Are my Lathe Chuck Jaws worn?

[M@tt]

So i've been having a bash at getting my lathe set up properly.

I still need to get myself a length of precision ground bar (unfortunately I didn't have the old printer i thought i did in the loft to smash apart for a suitable length so i'll have to visit Cromwells at the weekend for a section instead)

However in my attempt to try and do something productive i found a length of solid highly polished bar that was actually part of an old gearbox selector mechanism, which looks on inspection that it might be up to the job

So i clamped that in the jaws of the lathe and got the dti on it as close to the jaws teeth as possible. However i found that i was getting a 12 thou swing in readings as i rotated the jaw.



i then put the DTI on the edge of the chuck itself and found i got almost zero movement as i rotate it.




So i guess its one of 2 things
1) the bar i used is inaccurate
or
2) the the jaws are worn and not holding the piece centrally
or
3) both of the above

once i get a known accurate bar for measuring it should hopefully narrow down the source of the inaccuracy

i'm guessing a 12 thou variance so close to the head isn't very good?

I fi go on the assumption that the bar is ok and when i test again with the precision piece and i find the same variance, what are my options?

thanks

Matt

 

[brightspark]

So i've been having a bash at getting my lathe set up properly.

I still need to get myself a length of precision ground bar (unfortunately I didn't have the old printer i thought i did in the loft to smash apart for a suitable length so i'll have to visit Cromwells at the weekend for a section instead)

However in my attempt to try and do something productive i found a length of solid highly polished bar that was actually part of an old gearbox selector mechanism, which looks on inspection that it might be up to the job

So i clamped that in the jaws of the lathe and got the dti on it as close to the jaws teeth as possible. However i found that i was getting a 12 thou swing in readings as i rotated the jaw.



i then put the DTI on the edge of the chuck itself and found i got almost zero movement as i rotate it.




So i guess its one of 2 things
1) the bar i used is inaccurate
or
2) the the jaws are worn and not holding the piece centrally
or
3) both of the above

once i get a known accurate bar for measuring it should hopefully narrow down the source of the inaccuracy

i'm guessing a 12 thou variance so close to the head isn't very good?

I fi go on the assumption that the bar is ok and when i test again with the precision piece and i find the same variance, what are my options?

thanks

Matt

thats a 3jaw chuck so its not going to be acurate put it in a 4jaw then clock it up with a dti and then it will be . make a test bar up and turn between centers

 

[shenion]

A decent 3-jaw will have about 4-6 thou runout.

Mine had about 10 thou when i got it (70-years young.) I reground the jaws and now I get the 2-6 as i would expect.

The scroll could be worn, or one jaw knocked out.

New jaws may help if you can find them. But a good part of the error may be in the scroll.

A way to grind the jaws is shown here. Here is where I did mine using a dremel in my QCTP:

 

[Rudi McAnichal]

Are the jaws in the correct slots? They are usually marked with dots or numbered 1-2-3.

 

[langy]

Are the jaws in the correct slots? They are usually marked with dots or numbered 1-2-3.

I was about to say the very same thing !!!

 

[brightspark]

I was about to say the very same thing !!!

if u got them mixed up they would be over an 8th of an inch out on a self centering 3jaw chuck .it doesnt matter what order on a 4jaw independet chuck

 

[M@tt]

Are the jaws in the correct slots? They are usually marked with dots or numbered 1-2-3.

Yep jaws are in correct slots, i double checked that

I might have a bash grinding the jaws then once i've ruled out the accuracy of the test piece

and i'll keep an eye on ebay for a 4 jaw chuck by the sounds of it. Any recommendations? RDG or Chronos seem to be popular choices but are there any particular brands to go for/avoid?

thanks again for all the advice

 

[brightspark]

pratt burnerd are the best. if not a cheaper one that is good is made by tos whatever u buy u may have to get a backplate to suit your lathe and machine it to suit the chuck . or purchase ready to fit . rdg chronos supply

 

[Shox Dr]

No point spending big money on a 4jaw independent. As it's you that controls the alinement

 

[citizen kane]

You are clocking the back plate not the chuck, the back plate should have been turned true and its next to impossible to mess this up. If the register in the back plate for the chuck has been turned undersize the chuck may not be mounted coaxially. Move the clock onto the chuck body and check again, also dont forget to clean the scroll plate and jaws to make sure you have no swaf jammed in there and check again.

 

[M@tt]

cheers for the pointers gents

i picked up a 22mm precision ground bar at the weekend from www.cromwell.co.uk and checked it again. It's still 12 thou out right at the chuck so looks like my original bar was true and the 3 jaws are out by that much

i called into a tool shop in Attercliffe in Sheffield (MA Tooling -it's like an Alladins cave of new and used engineering tools!!) and they had a 6.5" Pratt Burnerd 4 jaw chuck for £40 but i think it will be too big for my 5" raglan lathe

am i correct in thinking that because i'm getting such out of true with the jaw it means that aligning the tailstock is going to be problematic?

I followed the procedure whereby i mounted my DTI gauge in a block of wood held in the chuck and swept the surface of the tailstock and adjusted as necessary but i was getting about 12 thou difference again (which would make sense wouldn't it?)

when i tried drilling a hole in the block of wood, i could visibly see the drill bit wiggling side to side, indicating that its out

So am i doing this right? What would the next step be to try and get some accuracy into it?
regrind the 3 jaw chuck jaws
fit 4 jaw chuck?

 

[shenion]

If you mean by 5" lathe it has a 10" swing, a 6.5" chuck will work fine. It is a good size. I looked up Raglan and it is 10.5" over ways.

I'd grind the 3-jaw. Only hard part is finding a way to mount a dremel or die grinder.

You really want both a 3-jaw and 4-jaw. 3-jaw for quick work where you can stand a few thousanths off and a 4-jaw for accurate or holding odd-shaped parts.

 

[M@tt]

Cheers Scott

I guess the trickiest part of getting the 4 jaw would be getting a suitable back plate for it? the one on the 4 jaw chuck at the moment is too big for the raglan (i measured the threaded centre hole diameter) and the existing one for my 5" 3 jaw chuck will be too small for the 6.5" one.

Are backplates a pretty common thread/size or are they all different depending on manufacturer?

 

[shenion]

Just get a plain backplate to match your spindle (I think it is 1.75" x 6 t.p.i according to lathes.co.uk).

Then you face the backplate, that helps get rid of any spindle error. You turn a register to match that of the chuck (the chuck will have a recessed area, probably about 5").

The only hard part is drilling the holes for the mounting screws.You can make them a bit oversize so they don't have to be exact.

The procedure is described here.

 

[M@tt]

brilliant cheers Scott!

 

[M@tt]

I called into the Tool place today with the intention of buying the 4 jaw chuck but then i spotted it was missing 1 of the jaw grub screws Does anyone know if these are available seperately or are they machined specifically for that chuck/jaw?

 

[hareng]

Interchangeable or reversable jaws?
If so conventional off the shelf HT ok.

 

[Wedg1e]

Matt, I'm with Citizen Kane... it looks as though you clocked the chuck's backplate. It's possible that the chuck's running eccentric on the backplate. Locate the screws that hold the chuck to the backplate, slacken them off then nip back up so it's just possible to move the chuck around on the backplate with a good thump. Then put your bar in, set up the DTI, note where the runout is and thump the chuck in the opposite direction. Finally tighten the screws.
It may work, it may not, but you have nothing to lose but 5 minutes of your life

 

[M@tt]

I re did the clocking on the chuck itself and bar it hitting a few surface pot marks it was clocking as true so definitely indicates the jaws are out

I made up some jaw tensions ready to try attempting to grind the jaws with my dremel as Shenion has illustrated but the only bit of stock i had availble in my stock was too wide and it meant they were holding the jaws open too high into the chuck, so i need to get my hands on a 20mm wide piece which will allow the jaws to protrude out far enough to grind

I'll try and get my hands on a suitable piece of steel to make some new shims and have another go

 

[MrFluffy]

If you make a plate like this from flat sheet about 5mm thick, you can grind right through the jaws in one pass. Lots of back and forth passes until it sparks out with tiny cuts because the bearings in the dremmel aren't the best compared to a toolpost grinder etc.

If after that the jaws are still untrue on your test bar, the error could be in the scroll inside, as grinding the jaws only puts them exactly right at the grinding diameter.