PDA

View Full Version : Acceptable Runout for a 4” Lathe Chuck?



Silverback
01-22-2009, 06:48 PM
So I bought some goodies on ebay, including a 4” 3 jaw lathe chuck with an R8 adaptor to use in my Bridgeport. Well, the retarded chimp that I bought the stuff from (I’m temporarily withholding names till I see if he makes any real effort to fix things otherwise, I kind of doubt it based on my experience with him/the company so far) took everything, chucked it in a flat rate USPS box with no padding between individual boxes/items to keep them from moving, opened the 4” lathe box, took the top foam packaging/padding out of it, bolted the R8 adaptor on it crooked, left the packaging out and folded the box flaps around the R8 stem which smashed the rest of the contents in the box in the process of getting to me. (the outside box got to me fine with no significant marks on it, but everything inside it is mangled, even the chuck box ended up with a big hole in the side of it and all it’s parts were scattered around the inside of the USPS box)

Anyway, first I put it in my bridgeport and checked it with a dial indicator- it’s .047” out… grrr… I can actually see daylight between one side of the adaptor and the back of the chuck, and not the other.

I took and loosened the mounting screws from the backing plate/adaptor to the chuck and carefully tapped it loose, cleaned up the burrs made whoever slapped this thing together, cleaned everything , made sure the mounting faces were as good as they were going to get and put it back together (and noted that either that either the treaded holes in the back of the chuck or the holes in the adaptor plate are miss drilled, and when you line one up the other 2 are a little off, also one of the recessed holes in the adaptor piece is not drilled as deep as the others so one of the allen cap screws sticks out a little). Anyway, I went back to the Bridgeport and dial indicator:

Measured at the R8 adaptor flange: less than .001” runout, the needle just barely wiggles as it goes around.
Measured at the machined side of the chuck, about 2/3 of the way down, between the adjusting head and the grooves that the jaws ride in: right around .004” runout
I chucked up a polished steel round that I know is pretty good and chucked it up and loosened and did it again 3x (and then had to go to work) and got:.007”, .004”, and .004”
(I don’ t know what the deal was with the one .007” reading, it shouldn’t have been dirt, I’m hoping that it’s not a sign that the jaws aren’t repeatable …)

So I know that part of the answer here is “well, how accurate do you need it to be?” but realistically, how accurate can I expect average new parts like this to be out of the box? I was hoping for better than .002”, but I don’t know if that’s really realistic.

BillH
01-22-2009, 06:56 PM
For a huflungdung chuck, .004 to .005 is average. Bison is a little better. You want better? use collets or a 4 jaw.

lane
01-22-2009, 06:59 PM
For a huflungdung chuck, .004 to .005 is average. Bison is a little better. You want better? use collets or a 4 jaw.


Yes I agree 100%

Just Bob Again
01-22-2009, 07:00 PM
.002? Not on a 3-jaw. Maybe on a brand new Buck or something equally expensive. You generally need an adjust-true type backplate to do any better. If any of my 3-jaw chucks stay within .005 I'm happy and it isn't unusual for older cruddier ones to be .020 out. The scrolls and jaws wear. The runout varies from place to place. Good in some spots, bad in others.

radish1us
01-22-2009, 07:09 PM
Hmm, me-thinks he is hoping for FAR TOO much, back up a little here, where did this chuck come from, ------- E-Bay.

So you bought somebody's throw away el cheapo chuck, ask yourself this, why did they chuck it away ?

YOU WANT ACCURATE, THEN PAY FOR IT.

oldtiffie
01-22-2009, 07:17 PM
Mark.

These pics are scans of the specs that came with a 125mm (~5") and 80mm (~ 3 3/16") Chinese chucks that I bought recently. The chucks were either on or within spec and work really well.

The dimensions etc. are metric.

0.04mm ~ 0.001"

As long as the chucks are to spec, run-out is not a concern as my chuck - as it seems does yours - mounts directly to the lathe spindle flange. I have about 3 to 5 thou radial "play" and use a bronze "dolly" to "adjust" my run-out (a sort of, but very effective "tap-true") process and then tighten the chuck up. If the job doesn't require any great accuracy as regards initial job run-out I don't always bother with the "adjustment".

I hope this helps.

http://i200.photobucket.com/albums/aa294/oldtiffie/3-jaw_chuck_specs1.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/3-jaw_chuck_specs2.jpg

BillH
01-22-2009, 07:17 PM
Yes I agree 100%
Thanks for qualifying my opinion.

Al Messer
01-22-2009, 07:25 PM
For what it's worth, I was always taught that the 3 jaw was for convenience of rapid chucking and the 4 jaw was for dead on accuracy. So is a collet set.

oldtiffie
01-22-2009, 07:46 PM
Not quite Al.

The 3-jaw is for convenience as you say. It is also for gripping anything that is round or has a number of faces that is a multiple of 3 - ie triangular, hexagonal etc.

The 4-jaw is more accurate but is better suited for rough surfaces and/or irregular shapes or anything (mostly) not suited to a 3-jaw chuck.

Collets while better as regard accuracy without any further adjustment may be limited as to how far the job to be gripped can vary or deviate from the collet nominal size (R-8, C5 etc.) but the ER series have a gripping range capacity of 1mm (~0.040").

Robin R
01-22-2009, 08:06 PM
You may find you can get better results if you unbolt it from the adapter and try moving it round one hole, then bolt it up again and re test. I'm guessing there are 3 bolt holes, so you have 2 more chances at improving the run out.

BadDog
01-22-2009, 08:09 PM
I just got a new in the box Bison 8" 3-jaw. Only problem, they dropped it on the back and bent 2 pins. I've been looking for a top-line chuck that I could afford in this range, so I hopped on it. In checking it out, various locations on the spindle, and using various pinions, clamping different diameters of TGP rod, it consistently came up with at most 0.006 TIR. And using the "0" pinion in the best mount position yielded pretty consistent results within about 0.0018 and 0.0024 measured TIR on a high quality tenth indicator (no temp control or anything, but that's the numbers indicated). There were a few outliers, but reclamp back in same location was near dead-nuts the same every time. Most measurements were taken near the chuck, but changed very little at 3-4" out (don't recall exactly, but not enough to care). There were a few "outliers" that I couldn't account for, but only by a thou or so at most. I must say I am VERY happy with this chuck (not an Adjust true by the way).

gellfex
01-22-2009, 09:06 PM
BadDog that's awesome. My 17 year old Bison 5" 3 jaw has .002 TIR.

Teenage_Machinist
01-22-2009, 09:08 PM
My chinese chuck has no more than 5 thou, have not checked it often

Silverback
01-22-2009, 09:53 PM
Thanks for all the responses guys.

I guess I was expecting more than I should have been. For those of you that missed it the first time, it’s a new chuck that I bought on ebay. Where it was made… who knows, I can’t find any labels on it. To be honest, if this was for a lathe that I was hoping to do some good work with I would have never considered buying a chuck this cheap, but this is a “lets see if I can rig something up on my Bridgeport” experiment.

Oldtiffie, looks like where I’m at I’m well within the tolerances of those instructions that you posted now (though I’ve only actually measured it in a few of the “a” positions on those diagrams, none of the “b” positions), and based on everyone else’s feedback I guess I should be happy where I’m at.

One thing that was mentioned, to get it close and then tap it into position and tighten the screws… the back of the chuck has a recess and the R8 adapter has a matching shoulder that fits _REALLY_ tight. I can take the screws out of the back and it will not come off, I couldn’t pry it off. It took a good deal of careful hammering to get it to move off the less then 1/8” high shoulder. I’m sure no amount of tapping with the screws out would change the radial run out significantly. That actually involved part of my thought process… I was wondering if somehow loosening that fit slightly (maybe a little emory to get it so I can move it and then some valve grinding compound or something like that), but now I’m wondering if that might be a big mistake.

I guess I’ll try to see if the jaws are repeatable and I might try mounting it in the 2 other positions and see which one is the best…

loose nut
01-23-2009, 11:38 AM
Even if you have a high accuracy 3 jaw chuck, if you aren't careful to use it gently the scrolls will get sprung and even with good care they wear, both will cause the level of accuracy to drop.

A chuck with poor concentric accuracy will do a good job if you use it the way it was intended to be used.

lazlo
01-23-2009, 01:45 PM
Not quite Al.

The 3-jaw is for convenience as you say. It is also for gripping anything that is round or has a number of faces that is a multiple of 3 - ie triangular, hexagonal etc.

The 4-jaw is more accurate but is better suited for rough surfaces and/or irregular shapes or anything (mostly) not suited to a 3-jaw chuck.

Many machinist texts, and Forrest Addy, who's practically a machinist text :) say that the 3-jaw chuck is for when you're in a hurry, and don't need accuracy, the 4-jaw is for making money.

Or, translated into golf terminology: drive for show, putt for dough ;)

I've heard from a lot of old timers that apprentice machinists were only allowed to use the 3-jaw after they had spent a year with the 4-jaw, and by that point were fast enough dialing it in that they didn't care to switch back to the 3-jaw.

BadDog
01-23-2009, 02:38 PM
Not that I'm qualified to comment on Forrest's suggestions, but I would amend that comment to add "or when you can turn the entire piece in one chucking". Of course that also implies "with no existing reference" to keep concentric, since that would by definition be more than one chucking. Sometimes a 4 jaw is the only way, sometimes it's just extra effort. But they are really not at all hard to setup, so I don't understand some folks reluctance to use them.

Silverback
01-23-2009, 03:02 PM
A chuck with poor concentric accuracy will do a good job if you use it the way it was intended to be used.

do you mean something more besides quick, dirty and without re-chucking the item?

BadDog
01-23-2009, 04:25 PM
It doesn't necessarily mean "quick and dirty". Concentricity is limited only by the spindle bearings as long as you do not unchuck. And, a lathe chuck with terrible run out may even maintain good concentricity when rechucked as long as you put the piece back in with the same orientation.

For example, I was recently checking over a pretty nice though old and well used Rohm 8" chuck. It had an average TIR on TGP of ~0.012 and it varied by only a few thou depending spindle orientation and pinion used. However, when consistently using the same pinion, and replacing the piece back in the same orientation to the jaws, the repeatability was (as I recall) within a thou or so. So actually it was a very good chuck with the exception of the runout. Grinding the jaws would correct it and make it as good and as concentric as any you would be likely find.

This is opposed to my 10" Chinese 3 jaw that came on my lathe. It MIGHT give you 0.002 TIR on one chucking of the same TGP. Then upon rechucking the same bar with the same orientation and using the same pinion it would give as much as 0.010 TIR. This chuck is largely junk unless you are doing low precision work OR you can do it all in exactly one chucking. But it's still fine for single chuck work, and I'll be saving it for doing big nasty gritty heavy awful mess that I need to reef on the pinion to tighten. It's largely beyond hurting, so I'll use it for things I don't want to risk my Bison on...

You mainly just need to know your chuck and use it where appropriate, switching to 4 jaw when/if warranted and desired.

pcarpenter
01-23-2009, 05:55 PM
I have posted here before about the 6" three jaw that came with my Chinese lathe and how I lucked out and it gave .0002" runout on a precision ground pin a few inches from the jaws at about 1.25" diameter.....blah blah. The point is that you can get lucky, but you can't bank on it. Even low runout now will only get worse as the jaws wear in.

I took my "lucky" chuck off here a while back because I wanted an 8". The new 8 has a couple thousandths runout. I leave it in place and thats good enough for most of the stuff you need to make. I will keep the 6" in good shape and minimize wear this way. And, of course as has already been mentioned, there is a 4 jaw.

I would figure that turning that face clean in place on the mill using a single point tool held in the vise would produce the best fit anyway....forget the burrs etc. Also, bear in mind that you are going to stick a lathe chuck in your mill spindle.....which is certainly doable (and done by others here) , but a bit Rube Goldberg in the first place.

Paul

Walter
01-23-2009, 07:14 PM
I had This Bison Chuck (http://www.use-enco.com/CGI/INSRIT?PARTPG=INLMKD&PMPXNO=952288&PMAKA=271-4072) and was exceptionally pleased with it, the advertised runout was << .001, actually .0008 and .0006, I lucked out and picked mine up at an auction for $50.00, IF i had need for a chuck this size I'd pay the Enco price in a heartbeat.

loose nut
01-23-2009, 07:29 PM
do you mean something more besides quick, dirty and without re-chucking the item?


Nope, turn everything or as much as you can in one operation, part off and if necessary re-chuck (collets preferably) to finish the other end, that way it doesn't matter if the chuck has a lot of run out, the part should still be concentric, if it isn't there probably is something else wrong with the lathe IE: bad headstock bearings etc.

If you are chucking something like hex bar and having it perfectly centered in a chuck with runout, thats a different matter, you will have to shim it so it is concentric to the axis, or better still use a hex collet and forget the chuck all together. You can also make a homemade collet that will hold hex or square stock, round stock too, that is machined in place and isn't removed until all the operations that you need it for are done.


Never assume that a 3 jaw chuck doesn't have runout or has parallel jaws, even expensive ones go bad. They make work go faster but you have to allow for their failings

oldtiffie
01-23-2009, 08:19 PM
Originally Posted by oldtiffie
Not quite Al.

The 3-jaw is for convenience as you say. It is also for gripping anything that is round or has a number of faces that is a multiple of 3 - ie triangular, hexagonal etc.

The 4-jaw is more accurate but is better suited for rough surfaces and/or irregular shapes or anything (mostly) not suited to a 3-jaw chuck.


Many machinist texts, and Forrest Addy, who's practically a machinist text :) say that the 3-jaw chuck is for when you're in a hurry, and don't need accuracy, the 4-jaw is for making money.

Or, translated into golf terminology: drive for show, putt for dough ;)

I've heard from a lot of old timers that apprentice machinists were only allowed to use the 3-jaw after they had spent a year with the 4-jaw, and by that point were fast enough dialing it in that they didn't care to switch back to the 3-jaw.

Its horses for courses.

The best lesson I got - while I was in pre-Apprenticeship mandatory schooling - as regards 3 and 4 jaw chucks, was each with a cold-rolled round bar in it. The teacher set out to show which was best-suited to differing settings-up. It was a good early lesson well done and well remembered - and applied.

First the 3-jaw. Easy to do as the job was relatively self-centre-ing. If the run-out was within tolerance and within the machining zone it was OK. But if a better run-out was needed it was shimming or nothing on a 3-jaw. Anything better required a 4-jaw chuck.

Next the 4-jaw. Took longer to set the round bar to the same run-out as the 3-jaw chuck but it could be set more accurately. Easy just use any pair of opposite jaws and keep halving the run-out (both pairs). It was a quite reiterative process.

"Dialing" was unheard of. I could and still can and do set mine up "by eye" using first a very visible run-out and then reducing it. I just "eye" the back of my lathe bed. For finer stuff, I use a tool in my tool-post, a good light and just halve the gap difference between opposite jaws - as before. I can get to 0.002" (0.004" TIR) easily.

If I need any better, I will use a dial indicator. I don't use a "fine" indicator for "rough" jobs and vice versa.

Rough-finish (ie castings) or jobs nor parallel get mounted in the 4-jaw, but with thin strips across (not along) the chuck jaw gripping faces.

I NEVER strain my chucks - ever. If I can't get it right with reasonable force the job comes off the machine until I re-think it and find a better way.

If an employer/supervisor is "thick" enough to indulge in that sort of behavior as quoted, it is time for that kid to go else-where - for his own good.

oldtiffie
01-23-2009, 08:36 PM
If you are really concerned about the run-out in a 3-jaw chuck, get one with "soft" (ie "machinable") jaws - and machine them to suit the job.

It will in effect be a custom-made 3-jaw collet that will be as accurate as you can bore it for size, and as "true" as your lathe spindle is to your lathe bed and cross-slide.

Everything that is manufactured for a market (just about everything) has a tolerance. Some have many, lathe chucks included, as finer/better tolerances "cost" (more).

Even "slip-guages" have tolerances even if they are in the millionths (of an inch or mm).

Soft jaws will cancel-out most or all of the chuck manufacturing departures from nominal sizes.

J Tiers
01-23-2009, 10:04 PM
The dimensions etc. are metric.

0.04mm ~ 0.001"



To be a bit more exact, it is 0.04 / 25.4 = 0.00157, or "about" 0.0016".

mm/25.4 - inches inches x 25.4 = mm

oldtiffie
01-23-2009, 11:38 PM
Touché'

Thanks for the "heads up" JT - while it was my typo it was and is my mistake - sorry.

1 mm/25.4 = 0.03937" ~ 0.04"

0.01mm/25.4 ~ 0.0004" ~ 0.0005"

0.025mm/25.4 ~ 0.001"

The 0.01mm ~ 0.0005" means that for many practical purposes that a metric 0.01mm indicator while closer to 0.0004" (4 tenths) is a practical approximation of 0.0005" (half a thou).

I use that approximation quite often when running both metric and inch - and this may not surprise you at all (it doesn't me) that I sometimes make the same cock-up as I did in the typo above.

My everyday use metric stuff (micrometers, digital stuff and dial indicators) are all calibrated to 0.01mm. I have "finer/better" stuff but is hardly ever sees the light of day (or the shop).

The Fixer
01-24-2009, 08:46 AM
Don't forget that it's a BP mill and they're not exactly the most skookum machines. I think they are rated for a max size cutter of 3" or so. Part of your misery could be related to the machine itself or you may find even more problems when you actually chuck something of size in it and push the go button.

DR
01-24-2009, 09:04 AM
I have to go along with Fixer on this.

Why the heck would anyone even consider putting a 4" chuck on an R8 arbor in a B'port mill?

A 3" cutter would be bad enough, much less the thought of holding a chunk of metal in a 4" chuck for gawd knows what purpose.

lazlo
01-24-2009, 11:29 AM
To be a bit more exact, it is 0.04 / 25.4 = 0.00157, or "about" 0.0016".

Not only that, but as someone here once said, the Chinese spec sheets are aspirational goals, and not accuracy tolerances :)

DR
01-24-2009, 12:36 PM
...................................

the Chinese spec sheets are aspirational goals, and not accuracy tolerances :)

Worse than that, A few years back at auction I bought some remnants of a failed machine tool dealer. Included were some unlabeled wood shipping boxes.

Each box had a taper attachment for some generic import lathe. Every one had a detailed inspection certificate. A close inspection showed they were all photocopies of the same certificate right down to the smudge of a thumb print.

S_J_H
01-24-2009, 03:59 PM
Just checked my old 5" 3 jaw Cushman used on my SB9A. At least 40 years old. Random checks show .0015-.004" runout. Pretty easy to tap into .001" out though.

Steve

oldtiffie
01-24-2009, 05:43 PM
Originally Posted by J Tiers
To be a bit more exact, it is 0.04 / 25.4 = 0.00157, or "about" 0.0016".


Not only that, but as someone here once said, the Chinese spec sheets are aspirational goals, and not accuracy tolerances :)

Unless each and every spec has been individually checked on each and every part, then it is a chance you take. If otherwise it may well be a report of a selected item in a batch which, based on probability is representative of the entire batch.

But accepting, for the sake of the discussion, that "the Chinese spec sheets are aspirational goals", I can say that I've checked most of my Chinese stuff - and I've got a fair bit of it by any measure - that it has all been within, and in most cases, exceeded or bettered the spec. I have not had a "failure" in that regard. I will say that the "finish" could have been a bit better and that there were some "niggling" short-comings, but none detracted from the performance and the vast majority were "fixable" - and fixed.

Perhaps I've been lucky. Perhaps too it might be that I am prepared to pay a premium price for acceptable products from a known Supplier who relies on repeat performance from the "Trade" and not from the general public or retail. In these "hard" times the Supplier has to try that much harder as the purchasers are a lot more fussy about the "bang for their buck" than they otherwise might. Buyers in the commercial zone are a lot more concerned about margins as well as reducing "down-time" so as to maintain efficiency.

Suppliers are concerned about margins and return business as well as "returns" and"call-outs" and customer complaints too.

So, perhaps I too get the performance I pay for and I gladly pay more to get more.

I accept that nothing is perfect and that any problem is just something that I have to deal with - and I deal with it - and make the most of what I have on the day.

By and large I do pretty well in what I need to do and with what I have to do it with.

Silverback
01-25-2009, 07:02 PM
I have to go along with Fixer on this.

Why the heck would anyone even consider putting a 4" chuck on an R8 arbor in a B'port mill?

A 3" cutter would be bad enough, much less the thought of holding a chunk of metal in a 4" chuck for gawd knows what purpose.

Never said I was doing something the right way...

oldtiffie
01-25-2009, 07:16 PM
Who said it was wrong - and why?

There is nothing to say that if due regard is paid to the capabilities of the machine, the job and the set-up, I can't see that it is wrong.

That someone or anyone else either hasn't done it, or has been told that it is wrong does not make it wrong in the hands of a good machinist.

People here use a chuck mounted on plain spindles held in lathes and that seems to be accepted as OK.

I would be more concerned about an out-balance fly-cutter at high speed and feed held in that same R8 collet in a mill than I would be about a well balanced slow(er?)-revving 4" chuck in the same collets.

If I recall correctly, "Boomer" does just that as he uses his vertical mill as a lathe (he doesn't have a lathe) and has survived so far and turns (sorry about the pun) out some very good work.

Silverback
01-25-2009, 07:32 PM
At this point I'm in no real position to argue if what I'm doing is being done the right or wrong way (the only real "machine" equipment that I currently own is my Bridgeport, and I'm quite happy/proud of it after years of rigging stuff up on the drill press..., definitly wasn't the right way but I've made some cool stuff that way), so I figured that if someone feels it's wrong... great, doesn't matter to me if it works for what I need and I don't destroy anything in the process (or if i do it with the understanding of what I'm going to destroy in the process).

I'm turning pieces roughly 2.5" and smaller, and feel that the extra mass of the 4" chuck actually helps steady things a little.

I'll have to see if I can dig up some of boomer's work.

BTW, oldtiffie, what's an old tiffie? Seems like you're always injecting some useful information and I'm replying to you fairly regularly, so I kind of wonder what I'm calling you when I'm referring to you as by your screen name.

The Fixer
01-25-2009, 08:54 PM
Just wanted to make sure it was understood that the BP is a 'light weight' machine in the milling world and a chuck could be a bit unsafe to someone unfamiliar with it's (the BP) limitations.
I agree there's no "wrong" if it's done with due considerations, and sometimes ya gotta be a little unorthodox in your approach to problem solving! I will admit to mounting a 6" 4 jaw on an R-8 in a King MillDrill with the head swiveled to 30deg for a very creative set up. Not to mention the 7" angle grinder clamped in the swivel vise.........Suffice it to say that it was a Saturday night solution that got us back on the racetrack for Sunday racing! Just keep the spindle speeds down!

oldtiffie
01-25-2009, 09:14 PM
................................................
BTW, oldtiffie, what's an old tiffie?

.................................................. ., so I kind of wonder what I'm calling you when I'm referring to you as by your screen name.

Thanks SB - easy to answer.

The answer is in two parts.

I am old by any measure - 72.

I am - among other things - an ex Ordnance Artificer in the OZ Navy. Artificers (there were several other branches: Engine Room, Electrical, Air etc - similar and derived from the Brit/UK Royal Navy - RN) were called "Tiffies".

Hence: old +tiffie

Silverback
01-25-2009, 11:13 PM
Fixer... i get it but you wouldn't believe how heavy duty a bridgeport seems when your only comparison is a delta drill press...
:eek: