lathe test bar

**Pete F** · 6th July 2014, 03:49 PM

Dean there are many, MANY ways to centre a lathe's tailstock, and it was the topic of rather a, errr, "robust" discussion on another unmentionable forum quite recently. Here I'm assuming all you're doing is just adjusting it back to centre and not rebuilding the lathe. I'll make that point from the start, lest the discussion progress the same way as the last time and people start mis-quoting texts they really have no idea what they contain. Don't get me wrong, you CAN use a test bar, nothing wrong with that at all, but most people don't have an accurate test bar and I don't think it's worth making one if this is all you're going to do with it.

At it's most rough and ready just shove a ruler between two centres, one in the headstock and one in the tailstock and clamp the rules between. It won't be especially accurate, but much closer than many people realise and will get you close enough at worst to refine it by other methods.

Once you're getting close, another method is to chuck an indicator holder in the headstock (it doesn't need to be accurately chucked) and use a dial indicator to sweep the bore of the tailstock. That's the method I normally use as it's reasonably quick and quite accurate. Ignore the top and bottom readings as the dreaded indicator droop, for which no blue pill exists, and you'll get a false reading top/bottom. However provided you measure accurately front and back there's nothing wrong with this.

The bore should be concentric to the outside of the quill. Assuming it is, then you can do the same as the above except sweep the outside of the quill. Again just consider front and back readings. If you're not sure if it is concentric, you can always pop it out and either dial it in a 4 jaw, or check it on a vee block and sweep the bore to check.

If you want to also check the height while you're at it (again assuming the tailstock bore in concentric), turn a diameter to exactly the same size as the tailstock quill. Then indicate between the two of them, this time you can check height, as you're then comparing the height of the turning to the height of the tailstock quill. The indicator is therefore drooping the same amount.

Instead of a test bar, accurately centre drill a regular bar and turn one end. Indicate that end then remove it from the centres and flip it over, while it's out traverse the carriage to the tailstock end and replace it so the turned end is at the tailstock. Indicate that end. That's "Pete's-Cheat" ™ method of doing a 2 collars test as you don't have to rough out all the material in between.

Unfortunately I find some people spend too much time with their noses in text books, or reading rubbish regurgitated by others on the internet, and not enough time with their noses instead in a machine giving some thought to what they're trying to achieve. If I see Schlessinger mentioned one more time in regard this type of thing I swear I'm going to lose it ... no wait, too late for that

What many seem to forget is the whole idea is to have the machine turning between centres without a taper. A machine can be measured until the cows come home, but it doesn't matter squat, what matters is what comes off the machine. In that regard I like to actually turn something between centres and measure the taper. Adjust the tailstock as required until it doesn't. Shortcut the process by making up a bar with 2 collars on it. I see books recommend turning down a bar and leaving the collars. The Old-Times must had more time/material on their hands in my opinion. You could just as easily shrink or otherwise fix a couple of bored out collars on to some stock, just as long as they don't move you'll be sweet. Otherwise another Pete's-Cheat ™ is to just turn down the end of a bar, remove it, traverse the carriage and turn down the other end. Measure the two turned areas.

The trouble with working alone and being self-taught is I don't really know what others do. But it seems to me that many people have forgotten the whole process of working between centres, and from what I've seen some think if it can't be chucked in a 3 jaw it can't be done. The beauty of working between centres is you can remove, flip, replace, remove again, flip, etc etc with no affect on accuracy. Just ensure your headstock centre is accurately cleaned up (why they're soft) and I will use a dead-centre in the tailstock for this type of work (another blast from the past many people don't seem to understand these days). Make sure the angles of both are the same and the centres are drilled accurately and identical. If somebody can't manage to do that then there's no real point in going to any real trouble with centring their tailstock as they clearly won't be able to use it to its potential anyway.

Hopefully that's given you a few ideas to start. Again, there's nothing wrong with using a test bar, and that's how it would have been done when shipped. There are many other methods to centre the tailstock also, but these are just a few I use.
Pete

**Pete F** · 6th July 2014, 07:52 PM

Dean I took these photos for you as an hour or so ago I needed to re-centre my own tailstock. In this case I wasn't turning between centres, instead drilling 1.5 mm holes in the end of 3 mm rod I'd turned (don't ask). I noticed that the drill was slightly off centre when I started the holes. Not much, but on small holes like this I could see it.

I didn't want to set this up with centres and turn things as I suggested above, it take too much time and wasn't necessary.

This shot shows how easy it is to get something about right, without even taking the work from the chuck in many cases. I'll be the first to admit it's a bit bodgy, it's reaching out a long way, and when rotated 180 degrees things aren't really symmetrical. But at least it will be better than nothing and it's very fast.

Mag Stand.jpg

Next we start getting a little more serious.

First indicating the outside of the quill. Easier than the inside. A much more rigid and Kosher indicator setup too.

Outside.jpg

Now indicating the bore itself. Better than above as you're indicating the taper itself.

Inside.jpg

Finally, you can always indicate something like this dead centre.

Centre.jpg

I was using a 0.002 indicator for this, and at the end of the day the limitations are more on the actual adjustment of the headstock itself than the setup to measure things. I think I got the alignment within 0.005 or so IIRC, but at this level any time you lean on anything, tighten it more or less than the last time, etc etc etc things will move. More than sufficient to finish the job and only took a few minutes. Far longer to write this up.

I hope that's of some help.

Pete

**.RC.** · 6th July 2014, 08:16 PM

The only way to do it properly is with three test bars, a sacrificial goat and venus has to be aligned with uranus..

**Michael G** · 6th July 2014, 08:20 PM

Slacker! Back in my day we used 5 test bars, a large roll of fine wire, 3 microscopes and had 14 apprentices to lift the machinery when it needed to be moved. Tell that to machinists these days and they don't believe you.

Michael
(channelling his inner Yorkshireman)

**cba_melbourne** · 6th July 2014, 08:38 PM

Originally Posted by morrisman

After watching tubalcaine's U tube video. I decided to make a lathe test bar , I followed his method to the letter ,buying a 400mm length of ground 3/4" stainless bar . After checking the bar on the carbatec granite plate, I accurately centred the bar each end and it turned out quite OK .

using the Verdict dti along the 400 bar between centres , The initial test on my old Sheraton showed the tailstock off set ( I did this adjustment some time ago) to be spot on

Moving the dti along the top of the bar 400mm , the bar drops .015" down to the tailstock

Bringing the tailstock up the the headstock where a centre is held in the 4 jaw chuck, I can see the tailstock centre is noticeably lower . Shims needed I think MIKE

A) 0.015" is almost 0.4mm low. That is a lot, it would make small diameter drilling impossible. The tailstock is supposed to be either exactly at center height, or something like no more than 0.02mm high according to the Schlesinger standards (sorry Pete F).

B) I am not sure what for you need a between centers lathe test bar, as there are other ways to measure tailstock alignment. A test bar just makes it slightly easier to measure. But it would have to be a precise test bar, just taking some ground stock and adding two centers does not make sense to me. But anyway, if you want to use a test bar made from ground stock, you need to make several measurements whilst rotating the bar, and take the average reading from the highest and lowest reading.

My preferred method to measure tailstock alignment is to clamp a dty into the spindle chuck, have its finger touch thge inside taper of the tailstock barrel, and then rotate the main spindle whilst ovserving the DTI needle. You may need a small mirror to read the DTI as it rotates. And you have to do this measurement twice, once with the TS barrel fully retracted and once extended. This is needed to confirm if the TS barrel moves parallel to the bed when extended.

PS: A lathe test bar that is much more useful to own, is one that fits inside the main spindle taper. Mainly used to align the lathe headstock to the bed. There are such Chinese test bars around, reasonably accurate and reasonably priced. Like this one from Chronos, MT3 including shipping from the UK for under AU$100: 3MT Lathe Test BAR 3 Morse Taper From Chronos | eBay
This type of test bar has, in addition to the Morse taper, also two centers to be used for the tailstock alignment. But it is also useful to align a mill head, or a dividing head setup.

**PDW** · 6th July 2014, 08:46 PM

Originally Posted by Michael G

Slacker! Back in my day we used 5 test bars, a large roll of fine wire, 3 microscopes and had 14 apprentices to lift the machinery when it needed to be moved. Tell that to machinists these days and they don't believe you.

Michael
(channelling his inner Yorkshireman)

I just use an FBH and if that doesn't work I have a welder & angle grinders.

PDW - metal butcher

**Pete F** · 6th July 2014, 09:00 PM

Originally Posted by .RC.

The only way to do it properly is with three test bars, a sacrificial goat and venus has to be aligned with uranus..

Well using the first setup you're not far from the truth actually! As I was packing up tonight I quickly set something like that up again and confirmed what I already knew was a reasonably accurately set tailstock. Even though it's a bodge, I could replicate the results of the more rigid way of doing it, but by sticking a sensitive DTI on the end of a piece of spaghetti like that you're always going to be asking for trouble. Set it up so everything on the stand and indicator are in the one plane, and then sweep the headstock to exactly 180 degrees and it will indicate accurately. However there's a lot of droop in that setup and if you're not exactly at the horizontal positions each time the readings will be out. Nevertheless it goes to show that no first born virgins need be sacrificed, and if you're in the middle of a job (as I was) and needed to bring the tailstock in to alignment it will get you very close indeed. Just be patient and use care.

Pete

**Oldneweng** · 6th July 2014, 10:38 PM

Originally Posted by Pete F

Dean I took these photos for you as an hour or so ago I needed to re-centre my own tailstock. In this case I wasn't turning between centres, instead drilling 1.5 mm holes in the end of 3 mm rod I'd turned (don't ask). I noticed that the drill was slightly off centre when I started the holes. Not much, but on small holes like this I could see it.

I didn't want to set this up with centres and turn things as I suggested above, it take too much time and wasn't necessary.

This shot shows how easy it is to get something about right, without even taking the work from the chuck in many cases. I'll be the first to admit it's a bit bodgy, it's reaching out a long way, and when rotated 180 degrees things aren't really symmetrical. But at least it will be better than nothing and it's very fast.

SNIP

I hope that's of some help.

Pete

Thanks Pete. A very clear and useful answer. Yes all I am thinking about is adjusting the tailstock back to centre if required. I would like to check it at some time. I have no idea what it is like at the moment. I like Petes Cheat. It sounds like my kind of deal. I have some (ok shed loads of) issues with wasting materials. I am quite the tight ar5se.

The pictures show methods that I should and will use to check my tailstock soon. Most people would probably think I should have done it already. Actually most people would have done it while the lathe was propped up on timber out in the implement shed, except that it would not have been level and it still isn't. I need to pour the concrete soon for the slab. A couple of days ago I realised that if I did not do it soon, I would have my very own wading pool right in my shed.

We don't run goats here. Would a sheep do at a pinch. My manager at work has a donkey. Could that be substituted? His kids would kill him, but I should care?

Dean

**WCD** · 6th July 2014, 11:28 PM

FWIW I read (forgotton where) this method of checking the horizontal alignment of the TS without a test bar.
Take a short piece of round bar a little larger than the TS quill. Chuck it, then turn down to as close to the TS quill size as you can.
A dial gauge in the toolpost will then show the horizontal alignment of the quill compared with the turned bar in the chuck. This assumes that the quill bore taper is concentric with the quill OD. Also assumes that the headstock spindle is correctly aligned.

One advantage is that you can check the alignment with the TS at any place along the bed.

Cheers,
Bill

**RayG** · 6th July 2014, 11:39 PM

Having run out of goats and apprentices....

Ray

PS.. I said half a tenth in the video, I should have said half a thou...

**Pete F** · 7th July 2014, 10:08 AM

Originally Posted by RayG

PS.. I said half a tenth in the video, I should have said half a thou...

... not that it makes any difference, as the units indicated on co-axial indicators are just that; "units". The actual amount will depend on the probe, angle, etc.

I don't believe they're calibrated in the same way as a DTI, which relies on a specific contact point length and angle (or corrections for the latter) to read correctly. I'm in the process of buying a bunch of replacement DTI points now, every brand is a different length

Quick tip BTW, increase the sensitivity of a probe by swapping a longer contact point, but then remember to swap it back or at least mark it, as the readings shown won't be accurate.

Anyway, using a co-ax is a good tip and quick.

FWIW I read (forgotton where) this method of checking the horizontal alignment of the TS without a test bar.

Maybe immediately above?

**Oldneweng** · 7th July 2014, 10:40 AM

Originally Posted by WCD

FWIW I read (forgotton where) this method of checking the horizontal alignment of the TS without a test bar.
Take a short piece of round bar a little larger than the TS quill. Chuck it, then turn down to as close to the TS quill size as you can.
A dial gauge in the toolpost will then show the horizontal alignment of the quill compared with the turned bar in the chuck. This assumes that the quill bore taper is concentric with the quill OD. Also assumes that the headstock spindle is correctly aligned.

One advantage is that you can check the alignment with the TS at any place along the bed.

Cheers,
Bill

Thanks for the suggestion Bill. One small issue is that my quill diameter represents a piece of steel that I would hoard in a big way. I would need to very lucky to find something to suit without it losing its usefulness. This is a one off test. once you take it out of the

**Oldneweng** · 7th July 2014, 10:42 AM

(Damn phone) chuck the piece becomes just stock.

Dean

**Pete F** · 7th July 2014, 12:31 PM

Dean the piece only needs to be long enough to get an indicator on it. 50 mm or less long and whatever your quill diameter is. It can be any material you please. We're not talking 6 m lengths of centreless ground tool steel here!

If you're concerned about raiding your stash for this, dual purpose it. Pete's-cheat ™ #628, once you've turned it to the quill diameter mark the #1 jaw on it so you know where it came out of the 3 jaw. Depending on how clapped out your chuck is, you can get jobs to repeat reasonably well (but not perfectly of course) if you put them back in exactly the same position in the chuck. Measure the diameter and then mill a step exactly half way through the piece. You now have Dean's new test piece for both setting the tailstock and your tool height. To re-set your tailstock, chuck it back with the mark on the #1 jaw and indicate off the round part facing you and compare with the tailstock. To set your tool height, rotate the chuck 90 degrees so the milled flat is facing down. Put the QCTP holder in but don't lock it. Bring the tool up to the centre of "Dean's Tester", as that way if you don't have it absolutely horizontal it won't matter. Lock your QCTP handle and with it locked nip up the holder's adjuster and lock it off with the nut. In about 5 seconds you've placed your tool exactly on centre. Because "Dean's Tool" isn't hardened it won't chip or damage your inserts, but eventually will get somewhat beat up. After a number of years, quite a number in my experience, it will get so beat up you'll have to raid your stash and give up another 50 mm length of steel. Just be aware that this won't be a really precise way to re-centre your tailstock, but more than adequate for things like drilling even small holes.

Still no virgins have been sacrificed.

**morrisman** · 7th July 2014, 12:49 PM

Originally Posted by RayG

Having run out of goats and apprentices....

Ray

PS.. I said half a tenth in the video, I should have said half a thou...

thats impressive !

Thread: lathe test bar

Thread Tools

Search Thread

Another approach

Another Way

IMRESSIVE

Similar Threads

Buying a lathe (Old massive Lathe, or New small C6 Lathe)

Tags for this Thread

Posting Permissions