Dr Al's Latest Folly

[Dr.Al]

The next big job for the tailstock was to make the frame that connects the tailstock to the bed and ensures it stays parallel with the bed axis and at the right centre height.

I've decided to make the base of the bed out of some 20 mm thick hot rolled steel. This is a lot more substantial than it really needs to be, but it means I don't have to bother with weld nuts (the steel is thick enough to tap directly) and it gives me a lot of room for adjustment: I'll build the frame over size and then mill down the bottom of the base plate to get the centre height right. If I made it out of 6 mm steel and had to take 2 mm off to get the centre height right then the plate would be getting a bit thin.

I've got one piece of 20 mm × 100 mm hot rolled steel with some thick mill scale on it, so the first job was to clean off a lot of that mill scale (firstly with a non-woven preparation wheel and then with an 80 grit flap disc). I then cut a section off with the horizontal bandsaw:

To make the ends of the tailstock frame, I got a bit of 6 mm stock and cut it into bits using the angle grinder with a cut-off disc (it seemed a bit small to easily hold for use of the metal cutting circular saw):

I then cleaned up the faces with a non-woven wheel and then clamped the two pieces together and used a flap disc to make them the same shape:

With that done, I tack-welded all the base bits together:

I then decided to pre-heat them a bit with a MAPP gas torch:

I've no idea whether that was necessary or helpful, but I've never tried to TIG weld anything this thick before and I figured it couldn't hurt. If I had any sense, I probably would have stick welded it, but I enjoy TIG welding a lot more (and I'm a lot better at it).

Once I'd got bored of pre-heating, I turned the TIG welder up to the maximum current (210 A) and went for it:

By the end of the weld, it was quite hard to hold the TIG torch due to the heat coming off the tailstock base! I'll now leave it to cool down.

I said at the start of this project that I was going to try to do it without buying anything. Well, I've failed! I've got a couple of single phase motors that I could have chosen for the power source, but then an old 1.1 kW (1.5 hp) three phase motor popped up on ebay relatively locally. I was the only bidder, so I picked it up for the auction starting price £10. To run it, I need an inverter, but I found a 2.2 kW one brand new for £53, which wasn't too bad considering. Hopefully they'll both work! All being well, this should be much better than a single phase motor as I can more easily get variable speed operation.

It's a big old beast (and somewhat out of proportion to the lathe if I'm honest), but there should be plenty of power, even when running it at reduced speed.

It's a 4-pole motor and hence only runs at 1420 rpm. Ideally I would have gone for a 2-pole one for the higher speeds that seems to be preferable for spindle work on a wood lathe (from what I've read). However, as it's such a powerful motor, I should be able to gear it up with careful choice of pulleys and that should bring the spindle speed up to where it needs to be. If I can figure out a way to modify the spindle to have a 19 mm (3/4") pulley mount, then I'll be able to swap pulleys between the motor and the spindle and pick between a 2:1 ratio and a 1:2 ratio with only two pulleys. The variable speed drive should sort out the rest, so that seems like the ideal thing to aim for. The only challenge will be sorting out the spindle's pulley-mounting thread, which is currently M20×1.5. If I could reduce it to M18×1.5, then I could just skim off the existing threads and start again. However, the through bore is 15 mm so that might be getting a bit thin. What I'll probably try to do is cut it as M19×1.5 and just do my best to align the new threads with the old ones. That's a challenge for another day though...

 

[Dr.Al]

We've had visitors staying this weekend, so I haven't had a lot of workshop time. However, I managed to make some progress on the tailstock before they arrived and after they left.

The next job on the tailstock base was to clamp it in the mill vice and skim the top surface flat and level. I wasn't aiming for a particular height here, as long as it was higher than I would eventually want it to finish.

With that done, the top pieces were ready to glue together:

I clamped them firmly together, using some scribed centre lines to get the top of the tailstock roughly in line with the base and then tacked them in place:

I could then weld around the more accessible areas of the join.

Once it had cooled down (and the visitors had left) I gave the welds a bit of attention with the angle grinder. It could probably do with a bit more, but it'll do for now:

To work out the exact centre height of the spindle, I measured the largest diameter of a Morse taper 2 dead centre and shoved it in the spindle. I then used a height gauge to measure the height of the top of the dead centre (and subtracted half the diameter, giving a centre height of 128.6 mm):

I then needed to bring the tailstock centre to the same height. I clamped the tailstock upside down on the table and used a dial gauge, running along the spindle, to check for level. I added some feeler gauge stock underneath the tailstock, tweaking the size of the shims until it the spindle was level with the travel of the bed. In the end I had a 0.6 mm feeler gauge piece at the end and a 0.3 mm piece in the middle (the middle piece was there so that clamping the tailstock down in the middle wouldn't bend the tailstock).

Once it was levelled on the table, I also aligned it with the X-axis of the table:

All clamped up and ready to mill:

I used my big carbide face mill to mill the bottom of the tailstock. I had to take about 1.8 mm off the bottom to bring it down to the right height (which I measured by zeroing the height gauge on the top of the spindle, measuring the milled surface and then adding half the diameter of the spindle):

Once the base was milled, I used a edge finder to find the centre of the spindle in the Y axis...

... and then drilled and tapped four M6 holes and drilled a central 10 mm hole. The 10 mm hole will be used for clamping the tailstock to the bed; the M6 holes will be used for some screwed-in pieces that will (hopefully) align it to the bed:

The last job on the tailstock was to add another tapped hole, this time in the back:

That'll be used (if necessary) for a brass screw that will hold the tailstock in place if I need to prevent it from moving while the lathe is running.

With that done and a couple of bits of box section plonked haphazardly on the bench, it's starting to look a bit like a lathe:

 

[Dr.Al]

The next job was to get on with the part I'm calling the "tool rest slide". This is the bit that will slide along the bed and back and forth and support the tool rest in whatever position I want it. I'm making this as a sort of "t-slot" thing so it can be clamped down in any position. The easiest way I could think of making that (with the bits of steel I have available) is to fabricate it out of some lengths of 12 mm × 35 mm and 10 mm × 20 mm EN3B I had in the drawer:

Rather than welding them together and worrying about distortion, I thought I'd just use M6×50 socket head cap screws. That meant I spent a little while drilling holes, a little while drilling and counterboring holes and a little while drilling and tapping holes...

All the bits with lots of holes in them:

This shows the pieces half assembled to give an idea of how they go together:

Once they were screwed together, I put the assembly in the mill's vice and used a long end mill to skim the ends flat:

Once both ends were flat, I mounted the assembly to an angle plate and drilled and tapped some holes for some end plates:

Two more holes were drilled and tapped for attaching the bit that the tool rest will fit to:

For the end caps, I cut a couple of slices off a bit more 12 mm steel:

That got squared up and milled to size and then some more holes mysteriously appeared:

This shows where the tool rest will sit:

At 57 mm high, it's a lot thicker than it (arguably) needs to be, but with a 128 mm centre height, I'll still be able to turn spindles that are up to 142 mm diameter, which is way bigger than anything I can imagine wanting to turn. If I ever get the urge to try bowl turning, the tool rest slide won't be under the workpiece (I think) so I'd still be able to turn bowls up to 256 mm diameter.

 

[Dr.Al]

I had a hour or two free this evening so I popped out to do a bit more on the tool rest slide. Following advice on tool rest height adjustability, I've decided to abandon the two tapped holes I'd originally planned to use for holding a tool rest support in place. Working on the other end of the tool rest slide, I used a spot drill to position some (very carefully thought-out) hole locations:

The holes were then completed. The outside ones were drilled and tapped M6; the middle one was drilled out to 16 mm, finishing with an end mill for final sizing:

I then flipped the tool rest slide over and milled a clearance slot at the bottom:

The eagle-eyed among you will note that I've also added a small drilled hole in the end plate (and a matching one at the other end).

Next up was a bit of 40 mm EN1A, which was mounted in the three-jaw chuck, drilled, bored and reamed to 16 mm.

After parting off, reversing and cleaning up the back face, it was then mounted (in a somewhat precarious way as I was being lazy and only drilling, not milling) and the axes zeroed on the centre of the bore:

Three holes could then be drilled and counterbored:

I did those holes in a slightly odd order: I spot drilled them, drilled them to 24 mm depth, counterbored them, then completed the through hole. The reason for that was that the vice mount was a bit precarious (it would have been better to have a v-block between the part and one of the jaws) and I wanted to leave the parallels in place until the last minute.

Once those holes were done, I shoved a couple of drill bits in the holes and used them to align it in the vice such that a drilled and tapped hole would go roughly central between the two tapped holes:

It could then be mounted to the tool rest slide and tested with a bit of 16 mm silver steel bar and a handy cap screw (which will get replaced with a knob of some sort in due course):

It was starting to feel like I was getting close to the finish line, so I thought I would check my ambitions somewhat by writing down a list of all the parts I could think of that still needed to be made. There were 44 of them! I'm still a fair way from completion.

Admittedly, some of those 44 are really simple things like the handle for the tailstock handwheel, which is basically a tube. However, some of them are complex (and possibly made up of multiple parts) like the motor mount and the bed/feet assembly, neither of which I've really started to think about yet.

 

[Dr.Al]

I've been working on the tool rest slide mechanism gradually in amongst other jobs, but I thought I'd wait until it was mostly complete before posting the photos and build process as I figured it would make more sense as a fait accompli.

The first piece to describe is the part I'm calling the clamp block. This sits inside the "t-section" of the tool rest slide and can move along to any position along the length. When the tool rest slide is clamped, this piece is the one that presses down on the tool rest slide and holds it in place.

I started with a length of 50 mm diameter EN1A. It's too large to go through the chuck, so I used a fixed steady to support the outer end while I turned the outer diameter down to a bit less than 46 mm and bored a (slightly under) 30 mm hole through the middle:

After cutting the end off with the bandsaw, I put it back in the chuck the other way round and faced the end. I also added a heavy chamfer to this end so that it was obvious which end had been faced in the same setting as the central bore. The chamfered end will be the top and it doesn't matter if that isn't perfectly square to the bore.

The part then got clamped in the mill vice and I drilled an 18 mm cross hole, 16 mm from the base. I did this with drill bits up to 13 mm and then 14 mm, 16 mm and 18 mm end mills to bring the hole to the final size:

It was then placed in the vice in the other orientation with the base facing upwards. I aligned it (very roughly) by lying a bit of 12 mm (the first piece that came to hand!) bar through the 18 mm cross holes and rotating the clamp block until it looked about right. Again, it would have been better to use a V-block for better clamping, but I didn't have one that was big enough and I'm only drilling so I didn't think it was important. With it clamped in the vice, I spot drilled and drilled two 3.3 mm holes:

Those holes then got tapped M4, but I did that by hand in the bench vice rather than trying to machine tap while the part was held rather weakly in the machine vice.

I then needed some modified M4×8 cap screws for those holes. The screws need to have a short 2 mm diameter section on the very end. To hold them for modifying, I drilled and tapped an M4 hole in an offcut of brass and then used an end mill to bore through most of the hole, leaving a flat bottom and about 5 mm of thread remaining. That could then be mounted in the chuck and I could tweak the end of the cap screw. I had to do this with the tool upside down and the lathe running backwards; otherwise the cutting force would have just unscrewed the cap screw from the brass bit.

This photo shows one of the modified screws on the bench and another fitted into the threaded hole in the clamp block:

The "clamp nut" is made from 30 mm EN1A. I'm calling it a clamp nut, but it's really a screw that happens to be made in two parts: the "nut" with an internal thread and a bit of threaded rod glued into that thread. I started by skimming the outside diameter until it was a nice smooth sliding fit in the clamp block:

I then drilled and tapped a blind M10 hole, about 15 mm deep:

The bar then got clamped in the mill vice and an 18 mm hole cross-drilled, in the same way as was done for the clamp block:

Finally (for this part), the clamp nut was sawn off the bar and held in a collet chuck for facing/chamfering the top end.

The next job was to make a couple of end caps / bushes. These started life as a couple of offcuts of 20 mm brass bar:

The first job was to shape the end, reducing the diameter to 15 mm for 6 mm length and skimming the outside diameter for a clean look. An 8 mm hole was drilled through and then opened up to 9.2 mm for a couple of millimetres:

After parting off, it was mounted in a collet (with the other one at the back end of the collet to help ensure that the collet held tightly on this short part) and the end was faced. I used my Hemingway Kits rotary broach to turn the 8 mm hole into an 8 mm deep 8 mm AF hex hole:

These parts need a cross-hole for a grub screw, so I mounted them together in the mill vice, using a length of 8 mm hex bar to align the internal hex hole with the jaws of the vice:

I could then drill and tap an M4 hole in each piece:

This photo shows how those end caps fit into the end plates of the tool rest slide and support the hex bar that runs along the length. This photo was taken before I drilled the extra holes in the bottom of the tool rest slide that give access for a 2 mm Allen key for tightening/loosening the grub screws that hold the brass bushes/end caps onto the hex bar. The brass bushes rotate freely in the holes in the end plates and hence the hex bar can be rotated freely.

continued....

 

[Dr.Al]

That brings us to the last piece of the puzzle. I started with a bit of 18 mm round bar and, after facing the end square, I used a 2 mm wide parting tool to cut two grooves in the part (and also to mark the length of the bar - that's the groove nearest the chuck):

An 8 mm hole was drilled through the part and then the rotary broach was used to turn it into an 8 mm AF hex hole, going as deep as the tool would allow (which is about 8 mm):

The resulting hole:

With the hole cut, I could work on the eccentric cam section. I added an 0.8 mm piece of feeler gauge stock to one of the jaws of the chuck:

I could then use a pointy tool to cut the central section of the piece. I wasn't aiming for a specific diameter here, I just kept taking passes until the tool was cutting continuously:

The final operation on this part (for which I didn't take any photos) was to cut it off the bar, mount it the other way round in the collet chuck and face the back end. I then used a 9.2 mm drill bit to drill most of the way through the part, leaving the final 8 mm-ish. The reason for that final drilling operation is that I want an 8 mm hex bar to pass all the way through the cam. My rotary broach will only cut about 8 mm deep, so the rest of the length has to be drilled wider to allow the across-corner dimension of the hex bar to go through. Theoretically this dimension is 8×2÷√3 = 9.238 mm, but I measured the size of my hex bar and it was slightly under 9.2 mm.

With the final part made, I could assemble the tool rest slide:

This shows the underside of the assembled tool rest slide:

The two modified M4 cap screws engage with the grooves in the cam and stop it from sliding out of the clamp block. Rotating the cam section lifts or lowers (under gravity) the clamp nut (and hence the M10 threaded rod). The big washer and the nyloc nut bear against the bottom of the bed.

As the threaded rod is round, the tool rest slide can be rotated to any angle. As the clamp block / clamp nut / cam assembly can slide freely along the hex bar, the tool rest slide can be moved back and forth across the bed (as well as along the bed). Rotating the hex bar (via a yet-to-be-made handle) by about 90° is enough to release/lock the tool rest slide to the bed.

This very short video shows the action of the lock (using an 8 mm spanner as a temporary lock handle and with the "bed" bodged as two bits of box section clamped round some 12 mm plate and then clamped to the bench).

I spent quite a lot of time thinking about how to make the tool rest slide (not least making a hex hole for a 45 mm long cam to slide along a hex bar) and I'm really pleased with how it's working. The action of sliding is a little bit rough, but that's just because of the rough finish on the box section. It should be very quick and easy to adjust the position of the tool rest with this mechanism.

 

[Migmac]

Really enjoying this thread, quite envious of your talent

 

[Dr.Al]

Thanks @Migmac

 

[sparkysy]

That is some amazing work, and an excellent project!!
Really enjoyed reading through your posts, thanks for sharing!

I do enjoy a bit of wood bothering, and I have a wood lathe, in fact that is working unlike my metal lathe that is still in bits and boxes from moving 10 years ago

 

[Dr.Al]

The next job, which I've been somewhat nervous about, is making the tool rest itself. In the excellent (from what I can tell, never having used a lathe!) book, Woodturning: A Foundation Course by Keith Rowley, he comments that most lathes have a less-than-ideal tool rest shape. This is the shape he recommends:

Making that shape seemed a bit too difficult to me, but I thought I'd try to make something that approximates it. After spending quite a long time thinking about ways to make something neat (as opposed to just roughing it out with an angle grinder and ending up with something shabby but workable), this is what I came up with.

I started by mounting a 300 mm long bit of 25 mm square bar in the mill vice and squaring the ends up. I suspect I'll eventually decide that 300 mm is a bit long (if it interferes with the tailstock and headstock too much), but I figured it would be a lot easier to shorten it later than try to lengthen it!

Once the ends were squared up, I drilled and end milled a flat-bottomed 16 mm hole, about 16 mm deep. Sorry for the blurriness of the next two photos...

I then fitted an end stop to the mill table, rotated the part through 90° and then drilled and tapped an M5 hole (and then took another blurry photo):

The end stop was used as the hole position wasn't that critical and it's a lot quicker to use that than to use an edge finder to find the middle of the bar after moving it.

I had hoped to drill some holes in each end using the milling machine, with the bar stood upright clamped to an angle plate. Unfortunately, the milling machine wasn't tall enough, so I had to just mark the holes out as carefully as I could and then do my best to drill them in the right place using the pillar drill, with a vice overhanging the table and clamped in place with a C-clamp:

The holes (matched on both ends) consist of a very small centre hole near the corner and an M8 tapped hole in the middle:

I used an offcut of steel to make a special live centre. This has a 10 mm hole drilled about 14 mm from the centre and then a 60° point was cut on a reduced diameter section to make the centre:

The tool rest blank got a small piece of M8 threaded rod screwed into the centre hole and that goes through the 10 mm hole of the live centre and acts as a drive dog, while giving access to the entirety of the outside of the piece.

This shows the whole tool rest blank in place:

I was very nervous about this as the centre holes are tiny (as they're only 3 mm from the each side, I could only really use a 4 mm centre drill), but it went okay. This was the first pass:

It took about 10 minutes to do each pass and there were rather a lot of passes, so it took a couple of hours to bring the tool rest to its final shape. About half-way through, the live centre (at the chuck end) snapped off in the part, so I had to redrill the centre hole (again using the pillar drill) and cut a new 60° point on the live centre before I could carry on. The second point lasted until the end.

After getting it to something that looked like the right shape, the fairly heavy-duty carbide cutter I was using (which I thought stood the best chance of surviving the interrupted cuts) had left a rather rough finish, so I swapped to a sharp rounded-tip high-speed steel tool and took a final light pass. In this video, you can just about make out the improvement in finish as the high-speed steel cutter works its way along the rest:

The finish after that pass looked a lot better:

This photo shows the finished tool rest profile (and the slightly mangled little centre hole, which had taken a lot of abuse!):

A while ago I made a filing machine (also known as a die filer), to a design by Harold Hall. That design called for some ground-to-size 16 mm EN1A bar. I had about 110 mm left over from the filing machine, so I figured that would do as a tool rest support for now. If it ends up not being long enough, it will be very easy to swap it for something else (such as the silver steel bar I used when testing the hole in the tool rest slide).

Preparing that bar only involved facing the ends and then drilling a hole for the grub screw to go into (to help make the grub screw be flush with the outside of the tool rest)...

... and that's the tool rest complete. I'm really pleased how this has come out as I was quite worried about whether that method of cutting the shape in the lathe would work or not.

 

[Dr.Al]

The first job for the lathe bed was to use a non-woven preparation wheel to clean up all the mill scale:

After clamping the two pieces together, my metal-cutting circular saw could then trim the two pieces to the same length, using this plywood guide:

The plywood piece ensures the saw cuts square; the fence was cut to length with the saw itself, so makes it very easy to see where the cut line will be.

Some end plates, made of 8 mm × 40 mm EN3B were also cleaned up and cut to length in the same manner, along with a piece to use to hold the tailstock bearing in place.

I drilled some holes in the plates. I'm not really sure what they'll be used for, if anything, but I needed at least one hole in each piece to allow air to escape when welding and I figured it would be a lot easier to drill and tap holes before welding the plates onto the ends of the bars.

They were tapped M10 by hand in the bench vice.

The bed pieces then got clamped down to a bit of 12 mm × 125 mm steel plate (to keep them level) and clamped around some bits of 12 mm × 50 mm flat bar to keep them parallel and the right distance apart:

To hold the bed upright, I used a G-clamp to hold a piece of aluminium to the bench and an F-clamp to hold the bed to the piece of aluminium, resting the end of the bed on a vice to raise it up a bit and make it easier to hold for clamping:

I then offered up the end plate and demonstrated that having a PhD in engineering and spending a considerable amount of my career solving complex differential equations doesn't necessarily equate to doing basic addition:

Those plates were supposed to be cut to the same length as the depth of the bed, but somehow I was 20 mm out. Cutting that plate was the last thing I did last night, so I guess that shows that I shouldn't try to engage my brain at the end of the day!

Never mind... I can sort that out later. Another couple of clamps were added to keep the bed top surface flat near the ends and then I tack-welded the end plates in place:

I could move the bed to a more comfortable position for welding:

The overhanging end plates meant I had to weld round the sides with the bed standing on end, but that wasn't too much of a problem:

I used the metal-cutting circular saw again to trim off the bit that shouldn't have been there...

... and then cleaned the ends up with a flap disk:

The bed is essentially done now, although I'll probably add some threaded holes in the bottom for attaching feet and/or the motor assembly.

With that done, it was time to play around with different layouts so that I could try to figure out how to mount the motor. One option is to mount the motor with the spindle pointing to the left. This would keep the overall length of the lathe to a minimum, but the motor would have to be a long way back in order to give clearance for the tool rest slide:

The support piece (currently a bit of 3030 t-slot extrusion) for the right-hand side of the motor also limits the left-ward travel of the tool rest slide.

The alternative (and the one I'm currently leaning towards) is to have the motor spindle pointing to the right. That allows the motor to be much, much closer to the spindle (it might even allow me to use the one and only drive belt I already have), but makes the overall length of the lathe much greater.

Other advantages of this layout are that it gives somewhere fairly logical for the inverter to go and it means the tool rest slide can go anywhere along the bed with no interference. Top view:

The main disadvantage of this layout is that it's going to need some thought to figure out how to attach the motor to the bed in a robust enough way to tension the drive belt. I quite like the idea of using the 3030 t-slot extrusion for the motor mount as it gives me a t-slot "for free" and I can use that to slide the motor back and forth to tension / loosen the belt. However, as-is, it would be completely independent of the bed and that might not be great for tensioning.

The simplest option would be to clamp the motor mount assembly (i.e. t-slot extrusion) to the bench and independently clamp the lathe bed to the bench. That would probably work fine, but I suspect it would be better (and easier to set up) if the two are attached in some way.

While I had the bed clamped to the bench, I also shoved a live centre in the tail stock and a (rather inappropriate) metal-work dead centre in the headstock and had a little play with the tool rest and how limiting the length of it would be:

 

[Dr.Al]

I've been carrying on working on the lathe and taking lots of photos, but I haven't been keeping up with writing it all up. I'll spread this weekend's work over a few posts rather than trying to cover it all in one go.

Another little job that I needed to do was make a T-nut for the tailstock. This will engage with the bottom of the bed and will be used to lock the tailstock in place. I possibly could have used a T-nut from the clamping kit from my milling machine, but I decided to make one a bit bigger and with a deeper upright section. If I decide in time to replace the bed (currently 40 mm × 40 mm × 2 mm box section) with something a bit stronger (e.g. 40 mm × 40 mm × 4 mm box section), it's likely that there will be a bigger fillet radius (box section outside fillet radius is typically 2×thickness) and hence the T-nut needs to extend beyond that fillet.

While I was at it, I decided to use the same "form factor" to make what is essentially a pair of washers for holding the headstock in place. I started with an offcut of 25 mm square EN1A and I used a little 16 mm carbide insert end mill (that I picked up at the Bristol Model Engineering Exhibition a few years ago) to mill the T shape:

The upright bit is 12 mm tall, so should easily pass any fillet on a box section corner:

I then drilled three holes and tapped one of them:

Finally, the t-nuts were split into two pieces:

After a quick tidy up of the sharp corners, they're done:

The spacing of the two holes on the "washer" for the headstock matches the spacing of the M8 holes in the bottom of the headstock. If I'd made them as individual washers, they would have had to be rectangular (or have one side cut off) to stop them interfering with each other, but making a single washer sorts that problem out.

My original plan for aligning headstock and tailstock was to use some eccentric pins. After pondering for a bit, I decided to do this just for the tailstock: the headstock is likely to be pretty much permanent attached to the bed, so it doesn't really need to be as easy to adjust. To get it in roughly the right placed, I made a couple of 12 mm pins with 6 mm holes through the middle. These attach to two of the threaded holes in the base of the headstock and are a fairly tight fit in the bed slot:

For the tailstock, I made four pins, each slightly under 12 mm. I used a collet block to mill a 10 mm AF hexagon on the end of each block:

I then used a bit of feeler gauge stock against one of the chuck jaws to hold the pin off centre. I could then drill a 6 mm hole all the way through:

This shows the pins fitted to the base of the tailstock:

They can be adjusted with a 5 mm Allen key and a 10 mm spanner:

The length of the pins was chosen such that the hexagonal section sticks out beneath the base of the bed and hence they can be adjusted in situ.

 

[Dr.Al]

The next job to work on was the method of attaching the bed to the motor.

I've got quite a lot of 3030 t-slot extrusion. A lot of it has holes/slots in various places (most of it came out of a skip having been used and discarded), but it's still perfectly usable. This seemed a good basis for the motor mount as the t-slots give an easy method of sliding the motor back and forth to apply tension to the belt.

To attach to the bed, I clamped the bed in the mill vice (with some 12 mm flat bar in the gap to stop the vice from bending the bed) and drilled some holes:

Rather than welding in some nuts (and risking distortion from the heat of welding), I decided to use M8 rivnuts for the bed attachment. These also have the advantage of being very quick to fit:

Some "feet" could be attached to the bed using those rivnuts:

There's one of these at each end of the bed. They raise the bed up by 30 mm, giving clearance for the t-nuts etc. I've spaced them 864 mm apart and drilled / counterbored some extra 8 mm holes out-board of the bed on a 192 mm pitch. Those numbers are both multiples of 96 mm, which corresponds with the dog hole pitch on my bench, so I have the option of screwing the bed down using those holes (although in practice I suspect it will be quicker to just use a couple of F-clamps).

I then built a simple frame out of t-slot extrusion, using some coupling joints that came out of the skip with the extrusion:

By undoing two of the M8 grub screws that hold the frame together, the motor mount section can be easily separated from the bed section, which is important as I don't think I'd be strong enough to lift the lathe and motor in one piece (the motor is very heavy).

To attach the motor to the extrusion, I cut a couple of lengths of 12 mm × 50 mm EN3B and used the lathe to drill/tap a couple of M10 holes and drill a couple of 6 mm holes in each:

Those plates got bolted to the bottom of the motor with some short M10 cap screws and some extrusion t-nuts and M6 cap screws used to hold the motor mount plates to the extrusion:

Loosening the four M6 cap screws allows the motor to slide easily back and forth on the relatively smooth surface of the aluminium extrusion. The two extremes of travel are shown in the next two photos:

General view of everything:

With that done, I thought I'd have a go at making some pulleys (and reminding myself why I hate turning aluminium). I had thought about modifying the spindle such that the pulley mount diameter was common between the motor and the spindle, but after thinking about it more I decided not to do this. The pulley for the motor probably needs to be longer (thicker?) than the one for the spindle due to the different mounting methods. The spindle pulley will be held on using a nut; the motor has a keyway and no thread, so the pulley is best mounted with a grub screw.

For the spindle pulley, I started with a big disk of aluminium, held in the outside jaws of the three-jaw chuck. I faced the end, skimmed the outside diameter down to 80 mm (fairly arbitrarily chosen) and drilled/bored a 20 mm hole through the middle:

I used the bandsaw to chop it off over length:

I then faced it to length, again held in the outside jaws:

To help with shaping the pulley, I used a bit of 30 mm EN1A to make a simple mandrel, with a 20 mm outside diameter and an M10 hole through the middle. I used the collet chuck to hold the mandrel as it gives a bit more clearance for top slide rotation (and to remove the chance of accidentally hitting the spinning jaws).

I then set the top slide at 70° to the lathe axis to cut one side of the pulley shape (the drive belt I have has a 40° included angle, so 20° on each side). I then rotated the top slide and did the same for the other side of the pulley shape. I tried to get a few photos, but messed up the focus on all of them; believe it or not, this was the best:

This was what it looked like when finished:

To make the motor pulley, I started with another disk of aluminium and turned a reduced diameter section (about 50 mm). This section is intended to be where the grub screw will be fitted for clamping the pulley onto the motor shaft.

Due to the overhang of the outside lathe jaws, there wasn't space for a normal tool, so I used a boring bar and turned the shape with the lathe running in reverse:

I then bored the 19.05 mm (3/4") diameter hole in the centre, using a bit of brass (that I'd previously turned to size) to check the size was (what I thought was) right:

The same mandrel as before (but turned for a longer section at a smaller diameter) was used to mount the blank and I could then turn the pulley shape. This photo gives more of an idea of why I detest turning aluminium!

To cut the keyway, I started by feeding a hacksaw blade through the slot and then remounting it to the hacksaw frame. I could then rough out the slot:

That got finished with a lot of tedious work with a hand file.

I then had a go at fitting the pulleys. The spindle one went on very easily and was clamped in place with another nut that I made in the same way as the spindle mount nut that I described earlier (I also shortened the spindle mount nut and made a second short one as they're now being used to pre-tension the spindle bearings and hence need to lock together to hold the tension).

The motor pulley was a lot more involved. I'd measured the motor spindle near the bearing and it was 19.048 mm according to my micrometer. That corresponds fairly closely to 19.05 mm, which is 3/4", so that made sense to me and hence that's the size I made the pulley bore. When I tried to fit the pulley, it wouldn't go on.

I re-measured the motor spindle and found that (rather bizarrely in my opinion), it's 19.048 mm near the bearing, but 19.35 mm at the tip! That makes for a rather awkward situation: if I enlarge the hole to be 19.35 mm, then it might end up being very loose by the time it's pushed all the way on (although I guess the grub screw would probably deal with that issue).

I've got a 19–21 mm adjustable reamer, so I used that to gradually increase the bore diameter. I was feeling quite nervous about making it too big (I really didn't want to have to start again with this part!), so I took very light cuts and kept trying to mount it on the motor spindle. After several cuts, I found that I could "persuade" the pulley onto the motor shaft with a mallet.

It took quite a bit of persuasion to get it to seat. Sufficient persuasion that it'll need a bearing pulley to get it back off! I had intended to cross-drill and tap the hole for the grub screw after making sure that the pulley fitted over the keyway, but given how tight it is, I've decided to leave it for now.

With an 80 mm (outside) diameter motor pulley, it should be easy enough to adjust speed (if I need more range than I can get from the variable speed drive) by either fitting a bigger spindle pulley (for slower speeds and more torque) or a smaller one (for higher speeds). I'm not sure which diameter to use for the speed ratio calculation (outside of the belt, inside of the belt, middle of the belt?), but with the variable speed drive, it probably isn't that critical.

For now, I've got 80 mm outside diameter pulleys on both the motor and the spindle, so the speeds should match.

With the pulleys made and fitted, I could fit the drive belt and slide the motor back to apply some belt tension:

I couldn't resist a quick spin test:

It's feeling much closer to being a lathe now that the spindle has been run, but there are still quite a lot of pieces to make before I can start cutting wood. My hope is that I'll be able to at least do some trial cuts by the end of next weekend, but we'll see what happens.

 

[Dr.Al]

I didn't have much free time this evening (and most of what I did have was spent clearing up aluminium swarf), but I did manage to rummage out my old speed gauge and used it (along with the centre pip in an MT2 drill chuck arbor) to find out what the spindle speed was with the variable speed drive set to 50 Hz:

Looks like 1500 rpm to me. I also decided to live dangerously and wind the drive frequency up to 120 Hz. Nothing sounded odd (which was nice!) and the speed read-out shows 3450 rpm:

That's not far off linear given the resolution of the gauge. I'm sure it'll slow down a bit when under load (i.e. doing some actual turning) and I'm sure the slip will be greater at the higher frequency, but it does suggest that it might be usable over a decent speed range.

The drive has a nice feature where you can programme in the RPM you get at 50 Hz and then set the display up to show how fast the lathe is running at any frequency. I doubt it's doing anything very intelligent to account for loading, but it might still be a handy guide when I'm aiming for an approximate speed.

 

[Dr.Al]

I had a couple of hours free this evening, so I thought I'd get on with another aluminium turning job. I think this is the last one and there isn't much material to remove, so it's not as unpleasant as making the pulleys.

I started with another bit of skip-dive aluminium. I mounted it in the three-jaw chuck and faced one side:

I then flipped it round, faced the other side and then drilled and bored a hole in the middle. I crept up on the final dimension, testing with the tailstock screw repeatedly until it went on with a snug fit.

With that done, I took the aluminium piece off the lathe and pushed it onto the tailstock screw, with a bit of superglue applied to hold it (temporarily) in place. The assembly could then be mounted in a collet chuck on the milling machine and a hole located right on the join between the two parts:

That hole (shown after spot drilling above) was then drilled 5 mm to a depth slightly less than the thickness of the aluminium disk and then it was tapped. A through hole was also drilled and tapped in a fairly arbitrary location near the rim:

With that done, a grub screw was tightened into the joining hole and the whole assembly put back on the lathe for a final light skim of the face (the facing cut was mainly to remove the superglue stains to be honest). The outside edge was also cut and some chamfers cut and the edges broken with a fine file:

The assembly was then stood upright in the bench vice and acetone poured over it to weaken the superglue so that the parts could be separated (the bearing retaining plate needs to go between the aluminium disk and the bearing so the parts can't be glued together).

I dug out a bit of 16 mm 303 stainless steel and held it in the collet chuck (the main reason I used the collet chuck was that it was already fitted to the lathe). I drilled a 7.6 mm hole all the way through, then reamed it out to 8 mm. I then used a 14 mm end mill to cut a shallow counterbore:

The part was then reversed in the chuck and faced to length, removing it and checking the length several times to get it just right.

Here's the finished stainless steel handle, along with a shoulder screw that will be used to hold it in place. The shoulder of the shoulder screw is very slightly longer than the 8 mm diameter section of the stainless steel handle, so when tightened into the aluminium disk, the stainless handle spins freely:

With that done, I could call the tailstock handwheel complete and fit it to the tailstock:

 

[DanZac]

This really is a brilliant little project with some great ingenuity and your usual fabulous attention to detail, and a great, informative, write up.
Keep up the great work, I'm really looking forward to seeing this little jem in action making wood swarf.

 

[Dr.Al]

Thanks @DanZac

 

[Dr.Al]

Some more simple-ish jobs done this evening. I started with a bit of brass bar and shaped the end before using a tailstock die holder to put an M6 thread on the end.

After parting off, I fitted it into its destination hole and marked the angle I wanted the cross-bar to come out at. I then clamped it in a collet block in the milling vice and drilled a blind 6 mm cross-hole:

It then went back to the lathe to have the end tidied up:

For a cross-bar, I got a bit of 6 mm stainless bar and put it in the collet chuck and rounded the end over with a file:

I repeated that process on a second knob and then glued the two stainless bars into their holes with some Loctite 603:

The first knob sits in the back of the tailstock and clamps the tailstock spindle thing in place:

The slightly longer one screws into the end of the tool rest slide and clamps the tool rest in place:

Next up was a bit more work with some 303 stainless bar. I didn't take many photos of the process as it was much the same as above, but I drilled and tapped an M10 hole in the bottom of some 25 mm diameter bar and then cross-drilled 8 mm. I also got a bit of 8 mm stainless bar and drilled and tapped an M5 hole in each end. They went together (without glue) to form the tailstock lock knob, using some M5 cap screws as bar ends.

I also glued a bit of threaded rod into the tailstock t-nut...

... and to be extra sure it wouldn't go anywhere, I peened over the bottom end:

This one is used for locking the tailstock in place. Using M5 cap screws for bar ends means they (and hence the sliding cross-bar) can be easily removed, which makes fitting the tailstock lock nut a lot quicker. Once it's screwed most of the way in, the sliding bar is inserted and the cap screws tightened and then the cross bar can be used to tighten / loosen the tailstock:

 

[Dr.Al]

I've had quite a productive afternoon, so I think I'll spread the work over a couple of forum posts, one now and one tomorrow morning probably.

The first thing I wanted to get on with was mounting the inverter (aka variable speed drive, aka variable frequency drive).

A couple of days ago, I 3D printed a little marking gauge to allow me to mark a pocket in a bit of 40 mm deep aluminium angle:

To hollow out the shape, I used a drill bit (not especially accurately) to get rid of the bulk of the waste and then filed it until it was looking okay-ish:

I then drilled some mounting holes, including a bonus extra one that I drilled in completely the wrong place!

Before mounting the inverter to the motor plates, I fitted what will be the final bit of wire to the motor and trimmed it to length. There isn't anywhere very obvious to attach the earth wire to the motor, so I'm still pondering what to do about that as it seems sensible to have a decent earth connection.

Using a multimeter on continuity test mode, it buzzes through okay from the spindle shaft to the inverter earth and all the way to the bed, so it's probably okay without that extra connection, but I still think I'll probably attach it to a screw hole somewhere in due course.

There was then a very tedious exercise of doing lots of cutting, stripping and crimping of wires. I'm an electronics engineer by profession, but I find wiring intensely dull.

With that done, I could finally mount the inverter through the motor mount plates (with slightly longer screws than were there before), so the inverter slides back and forth with the motor.

Another little 3D-printed creation was a little control panel to sit at the right-hand end of the bed, connected to the inverter with an old 9-way serial cable extension lead (held to the bed with some M4 rivnuts and some cable tie mount things):

It all seems to be working okay when powered on. It took me a while to figure out how to configure the remote controls. The manual for the Jaguar Cub inverter I have on my metalworking lathe is 171 pages long. The one that came with this cheap unbranded inverter is two sides of A3 and I've already found a couple of mistakes in it! Anyway, after fiddling and experimenting for a while, I got it all set up okay.

There are a couple of things that don't work quite as planned. The little LED on the control panel is one with a built in flasher. I fitted one of those to the control panel on my metalworking lathe, wired up such that it flashes when the inverter is configured to run the lathe backwards. I wanted to do the same here, but the LED needs 24 V to come out of the inverter and this one only supplies 12 V. I haven't decided whether I think it's worth the effort to do anything about that.

The other thing which doesn't work quite right is the speed control. I've got the inverter configured for maximum frequency of 120 Hz. If I set it up to derive the speed from the control panel potentiometer, it runs from 0 to 120 Hz. If I switch it to use the remote potentiometer, it runs from 0 to 60 Hz. The manual shows the potentiometer supply voltage as "10 / 5 V". My inverter supplies 5 V for the potentiometer, so I wonder if the power supply output is 5 V but the potentiometer is expecting to run from 0 to 10 V!

The next job I got on with was making a lid for the headstock (to stop it filling up with sawdust). I started with a bit of 2 mm plate (probably stainless steel, but I'm not sure) and chopped it roughly to size with the metal cutting circular saw:

I then used some transfer punches to mark the location of the mounting holes and measured the location of the hole needed for the tommy bar (used to lock the spindle and allow a chuck or the pulley/bearing nuts to be tightened/loosened). They were drilled and then the lid was attached to the headstock and ground to fit with a flap disk in an angle grinder, not worrying too much about perfection!

There's still a couple of holes in that lid, which sawdust could get into, so I mounted a bit of 20 mm 303 stainless bar in the four-jaw chuck, rather dramatically shoved off-centre. I then faced it and turned it down to 4 mm:

I used a die to cut an M4 thread on the reduced diameter bit:

After cutting off with the bandsaw, I then used my home-made soft-jaws to hold the part the other way round to tidy up the cut-off face:

The finished part, along with the nyloc nut that will hold it in place:

This is what it looks like in situ:

It can be rotated out the way to allow tommy bar access:

 

[Dr.Al]

The next job I decided to work on yesterday is one that is arguably a bit pointless, but I thought I'd do it anyway.

I started with a bit of 25 mm silver steel in the chuck. I faced it off and then drilled and tapped an M10 left-hand thread:

I then transferred it to a collet block in the mill vice for shaping:

The first job was to remove the stock to slightly under half-way, before removing all the way to the midway point on just one side. I then flipped it over and repeated - note that I'm using a slightly blunt carbide tip because I don't really care how rough the finish is on this part:

I then held the collet block at a rather arbitrary angle in the mill vice (the angle was mostly dictated by the length of the bar!) and used the same cutter to turn those prongs into pointy bits:

The part was then sawn off with the horizontal bandsaw and then I used the lathe again to clean the rear face up:

The other half of this part was made from an old MT2 blank end arbor, which had previously been used for a tailstock die holder for my mini-lathe. I never use the mini-lathe now that I've got the Harrison M250, so I figured this part was fair game. To hold it in the spindle, I used a stack of adaptors. The M250 has a rather unusual Morse Taper 4.5. A while ago I made myself an adaptor to go from MT4.5 to MT3 (the "proper" adaptors of the sort that would have come with the lathe seem to be like rocking horse droppings). This, in combination with an off-the shelf MT3 to MT2 adaptor allowed me to hold the part, with a bit of M10 threaded rod all the way through the spindle acting as a drawbar to stop the arbor from shifting.

It was then just a case of cutting the matching M10 left-hand thread...

... and then making the end pointy:

That's it done for now. I should probably harden and temper the silver steel part, but I don't think I'm going to bother. The two parts together go in the headstock spindle thus:

Any turning forces will tend to act in the direction that causes the thread to unscrew, which will drive the prongs further into the wooden workpiece.

It's probably a rather unconventional design for a drive centre, but it felt like it was the easiest way to make one given the bits I had to hand.

The reason that I'm not going to harden it and that I said at the start that it's a bit pointless is that I'm planning to buy a proper commercial drive centre fairly soon (recommendations are welcome). I only really made this one to prove to myself that I can get to the point of turning a chisel handle with minimal purchases involved. As soon as I've proved this works well enough to be functional, it'll probably go in the metal scrap bin!