Looking for a test bar

[ibuildstuff4u]

I have a 9X20 asian lathe and I want to dial it in so it's more accurate to use.  I'm looking to buy a good test bar but so far I haven't found too many on the web.  I really want one with a MT#3 on one end, but I would be ok with just a plain bar.

Does any one know of a good source for test bars?  Also it can't break the bank and needs to be $100.00 or less.

Thanks for the help!

Dale P.

[fixerup]

Hi Dale,.
If this  test bar is to dial in your tailstock,here is my keep your money for something else solution . Just mount a long rod in your chuck and support the other end in your live center in the tailstock. Take a  small cut close to the chuck about .5" wide. Note the reading on your cross-slide. Move over to the tail stock end and make the same cut. Measure both cuts, if they are not the same, adjust the screw in the base of the tailstock to make the correction.
TIP:You can mount a dial indicator on the rod at the tailstock end to give you an idea of the correction needed , while adjusting the screws. 

Repeat: Cut a little off both ends again and check, until they match.

Phil   
 

[Chazz]

What fixerup said,I used a 5/8"dia x 18" piece of Precision Ground Drill Rod (0-1 in my case) and 'Rollies?' Dads' method of dialing in. 

Cheers,
Chazz

[Fergus OMore]

Might I say 'Why?'
As far as I can recollect, the 9x20 has only two mounting points and consequently should not be 'in wind' or twisted if reasonably mounting procedures are followed.

Therefore, it is only the tailstock that is likely to be 'out' of alignment and requires adjustment.
Whilst the advice given is OK, I would simply put a good centre in the spindle and tailstock, being the centres together to hold a thin sheet of steel such as a Stanley knife blade or safety razor blade. If it tilts one way, simply adjust the tailstock until it is held square in all directions.

Have a nice Christmas-- and stop looking for canaries in the sparrow flock.

Fergus O'More

[Bernd]

Best bet is to use a rod about 12" long between centers. Not in a chuck and a center in the tailstock. Reason the headstock could be at a slight angle and the tail stock dead center, yet when you cut just using the chuck you'll have a taper.

To make a test bar you'll need a steady reast to drill a center in the end. Clamp the bar in the chuck by about 1/4" and the other end of the bar in the steady rest. Have it sticking out about 1/2". Drill you center hole. Reapeat for other end. Now set up machine with a center in the headstock and one in the tailstock. Take a cut at each end as described in the about posts. Much more accurate.

Bernd

[andyf]

Real cheapskates like me use a length of threaded bar, centre drilled at each end, with two  washers clamped between nuts, one at the headstock end and the other near the tailstock. Take cuts on the washers and adjust until both have the same diameter.

Andy.

[Fergus OMore]

Best bet is to use a rod about 12" long between centers. Not in a chuck and a center in the tailstock. Reason the headstock could be at a slight angle and the tail stock dead center, yet when you cut just using the chuck you'll have a taper.

Bernd

On a 920? Sorry but the headstock  isn't built like that. It bolts into the front vee.

If it wasn't a 920 and (say a Myford), the head stock is tilted in so that it faces concave and not convex.

Back to my orginal comments about 920's. It's the tailstock on which to concentrate. Been there- got marks on me teeshirt to prove it.

Had the head stock casting off -if push comes to shove.

Of course, you can read Schlesinger's Limits!  I'll get me kilt and plaid

[Pete.]

Warco do a MT3 test bar for £35

http://www.warco.co.uk/other-measuring-equipment/331-morse-taper-test-bars.html

[Fergus OMore]

Again, this a 920 or in British parlance,  a 918. It has a prismatic bed and a raised vee which acts as a locator to both the headstock and the tailstock base. The bed is machined with a single fixing at both ends- and not 4. The bed is hardened and wear should be negligible.

Plenty of other things to fault a 920 but we had the much the same old argument about the 920's little sister just a week or two ago.

Regards

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Post Script

It is some 18 years since I ran a 9180 which was one of the variants which accepted much of the Mydord tooling. My memory at over 81 is sometimes a bit sketchy but ALL the words and music on the 920 have been traced to

'Re-building the Asian 9 x20 lathe'.

Several comments suggest that the risks of 'playing hopskotch with Rollie's Dad' are minimal and my alignment using two( I hope) accurately ground centres is shown.

[denncarm]

Try a roller bar out of a scrapped laser printer, some of those are quite accurate.

[Miner]

Your idea in principle is fine. To be honest, There's not enough tuning of machine tools to make what you buy the best you can possibly get. But your also making a somewhat semi educated guess that your headstock Morse taper is 100% dead on in alignment with the center of rotation to the spindle O.D. and the bearings. I'd recommend spending that $100 on the Connley book "Machine Tool Reconditioning". You'll never look at, Or treat a machine tool the same way again. To be totally honest, That book really should be required reading for anyone who owns machine tools. I had a lot of ideas before I bought my copy of this book. Almost all of those ideas were wrong.

Pete

[Fergus OMore]

Hi Pete,
              I deliberately avoided the mention of Connelly for the sheer enormity of the work-- and the likelihood of falling asleep reading it.
I actually am a glutton for punishment and not only read it- but overhauled a few small lathes!

Georg Schlesinger in -I think, Testing of Machine Tools goes into the matter of testing which is where we came in- rather than cover our sleeves with iron dust mixed with cobalt blue and  spots per square inch.

I therefore assumed that the poster's 920 was fairly new and from my own hands-on knowledge, was unlikely to be affected by 'in warp' problems. Frankly, if the lathe was permanently twisted there would be bugger all that an amateur could attempt. Again, a test bar would only cover part of the length of a twisted lathe.

Again, cutting a pair of disks until they were equal- cannot work in a twisted 920 because it has only 2 mounting points and not the conventional 4 or larger multiple of two. One can cut all day, but miracles only come from the East!

Anyway, our poster has something to go on.

[Bernd]

On a 920? Sorry but the headstock  isn't built like that. It bolts into the front vee.

What if the vee's in the headstock are not parellel with the spindle in the headstock?

[Fergus OMore]

I deleted my first reply because-- 920 lathes come with test certificates!

[Pete.]

On a 920? Sorry but the headstock  isn't built like that. It bolts into the front vee.

What if the vee's in the headstock are not parellel with the spindle in the headstock?

If you worry about that then where do you stop? Are the tailstock vees parallel to the ram? Is the cross-slide dovetail perpendicular to the ways?

[DaveH]

Hello Dale,

Well if you were not confused, I think you sure are now

I think we should try and answer the OP's question.  At the very least read it.
 
DaveH

[Pete.]

I answered his question exactly by suggesting a source at half his budget.

[DaveH]

Pete,

Yes you did, and it was a very good answer
 
DaveH

[Fergus OMore]

Unfortunately, buying a Test bar is as useful as an ashtray on a motor bike!

The headstock and whatever have been aligned by far more accurate instruments than we can afford.

Again cutting two disks or whatever which some books suggest is not usually necessary as there are only TWO securing points. The books assume that there are FOUR. Let me assure you that to fasten them down only involves two bolts and unless you want to snap the lathe bed, it ain't going to twist.

That leaves the tailstock which has only ONE degree of freedom- and that is sideways.

To adjust the tailstock/poppet/ whatever you want to call it, you measure the diameter of the poppet in tenths of a thous.
Put a bar in the three jaw and turn it  'precisely' to that dimension.

You THEN clock with a very accurate dial gauge mounted on the saddle- and adjust and clock from the newly machined  bit along to the tailstock poppet until the readings agree to 1/10th of a thous.


OR

Use a Stanley knife blade as earlier intimated-- and go off to the boozer with the savings.

[mgnbuk]

I deleted my first reply because-- 920 lathes come with test certificates!

And you believed it ?

Without checking ?

On a Far Eastern machine ? (AKA "A kit of part machined castings, loosely assembled for ease of tranport")

The headstock to bed alignment check is a lot easier with a test bar directly in the spindle taper. How else can you check that the spindle axis is pointing slightly upwards & slightly towards the tool (as specified in National Standards (British & DIN in my experience)) ? And this is normally checked after levelling the bed to ensure it is straight & not twisted with a precision level (again, as specified in National Standards - and usually the first check). While the test bar is mounted, you can also check that the taper is running true (part of National Standards.....).

The same goes for the alignment of the tailstock barrel - it should point slightly upwards & toward the toolpost. Yes, you can clock over the top & side of the barrel, but drills locate in the taper !

I used to work for a CNC machine tool rebuild company & we always used a spindle test bar to check that the headstock was correctly aligned. On large bore machines where the spindle "bung" was missing, we had replacement bungs made (at great cost, usually) to allow the use of an MT 4 or MT5 test bar. We also had special bars made - one was to check the turrets of Webster & Bennett vertical borers. Once the headstock was within spec. on a levelled bed, the saddle would be scraped in & part of that was to trammel the front cross-slide shear from the spindle to ensure that a slightly concave cut would result.
(Concave turning is not achieved by throwing the headstock, it is achieved by throwing the saddle).

So I would suggest that the original poster is taking a sensible course of action to check for himself that his machine is up to scratch. I will be doing the same when I get round to CNC converting a Chester mini-lathe - even though it is new, I want to see for myself that the headstock is correctly aligned.

The headstock and whatever have been aligned by far more accurate instruments than we can afford

A DTI & test bar are not expensive. And how much time do you think that a Chinese factory is going to spend checking & adjusting a headstock ? (or tailstock, or saddle etc.) For a machine that sells here (after shipping costs & re-seller profit margins) for the kind of money a 920 lathe sells for ?

unless you want to snap the lathe bed, it ain't going to twist.

         That is so far off the mark !

That leaves the tailstock which has only ONE degree of freedom- and that is sideways

The base of the tailstock doesn't wear as you slide it up & down the bed ? Are you sure you don't want to check that ?

To adjust the tailstock/poppet/ whatever you want to call it, ........

Or just mount you test bar & clock round that. And the bit that slides in & out of the tailstock body is the barrel (or chute).

Entertaining thread !

Regards,

Nigel B.

[Rob.Wilson]

Or maybe we could just use the search function on the forum ,,,, works well 
http://madmodder.net/index.php?topic=6146.msg65130#msg65130   

Rob

[Fergus OMore]

I'm sorry but 'We are NOT discussing precision tools'- At least I am not.

I was simply describing a fairly straightforward method to re-align a cheap lathe.

Of course, I know how to align a lathe and machine and scrape-- but this was an answer designed to correct mis-alignment- with tooling which the poster might reasonably possess.

I've sent Nigel a PM which should clear some misapprehensions.

[Miner]

Norman is of course 100% right about his comments regarding the Connelly book. It is a tough to read book. But depending on the level of interest it's well worth having. There's also far more knowledgeable people than myself posting their experience in this thread, So I think I'll just read how the rest of this goes.

Pete

[spuddevans]

Let's keep it all friendly here folks  It would be a shame to have to lock this thread if it gets out of hand.

Remember that, as in the case with many area's in life, there is more than one right way to achieve this, and there have been some great replies to help Dale P (the O.P.) with both sourcing a test bar, and some alternatives that can also achieve similar results using other equipment and techniques.

So let's keep the fine suggestions coming, but please keep it nice. Tell us how you would achieve greater accuracy with your lathe, it may be different from what I would do, but we all can learn something from each other when it is told by friends

Tim

[Pete.]

TBH if the test bar has MT3 fitment AND is drilled for centers it can be used to check everything to a basic level. Put it in the spindle and you can check the headstock is aligned with the direction the saddle is moving then put it between centers and you can check the tailstock for nod and sideways alignment.

All I've ever done is turned a bar between centers and mic'd both ends then adjusted the tailstock set-over to suit. That gets me a parallel cut and I can't warrant getting any further into it than that.