Lapping a Unimat 3-jaw chuck

Hello:

The OEM 3-jaw chuck that came with my ca. 1980 Unimat 3 does not seem to be in good shape.
Granted, it is 43 years old and it may have been subjected to some abuse, no way of knowing just how much.

While attempting to diagnose the source of chatter I could not find, I noticed this:

A so-so photographer and a a so-so camera makes it look worse than it is but there is a good amount of axial play there, probably one of the sources of the chatter I often (too) get as well as deviations when drilling a small diameter length wise hole with a drill bit.
 

I then disassembled it to see what I could find inside and put the pieces on my scanner's bed (!) to get accurate images of each one's critical surfaces.

These are those images:

1. Jaw surfaces do not look too good.

I probably have some responsibility for part of this damage as once or twice a large-ish piece flew off the chuck, probably because the jaws were already in bad shape and did not hold it properly.  

2. Asides from the scanning software artifacts, the scroll teeth seem to be (?) in decent shape.

3. The same could be said for the scroll itself.
 

To my untrained/unexperienced eye, what looks bad are the jaw surfaces, the rest being in acceptable shape.
ie: I cannot make out any issues with the scroll or the teeth running on the scroll but you chaps know better.

As to how to fix this rather expensive problem, all the methods I have seen (on-line) are with the jaws in place and a rotating cylindrical element grinding each jaw as the chuck rotates, with grinding element mounted on a Dremel type tool attached to the tool post.  

ie: jaws ground one at the time and preloading then chuck. This requires some sort of jig to get them tight/with no slack while holding nothing to let the stone do its work.

It seems that the success of this method is very much dependent on how well  the chuck is set up with whatever is used to preload it.

I have only come across one other method in the January 1924 (p. 86) issue of Popular Science and mentioned in the unimat forum at io.groups:

Edited By Julius Henry Marx on 17/01/2023 11:41:29

No picture extant. There is a post somewhere about how to post pics but it escapes me. Basically, upload pics to your forum album, then click on the camera icon at the top of your post and select the pic you want.

Sounds like you will need to regrind your chuck jaws. There are previous threads on it and a good(ahem) article in MEW a couple of years ago, author Pete Barker.

Hello:

> ... a post somewhere about how to post pics ...

Indeed ...

It took me a good while to figure out how to do it properly.

But the editing time-out is too short ... 8^/

Admin: would it be possible to have it pushed to at least 45/60 minutes?

Continuation:

> ... other method in the January 1924 (p. 86) issue of Popular Science and mentioned in the [email protected]:

Image courtesy books.google.com

It may be possible to avoid having to use the pilot bushing by fitting the arbor to the size of the Unimat's spindle bore but maybe this method is dependent on proper headstock/tailstock alignment.

I think have the bed ie: headstock/tailstock reasonably well aligned.
After a couple of surface passes on a steel rod mounted between centers, I get a 0.05mm deviation when I run a comparator dial from point 'A' to point 'B' 125mm away.

Is it good enough?

I'd appreciate comments on how to fix this chuck problem and which method would be most suitable.

Thanks in advance.

Best,

JHM

Edited By Julius Henry Marx on 17/01/2023 12:00:26

Posting photos on the forum is easy enough when you know how, but far from obvious!!! Instructions here.

Not unlikely that a 43 year old chuck is worn. Condition depends on how the tool was used in the past. Many second-hand hobby lathes were gently worked and carefully maintained by sympathetic owners, and are in almost 'as new' condition. At the other extreme, some were drubbed for years on end by careless brutes working against the clock! Most are somewhere in the middle, unlikely to have been either mollycoddled or thrashed, but maybe somewhat abused by Learner Drivers, or kept busy for years by a keen modeller. Old chucks are often bell-mouthed or worse. A light regrind as per Hopper fixes moderate wear, otherwise new jaws, or a new chuck if the scroll and drum are damaged.

Dave

That looks really bad. You won't fix that much wear by lapping. Lapping is only to remove tiny amouts of metal, like. .001". Even grinding it would be difficult. Can you stick a feeler gauge in there and measure how much wear there is? It may be the chuck body is distorted. In which case a new chuck is the best solution.

Otherwise you could try grinding it with a Dremel tool mounted in the toolpost. With a suitable clover leaf plate to push the jaws outwards at the far end while grinding.

Edited By Hopper on 17/01/2023 11:57:53

Hello:

Edited By Julius Henry Marx on 17/01/2023 12:08:11

I did successfully improve the wear / bellmouthing on the jaws of the OE 4" Pratt direct mount chuck that came with my mid-60s ex-school Super 7 by employing a similar method to the Popular Science article.

I held a suiable length of 1/2" silver steel in the tailstock chuck & used fine valve lapping paste between the silver steel & the 3 jaw chuck jaws. IIRC this required a fair bit of pressure on the 3 jaw chuck jaws to get it to cut & I pushed / pulled the tailstock body with the spindle rotating to move the silver steel in and out of the jaws. It took a while & a couple of recharges of lapping paste but, eventually, there was a fresh witness mark on all 3 jaws right to the front edge. Checked by gripping a piece of 1/2" SS, runout was with a couple of thou TIR close to the jaws.

I felt it was a worthwhile excercise at the time, but do have a new Chinese 100mm chuck & backplate to hand to fit at some point - the Pratt is the better part of 60 years old now & nothing lasts forever !

Nigel B.

Hello:

Posted by Hopper on 17/01/2023 11:57:02:

> That looks really bad.

I was rather hoping it was not too bad.

> ... won't fix that much wear by lapping. | ... stick a feeler gauge in there and measure how much ...

I haven't tried yet.

> ... may be the chuck body is distorted.

That I can check with a comparator or test dial but it does not look distorted.

> ... could try grinding it with a Dremel tool mounted in the toolpost.

I am planning to make one of those this month.

> ... clover leaf plate to push the jaws outwards ...

Don't have one but I have seen other methods to get that done but I don't know how efficient they are.

Thanks for your input.

Best,

JHM

Efficient enough to totally eliminate your bellmouthing and get your chuck running true within .002", in my experience.

You can achieve the same effect as the cloverleaf plate by inserting three pieces of flat bar between the jaws. A hacksaw cut halfway down each piece gives it some spring.

Your bellmouthing will not be caused by the scroll plate or the teeth on the jaws. It will be jaw wear, jaw distortion or chuck body distortion.

Hello:

Posted by Hopper on 17/01/2023 11:57:02:

> Can you stick a feeler gauge in there ...

Yes. I did that earlier today, we are talking about a ~ 0.05 gap, maybe less as it is not uniform on all three jaws.

Had to measure the gauge with the micrometer as the markings have worn off.

> ... may be the chuck body is distorted.

I took the chuck apart again and mounted it on the spindle, as usual but without the scoll/jaws.

I then set up the clock and after cleaning the front surface of the chuck, set the tip to the outermost diameter of the body and set it to nought.

The variation with other points (90°/180°/270° was randon and at the most, less than half a marking (0.01), so we are at ~ 0.005mm, which seems reasonable (?) for a chuck of this OEM quality and age.

Can I conclude that now it is just a matter of grinding the jaws?

If we can conclude that it is just jaw wear, it will be fixable once I get the Dremel tool holder made or get the necessary stuff to try the other method which is slower.

And because of that, maybe less prone to error.

Thank you very much for your input.

Best,

JHM

Edited By Julius Henry Marx on 17/01/2023 18:40:09

It might be a good time to think of a replacement chuck, small ones are not expensive and someone should be able to advise which would or could fit.

Edited By old mart on 17/01/2023 18:55:00

I made the petal plate when the original article was published and ground two 3 jaw chucks which have so far been absolutely brilliant, and in my o[inion the better option than an import. it has been loaned out several times and loaners are well pleased with the results, plus another skill learned.

Interesting as this is something for me to bear in mind with my ageing machine-tools.

The petal-plate, or Hopper's separate plates, are new to me. I had previously seen chuck-grinding described using only a simple ring gripped by the outer jaw steps, so presumably these are an improvement on that, or overcome a problem with it.

Could someone please explain this point?

Putting a ring around the outside steps at the outer end of the jaws then tightening the jaws outwards against it has the effect of pushing the ends of the jaws inwards, meaning when you grind them, you end up grinding more metal off the end of the jaws where the bellmouthing is the worst -- the exact opposite of what needs doing.

The clover leaf ring -- or the three strips of flat bar -- restrains those outer jaw ends as you tighten the jaws inwards against it, duplicating the forces in play when gripping a job in the chuck. The bellmouthed outer ends of the jaws are pushed outwards, so the metal removed by grinding is in the smaller diameter area up inside the chuck body. You end up with nice parallel jaws that way.

JHM: It sounds as though your problem at .05 is simple bellmouthing wear/distortion of the jaws. Not uncommon and fixable with the Dremel method. Let us know how it goes.

Hi Julius I bought a chuck with the jaws out by 0.020" and that ground very well using the clover leaf plate and toolpost grinder and after over a year still grip true within a thou. I also ground the outside of the jaws by using a couple of rings on the outside of each step whilst I ground the other step.

https://youtu.be/2VKR45MYVEw

David

Hello:

Depends.

I had a very hard time finding a 4-jaw chuck for my Unimat 3, prices were all across the board with very little options to choose from.

The fact that used OEM Emco 4-jaw chucks for DB/SLs as well as U3s are still going for absolutely silly money so many years later means that suitable after-market alternatives have never really become available for U3 owners and a thorough on-line search confirmed it.

Sherline markets a really very nice looking one which would have set me back a cool US$131.00 plus shipping which in my case would have been a total of US$175.00 + import taxes on that.

On the other end of the spectrum were the rather shabby looking ones from Asia (India?) available for ~US$65.00 which I decided not to risk my money on.

There was nothing in the middle as other OEMs, who manufacture good quality chucks for very competitive prices (eg: San Ou), do not make the type of chuck the Unimat 3 needs.

I finally found an OEM Emco on eBay for a total of ~ US$150.00.

Thanks for your input.

Good to know it works well.

I am now seeing about getting one laser cut in 1/4" hot rolled plate, just have to make sure the dimensions are correct as this is for an OEM Unimat 3 chuck which is a small thing as chucks go.

Next comes the Dremel holder tool.

Thanks for your input.

Best,

JHM

Edited By Julius Henry Marx on 18/01/2023 11:21:31

Hello:

Happy to hear that.

I am thinking of getting a nut to turn and insert on a laser cut piece to make the Dremel holder but am having difficulty getting the thread spec right.

My Dremels are rather vintage, the best one is a 150W US made Moto Tool "Model 328 Series 1 07" and the other one is newer albeit much lesser quality 125W "Model 395" made/assembled in Mexico.

FWIW, all the US made 395s I have come across seem to suffer from the same quality issues this one had. eg: rotor blowing out after eight months use, speed selector breaking down after a couple of month's use, plastics and rubber coverings deteriorated in a year's time, etc.

I can put on an M20x2.5 nut on both of them (at least that's what the hardware chap said it was) but it seems a strange spec for a US made/designed tool.

My caliper says it is ø 19mm and my thread gauge says the pitch is 2.0mm, which would mean it is 3/4" but I cannot figure out the pitch. Could it be be 12.7?

Anyone here have the right specs for these Dremels?

Thanks in advance.

Best,

JHM

Edited By Julius Henry Marx on 18/01/2023 12:03:22

I had this bell-mouth problem on my 6" chuck. A member of my model engineering club had donated a 'kit' he had made for truing 3", 4" and 5" chucks. When I borrowed it I found that, naturally, the 5" kit was too small for my 6" chuck. A swift phone call resulted in a package containing the missing 6" device, which he loaned to me.

Each kit comprises two components: a device to allow the chuck jaws to be partially closed under tension with no backlash, and a diamond lap that is held in the tool post. The lap a small piece of diamond impregnated matrix silver soldered onto a bar rather like a boring bar.

The chuck is closed as far as it will go against the tensioning device and the lap is fitted to the tool post. The lathe is then started and the lap positioned until it is just touching the 'lowest' part of the chuck jaws, when a faint chirping noise is heard. As the lap is slowly withdrawn the chirping is reduced, which is a clear indication the the chuck is bell-mouthed.The cross slide is adjusted and the lap moved left and right until the chirping is constant right along the chuck jaws. It took about 10 minutes to true my 6" chuck. No sign of bell-mouthing now.

Here's a photo of the complete kit of four devices. I didn't notice that the diamond 'flag' wasn't visible in the large lap, but it's just like the two smaller ones, only bigger.