Hi Dave,
You do not say if the chuck was supplied ready to fit the lathe or if you machined a backplate to fit the chuck to your lathe. I would remove the jaws and check if the front face of the chuck is running true. You could also swap over the suspect pair of jaws (WITHOUT ROTATING THE CHUCK.) and note if the direction of the runout changes or remains the same. If it changes then I think the jaws are at fault in which case you could machine the jaws. To do this hold a ring of a suitable size from the inside, true it up and then skim the inside of the jaws. I would wait for some more replies before trying this suggestion.

Les.

I have a Harrison M300 and an 8" 4-jaw chuck that came with it. I can't see a makers mark on the chuck, but there is a small label in a circular indent on the face of chuck which has a stylised letter 'h' on it. I assume that means the chuck was supplied by Harrison?

I've measured the runout of the chuck body on the outer edge of the front face of the chuck, TIR is 0.05mm. That agrees roughly with the variation in thickness I would expect when 'snugging' a part against the body of the chuck. If I face off a part that is 'snugged' up against a step on the jaws I usually get a variation in thickness that is a little worse. On a 4.4" OD gear blank that I machined last week using the 4-jaw chuck the thickness variation is 0.04mm, so I must have been lucky that day.

I rough machined the crankshaft bearings for my traction engines using the 4-jaw chuck. Never again; the finish might have been good but they were not quite square or parallel. If I want a part square and parallel I now use the milling machine.

Regards,

Andrew

.

.

So when my Ipad goes tits up in 5 1/2 years , Apple will give me a new one.?

That's brilliant.

But surely they expect it to light up when you switch on ? if it doesn't do i get a new one?

6 years dowen the line EVER iPad will still be working ? If not Apple with replace it.?

Pigs, theory of flight, comes to mind.

So take this chuck then.

Has the seller actually said what it does. It's a chuck after all.

I have a 4 jaw on my welding positioner, I expect it to be accurate to plus or minus the odd post code and that plenty good enough.

Hi Dave C ,

You should bear in mind that ordinary grade self centering and independent chucks are only really very crude workholding devices . To put work randomly into such chucks and expect work to be set true by default is an unreasonable expectation .

In case of a four jaw independent which does not have any particular faults a nominal cube for instance can be out of position relative to the jaws by several degrees in rotation and up to a couple of degrees in tilt in two axes .

To set up a nominal cube properly requires positioning best guess in jaws and then clocking reference faces true in one , two or just sometimes three axes and progressively making small adjustments and tightening jaws at the same time .

Further notes :

Jaws are never going to be exactly square to chuck – might be somewhere near on a new and good chuck – but in general not .

The work will generally move as the jaws are tightened . You can keep track of movement and compensate while setting up as above but it helps to have some restraint where it can be arranged . One useful thing to do is to use the tailstock to prevent the work moving forward .

In many ways setting work in a four jaw independent chucK is the same as setting work in a milling vice – and its quite common practice to tap the work home as vice jaws are tightened .

For any particular chuck learn its defects – if you know that for instance that work is always cocked away from jaw no 1 then you can always bias the initial setting up of work to compensate .

Actuall knowing the defects of any particular machining set up is always useful and sometimes allows quite astonishingly accurate work to be done on dodgy equipment .

Regards ,

MikeW

Edited By MICHAEL WILLIAMS on 05/07/2013 13:42:43

> I am holding the parts far back as near to the chuck body as possible

If you can't actually snug the parts up against the chuck body or the steps on the jaws, all bets are off – you can't blame the chuck.

Neil

An interesting point. The component in question is an axle box, and my experience does suggest that it's very different trying to grip a square (or round) component with only the outer part of the jaws to gripping a bar threaded right through the chuck. When I have to turn such a short part (in 3 or 4 jaw) I always insert a temporary parallel between it an the chuck face until the chuck is tightened up.

Neil

It's common practice when setting a round bar to run true in a four jaw independent chuck to clock the bar in two places – one very near the jaws and one as far as possible outboard .

If you want consistently true workholding without fiddling then you have to have very good chucks or start in a different place and use collets or things like V angle plates .

MikeW

Dave

PM sent.