View Full Version : Centering Work in 4 Jaw Chuck

02-02-2003, 09:21 PM
What is the best way to center something in a 4 jaw chuck? Is there a system or method of some type? Assuming that each job is a bit different, how about the best way to center a piece of bar stock within +/- .001.
thanks randy t

02-02-2003, 09:59 PM
I eyeball the concentric rings first to get the jaws in the ballpark. Then put a mag. base indicatator on the work; spin the chuck by hand and loosen the low side and tighten the high side. That's all there is to it.

Forrest Addy
02-02-2003, 10:12 PM
What you ask is far too complex to compress into a four paragraph reply but here's a place to get started:


The presentation is a bit simplistic but the illustrations are very clear.

I seldom use a three jaw chuck. Over time I've become so practiced that a four jaw is about as quick. Futher a four jaw is far more accurate and has a much better grip than a three jaw.

The best kept secret of four jaws chucks is to grip only a short amount in the jaws. This allows the jaws to bite in and you can bang the work straight so it dials in on both ends. If you swallow the work you can't move the overhang because the jaws prevent the work from moving.

Four jaw chucks are limited in accuracy only by the skill of the user. Using a very old Pratt and Whitney electronic test indicator (vacuum tubes an all), I once dialed in the spindle sleeve of a G&L horizontal boring mill to 0.00005" (yup, 50 millionths) on the bearing fits so I could internally grind new bushings.

Three jaw chucks are quick and convenient for rough and single ended work. Secondary operations where end to end concentricity is a essential requires you to set up and bore soft jaws or have the expensive adjust-tru feature. If you work out of the hard jaws on a typical three jaw, you'll be battling residual error every inch of the way.

[This message has been edited by Forrest Addy (edited 02-04-2003).]

02-03-2003, 12:10 AM
Forrest and Snorman are right; the more you do it the easier it becomes. When you've got it close enough to indicate, pay closest attention to the readings directly over the jaws. Let's say jaw 1 reads .050" higher than the reading at jaw 3, then the piece needs to move half that amount toward jaw 3. Once that plane is close, switch to 2/4. It might need tweeked a bit after that, but that'll get you real close real fast. You can do it quicker than I can say it, once you get the hang of it.

02-03-2003, 08:23 PM
1)Use the rings on chuck and the corners of jaws to get close.
2)Using te tip of cutting tool to "indicate" adjust either #1 and #3 jaws or #2 and #4 jaw really close and then do the other set. At this point you should be within .003 to .005.
3) Use a dial indicator or "last word" to finish the job. Again do only one set at a time.

Before you start you might want to check if the part is "round". If the part is not round then you can only get relativly close. On square parts I use the flats not the corners. Last Saturday I dialed in four parts and on step (1) I was within .030 and step (2) within .003 in two steps each jaw.
Then I used the "Last Word" to get it accurate.


02-03-2003, 08:45 PM
"...and you can bang the work straight..."

Would you elaborate on that a bit? On the face of it, that doesn't sound like anything I'd want to be doing on a lathe I owned.

02-03-2003, 09:44 PM
Round stock is easy odd ball work pieces are more fun.I usually center punch the center I want to work around then back the jaws out to clear the work piece ,then using the tailstock with the dead center installed use the center to locate the part on the chuck face, then advance the jaws in until they touch the part on all four sides snug them down and check the center with the indicator bar and adjust as needed . http://bbs.homeshopmachinist.net//biggrin.gif

02-03-2003, 11:25 PM
Thanks for the above - None of the methods worked for me until I figured out that my 4-jaw chuck had only 3 jaws.

Mike "1 jaw short"

Forrest Addy
02-04-2003, 02:03 AM
deleting double post

[This message has been edited by Forrest Addy (edited 02-04-2003).]

Forrest Addy
02-04-2003, 02:12 AM
SGW: Re "...bang the work straight"

Machine tools are designed for a certain amount of shock and vibration. They'd be useless if they weren't. A little action with a copper or babbitt hammer is necessary to bring extended diametral features into concentricity with the chucked end. Sledge hammer, no. I lb copper hammer certainly.

Use common sense and adjust the weight of the hammer to the size of the machine. I've used full overhand blows of a 16 lb babbitt hammer knocking in a bronze ring casting on a 48" American lathe and on a EE a tapper consisting of a 3" piece of 5/8 brass round stock.

You have to lift the dial indicator or otherwise isolate it from the shock. The lathe suffers not at all when subjected to shocks only a few percent of those encountered in machining casting risers or flame cut edges full of hard spots.

My 10 HP lathe once made 8 lb of chips per minute. I was showing showing the nephew what max rate stock removal was all about on my 17" lathe as I machined him a pair of weld-in rear axle housing ends (whatever they're called). Since I was starting from 7" dia stuff there was a full inch to come off. Randy timed and caught and weighed the chips for lack of anything better to do. I made the rough down in two cuts. The figures are on my wall. 8 lb, 1 oz of chips in 1 minute.

Machine tools are tough. Make them work to earn their keep.

[This message has been edited by Forrest Addy (edited 02-04-2003).]

Ragarsed Raglan
02-05-2003, 02:04 PM
Gotta say I agree with 'weirdscience' about round stock being easy. Now for irregular stuff I mark out my centre point ....and (as I've said before in previous posts) use a centering 'scope mounted in the tialstock. Best piece of equipment this side of a door handle.


kap pullen
02-05-2003, 10:33 PM
Mr Addy is right.

The Bull (of the woods) told me once, "the machine's suppose to work, not the machinist."

"Make time roughing out, so you have time to finish up."

"Run all three heads on that boring mill, planer, or slab mill, that's why they's there"

Never really managed to achieve that goal (machinist not working).

Guess he was better'n me.

We called him Caliper Al, The working mans Pal.

He could turn a shaft straight with in .005
using those old tap calipers without even shutting off to measure, or adjust the size (or so he said).

Just set backlash in on the slide, and out on the compound he said, and set her on the run.

A hero in his own mind he was.


02-15-2003, 07:17 AM
Another way to center a large block on a 4 jaw is to make a long rod with a point on the end. Chuck this up in the tailstock and put your indicator on it. It will run out of round until you get the chuck adjusted. As everyone else stated, once you get the hang of it you can adjust a 4 jaw almost as fast as closing up a 3 jaw. I have found that this works rather well and will get you within .001 or better. This is a rough sketch but you get the idea. http://hotrodders.com/photos/00003338/Indicate.jpg


Dr. Rob
02-15-2003, 07:39 AM
Right on, Kevin! What a great sketch! Terriffic.

And Wierdscience trick with the tailstock is what works around here. Gets the job pretty darn close, fast & easy.

Otherwise, my four-jaw seldom gets used at all; I'm just not as proficient with it as the rest of you. I found that careful planning of the order-of-operations or other trickery gets me by. Maybe I should learn more.

02-15-2003, 11:38 AM
I forgot about that way. I have used that numerous times also. And if the piece is smaller than the hole in the chuck you can back it up with a couple of small parallels until it is tightened up. DON'T FORGET TO REMOVE THE PARALLELS before turning the lathe on though. When I show teach someone something at work I like to teach them the more difficult way of doing things first then after they get pretty good at that then I'll show them the shortcut. The very first thing I always ask someone is if they have ever trammed a mill (usually Bridgeports or Alliants) before. If they haven't I'll kick the mill out of tram and show them how to do it. I have let them go 2-3 hours before working at it util they get it right. After the hard stuff is out of the road then they get the enjoyment out of machining.


02-17-2003, 05:40 AM
I'm not very experienced in machining, but I center up my 4 Jaw chuck by figuring out how many thousandths an 1/8 of a turn of the chuck key will make. I tighten up the item on all four jaws and rotate the chuck while measuring the offset from peak to valley with a dial indicator, only measuring in the plane relative to the Jaws you are adjusting. Figure half the travel from the peek to valley, and loosen on the opposite side that amount and tighten on the peak side. After a couple of rotations it is easy to dial it in. I only have two chucks so it isn't hard to figure out.

Ian B
02-17-2003, 08:40 AM

You mentioned tramming a turret mill head square; I do this with the outer ring of a large scrap roller bearing laying on the table and a dial gauge held on a bar in a boring head, then adjust he 'nod' and 'roll' of the head until it all seems square. As my turret mill (Beaver) is a fairly recent acquisition, I've only done it a couple of times so far.

How do you do it, any tips? I saw Thrud's link to the cross between a precision ring and a milking stool (drool, but too expensive for me) - what do you use? Do you tram to the table, the vise,...



02-17-2003, 10:22 AM
I tram my mill either way. My department that I run has surface grinders so the first thing I did was to tear my vise apart and precision ground everything. It was a used vise so I reconditioned it to new. That way I know every characteristic of it. I then trammed the head of the mill off of a large ground plate (1/2" thick tool steel) and the mill table. After it was trammed that way I secured the vise, put two 1-2-3 blocks in the vise, put the plate on top of the blocks, then double checked the tram. If it is different then I know that the vise probably has a chip or dirt under it. Please note the vise was checked with an indicator, height gage and a surface plate. I like to double check everything and by that I mean check everything differently twice to insure everything is accurate. Always double check yourself. Where I work everyone in my department has their own mills to run. This way everyone has to learn to read their own mill, know it's characteristics, and how to compensate for any errors that there may be. We make precision aircraft parts so accuracy is a must. Most people do not know how to read a piece of machinery and they just assume that it will make what they want. You would be surprised at how many thousandths a table will drop by moving from one end to another. The tool designers we have jus assume that because a mill can run 26" in the "X" axis that we can precision mill something that long. By that I mean within .005 thousandths for the total length. Even on a new mill that is almost an impossibility. But anyways about the original question...I tram either way. If you have an old vise then clamp up something you know that is precision and tram the head to the vise because that is what is going to be holding the part.


02-18-2003, 12:29 AM
Ian B:
Milking stool. I laughed and laughed! http://bbs.homeshopmachinist.net//biggrin.gif