One very large can of worms!

Hi

After checking over a used but cheap and local machine I eventually bought a 'Grayson'.

I watched it spin up and checked it over before buying it and was very pleased with the overall condition although there is some wear and backlash evident.

Now it is at home and setup on a bench I have realised that actually getting it all true is going to be a problem, I will try to explain why.

Whilst using a dti to check the chuck for run out (0.7mm) I noticed that the headstock does not appear to sit 100% parallel on the bed, the headstock casting is bolted to the bed casting and there is a small amount of adjustment available.

Q1 Where should I start in ensuring that everything is actually running true?

Please bear in mind that I don't need absolute accuracy and will mainly be turning aluminium bar of no more than 120mm length and 8 - 20mm dia.

Q2. I need the tail stock will need to be aligned to the head as Iwill want to drill and ream holes which will need to be as accurate as possible.

Any ideas would be very helpful and gratefully received.

These lathes were built to a price and you're lucky that you have a separate headstock - at least it can be adjusted!

I'd start by checking the lathe's spindle nose and internal taper for truth to each other. Ideally then you'd need one of those test bars which you can insert into the carefully-cleaned spindle nose taper; what is it - 1MT? You can then check that the headstock is pointing in the right direction using a clock gauge in the toolpost, running it along the top and on the front of the bar. If it's "out", you can see what you need to do to make it right.

The tailstock alignment is another problem; it can be worn on the base (an old ML7 I acquired had 0.025" wear here!) so that it no longer lines up vertically with the headstock spindle, or (assuming it has the feature) offset so that it turns taper. You can put a clock gauge on a piece of bent bar in a chuck fitted to the headstock spindle and use that to check the concentricity of the tailstock taper to the headstock. Again, the measurements should tell you what to do to correct any errors.

The test bar can also be used to check that the tailstock is pointing level and in line with the lathe bed in the other plane.

0,7mm chuck runout is a tad excessive. I think you'll be forever turning up split collets to hold second-operation work until you get a new chuck or grind the jaws of this one.

I think you'll be spending quite some time and swear words on this one...good luck.

Edited By Nigel Bennett on 07/09/2016 10:03:09

Nice little lathe like an early Drummond and worms are good for the soil.

Some people will tell you to get a morse test bar but you don't want the expense and the taper in the spindle may be less than perfect anyway. So.
Find a bit of tube like old water pipe a few inches long and chuck up firmly without tailstock support. Turn a small section near each end to be the same diameter moving the saddle and topslide as necessary to achieve this. You now have a temporary test bar that is by definition dead in line with your spindle.
You can use this to align the spindle to the bed though (not allowing for bed twist), and with some thought to the tailstock aswell.

Then search for 'levelling a lathe' for the next stage in removing bed twist but with a small lathe if you don't bolt down the tailstock and induce some twist you can start off by ignoring this.

+1 for Bazyle's suggestion but even turning a section at the end of the bar isn't strictly necessary if you use an indicator and compute an average reading of max and min values obtained from a full revolution of the test bar.

I was recently provided details of this average method by a helpful member on the Myford users forum and used it to level my old Myford.

I'll try to find the instructions and post them.

Jon

This may be of use at some point.

**LINK**

I agree with the keep it simple brigade as has already been suggested.

Essentially what you want out of a lathe is that it turns parallel from the chuck and that it turns parallel between centres. Turning is always the ultimate test.

Test bars and the like are useful short cuts when adjusting but are not necessary if you don't have them.

regards Martin

I think the test bar would be worth it, you can't always trust barstock to be totally right or that the cut will be representative of what is going on with the lathe, for e.g loose gibs can cause a tapered cut. Not necessarily anything to do with the headstock.

Michael W

DP

Edited By Michael Walters on 07/09/2016 12:39:11

That was precisely the document I was referring to.

Thanks Vic

Jon

"Q1 Where should I start in ensuring that everything is actually running true?"

Easiest way to start off with is to chuck up a piece of aluminium the length you say you will be working with, no tailstock centre, and take a cut over the full length. Measure it and see how the lathe is cutting - parallel or tapered. If you get lucky and it's turning parallel, no need for further mucking about with headstock or bed.

To check the tailstock, mount the piece of aluminium between centres and take a fine cut over it. Measure the bar after cut and if its tapered, adjust the tailstock to suit until it turns parallel. Double check for vertical wear/misalignment by bringing the tailstock and headstock centres together and put a steel rule in between. It should stand up vertical if alignment is ok.

Then, at least, you have a starting point.

Before you do the above, the lathe is best bolted down to the bench. Preferably a steel bench, as solid as possible. When bolting the lathe down, check all four feet with feeler gauges. If you can slip a feeler gauge under a foot, slip a piece of shim of that thickness under the foot. This way, when you tighten the four mounting bolts down, they will not pull the bed out of shape. Again, this provides a starting point where you know where you are at. You can add more shims later IF it turns out the bed requires levelling.

Otherwise, it can be very easy to spend a lot of time chasing your tail aligning one part of the lathe then another, which throws out the first and so on.

Your .7mm runout on the chuck is way excessive by a factor of about 10 or so. You should look at what is going on there before doing any test turning. Check the runout on the actual spindle. Then on the chuck backing plate. Then on a known round object held in the chuck, such as a gudgeon pin or at least a good piece of round bright mild steel.  That way you can determine if the problem lies in the spindle, the backing plate or the chuck jaws themselves. Runout of the actual body of the chuck does not necessarily tell you much.

Edited By Hopper on 07/09/2016 14:07:18

Edited By Hopper on 07/09/2016 14:15:54

Edited By Hopper on 07/09/2016 14:17:33

Edited By Hopper on 07/09/2016 14:18:15

Just a comment on "twisting the bed".

I have never bought into the idea that adjusting the raising blocks on a Myford stand was about twisting the bed but rather the opposite. A newly ground bed when placed on a level surface should be true and turn true. If when clamped down to the stand it doesn't turn true it means the bed is twisted because the blocks don't all lie on the same surface. Adjusting the block to get it turning true again is about leveling the feet so they are on the same surface and therefor no stress is put into the bed. No stress = no twist and you are back at the condition when the bed was on the surface grinder. In fact when I moved workshops about 7 years ago I set my Myford up on the stand but had not bolted it down. Some weeks or even months later after forgetting about the bolting down and having used the lathe for some time turning as true as I was happy with I realised my 'ommission'.

Tightening the hold down bolts threw it out until I had properly adjusted the raising blocks and I was back to a non stressed condition.

I'm sure someone will dis-agree but there you go.

Martin

I don't Martin. I have used large lathes on very heavy cast iron stands that don't even have cupboard in them. The basic idea was that the lathe wouldn't distort due to bolting down and the stand could just be stuck to the floor with the usual pads and didn't need a dead level floor. That was a long bed DSG toolroom lathe. There were others much the same.

Personally I make up jacks, adjust using my fingers and also only tighten down finger tight. As I thought Myfords were a bit flimsy I used rubber washers under those. After all the lathe isn't going to go anywhere of it's own accord the biggest risk with some is that they may tip over.

I thought my Boxford didn't need the rubber washers, just carefully adjusted nut and bolt made up jacks and finger tight nuts and washers to hold it down. It been like that for a long time now. I suppose I could use nylocs and feeler gauges but the arrangement has been fine on the lathes I have owned. The mini lathe I currently own is mounted on strips of 16mm mdf as the bed screws in from underneath the feet. That way I can fix from the top. Actually it's just sitting on them. And should add they are more like pads than strips.

John

-

Edited By Ajohnw on 07/09/2016 14:49:13

You are correct when you say they are flimsy compared to 'big jobs' although proper bolting down on a good stand does add a certain amount of rigidity However usually if you want to take a last whisker off all you have to do is lean on the tailstock when you take the same cut as the last time. For it's size it's phenominal what you can do on a Myford and they are certainly 'flexible' in what they can do as well. (Pun recognised)

In reality everything flexes to some degree it just depends on what you want to achieve. I am sure the sub millionths of an inch boys take great care in designing not just the machines they use but the structure of the buildings they are housed in. Our electron microscopes are vibationally isolated from the rest of the building and the microscopists still come in at 3 in the morning to get the best images because everything is so still. But then they are achieving near atomic resolution.

Recently I have been revisiting the late George Thomas's writings with great profit and I note his comments on lathes that turn taper rather than true. It matters a lot less when you know exactly what it does do, you can always compensate for shortcomings if you know they are there.

regards Martin

The easiest answer sometimes is to set the compound slide. It can take a while and a dti to see just how much it is moved is handy. Some engineers at a place I worked at had to do this on a pretty large lathe even to account for bearing wear but that needs a very even feed. When they moved to new premisses and had a new lathe they laughed there socks of because one of the directors had it for home use thinking it was all ok.

I know of someone who mounted their lathe on a vary large chunk of polished granite. It did so I hear improve finish. I thought I would give MDF a go but haven't finished yet. There are lots of things about that are dead vibration wise. I might try corian at some point and have wondered about really thick slabs.

John

-

I must bolt my lathe down some time...

N.

Slightly O/T but my first lathe was a Grayson, cracking little lathe, had powered cross feed, which was a nice feature for a 3.5" lathe. Accurate enough for my needs at the time (car building) if a touch small, and a great starter machine to learn on.

Hope you manage to get it within acceptable tolerances.

Wow thank you to all contributors, so much useful information. The bed has been bolted onto two heavy steel rails which in turn have bene bolted firmly but not excessively onto several layers of MDF and chipboard (like a laminate). The bed itself seems to be stress free and very level.

The nose spindle (2mt) has very little run out (0.02mm) according to the dti, must admit that I should be using thous as that is what I was raised on. Not sure how true the internal 2mt is as I have not tested or used that yet.

The chuck backplate also has very little run out so I am guessing that either the chuck body is not 100% true. I am going to try remounting it and checking it using each of the three positions available.

Once again, thank you so much for the help.

If you turn a few trial pieces and find it is turning tapered, you might like to loosen off the mounting bolts, try sliding feeler gauges under each foot and then add shims if any gaps are found.

Another thing that I think has not been mentioned so far is that before you try to twist the bid to set up alignment etc etc etc, it's helpful to make sure first that the spindle bearings are in good shape and correctly adjusted. Not sure if the Grayson has split bearings like a Myfrod or tapered bronze bushings like a Drummond, but either way loose bearings can cause tapered turning and should be eliminated before any other alignment work is undertaken.

The bed is probbly a bit thin, like an Ml4, a 3.5 inch M series Drummond for example is about 107mm across the ways, there's a lot of metal in it.

However, if you can use your steel rails to produce a box section, like a pultra

then you can at least stiffen the bed part of the equation.

The cross slide will still be lacking but it's a start, and as others have mentioned, with practice, the unit can perform

The Pultra tray is cast iron probably because in it's day that was a very cheap way of producing it.

I've heard from people who are no longer with us that certain grades of cast iron are more suitable for lathe beds all down to how well they transmit vibration. More recently when Taiwan was the source of small model engineering lathes some complained about the grade of cast iron that was used. What I suspect they really meant was that higher tensile stuff was being used and lower with more mass would be better. After I bought my horizontal band saw I was inclined to agree even just on the basis of the noise it makes.

The Taig in some ways is one of the best lathes I have used. When I had it there was no problem obtaining a very high quality finish and very round work. The bed on those is an aluminium channel with some sort of concrete slurry in it. Several design problems and they loose accuracy over time but in terms of initial tolerance and finish nearly as good as anything I have ever used.

John

-