Spindle not running true

I have recently stripped down to paint etc my old Grayson lathe. Having just built it back up, I have noticed that when i place a dead centre in the spindle bore it oscillates by up to 3mm! There is no burrs and on looking in towards the spindle when running the bore appears to be oscillating which it wasnt doing before. I placed a dti on the nose itself which is showi g a runout of just 1/2 thou albeit when running a chuck its runout is around 4 thou! Is the spindle somehow became slightly bent?

Forget the spindle bending, that can't happen spontaneously, but what else has gone wrong I can't think other than you now have grossly slack bearings which are allowing the spindle to flop about

Sorry

Brian

1. Most likely taper in your spindle is not made correctly - will need reboring. This is to address horrible runout of dead center. If it is really 3mm spindle may actually need *remaking*.

2. Bearings will need replacement and/or rescraping. Installation of chuck increases runout - worn out bearings.

Bent spindles are very rare and 3mm runout on dead center in the spindle taper combined with relatively slight runout on spindle nose suggests something else.

Martin

Thanks for the quick replies.

Maybe the 3mm is exaggerated by myself but it is certainly way excessive on the mt2 internal taper This old lathe runs in split brass bushes and there dossnt seem to be wear or play however i take it on board they may have.

Would it make any difference using a mt2 reamer to clean it up?

I never had any noticeable problems before. It isstrange that the internal bore is seen to be oscillating. 😥

Something silly has happened. If it was running well before and doesn't now, by what is obviously a gross degree of error then it cannot be a fault with the spindle which will have been set aside while painting took place.

You have forgotten to fit something perhaps, Can you shake the spindle about or has it collected some foreign object in the bore in the interim? Common sense says it has to be something of that kind, major faults just don't happen in this way

I would definitely hold off on reaming or anything drastic, I'm sure such work is completely unnecessary

Brian

Edited By Brian Wood on 24/03/2018 19:20:49

Just spent 2 hrs there with various dti gauges and the mt2 taper is running untrue. Not missed anything there is very little to this lathe.

If you have not missed anything during re-assembly and it's set up correctly maybe the question is have you included something that shouldn't be there like a bit of swarf or some other foreign matter? As Brian suggests.

Edited By Martin W on 24/03/2018 23:21:21

Have you tried a clock gauge on the surface of the female taper, ie without the dead centre fitted?

For real precision you can always skim the dead centre in situ, the one at te headstock end should not be hardened.

If everything was running true before disassembly obviously something has gone awry on reassembly .

Are you 100% certain the bearings went back in exactly same way / orientation they came out ? Was there any operations on the spindle that required a press like having to press a pulley off or on etc ? Did you use any substantial amount of heat on anything ? I would set the spindle up on some v blocks and check it for straightness and if the bore is running out .

Before you go machining anything double check everything as machining the spindle bore should be a last resort .

Is there any tight spots when you rotate the spindle by hand ? If it were bent i would think there would be one hell of a tight spot !

When you indicate the taper in the spindle bore push the spindle back towards the rear of the headstock with with your hand as you rotate the spindle to ensure any endfloat is taken out of the equation .

Lastly the centre you used have you used it before ? If not is there the possibility that it came of a different lathe and may have been machined to run true on that machine ?

Have you clocked both the end of the spindle taper and further down? The nose could be clocking true but if out of line further down you will then see wobble.

If it was OK before then don't start thinking of taking any metal off the spindle at this stage

Is your Grayson one of the models where the headstock is held by three bolts and detachable from the base?

If it is, and you dismantled it, are you sure the spindle is on the centreline of the lathe when you put it back together?

The only correct way of establishing for sure that the spindle is still true throughout is to lay it on it's journals in a pair of identical Vee block grooves and index it round while clocking all the relevant parameters. That will prove once and for all that the spindle is either bent [most unlikely] or just as it should be [much more to be expected]

In manufacture it would have been made as a machined component mounted once with all the nose detail plus taper machined at one setting so that everything was concentric to everything else.

I don't like to correct misconceptions, but David Standing's suggestion will only show an offset on the spindle for a misaligned headstock and not reveal variable run out, it will still be guided by the bearings.

XD 351 and Martin W have a better forensic approach and I am sure this strange error will be explained in carefully reviewing all the rebuilding. It is a simple lathe and there is apparently little to go wrong at first sight, but somehow that has happened here

I am sure I am not alone in wanting to know what has gone wrong here; how can taking a working lathe like this apart for painting result in a 'fault' of this kind on rebuild? It makes no sense at all. Something has changed dramatically in the course of those operations.

Regards

Brian

Have you cross-checked with a different centre?

Brian

No misconceptions here

I wasn't suggesting that the spindle centre line, if out, would be the cause of the apparent offset spindle bore because obviously it wouldn't. I was suggesting that if it was out of line with the lathe centreline, it would make measuring and finding a solution more difficult - it would take one variable out of the equation if it is known that the spindle axis is in line with the lathe centreline.

Shafts can bend spontaneously due release of stress's. I worked for a large sugar mill and we removed a very large old lathe, which twisted to hell when we undid all the bed bolts. Also long carnakshafts have to be stored vertically to stop them twisting.

OK, I see where you were going David, I misread your thinking. Sorry.

Vintage Engineer, The spindle on this lathe is a short relatively thick walled steel tube maybe one foot long that we are told was straight before it was taken out for machine painting and now runs out of true; surely not in the same league at all I venture to suggest as big crankshafts and the bed twisting you describe.

By the same token, big power station turbines have to be kept moving round slowly by barring gear to prevent them from sagging under their own weight during shut downs.

Brian

No problem Brian, I probably didn't make myself clear anyway

I'd agree with Jason - if it was OK before and you didn't brutalise anything when it was apart (includes misplacing shims and so forth), then I'd look at what has changed.

I would check the assembly very carefully, and I think I'd have a scout about to see if there is a blob of paint interfering with anything, or preventing some housing from closing up as it should.

I think it would be most unwise to start hacking at metal until you identify what is wrong.

Edited By mgj on 30/03/2018 18:22:43

Thanks for the replies .after much ado, I have cleaned out the female mt2 using a mt2 finishing reamer and its certainly helped. I think that there had been slight damage at the start of this taper throwing it out ie the centre etc not seating correctly.

I now have it within a thou on the nose.Thanks again for the suggestions

Hello Aitken,

Thank you for completing the story, it is always useful to others to know what you did to put it right and clear up what at seemed to be a baffling problem for the rest of us when you first called for help. You might even get the residual error down to near negligible if you persevere.

Regards

Brian