Setting up a wood turning lathe

[OldWood]

I expect this has been addressed before but I'm not finding anything in a search.

I've got a good "old fashioned" CI base lathe (Wadkins Bursgreen). I've had it for a year or so (first lathe) and was slightly surprised the other day when a shaft I was turning wouldn't go back onto the centres and run true - it wasn't at all important as it was a small turning off operation.

Anyway I put a spirit level across the bed and packed up the head end by 7mm to get a near level across the bed, and found the tailstock end only wanted 1.5mm. Now I can access a super sensitive spirit level - I've used in the past when setting up my metal working lathe - but I'm sure there's a more woody related technique for getting the bed true and the centres to align over its length - question mark !!

Rob

 

[CHJ]

..... I've had it for a year or so (first lathe) and was slightly surprised the other day when a shaft I was turning wouldn't go back onto the centres and run true - it wasn't at all important as it was a small turning off operation.
.....Rob

Although I fully agree on the desirability to have any lathe or machine set up with its main reference surface level and without any twist I'm at a bit of a loss as to why a centres mounted spindle should not run true.
A traditional Pole Lathe is maybe the most unlikely animal to have constant end centre references but anything pivoting between them will be bound to turn true.
Are you sure that the run out when remounted was not due to the wood moving whilst off the lathe due to stresses released by the turning.

 

[OldWood]

Although I fully agree on the desirability to have any lathe or machine set up with its main reference surface level and without any twist I'm at a bit of a loss as to why a centres mounted spindle should not run true.
A traditional Pole Lathe is maybe the most unlikely animal to have constant end centre references but anything pivoting between them will be bound to turn true.
Are you sure that the run out when remounted was not due to the wood moving whilst off the lathe due to stresses released by the turning.

Chas
I will go back in the morning and take the packing pieces out and repeat my run-out measurement, but it is now +/- 5 thou. over 25 inches. The longest piece of gash wood I could find this afternoon was an old sledge hammer shaft (therefore likely to be ash), which I centred at both ends and turned a couple of inches true at both ends. I then remounted it the other way round and measured the run out.

In bowl work and the likes such an inaccuracy is insignificant, ie +/- 1 thou over 5 inches, but with longer work pieces I was pursuing what was regarded as acceptable, and how 'professionals', which includes committed amateurs, went about minimising the offset.

The 18" x 2" piece of wood I'd been working on was well seasoned cherry that has been on and off the lathe over a period of about about 8 weeks and it was only when turned round, more by chance than anything else, and re-centred to do the final parting off of a spigot, that the very noticeable eccentricity became apparent. If it was wood movement then the irregularity would have been at a maximum towards the centre of the piece, but in fact this maximum was at the back centre.

Rob

 

[CHJ]

Don't go messing your machine setup about if you have it levelled and true, that's a retrograde step.

But any object supported between two good centre references should spin true no mater which end is presented to the the headstock end. On metal turning lathes this is the way to ensure you can do just that for accurate turning.
On a metal lathe inaccuracys of alignment will only result in a taper end to end but not run out of true if turned around, you don't even get this on a wood lathe because the tool post, (your hands) move the tools to compensate.

I presume the headstock drive was via a 4 pronged drive centre, not in a chuck.
How did you engage your drive centre in the old tailstock end to get drive function. (assuming you are not using Steb centres)
Hitting a pronged drive centre into a plain tailstock centre may have drifted off due to grain presentation.

 

[OldWood]

Don't go messing your machine setup about if you have it levelled and true, that's a retrograde step.

Hi Chas

But that with all due respect is where I came in - I'm really asking the question of how does one level and true the lathe without things like an ultra sensitive level ? That actually is what my original question was !! :? But then perhaps, unlike a metal working lathe where the tool is required to travel absolutely parallel to the workpiece, this isn't important in wood turning.

Unfortunately we've ended up discussing something which may have been a faulty analysis on my part, but did lead me to investigate the machine set-up and finding what would appear to be a degree of twist in the bed.

What is interesting is that a workpiece running between centres cannot be guaranteed to run true if turned end to end. This would suggest that my test piece of today with it's +/- 5 thou error is pretty good particularly as the piece that started all of this was well over 1 mm out - ie +/-20 thou.

Rob



Rob

 

[CHJ]

...Hi Chas

But that with all due respect is where I came in - I'm really asking the question of how does one level and true the lathe without things like an ultra sensitive level ? That actually is what my original question was !! :? But then perhaps, unlike a metal working lathe where the tool is required to travel absolutely parallel to the workpiece, this isn't important in wood turning.

Not as important as you say, a standard reasonable quality building level should be adequate for a wood lathe, just check it end to end on a flat surface and note any bubble errors before you use it.
The only error that is likely to manifest itself if the bed or bars are way out is in boring holes from the tailstock as mis-alignment of tailstock quill could result in drilling out of line with the headstock, and this can vary at different points along the bed if it's twisted.

...What is interesting is that a workpiece running between centres cannot be guaranteed to run true if turned end to end. This would suggest that my test piece of today with it's +/- 5 thou error is pretty good particularly as the piece that started all of this was well over 1 mm out - ie +/-20 thou.

Unless you are reinforcing the centres in the wood in some way and applying the tailstock pressure to exactly the same pressure then I would not be at all surprised to see 10-20 thou. difference, you may have better register swapping between Steb centres but this is wood turning after all, wood left overnight is likely to move more than the tolerances you are talking about.