How much out of the chuck?

I want to mount a piece of 3/4" round bar to take a test cut. Is there a rule of thumb that says what length it is safe to have sticking out of the chuck unsupported?

And should I use the 3 jaw or the independent 4 jaw chuck?

.

What is the purpose of the test cut. i.e. what information do you hope the glean from the cut.?

Nick

I was always lead to believe that 3x the diameter was ok, but I could be mislead.

Gordon.

Tubal Cain in his book Workholding in the Lathe (WSP15) suggests using 19mm to 25mm diameter, well ok, 3/4in to 1in, in a three-jaw self-centring chuck with about 180mm (7in) sticking out unsupported, and 125mm (5in) inside the chuck jaws and the hollow mandrel. This is when checking/setting up a lathe. He says that you can use the 4-jaw if you wish.

Using a 4-jaw independant chuck will require you to use a "clock" or equivalent to get the bar more or less central but does allow the security of the extra jaw.

Hope this helps,

Peter G. Shaw

Hi Nige I would ask you to do three things.

1 put a piece in to the three jaw chuck holding it with the full width of the chuck and having it sticking it about 1 1/2 inches. With a nice sharp tool, try various cuts start with a small fine cut and measure the diameter at different places along its length and see if it is parallel. write down the results. Then take a larger cut say .05 and repeat the results.

2 try the same procedure with a piece sticking out about 3 inches.

3 Take an 8 inch piece stick it as far back into the chuck as far as it will go, face the end and centre drill the end with a No 3 or 4 centre drill. Pull out the bar so you are holding it with about 1/2 an inch, support the end with a revolving centre and lock up the tailstock. Repeat the same cuts as before making sure that you clear the side of the tailstock when you wind in with cross slide and that the bar isn't pushed into the chuck losing the support of the centre.

reduce any cuts that seem to go badly and record and learn if you can cut parallel and smooth etc.

David

I was told the same thing.

Michael W

I presume that this follows on from your previous thread and relates to headstock alignment. If so you don't need the chuck to be accurate nor do you need to take test cuts. All you need is a truly parallel test bar such as a length of half inch silver steel in the chuck and a DTI in the tool post. With the DTI at centre height clock the test bar near the chuck, rotating it by hand and note the highest and lowest readings. Take the average of the two readings. Without touching the cross-slide handle move to the far end of the bar. Repeat the measurement there. The difference between the two readings will give you the spindle alignment error over the distance between the two measurement points.

Taking test cuts will deflect the bar so needs to be done very carefully with a sharp tool and very fine cuts. Don't, as someone suggested use the tailstock for support as it defeats the object of the test.

Russell

Hi Russell The object is to see the results of holding bars differently and getting used to a lathe and what results you get for this particular lathe and it's sturdiness.

David

"3 Take an 8 inch piece stick it as far back into the chuck as far as it will go, face the end and centre drill the end with a No 3 or 4 centre drill. Pull out the bar so you are holding it with about 1/2 an inch, support the end with a revolving centre and lock up the tailstock."

I did something similar to this when I was learning to use my lathe. After facing and centre drilling a round bar, I loosened the three jaw chuck and pulled the bar out three or four inches. I then supported the end of the bar with a revolving centre. Unfortunately the drilled centre in the bar was no longer running true, and I ruined a nice little revolving centre that came with the lathe. that was an expensive mistake, I don't forget those. If using this method for testing make sure the drilled centre is running true before engaging the running centre.

Nick_G: This advice, which I have had more than once, is what I am doing.

" Why not put a piece of 1" bar in the chuck and take a light test cut along the length of it, say about 4 inches? Measuring the diameter at each end will give you a pretty accurate indication of spindle to bed alignment."

I realise it is yet another method of checking alignment and to be honest I don mind using a multiple of methods as at my level of ignorance every one of them has something to teach me

I need to align my head stock with the lathe bed which is hopefully coincident with the axis of the carriage travel. In doing this I need to test alignment. I can do that before and after the head alignment and both will give me info on how things are set up. Whether or not I am able to make best use of that info is another question.

Thank you everybody for the advice, all of which I appreciate, I now have to get my head round it but I think I am there with the way I will proceed

Hi Nige,

I think Russell may be referring to what is known as "Rollie's Dad's method" of spindle alignment. . If you search for this on 'tinternet there are several articles that describe it.

For the parallel test bar I use one of the guides from an old printer as these are manufactured to a tight tolerance.

No metal cutting is involved, but you can always carry out a turning test as a final check.

Gordon.

With a 2 MT Headstock bore, you won't get anything much bigger than 1/2" through the bore.

Presumably, you are about to try checking the bed for twist, in the manner advocated by Myford, for the Series 7 lathes. Which would probably lead you on to shimming under the feet until the bed is free from twist.

But your lathe, being a cantilever bed is probably less sensitive to disparity in the mountings, and you may have difficulty doing anything about any twist that may be detected.

FWIW, get the Tailstock aligned first, (now that you have the Headstock about as good as you are going to get) by checking the Alignment bar between centres.

Howard

Hi Howard: I think the cantilever design of the ML4 bed makes it most unlikely I could effect any twist in the bed, I believe it will be far too stiff. I will get the tailstock lined up tomorrow and then try some test cuts. Fortunately I have a very nice live centre on a MT1 arbor that will fit the tail stock !

The bore on the ML4 is 19/32" so you are right nothing much bigger than 1/2" is going through it and of course modifying the mandrel to take 7 series chucks, back plates etc. wont change that aspect but would make fitting modern chuck tooling, collet chuck etc. much easier. Once I get some actual turning under my belt I will have a much better idea if the ML4 is worth keeping or whether I would be better off selling it and putting the proceeds towards something bigger and/or newer. Have to get on and do some turning

Will I see you at PMES on Monday night Howard?

Not much point in aligning the tailstock until you have finished aligning the headstock. Any movement of the headstock spindle will throw the tailstock alignment out and you will have to do it all over again. Tailstock alignment is a totally separate operation from headstock spindle alignment and is best kept as such.

Myford's ML7 user manual (get a PDF copy free off the net, it's everywhere) suggests a piece of 1" diameter bar from 4" to 6" long. You would want an extra 1.5" or 2.5" or so to grip in the chuck. It is usual to rough turn the centre section of the bar down smaller so your light (.002" ) test cut is made only over the two short sections on the ends, at the same cross slide setting of course.

Don't use smaller, eg 1/2", bar as it will flex too much. And you don't need it to go up inside the spindle, just inside the chuck jaws is all that is needed. .

Edited By Hopper on 06/08/2017 05:06:54

Edited By Hopper on 06/08/2017 05:07:20

Absolutely! You need to get the headstock aligned first. Try my method outlined above it's much easier to do than describe and the test only takes one minute once set up. Much quicker and more accurate than taking test cuts which you could do afterwards if you wish.

Yes, it is sometimes called Rolly's Dad's method but I think I was using it before it was called that!

Russell

Nige: If it were me I'd stop worrying about whether headstock is, or isn't aligned, and just cut metal. There is a huge amount to be learnt about tool grinding, feeds and speeds that are not affected in the slightest by any alignment issues. The converse is true, if you can't get a good finish then it is going to be very difficult to make any useful measurements regarding alignment. It's only too easy to end up chasing your tail.

I spent (wasted) a lot of time early on trying to get my lathe to turn parallel near the headstock. Turns out that the gap piece wasn't installed properly. Darn good job I didn't mess around with the headstock alignment. It is not easy to make measurements that highlight one, and only one, alignment parameter. In 'adjusting' one aspect you may well mess up another one.

Andrew

Hi Nige

From time to time you may need to extend work beyond the normal diameter x 3 to 4 to one, well that's my rule of thumb. For a 19mm bar that should work fine as long as you are taking light cuts.
There should not be much deflection due to tool cutting pressure. there will however be some deflection and the part diameter may increase towards the unsupported end due to the work springing away from the tool.

For any beginners reading this, extending the work excessively can lead to the part bending due to centrifugal forces, There have been some serious accidents where the operator was whipped by the flaying bar.
Polishing a long bar held in the chuck even using the tail stock at high speed is particularly dangerous. If you dare Google lathe accident.

As far as the 3 jaw versus 4 jaw? Like most people I started using the 3 jaw and was annoyed to have to put on the 4 jaw because it was more difficult to set the work true. However the three jaw can never be as accurate as the 4 jaw (independent jaws type) chuck. With practise you can set the work in a 4 jaw within a tenth or two within a minute or so.

Now the only time I use the three jaw is when I have to make a lot of small parts the same from bar stock, the kind that are turned all over then parted off. If I had a collet set I could also use that.

On the other hand if the part was say a ground bar that needed to be accurately turned down to a shoulder It would best be done in the 4 jaw chuck having set it dead true. (and using copper packers to protect the finish on the bar)

Indicators.

I have a couple that sit next to the lathe.
A ten thousanths of an inch .0001" about .0025mm per division the other a "thou" per division .001" about .025mm

For cold rolled bar there is not much point using a tenth indicator, it will jiggle all over the place unless it has a ground or rolled finish. is is also more fragile and best kept for finely finished surfaces. For average work a thou indicator will be fine and you can interpolate between the division lines anyway.

A while back I purchased a new magnetic stand, the type that only has one knob that will tighten it in one go,**LINK**
These are a a great time saver.

Across the Atlantic pond Keith Fenner and Abom78 on you tube both show very clearly how quickly work can be set in the 4 Jaw.

Cheers
John

Edited By John McNamara on 06/08/2017 16:28:59

Russell: I used your method in the end and got mostly sorted as described in my' Mandrel thread', thank you.

Andrew: Now doing exactly that, cutting stuff

John: Thank you, a lot of good advice there which I am following up slowly

Again, thank you everybody