Chuck jaws off axis – what to check

[Paul FParticipant @paulf66363]

I have recently purchased a new 4" 3 jaw chuck and installed it onto my Clarke lathe (which I have recently purchased second hand)

The jaws appear to be off axis with the lathe spindle – when I install a long bar of solid, at 150mm away from the chuck I have 1.5mm of run out, the run out reducing as I bring the DTI closer to the chuck.

This is what I have checked so far – any ideas what further steps I should check

I have checked both sets of jaws with no improvement.

If I rotate the bar in the chuck, the runout stays in alignment with the chuck.

The chuck appears to have minimal runout on its circumference

The face plate has has virtually no run out on either the circumference or the bolt mounting flange. I have deburred the bolt mounting holes.

Rotating the chuck to a different mounting flange does not correct the issue (I will check later to see whether the position of worst runout moves with the chuck or stays with the faceplate).

Any ideas would be gratefully received

[Paul FParticipant @paulf66363]

[John HaineParticipant @johnhaine32865]

It's useless, send it back, don't even try to correct it!

[JasonBModerator @jasonb]

Have you actually machined the backplate on the lathe or just bolted the chuck straight onto it?

J

Edited By JasonB on 28/09/2014 13:19:31

[John ShepherdParticipant @johnshepherd38883]

As Jason says first check the jaws are in the right order.

Then:

1. Check the test bar, it is unlikely that a piece of MS bar is round or straight. If you have any ground bar or silver steel that will be a better bet.

2. With a decent test bar check for run out near to the jaws.

3 . You say the outside diameter of the chuck and it's backplate are true also check the face of both backplate and chuck.

4. Make sure the backplate is clean and the mating recess in the chuck is clean and free from burrs and that they are the same size.

5. What was on the lathe before, did that run true?

6. Turn a short length of bar and measure the diameter at both ends and check they are both the same.

Hope that helps to decide what is wrong

[Paul FParticipant @paulf66363]

Thank you for the advice.

I haven’t machined the backplate – but as neither the face nor the circumference appear to have any runout then I didn’t think it was necessary ( and didn’t feel confident in doing that yet)

The jaws are definitely in the right order – and I have tried both the internal and external ones.

I have now checked the face of the backplate and the chuck and these are both true.

I had a ground bar and so tried that – still the same amount of runout.

I put a pen mark on the chuck and the back plate at the point of greatest runout. I then unbolted the chuck and refitted in an alternative position. The point of greatest runout was aligned with the pen mark on the chuck.

I think the only thing left to check is for burrs on the back of the chuck

[Ian PParticipant @ianp]

Is the 'minimal runout' on the body of the chuck the same value at both end of the chuck body? In other words does the axis of the chuck show the same symptoms as your test bar.

If the chuck body is running true than I would check the jaws one at a time just by holding them against an engineers square. Depending on the quality of manufacture you may need to stone off any tiny burrs where the scroll castellations are. The amount of error you observed sound as if its caused by the jaws not being square.

If the jaw edges are not square and you lightly grip a well finished (ground or polished) test piece you will be able to feel whether the jaws are gripping equally along their length. I have refurbished chuck jaws that did not grip concentrically and/or squarely just using an oilstone (diamond type).

Ian P

[Ian PParticipant @ianp]

John

0.0002" runout is a very commendable result but I wonder how much it is due to luck?

I'm not decrying the method, I've done similar things myself and have no fear of modifying something so that it is no longer 'original'.

Your method relies on the chuck jaws gripping your mandrel accurately in the first place.

With a self centering chuck the one part that ultimately determines the accuracy is the scroll. I dont know the correct term but the scroll should revolve around its 'centre' and it must be a good fit on its journal. That journal is the thing that needs to be concentric with the lathe mandrel. Presumably when the chuck is manufactured the scroll journal will be machined at the same setting as many other features of the chuck body (most likely including the chuck OD.)

The chuck body itself is mainly a carrier for the scroll and the three jaws, it has to hold them 120 degrees apart and guide them at 90 degrees to the lathe axis. The precise location of the pinions and runout on external parts of the body do not affect accuracy, although useful for set up purposes.

Ian P

[blowlampParticipant @blowlamp]

I've used cheap diamond files to correct the jaws of a couple of my lathe chucks, the great advantage is being able to control where and how much material is removed.

It's quite easy to 'pull' things to get the best result, just by nipping a bit off here and there, until it's as accurate as you want it.

Martin.

[Paul FParticipant @paulf66363]

I contacted the supplier this morning via email. They have already dispatched me with a replacement – so I guess they have had some other faulty ones – I have had it 6 weeks but only just got round to mounting it.

I will let you know how I get on with the new one.

[Ian PParticipant @ianp]

John

I should have expressed better what I was trying to say. The OP mentioned that his runout increased as he measured further away from the chuck which seemed to point to the the chuck (jaw) axis being tilted. If that chuck was trued using the method you used then it would then grip accurately with that one set of jaws. If the internal gripping jaws were then fitted its likely they would grip the job tilted.

Ian P

[Paul FParticipant @paulf66363]

The new chuck arrived on Tuesday but I have only just had the chance to try it. This one has a build date of a few months later than the first and appears to be a lot better quality. This one has 0.08mm run out at 50mm as per the spec.
Thank you all for the advice

[Author]

Posts