How to tighten a 3 jaw?

[Bamban]

What Marcel says actually does work for very small adjustment and wont hurt the chuck cause you ain't smacking it with a 5 pound sledge. You give it a sharp heavy tap. I agree with his method , as i have done that on occasion.

Now on to your question.:

I chuck up a 1" ground bar a few inches long ( a piece of drill rod or turn a 1" bar). I loosen the bolts holding my chuck to the back-plate. (my chuck has three dowel pins in the back-plate locating it. I think the register is bad. God only knows what this thing has been through all these years.) I get my chucked up rod indicated within as close to perfect as possible. 1 or 2 tenths at least. I then tighten the bolts and reamed and re-pinned my back-plate and chuck. Because of parts fit, wear and manufacture accuracy The chuck will generally repeat within a few tenths AT THAT DIAMETER ONLY. Because of the factors I just mentioned, when you chuck up at any other diameter there is going to be error. How much error depends on the condition , wear , and quality of your chuck. I was fortunate that this Union chuck was a very good chuck and the worst run-out I get is .0025", but if I chuck at 1" diameter, I can get it to a few tenths run-out on a turned bar.

Take note: To get the repeat-ability I have to sometimes chuck,and un-chuck a few times, and at any other diameter, I do the same till it gets as close as possible to true. If I just throw a piece in the chuck and tighten it, It may be out .002" to .003". Also you can only get accuracy on a piece that is true round. If you throw a piece of CRS or HRS steel rod in the chuck , you cant expect any accuracy till you turn it. I hope I explained this so it is easy to understand.hew:

Mark,

Thank you.

It makes sense the way you laid it out, I think I can follow the procedure. Got an Asian 3J I can try this on. It has 0.006 run out on the 1 inch test bar I picked up from PTG.

 

[Mark_f]

I dont think it is that simple. As I have heard it explained, the scroll has a small amount of play in it due to various factors (tolerances, mass production etc... ). With this in mind it stands to reason that one of the drive sockets will use this play to give less runout than the other two.

Since this play is not designed in it cannot be determined beforehand which of the drive sockets will give the best runout. Some manufactures then test for the one with best runout and mark that one.

Sent from my GT-I9505 using Tapatalk

My reasoning for using all three sockets is because of the "play" , very tiny amount that it is, is present in the scroll grooves, jaw grooves, and socket grooves. Let's face it, these things cant be a perfect fit ,especially being massed produced. When you tighten one screw , it locks the screw, jaw and scroll taking up all the slack and also at this point is taking all the tension to hold the part. When you tighten the other two screws, you take up any slack they had and also lock the part in the chuck with them , distributing the load to all three screws instead of one, thus also gripping the part tighter. This is MY reasoning for tightening all three sockets. It maybe hogwash , but it makes me feel better, so I do it. :nuts:

 

[darkzero]

Ponder this; If the chuck manufacturers wanted you to use the same pinion all the time, they would save the manufacturing cost of the other two pinions and holes and leave them out.

Not necessarily. From what I've seen chucks with only 1 pinion have less max rated speed than a chuck with 3 pinions (but there are other factors too). Think about the imbalance issue at high RPMs if not dynamically balanced but if your lathe can't hit those speeds it doesn't matter much.

I agree with f350, I also hear that the marked pinion is the master that was used for tightening when the chuck was ground. And whatever pinion you use if you regrind the jaws then becomes your master.

 

[Bamban]

What Marcel says actually does work for very small adjustment and wont hurt the chuck cause you ain't smacking it with a 5 pound sledge. You give it a sharp heavy tap. I agree with his method , as i have done that on occasion.

Now on to your question.:

I chuck up a 1" ground bar a few inches long ( a piece of drill rod or turn a 1" bar). I loosen the bolts holding my chuck to the back-plate. (my chuck has three dowel pins in the back-plate locating it. I think the register is bad. God only knows what this thing has been through all these years.) I get my chucked up rod indicated within as close to perfect as possible. 1 or 2 tenths at least. I then tighten the bolts and reamed and re-pinned my back-plate and chuck. Because of parts fit, wear and manufacture accuracy The chuck will generally repeat within a few tenths AT THAT DIAMETER ONLY. Because of the factors I just mentioned, when you chuck up at any other diameter there is going to be error. How much error depends on the condition , wear , and quality of your chuck. I was fortunate that this Union chuck was a very good chuck and the worst run-out I get is .0025", but if I chuck at 1" diameter, I can get it to a few tenths run-out on a turned bar.

Take note: To get the repeat-ability I have to sometimes chuck,and un-chuck a few times, and at any other diameter, I do the same till it gets as close as possible to true. If I just throw a piece in the chuck and tighten it, It may be out .002" to .003". Also you can only get accuracy on a piece that is true round. If you throw a piece of CRS or HRS steel rod in the chuck , you cant expect any accuracy till you turn it. I hope I explained this so it is easy to understand.hew:

Mark,

My sincerest THANK YOU for sharing your procedure.

I initially thought I would do the procedure on my ChiCom 3J, but changed my mind and did the procedure on Bison 6 inch 3J. I actually did the procedure a couple of days ago, but have not reported on it. Today after I got frustrated with my new to me old Buck set thru 6 inch 6J chuck, I decided to install the adjusted 3J back in the lathe and do the same test I did on the 6J, that is read the run out at the end of test bar and moved the carriage in close to the chuck and read run out. I only have the 0.001 DI, but as you can see the needle is fairly steady.

Thank you, again.

Sorry, boss, this 3J gives me the best repeatability by using a particular pinion, and I tested all of them and any combination and I keep on going back to that single pinion approachView attachment 92007

 

[Rick Leslie]

My lathe chuck is a quality make but very worn, so repeatability is hit or miss. I've never looked at the keys for numbers but I will now. My drill chuck is the same; very worn. I do find tightening all three key holes makes a difference in holding force with it. I would never have thought of such but read it (I believe on here) and it works. (For those of us too frugal to buy new and/or quality stuff.)

 

[Round in circles]

There is of course the real big elephant in the workshop that most apprenticed machinists will know and follow if they are worth their salt .
It is:-
Use a three jaw to turn to some where near
use a four jaw to turn something reasonably accurate
Turn centre drilled items between centres using a dog & face plate if you want precision & repeatability .

This is why the accuracy of a three jaw & its repeatability is not so much of a problem if you have a face plate and dogs or you can make the driving dogs.

 

[gregg-k]

I read somewhere that the marked socket was the one used at the factory when the jaws were ground. Its supposed to give the least runout. True or not I can't say and have no recollection where I read that.

Greg

I agree with Greg ... that's exactly what I learned as well ... over 50 years ago, and I've always done it that way.

My 8-1/4" Buck 6-jaw Adjust Tru chuck has one pinion socket to tighten the jaws, so there is no choice but to use that one.

.. Gregg

 

[gregg-k]

Ponder this; If the chuck manufacturers wanted you to use the same pinion all the time, they would save the manufacturing cost of the other two pinions and holes and leave them out.

Good point, and yes, and my Buck 6-jaw chuck is made exactly that way ... it only has one pinion.

.. Gregg

 

[Round in circles]

Good point, and yes, and my Buck 6-jaw chuck is made exactly that way ... it only has one pinion.

.. Gregg

Is the jaw operating device the same as a bog standard chuck back plate ?

 

[hermetic]

Anyone having any doubts that the marked pinion is the one to use for most accuracy and repeatability should read through this catalogue of Pratt-burnerd chucks. It is known as the "nominated pinion. It is not an index mark for backplate alignment as when you buy a brand new chuck, you don't usually get a backplate with it! Doesn't stop anyone fitting a chuck to a backplate using a 0 to also align the backplate to the chuck.

Phil.