PDA

View Full Version : Alignment, how much does it matter?



hughie
6th July 2021, 10:22 AM
Ok this comes up from time to time the alignment of the tailstock and head stock.

So the question is 'If this matters, why?'

Ok from an engineering perspective, yes, its poor machining I agree. My basis for this agreement is 50 years in the engineering game in half dozen countries and a dozen industries or more. Its seems this problem is not always at the lower end of the market. It seems to effect all make and models at some stage, I have seen it Stubby lathes as an example.

But seeing that woodturning is a free hand craft I'm darned if I can see at the end of the day how it effects the outcome. I run two lathes both hand built by myself, one is a bowl lathe 700 bed swings 800+, the other with fully extended bed 1.5m swings 800+. The tailstock on the bowl is the most basic as you can get. Is it aligned hell no. Its cobbled together out of angle iron and threaded rod to get me out of trouble one day, do I still use it? Yes, when ever its required. The big lathe I have as yet to build the tailstock, it will be a great deal more sophisticated than the other. Will be aligned? well as much as my very basic engineering shop can produce. I certainly wont be losing any sleep over it if its out.

So ladies and gentlemen your thoughts

LanceC
6th July 2021, 10:40 AM
I think that if you're turning between centres, when placing your rest perpendicular to the axis of rotation, I can't see there would be an issue.

I do see an issue when using a chuck with tailstock support. By fixing the stock to one end (the chuck) it is locked to that axis of rotation. An off-axis support will be unable to hold the stock steady. The tip of the tailstock centre would scribe a circle on the other end of the stock. The effect will be more pronounced as the stock length increases.

Blackforester
6th July 2021, 11:20 AM
What Lance said. But the other issue is when it comes to boring holes, particularly long ones. The tailstock spindle axis has to be an extension of the headstock axis both vertically and horizontally with no "kink" in it, otherwise the drill will not stay centered and will drift.

Skew ChiDAMN!!
6th July 2021, 11:50 AM
I think that, like most tasks, it depends on what you're doing.

It's a simple fact that the greater the distance between the head- & tail-stock, the less any misalignment matters. I'm fairly sure that a 1mm misalignment will cause about a 0.5° "wobble" at 100mm spacing and only around 0.1° at 500mm. (I'd have to do the math to be sure.)

That is negligible when it comes to most forms of turning; where it starts to matter is when turning short items. eg. pens, lace bobbins or shorter.

Even then, your choice of method probably matters more. I've found that pens turned between conic centres tend to be more accurate than pens turned on a fixed mandrel. Given the same degree of misalignment, of course! I put it down to the mandrel trying to flex and bow, exaggerating the existing error. (Ever tried turning a trembleur w/out steadies? :rolleyes:)


Me, I don't worry about it. My current set-up has around a 1mm misalignment, although when I bought it I'd checked (multiple times) for a perfect kiss. I suspect it's more the cheap tail spur I'm using rather than a tailstock issue, but I don't care enough to chase the matter down.

LanceC
6th July 2021, 12:07 PM
That is negligible when it comes to most forms of turning; where it starts to matter is when turning short items. eg. pens, lace bobbins or shorter.

I'm struggling to understand this. Can you explain it in a little more detail?

In thinking about this some more, I think it matters to different extents bases on the operation.

Turning between centres: not at all.

Chucking and using the tailstock for support: We don't need to align both axes. All we're looking for is that the tip of the tailstock centre falls on the headstock axis, as the support is just a point in space.

Boring: Head and tailstock need to share a common axis as the cutting point of the drill will be moving linearly, and if not following the headstock axis will bind or drill an oversized hole.

Further to Skew's post, it would be interesting to know the tolerances required before any perceivable issues arise.

chuck1
6th July 2021, 03:12 PM
With turning long spindles like balusters and posts as long as you can get the timber to run true is doesn't matter about the centres being aligned.
I get the square to run true as possible by eye, so having timber dressed square is more important where the turning has squares top and bottom.

But turning pens on a mandrel close enough isn't good enough but there more experienced pen turner than myself

Mobyturns
6th July 2021, 03:51 PM
As Skew mentioned - it depends upon the task.

Not all spindle turning is turned "between centers" as in using a drive center and a cone center.

I regularly use ER25 & ER32 collet chucks in conjunction with a quite small (< 6mm) custom made cone center for pen blanks, spin tops, and delicate finials. Any horizontal or vertical mis-alignment is very noticeable as it tends to induce a flex into delicate pieces which then actually end up elliptical in cross section. Using a pen mandrel on a lathe with poorly aligned centers will cause the same issues.

The center on the fixed end (head stock) remains relatively accurately centered, however the axis of the blank is not "straight" between centers i.e. the turning axis and blank axis are not collinear. The shorter the blank and on more delicate work, the more noticeable the error becomes.

For pen makers imo it is also the number one cause of poorly fitting / matching pen tubes to pen components on assembly. Careful measurement of tube + blank wall thickness often shows an easily measureable difference around the circumference of the tube/blank.

I drill all my custom made inlay pen blanks on the lathe (1992 Woodfast M908) using the ER collet chucks (Vermec) and can achieve quite accurate concentricity when drilling from both ends, to prevent blowout & other issues. The Woodfast lathe has excellent alignment and the fitment of a "new" tailstock recently purchased on this forum also has excellent alignment.

Poorly aligned centers are the result of "cheap" manufacturing processes and cutting costs, which I might add are often not reflected in the purchase price.

powderpost
6th July 2021, 09:49 PM
I regularly make walking sticks in one piece, as well as two and three pieces up to about 850mm. The head stock and tail stock centre do not meet perfectly. I have not yet found one become anything other than round. In my opinion for spindle turning it does not matter one bit if the centres are not perfectly in line. I also make some fairly complicated urn or vase shapes up to about 300mm and do that in two halves. I use a threaded tail stock to rejoin the two halves. Even though it is a Woodfast M908 the centres do not meet perfectly, but I can still achieve a reasonable result.

My conclusion is that perfect alignment is not absolutely necessary, there can be some tolerance, how much? I have never bothered to measure the difference, I have never had the need to. Getting the centres accurately marked is by far more important. And, I do turn lace bobbins out of bone as well as a lot of pens on the same lathe.

Jim

Mobyturns
6th July 2021, 11:38 PM
Jon Siegel's article on "Tuning Your Lathe" and Figure 1 may help to understand lathe alignment errors. Articles by Jon — Big Tree Tools (http://www.bigtreetools.com/articles)

Jon gives a very good explanation of the errors and some guidance on how to correct them.

The best way to understand how the miss-alignment of the head stock and tail stock axes will affect your turnings is to make a drawing that exaggerates the errors.

Instead of trying to understand how a 0.5 mm offset affects things, make the offset 5 or 10 mm in the drawing, or as Jon Siegel has done in Figure 1.

A very good illustration of how turners have used the eccentric turning center (essentially an exaggerated miss-alignment) is for elliptical turning using Volmer's elliptical turning chuck. OVALTURNING (volmer---ovaldrehen.de) (http://www.volmer---ovaldrehen.de/Ovalturning2A.pdf)

Being a surveyor I have a very good understanding of theodolite "instrument errors," how they affect observations, and how to minimize the errors. Similar principles apply to lathes. Some errors can be compensated for, some can't. Eccentricity of the turning axis cannot be compensated for, (it can be minimized in some instances) but it can be employed to advantage in some instances.

If you place a bowl blank tenon in a scroll chuck then bring up the tail stock that has a 1.5 mm offset in either the horizontal or vertical plane (or both) you will get vibration. This occurs because the tail stock center is trying to force the outboard face of the bowl blank to revolve around its center and not the scroll chuck center, but the bowl blank is quite rigid.

With spindle projects held in a scroll chuck or collet chuck or old fashioned "cup chuck" in the head stock (constrained end), the offset at the tail stock end will more than likely force the spindle blank to bend and "run" on the tail stock axis (free end). That bend will become more pronounced as the blank becomes thinner. Similar errors/effects occur with a pen mandrel, the MT2 end is "constrained" the tail stock cone center "free". Pen mandrel savers complicate things slightly.

When turning between a center in both head stock and tail stock the "bending moment" does not exist, i.e. the blank and turning axes for all practical purposes are the same, similar situation turning a pen blank between "cone centers" - both ends running "free."

Drilling accurately on the lathe becomes near impossible with any significant offset in either the horizontal or vertical plane (or both) and / or the head stock and tail stock axes are not collinear (lying on a single line.) This is because both ends are effectively "constrained." When things are "constrained" something has to give, that something may be the drill bit trying to bend to conform with the head stock axis, or it becomes vibration etc.

Ironwood
7th July 2021, 09:14 AM
How much it matters, totally depends on what you are turning.
For roughing down a big out of balance bowl blank using tailstock support, it wouldn’t matter too much at all. But try to make kitless pens on a lathe that is misaligned, and things start to go pear shaped pretty quickly. This is why many people prefer to use a metal lathe for this type of work, they are usually more accurate.

artful bodger
7th July 2021, 07:21 PM
Don't forget that some of the Wadkin patternmaking lathes had tail stocks where you could purposely move the dead centre away from the centre line so you could turn nice straight tapers with the saddle.
If you really want to prove the point that it does not matter, you could move the tailstock centre as far off centre as possible and still turn a perfect cylinder (by hand) by adjusting the tool rest(not the saddle)to suit.

hughie
7th July 2021, 09:27 PM
I regularly make walking sticks in one piece, as well as two and three pieces up to about 850mm. The head stock and tail stock centre do not meet perfectly. I have not yet found one become anything other than round. In my opinion for spindle turning it does not matter one bit if the centres are not perfectly in line. I also make some fairly complicated urn or vase shapes up to about 300mm and do that in two halves. I use a threaded tail stock to rejoin the two halves. Even though it is a Woodfast M908 the centres do not meet perfectly, but I can still achieve a reasonable result.

My conclusion is that perfect alignment is not absolutely necessary, there can be some tolerance, how much? I have never bothered to measure the difference, I have never had the need to. Getting the centres accurately marked is by far more important. And, I do turn lace bobbins out of bone as well as a lot of pens on the same lathe.

Jim

This is pretty well where I sit as well, nothing I do with my lathes needs any significant alignment that I expected to be there in my engineering side of life. But having said that it doesn't mean there isn't a need for perfect alignment at some point. I am just very doubtful if I will find it .

Mobyturns
8th July 2021, 10:36 AM
This is pretty well where I sit as well, nothing I do with my lathes needs any significant alignment that I expected to be there in my engineering side of life. But having said that it doesn't mean there isn't a need for perfect alignment at some point. I am just very doubtful if I will find it .

Ain't that the truth with all the "budget clones" on the market. "Perfect Alignment" - does it exist on any lathe? Even the best metal lathes manufactured to the highest manufacturing and rigorous quality control standards still require "tuning" for optimal performance.

In my surveying career, we studied and recognized the sources of potential errors, with the instrument, observer, environment, ancillary equipment etc and put in place standard procedures to eliminate them, or minimize the effects of them.

At some point there is a limit to the accuracy that can be attained with the equipment and processes and there is not much point in "pushing" to get more accuracy. That's why surveyors introduced "class" and "order" for equipment and processes. A cheap builders level or theodolite (Class D 5th order) produces acceptable results in that application, but if you want "Class A" "00 Order" results - the highest precision and accuracy attainable then it comes into the realm of highly specialized staff, and very high quality, extremely expensive gear.

As I like making "small stuff" I knew that finding a lathe with the "least amount of problems" was the way to go. Fifteen or so years back I deliberately went searching for my Woodfast M908 and got lucky with a well cared for lathe. When I purchased the "new" tail stock from the WWF Marketplace it was with some trepidation, but I had a fair degree of confidence in Australian Manufacturers of that era! No issues at all and to be honest not any significant difference in alignment to the "old" original one other than fair wear & tear - I do a lot of drilling on the lathe.

Our old Nova Mercury was OK but no where near as well aligned as the M908. Both the VL100 we purchased of WWF Market place and the new VL150 do not have any significant issues though I have heard reports that others have had some niggly alignment issues.

Horses for courses, but you have to recognized the limits of what you have.

hughie
8th July 2021, 10:49 AM
Ain't that the truth with all the "budget clones" on the market. "Perfect Alignment" - does it exist on any lathe? Even the best metal lathes manufactured to the highest manufacturing and rigorous quality control standards still require "tuning" for optimal performance.

In my surveying career, we studied and recognized the sources of potential errors, with the instrument, observer, environment, ancillary equipment etc and put in place standard procedures to eliminate them, or minimize the effects of them.

At some point there is a limit to the accuracy that can be attained with the equipment and processes and there is not much point in "pushing" to get more accuracy. That's why surveyors introduced "class" and "order" for equipment and processes. A cheap builders level or theodolite (Class D 5th order) produces acceptable results in that application, but if you want "Class A" "00 Order" results - the highest precision and accuracy attainable then it comes into the realm of highly specialized staff, and very high quality, extremely expensive gear.

As I like making "small stuff" I knew that finding a lathe with the "least amount of problems" was the way to go. Fifteen or so years back I deliberately went searching for my Woodfast M908 and got lucky with a well cared for lathe. When I purchased the "new" tail stock from the WWF Marketplace it was with some trepidation, but I had a fair degree of confidence in Australian Manufacturers of that era! No issues at all and to be honest not any significant difference in alignment to the "old" original one other than fair wear & tear - I do a lot of drilling on the lathe.

Our old Nova Mercury was OK but no where near as well aligned as the M908. Both the VL100 we purchased of WWF Market place and the new VL150 do not have any significant issues though I have heard reports that others have had some niggly alignment issues.

Horses for courses, but you have to recognized the limits of what you have.

Yup I rebuilt from the ground up a 1985 Woodfast 400, New three phase motor, VFD , remote control. very well built couldnt fault it. I actually posted the lathe here. Finally moved it ON as the swing was a bit small for me. But as you say from that era well designed well built to last a life time, no planned obsolescence. If think if I had the chance to pick up a 410 model I probaly would jump at it.

NeilS
8th July 2021, 11:06 AM
I had a go at the topic awhile back... there are a number of types/causes for misalignment and potential fixes (or not).

https://www.woodworkforums.com/f8/checking-centre-alignment-secondhand-lathe-229871#post2153205

I have found the misalignment on the lathe at the local Mens Shed a real PIT*. I had a go at realignment but it still gives oversized holes when drilling.

The only issue I have had with the Woodfasts in my workshops was when I got twist in one lathe from an uneven floor. Shimming up the low leg/s sorted that out and it runs in perfect alignment since doing that. Sighting along the full length of the bed through a hollow tail centre placed in the headstock can diagnose the degree of misalignment and adjustment needed. Twist will not necessarily be evident when testing by kissing centre points.

IME, unless the misalignment is significant it isn't an issue for many turning tasks.

Anyway, I do most of my turning outboard without a tailstock so alignment is not such an issue for me for most of what I do.

Mobyturns
8th July 2021, 11:50 AM
I had a go at realignment but it still gives oversized holes when drilling.

That is probably the most commonly encountered "issue" with lathe misalignment. It amazes me how many turners just push on and ignore the problem.

I had a read of your thread, excellent.

Skew ChiDAMN!!
8th July 2021, 11:57 AM
AFAIK, there are three types of misalignment for tail stocks:

Simple misalignment, when every time you slide the stock away and back, the misalignment is the same every time.
eg. Xmm too high & Ymm to the left, etc., etc.
Often this can be remedied by re-aligning with either adjusters or shims. Slop in the ways. Normally any vertical misalignment will remain constant, but it may vary in the left/right (or back/forth if you prefer) direction.
eg. Xmm too high, but varies between Xmm too far left and Xmm too far right.
This typically needs the gap between the ways to be re-machined parallel and the tailstock re-adjusted/modified to suit. It's also a PITA, and is WHY we often recommend the kiss test when looking for lathes to buy. You do not want this lathe. :no: A poorly made live tail spur. This is obvious, as all you need to do is bring the points up to kiss and then rotate the live spur. The point will orbit around the drive spur point in a circle of fixed radius.
Easily fixed... replace the spur if it bothers you that much. Just don't mistake this for a tailstock issue, as you can spend a lot of money trying to 'fix' a non-existing problem.

As I alluded to in my earlier post, I suspected the latter in my case. As a matter of curiosity, I checked it yesterday and yup. It's the culprit.

It is of the type where it's a stamped cone pressed onto a bearing... and apparently it's a bit 'cam shaped.'

'Twas about 1mm out of true on the lathe, but I could bring it back to true by removing the cap from the bearing, rotating it about 45° and pressing it back on. Admittedly it took a few tries to get it positioned correctly.

I anticipate that it will rotate out of position again in the future, but as I also said earlier, that doesn't worry me. This particular spur is one I only use on long spindles because of it's short length. Using it, I can fit spindle blanks up to 3cm longer than I can with my good double-bearing spur on my existing bed w/out the need to eff around with an extension bed. (Been doing a lot of copy turning of balusters. Booooooooring!)

Fumbler
16th July 2021, 06:20 AM
I'm not going to follow up with exactly what everyone else has said except 1 point. If something is not square, it will all come out in the wash, if no changes to the machinery is made in between cuts. Here is a rolling pin, made with (post cut discovery) a saw blade that cut slight off 90deg. after glue up, and each subsequent rotation from the same start point, the celtic knot is absolutely perfect in joint size and alignment (due to exact replacement of removed material) and the lathe slightly off in alignment 0.5mm. not noticable over the 900mm length.

498126

Fumbler
16th July 2021, 06:22 AM
That is probably the most commonly encountered "issue" with lathe misalignment. It amazes me how many turners just push on and ignore the problem.

I had a read of your thread, excellent.

yes me, i just use a smaller drill bit.

Mobyturns
16th July 2021, 08:37 AM
I'm not going to follow up with exactly what everyone else has said except 1 point. If something is not square, it will all come out in the wash, if no changes to the machinery is made in between cuts. Here is a rolling pin, made with (post cut discovery) a saw blade that cut slight off 90deg. after glue up, and each subsequent rotation from the same start point, the celtic knot is absolutely perfect in joint size and alignment (due to exact replacement of removed material) and the lathe slightly off in alignment 0.5mm. not noticable over the 900mm length.



Yes, that is correct to a point. However the nature and scale of the error/s is in your favour.

The error in verticality of the saw cut, will not have any appreciable effect (once turned) because in reality all of the inserts are square to each other, IF as you correctly point out - "no changes to the machinery is made in between cuts" & OR the blank is not flipped end for end for any of the cuts.

The error in lathe center alignment is 0.5 mm in say 450 mm or about 1 : 900. Not really going to be noticeable on a 30 to 45 mm cylinder to anyone but the most fastidious craftsman.

Now those same errors on a 12 to 15 mm diameter cylinder 35 to 60 mm long (a pen blank) will be quite apparent to most viewers. The shorter and thinner the work piece the more apparent the error.

For an experienced craftsman on architectural turnings a 0.5 mm center alignment error wouldn't even raise an eyebrow, but for thin spindles, i.e. finials etc, its a real problem that has to be accommodated.

NeilS
16th July 2021, 10:30 AM
yes me, i just use a smaller drill bit.


..................................:roflmao:

Fumbler
17th July 2021, 02:04 AM
Now those same errors on a 12 to 15 mm diameter cylinder 35 to 60 mm long (a pen blank) will be quite apparent to most viewers. The shorter and thinner the work piece the more apparent the error.

Totally agreed, i found that out the heard way very early on, and the same with not using a mandrel saver, i was tightening the tailstock so much it was flexing the shaft and making the pens oliptical.