Myford ML 7 run out – spindle pointing up ?

[Matt Homer 1Participant @matthomer1]

I have been restoring and trying to fine tune a Myford ML7 lathe I bought in the Summer.

Done the tail stock alignment with test bar etc and got the lathe bed as level as I can.

However getting a serious amount of run out over 1 inch (2 thou) bigger at the end and smaller nearer the chuck.

tried the 4 jaw as well as the 3 jaw tru grip chuck but still the same result.

When using a dial indicators on the horizontal plane I get about .5 thou over 8 inches which I think is pretty good. However when I add another dial indicator to the top of the test bar it rises quite a bit….low near the chuck and slopes upward. Dial indicator on the rear of the cross slide.

I did mess around with the levelling feet on the ml7 stand when trying to get everything level so not sure if still a twist /bow or that it is the spindle that is out a few thou and pointing up if at all possible.

Also get the same result if use a centre in the spindle with the test bar and it appears to rise up the further you move along the carriage / go away from the chuck.

Using a solid toolpost I made and I understand about the backlash on the crossslide (arc euratrade kit purchased ready to do that)

Thanks

Matt

[Matt Homer 1Participant @matthomer1]

Got 2 thou run out over 1 inch / 25mm …. vertical plane out 2 thou

[Dave HalfordParticipant @davehalford22513]

You aren't the only one Matt link.

[Howard LewisParticipant @howardlewis46836]

Just a few quick thoughts.

From your description, it sounds as if the Tailstock is higher than the Headstock

If you levelled (removed twist ) by machining a "dumbell" you should be able to rule out twist in the bed, which we hope is not badly worn at the Headstock end.

Avoid using chucks when checking alignment. They contain their own errors which you do not want them introducing extra inaccuracies into your measurement.

It is possible that, at some time in the past, a different Tailstock, might have been fitted to your lathe, so that the centre is higher than the Headstock centreline.

I am assuming that you did check / clean the Headstock 2MT, bore and then trim up the centres before carrying out the between centre checks?

IF the Headstock is out of alignment, it means checking that there is no dirt / swarf between the Headstock and the bed proper, before trying to eliminate any error. As you already know, a minute error at the Headstock end will be magnified as you move towards the Tailstock.

Once the Headstock has been disturbed then it will need to be aligned , anyway..

To check the Headstock, for being level, you need to put a 2MT alignment g bar into the clean, undamaged bore of the mandrel. That will tell you if it the Headstock that is pointing up, or whether the Tailstock is high.

Howard

[Brian WoodParticipant @brianwood45127]

Matt.

I note that you have the dial indicator you are using to measure rise on the headstock mounted on the cross slide.

Have you been able to eliminate a wear hollow in the bed [which is often more pronounced at the headstock end] and what you are measuring is the hollow in the bed rather than a genuine headstock error?

Regards

Brian

[Dr_GMJNParticipant @dr_gmjn]

Posted by Dave Halford on 08/11/2020 16:28:08:

You aren't the only one Matt link.

Please note the corrected figures!

With a Myford 8" test bar mounted in the spindle taper, I'm getting a vertical rise of 0.0015" over the cylindrical part of the bar.

With the test bar mounted between centres, I'm getting a similar rise.

However, with the two centres point-to-point, the tailstock point looks lower.

I'm getting a 0.002" runout at the end of the bar, with it mounted in the taper.

Test bar between centres – 0.0015" difference along the top:

End of test bar in spindle taper, 0.0015" difference along top:

Centres, point-to-point – tailstock looks lower, yet the bar rises left to right. Could be something to do with the bed?

Here's the corrected Sheet:

If you want me to re-do any of the checks, just let me know.

[Dave HalfordParticipant @davehalford22513]

Looks like you never did have a problem.

If you look closely the top of the tailstock centre point is a little rounded – has no effect as it's inside the centre drill point hole.

[Dr_GMJNParticipant @dr_gmjn]

Posted by Dave Halford on 08/11/2020 21:03:10:

Looks like you never did have a problem.

If you look closely the top of the tailstock centre point is a little rounded – has no effect as it's inside the centre drill point hole.

Of course those figures are for a larger Boxford lathe I think, and my test bar is only 8" rather than the 12" specified.

Is there an equivalent table for an ML7?

[Matt Homer 1Participant @matthomer1]

Thanks guys…day off tomorrow so will have another go at it. Probably work through that sheet as a starting point.

[HopperParticipant @hopper]

Guys you need to relax and stop chasing perfection on a 50 year old low-cost lathe.

Most lathes of that era were manufactured with the tailstock centre a thou or two higher than the headstock spindle. This was to allow for wear. As the base of the tailstock wore, it came down into perfect alignment before it started to go out.

A small amount of vertical misalignment (And note, the term here is not runout, Runout refers to eccentricity under rotation) will not affect the operation of your lathe by any measurable amount. Adjust the tailstock horizontal offset so a 6" turning test piece between centres turns parallel along its length and be done.

Bed wear will also cause a carriage-mounted dial indicator to read a few thou variation along a ground test bar between centres and add to the confusion. The turning test tells the truth in terms of the end result.

However, if you do have excessive vertical misalignment of the tailstock of more than a few thou, the most common cause is swarf, paint or burrs either on the base of the tailstock or on the two mating surfaces that allow for the horizontal movement for adjustment between the two parts of the body. Easily fixed with a fine flat file.

They ought to ban the sale of those damn ground test bars. They are the cause of more unnecessary angst than they are worth. Adjust your lathe so it turns straight and true under working conditions and carry on lathing.

 

Edited By Hopper on 08/11/2020 22:53:30

[Dr_GMJNParticipant @dr_gmjn]

Hopper – I don't think we're chasing perfection. I'm certainly not. From my point of view, I'm not too familiar with lathes, and mine is old (1966). I simply want to know if it's within about the right ball park for accuracy of all the relevant alignments. If it's not, I want to know if I can improve it until it is. If it can't be improved within reasonable time and cost limits, it's fine – I'll live with it. I already know it can produce good enough parts for a small steam engine that works, but there's nothing wrong with wanting to get something to work as well as it can.

[Pete RimmerParticipant @peterimmer30576]

Posted by Brian Wood on 08/11/2020 18:19:25:

Matt.

I note that you have the dial indicator you are using to measure rise on the headstock mounted on the cross slide.

Have you been able to eliminate a wear hollow in the bed [which is often more pronounced at the headstock end] and what you are measuring is the hollow in the bed rather than a genuine headstock error?

Regards

Brian

If the error were due to a hollowed bed the test bar would read higher on the dial gauge where the hollow is and lower at the less worn end. It reads like he has the opposite of this.

[Dr_GMJNParticipant @dr_gmjn]

Posted by Pete Rimmer on 08/11/2020 23:00:57:

Posted by Brian Wood on 08/11/2020 18:19:25:

Matt.

I note that you have the dial indicator you are using to measure rise on the headstock mounted on the cross slide.

Have you been able to eliminate a wear hollow in the bed [which is often more pronounced at the headstock end] and what you are measuring is the hollow in the bed rather than a genuine headstock error?

Regards

Brian

If the error were due to a hollowed bed the test bar would read higher on the dial gauge where the hollow is and lower at the less worn end. It reads like he has the opposite of this.

Depends to a degree where the saddle is relative to the DTI plunger. My maximum bed wear was about 6-7” along the bed from the chuck end:

**LINK**

I suppose you could orient the DTI base arm so that the saddle was on a less worn area, take a reading along the bar, then try again with it orientated so the saddle moves over the worn part, and see if there’s a difference?

[HopperParticipant @hopper]

Posted by Dr_GMJN on 08/11/2020 22:54:56:

Hopper – I don't think we're chasing perfection. I'm certainly not. From my point of view, I'm not too familiar with lathes, and mine is old (1966). I simply want to know if it's within about the right ball park for accuracy of all the relevant alignments. If it's not, I want to know if I can improve it until it is. If it can't be improved within reasonable time and cost limits, it's fine – I'll live with it. I already know it can produce good enough parts for a small steam engine that works, but there's nothing wrong with wanting to get something to work as well as it can.

Well at .0015" rise between centres you are within the ball park, ie it won't affect your lathe's performance. It's not worth trying to machine/remove .0015" off the base of the tailstock for that.

Edited By Hopper on 09/11/2020 05:11:47

[HopperParticipant @hopper]

Posted by Matt Homer 1 on 08/11/2020 22:16:18:

Thanks guys…day off tomorrow so will have another go at it. Probably work through that sheet as a starting point.

Some pics of your set up when getting the errant readings would be helpful to make sure your set up is good.

Edited By Hopper on 09/11/2020 05:09:45

[Matt Homer 1Participant @matthomer1]

Right…think we have a combination of….

1. 0.04mm runout on the actual MT2 test bar !… Stuck this in between centres on my Warco 250v and get the same on amount of runout at the end with dial indicator ontop of the bar. Considering it should only be 0.0002 inch run out looks like something has gone wrong with that bar ?? Still struggle to understrand it as when you put dial indicator on the horizontal i.e check tailstock alignment, very little run out at all on the Warco using the Test bar.

2. Myford Spindle outer and inner (inside the MT2 hole) very little run out at all.

3. Carriage gets stiffer as your move it towards the tailstock end so presume that is wear. However did some measurements with micrometer and only a few thou out with thickness and width of the lathe ways from what I could see.

4. with 4 jaw on and dialed in on a piece of Aluminium I managed to get 1 thou difference over 1.5 inches. Taking very light cut at the end to avoid tool pressure etc etc.

5. Will try again with a longer piece between centres and see what I get at a later date as this was the initial thing that got me looking at it trying to machine a 12 inch piece of aluminium between chuck and tailstock and getting some very weird readings all the way along.

Cross slide thrust washer kit to put on and some more cleanup/tinkering to do over coming months so hopefully might get better results or spot something.

Will probably use the Myford for light work with Aluminium and Brass and stick to the Warco WM250v for more accuracy.

As stated from some comments the Myford ML7 is 84 years old so maybe I am expecting too much and the last owner used it for Wood turning as it was in such a mess !

Many thanks for informative comments.

cheers

Matt

[Dr_GMJNParticipant @dr_gmjn]

Only 0.002" wear on my ML7 bed was enough to lock it solid towards the far end if adjusted as a compromise at the chuck end. As I said previously, the most wear was about 6"-7" from the chuck, tapering towards the tailstock. You're not going to get the feel right without addressing that somehow.

I converted my ML7 to wide-bed, based on Hopper's method (but a bit simpler), and filed the wear off the front bed. It seems like mine was an anomaly, and usually the wear is on the inside face of the bed, which would have eliminated the need for filing. Anyhow, a pretty comprehensive account of how I did it is here:

https://www.model-engineer.co.uk/forums/postings.asp?th=168005

[Matt Homer 1Participant @matthomer1]

Thanks Dr_GMJN….will have a read.

by the way that workshop of yours looks to clean going by the picture of the ML7.

Cant wait to retire and get me a double Garage that might look like that.

Cheers

Matt

[Dr_GMJNParticipant @dr_gmjn]

Posted by Matt Homer 1 on 09/11/2020 15:54:04:

Thanks Dr_GMJN….will have a read.

by the way that workshop of yours looks to clean going by the picture of the ML7.

Cant wait to retire and get me a double Garage that might look like that.

Cheers

Matt

Ha ha thanks, that photo was taken back in 2007 when I'd just bought and cleaned the lathe up.

Even in that un-fettled state, it did a good job of machining the parts for my first ME project earlier this year, which was a Suart 10V steam engine (there's a build log in the appropriate section here).

It's worth doing a thorough evaluation of bed wear. You need two sizes of micrometer. Once you know that, you can decide what to do. Hopper's article in ME is very good for this.

[Andrew TinsleyParticipant @andrewtinsley63637]

Hello Hopper,

Not quite sure if I agree with you re test bars. IF you know what you are doing, then they are very useful. However I do take your point that in the hands of the inexperienced, they can lead the owner a merry dance.

Andrew.

[HopperParticipant @hopper]

Posted by Andrew Tinsley on 09/11/2020 16:17:36:

Hello Hopper,

Not quite sure if I agree with you re test bars. IF you know what you are doing, then they are very useful. However I do take your point that in the hands of the inexperienced, they can lead the owner a merry dance.

Andrew.

Especially if they are so poorly made they have 0.04mm runout like Matt posted above. Once upon a time such things were made by the likes of Moore and Wright and Starrett — and cost ten times as much — you could bet your life on them. Not so with today's cheap Chinese and Indian-made hobby tooling. Might be ok. Might not. It certainly pays to check before using, as Matt found out.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Hopper on 10/11/2020 02:38:14:

Posted by Andrew Tinsley on 09/11/2020 16:17:36:

[…]

Especially if they are so poorly made they have 0.04mm runout like Matt posted above. Once upon a time such things were made by the likes of Moore and Wright and Starrett — and cost ten times as much — you could bet your life on them. Not so with today's cheap Chinese and Indian-made hobby tooling. Might be ok. Might not. It certainly pays to check before using, as Matt found out.

.

This might be a good time to mention Dr Schlesinger’s thoughtful design

See p9 of ‘Testing Machine Tools’ [7th edition] … which is now freely available on t’internet

… So much for the exclusivity of owning an original printed one

MichaelG.

[Author]

Posts