Lump in Lathe MT3 Spindle

So I decided it was time to have another tweak of my lathe. The spindles a bit off after I put it back together (without realising how to tweak it) and the tailstock is about .25 mm too high.

I had the bright idea of getting an MT3 test bar, which I did (Ebay).

However, when I put it in the headstock spindle it rattled. Even I know this is wrong.

So I tried my MT3 centre. That rattled too, though not so much.

After sleeping on it, I blued up the Centre and checked the results.

The main issue appeared to be right at the mouth of the spindle. I had a bit of a go at skimming off an tiny amount from that, but it still rattled.

I checked with an indicator and found that the rattle was about .01mm just by the edge of the spindle and about .05mm 20mm further out. A bit of homework suggested that the fulcrum (ridge) is about 5mm in from the edge. Pretty much where the bluing suggested.

Next I run the indicator into the spindle. Just inside before end of the taper, i'm getting a peak at 0.03mm roughly over 45 degrees. The rest of the spindle is round to considerably under 0.01mm.

5 mm further in I get a much smaller tick leading me to think that its a defect right on the edge of the spindle.

However, I also see a couple of apparent high spots right at the other end of the spindle, though these don't see as repeatable, so if could be some dirt.

I'm a bit scared to try anything without some advice.

Cheaply I could set up the topslide on an MT3 taper and try and skim the edge. Good aspects to that are cheap (did I mention that) and quick. Also if it is just the last 5 - 10 mm of the taper and I cock it up then there is still quite a bit of taper left. Bad side is that it's not going to address the high spots further inside, though they may not actually be having a detrimental effect.

Option 2 that I came up with is to get an Mt3 taper reamer and apply that. My concern with this (apart from the not cheap aspect) is that, given there is a lump it may push the whole taper slightly off center (that is it wil take a little off the opposite side to the lump as well as the lump itself. Most likely this will be of the order of microns and not really affect the accuracy of the lathe (not in my hands anyway)!

often a cry for help ends up with an option 3 (and often 4 - 237) so I look forward to your feedback!

Iain

Firstly let me state I am not a machine reconditioner so this is personally what I would do as a non professional.

1. Clean he spindle taper in as gentle a was as possible. Either make up a wooden scraper from a suitable piece of mdf shaped to the correct taper, or fit a 3MT taper with 3 thin strips of scotch brite attached with double sided sticky tape. This should clear any crud.

2. Do your blueing test again with the test bar (I am assuming its a good one)

3. Scrape any high spots locally with a suitable scraper.

I would shy away from the reamer option on a spindle unless used really gently by hand to clean rather than cut as per my option 1.

When you do blue it up shine a light down the other end to make things easier to see.

No doubt others will have comments to make too so don't be in a hurry.

regards Martin

Martin's advice looks good to me.

Use a torch and a mirror and you should be able to see the high spot marked with blue.

It may just be dirt, but it's almost certainly a 'bruise'. If the rest of the taper is OK then stoning, scraping, or even filing out with a riffler file. the small raised patch will have no impact on the rest of the taper and should restore accuracy.

I would be very wary of any approach (i.e. a reamer) that could result in damaging the rest of the taper surface unless it was really chewed up.

Neil

.

Option 2 could be potential problem. Before this was done one would surly have to be 100% sure that the headstock and the tailstock (and anything you placed into the tailstock) were in perfect alignment. - Which as you have said they ain't.

Nick

Martin - thanks.

Option 3 looks particularly good!

If the bruise is just round the big end, this is something I can do (usual disclaimers of only vague competence apply).

If I find that there are issues near the small end, are there any tricks or tips for getting the scraper (or other tool) down there with some accuracy. 4 or 5 inches isn't so far on the outside, but in a 20mm wide bore it's more challenging!

I've got a set of small scrapers and was planning to use the triangular one for this. Is that the right one?

Iain

"'planning to use the triangular one for this. Is that the right one? "

Sounds like it. Just so long as you can remove small amounts of material where the blue has indicated. Go gently, it doesn't matter if you have to go at it several times. Blue, scrape repeat. Like Neil says the rest of the surface is correct so you don't have to touch it.

It's reasonable to assume that the spindle socket was correct at one time in it's life. The high spots are most likely to be impacted material from not cleaning the socket well enough before fitting a centre. I cannot see how a bruise would be caused but maybe others can. If it is impacted metal then it should flake off reasonably easily. You should be able to tell more from the shape of the raised portion. Be your own detective.

Martin

OK.

I've done some scraping,.

The good news is that the socket no longer wiggles! Hoorah!

So I've put the test bar in and measured the run out near the end (around 200mm) which is around 0.18 mm.

By 'measured the run out' (in case I have my terminology as wrong as usual) I mean I put a dial indicator on the bar and rotated the spindle carefully, seeing a deviation of 0.18mm. I suppose that means that the bar is off-centre at this point by 0.09mm or a bit under 4 thou.

Now there is still blue when I put the indicating taper in, but it's moving back in the spindle and spreading out. Still mainly on the same side. However, the blue is on the side where the rotation pushes the test bar further out, which says to me that scraping more will make the run out worse...

I'm not sure if this runout is reasonable for a mini-lathe. And I'm REALLY not sure if I can do anything about it.

The collective wisdom will be welcome!

Iain

Mine hasn't got any noticeable runout on the spindle or its taper.

The spindle may just be resting at three or four points...

Neil

which most likely means a spot right inside the taper where it's hard to see and reach..

But, sadly, that's a job for another day. Family are back from holiday which means it's time to cook tea,

Iain

You could have a go at checking the spindle itself with your indicator. Run the probe at a few points down the taper socket and note the results. Check the outside too. If the outside is OK your chucks are going to run true (or at least within their limits). That just leaves you with the socket. If you don't do much turning between centres you could just leave it at that and skim your soft centre when you do use it.

regards Martin

Hi, Martin.

I've found this a bit tricky. I can't actually see the blue at the end of the taper - not enough to be sure where it is and it's equally hard to see where the scraper is scraping.

After scraping away a bit more, I've got to the point where there are somewhat more bearing points, judged by where the blue is removed and the considerably increased coverage of the blue inside the taper. The test bar is now swinging a bit less ( .12 mm) and is quite rigid, so an improvement.

However, I've tested the face and outside of the chuck mount and there is definitely an offset. it's not much - 0.01 - 0.015 mm. Similarly an offset on the face.

I'm thinking that the spindle is actual out of line a little and the readings I'm getting more are more or less in line with a .15 mm run out 200mm further out.

I have to admit that there was a certain amount of force ('engineer's friend' involved when I had to remove the spindle to fit new gears, so it could well be my fault.

I've already turned my MT3 centre down when I started using it (and marked where to put it). Sounds like I may need to do it again now I've tuned things up a little.

Any suggestions or thoughts more than welcome.

Iain

Are you 100% sure the new bearings are fully seated? Have you adjusted the pre-load on the bearings?

Where are you measuring runout on the spindle? You should measure it on the chuck register, not on the flange. Schlesinger's limit for the chuck register is 0.01mm by the way.

Neil

The spindle is a VERY VERY tight fit. You may recall recommending a puller to extract it from the headstock many moons ago. that failed because the angle iron crossbar deformed rather than pull the spindle out. If I was to do this again I'd build a proper puller, but at the time...

In the end it was severe agression with a hammer that got it out (albeit with some buffer material). severe enough that the threads on the gear end got mashed I had to mill off 5 mm or so to refit the nuts.

I have no particular expectation that the roller bearings are correctly loaded. I was just damned please that I got the thing back together at all ( a lot of swearing and grunting to get the spindle anywhere near pulled in).

I was, by the way, replacing the gears not the bearing. The originals were put back on.

I'm not sure what else I can do.

Iain

YIKES! That is abuse.

Mine call came apart without any drama.

I wonder if you had some cold welding happen?

Do you have roller bearings or the original ball bearings?

Neil

Ball bearings. not changed them. If I were to replace the spindle, then I would probably replace the bearings too.

It was just about as hard to get the thing back in. Lots of clicks as it moves .5mm. Hard to tell when it's in place .

Iain

The "VERY VERY tight fit" of the spindle in the bearings may well have distorted the internal taper, if the taper extends back to where the front headstock bearing sits. If so, you probably won't do much further good with a scraper unless you are an absolute legend.

The ridge your blueing found a few millimetres into the mouth of the taper hole could perhaps also be the result of pounding on the end of the shaft to get it fitted into the tight bearing fit, either at the factory or in your subsequent dismantle/reassembly.

If it were me, I'd buy a new spindle and if necessary lightly emery tape down the spindle to make it a correct very light press fit in the bearing inner races.

While there, fit some taper roller bearings for a better result all round and refer to the manufacturer's data sheet for the exact fit information. If you want to do the best practice job on it, the bearings should be heated in oil to expand them and then basically slipped over the shaft.

Edited By Hopper on 26/10/2016 05:11:49

On the other hand, if you are not planning on using that internal morse taper for anything other than fitting a test bar, and you have no plans to use a MT3 collet chuck etc, you can test the headstock alignment without fitting a test bar into the Morse taper.

Simply put a piece of 1" diameter bar in the three jaw chuck, sticking out about four inches, with no tailstock centre. Then take a light cut over the length of the bar. Measure if the bar has a taper end to end and adjust headstock alignment until you can get a nice straight cut with no taper.

If you had to pound on the spindle with a hammer to get it back together, your bearings are toast. So you might as well go the whole new spindle and tapered roller bearings route. See the ARC Eurotrade website in the section called "ARTICLES" for a how to on changing the bearings.

Edited By Hopper on 26/10/2016 05:49:41

Edited By Hopper on 26/10/2016 05:51:39

Thanks, Hopper.

The bearings were super tight in the headstock as well and indeed the bearings support the taper 10mm or so inside the mouth, so any force on the chuck end could have provided some distortion.

Most of the aggression, though, was on the thin end when trying to get it out. I think I used a better padding on the way in - but I think it still needed some welly to get it in at all.

Will roller bearings fit more easily onto the spindle and headstock do you think? I'm aware of the bearing swap from ARC. Their description is how I knew how to dismantle the lathe.

Finally, I was planning to use 'Rolly's Dad's Method' of alignment. Which, simply put, asks you to rotate a bar which may be out of line and take the average of min and max readings at each point. It's a lot quicker than skimming a test bar. which I'm not so sure will work anyway when there is significant run-out.

Iain

This is mere idle speculation on my part, but : Is it plausible that the spindle-wall around the MT3 is rather thin, and that the beating it received has 'concertina'd' the spindle

I think I would carefully measure the outside diameter before proceeding much further.

MichaelG.