Parallel turning on a Myford ml4

Hi , last week I needed to turn a shaft out of a length of 6082 aluminium 200mm long . I held 40 mm in my Pratt & Bernard 3 jaw chuck and had 200mm sticking out o the chuck supported by a revolving centre ,I proceeded to turn the bar and had a run out of 0.5mm ,not good.

I did a lot of reading up on here and on the Internet and decided too see if I could get the lathe too turn as parallel as possible ,bearing in mind my lathe is a Myford ML4 so it's approx 70yr old .

All my jib strips are tight and there's no play in the bed and the bed looks in good condition ,I then decided too hold the bar in my 4jaw chuck again a P&B chuck as I thought the 3 jaw may not be as accurate as the 4jaw and I haven't got a centre for the headstock.

I used a dti to true up the bar and also checked the run out at the tailstok end which is supported by the revolving centre ,I manage too find a happy medium between the two.

I then proceeded too turn my bar and I still had a run out of 0.5mm ,I decided to place the dti on the tail stock end and adjust the tail stock as I felt my tail stock was a long way out ,after many adjustments I got my desired results ,I now had a run out of 0.05mm checked with my micrometer not bad for an ml4

Pics of the bar measured with my vernia 

Edited By Kevin Fenrich on 05/01/2013 15:20:18

Edited By Kevin Fenrich on 05/01/2013 15:21:13

Edited By Kevin Fenrich on 05/01/2013 15:22:02

What does it turn like between centres? thats how you normally set up a machine.

_Paul_

A good revolving centre in the tailstock and an adjustable tailstock can save a load of hassle with these old hobby units

Tweak and go, great fun

The original MLs has unadjustable tailstocks...yuck, this was eventually sorted by myford (see bottom of page) and makes a huge difference for a hobbyist

Thanks for the replies , @ Paul, I don't have a centre for the head stock ,so I can't answer that question , but overal Im pleased with the results .

Kevin,

Place a piece of bar in your chuck and turn a 60 deg taper to a point, this will be your centre at the chuck end.

Martin P

Kevin

You dont need a centre for the headstock, just turn a 60 degree cone on a bit of scrap bar held in the chuck.

Ian

Thanks , I don't know why I didn't think of doing that in the first place

Edited By Kevin Fenrich on 05/01/2013 21:27:36

Kevin,

I'm not familiar with the ML4, but I assume its basic structure is similar to the 7 series - i.e. it has a foot at each end of the bed, with holding down holes at front and back of each foot.

If so, then you need to go through several steps on setting up. The first essential is to know that the bed is not twisted - if it is, then the saddle will not move parallel to the spindle axis, so all turning held in the chuck will be tapered. You will be surprised how easy it is to hold the bed down in such a way that the bed is twisted. The only reliable way to correct this is to chuck a substantial bar - at least 1" diameter and 6" long - with no tailstock support, and turn small bits at remote end and headstock end at the same cross slide/topslide setting (it helps if you relieve the bar slightly between the two ends to remove the need to move the tool when switching between the two). The two lands you turned should be identical in diameter - a couple of tenths out should be acceptable.

If the remote end is larger, that means the bed is tending to move away from the tool point as you move away from the headstock; slacken off the rear bolt of the tailstock foot, and shim it slightly under the foot and re-tighten. Try again, should be better. Repeat until happy. Obviously, if that end is smaller, then shim under the front side of the foot instead.

Some people will tell you that you can do this with an engineer's level. I have an engineer's level, and I've tried it, you can't get the same accuracy (and the levels cost); the way I described is (a) more accurate and (b) free.

Then, when happy the bed is not twisted, check the tailstock is in line with the headstock spinlde axis. The proper way to do this is with a test bar and a DTI, but simply aligning centres in headstock and tailstock is fairly good.

Then you can turn between centres and know that you will get a parallel result. For holding in a chuck you need to be sure your chuck is acceptable. Old 3-jaw s/c chucks are notoriously poor in this respect. You can use a 4-jaw independent, use a collet, buy a Griptru chuck (pricey) and learn to use it, or remedy your old chuck, perhaps by re-turning the register on the backplate (or even providing adjusting screws) and possibly grinding the jaws if they are bell-mouthed.

David

Edited By David Littlewood on 05/01/2013 23:43:20

Hi David

your recommended way of setting up puzzles me. In't it the opposite of what Harold Hall recommends?

Regards Thor

Hello Kevin,

I inherited my late father's ML4 in 1963, it was aleady about 20years old then and I ran it for a further 20 years before selling it on to buy a friend's ML7.

I suspect you have the model with the plain non-adjusting tailstock, a real pain to set up both in terms of alignment across the bed and true aim along the bed. With the single foot mounting for the whole machine, leaving the tailstock end of the bed as a cantilever, the advise above from David Littlewood about bed twist and correction doesn't apply.

A relatively quick way of setting things true to at least a reasonable approximation, and assuming you can trust your 3 jaw chuck, is to do as follows:

Slacken the tailstock foot adjusting screws and mounting bolts along with the bed clamp on the rear of the tailsock. Wind out the barrel to about 3/4 reach, clamp it with the barrel clamp and then grip the front end in the chuck. Now that at least is aligned with the headstock centre-line and with it the tailstock body.

Now you can reset the tailstock front shear dovetail and carefully tighten the bolts on the tailstock holding the dovetail piece to the tailstock, checking with your DTI that you are not moving the tailstock body as well; this may take a while.

When you are happy with that and the bolts are tight, test again as you tighten the bed clamp for the tailstock, it should not deflect more than a few thou if you have got the previous operations right.

If you have some 1MT tooling with a plain parallel section, grip that in the chuck instead of the barrel, the 1MT barrel bore is the definitive centre for the tailstock.

You might still be able to get a handbook from Tony Griffiths, visit [email protected] he has a wealth of handbooks for all sorts of machinery, it will be very useful for you.

I hope that helps

Brian

Thor,

You are quite right, I said it the wrong way round. If the cut at the tailstock end is larger, the front of the tailstock foot needs to be shimmed up, to bring the tool nearer the piece. Thanks for correcting me.

Serves me right for writing late at night after sharing a bottle of wine with dinner!

Brian,

If the ML4 has two mounting feet, then I disagree with you; it is IMO far more important to remove twist from the bed than it is to align the tailstock by twisting the bed. The former affects everything, the latter only when you use the tailstock. It is in any case hard to see how a non-adjustable tailstock could get out of adjustment, but if it is out I believe it would be far better to put it right properly by scraping and shimming the locating part. Correcting one fault by deliberately introducing another (twist) is hardly the way to getting things right.

David

Edited By David Littlewood on 06/01/2013 12:06:06

David,

Brian's method is correct. Harold is writing about a Myford with the need to adjust for possible  twist due to incorrect mounting,  the ML4 is, as Brian says, cantilevered and so the shimming does NOT apply.

Edited By KWIL on 06/01/2013 12:06:52

KWIL,

OK. I think was confused by Brian's reference to a "tailstock foot adjusting screw" as I thought that meant the bed had two feet. The danger, I suppose, of commenting when I don't know the machine in question. The advice (corrected!) is good for bipedal lathes.

BTW, there is no guarantee that a cantilever lathe bed is not twisted - maybe through use of unseasoned castings - though how one corrects it is a different matter.

David

An ML4 bed will twist in just the same way as a Myford 7 series lathe if secured to an uneven surface and so the same precautions should be adhered to for both types of machine.

As has been stated numerous times before: Level the bed with a engineer's level to ensure all twist is eliminated, then take a test cut on a substantial piece of free machining metal bar, using a sharp tool.

If the bar is unacceptably out of parallel, then you need to look for the root cause, such as headstock misalignment or wear etc. If the bar is out of parallel by a thou' or two, then you could tweak the mounting bolts by shimming or other adjustment until you're happy.

The first rule though is to ensure there is no twist in the bed - particularly with a fixed tailstock machine.

Martin.

Thanks for all the replies intersting reading , I checked the bed with an engineers level before I adjusted the tail stock and shimmed accordingly , my ml4 must be a late model as I can adjust the tail stock by loosening two nuts and then I can adjust using the 2 small screws on the side of the tailstock ,kind of like a jib strip , I'm fairly happy with the results 0.05 mm over 200mm is quiet good for a machine of this age , at the tail stock end of the bed I have d1992 stamped ? Any info ?

Hello Kevin,

D1992 will be the serial number. Mine was D2382, bought new by my father in 1945

From what you say, your model sounds like the non adjusting tailstock, the later Myford mod. had a tongue built in across the tailstock base [as in the ML7 which replaced it] so that it was possible to offset PARALLEL to the lathe bed when turning tapers.

Before then, with just the dovetail jib location, there was no control on the alignment at all and it was possible to set up to turn parallel and then find you were drilling at a skew from the tailstock. If you have a longish straight bar you can trust, put that in the tailstock chuck and measure displacement down its length with your DTI. If it isn't too far out I would call it a day and heave a sigh of relief!

Incidentally, the feedscrews for both X slide and top slide are 12 TPI; I know the dials are marked in 100 divs, but one full turn moves the tooling by 96thou.

I modified my headstock spindle with a close fitting collar to bring the register diameter up to 1.25 inches so that ML7 chucks and accessories would fit. As supplied, the nose was 1.125 inch register diameter with a 1.125 inch x 12 Whit thread and it became difficult to find things made to that size.

Brian

Edited By Brian Wood on 07/01/2013 10:38:23

Thanks Brian ,the info you supplied is much appreciated .

Hello ML4 club members. My machine is D2017, which I recently bought from a club member. I have only just finished setting up my workshop and your post has prompted me to reply. I carried out the rollie's dad method of lathe alignment and after some adjustments I am happy with the accuracy. I checked against any potential twist using a fancy new electronic level device when I fitted the lathe to its bench top. I am lucky to have the adjustable type tailstock. Which I found, after checking had been offset to parallel?

My method to overcome the register diameter issue was a internal bush fitted inside a new purchase ML7 backplate. This allowed me to use the three jaw chuck that fitted the register and a four jaw chuck was fitted to the new backplate with its internal register now at 1.125". I will most likely do the same for any other backplates I get in the future.

Somebody once said there is life in the old dog yet. For our venerable lathes this seems to be the case.

Errrr, I make 12tpi to be 83.3 thou per turn?

Regards,

Andrew

It was a remark of my father's when I was 10 years old, I never questioned it at the time, or checked it later, so I could be mistaken.

In those far off days we didn't have independent means to verify actual movement compared to iindicated.

I usually measure everything anyway without depending on what the dials say

Brian