By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by Allendale Jan 24th

3-Jaw Chuck not running true - how to fix

All Topics | Latest Posts

Search for:  in Thread Title in  
Avon20/03/2021 17:59:33
71 forum posts
26 photos

I've discovered that my 3 jaw chuck doesn't hold work 'on-axis', I've put a DTI on various diameters with the following total runout measured:

1. With a ground 1/2" steel bar held in the internal 3 Jaw Chuck: 0.25mm

2. Same bar held in 1/2" collet: 0.04mm

3. Mandrel, just next to headstock bearing: 0.0mm (no discernable movement)

4. Mandrel, inside the collet taper: 0.0mm (no discerable movement)

The run-out using the chuck is too much to allow me to turn something and then reverse it to complete machining.

What do I do to fix it? Is it possible to fix it?

This is on a Smart & Brown, Model M, Mk 1.

Grateful for your advice.

old mart20/03/2021 18:10:30
3416 forum posts
210 photos

That runout is common on three jaw chucks. You would be best getting a four jaw independent, they can be dialed in to whatever you want. With the chuck you have, the easy way to loose most of the error is to reduce the fit between the backplate and the chuck register by 0.3mm so you can get it running better before the bolts holding the chuck on are tightened.

Avon20/03/2021 18:15:35
71 forum posts
26 photos

Hi Old Mart,

Just to avoid misunderstanding your statement 'reduce the fit between the backplate and the chuck register by 0.3mm' means to allow more of a gap between the locating circle of the chuck and the raised portion of the back plate: I would then nip up the bolts once I've checked the chuck runout.

Apologies for what is probably a dosy question!



Derek Lane20/03/2021 19:05:14
569 forum posts
112 photos

Check out Joe Pieczynski video

old mart20/03/2021 19:07:41
3416 forum posts
210 photos

With a little clearance in the register fit,you can gently tap the chuck body into line while the bolts are loosened slightly. It is a hit and miss method, and needs to be done after the workpiece is clamped tightly in the jaws. The runout should be rechecked after tightening the bolts. A different diameter of workpiece may run differently, depending on how the scroll is worn and wear generally. At the museum, we have a Smart & Brown model A which has the same size spindle fitting as your machine, and over the years I have ammassed about 6 three jaw scrolls and 2 four jaw independents. Only the four jaws and one of the three jaws have a tight register, the others can be adjusted if necessary. The ones that can be adjusted are all very good and have only been adjusted once or twice, their runout at different diameters is well within my idea of a good chuck.

Watch the video before doing anything to the chuck.

I have watched that video some time ago, and it gives a lot of very valuble tips on how tighten up work properly.

Edited By old mart on 20/03/2021 19:11:26

Edited By old mart on 20/03/2021 19:15:46

Dave Halford20/03/2021 19:21:07
1820 forum posts
19 photos

Before you do any of the above try a larger piece of turned metal and see if it's the same.

I've got a 5" chuck that throws out 15thou at 1/2" dia, but only 3 thou over 1 1/4" bar. You can feel the scroll go tighter as you wind the jaws out.

SillyOldDuffer20/03/2021 19:29:01
7695 forum posts
1699 photos

Another possibility is the jaws are worn. New jaws, or try regrinding them by clamping the jaws against an outer ring and truing up the ends. I expect someone else will be able to explain better! If the chuck body is worn or damaged, replace the chuck

Before doing anything too wild though, strip it down and for a good clean. Might be swarf in the works, especially at the register surfaces


old mart20/03/2021 20:13:37
3416 forum posts
210 photos

As S O D mentions, a good strip down and clean can work wonders.

not done it yet20/03/2021 20:15:35
6440 forum posts
20 photos

First, do your jaws have the same number stamped on them? Next, does it make a difference if you change the jaw positions by one slot each way (not all chucks have the slots numbered).

Then try the ideas above. There are very few self centring chucks where the workpiece can be removed and remounted without losing concentricity. Collets go a long way towards, but turning between centres is the time-proven method for items which need dismounting and remounting.

old mart20/03/2021 20:39:45
3416 forum posts
210 photos

The other thing we all forgot to mention is which of the three key positions gives the best results. Only testing with different size work, preferably precision ground on the od, using each of the key positions will find the sweet spot, and a small mark made to remember which one. Any of them can be used for undoing the chuck, but always use the best for tightening up.

For the best critical work, soft jaws bored correctly or four jaw independents are unbeatable.

Edited By old mart on 20/03/2021 20:42:04

martin haysom20/03/2021 20:49:22
58 forum posts

i too have a smart and brown model m mk 1 i get similar numbers to you when that is not good enough i use the 4 jaw

bernard towers20/03/2021 22:24:38
338 forum posts
89 photos

Why not try to resurrect it by grinding the jaws. In a previous issue of MEW it shows how to make a petal plate and from personal experience it’s nothing short of brilliant. I have done 3 chucks and have also lent it to 3 others who were really pleased with the results, what have you got to lose?

Neil Lickfold20/03/2021 22:52:33
720 forum posts
127 photos

You want to check a lot of different sizes. The scroll on some early lathes is 8tpi so every 1/4 inch size will likely be the same runout. So 1/4 bar is often the same runout as 1/2 inch. But 5/16 and 3/8 are on a different part of the scroll. Is it the same when held by the full jaw length and at 1/2 the jaw length. Older chucks can have damage to the chuck body by gripping too tight on short pieces, and racks the jaws track in the chuck body.

Another option is to loosen the holding screws, and the spigot, then put Belleville washers under those screws or nuts. It will allow you to make small adjustments to the part being held , a bit like a grip true does, but without the adjuster cams.


Mike Poole20/03/2021 23:13:50
3095 forum posts
72 photos

This old chestnut goes round and round, if three jaw scroll chucks were capable of extreme accuracy then why make collets and 4 jaw independent chucks. A scroll chuck can be made to a highly accurate tolerance and when new and unabused may perform quite well. The day will come when the chuck is over tightened or a jam up strains the chuck and it will now have lost whatever class of accuracy it was manufactured with. I was taught to not regard the scroll chuck as accurate and to plan operations to avoid removing the work from the chuck, if removal cannot be avoided and concentricity is important then a collet or 4 jaw independent should be used when remounting the job. If after a thorough clean and inspection the runout persists then check at different diameters and if you have a similar runout on all diameters then grinding the jaws may improve things but this improvement may not last forever and certainly a collision could undo it all in a moment. The mechanism of the 3 jaw is not well protected from dirt and swarf so it can eat a chip at any time and adversely effect the accuracy. Unless you clock job then it could be unexpectedly out at any time. I couldn’t tell you what the runout of my scroll chucks is because if accuracy is important then I won’t be using it.


MadMike20/03/2021 23:27:29
226 forum posts
4 photos

Mike Poole is absolutely spot on. I struggle to believe that so much has been written here about backplates, grinding jaws etc on a three jaw chuck.

When yo turn something in your three jaw chuck does it produce correct size concentricity, straightness etc? If it does then you are doing fine.

Having part turned your piece, turning it in a three jaw chuck is a totally pointless exercise. You have no chance of the second end running true to match the first turned end. So either use a collet on both ends, mount the piece between centres, turn the piece complete and merely part it off leaving a simple facing operation to finish, or if real accuracy is not required bore out a set of soft jaws to accept the first turned end. Yes you could turn the part and mount it in a four jaw chuck but the methods I have describes are more consistent. HTH. Happy turning.

Hopper21/03/2021 06:16:54
5505 forum posts
137 photos
Posted by bernard towers on 20/03/2021 22:24:38:

Why not try to resurrect it by grinding the jaws. In a previous issue of MEW it shows how to make a petal plate and from personal experience it’s nothing short of brilliant. I have done 3 chucks and have also lent it to 3 others who were really pleased with the results, what have you got to lose?

Issue 267 of MEW has the full article on regrinding chuck jaws. There is more to it than a simple ring. The "petal' or clover leaf plate is very simple to make and a Dremel grinder held in the toolpost with a simple bracket will do the rest.

Three jaw chucks should hold true within about .003" or 0.07mm so the OP's 0.25mm is clearly unacceptable. First step is to give it a good clean out and check over. Check mountings on backplate etc. And check to see if the jaws are bellmouthed by looking for a gap near the front of the jaws when lightly gripping a ground bar etc.

The several Myford and Drummond chucks I have reground now have a runout of less than .003" at any given diameter of job. Very happy with the result.

picture 9. chuck regrinding. sparks fly as grinding starts..jpg

picture 4. chuck regrinding. overview of plate fitted to chuck..jpg

picture 11. chuck regrinding. stop grinding when all three jaws are evenly ground..jpg

David George 121/03/2021 08:02:46
1675 forum posts
497 photos

I bought a 4 jaw self centering chuck with jaws that were mixed up and didnt match the body but we're numbered 1 to 4 with diferent set numbers. It was cheap at the Doncaster show. I mounted it up on a new back plate and with a test piece there was about 0.65 mm run out and needed grinding. I made a set of holders for the jaws, a four hole cloverleaf plate for the inside grinding and a set of rings for the outside faces. After grinding the run out was less than 0.02 mm. You have to load the jaws in the direction of his they are used and direction of clamped.


I borrowed a toolpost grinder from a friend and ground both inside and outside faces on the jaws.



Edited By David George 1 on 21/03/2021 08:05:10

larry phelan 121/03/2021 09:36:01
1119 forum posts
14 photos

I used to think 3 jaw chucks were dead on, not anymore !

But then, I used to believe in fairies at one time too.

The 4 jaw is your only man.

Paul M21/03/2021 10:07:57
70 forum posts
4 photos

Never used a 3 jaw chuck that is spot on accurate. To ensure concentricity I always use over size bar and turn down work to diameter if in a 3 jaw. Anything else is first clocked in a 4 jaw. Others may disagree but I reckon it's a waste of time fiddling about trying to improve the run out of a 3 jaw chuck apart from occasional grinding of the jaws when worn, if you have the kit.

SillyOldDuffer21/03/2021 10:26:25
7695 forum posts
1699 photos
Posted by MadMike on 20/03/2021 23:27:29:

Mike Poole is absolutely spot on. I struggle to believe that so much has been written here about backplates, grinding jaws etc on a three jaw chuck.

When yo turn something in your three jaw chuck does it produce correct size concentricity, straightness etc? If it does then you are doing fine.


Certainly don't disagree with Mike and MadMike comments, but Avon's 0.25mm error suggested his problem is worse than usual.

My Chinese 3-jaw measures about 0.05mm 50mm away from the chuck jaws. If Avon's chuck has a 0.25mm error at the jaws, I think something is wrong.

But provided work isn't moved during the job, the absolute accuracy of a 3-jaw often doesn't matter because cuts are taken relative to the lathe's stable spinning axis, not the imperfect chuck. When work has to be moved, collets and 4-jaws are the answer when accuracy matters. Expensive 3-jaws provide an enhanced level of safe interchangeability, which can save a lot of time, but no 3-jaw is entirely trustworthy for spot on work. I suspect for many amateur purposes, an ordinary 4-jaw is better value than a costly super 3-jaw. The 3-jaw pays off in a busy workshop - hours of turning, day after day, not grandad firing up his Connoisseur for a Royal visit!



Edited By SillyOldDuffer on 21/03/2021 10:28:01

All Topics | Latest Posts

Please login to post a reply.

Magazine Locator

Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!

Find Model Engineer & Model Engineers' Workshop

Latest Forum Posts
Support Our Partners
rapid Direct
walker midge
Eccentric July 5 2018
JD Metals
Eccentric Engineering
Subscription Offer

Latest "For Sale" Ads
Latest "Wanted" Ads
Get In Touch!

Do you want to contact the Model Engineer and Model Engineers' Workshop team?

You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.

Click THIS LINK for full contact details.

For subscription issues please see THIS LINK.

Digital Back Issues

Social Media online

'Like' us on Facebook
Follow us on Facebook

Follow us on Twitter
 Twitter Logo

Pin us on Pinterest