Tailstock Alignment Puzzle

[Bryan]

I'm not surprised that my tailstock bore is lower than the spindle, nor that the quill slopes down toward the headstock. What is a bit baffling is the numbers. If I shim the front of the TS up till the quill is level (actually parallel with the TS ways), the bore ends up high by .6mm. Wear can't account for this, and I know this machine had much tighter tolerances than this when new. This tailstock was welded, and no doubt some warpage occurred, but I'm struggling to see how that could explain this anomaly. It's not like material was added to the underside. The only thing that makes sense to me is if the tailstock has been replaced at some time and not fitted properly.

So to get the quill parallel and at the right height, I will have to machine the underside of the TS body. Somehow. I sure won't be scraping off .35 mm (.013")!

[Stustoys]

Hi Bryan,

I'll leave it to others re welding moving the casting about.
Remember someone else had a go at fixing it before you, no telling what they did.
Did you check how flat the bottom of the TS was after welding?

Stuart

[Bryan]

Stuart, yes I slapped it on the surface plate, but not till the other day. It showed 3 small areas of contact. A day and a half of scraping has increased the footprint, but made no real progress with the alignment. The shaper is the tool for that job. I just get this nagging feeling I may regret carving it up. I need to go over all my tests and measurements again. DTIs can be so confusing. And the fact I can't understand how it got that way makes me doubt my findings more.

[Dave J]

Sounds like you have some work ahead of you.
The specs for a tailstock bore are 0.0005"inch at the front opening above the headstock spindle centre line, and 0.0005" inch per foot rise towards the headstock.
Just remember gravity affects the readings.
You may or may not get this close or even want to, but I thought I would put up those numbers for something to aim towards.

Dave

[Stustoys]

How are you testing it?
Pictures?

[Bryan]

Thanks Dave, I'll let you know. :) I'd be pretty stoked to hit numbers like that, but it may be optimistic for a 50 year old machine. Way wear is going to be a party pooper. You could get it smack on in one spot, slide the TS along a bit and be out again. But I'm sure I can get it better than it is.

Edit: Stuart, documenting will be a bit of work, which I'm too bushed to contemplate right now. But it would be a good exercise. Will see if I find time/energy this week.

[Stustoys]

Maybe its in the base? Still it will give you something to work on after next weekend.

Stuart

[Dave J]

At least you will pick up some tips on the weekend about how to check it out and what to use to measure it properly.

I do have one easy way to fix it, send the whole lathe to me and I will do it for free, it just might take me 10 or more years to do it (or so I will tell you, LOL)

Dave

[simonl]

Question, How do you measure the tailstock/headstock spindle height? Is it simply a matter of putting a dead centre in the tailstock and measuring the top part that is not tapered, then doing the same on the headstock?

PS I will leave any suggestions to the qualified/experienced members of this forum!

Simon

[pipeclay]

An initial test to see if the centres are aligned can be done with a centre in both spindles.

You can bring them together and visually look at them to see how well the ppoints of the centres line up horizontaly and virticaly.

You can place a thin ridgid pice of material between the centres points,if this material stays in the vertical and horizontal position with no visible tilt/angle either way you would say that the centres are aligned,if it kicks one way or the othe it indicates that more than likely the tailstock is sitting low/high or offset to the left/right,it is not often that the headstock is out of alignment.

If you wanted to check centre height you could use a parallel,vee block,inside mic,outside mic,depth mic.

Bit hard to describe easily in words but you would or could turn a piece of material in the lathe about 1"/2" long (dosent matter about dia,but a round number might be easier) with a good finish,place your parallel on the lathe bed,(preferably on the flats rather than the vees) you may need to use more than one paralle or even some pieces of tool steel or any material of a known size as long as there both the same size,(lay these along the flats but they must be higher than the vees,then bridge between these pieces with a parallel.

You would then in my case use your inside mic and measure the distance between the parallel you have across your lathe bed and the underside of the piece you have turned.

You will have to do some simple calculations to determine the centre height of the spindle from the flats of the lathe,once you have done this you can then do the same measurement with the tailstock (except in this case you will be measuring to the underside of the tailstock spindle,best to have the tailstock clamped to the bed and the spindle locked.)You can if you wish move the tailstock along the lathe bed and check centre height at numerous positions.

The calculations that you would need to take into account would be the height of total material to get you above the vee ways (this would be added to the measurement of the inside mic),the diameter of the piece of material you have turned (this would be halved and added to the inside mic measurement).

So at the end it would be PACKING + HALF DIAMETER MATERIAL TURNED+INSIDE MICROMETER MEASUREMENT= CENTRE HEIGHT OF SPINDLE,allways check the measurement of an inside mice against an outside mic unless the inside mic has been calibrated.

These checks can also be done in the reverse by working down from the material using a depth mic.

[simonl]

I see. So you wouldn't consider taking measurements from a DTI which is magnetically clamped to the saddle? That way you can move up and down the ways to take measurements at any point you like?

Or would this add unnecessary errors into the measurements because you are taking the readings from the saddle which sits on the ways?

Simon

[pipeclay]

It depends on what you want to measure,I thought you asked about checking centre height.

[Bryan]

Briefly, I used 2 tests, both using a .01mm dial test indicator (DTI). For those not familiar, this is the lever type of indicator.

The height test was done with the DTI magged onto the spindle face, reading either the bore or the outside of the quill. The spindle was then rotated. This can also be used for lateral adjustment of the TS.

The slope test involved the DTI fixed to the compound so it read the top of the TS quill. The TS was then advanced along its ways and another reading taken. The carriage stayed locked and the cross slide was used to find the top of the quill for each reading. The quill stayed extended and locked.

Stuart, the base actually slopes up, but the body slopes down more. :(

Dave, thanks for the offer. Don't hold your breath. :q

[simonl]

I possibly didnt explain it right. I meant a means of measureing centre height purley so you can compare headstock spindle to tailstock. For the purposes of alignment does it not make any difference where the measurements are done from as long as it is repeated for both, ie the top slide or saddle?

That was more my question.

Cheers,
Simon

[Stustoys]

Bryan,

Your tests seem "okish" except the first one. The lateral test could be ok but the vertical test like that any deflection if doubled due to gravity.(I'll go test it and get back to you.)


Stuart

[Stustoys]

OK so no cheating by looking at the fourth picture.
How much movement would there be on the DTI between the first and second pictures?
Got your answer?



Did you say 0.005"?
(in fact its closer to 0.004" due to the angle of the DTI arm"?".)
Depending how rigid your set up is it maybe more or less than this. Also I haven't worked out if this would make your alignment problems appear worse or better.

Also I've changed my mind about the lateral test, it depends how you have the DTI set up. You could still have a deflection issue but the effect, if any would very small as long as the Dia of what you are checking isn't tiny.

Stuart

[Bryan]

Stuart and Dave, you're right. I just did a sag test and got .2mm. I'd read of this issue but for reasons unknown didn't take it seriously. I do now! It doesn't make the problem go away but it does reduce its magnitude considerably. Thanks. Now what holes can you find in my other test? :)

PS: Stuart, thanks for taking the time to do this.

[Dave J]

I think it was last year I posted this from another forum.
http://i990.photobucket.com/albums/af26/top720/Sag.jpg

http://i990.photobucket.com/albums/af26/top720/Sag1.jpg

http://i990.photobucket.com/albums/af26/top720/Sag2.jpg

Dave

[Stustoys]

After you scraped the bottom of the TS did you spot it to the base to make sure the base was flat?
One other thing, is the arm of the DTI square(tangent) to the surface you are measuring? There is some math about to allow you to work out the true length of movement taking into account the arm length and angle but I've never used it. Just remember if you dont you arent talking true length.

I'd be tempted to repeat the slope test on the bottom of the quill and see if you come up with the same numbers as the top(but I'd guess there will be a key in the way?).

I think there maybe better(maybe not better, easier? not sure lol) ways to test the slope but you would need a test bar.

Stuart

[Dave J]

Dial indicators have a much stronger spring in them over a dial test indicator, so with a good solid mount fabbed up to go into the chuck, should be fine for the home shop.
I would suggest using a 4 jaw to off set and adjust it, but I don't think you have yours mounted yet.

Another idea,
If a shaft was machined in the chuck to exact diameter of the tail stock quill, it could be used as a guide for the tailstock quill so the indicator could be run from under it to under the ram and the same as the top and sides.This way the indicator would be in the same position and gravity would not affect it.

I bet your glad your doing the scraping class now.

Dave

[simonl]

Dial indicators have a much stronger spring in them over a dial test indicator, so with a good solid mount fabbed up to go into the chuck, should be fine for the home shop.
I would suggest using a 4 jaw to off set and adjust it, but I don't think you have yours mounted yet.

Another idea,
If a shaft was machined in the chuck to exact diameter of the tail stock quill, it could be used as a guide for the tailstock quill so the indicator could be run from under it to under the ram and the same as the top and sides.This way the indicator would be in the same position and gravity would not affect it.

I bet your glad your doing the scraping class now.

Dave

Is this method similar to the one you guys are talking about?

Part 2 TAILSTOCK ALIGNMENT on LOGAN HARDINGE South Bend Lathes - YouTube (http://www.youtube.com/watch?NR=1&v=1EZh8SUoA4o)

Simon

[Stustoys]

Dial indicators have a much stronger spring in them over a dial test indicator
Are you saying this is a good thing or a bad thing?
I'd think it would make things worse(I'll test it and find out).
Still like you say, deflection goes up as the cube of distance so a short holder is the way to go.

Stuart

[Stustoys]

Simon, that's fine for lateral, its likely pretty good for vertical as well.

In fact you've just reminded me of something else. Vertical alignment doesn't matter anything like as much as lateral alignment.



Stuart




Is this method similar to the one you guys are talking about?

Part 2 TAILSTOCK ALIGNMENT on LOGAN HARDINGE South Bend Lathes - YouTube (http://www.youtube.com/watch?NR=1&v=1EZh8SUoA4o)

Simon

[Dave J]

Are you saying this is a good thing or a bad thing?
I'd think it would make things worse(I'll test it and find out).
Still like you say, deflection goes up as the cube of distance so a short holder is the way to go.

Stuart

For gravity I think it is a good thing, as the DTI has the problem of having soft spring. This would of course only work if the mount was solid.

Dave

[Dave J]

Simon
That one will work. There is another way to got it spot on for a curtain distance, and thats the 2 collar test. Depending on your lathe sometimes what is spot on near the chuck wont be the same when the tailstock is down the bed a bit.
You place a shaft between centres with a drive dog driving it, under cut the middle leaving 2 collars on either end. You then set the cross slide and take a cut of the first collar, then without touching anything move the carriage along and cut the next collar and then measure both and adjust the tail stock to suit.
If you have a precision job and have a 2 collar test bar the same length it will be set up spot on.
Another way of making a 2 collar bar is to tack weld 2 collars onto a bar.

These can be made up any length.

Before doing any of these tests, you need to make sure you lathe has no twist by using a machinist level which can be purchases from CTC tools for around $80 posted.

Dave

[Stustoys]

As close as I could get to the same test my 0.001mm dial gauge says 0.13mm, my 0.01mm one says 0.2mm. "Roughly" (I cant rotate the spindle a full 180 with the dial gauge on there unless I take the gap out).

Stuart

[Dave J]

Would that be because of your dial base and arms? As I couldn't see the dial indicator moving that much just turning it upside down say in your hand, though I will try when I go down the shed in a minute.
This is why I said to make up a strong mount to hold the dial indicator. Something like a 30mm solid bar to go into the chuck with another piece welded to the side of it at a slight angle to get on the outside of the tailstock quill and hold the indicator.

Dave

[Stustoys]

Would that be because of your dial base and arms?
Yes certainly. That was my point(although not very well made lol)
Once the long arms of the dial gauge base are removed (i.e. mounting the DTI in a piece of wood) most of the problem will go away and as the vertical alignment isn't so important it would certainly be close enough (well for me anyway).

Stuart

[Bryan]

Stuart, your test rig was much more elegant than mine. Very clever. In my defence I was trying to duplicate the exact setup I had used on the lathe.

http://www.woodworkforums.com/members/59697-bryan/albums/graziano/7084-pa041637.jpg
http://www.woodworkforums.com/members/59697-bryan/albums/graziano/7083-pa041636.jpg

Dave, great minds think alike:

http://www.woodworkforums.com/members/59697-bryan/albums/graziano/7086-pa041645.jpg

So now I think I have a more reliable number. I get .46mm low. When I shim to level, it's .2mm high. That's still a lot of scraping. But I'll be a better judge of that by next week hopefully.

Stuart, I did spot the TS to the base. Neither is flat yet but there is plenty of contact for stability. I kept my indicator pretty flat. If I get down to the last micron I'll look up the trig. And yes, there's a key slot on the bottom of the quill so I can't get a reading there. BTW, the reason I'm using a sleeve is I was getting erratic results from the bore, so I guess it's a bit worn/damaged at the mouth.

[simonl]

Hi Bryan,

When you say you will scrape the base of the TS to get better alignment with the headstock, do you mean scrape the v and flat that sit on the actual ways, or, the flat part of the base that can be separated from the tailstock?

I assume that if both height and angle adjustments are needed then scraping a flat surface would be easier than a V. Assuming you are happy with the V as it stands I suppose...

Can you explain this? I hope I have worded my question adequately.

Cheers,

Simon

[RayG]

Hi Bryan,

What about a coaxial indicator, running in the bore of the tailstock morse taper?

EDIT: Ok, here's a short video, showing what I mean

Tailstock Alignment Check - YouTube

Regards
Ray

[Stustoys]

Hi Bryan,
You're just showing off how big your lathe is...... I didnt even notice the 3 jaw the first time around ;)

Is that angle iron you are using in your test?

How far out of the TS is the quill when you are getting 0.46mm low?


Stuart

[Bryan]

Hi Bryan,

When you say you will scrape the base of the TS to get better alignment with the headstock, do you mean scrape the v and flat that sit on the actual ways, or, the flat part of the base that can be separated from the tailstock?

I assume that if both height and angle adjustments are needed then scraping a flat surface would be easier than a V. Assuming you are happy with the V as it stands I suppose...

Can you explain this? I hope I have worded my question adequately.

Cheers,

Simon

Simon, I wasn't planning to mess with the sliding surfaces. If I were doing it by the book that would be done first. But V ways are a more complex alignment issue, and I'm not trying to recondition the whole lathe, just make some adjustments. So I will be scraping the bottom of the tailstock body and the top of the base. I'm not even too fussed if those surfaces are dead flat or level, as long as they mate well and result in adequate alignment of the quill. Let's say .02mm. That's a nice round number. About a thou. Should do for an old warhorse.

Ray, I have heard of coaxial indicators but never really understood them. Haven't paid too much attention I guess because they're usually used on mills. What would the advantage be? I ask only out of idle interest, as I'm having to make do with what's on hand at the moment.

Edit: Ok Ray, that is pretty cool.

Edit: Stuart, the quill was out about 120mm. Full travel is about 150. If you're thinking the number would be different with the quill retracted I think you're right. It should about halve the error. I will test that tomorrow.

[Dave J]

Sorry Ray, I have different thoughts on that tool.
Bryan, don't waist your money on a coaxial indicator, they are not real accurate. Even the Blake's don't get a good review on PM.
I bought one of the cheap ones and had to totally rebuild it and regrind all the critical surfaces and even after doing this it's off compared to my DTI used on my mill.
They are good enough for rough set up like with in a thou or 2,(even though they claim .0005 accuracy) but they say the Haimer (taster type) type are superior, but out of our price range for this type of tool. Mine sits in the cupboard ever since I rebuilt it and will stay their.
Have a read here
http://www.practicalmachinist.com/vb/cnc-machining/coaxial-indicators-blake-haimer-233038/ (http://www.practicalmachinist.com/vb/cnc-machining/coaxial-indicators-blake-haimer-233038/)

You are going about it the right way using the tools you have, and the only error you have to watch out for is gravity as we discussed earlier.

Dave

PS
If it was any good I would lend you mine for the postage cost, but it wouldn't be worth the postage, as their not as accurate a dial or dial test indicator.

[Dave J]

Also forgot to add the coaxial indicator is only to compare movements on the dial, not to actually measure. The numbers on the dial could be replaced with anything as they are not the true reading because of the angles, lengths etc.

Dave

[RayG]

Hi Dave,

All good advice, I think you are correct on all counts, (as always) :)
I've mostly used mine on the mill for lining up holes, where it does the job nicely.

For what Bryan is doing I'm sure what you say is correct, a DTI would give more accurate absolute measurements.

I think Connelly might have a section on measuring tailstock alignment. I haven't checked. Bryan, do you have a copy of Connelly?

Regards
Ray

[Dave J]

Hi Dave,

All good advice, I think you are correct on all counts, (as always) :)
I've mostly used mine on the mill for lining up holes, where it does the job nicely.

For what Bryan is doing I'm sure what you say is correct, a DTI would give more accurate absolute measurements.

I think Connelly might have a section on measuring tailstock alignment. I haven't checked. Bryan, do you have a copy of Connelly?

Regards
Ray

High Ray,
Not sure about as always.
Give it a try yourself next time you have it out and after setting it up on a hole in the mill, change over to a DTI and see if yours runs true.
This is the test I gave it and I also did it in reverse and got the same results.

When I received mine the needle would move back and forth while running and touching nothing. The seller took care of it buy compensation, so I took it completely apart as I had nothing to loose and ground the main body base in the lathe, under the top hat type piece, and trued up the shaft that goes into the collet to be in alignment with the main shaft. I also emery papered everything smooth on the rocker piece.
I then took it back to the mill after a few days work and it was running perfectly true touching nothing, but was out to my DTI and I tried a few in disbelief.

I am pretty sure Bryan got a copy when we where all taliking about in a thread.

Dave

[Bryan]

With the quill retracted the figure is .07mm low. Shimming to centre height in the almost retracted position would probably be fine for centre work, but for drilling the slope would be undesirable. Getting it pointing in the right direction seems worth spending a bit of effort on.

Ray, yes I have Connolly. He puts the tailstock in the context of a full rebuild, basically starting from the ways and working up. A lot of his tests I'm not equipped for. His final quill alignment test uses a test bar and assumes perfect ways. However he does use an interim test for alignment of the quill to the underside of the body. You put the body on a surface plate and slide an indicator across the top of the quill at different positions. I did that test and found it sloped down about .2mm/100. Then I did essentially the same test but with the TS installed on the lathe and the cross slide holding the indicator. This showed a slope of about half that. Turns out the base slopes up and partly compensates for the body sloping down. Way wear is the only explanation I can think of for that, though I don't really see why it would be so uneven.

If I did it by the book and made everything level, I would end up with a very low quill. Therefore I intend to let those mating surfaces (base/body) be whatever crazy angle they want to be and remove the minimum necessary to get the quill level. Then I will adjust the height as needed.

Or maybe I'll just shim to height as is so I can get on with half-arsing the next thing. :)

Thanks everyone for your interest.

[Dave J]

I would just shim it for now and come back to it when you have gained a bit more experience with scraping. I am not saying you wouldn't be able to do it now, just with so many variables with the ways etc, it would be better to have a bit of experience scraping and maybe consult a few people like at the scraping class before removing any material.
I see often Forrest Addy recommending shims between the base and the tailstock, so it is a common fix that can be adjusted or undone or at any time in the future.

Dave

[Stustoys]

Hi Bryan,
If I'm understanding your numbers correctly, don't they meant you have to scrap the back down? The front will come up a little as you do due to the over hang.

Still, best wait until after the weekend ;)

Stuart

[Bryan]

Yes, yes and yes. In that order. :)

[simonl]

Hi Bryan,
If I'm understanding your numbers correctly, don't they meant you have to scrap the back down? The front will come up a little as you do due to the over hang.

Still, best wait until after the weekend ;)

Stuart

Would that then remedy the tailstock alignment in terms of running parallel to the ways but you would still then have to deal with any height difference with respect to the headstock? Would you then shim it to bring it up in line with the headstock?

Hope you don't mind the questions from a novice.....

One other question, do you necessarily have to scrape? Can you simply run a fine honing stone over it? I was under the impression that scraping was more for conditioning two surfaces that move across each other and that the main benefit was the allowance for oil pockets between the two surfaces for lubrication but in this case the two surfaces will not be moving.

I really should have signed up for this scraping class!

Cheers,

Simon

[Stustoys]

Hi Simon,
It all depends where it ends up after scrapping it parallel(or the end of quill rising a little as it comes out, from memory). It may need to be shimmed but if I have the numbers right in my head it may need to be scrapped down to center height(or a little high)

I'd think parallel is more important than center height. Also need to make sure it stays parallel to the other plain.

Keeping the surface truly flat with a hone isn't easy(some would say not possible). Also in this case you want to keep the surface flat but remove metal from one side but not the other(and everything in between in between lol). Scraping can be used as you say, but it can also be used to fit two surfaces that are clamped together(in fact I'm pretty sure its easier as you need less spot per inch).

All scraping advice subject to change :D


Stuart

[simonl]

Hi Simon,
It all depends where it ends up after scrapping it parallel(or the end of quill rising a little as it comes out, from memory). It may need to be shimmed but if I have the numbers right in my head it may need to be scrapped down to center height(or a little high)

I'd think parallel is more important than center height. Also need to make sure it stays parallel to the other plain.

Keeping the surface truly flat with a hone isn't easy(some would say not possible). Also in this case you want to keep the surface flat but remove metal from one side but not the other(and everything in between in between lol). Scraping can be used as you say, but it can also be used to fit two surfaces that are clamped together(in fact I'm pretty sure its easier as you need less spot per inch).

All scraping advice subject to change :D


Stuart

Thanks for the explanation.

You guys are gunna have a ball this weekend!

Simon

[Metalman]

Hello Bryan, I have a tailstock alignment problem as well, my problem has been caused by my tailstock not being the original that came with the lathe. The 'new' one is lower than the headstock so I have temporarily shimmed it up. However I eventually plan to use the lathe to bore out the quill barrel and and fit a bronze sleeve. This might be an option for you especially if the quill assembly is worn as mine is. I am going to get the quill lightly ground to eliminate any wear on this before the sizing of the bush.
Mm.

[Dave J]

Would that then remedy the tailstock alignment in terms of running parallel to the ways but you would still then have to deal with any height difference with respect to the headstock? Would you then shim it to bring it up in line with the headstock?

Hope you don't mind the questions from a novice.....

One other question, do you necessarily have to scrape? Can you simply run a fine honing stone over it? I was under the impression that scraping was more for conditioning two surfaces that move across each other and that the main benefit was the allowance for oil pockets between the two surfaces for lubrication but in this case the two surfaces will not be moving.

I really should have signed up for this scraping class!

Cheers,

Simon

Hi Simon,
If you don't ask you will never know. I think I speak for most of use guys on hear with a bit of experience that we would rather you ask a question, than do something you don't know about and stuff it up. After all we all come hear to learn and help where we can.

Another option to scraping is to mill the top of the base and the bottom of the tailstock, but scraping is far superior, and for a quality lathe like Bryan's well worth the effort as it would have been done like that from factory. Most of the smaller Chinese lathes are only milled on these surfaces, so milling does work if done accurately.
Mine has scraping marks, but I wonder if that was just for show after grinding or to hold a bit of oil in the pockets.

As for using a stone, that is fine if it has a few nicks sticking up, but not to remove metal as Stuart said.

Dave