PDA

View Full Version : 4-indep-jaw chuck on rotary table: Centering not critical, right?



Bolster
12-19-2010, 01:22 PM
If you attach a 4-independent-jaw chuck on a rotary table, the centering of the chuck to the table is not critical, right?

I mean, you want to get close, but no need to use a DTI to get the chuck concentric on the rotab is there? Because the four jaws are independent?

moe1942
12-19-2010, 01:24 PM
Not until you rotate the table..:)

Bolster
12-19-2010, 01:30 PM
Hm. OK, how about this as an example.

The rotab is centered (exactly) under the quill, with DTI used on the rotab's center hole to get it exactly concentric. Then put on the 4-jaw-independent chuck. Now assume that chuck goes on sloppy, .1 out of round, let's say. Next I mount the part in the chuck and use the quill-mounted DTI on the part until it's concentric (or, alternately, that I'm picking up a punch mark on the work so that it's centered to the quill...whatever). The PART and the ROTAB are concentric, but the chuck is not...isn't it somewhat irrelevant if the chuck is not concentric, aren't the differential jaws compensating and making a "correct" center, if I am aligning the part concentric with the quill?

I guess what I am saying is, if I am going to "sweat" (1) getting the rotab exactly centered under the quill and (2) getting the part exactly centered under the quill, is it necessary that I also "sweat" getting the 4-jaw independent chuck exactly centered on the rotab? (Obviously if it were a scroll chuck of any sort, it WOULD matter a great deal.)

danlb
12-19-2010, 01:31 PM
Logic says that you will adjust the jaws till the part is centered to the rotab's rotation. So no, the chuck body should not need to be perfectly centered.

Dan

Bolster
12-19-2010, 01:37 PM
Well that's where I'm netting out. But I get a lot of people responding like Moe above, who say that even a 4-jaw independent chuck must also be DTI'd into concentricity.

Arthur.Marks
12-19-2010, 02:45 PM
But I get a lot of people responding like Moe above, who say that even a 4-jaw independent chuck must also be DTI'd into concentricity.

Don't listen to 'em. Better yet---set it up; make your part. Prove them wrong.

gcude
12-19-2010, 03:23 PM
I guess another way to think about it is to look at mounting a 4-jaw chuck to a backplate for a lathe. If you attach a backplate, it doesn't have to be dead on concentric with the chuck, but you probably don't want to have it off by 1/2" and have your lathe walking and dancing around the shop. If the 4 jaws on the lathe are concentric with the part held there, you should be good to go, even if the chuck is doing the whorley-gurley. So, no the centering on the rotab on the mill should not be critical, only the 4 jaws holding the part.

moe1942
12-19-2010, 03:32 PM
Don't listen to 'em. Better yet---set it up; make your part. Prove them wrong.



That's what I would do.

Bolster
12-19-2010, 03:46 PM
The reason this is an issue, is because I expect to move the chuck on and off the rotab frequently. So it would save a lot of fussing if I knew that concentricity of chuck-on-rotab wasn't critical.

And of course, since it's a rotab, I don't have any issues of a heavy part wobbling at high speed. My RPM on a rotab is what, one rotation every three minutes or so.

moe1942
12-19-2010, 04:47 PM
I missed #3 post. That will work but an accurately located three jaw really speeds things up.

Benta
12-19-2010, 05:17 PM
Bolster, the thing is to get your workpiece to run concentrically with the rotab's _true_ axis.

A lot of people have an idea of indicating the hole in the rotab to center it, not knowing whether the hole is actually running true. Bad idea.

Instead, first, mount your workpiece on the rotab/in the chuck and use a dial indicator to get it true by turning the rotab (dial indicator fixed to the mill table) and adjusting the workpiece position in the chuck ("bumping it in").

When that's done, get your workpiece to turn concentrically with the spindle by placing a dial indicator in the spindle, and turning the rotab while moving the mill table until your dial indicator is stable.

Did I mention centering the chuck? No, because it's not necessary!

Benta.

Bolster
12-19-2010, 05:32 PM
I recognize that sequence from Harvey's book (Machine Shop Trade Secrets)! Sounds good to me, thanks.

Fasttrack
12-19-2010, 05:39 PM
Bolster, the thing is to get your workpiece to run concentrically with the rotab's _true_ axis.

A lot of people have an idea of indicating the hole in the rotab to center it, not knowing whether the hole is actually running true. Bad idea.

Instead, first, mount your workpiece on the rotab/in the chuck and use a dial indicator to get it true by turning the rotab (dial indicator fixed to the mill table) and adjusting the workpiece position in the chuck ("bumping it in").

When that's done, get your workpiece to turn concentrically with the spindle by placing a dial indicator in the spindle, and turning the rotab while moving the mill table until your dial indicator is stable.

Did I mention centering the chuck? No, because it's not necessary!

Benta.


Perfect reply. For sloppy work, you can center the rotab by sweeping the center hole but if you need to hold a close tolerance, Benta's got the procedure to do it. :)

I had a few scrap parts before I finally started doing things the way Benta described. ;)

oldtiffie
12-19-2010, 05:44 PM
If you attach a 4-independent-jaw chuck on a rotary table, the centering of the chuck to the table is not critical, right?

I mean, you want to get close, but no need to use a DTI to get the chuck concentric on the rotab is there? Because the four jaws are independent?

There is rarely if ever a need to get a chuck concentric to a rotary table with the (probably) single exception that a job is concentric with the chuck and the job was centred in which case the concentricity of the chuck was a chance by-product of centreing the job on/to the rotary table.

I never use a 4-jaw chuck on a rotary table unless I cannot effectively grip the job in a 3-jaw chuck or a collet.

Not all "centres" (ie morse tapers) are precisely or at all accurately either concentric with the table axis or have no "conical" misalignment either.

Here are the test piece figures for the concentricity at the table and 4" out from it:

6" Vertex rotary table: 0.0008" (0.02mm) at the table and 0.0012" (0.03mm) 4" from the table.

8" Vertex rotary table: 0.0008"

Here are the inspection sheets for my pretty good "Vertex" 6" and 8" rotary tables:

http://i200.photobucket.com/albums/aa294/oldtiffie/Vertex_Rotab_hand-book/Vertex_rotab6_8.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Vertex_Rotab_hand-book/8Vertex_Rotab_Manual11.jpg

There are really mainly (only) two "centreing" situations that pertain to rotary tables:

1). getting the axis of a job and the table to be co-incident/aligned; and

2). getting the co-incident axis of the table and the job to be co-incident with a mill quill axis.

Here is my 3-jaw chuck and my ER-32 collets on my rotary table:

http://i200.photobucket.com/albums/aa294/oldtiffie/HF-45%20Mill%20misc/HF45-5.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/HF-45%20Mill%20misc/HF45-4-1.jpg

I can centre the job to my rotary table and then centre both the job and the rotary table to the mill quill axis to within a Total Indicated Run-out (TIR) of 0.01mm (~0.0004") - or better if required - within a manner of a couple of minutes - and all with just a Test Dial Indicator (TDI) and neither need nor use a "centreing" device/adaptor.

I will cover the methods later.

Bolster
12-19-2010, 07:50 PM
Thanks for the replies...please note I will be inserting non-circular work into my chuck. Seems most replies so far are assuming circular work.

Fasttrack
12-19-2010, 07:52 PM
Thanks for the replies...please note I will be inserting non-circular work into my chuck. Seems most replies so far are assuming circular work.

I did not assume that the work was circular. I did assume that you had something to indicate off of, though. You need a hole or a pin or something on your work piece to be able to indicate whether or not the workpiece is centered correctly.

SVS
12-19-2010, 09:10 PM
[QUOTE=Benta] When that's done, get your workpiece to turn concentrically with the spindle by placing a dial indicator in the spindle, and turning the rotab while moving the mill table until your dial indicator is stable.[/QUOTEec

I think you made a typo....

Need to turn spindle for the second step. Otherwise you are just double checking the first step.

DR
12-19-2010, 09:14 PM
.................................................. .................................................. ..

http://i200.photobucket.com/albums/aa294/oldtiffie/Vertex_Rotab_hand-book/Vertex_rotab6_8.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Vertex_Rotab_hand-book/8Vertex_Rotab_Manual11.jpg

.................................................. .................................................. ....




Anybody else notice something funny about these test reports?

I bought several crates of offshore machine tool parts at the closing auction of a local machine tool dealer. Included were 4 taper attachments for JET lathes. Each attachment had a test report similar to the above. Big surprise, every report was identical right down to the finger smudge in one corner. So much for test reports.

oldtiffie
12-19-2010, 10:15 PM
I don't have any concern about the report as long as it says the rotary table is within those limits.

I have no reason to believe that mine are not.

I'd agree that my "copy" is just that - a generic copy - as is used with every rotary table.

I'd agree too that is would be nice if the specific sheet was indicative of a specific rotary table - ie mine.

But I've checked mine and the "errors" do not exceed the limits and in most cases are within them.

A lot of stuff - and not just "Chinese" either - specifies that the item/s is/are in accordance with "Standard XYZ" etc. and if you want the actual measurements/sizes/limits etc. of that Standard. you have to go hunting and/or pay for it.

Surface plates, straight edges and slip guages are good examples.

DR
12-19-2010, 10:19 PM
What I was referring to is there's no serial number on the report, also no date.

So what good is it? How do you know it even applies to your tools.

oldtiffie
12-19-2010, 10:31 PM
I don't.

My tool number is not on it.

My tool is within the limits and if it wasn't I'd have the rotary table and spec in my hot little hand as I hot-footed it off to my supplier for a new item or a refund - and I'd get it.

But as I said, I haven't had to and don't expect to.

Bolster
12-20-2010, 01:25 AM
Can someone please confirm the following?


[QUOTE=Benta] When that's done, get your workpiece to turn concentrically with the spindle by placing a dial indicator in the spindle, and turning the rotab while moving the mill table until your dial indicator is stable.[/QUOTEec

I think you made a typo....

Need to turn spindle for the second step. Otherwise you are just double checking the first step.

oldtiffie
12-20-2010, 01:45 AM
Its wrong.

I will explain later.

I will explain both the logic and the procedure/s. Then the procedure will seem - and be - logical.

I don't like rote learning.

Ian B
12-20-2010, 02:46 AM
The centring of a 4 jaw on a rotary table isn't critical, but usually nice to get somewhere close. How about making a plug that fits in the centre hole, and also registers in the back of the 4 jaw? This'll locate it quickly, and make the whole setup less likely to slide sideways under cutting loads.

Certainly for a 4 jaw, eccentricity of the centre hole is unlikely to be an issue - or you must have a *really* bad rotary table!

Ian

Benta
12-20-2010, 03:52 AM
Originally Posted by Benta
When that's done, get your workpiece to turn concentrically with the spindle by placing a dial indicator in the spindle, and turning the rotab while moving the mill table until your dial indicator is stable.[/QUOTEec

I think you made a typo....

Need to turn spindle for the second step. Otherwise you are just double checking the first step.

My Bad, it was late last night. It should read "and turning the spindle", not rotab.

All of above assumes that you have some kind of reference (hole, pin, whatever) to indicate to on your workpiece.

Benta.

winchman
12-20-2010, 03:53 AM
I can envision doing some machining on the lathe with the part mounted in a 4-jaw (or a 3-jaw) chuck. When the lathe work is done, you could transfer the chuck (with the part still in it) to the rotab on the mill.

The part is already centered on the chuck, so you'd just dial in the chuck.

Better yet would be to have a mount for the chuck already dialed in on the rotab.

All operations you did with the mill by turning the rotab would be concentric and square to the axis of the part, the chuck, the rotab, and the lathe spindle. You could also switch back to the lathe if necessary.

John Stevenson
12-20-2010, 04:37 AM
At the end of the day you are working on the part, not the chuck.
The 4 jaw could be replaced by a jig plate of a shape that can't be dialed in.

No one says that the part holding device has to be concentric.
I have a 3 jaw chuck on my slotter with about a 2 1/2" diameter hole in it. so if I'm slotting say a 3" hole unless the work is packed up, which reduces the holding grip, the tool will hit the chuck.

My most popular dodge is to insert the jaws with two of them off set by one tooth. this moves the centre out of true and lets the tool go into the 2 1/2" centre hole.
It still holds the work safely and puts the part where I want it.

oldtiffie
12-20-2010, 05:41 AM
Let's see what happens when you centre stuff in a 4-jaw chuck on a lathe.

The chuck, on its back-plate which is fixed to the lathe spindle is fixed relative to the lathe.

When centering the job in the lathe with the 4-jaw chuck the job is moved relative to the chuck, which being fixed, means that the job is moved relative to the lathe axis as the job is centred. The centering is measured by rotating the job which is rotated on the lathe spindle.

The error or off-set is measured as a radial deflection of a Dial Test Indicator. It is corrected by moving the chuck jaws.

That's pretty basic.

http://i200.photobucket.com/albums/aa294/oldtiffie/HF-45%20Mill%20misc/HF45-5.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/HF-45%20Mill%20misc/HF45-4-1.jpg

If I bolt my front-face mounted 3-jaw chuck or my ER-32 collet adaptor to the table of the rotary table with my job held firmly, I can use a TDI or a "Blake" co-axial indicator or similar:

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Dialindicator1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/measuring/Co-axialindicator5.jpg

What I do in this case is to keep the axis of the rotary table axis stationary in "X" and "Y" terms on the mill table. The indicator in the mill spindle (or anywhere else off the rotary table) is kept still/stationary/steady.

The indicator is placed on the job and the job (in the chuck which is fixed to the table of the rotary table) and the rotary table and the job are rotated relative/about the rotary table axis. Any errors in the indicated run-out of the job axis relative to the rotab axis are removed by moving the chuck relative to the rotab until all run-out is removed.

The job is now concentric with the rotab axis and all or any errors in the chuck or the rotab MT taper bore are by-passed or negated.

On both the lathe and the rotary table the aim was to get the job concentric with a rotating axis - in the case of the lathe it was the spindle axis and in the case of the rotary table it was the table axis.

Getting the axes of the concentric job and rotary table under and concentric with the mill quill spindle axis is simply a matter of moving the rotary table axis by use of the mill table "X" and "Y" slides under a TDI or "Blake" in the mill spindle, lowering the quill and moving the TDI to contact the job.

Now rotate the mill spindle - by hand for a TDI or slowly for a "Blake" - and adjust the mill "X" and "Y" slides until the deflection difference on the indicator is zero.

At that point the axes of the job, the rotary table and the mill quill spindle are concentric and co-incident - that is to say that are all in line.

This takes quite some typing buy I can do the job in less the five minutes.

And again - not an adaptor in sight or needed.

Practice it and it will soon become evident that it not only works but perhaps more importantly - why it works.

Benta
12-20-2010, 08:18 AM
Oldtiffie, we agree completely.
First get the workpiece centered on the rotabs true axis.
Second get the workpiece concentric to the spindle.

Only reason I use an indicator mounted to the mill table for the first centering is because I'm not able to brake my mill spindle and it moves easily.

Cheers,

Benta.

oldtiffie
12-20-2010, 08:48 AM
Thanks Benta.

In that case we are both either dead right or dead wrong.

I just set out to show that a job in a 3-jaw chuck can be centred just as well as the job on the lathe in a 4-jaw chuck.

It is my "tap-true" version of the "Set-true" feature on lathe chucks.

It actually gets better for me as that front-face mounted 3-jaw chuck and my ER-32 adaptor fit my flange-mounted lathe. I have about 0.003">0.005"diametrical" clearance between the locating spigot on my lathe and mating bore/recess in my back-plates as well as ample clearance on my chuck-to-flange securing bolts. I use that as a "tap-true" for my 3-jaw chuck and my collet adaptor as well.

It means that a job in a 3-jaw chuck or collet can go back and forth between my lathe rotary table and mill table as well as my tilting table.

http://i200.photobucket.com/albums/aa294/oldtiffie/HF-45%20Mill%20misc/HF45-4-1.jpg