Lathe Headstock Alignment

[GregNixon]

How then would you check the alignment of the HS and then proceed to correcting twist? ie how do you tell if the misalignment is from the headstock or a twist in the bed.

 

[Neil Lickfold]

I recently worked on a lathe that the Headstock was not true to the bed. To get the bed level, it was effectively twisted to get the headstock end of the carriage level or reading the same error of the level , and then have the same error in the level at the tailstock end , and finally checked in the middle of the travel as well. After doing that, the lathe was turning a taper over 150mm. I used a piece of Ø38mm 7075 ali bar and a 0.2mm radius DCMT11 insert. I trimmed the end of the bar until it was turned and cleaned up for a distance of 10mm to take a reading, and set the dial to zero. Did the same at the chuck end and the diameter difference was 0.15mm. This lathe had Jacking screws at the back between the motor and the headstock casting. It looks to me that the Optimum does not have the adjusting screws, but I can't actually see your lathe. It appears that the access to the rear screws is from the back with the motor removed. In any case, you need to set up something that is turned over some distance and then put a DTI on that workpiece before undoing any screws. Have a horizontal indicator as well as one for the vertical is good if you have two indicators. Loosen off the 4 headstock screws, and see where or how much movement is of the headstock on the bed. Some have the V area not as a full register but is relieved and the headstock is actually sitting on the flat areas. You may be able to carefully tighten the headstock hold down screws and get it back into the proper alignment. Look for some sort of locking side screw that is designed to prevent the headbolt screws from coming loose as well before you just try to undo them. I am thinking that on your machine, the motor will need to be removed to be able to access the hold down screws or bolts. Keep in mind that if you have moved the machine, all the bed leveling will need to be done all over again to get it correct. It can take upto a day to level a lathe . I always start with the bed, then the headstock and last is the tail stock alignment. If your level is only accurate to 0.4mm/m it is a lot harder to get it to level , than if you also have a level that is 0.05mm/m. It does not have to be level like oil just stays in one spot over and area that has been silicon sprayed. Levelling means that the bed is in it's true aligned state. sometimes it may mean that some twisting load is required to get it there. On big machines, the variation from true level can influence the alignment of the headstock to the bed depending on how much weight is having a twisting motion or the type of stand or frame the machine is on or how its constructed. A lot of casting can also move over time since being made as well. Some lathes the feet at the back are used to twist just the headstock section of the lathe and have hold down holes on the back and front of the headstock like yours.
If it was me, I would undo the M12 bolts holding the headstock on and see what is going on. Pictures of the actual machine will help a lot.

 

[petertha]

How then would you check the alignment of the HS and then proceed to correcting twist? ie how do you tell if the misalignment is from the headstock or a twist in the bed.

That's a great question & maybe wordy answer, but I will try. Have a look at the picture I included in post#4 & the related description of how a test bar can be used. I attached picture of mine. Depending on your lathe you can use an MT sleeve adapter or just just buy the appropriate MT taper ended test bar. My spindle is MT5 so I use the MT3/MT5 adapter which I have confirmed is accurate. This allows me to use the MT3 test bar for subsequent tail stock work.

Pre-qualifiers: this relates to lathes of the type with removable HS where HS deviation is possible. This also assumes a newish machine with straight ways. The beds have not eroded away a hollow close to chuck. If that's the case, all bets are off, setup can only be a tradeoff compromise relative to chuck distance. Its an additional problem. And this discussion has zero to do with tailstock alignment. The TS must not be integrated into this initial test.

Personally, I think its best to get the bed as reasonably close to begin with even if it means somewhat undoing a prior setup. Why? because this is the very same datum surface that our DTI will ride along via the carriage when measuring the test bar pointing out into space completely independent of the ways. So as best you can, confirm the bed is not bending upward or downward relative to HS when viewed from side. And also is not twisting CW or CCW relative to HS when viewed from end. Unfortunately ($) to do this right is best accomplished with a precision level.

The test bar locates snugly into the HS MT socket and the extended portion exaggerates the HS axis relative to the 'mean' initial bed axis. Traverse an indicator down the length of the bar, both along the horizontal plane & vertical plane. We are attempting to discern if HS spindle axis is pointing inward/outward viewed from top and/or upward/downward viewed from the side relative to ways. If significant axis deviation is noted and dominated by the HS being mechanically out of alignment, that issue needs to be resolved first. If its a small deviation where we are satisfied HS alignment is as close as we will ever get, then we have arrived at the point of what I'll call the where classic methods like RDM come into play. The remainder of work is addressing lathe bed twist and ideally concluding with real cutting conditions. BTW if you doubt the significance of HS deviation as a source contribution, do the simple trig calculation for yourself. Consider a measly 0.001" over the length of HS block & extrapolate to the end of test bar. It translates into a significant amount of taper that actually represents a lot of equivalent lathe bed twist to accomplish. And something as measly as 0.001" HS rotation is nothing on a set screw, mere degrees of screw rotation depending on the thread pitch.

Here is my own summary. Others may have a different perspective. Taper cutting is a function of 2 somewhat independent sources. It could be 100% HS related, or it could be 100% lathe bed deviation related, or it could be some % blend of both. Unfortunately the sources could serve to counter/mask one another or they also could cascade & exaggerate one another. Adjustment may well be an iterative thing, but hopefully by standardizing one deviation source (ideally the one with lesser +/- limits) in order to conform the more dominant or significant source, we can proceed logically & efficiently & not chase our tails too much.

 

[HMEL]

Does anyone know how to adjust the Headstock Alignment on an Optimum TU2807V Lathe

Before I would do anything with the headstock I would check the tail stock. If not using the tailstock, then check the chuck, You can lift the headstock on most lathes maybe clean the ways and carefully reset it. This requires a lot of work. And I would try to get a hold of the manufacture and ask what adjustments and checks they recommend. Its easy to really mess an alignment up without the tools to set it up.

 

[colin james]

I kind of chuckled when you use the word 'just'. This entire post is actually quite typical of many similar 'lathe levelling' posts one encounters on hobbyist forums. Its a familiar script. Someone notices they have taper cutting issue & the immediate solution offered is to jack the feet & twist the bed. The lathe type often does not even come up in in discussion as a qualifier. Or if it does, few people recognize that the HS is bolted to the ways & any misalignment there may well be the dominant if not entire source of taper cutting discrepancy. Then there is mysterious absence of the operator/parts manual even mentioning adjustment screws being there. If one has a larger lathe, say in 14" range, pulling the HS off & shimming is not exactly a trivial job although I recognize hobbyist skill sets vary. I'd certainly want to be sure before embarking down that path. Which is is exactly why a MT test bar is a valuable asset.

Just a guess on my part but maybe 20-30 years ago the chance of a hobbyist lathe being of the 'integrated' type (where no spindle alignment correction is possible) was certainly much higher than today. And yes, in that case you are stuck with what you have. Lathe twist is the only option & this presumes other factors like wear are good or minimal. But I would venture to guess a much higher percentage of hobby lathes in shops today are of the bolt-on style. Here is a popular vendor, quite typical of others including say Grizzly or other Asian importers. I count a dozen lathes from the smallest up to and including 14" swing which are bolt-on HS. Now if the factory did their job 100% and/or we have verified ourselves that HS is perfectly aligned in both vertical & horizontal plane, then we have arrived at the equivalent point of an 'integrated' lathe and hopefully relatively minor bed twist alteration may remedy remaining slight taper cutting. But if you have HS alignment problem & ignore it, and proceed to lathe twist, you are possibly attempting to fix the wrong problem or overcompensating which isn't the best solution. Of course this is a hobby so everyone is welcome to proceed as they see best. I'm just offering another perspective to the discussion.

https://www.precisionmatthews.com/product-category/lathes/12-14-swing/

I checked out the link and had a look at all their Lathes and mine appears similar in construction and size to the Precision Matthews PM-1030V
I have had a further look and I have had to remove the Electronics to find that I also have to remove the electronics cabinet from the rear of
the Lathe to get access to the 2 off Socket Head Screws which secure the headstock to the Rear of the Bed.
And to access the front 2 off Socket Head Screws, I have to remove the front cover to get to the screws inside the front of the head stock.
This is poor design, it should have been designed so all the screws could have been accessed from outside the Headstock, and not inside
as this makes this a major job, and not a small easy adjustment as some other machines.
Thanks for your help.

 

[colin james]

That's a great question & maybe wordy answer, but I will try. Have a look at the picture I included in post#4 & the related description of how a test bar can be used. I attached picture of mine. Depending on your lathe you can use an MT sleeve adapter or just just buy the appropriate MT taper ended test bar. My spindle is MT5 so I use the MT3/MT5 adapter which I have confirmed is accurate. This allows me to use the MT3 test bar for subsequent tail stock work.

Pre-qualifiers: this relates to lathes of the type with removable HS where HS deviation is possible. This also assumes a newish machine with straight ways. The beds have not eroded away a hollow close to chuck. If that's the case, all bets are off, setup can only be a tradeoff compromise relative to chuck distance. Its an additional problem. And this discussion has zero to do with tailstock alignment. The TS must not be integrated into this initial test.

Personally, I think its best to get the bed as reasonably close to begin with even if it means somewhat undoing a prior setup. Why? because this is the very same datum surface that our DTI will ride along via the carriage when measuring the test bar pointing out into space completely independent of the ways. So as best you can, confirm the bed is not bending upward or downward relative to HS when viewed from side. And also is not twisting CW or CCW relative to HS when viewed from end. Unfortunately ($) to do this right is best accomplished with a precision level.

The test bar locates snugly into the HS MT socket and the extended portion exaggerates the HS axis relative to the 'mean' initial bed axis. Traverse an indicator down the length of the bar, both along the horizontal plane & vertical plane. We are attempting to discern if HS spindle axis is pointing inward/outward viewed from top and/or upward/downward viewed from the side relative to ways. If significant axis deviation is noted and dominated by the HS being mechanically out of alignment, that issue needs to be resolved first. If its a small deviation where we are satisfied HS alignment is as close as we will ever get, then we have arrived at the point of what I'll call the where classic methods like RDM come into play. The remainder of work is addressing lathe bed twist and ideally concluding with real cutting conditions. BTW if you doubt the significance of HS deviation as a source contribution, do the simple trig calculation for yourself. Consider a measly 0.001" over the length of HS block & extrapolate to the end of test bar. It translates into a significant amount of taper that actually represents a lot of equivalent lathe bed twist to accomplish. And something as measly as 0.001" HS rotation is nothing on a set screw, mere degrees of screw rotation depending on the thread pitch.

Here is my own summary. Others may have a different perspective. Taper cutting is a function of 2 somewhat independent sources. It could be 100% HS related, or it could be 100% lathe bed deviation related, or it could be some % blend of both. Unfortunately the sources could serve to counter/mask one another or they also could cascade & exaggerate one another. Adjustment may well be an iterative thing, but hopefully by standardizing one deviation source (ideally the one with lesser +/- limits) in order to conform the more dominant or significant source, we can proceed logically & efficiently & not chase our tails too much.

The Lathe has hardly been used so their is no twist or wear in the bed as I have already checked this with a precision machine level, also the head stock is removable and as far as I can see the problem is only the alignment of the head stock which appears to have moved.
It is just disappointing that all the hold down screws are located in the front and rear compartments of the head stock, behind the front cover at the front, and behind the electronics box at the back, which can only be removed after removing the electronics.
I am afraid this design leaves a bit to be desired, a lot of extra work which could have been avoided by having the screw go in from under the bed, instead of from inside the 2 compartments inside the head stock.

 

[Steamchick]

Just reset my Unimat - by inserting a piece of true bar in the tailstock chuck, then clamping in the headstock chuck, while the headstock was free to rotate. I then tightened the clamp bolt on the headstock, inserted a couple of old fine drills, and tried to take a picture. It was noticably not true beforehand, but looks better now (through a lens). Sorry my camera won't focus that close, but you should "get the picture".
K2

 

[retailer]

I'm with petertha on this one, my Taiwan 12x36 lathe was very turning slightly tapered - around 1.5 thou in 6 inches (0.035mm in 150mm) - to fix the issue I listened to all of the internet experts and purchased a precision level to take the 'twist' out of the bed - the level was seriously sensitive - just using the tool post to steady myself as I looked over the level was enough to send the bubble to one side, 2 full days and evenings later I achieved nothing and gave up on the level and then tried rollie-day-method with the same result, so I looked at the headstock alignment.

My lathe has a bolt on headstock and 2 adjuster screws that are quite hard to get to, I had to make up a special long hex key that fitted in my ratchet socket spanner just to turn the adjusters, - peterthe is right when he says the adjustment is very sensitive, but I managed it in one evening and my lathe now turns almost perfectly at just over 0.5 thou (.014mm) over 12 inches or 300mm.