Hi Jan

Can we first get this clear, after turning the test piece it was smallest at the free end of the bar, that is farthest from the chuck holding it.

I must say though that 250mm projecting from the chuck at 20mm diameter is too much. I would suggest no more than 100mm with a 25mm diameter bar would be better.

You do not say on what the lathe is mounted as whilst the packing is 3mm thick much of the twisting will probably be in the mounting surface. Twisting the bed by that amount would seem very extreme.

Turning a test piece of 250mm long by 20mm diameter one would expect some deflection so the actual situation will be worse than the amount you have measured. Do carry out the test with a shorter larger diameter bar as above.

Harold

Your Myford cabinet will certainly not be a strong as your lathe Jan so the thickness of any packing you place under one foot will mostly be accommodated by the cabinet twisting. So if you put 3mm packing under one foot do not worry that you will be twisting the lathe by that amount.

My Myford cabinet (not industrial) has an angle iron frame below the drip tray, I would imagine the industrial stand does also. As you have a super seven on an industrial stand, do you not have jacking screws at each foot. See my website here for an example Photograph 1

I Think you are saying that the lathe is level at the tailstock but requires a piece of packing, 0.25mm, under the rear of the level to make it read level at the chuck end, that is far too much. It is getting late so I may be getting it wrong, but I think that error is at odds with the error you are getting in the test piece. I will think more about it in the morning.

I know I will be shot down for saying this but being level is not vital, not being twisted is the important requirement, if you needed 0.25mm at both ends that would be almost as good. In any case, a very small twist can sometimes be needed to offset manufacturing errors in the lathe itself.

Harold

I will second that.

As long as the bed twist is under control, being level is not that important.

I have mine set slightly high at the tailstock end, the reason being that the coolant drains from the tray a lot better.

Phil

Unfortunately, Martin, 99 percent of lathe owners, almost certainly more, don't know that, ask a builder who has a lathe. I do agree though that if an engineers level is available It is not only the simplest method, it is by far the quickest, especially if the lathe has jacking screws for its mounting. But even then a test piece must be turned so as final adjustments can be made if necessary, so that the lathe turns parallel. Also, engineers levels are very expensive in most peoples estimation, OK, they can be had for around £60 so I am told. Even so, many have purchased a machine and then found they do not have the cash for the essential extras, let alone a spirit level. I know, I've been there, which is why I got so much into making workshop tools.

Now to Jan's problem. As there would seem to be some conflicting facts in the data we have, in that if the bed is lower at the rear at the chuck end, especially by the amount given, then the test piece should be larger at the free end, not smaller.

I would suggest therefore Jan, that you forget the spirit level at this stage and make adjustments purely on the basis of the result of turning a test piece.

Harold

Hi Jan, have you got a copy of the S7 manual? if not it is a good idea to get one and well worth the cost! I have set my S7 several times using the cut and try method described in the Myford manual with no problem. don't worry about how much shim you need as the stand etc is only a welded box and cannot be expected to be a reference surface. When the lathe is cutting parallel the lathe bed isn't twisted. Obviously a used lathe will be worn to a certain extent so you will have to get the best out of it you can.

Tony

I can state that it is possible to align a Myford lathe (made in a country where they drive on the left side of the road) accurately in the Netherlands.

A slant bed lathe, like the Dutch made Ergonomic, proves the point that as long as the lathe bed is not twisted it will turn parallel, however it is a curious experience when you are standing behind it for the first time.

Niko.

Hi Jan/Gents

For info to those whom may be new to the sport of "measuring twice and cutting once"

I've just refurbed a ML7 (1954) once rebuilt I found a similar thing, when I checked the center alighnment from Chuck Spindle to Tailstock Spindle, these where out of line and hence the tapering, I corrected this with the use of two small capacity MT2 tapered chucks (0-10mm opening) one inserted in each spindle both with a small sewing needle in, brought the two points almost together and adjusted the tailstock adjustment screws to compensate which solved the mis-alignment cross the bed, the then problem I encountered was height needle to needle which appeared to be 0.009" low at the tailstock, I parted and shimmed the front edge of the tailstock between the foot plate section and spindle casting with two 4mm wide strips cut from an old beer can which 0.0045" thick across the whole width of the tailstock, I then rechecked ( sherlockholmes magnifying glass and a small torch) with the needles by moving the tailstock backwards at intervals until tailstock was fully extended, at fully extended the needles just started to glance tailstock downwards (don't know how to quantify this deflection let alone measure it.

Next I re-intalled the chuck and with a piece of 5/8" barstock 18" long (already centered at both ends) fitted between chuck and tailstock dead center. I layout blued the bar down the surface nearest me, traveresed the saddle to the chuck end, with a ready centered cutting tool (on this occasion a radius ended one), I advanced the cross slide in until the tool tip barely touched the bar I then manualy rotated the bar backward and adjusted the tool in until it scuffed the bar surface with the saddle being engaged to the leadscrew, I then manualy turned the lead screw hand wheel sliding the saddle and in turn the cutting tool along the surface of the bar, which I felt at the points of bed mis-alignment the tool would attempt to deflect the bar or score deeply into it and at these points the distortion could be marked for accurate measurement, fortunately none of these deflections or scoring occured, when I reached the tailstock end of the bar I manualy turned the chuck backwards to see a similar scuff to the one at the chuck end and was satisfied that all was ok.

It's worth noting that I had initialy when re-fitting the head stock squared this by fitting the 7" back plate and using a set square verticaly up the plate to the bed surface and across the bedway to the plate, Also I have not yet required a spirit level nor wasted a bit of bar stock (which may come later when I plug it in).

I hope this will help with setting up without the need for expensive kit and complex three dimesional Trig

Bazza

Hi Gents

Correct me If I'm wrong, but If you re-grind the surfaces of the bed dose that then not require all the associated equipment Travelling on the bed to be then re-aligned, new gib strips fitted etc as the new lesser bed wil undoutedly make al the associated equipment looser in contact of the beds surfaces.

and Also am I right in that if you require to make a deliberate taper, then you set over the tailstock center nearer to the cutting tool for a taper reduction to the tailstock and conversly set over the tailstock away from the cutting tool to taper reduction at the headstock end using the tailstock alignment screw and the gradutated tailstock plate.

Regards

Bazza

I tend to cringe when I think of what can happen when cutting a (long) length of a small diameter bar which is sticking out of the chuck unsupported.

Therefore, I think it’s worth noting that for a round bar protruding from the lathe chuck and unsupported at the tailstock end (ie. cantilevered), it’s resistance to bending is related to the fourth power of its diameter and the third power of its protruding length.

In other words, for the same applied (cutting) load, a 15mm diameter bar will deflect (bend) 16 times more than a 30mm diameter bar.

Likewise, a bar protruding 200mm from the chuck will deflect (bend) 8 times more than a 100mm bar when subjected to the same load.

So keep this in mind whether you are running tests or making something more important.

Regards to all,

Sam

Edited By Sam Stones on 10/11/2012 22:20:39

The bottom line is you want your lathe to turn parallel? If so follow the Myford manual to achieve the desired result, it works so why re invent the wheel? If the bed is worn you will need to decide whether to get it re ground or live with the inaccuracies. Your work will follow the machines wear patterns.

Tony

Gray,

At the risk of inflaming the argument, I think that the ship analogy is to demonstrate your point that the lathe can be aligned yet does not have to be level with the earth. The point that seems to be difficult to get across is that when a spirit level is used across the shears of a lathe, the bubble does not have to be in the middle but it does have to be in the same position at both ends of the bed and for this you really need a precision level with graduations on the vial.

I think quite a few very good locos have been made on shipboard lathes.

regards,

Rod