wobbling on the saddle/slides/etc. Absolutely everything I've eliminated I think.

The chattering noise is coming from the headstock. The poor surface finish does not improve the closer to the chuck it is. The bearings I adjusted tighter and tighter but that didn't help. I have about 0.007mm of end play in the spindle (if I chuck up a big bar and pull on it).

I'm wondering if the bearings are knackered? They're timken taper rollers and the races look okay to a laymans eye, but I don't know for sure how to test them?

Would appreciate any advice as I'm a bit disappointed at this stage!

Thanks,
Rich

[John Swift]

may be not enough pre-load on the spindle bearings
excessive overhang of the tool and play in the gibs will add to the problem
try using a centre to support the end of the bar
if the chuck jaws are slightly bell mouthed it would not help having so long a piece unsupported

   John
PS
having another look at the video the tool length is not the problem

[loply]

As I say I tightened the spindle bearings a lot and it didn't really help. The official spec is for no preload on them.

Had the part chucked tight as hell though as I say the problem happens even right next to the chuck so I don't think it's a workholding / rigidity issue.

[Arbalist]

What type of tool is that you're using, it's difficult to make out for sure.

[loply]

Tried everything - all the same result - two types of carbide insert, HSS, really sharp HSS, and sharp brazed carbide...

Best results came from tools with big back rake - like 35 degrees - but these won't last on steel. Still could only take 3 or 4 thou without chattering.

[jb3cx]

I think I would agree with john about using a back centre ,and try different speeds ,that's a fair lump of Steel hanging out of the chuck
Peter

[Pete.]

Put a DTI on the chuck and wrench the bar up and down. If it moves the DTI your problem is in the bearings. Where did you get the spec for the pre-load? My 280 is set by running the lathe at high speed and measuring the headstock temps.

[Arbalist]

Tried everything - all the same result - two types of carbide insert, HSS, really sharp HSS, and sharp brazed carbide...

Best results came from tools with big back rake - like 35 degrees - but these won't last on steel. Still could only take 3 or 4 thou without chattering.

But what is the tool in the video?!

[chipenter]

I recently had to scrape my compound slide to my toolpost . as there was a gap at the front of about 1 thou made a big diference to the finish .

[SteveT]

Looking at your video the chatter is uniform along the length of the bar, and the tool looks like its about 10mm wide and flat to the cutting surface the tool should have a clearance rake so you are only cutting at the tip.

1/ have you assembled the taper bearings correctly I think they should be back to back? also depending on the bearings set-up are the bearings cross located
that is the bearing assembly needs space to expand as it heats up, a common fault on newly assembled bearing units is the shims are but in the incorrect place       
so locking the bearings.
 
2/ From the video the lath sounds very rough which points to the head bearings

If you could post an assemble drawing of the head it would help a lot.

Hope this helps

[RussellT]

I haven't watched the video but it might also be worth checking the fit of the chuck to the spindle nose - a bit of swarf could cause this sort of problem.

Is this the original or the spare headstock.

Russell

[Arbalist]

The tool looks like its about 10mm wide and flat to the cutting surface the tool should have a clearance rake so you are only cutting at the tip.

That's what I was thinking. I'm also unsure about the height of the tool as well.

[DavidA]

Loply,

According to the manual for my 280 Synchro there should be no end float at all.

It instructs one to adjust the preload but not to allow excessive heat to be generated in the bearings.

it reads...

The machine should run at top speed and the bearings should not exceed 65 C.

that may be your problem; loose taper bearings.

Dave.

[hermetic]

To sum up everyones comments! The bar is far too long and large to be not supported at the outer end., the tool angle is far too obtuse, I don't think the cutter is on centre height, and the machine speed is too fast, and also there is probably some flex in the QCTP. Watch doubleboosts early videos on youtube using a similar lathe. Then put a centre in the bar and support with the tailstock centre, re sharpen the tool, set on centre height. Slow down the lathe, and slow down the feed, and that should cure the problem

[loply]

Hi guys,

Thanks for all the advice. I appreciate all the comments although some of them have missed some of my original points.

Regarding the spindle bearings, if I grab a long bar e.g. the 50mm x 150mm piece of steel in the video, I can deflect it's ends up or down about 0.03mm quite easily with just a bit of hand pressure. I've no idea whether that's normal or not as I've never tried it on another lathe before. Could somebody else confirm what they get on their lathe for a similar size bar?

If I put a really sensitive indicator on a ground MT3 test bar and spin it by hand, the needle does jump around randomly a bit. The indicator has increments of 0.002mm (ie about a tenth). It doesn't run out by a consistent amount. That might just be normal though as the indicator is so sensitive. With a normal 0.01mm indicator it looks pretty smooth.

I initially had no preload on the bearings, then gradually tightened them until I couldn't move the nuts any tighter with an 8 inch spanner. Strangely the bearing friction didn't increase noticeably. They don't seem to be getting hot when ran at full speed though I can only check by touching the casting.

I will double check the chuck-spindle connection carefully, that is a good point, thanks. I did have the chucks on and off quite a few times today mind and it was the same each time, maybe there is a ding in the thread which I didn't notice though.



To address some of the points which I don't think are relevant (though I could always be wrong!) -

The workpiece isn't the issue, the bar is a traditional size for a test bar, ie 50mm steel bar 150mm long, so it's only sticking out by 3x it's diameter, no need for tailstock support. Indeed that would defeat the entire point since it's meant for testing the headstock alignment. This is a fairly normal test you would do on any lathe of this size (albeit often with two 'collars' by relieving the middle somewhat). Besides which the chatter occurs even right next to the chuck so it's nothing to do with the length of the workpiece.

The tool is also irrelevant, it does the same with any type of tool. In the video it's a high back rake HSS tool - about 12 degrees on all sides but about 20 on the back rake, and razor sharp. Same thing happens on carbide inserts and brazed carbide. The exact same tools on my previous lathe cut perfectly. Definitely on center by the way - checked that several times and tried slightly above and below too just out of desperation. I get better performance with really high back rake (like 30+ degrees) but that shouldn't be necessary.

I tried two different chucks too by the way, a 4 jaw and the 3 jaw in the video. Cranked both of them really tight to be sure.

Cheers,
Rich

[loply]

By the way the depth of cut in the video is 0.06mm, that's all it can manage, any more and I think it's dangerous (sounds like it might rip the part out).

For RPM, tried everything from about 150 with HSS up to about 800 with carbide, and everything in between. Couldn't find a sweet spot anywhere.

[awemawson]

I wonder if the bearing adjustment isn't working quite as you think? If you are swinging on an 8" spanner I'd expect some noticeable friction and them running warm. Indeed in the past I've adjusted head stocks so that there is just a bit of noticeable friction, then backed them off a tad. Leave them running while having your tea, and if they are not warm when you return they are too loose. If you can't keep your hand on them they are too hot and need a slight slackening.

Crude. But it works.

[philf]

Loply,

Is there a sleeve over the spindle between the inner races of the taper rollers?

If there is you might just be tightening up against the sleeve and no amount of effort on the spanner will make any difference.

Phil.

[loply]

Hi Phil,

Nope, no sleeve. I know what you mean. It's just a simple bearing-spindle-bearing arrangement though.

Awemason, I am a bit perplexed as to why tightening the bearings so much isn't increasing the drag noticeably. I literally can't tighten them any more and they're only just dragging. I know the way it's built inside out though and there's nothing complicated about it. The only thing I could think is that the rear bearing is stuck on a ding on the spindle so won't move, but I gave it that much force I don't think that's likely.

Just been out and checked something, it's actually about 0.07mm up or down I can deflect the bar by, and I notice the chuck also moves - by about 0.03mm up or down.

I'll check whether the spindle nose itself can be deflected tomorrow, too late to start taking it apart now though.

[DavidA]

Loply,

...Hi Phil,

Nope, no sleeve. I know what you mean. It's just a simple bearing-spindle-bearing arrangement though.


Are you quite sure about this ?

I have in front of me the machine manual for the 280 Synchro, which also has two taper roller bearings on the spindle at the chuck end of the headstock.

There does appear to be a spacer in between the two outer rings.

Dave.

[loply]

Hi DavidA, absolutely positive. Had the whole thing apart no more than a few days ago, plus, I have a complete spare spindle + headstock sitting on my workbench Unfortunately it has no bearings so I can't compare or use it...

The 280 is a lot more complex and has more bearings (plus I think a taper spindle nose?). They're much better machines I think. The threaded spindle nose is a real weakspot.

These older lathes only have one bearing in the front and one in the rear. Originally there would have been a sleeve on the spindle but it purely carried some gears - it wasn't a bearing spacer - but it's gone anyway as I have a taperlock pulley on there now instead.

Here's a video - the more I think about it, this definitely doesn't seem normal? I'm pushing quite hard mind. I'd appreciate it if anybody else could make a similar check on their lathe.

[DavidA]

Loply,

You are correct, they are very different.

Hope you get it sorted. i am still having teething problems with my 280.

Anyway, if all else fails, ask the company. They are still there and will probably be happy to advise you.

http://website.denford.ltd.uk

Dave.

[PK]

Do both bearings have the same ID? If not, could there be a shoulder on the spindle the rear bearing is hitting before it loads up against the headstock. (yes this is a stupid question, my experience in these things is that my own mistakes are often illuminated by them).

[Pete.]

Hi guys,

Thanks for all the advice. I appreciate all the comments although some of them have missed some of my original points.

Regarding the spindle bearings, if I grab a long bar e.g. the 50mm x 150mm piece of steel in the video, I can deflect it's ends up or down about 0.03mm quite easily with just a bit of hand pressure. I've no idea whether that's normal or not as I've never tried it on another lathe before. Could somebody else confirm what they get on their lathe for a similar size bar?

If I put a really sensitive indicator on a ground MT3 test bar and spin it by hand, the needle does jump around randomly a bit. The indicator has increments of 0.002mm (ie about a tenth). It doesn't run out by a consistent amount. That might just be normal though as the indicator is so sensitive. With a normal 0.01mm indicator it looks pretty smooth.

I initially had no preload on the bearings, then gradually tightened them until I couldn't move the nuts any tighter with an 8 inch spanner. Strangely the bearing friction didn't increase noticeably. They don't seem to be getting hot when ran at full speed though I can only check by touching the casting.

I will double check the chuck-spindle connection carefully, that is a good point, thanks. I did have the chucks on and off quite a few times today mind and it was the same each time, maybe there is a ding in the thread which I didn't notice though.



To address some of the points which I don't think are relevant (though I could always be wrong!) -

The workpiece isn't the issue, the bar is a traditional size for a test bar, ie 50mm steel bar 150mm long, so it's only sticking out by 3x it's diameter, no need for tailstock support. Indeed that would defeat the entire point since it's meant for testing the headstock alignment. This is a fairly normal test you would do on any lathe of this size (albeit often with two 'collars' by relieving the middle somewhat). Besides which the chatter occurs even right next to the chuck so it's nothing to do with the length of the workpiece.

The tool is also irrelevant, it does the same with any type of tool. In the video it's a high back rake HSS tool - about 12 degrees on all sides but about 20 on the back rake, and razor sharp. Same thing happens on carbide inserts and brazed carbide. The exact same tools on my previous lathe cut perfectly. Definitely on center by the way - checked that several times and tried slightly above and below too just out of desperation. I get better performance with really high back rake (like 30+ degrees) but that shouldn't be necessary.

I tried two different chucks too by the way, a 4 jaw and the 3 jaw in the video. Cranked both of them really tight to be sure.

Cheers,
Rich

1 thou up and down on the spindle is miles too much. I think you have a problem with your spindle fitment, borne out by the fact that as above you should be able to bind the spindle with heavy adjustment.

Have you got a photo of the spindle assembly out of the headstock?

[loply]

PK, I understand what you mean, but no, the spindle is one diameter where the rear bearing is, with no shoulder.

I've just been out and took the chuck off to eliminate the chuck:spindle connection. I can move the spindle up and down by about 0.01mm by pushing hard with my fingers on a 200mm long bar in the spindle taper. Again I don't know whether that's normal or not?