Inside Doc's Machine Shop

[DocsMachine]

I, ladies and gentlemen, am an idiot. I know this doesn't exactly come as a surprise to many of you.

In this particular case, as has happened all too many times in the past, I got too focused on one "solution". Focused on trying to execute that one idea, I didn't bother thinking of other options.

My problem was I needed a way to hold my parting tool further out than my basic tool post could. I had an extension, that didn't work. I had a different tool holder entirely- that turned out to be slightly too short, and oriented the tool in the wrong direction. I hit upon making an extended tool block to fit my existing tool post, and as already recorded for posterity, was stopped by a lack of suitable tooling.

This morning, in the sober light of noon or so, I realized that yes, indeed, I was an idiot. There was a way I could do it right and do it fast.

So, I went to my metal supplier, and bought 5" of 1-1/2" (38mm) square steel stock.



After a quick deburr in the belt grinder, I clamped it end up in the vise, and gingerly cut a 1/2" square notch in the end.



That got lightly chamfered for both looks and to reduce blood leakage...



And then drilled and tapped for 5/16"-18:



The soon-to-be underside got a notch and a flat to make a "key"....



And then two carefully-measured and located bolt holes to mate up with the 'key' on the underside.



The bolts wound up just slightly long, so I threw a collet in the little Hardinge and shortened 'em up by about a sixteenth. (21 millifarthings.)



The whole thing bolts on, nice and solidly, like so:



Including the trip to the metal shop, the whole process took roughly three hours. (There was a lot of fiddling, fettling and fitting I didn't show. )

Anyway, I finished up programming the part, and took 'er for a spin:





You can see the "parted" piece falling towards the back, at the bottom center of the frame.

And, since I had the raw stock already cut and ready, I ran the first two pieces, for a total of 24 parts so far.



I'll be able to turn more over the rest of the week, which will put me- at least on this one task- three or four days ahead of where I'd be if I'd just waited for the cutter for the other block.

I'll still finish up the other block- it'll have its uses- but at least I can make a little quicker progress than I'd anticipated.

Doc.

[DocsMachine]

With all the stuff I have going on at the moment, it's time to wrap this little project up, if I can, and get it out of the way. Projects are fun, but there are times when it's better to have a working machine, than to be working on the machine.

In last week's exciting episode, our heroes had just made a new disc to act as the spindle lock. The next step was to bore the pin-receiving holes in it, for which the first step was to mark them. I gutted the locking-pin assembly, and used just the collar and a transfer punch, to lightly mark the OD of the disc.



I then set up the dividing head on the green mill, found the axis center, and located the punchmark as close as I could, with a point turned in a 1/4" steel rod.



The idea here is to try and keep some semblance of balance to the new disc- so I needed an easy way to make sure the two holes are exactly 180° from each other- within the limits of a budget-priced Japanese dividing head, anyway.

A simple set of the quill travel stop also assures that both are the same depth. I pilot drilled them, and then bored them smooth with a 3/8" endmill.



After that, the disc needed a keyway. I'd ordered and received the broach I needed, but I didn't have a guide bushing. So I found a likely-looking section of 3" round 'looneymum (or for you Brits, 'looneyminimum ) and turned it down 'til she fit through the disc.



Swapping back to the mill- which at this point desperately needs a cleaning- I located center and milled a groove of the proper width and depth to accept the broach.



Moving over to the hydraulic press, I carefully broached the keyway, trying to keep it 90° to the two pin holes.



Et Voilá! (Which, I have on good authority, roughly translates to "Hey, check this sh*t out!" )



One last-minute thing: Because the collar is in under a casting, a proper setup is to provide an easy way to remove it- in this case, since you can't get a gear puller in there, a pair of threaded holes lets you grab it with something like a steering-wheel puller.



Again, centered and evenly spaced, to try and maintain a decent balance.

This slips in like so, along with a new key (since the one I'd made before is now too short.)



The original pulley goes over that....



And finally the locknut and it's retainer.



But, we're not done yet!

The locking pin assembly needs to be reworked. It had a spring-loaded detent, that both held it up out of the way, and held it down in the 'locked' position. That meant it was possible to lock it... and then forget to UNlock it. This DC drive setup has a fair bit of low-speed torque, and it's entirely possible that doing so could damage something.

Point in fact, I suspect that's why the factory lock-disc was missing- either some operator kept forgetting it was on, or to keep such an occurrence from happening.

So I set about redesigning it a bit. I found a chunk of 3/8" round stock in my scrap bin, that turned out to be 304 stainless. Not ideal from a machining standpoint, but it was a nice, clean, accurate bar, so I went with it.

It got turned to length, grooved for the snap ring, and one end threaded for the original ball-knob.



I then found a suitable spring in my Suitable-Springs Stashes®, and assembled it like so:



And that assembly simply tightens into place just as the original did:



Where it, of course, works like a champ.

The spindle is now back up to 100%. We just need a motor. And the collet closer. And the turret still needs some fiddling, and...!

Doc.

[Cal440]

Nice work as always Doc,the finish line must be close, you can make it!

[DocsMachine]

NOW we're getting somewhere!

I rolled into the shop about two in the afternoon, having had some prior engagements on my schedule, but with the intention to finally get the motor back into place and wired.

I'd gotten the motor back from the shop last Friday, but was unable to clear enough time to do much with it. They did a great job, and while it was a tad more than I was hoping to spend, they gave it a full workover, including new bearings.

But before I could attend that, I needed to patch up the flexy conduit, and double-check the wires in there were okay.



Turns out the only real damage was to the outer plastic sleeve- which was old enough it cracked when bent too far. The inner steel armor held up nicely, and the wires were fine.

The ends are older, and likely obsolete, fittings, but I was able to extract the sealing ring from the bad end, bandsaw off the 3" or so of damage, and carefully resinstall everything.

But first, the motor!



New bearings, dipped and baked rotor, new field and armature wires, and they checked the comm and brushes. I didn't know paint was on the agenda, but I ain't complainin'.

I cleaned the junction box, and snipped a sort of "gasket" from some of the leftover tool-drawer liner...



And fixed it in place with fresh bolts and washers.



Bolted on the pivot/hanger...



And with remarkably little cursing, got it hung up, the belt installed, and the tension adjuster bolted in place.



Since the spindle pulley was now in it's final resting place, I was able to slide on the jackshaft pulley, and with the aid of a piece of string and a weight, align it to the spindle and lock it down.



This one's a taper-lock pulley, likely an aftermarket, as the few diagrams I've seen online- including the paper one glued inside the door- show a stepped pulley. Two sets of two pulleys, to give two different speed ranges- one normal and one slower, for more torque.

I had again used a buddy's heat-shrink printer thingy, and made a new set of ID tags for the new wires.



The shop had put their tags out at the end of the long wires- not unreasonably- but in my case, I wanted to snip about 4" off each one, so they'd fit better inside the junction box. Anyway, simple matter of double-checking, sliding, shrinking and snipping.

I similarly relabeled the wires in the conduit, slid that into place, and got it connected up at both ends.



With the wires wire-nutted in place and in proper order... it was time to test this puppy!



And it works like a champ! There's a touch of vibration, probably from the big iron step pulleys, but plenty smooth enough, and at this point, with just the motor and jackshaft, damn near silent. The motor-generator pair makes more noise.

Throw the two spindle belts on...



And take that for a spin:



Smooth, quiet.... A significant difference from the as-delivered sound level. The original setup didn't "howl", but as noted earlier, it made more noise than I thought it should.

All this was a nontrivial extra expense and expenditure of time, but it was worth every penny. I'm actually wondering now, if the motor-generator pair couldn't stand a set of bearings...

All the way up to top speed- in this range, about 1600 RPM- and still nearly no noise save for the M-G setup. I mean, it's not library-silent, but the fans on the VFDs on some of my other machines, make more noise than this spindle and drive.



And finally, the back covers can go back on, more or less for good at this point.



Over the course of the day, I snuck in a few other little fiddly bits as well. One item was the electrical enclosure had no latch- it used two bolts to hold the door shut. I'd imagine that's in part to keep operators from poking around in there, but in my case, I didn't like having to get a wrench when I wanted to fiddle with it.

So I found a hunk of 1" 'looneymum, gouged a curvy bit into it, drilled and tapped it...



Gave it a quick knurl with a handheld knurling tool...



Parted it off, faced and chamfered it...



And somehow ended up with two of them.



These simply go in place of the bolts, and let you open the enclosure, if necessary, by hand. There's a gasket on the door, to give some 'crush', so the knobs are unlikely to vibrate out.



I also bolted the back cover... er, back on, including adding a middle bolt at the top edge, as the cover tended to "oil can' a little, and rumble slightly under vibration. I had to use the large washer there, as there wasn't enough depth to the cabinet edge to put a screw through both.



I may wind up putting a screw or two at the bottom, too.

And finally- finally!- I remembered to cap the sump drain.



I'm still not sure if I'll ever use flood coolant or oil- this thing doesn't have enough splash guarding- but I figured if I ever do, it'd be just my luck I'd have forgotten I'd never capped it.... after I dumped two gallons of oil into it.

Stand by, we're nearin' the end!

Doc.

[DocsMachine]

Heh. No, I don't yet have an X/Y slide, I haven't mounted a chuck yet, or reassembled the collet closer. I figure I'm still at least a week from "first chips".
Doc.

[superman]

What's taking so long.... i kid, i kid.

[DocsMachine]

Money, and too damn many other things going on.

Doc.

[DocsMachine]

The last major subassembly: the collet closer!



The first bit is the handwheel, and it's notched... locking ring... thing.



The handwheel is to let the operator turn the spindle by hand, of course, and the locking ring is what helps keep the threaded drawtube from unscrewing from the collet. Fortunately, this unit only needed a good scrubbing and some paint.

One interesting bit was this key, which I'm not actually convinced is original:



The setscrew coming in from the outside is tapered, and wedges those two blades apart, locking it into the keyway. Odd indeed, but it works.





After that quick check, I split the parts, scrubbed everything thoroughly....



And gave the handwheel a quick squirt of gloss black.



While that was drying, it was time to look at the rest of the locking mechanism.



It had puzzled me for a bit, as this style of collet closer had no obvious "lock"- a way, as noted above, to keep the drawtube from unscrewing itself from the collet. Which could throw off dimensions, cause the work to come loose, etc.

Once I finally sat down and studied it, it seems this arrangement uses a sliding collar, rather than a small lever-shaped lock. The knurled collar in the middle of the assembly, pulls to the left, and unlocks two little studs from that notched ring shown earlier. The collar is spring-loaded, and has a detent to help hold it 'locked'.



One of the problems was that those little studs were missing.



I'm guessing that the larger holes (there's two sets, 180° apart) were the original studs, that long ago stripped out. And at some point, somebody drilled and tapped two more, for some 10-32 screws.

Either way, I pondered way to replace them, but we'll come back to that.

Continuing the disassembly, the worn-smooth aluminum handwheel at the very end came off easily, but had clearly at one time been keyed in place- the setscrew had mashed the Woodruff key slot into an unusable mess.



Fortunately, this wheel is just there to turn the drawtube while installing or removing a collet, so it doesn't see a lot of horsepower. I'm going to try cleaning up the ring and maybe re-knurling it, but probably not 'til later.

Removing the badly-damaged snap ring lets bearing carrier slide off- and I was happy to find that bearing, at least, was in excellent shape.



Less so, could be said of the two screws that retain the locking collar. I got 'em out, but they're badly buggered, and were very likely not original.



That revealed the inner locking mechanism, which thankfully just needed a good scrubbing in our favorite non-leaded solvent.



The bearing carrier assembly came apart easily- though you can see some ham-handed types have beat up this locknut, too.



The inner cone pressed out easily, and then, once the snap ring was out, the bearing pressed out easily too.



With most of it all apart, and thoroughly cleaned, it was time to do a little spiffing up. The bearing carrier took a ride in the other lathe...



And after filing down some of the more egregious nicks and dings, took a nice buff from some Scotchbrite and WD-40.



Which, Ladies and Gentlemen, marks the Three-Hundredth photo in this project series.

And we're still not done yet.... to be continued.

Doc.

[DocsMachine]

I can't put it entirely together just yet, but thankfully most of this assembly just needs a cleaning.

One thing I was looking at, was the back of the headstock- the belt cover casting. There's these four little threaded holes around the opening...



That I suspect once held a thin sheetmetal cover. Meant, most likely, to be a splash guard, so oil, coolant or chips don't get down into the belt area, and/or drip on the motor.

With the handwheel in place...



There's a gap there, about 1/16" wide. So the cover was probably something like 18 gauge. I'm toying with the idea of making a new one... If I ever do run flood coolant or oil on this thing, it'd help keep it off the belts.

Anyway, after the wheel had had a chance to dry a bit, I bolted it and the locking ring thing back together, and clamped it back in place.



For those two little lock studs, I decided to keep it simple- I turned a pair of stainless button-head screws to just .350" OAL, degreased everything, and screwed them in to the 10-32 holes with a touch of blue Loctite.



Those mate up with the notched ring like so:



With that done, I used a little grease to stick the three springs in place, so I could slide that collar back on.



That round hole in the center is where the detent ball goes.

To replace those buggered screws, I took two more buttonheads, turned about .020" off the edge so they'd fit cleanly- and still retain the collar- slid everything more or less into place, and snugged 'em down.



I slid the whole mess back into the headstock to finish the assembly. Note the empty Woodruff key slot- that keeps the locking 'cone' from trying to spin on the drawtube- which it's not designed to do, and could damage both parts if you let it do so for too long. That key was missing, and probably has been for some time.



I had a replacement key, but it needed to be cut down quite a bit before the parts slid together smoothly.





Unfortunately, the key notch in the drawtube was bigger than the keyway in the sliding cone- which, for anyone who's used Woodruffs will know, means that allows the key to "rock" as you try to slide the parts together. Once it rocks upward, you can't assemble the pieces.

I finally managed it, but it was frustrating- and I probably didn't assemble the whole thing in the factory-specified manner.

Anyway, with that fitted, I of course had to slide the 'cone' back off, and assemble it into the bearing carrier.



Aaaand go through that little but of frustration again, this time with the carrier's dust shield making the key all but inaccessible.



But, after all that, here it is fully assembled (save for the actuating arm) and "locked".



For the first time in the nearly two months (!!!) I've had this thing, it has a collet and a workpiece properly locked into place.



Stay tuned, we're on the home stretch!

Doc.

[DocsMachine]

We're damn near done, ladies and gentlemen! At least with the major bits, anyway.

I had to make a parts run this morning, and among other things, stopped to get a new section of flexy conduit for the pump connection.



The old stuff was getting brittle, and I didn't feel like trying to scrape off the old green paint. I was again able to save the old-style conduit ends, and it just took a little finagling to get it all back into place.

It's not actually wired to anything, as I don't have a source of 3-phase on this machine apart from the VFD, but as I've said before, I'm still not sure I'll even use flood on this thing. At least not right away.

Anyway, after that... it was finally time to finish the collet closer, the last major assembly in this project.

The last piece was the yoke, a casting with the actual handle the operator... er, actuates. In this machine's case, the casting had been damaged at some point in the past, stripped and repaired, and the handle itself was missing.



I suspect the original setup had a short rod as a handle, with the typical black phenolic ball-knob, and which had at some point become stripped out of it's threads. Or some damage occurred and the casting got cracked. Either way, somebody had repaired the crack via brazing, and also welded on an extra nut to give the threads more support.



The braze seemed solid enough, and I saw no reason to try and re-do it, so I just ground on it a little to smooth it up slightly, and then lightly sandblasted it, to get the last of the old paint off.



The trick now was to mount a handle. I'd picked up a "long teardrop" shape phenolic handle on one of my earlier McMaster orders, so I just needed to make a rod that goes between it and the casting. I rooted through the junkbins and found a suitable hunk of 7/8" hot-rolled mild steel.



One end I threaded to 5/8"-18, which is what the yoke casting had been tapped to. (I have no idea if that was the original thread- I suspect not.)



The other end I turned down and threaded to 1/2"-13.



That gave me my mounting shaft, but it still needed a little something.



So I turned a "wasp waist" into it, with a couple of smooth curves to the ends. 100% for looks, but hey.



Then, using a square collet block in the mill, I cut two wrench flats on the "fat" end.



Which of course lets me firmly tighten it into place.



After that, I took it back apart, and masked and painted the yoke.



The last bits of this little job were to cut down some long 1/2" setscrews, and turn a smooth 3/8" "dog point" on the ends.



Then, using a different piece of 1/2" threaded rod as an arbor, I thinned and flattened two 1/2" nuts to go with them.



These are the 'pivot pins' for the yoke to connect to the bearing carrier- which of course stays still (non rotating) as the spindle is turning, and lets the handle pull and push on it and the drawtube.



And done, ladies and gents!



Assembled, adjusted, lubed and 100% functional. It opens and locks smoothly, and there's only a hint of shake as the spindle turns. (That's common for this kind of closer.)

So what's left to do?

Just a few detail things, a clip here and a snap-ring there. I need to finish tweaking the turret, and at some point set up an X/Y slide on this thing. That won't be cheap, but it'll make the machine much more useful.

That and getting a first batch of tooling set up, will get us actually cutting chips!

Down the line sometime, I'll replace that wooden slide-out collet tray. The old wood is, I think, a little too thin, and I can make the pattern denser, to hold half again as many collets. I'll eventually make a tool rack for up top, for the often-used stuff, and eventually start considering splash guarding.

But... after almost two months- two months of time I really couldn't spare- she's damn near done!

And now... I really, really need to clean the shop.

Doc.

[DocsMachine]

Today was Shop Cleanup Day- probably the first of several over the next couple of days.

I've been going pretty much flat-out since at least May, and certainly since mid-June or so. And, as a result, the shop is a wreck. And I'm a slob, so that's saying something.

Tools, parts and leftovers are scattered literally everywhere, I have bins of in-progress CNC parts on damn near every available flat surface, and almost every machine is choked with swarf and dripping with oil. All the trash cans are full, there's extension cords and air hoses everywhere, and I have boxes and tools piled on some of the unused machines.

The floors look like there was an explosion at the glitter factory, all the workbenches are too crowded to work at, and thanks to the new machine displacing a non-trivial amount of shop space, I have a bunch of things, like the big vacuum, that no longer have a 'home'.

Much as I wish I had a nice, clean, well-organized CNC shop like you see on all those YouTube videos, that's just not in the budget.

So I took a chunk of day and tried to wring a little order out of this five-pound sack full of fifty pounds of packrattery.

Here, for example, was the Sheldon lathe at the start of the day:



And here it is about an hour later.



All the collets are back in the drawer, all the lube points have been lubed, and all the chipped inserts have been replaced, too.

I also cleaned out the chuck. It's an integral-back Bison, and was almost new when I bought this machine back in 2008. I have since put a bloody great lot of miles on it, and have strove to keep the chuck in as good a condition as I can- I don't overtighten it, I've never had a real crash, and I used to open it up, clean it, and regrease about once a year.

Lately it's been binding- a sure sign there's too much swarf in the scroll. And it occurred to me... I couldn't really remember when I'd cleaned it last. It might have been pre-pandemic.

So I popped it off of there, opened it up, and let it marinate for a while in Mr. Wizard's Amazing Lead-Free Universal Solvent:



Thankfully there's no apparent damage and no real wear, so I just scrubbed off the old grease, rinsed out all the swarf- and there was a lot- and blew it all dry.



After that, I reassembled it with some fresh grease...



And put 'er back on the also-freshly-cleaned lathe.

The whole shop is looking a little better- somewhat less like a fresh bomb crater than it did yesterday. Still have a lot of tidying up to do, and as always, I still need to find a 'home' for some of this junk, but like the cleaning itself, that's a neverending process.

Doc.

[DocsMachine]

One quicky little bit today: To get it up off the floor, I remounted the lever cross slide, and since the accessory bits looked a bit shabby, I gave 'em a quick coat of black.





That and a teeny dab of some RTV to seal a couple of gaps where the headstock casting meets the bed, in order to seal it against oil or coolant intrusion, is all I did to the Rivett today.

More later.

Doc.

[DocsMachine]

She's done enough for a "before and after" set.






(Edited background, of course. One was taken outdoors, the other indoors- I swapped the background to make them more comparable.)

Doc.

[Cal440]

Wow Doc, you do mighty nice work that is one sharp looking piece of equipment! Way to keep at it and over come all those obstacles. Your attention to detail is outstanding,
congratulations on a job well done.

[DocsMachine]

Thanks! I hope to have some actual product out of it before too long.
Doc.

[DocsMachine]

Sometimes I hate my brain.

I got an idea on how to make the 'dust cover' to go under the collet closer- an item I don't really need, that the machine works just fine without, and that I don't really have time for. But that rabid ferret in my head wouldn't let the idea go, so... of course I had to do it.

[sigh]

So, I stripped off the closer, and removed the back plate of the headstock.



And found a scrap of 1/4" aluminum plate, used the cover to roughly mark a blank and bandsawed it out.



That got chucked up in the freshly-cleaned lathe, the center trepanned out, and the OD thinned as best I could with that setup.



I got the size right...



... but it was getting too thin for a solid hold, so I needed to go a different route.

I plunked the rotary table on the also-freshly-cleaned mill, fitted my ratty old workplate, surfaced it, made and fitted a center plug, and turned that to size.



After that, it was easy to thin down the flange to the proper depth with a nice sharp 4-flute.



With that done, we needed mounting holes. The new plate would index to the round opening, so I just fitted short setscrews, slightly proud, to the four screw holes.



And a light rap with a rubber mallet made a perfect mark for each one.



The holes got drilled and countersunk, though not as cleanly as I'd have liked due to how thin and flexy the flange is- only about 50 thou. And, as I suspected, that also meant that the screws needed to be slightly less than 50 thou thick at the head.



And she fits, and clears- though just barely- the closer handwheel.



One thing I didn't anticipate was that the backplate's machined feature, isn't quite perfectly concentric with the spindle- which in retrospect makes sense, as they're separate castings and machined in separate operations.

Luckily, I'd made the spindle opening a little oversize, and despite being slightly offset, there's still clearance, and no rubbing.

With everything mounted up, I scribed the outline of the casting- which isn't truly circular...



And used the bandsaw to remove the bulk of the extra material, then the belt grinder to finish the edge to the scribe line.



A little filing, sanding and deburring, and she's basically done.



Installed....



And with the closer back in place and up to full speed.



No rubbing, no extra noise, I think it's good. And I can mark that little bit off the ol' mental checklist.

Doc.

[Cal440]

Nice job Doc on figuring out that final piece looks factory 🏭 👌