Elliott 3 inch Dividing Head

[RobCox]

This arrived courtesy of DPD the other day:

I bought it off @skotl. My hope was that as it is a 3" dividing head, rather than the 7" I already have, it would be a bit smaller and fit on the shaper a bit easier.

Wrong! Having taken it to bits to start a full refurb, I offered up the base casting to the 7" mounted on the shaper. Turns out that the head that this belongs to

is exactly the same as this

Swing being the difference, oh and ignoring the extra 1/2" . Never mind, it's going to get the full treatment before I sell it I guess. Don't need two.

So work on it has begun. I stripped it to bits, except that the chuck was stuck fast to the spindle, really tight, even after cooking it with a propane torch, gripping the spindle in my bench vice and giving the backplate a few whacks on some screws placed in the chuck mounting holes (until one screw sheared off and I gave up). I resorted to the lathe, parting off the backplate close to the spindle:

I managed to avoid the spindle itself and once I'd broken through the backplate came off easily.

This is the collection of bits after disassembly, after a wire brushing and a bath in white spirit. It was filthy:

I started on the chuck first, which I'll probably keep as its in reasonable nick apart from the "clumsy b*****d" marks. More wire brushing, filing the dings flat (it's not the same shape as an anvil, so why has it been treated like one?), removing burrs and oiled (nearly said "re-oiled", but it hadn't seen any in a long time) it looks like this (with the old screws for comparison):

 

[RobCox]

And so on to today's little job, fitting the body into the base with two new screws I'd made. Except it didn't fit! Turns out that one of the original long screws was missing because the holes didn't line up properly and one of the securing straps wasn't original. Only one of the straps fitted. This is one of the holes in the replacement strap offered up alongside the base:

You can see the bushing has been drilled off centre. Turns out the bushing was very eccentric:

After a bit of pondering I decided to plug the holes in the ill fitting strap, redrill, tap and counterbore the holes to make it fit properly. I threaded the clearance hole side of the strap to ensure the plug I fitted would stay put, loctited to be sure, and milled the plugs flush with the mating surfaces of the strap:

One end of the base has a through hole so I used a transfer punch to mark the new hole centre that end. The other end of the base has a blind threaded hole so I made a transfer screw from some drill rod, hardened the point and ground it sharp:

With the body of the dividing head on the base I could then correctly line up the strap and mark the new hole centres, miles off from the original holes as the plug outlines show:

After being set on the mill, drilled, tapped and counterbored here is the strap:

I also milled off 12 thou to ensure it would actually clamp the body. This is the strap properly fitted, with new (concentric!) bushings lining it all up:

I thought this was going to be a straightforward refurb. Clearly not.

Rob

 

[skotl]

Good work, Rob! I'm glad you got it apart but I'm even more glad that it wasn't just sheer incompetence on my part - it really was stuck!

I think the really depressing thing about the poor thing is that it came from an industrial training centre and, if that is the state they let their kit get into, you have to wonder how good the instructors were. This was the auction that we all followed for a few weeks, and there were a lot of mills, lathes, etc so goodness knows how well kept they were.

Glad you're making progress - it's clearly in the right hands now

 

[RobCox]

I've been working intermittently on this. The progress to date is:

A new locknut for the spindle bearing arrangement. I could have used the original but it had been severely abused and was missing its grubscrew. This is the new one before I cooked it gently with the propane torch to give it a blued appearance like the original:

The little soft copper washer stops the spindle threads getting mangled. The original had one too.

I got the burrs cleaned up on the worm wheel. This is a split design so backlash can be minimised. The bolts that hold the two halves together were a bit chewed too. I cleaned them up with a file and heated them and dropped them in oil to make them look like they would have done originally:

This is the spindle arrangement - if you're dismantling one of these dividing heads this might help you with what you'll find:

The thrust washer next to the worm wheel has a smooth and rough machined side and needs to go in the correct way round.

At this point I assembled the spindle into the body casting, which is a fiddly job as everything goes in the access hole at the sideand must be slid onto the spindle as it is fed through the body. After I'd got it adjusted I realised I'd not assembled the universal drive idler shaft . This is impossible to fit with the spindle in place so out it came again. The idler shaft holds the detent which allows or stops the universal feed from moving the dividing plate. This shaft had also received a beating so I skimmed off a couple of thou on the lathe. The anti rotation pin had also been previously sheared off so I drilled it out - breaking a 3.3mm drill in the process. To remove the debris I rotated the shaft 180 degrees and drilled a 2mm hole the other side and used a masonry nail ground down to fit this tiny hole as a punch to drift the detritus out. I then turned a new pin and fitted it:

This is it up close being slid into the body of the div head. The nut which holds this in place goes through from the access hole and is similarly fiddly to fit:

This shaft captures the idler gear when fitted:

Then the spindle goes back in:

This photo was from the first fitting. Before final assembly I lined up the two halves of the worm wheel teeth and snugged the bolts while I could get at then. These need a 1/4" AF spanner which I don't have.

Then the worm shaft was cleaned and fitted. It's going to get new locknuts as the ones that were fitted looked really shabby, but that's still WIP. Two views on the state of play so far:

 

[skotl]

Fantastic work, Rob - sounds like it was really a basket-case.

The bolts that hold the two halves together were a bit chewed too

Yeahhh.... that would have been me...

 

[RobCox]

Next instalment...

The worm shaft now has two matching locknuts. As with much of this abused dividing head, I'm taking advantage of having a twin with all of the bits intact, so these match what's on the twin:

Again, they've been heated and quenched in oil to colour them. Next comes the spindle lock. The whole handle was missing and had been replaced by a 1/2-13 UNC SHCS. Here's the twin and the new one:

The original looks a lot rougher than the new one here. I fitted a brass plug which bears against the spindle:

The spindle lock threads in the body are as rough as a badger's backside and are a sloppy fit on a nominal size thread, so I cut the thread a few thou oversize so it's a better fit.

Next comes the plunger that locks the dividing plate when it's not being driven from the universal shaft. The retaining screw was missing - it had probably been intentionally removed as the plunger had been bashed so when I'd made a new screw and fitted the plunger locked solid - so after the screw came a new plunger too. The screw - this is a 10-24 thread and feels like micro-machining for me:

The abused plunger:

And it's shiny replacement fitted. The pin is a press fit. To ensure it wasn't crooked I turned the 3 1/2deg. taper on it once it had been fitted to the plunger:

The dividing plate has been cleaned up along with the sector arms. The dividing plate was only held on with one screw and the only 10-24 csk head screws I have are a bit long. When I cleaned the sector arm locking screw I found I'll have to make one of these as well. It's a 2BA thread:

And finally, the state of play so far:

That's all folks!

 

[RobCox]

A bit more progress. I've remade the entire indexing arm assembly as the original was beaten up, worn and mangled and it was easier to start from scratch. This is the wear on the back of the indexing arm:

I could have milled it a bit smoother, but other parts of the assembly were worn badly, so the whole lot got re-made. Fortunately I had the other working dividing head to compare things to. The wear above was caused by incorrect assembly. I put it back together in the same way it came apart, and when I tightened the nut that secures the indexing arm onto its shaft, the whole mechanism bound up. Here's the cause:

Dividing head undergoing repairs:

Working dividing head:

Notice the shoulder on the spindle doesn't protrude, so the arm ends up tightening against something that's not supposed to rotate! I took the worm shaft out, modified the spacer positioning and thickness, and all was well. Here's the new indexing arm assembly components in the same shot as the knackered ones:

And assembled onto the dividing head:

The sector arms need attention now. I've got to re-make the locking screw, and they rotate too freely. That'll be the next part of this job.

 

[RobCox]

I got the sector arms sorted today. First job was to find a donor washer, mill out the ID to 1.000"+ and thin it down to 15 thou so it would cause the circlip to hold the arms tighter against the dividing plate and not flop around. Fun and games on the surface grinder thinning it down as the poles on the mag chuck are widely spaced and wouldn't get enough of a grip of the washer, so I ground a piece of scrap flat and used some green loctite to hold it on to that. Not the most accurate but it worked. Bit of acetone and a Stanley blade to free it - job done!

Mixed results. Everything tightened up, so I tried removing the circlip, which of course you have to be able to do if you want to swap dividing plates. 15 minutes of bad language later, I'm left with the conclusion that this circlip is another bodge that this div head's been subjected to. Time to replicate whats on the other div head.

The clip on my original was about 22 thou thick. Closest shim I had was 15 thou, so that had to do. Scratch the outline of the existing clip on the shim, roughly tinsnip the outside and as much of the inside to shape, then attack it with the fine CBN wheel on the outside and a diamond burr on the inside (I have an Oregon chainsaw sharpener which takes 3mm shank burrs and runs off 12V and is perfect for this. Here's the result after a cleanup, original for comparison:

Next give it some curvature so it will push against the arms and give the required feel. A piece of 2" bar and a soft blow hammer did the trick here:

Then off to the propane torch. I got it red hot and quenched in oil (I had no idea at this point if the shim contained enough carbon to get hard, but it seemed to). Then a polish with 180 grit, back to the torch and warm it (easier said than done when its only 15 thou thick) to blue, so its not glass hard. Here's the div plate with the spacer washer (which I left in), the sector arms and the old and new clips:

I put it all together. Result! A shade tighter than the other div head, but moves easily and stays where it's put. I fitted the replacement securing screw for the arms, that too needs re-making as its too thin and as I pulled the 2BA die up to the back of the head one of the teeth of the die gave up the ghost. I might have overdone it but the die was a cheap carbon steel one with huge gullets. My excuse anyway.

One step closer to being finished.

 

[RobCox]

More progress on putting this back into a fully working condition. Firstly a nice shiny locking screw for the sector arms:

The arms need to come off again as the top one has a burr which, when removed, will let the locking screw sit flatter. The nameplate which covers the internals needed some new screws. I looked for some #8-32 UNC round slotted head ones, couldn't find any so made some instead. I've cut out and filed to shape a new plate which may or may not replace the original depending on whether I can successfully print the legend on it:

I'm going to try laser printing the image onto toner transfer paper, transferring the toner to the aluminium plate, then using a polyurethane varnish to protect it. I'm hoping the technique works as I could do with a speed plate for my M300.

And so on to the universal/differential drive. The shaft which carries the helical/spur gear pair has several issues - the thread is galled and mashed up, the register behind it which fits in the casting is worn or is a few thou undersize and needs to locate properly to get the spur gears to mesh nicely, and finally the anti-rotation key has been mashed up:

It was easier and would produce a better result to simply scrap this and make a new one, so that's what I did. I cleaned out the locating hole in the casting and made a tiny round key to fit in the shaft:

Sorry for the blurry image. The 3mm end mill in an ER32 collet didn't have enough clearance to reach so at the top of the pic you can see the home-brewed endmill holder to give it more reach. The new shaft sits snugly in the hole and the pin stops it rotating:

And on to the last shaft, the universal drive input. Somebody's beaten this up too:

I used a hammer and punch to try and move the metal from the dings in the shoulder back into place, then skimmed a couple of thou off each of the shoulder surfaces to clean them up. The 0.750" 20TPI thread I picked up and cleaned out on the lathe:

The other end of this shaft has a thread for a large headed screw to secure on the input helical gear. The screw of course is missing, although the remaining debris of unfitted parts probably contains the bodge that replaced the original. The overall assembly on my other working dividing head looks like this:

And as luck would have it, the thread on the shaft of the head being restored doesn't match the other one. The other head is #10-24, this one I found a 0BA tap fitted without removing too much material. Definately 0BA, not M6x1.0, as an M6 thread fits but is sloppy. And this is the assembly now, needing the screw slotting:

A few odds and ends to clean up now to finish the mechanical restoration, then it's all coming apart for the cosmetic job on the castings, primer, minimal amount of filler, primer and enamel, colour TBD at the moment.

 

[skotl]

That's gonna be better than new when done - great work!

Are you going to keep this one then sell your other one?

 

[Agroshield]

When you have finished, could you gather together all the mangled bits and make a 'photo of shame' for all to see?

These days, you can find screw head proportions online if you need to replicate an original (e.g. https://torqbolt.com/asme-b18-6-3-s...ne-screws-dimensions-standards-specifications).

 

[RobCox]

A couple of odds and ends have been finished today. The helical drive gear fixing screw has now got a slot. I haven't made a key to stop this gear slipping on the shaft yet:

The helical/spur gear shaft now has a larger diameter nut. The previous one was only just bigger than the hole size and the edges of the hole in the casting has started to deteriorate. Also, I've made a square washer and fitted a screw to the eccentric sleeve which adjusts the engagement of the helical gears (both bits were missing when I got this dividing head):

Finally, I've got some paint on order, so time to take it all to bits again!

 

[RobCox]

Are you going to keep this one then sell your other one?

This one isn't as smooth as the original one, despite all my efforts, so I'll keep my original.

When you have finished, could you gather together all the mangled bits and make a 'photo of shame' for all to see?

Oh yes. They're in a pile on the workshop floor at the mo. They'll make a nice little horror show!

 

[RobCox]

I've worked a bit on the universal shaft with some slip stones and skimmed a couple of thou off the inside face of the shoulder on the shaft and it now rotates quite smoothly. Happy with the mechanical restoration, time for some cosmetic improvement. There are four cast parts to paint, so I dismantled the head, scraped all the old paint off (not that there was that much to remove), then gave the castings a bath in white spirit to degrease them:

So this is the starting condition. Then a coat of grey phosphate primer:

Its still wet here, but the grinding/fettling marks on the casting don't really show through so if it looks good when dry I'll skip adding any more filler.

Some advice required please. The paint I ordered, Boxford Water Blue, looked quite dark on the PC screen, certainly nothing like the colour when I took the lid off for a peek...

Bit bright. More Axminster Tools blue. The question is, would you use this?

 

[Seadog]

Dove grey for my Boxford (I think)

 

[RobCox]

Progress has been slow, but I have managed to put some lipstick on the castings:

This is after the second coat. The main colour is going to be J&S surface grinder blue, as suggested by @bR39Da. I couldn't bring myself to paint it Axminster Blue (which is what that other paint looks like). This dividing head has already suffered enough abuse...

It only came with one dividing plate, plate A, so I checked what the going rate for a used plate would be, something around £50, so I figured I could make one for far less as my time is effectively free, given this is a hobby. I got a chunk of 140mm dia EN1A and took a slice off with my bandsaw. Apparently the capacity of this saw is 125mm for round stock, but it managed this in 3 cuts ok.

I have a stack of shims the width of the sawcut to put in the kerf for the last cut. The cuts even lined up well, so the saw must be aligned reasonably:

I cleaned it up and reduced the thickness to 0.325" on the lathe, made an adaptor to fit a backplate on the other dividing head, which I will use to do the indexing, rather than XY coordinates on the DRO, as I'd still be doing it at Christmas. I've got as far as doing the first ring of 16 holes. Only another 250 holes, plus the one on the back, to go.

 

[Milkybars]

@RobCox whats the plate dimensions and PCD on the holes ..I have a few orphan ones knocking about !

 

[RobCox]

Plate dimensions are 1 inch bore, 3 csk holes on a 1.450in pcd, inner ring of holes on a 2in pcd, holes are a tight fit on a 3mm drill shank, 16,17,19,21,29,33,39,43,49 hole circles, plus a hole on the back to lock the univ drive when not differential indexing or helical milling. The outside of the outer ring of holes touches a 5 inch circle, so the plate is an 1/8" or so bigger.

I should have asked before I started drilling! The going rate on ebay seems to be around £50.

 

[Milkybars]

@RobCox None of the ones I have meet them dimensions !