Aligning a lathe to turn parallel

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

pete

Well-Known Member
Joined
Mar 12, 2008
Messages
480
Reaction score
89
Ok,
Andrews thread about "Lathe recommendations in Canada" started to go a bit off track and we were approching a hijack. Please see the last few posts about lathe alignment on page 3. Does anybody here have anything to add. Whats your approach to aligning a lathe to turn parallel? Methods? Even links to some specialized ways of doing it?
IMO, It's a subject that is really not stessed enough in regards to importance.


Pete
 
Pete,
I presume you are talking of a lathe that does not turn paralell from the carriage - ie the headstock is out of line with the lathe bed.

Any realignment method is dependent on how the manufacturer attached the headstock to the bed.

In the case of a Colchester it is via tapered pins. I whacked one out of line when I was an apprentice (left the chuck key in - damn fool) - to straighten it out I had to pull the pins (replacing them did not entirely correct the error) and via a process of running a dial gauge up and down a precision shaft (a bottle square with precision centres) between centres and then whacking the (lightly bolted but not pinned) headstock with a mallet and measuring with a dial gauge to get it back into line. I then had to re-ream the tapered pin holes and re-install the pins.

The whole process required several methods - shaft in 4 jaw - check tailstock - precison shaft between centres and the final proof was an accurately turned up piece of bar by the best turner in the shop who had to "sign it off" - I remember the tongue lashing I got for buggering the lathe up in the first place as well as a few more each timed he deemed my realignment "not good enough".

My mini lathe has a headstock profile that matches the lathe bed so any error here could only be corrected by scraping the underside of the headstock (horrors) or reboring / sleeving the main bearing holes in the headstock.

etc. etc.

You need to first measure to determine where the error lies and then figure the best way of correcting it.

So I don't think there is any specific way to fix this problem as each machine design will have its own peculiarities.

Question - How did the lathe get out of whack ?

Ken
 
Ken,
Nope my lathe is just fine, I have my own methods for lathe alignment and think I get pretty good results. I started this thread as maybe a way to show guys that just bought their first lathe that there's more to it than just bolting a lathe to a bench top and then thinking their good to go.

Pete
 
I would like some more info on this if possible!

For example:
The lathe is on a sturdy bench that is level to the concrete floor.
The lathe is bolted to the table.
I would need to check the lathe bed with an engineers level to get it perfectly level.(?)
If all that is good, how does one tell if the headstock or tailstock is out of alignment?
Do you require a precision cut bar to ensure it is in alignment?
I have read of bars that you can buy that have morse tapers cut at each end to fit the headstock and tailstock,
is this the ideal method?
Or would using a D.I. be sufficient?

Sorry for all the noob questions, but that is what i am!
I am reading to try and gain more understanding of techniques and methods.

Andrew
 
Pete,
Yes - I read the original thread - jumped to the (erroneous) conclusion that it was something more fundamental.

Lathes can get twisted out of shape by injudicous bolting.

Most of the larger industrial size lathes I have had mounted on anti-vibration mounts so simple leveling generally sets them up right.
When rigidly bolting to a foundation you shim so that it is standing firmly before bolting - during bolting place a long shaft in the chuck and a dial gauge in the toolpost about halfway down the carriage - if the dial gauge starts moving as you tighten down - check your shimming - some movement is normal but the dial gauge should be where it started once fully tightened.
Some machines have jacking bolts adjacent to the hold down holes and you can work them against each other so as not to "bend" the lathe.

Agreed it is never a simple bolt it and forget it - that can get you a permanently twisted lathe.

Andrew - leveling is always good practice but I wouldn't lose too much sleep over it - I'd be more concerned with ensuring that my bolting is not twisting the lathe.

Ken
 
As Pete says, we were discussing ways of securing lathe bed to bench.

1 - our small lathes are very light, our benches are often not as solid as they should be and workshop floor is often flimsy timber. However much effort you out into securing the lathe without distorting it, it is often all in vain because everything moves, twists and distorts.

2 - all the lathes with more than three fixing points - by default it is difficult to achieve flat plane that avoids distortion. If you have one fixing point (like on watchmakers lathe with pedestal under headstock) you cannot distort it by fixing to bench. Similarly with two fixing points (again mostly small lathes) you are not twisting the bed (as long as you discount the width of the pedestals). With three fixing points - you achieve flat plane by default. And yet most of the lathes have at least four and often more bolt downs.

3 - however carefully you adjust all the multitude of legs/fixing points, unless you leave one end of the bed free, you will be affected by thermal expansion distorting the bed fixed rigidly to bench

Solutions:

Some people fix the lathe to something substantially stronger and heavier than the lathe (eg granite block).
I was in turn extolling virtues of the idea used by Hardinge in their HLV-H lathe. Three point support (two under headstock, one under tailstock), tailstock end allowed to expand as well as preventing bed from twisting (single point). All three points sprung loaded. One could adapt similar idea (using intermediate plates between lathe and bench) to our little lathes.
Advantage of this arrangement - it is impervious to flimsy floor, weak bench, not affected by thermal distortion and not bothered by owners adjustment.

Chris

ps mine is Prazimat DLZ
 
lazylathe said:
...I would need to check the lathe bed with an engineers level to get it perfectly level.(?) ...
Doesn't matter at all whether it is level or not - levelling is simply an easy way to make sure that the bed is not twisted i.e. if both ends are level than it means that both are the same (not twisted), but you can incline the bed towards you (for example) and providing that both ends are inclined by the same angle - you are fine.
Of course you will find both internet and hobby magazines full of people who will tell that it will never work unless you have it absolutely level.
:big:
Don't let them fool you ;)

Chris
 
My last hour of single finger typing about this has apparently disappeared, Till I cool off I'll sit back and read all your points.

Pete
 
I am sorry to hear about your misfortune Pete.
I am sure it has happened to all of us at some time, usually after some long typing session ;)
My trick to prevent this disaster, is to select just typed text and "copy" it, before using either "preview" or "post" buttons - this way if anything goes pear shaped, I simply use "paste" to recreate my lost posting.

Chris
 
Chris,
Thanks, That's what I get for trusting the computer gods.

Pete
 
OK, Let's try this one more time.

Andrew, Yeah those accurate test bars with the Morse taper on one end do work, But won't do everything. Your headstock and tail stock tapers must be spotlessly clean, Properly ground, and undamaged. A .0001 reading DTI would probably be the best to use, As a .001 reading dial indicator IMO doesn't give you a fine enough resolution.Slide the test bar into the headstock M/T and give it a fair tap to seat it into the taper. Fixture your DTI so the probe tip is really close to the C/L of the test bar. Use your carrige travel to run back and forth on the straight portion of the test bar. Reset the probe tip to the top of the test bars C/L and repete the above test. Do the same testing setup for the tailstock. This will show just how well both the headstock and tailstock are aligned with the lathe bedways. But it won't show how well the headstock and tailstock are aligned to each other.

All the postings about lathe leveling are ok, But a dead level lathe is just an easier way to get your lathe real close to what you really want. Those postings forget that this is a static condition. Due to various tollerance stacking ect. you want to test your lathe under cutting conditions. It's a lot better to have a lathe that turns parallel end to end than one that's just level. You could set the lathe up to be 45 degrees out of level and if the lathe is well aligned it will still turn parallel.

To do this I use a 2" diameter bar of even scrap material around the same length as your lathe will take between centers. I turn between centers so in the future I can do testing by lightly machining the bar again. Far faster to set up than to accuratly set up the bar in a 4 jaw chuck. Once your scrap bar is machined end to end so it's fully cleaned up you make a light depth of cut pass the full length of the bar. Measure each end with a accurate micrometer. Any differances in size, Then your lathe needs aligning. Since my Emco lathe has a tailstock that can't be adjusted to turn tapers, Then what needs to be done if the test bar isn't parallel is twist the tailstock end to match the headstock end. You do this by using shims or jacking bolts. Say if your test bar was slightly smaller at the tailstock end. You need to twist the lathe bed AWAY from the cutting tool tip. I'm not 100% sure of the exact order lathe alignment is done with a lathe that has a tailstock that can be adjusted side to side for tapers. So I'm hoping that someone who does will post that here.

There's a lot more to aligning a lathe than I've posted here, I didn't even mention what to do with shims or jacking bolts, scraping ect. if your headstock is missaligned with the lathe ways, but this is getting long enough. Obviously when your getting down to .000? tollerances then EVERYTHING you do has a effect somewhere. If your next part you want to machine on the lathe is a harder material, Different material, Larger depth of cut ect. Then this can and will show up to greater or lesser extent when trying to turn truly parallel. The bench the lathe is bolted to, Method of bolting it down as mentioned by Chris, How to let the lathe compensate for temperature fluctuations ect,ect,ect. Needs to be addressed too. This is why I started this thread because it's a lot more complicated than a lot of people think. Is it nessisary to align a lathe to .000? tollerance? Only the user can make up their own mind about that.

Oh yeah, I forgot to add that I think it's a good idea to test any lathe. If you don't test it then you don't really know just what your actually able to achieve as far as tollerances for any machined part.

Pete
 
Hm-mm, Over 450 views and nobody wants to add to this. I was hoping I could learn something too. Will have to check this thread when I get back from work in a couple of weeks.

Pete
 
Pete,
Most times I use the shave the bar method, as you describe above.
But a very neat alternative is Rollie's dad method. A lot written on it, so I give links.

igor.chudov.com/manuals/Rollies-Dads-Method-of-Lathe-Alignment.pdf

And if you like video

[ame]www.youtube.com/watch?v=ZaIuHNFIFNw[/ame]


Take care,

tom in MA
 
There is a lot of info in this post for me to try and digest!
I hope there is more to come!

A lot to contemplate and figure out.

I can see the ladder in the distance but i am so far from the first rung it is not even funny! ;D

Andrew
 
When I refurbed my old Myford ML2 or 3, it was just a matter of slackening the head hold down bolts, and gently tapping the head into correct alignment.

The Atlas on the other hand was brought into alignment using a cold chisel and a file on the guide that sat in the bed channel. By careful manipulation, you gradually 'kicked' the head around until you got it turning perfectly parallel.

When I came to fix a mini lathe, I actually used cooking foil shims on the bedways under the head to achieve the same thing. Bottom part of this post.

http://madmodder.net/index.php?topic=627.msg5193#msg5193

On the new lathe I have now, there are jacking bolts that can be used to kick the geared head around if ever it starts to turn tapered.


Bogs
 
Thanks Bogs!
I had read through the whole posting on fixing Darren's lathe and it was an excellent read!
That was some amazing work and very logical thinking of how to progress with such a fix!

Andrew
 
I am very tempted to try out the Rollie's dad method on my Mini Lathe, but rather scared to find out what the results would be. I already know it turns a slight taper, but not knowing a value almost makes it easier to ignore. ;)

Seems to me that having a value attached to it is going to get me down a dark path of chasing my own tail forever ::).
 
Getting it adjusted to perform best is one thing, keeping it this way without constant tweaking of your jacking screws and bolt downs is completely another ...
So if you start with timber floor, timber bench etc you are might as well not bother - all these things are affected by both humidity and temperature, constantly twisting, bending and changing.
You are better off with concrete slab for a floor, solid metal or granite for a bench etc.
But ... (yes I know I am boring ;D) why confront the problem instead of avoiding it - let me repeat here (because it's the best place for it I think) what I have already described in Andrew's thread - arrange three point spring loaded boltdown and you will never worry about twisting your bed or needing to adjust it.

Chris
 
I used the dad's method to initially set up my lathe. This got it close but seemed difficult to zero in and wasn't perfect. I recently readjusted it, this time turning a 9" long piece of 2" diameter aluminum and measuring taper with a 10th's mic. I was able to bring it true like this. My Clausing 4900 stands on 4 anti-vibration pads with adjusters on concrete. I didn't have access to a machinist level so I'm not sure how close mine is to being level.

Greg
 
bambuko said:
Of course you will find both internet and hobby magazines full of people who will tell that it will never work unless you have it absolutely level. Don't let them fool you ;)
Never work? I doubt it, . . work better maybe. One of the things I've done since I entered the hobby is read. Read read read read read - everything I could get my hands on going back before the turn of the previous century. I've also talked with many people with years of machine tool experience and read dozens of accounts or specifications for equipment installation. I was involved in design and construction of two automotive plants in the South (Nissan, Saturn) including the construction of concrete machinery foundations and subsequent machinery installations and absolute LEVEL was of primary concern. I was also similarly involved with the foundations for a major MRI installation at Vanderbilt Univ Hospital where there was even more concern with absolute level and elimination of vibration. Nowhere have I ever read, or heard, even once that leveling didn't matter, even for the garage guy, until today so you'll have a lot of 'splainin' to do to convince me otherwise. Personally I doubt if all those people and sources are wrong but then, what could it hurt? The problem for most of us is that a level of the level of precision necessary to be of any benefit is very difficult to borrow (and very expensive to buy.) We can't just run over to the tool crib and check one out so we do without.
 

Latest posts

Back
Top