Runout on lathe....
I was wondering if somone could tell me why I would have runout on my 3 jaw chuck. I have .005 inch runout on my lathe chuck. There doesnt seem to be any play in the spindle. The runout seems to be with the chuck itself. I have a 16 in Standard Modern Utilathe with an 8 inch Pratt 3 jaw chuck D1-5 mount. Is there a way to fix the jaws so as to eliminate the runout? Is the problem in the jaw threads? What about if I take off the chuck and rotate it to the next pin hole on the backplate? Would my problem be worn jaws? They seem to be in good condition. I am new to lathes but this runout thing is driving me nuts. Any comments or suggestions????
There have been many good threads about this. Use the forum search
But, I'd say .005 runout on a 3 jaw is pretty good though.
its the nature of the beast, that beast being a 3 jaw chuck. Hobbyists like and defend them, but the fact remains they dont hold as well and will NEVER cut run out down to 0.0005 over its griping range.
You COULD get a set-tru chuck and be able to adjust for a given range of diameters, but whats the point? You might as well put that 4 jaw chuck on and adjust it anyways. 4 jaws are far more versatile, and accurate.
Your idea of rotating to the next pin hole is a good one, as alot of chucks and old spindles have a sweet spot were the run out is the least.
If you check the runout at multiple diameters from 1/4" up to maximum you'll find the runout will vary quite a bit if your chuck is at all typical. The scroll is not particularly accurate and becomes less so over time, depending on how hard you crank the chuck handle.
This of course means that if you grind your jaws true they will be true only for the diameter at which you ground them. I'd take a 0.005" chuck, but one shouldn't expect repeatability from a three-jaw. A Set-Tru chuck can help some.
Some time ago there was a post on 3 jaw chucks and a fix for making them more accurate. If you have a 3 jaw chuck that bolts to a backing plate, remove the 3 jaw chuck and machine the backing plate alignment lip about 0.010" smaller diameter . machine the backing plate bolt holes larger. And while your at it check the backing plate mounting face run out and machine as neccessary.
Reassemble the backing plate and chuck and remount them to the lathe spindle.
Now if you need to hold less then 0.005", install your part in the chuck, lock it down, then mesure the run out. if it is more then what you want, loosen the mounting plate bolts(not by much, they need to hold the chuck) and tap the chuck until you get it close to zero runout, retighten the bolts and check run out again to verify it held.
FYI: I have done this to all my 3 jaw chucks now and it works really well.
If you are doing alot of repeat parts, think about a collet closer also.
A 3 jaw isn't just for hobbyists
If you need to do a lot of pieces that are all alike, and too big for your collets, a 3 jaw is nice....
An adjust-tru can be set up for the first, and should stay good for the rest.
or if you just do all the work without removing the part until it's done, it's also fine, unless you use the original OD of the piece and need it concentric.
or bore soft jaws to hold the work, and most all of them will be right, as with adjust-tru. (might drift, wear, etc)
Like any tool, it has limits, and absolute accuracy is the problem with 3 jaw chucks.
The problem with 4 jaw chucks is that they are ALWAYS less accurate than a 3 jaw......... when you start they are usually WAY off for your present job.
The good news is that they can be tweaked-in to about perfect, that's one thing they are made for.
The bad news is that you will need to do that for each of the 750 parts you are running, because loosening a 4 jaw generally throws all of the adjustment off. Even carefully loosening only two jaws doesn't do it.
Hence the 3 jaw, soft jaws, etc.
You might also find that each of the three wrench sockets will give better results at differant diameters. On mine I find that the #3 gives significantly less run out on 1" diameter work.
I was messing with my 6" 3 jaw D1-4 chuck last night trying to improve the runout. I was rotating the chuck to different mounting positions on the D1-4 mount. Runout went from .007 (worst case) to .0025 (best case). I can live with .0025 on a three jaw.
Best thing I did was learn how to dial in a 4 jaw. Now my 8" 4 jaw is my chuck of choice and, for the most part, stays on the lathe.
I have a Rohm 4 jaw scroll on one lathe. I does the job of 4 chucks on my other lathe. I prefer it over my adjust a tru. You can't get by without an idicator or you will have run out as I check each jaw for tightness. Kenny
I had the exact same problem on a 15" leblond I just purchased with a 3 jaw chuck. Put a 1" round in the chuck - .007 runout but check the spindle, .0005". WTH! Then it occurred to me, someone had turned the jaws around and put them back in the wrong place. Take your jaws off and check. The " stamped on the chuck should match the " on the jaw. Sure enough 2 jaws were switched. Put them on right and .0005 on the part!
I see this positioning advice about chuck jaws quite often. On the admittedly few chucks I have experience with if two of the master(or one piece) jaws are switched the massive error is immediately apparent. When I reassemble a scroll chuck I look at the back of the jaw and start the one with the thread closest to the tip first and the jaw with the threads farthest from the tip last. I don't pay any attention to the number(if any) stamped near the the slot. Were these chucks selectively assembled or the jaws ground in place? Seven thousandths seems a large error in scroll centering. Also when switching the gripping surfaces on master jaw chucks I have no set procedure to ensure the same top and master jaw remain together. I haven't noticed any problems from my slapdash practices. Am I lucky or unobservant?
Also check that the chuck is properly seated on the spindle nose. I have two lathes with D1 spindles, a D1-3 and a D1-6. They are better than screw on chucks, but I don't especially like them because they require that the recess in the backing plate exactly match the spindle. If it is tight, you have to go around the locks progressively, tightening each little by little while watching runout on a test bar in the chuck. An especially bad problem is when the backing plate face is not completely seated on the spindle and the runout gets worse the farther out on the test bar you check. I converted the 3 jaw chuck on the D1-6 lathe to an adjust tru, which helps in centering but not on the wobble.
Don't forget soft jaws - wonderful invention
Help me understand. You don't pay attention to matching the " on the jaws with the " on the chuck? Don't those "s 1,2,3 correspond to the position they were in when the chuck was machined? If you switch one your screwed. If they are replacement jaws or you're dealing with a 4 jaw I see your point. I've been in the metal industry for 32 years and it's been common practice to always match the numbers. I'm also wondering about some of the post concerning tolerances. If a lathe wouldn't hold .0015 I wouldn't own it. Just my opinion. Let me say I really enjoy this sight and the information and help. As i said I've been in the business for 32 years but my tooling was always preset and I'm learning alot about indexible tools here.
Originally Posted by Larry on Lake Superior
Whoa there big boy! Not very observant? That's kind of arrogant isn't it? I'm not talking about disassembling anything, just taking out the 2 socket head bolts on the reversible jaws and turning them around. That's the normal procedure for chucking something a large diameter part. As I said, I have a 15" leblond lathe with a three jaw chuck. .007 runout on a 1" round in the chuck. .ooo5 on the outside diameter of the chuck. Checked with a dial indicator. Jaw 1 and 3 were switched. After I corrected them the runout on the shaft was .0005. There you have it. Just a fact. Huff and puff if you like but it changes nothing. If you don't believe me that's fine. Could be you've been working with junk. By the way it doesn't have to be
Originally Posted by Phil Burman
.005 to be visually obvious. I'm just trying to help this guy out. You want a picture I'll send you one.