Rebuilding an ML7

[Andrew TinsleyParticipant @andrewtinsley63637]

At long last I am reassembling my overhauled ML7. Now when it was dismantled. I noticed that the shims on the headstock were a bit weird, The rear bearing had equal shim thickness back and front and the front bearing had significantly more shims on the rear than at the front.

Now this leads me to the suspicion that the headstock isn't aligned to the lathe axis? True of False?

I can obviously check the headstock alignment by turning a bar and using a DTI on the saddle to check the bar on its side and top.

Now here is the rub. This same test is used to see if the bed is straight. so how do you tell the difference between the two?

Andrew.

[Andrew TinsleyParticipant @andrewtinsley63637]

[Neil WyattModerator @neilwyatt]

Posted by Andrew Tinsley on 06/09/2017 11:50:21:

Now here is the rub. This same test is used to see if the bed is straight. so how do you tell the difference between the two?

You can't but it doesn't matter…

[Ex contributorParticipant @mgnbuk]

Now this leads me to the suspicion that the headstock isn't aligned to the lathe axis? True of False?

I would suggest false. The shims set the bearing clearance, not the alignment of the spindle axis with the bed guide faces.

Now here is the rub. This same test is used to see if the bed is straight. so how do you tell the difference between the two?

You dont. Now you can see why machine tool builders use a precision level (or other means, such as a laser interferometer) to set the bed "level " (i.e. without twist or bow) first. You now have a reference to check the headstock alignment using a test bar in the spindle (dial gauge on the saddle – check in both vertical & horizontal planes).

If there were cheap & easy shortcuts to building machine tools, be assured the machine tool builders would have found them by now – the Yorkshire based companies certainly would !

Nigel B

[Andrew TinsleyParticipant @andrewtinsley63637]

Thanks for that answer Nigel. I do have a decent level to check for twist in the bed. But I always assumed (perhaps wrongly) that the final check was that from turning a bar and checking with a DTI.

I can certainly use my 12" Rabone level to check for bed twist. Once that is correct, I can then infer that a turning test would reveal any misalignment of the headstock spindle?

I was suspicious of the difference in shimming of the front bearing, if I were doing the job from scratch, then I would shim equally front and back of the bearing shell housing. The fact that they were unequal makes me suspect that a previous owner bodged the job of adjusting the bearing to overcome wear. Is this a reasonable assumption? Or is it quite common to have different shim thicknesses front to back?

Neil, why should you say "it doesn't matter" or is this simply being tongue in cheek?

Thanks,

Andrew.

[Tim StevensParticipant @timstevens64731]

If you level your bed, and then machine a test bar with the tool in the conventional position, – at the same height as the head and tailstock centres, supporting the bar only in the headstock, you will be able to check the horizontal alignment of the headstock with the bed of the lathe. The headstock can still be out of line in the up and down sense.

If you find that your test turning produces a truly cylindrical bar, the headstock is parallel with the relevant surfaces of the bed. Now try the same test with the bar held between centres. If this is also a parallel result, your head and tailstock are aligned horizontally.

If you can arrange the toolholder so it runs with the cutting point directly above, or below, the bar, – not easy in practice – similar tests will show how the headstock etc align vertically. Of course, holding the tool in this position will put loads in ways the machine was not designed for, so the recommendation is to take a very small cut, keeping the loads low. And do it carefully, as the feed knobs will not move the tool as you might expect.

There are usually adjustments that can be made to correct any horizontal out-of-line findings, but on the modelling lathes I have worked with, the up-and-down condition depends on the factory, and changing anything means a serious regrind of various parts. So most of us don't bother.

Hope this helps

Tim

[Martin DowingParticipant @martindowing58466]

Hi,

I am now in the process of doing very much the same (rebuilding or rather improving precision) of old ML 7.

1. Spindle bearing. This one usually need scraping in.

When I got my replacement set, I still have bought new set from Myford Nothingham.

I was advised that the set is carefully made and and it is unlikely that futher scraping will be needed once installed. All what I had to do is to sdjust shims.

I was recently wiggling test bar homed in socket and there was no more than 0.5 thou of DTI movement as checked on side of spindle register.

That is very good.

2. Leveling.

Precision level with accuracy better than 1 thou per foot (0.05mm/m) was used across bedways at headstock and tailstock. This is a bit overkill, lower grade level, say 3 thou per foot is good enough but I only have much better one.

Leveling was done up to 1 thou per foot accuracy. A lot of piddling here…

3. Headstock allignement.

Once lathe got levelled I have done a turning test on it,

Decent 1 1/2 inch diameter MS bar with 2 bobins od 2 inch diameter F/C steel spaced 6 inch was installed on 4 jaw chuck, centered properly and turning test was done and measurements taken

Unfortunately that was much off, something like 8-9 thou a foot off!

This sad state of affairs had a long legacy. When I have bought my lathe, it was worn. Bedways, saddle and tailstock base were rescraped by professional but dmaged spindle bearings was left as they were because I didnt have a new set yet. Hence said professional couldnt allign a headstock and now I had to do it. I had to reallign machine once new bearings and spindle have been fitted.

Fortunately it was not as difficult as many legends are stating.

So 4 high tensile strenght screws bolting down head to bedways were loosened.

No, usual Allen key wont do it, had to buy a proper one with a socket.

Then 2 screws pressing headstock to a shear was losened and I hoped to be able to move the headstock but I couldn't.

Some moron, sorry, pevious owner, filled everything with paint. I had to use (carefully) some lever to disassemble "paint welded" headstock.

All the paint was now cleared from vital surfaces.

There is some play between headstock and shear, allowing for a tiny twist.

Adequate shimming was inserted from rear side and headstock was bolted back. before final tightening 2 screws pressing head against a shear were also tightened.

Some maths is handy here and it helps greatly to calculate thickness of shims before scewing it back. Armed with basic trigonometry I needed to make final adustement only once by *axially* moving already installed shims less than a quarter of inch.

Final accuracy by turning test better than 0.0005 inch a foot.

Not bad… but it is not all.

Now *vertical allignement had to be checked… and turning test is not very good for it (less sensitive).

So I removed test turning bar from the socket and inseted hardened 25mm ground bar used in linear bearings and centered it very well in 4 jaw.

First I reconfirmed horizontal allignement, few times, turning chuck a bit each time to make sure the bar is not bent.

I was really pleased to record better than 0.0005 inch a foot allignement again.

Now I checked vertical allignement by applying DTI from the top.

Oops… head was watching up about 1.5 thou a foot.

This was remedied by loosening bolts like before and inserting shimstock under back of head (trigonometry help here, first attempt was the last…

All described drill were repeated, final outcome was better than +0.0005 inch a foot horizontally and no measurable deviation with my dti vertically.

But what about facing?

To check error there I needed a decent *flat* faceplate… and all of mine are concave with concavity exactly like my lathe produces.

So I grabbed one and gave it to professional guy on surface grinder… and he ground it flat.

I have installed it on my machine and checked with DTI for "wobbling". It seems that professional guy done his job right as it didnt "wobble".

Now I have run DTI across. Just below 1 thou a foot. Thanks God… Crosslide/saddle alignement is good enough.

Unfortunately this is not all… You remember I was using hardened 25mm linear bearing bar held in 4 jaw for allignement testing. Meantime I have ordered one to use with Morse socket and it came.

And guess what… once installed it wiggles about 4-5 thou per foot of lenght. I started to look on this test bar with utmost suspicion, after all it came from RDG… However centered in 4 jaw it shows very good allignement like my previous test bar did. Unfortunately 12mm hardened linear bearing bar held in 12mm hardly used original Myford collet also shows bad runout.

It seems that taper need regrind. Now I am making substantial cross slide mountable holder of my machine mountable internal grinder. It is Milwaukee tool and can rotate up to 30000rpm. My ML7 has taper turning attachement so I should succeed.

I intend to but hardened *master MK2 male taper* to accurately allign everything befor this tiny 0.2 thou cut, grind few sockets to make sure I have learned to do it right and then fingers cross and hope for the best…

I will probably ask few question before I actually do it but I welcome any suggestions now.

Also who sells this *MK2 male master taper* with reasonable price?

[Author]

Posts