[TangentAudio]

Greetings, Zoners.

I've had my converted Sieg X3 for four years now. I used an off-the-shelf kit for the mechanicals of the initial conversion, sourced from the now-defunct ProMiCA in Australia. I did this primarily to focus on getting a machine running ASAP. My official background is electronics and software engineering, so it was a natural fit to do those things myself. I have a thread over in the X3 forum that describes some of the details, including some links to videos that give an overview of the conversion and some of the unique things I did in my build.

It has been a reasonably good machine for me and I've learned a lot in the process. I've been a life-long tinkerer and maker and it's been awesome to finally have a tool to help me realize the ideas in my head. So far I've mostly used the machine to make functional prototypes for ideas relating to my various other hobbies and interests. I've made motorcycle parts, mountain bike parts, photography/videography related parts, etc.

Lately I've felt like a new machine might be nice, to overcome some of the limitations of the X3 (and simply to have a new toy to play with and learn). Tormach/Novakon/IH/etc. seems like the logical step, but once you're talking about $10K+, a used commercial machine starts to look appealing as well. I am operating out of my walkout basement shop with a 36" wide door, and no 3 phase available on my street. Moving a 3 ton machine into my basement just doesn't seem like it's in the cards any time soon. Moving a Tormach or any of the other hobby class machines would be easy, but ultimately I decided waiting at least another year might be a good idea to see if the market offerings change.

By chance, I ran into this awesome build thread on this forum. The really interesting discovery to me was that Tormach actually sells their spindle heads as complete units. They're not exactly cheap, but once I discovered that they're very close in size to the X3 head, got really interested in the idea of grafting one onto my X3. This would get me a 10K RPM spindle and a 3 phase AC motor with belt drive. Having 5 times the spindle speed and getting rid of that rattle-trap gearbox on the X3 seems really appealing.

I did some measuring and modeling, and it seemed like it was possible with some work. The main part that needs work is the mounting to the Z slide. The slide on the X3 is a bit too narrow, but it's actually the same height as the Tormach spindle head. Sometimes I just can't let go of an idea, even if it's a bit outlandish. So, I'm aware that this is a pretty silly project, but I think it will be entertaining and educational.

Maybe it's a bit like this?

Attachment 208614


Anyway, I'm starting a new thread over here because the X3 forum seems to be completely dead these days. Not only that, doing mad scientist head transplants seems a bit out of scope for that forum anyway.

[TangentAudio]

Mating the Head to the Z Slide
Attachment 208616

Looking at the mounting of the Tormach head vs the X3 Z slide, it's a bit eerie how close these are. Coincidence? Creative Chinese copying? Who knows.

Machining a new Z slide seems like one possibility, though the dovetails are 55 degrees instead of 60 degrees. This would be a lot of precision work, and I think it is probably a bit out of my reach in terms of my skill and ability. I'd probably have to do it at least twice to get it right, and even then maybe I wouldn't nail it. I don't have much experience working cast iron or steel, so it would be a pretty steep learning curve.

Another possibility is to offset the entire Tormach head left or right by 8mm - at that point, two of the holes actually appear to line up. Then I'd make an extension for the opposite side that would bolt to the side of the Z slide, and to which the Tormach head would bolt from the front. It would be "L" shaped and wrap around to the back of the Z slide for additional support. The nice thing is that this doesn't require any significant modifications to the Z slide that would prevent putting the X3 head back on if I decide this whole project is stupid. The downside is that it doesn't seem to be anywhere near as rigid as other approaches.

Attachment 208618

The sketch above represents my current line of thinking. It does require some Z slide modifications, as well as some non-critical modifications to the side plates that attach to the Z slide.

Basically I'd make the two parts shown in red with the X3, before disassembling it. Then I'd tear it down, beg/borrow some manual machine time somewhere, and make the modifications to the Z slide and side plates. Then put it all back together, mount the new Tormach head and get to tramming.

The idea is still a work in progress. I think I should probably account for the middle bolts on the Tormach head, using a similar notch-in scheme. My initial reservation was that if I do that, I lose the two registration dowel pins for the X3 head. Not the end of the world I guess, but it is a commitment of sorts.

I also think it would be nice to include some kind of tramming aid. Maybe just a jackscrew arrangement on one corner to help me tilt the head left-right while I'm setting it up. Head nod would still have to be corrected with shims (either between the head and the Z slide, or under the column depending on the source of the error).

Speaking of dowel pins, the Tormach head has provisions for two pins. They're arranged differently than the X3's (which are drilled in from the side, in between the sandwich of the head and Z slide). I am assuming the correct procedure for setting it up will be to put one of those in in one corner initially. That will give the head a pivot point. Then I'd get the head in tram left-right, and while the mounting bolts are snugged down, transfer punch and drill the other one in to lock it in place. Does this seem reasonable?

Spindle Nose to Column Distance

The stock X3 measures approximately 215mm from spindle nose to the mounting surface of the head. The Tormach 770 is 232mm according to the Tormach drawing.

The Tormach spindle would be 17mm (0.669") further from the column assuming a direct mate of the new head to the existing slide. This is probably close enough to work, though it does mean the center of the table would no longer be directly under the spindle when the Y axis is 100% supported by their ways. My ProMiCA CNC conversion added some Y travel over the stock X3, so it does help this situation.

[TangentAudio]

So one of my buddies that owns a pickup truck gave me a hand, and we picked up the spindle head at the local freight depot last night. The bill of lading indicated 220lbs, but I think that's overstated. I will be weighing everything soon, and comparing it to the weight of the X3 head.

Attachment 208630

Here it is, out of the crate and up on the bench. Looks like it survived the trip halfway across the country.
Attachment 208632Attachment 208634

Some hardware for mounting was included. Note the two tapered pins. It's worth noting that the drawbar is not included. I will have to order that separately.

Attachment 208636

The fit and finish is a bit... disappointing. The casting is pretty rough, and it's obvious there is lots of body filler being used. The large chamfer detail on the front of the head looks almost like it's hand shaped and made out of play-doh. I'm also really not digging the idea of body filler under the heads of the mounting bolts. There are a few examples of paint overspray here and there as well. It really makes me wonder if the Tormach factory in China uses spindle head assembly sales as a way to get rid of "B Stock" or lower build quality assemblies. I hope the fit and finish of a full Tormach machine is at least a bit better than this. It's harsh to say, but the outward appearance of my Grizzly X3 is considerably more refined. I'm sure my X3 is loaded with body filler as well, but it seems like it was at least done with more care.

Attachment 208638 Attachment 208640 Attachment 208642 Attachment 208644 Attachment 208646

I've started to tear it down a bit, but so far I haven't been able to get the spindle pulley retaining nut off. The pin spanners I have on hand do not seem quite up to the task.

Attachment 208650Attachment 208648

Here is the plate from the motor. To drive it, I sourced the same Emerson/Control Techniques Commander SK series VFD that Tormach specs for the 770 mill. The model is SKB1100075, and you can find them on ebay for considerably less than Tormach charges.

Attachment 208652

[Fenza CNC]

Interesting build, I'll be watching this closely.

Shame to see the quality of the casting in that state, one would expect better from Tormach.

I'd defiantly remove the paint from where the bolt heads will mate, maybe even flatten the surface if needed, hard to tell in the picture.

[TangentAudio]

I agree, Fenza. I am going to see if I can work out a way to clamp down and true up the head casting onto my mill table and do some side milling of the mounting flange surfaces. I can't think of any other way to get a tool down there due to the shape.

[TangentAudio]

New part for the build arrived today. 70 ohm, 300W braking resistor, for the VFD. Cheap and Chinese from ebay. So cheap that one of the end caps was dislodged, and the shipping bag was filled with fine sand from inside the resistor. I contained the mess and put it back together, but I'll probably need to RTV the ends shut. Thankfully it did still measure out to 70 ohms, so I don't think there's any real damage.

Attachment 208702 Attachment 208704


I also continued the teardown process. I was finally able to get the lock ring off of the pulley with a little more effort, using a pin spanner from my bicycle toolbox and an adjustable wrench to brake the spindle using the flats that are milled into the shaft. But, I couldn't budge the balanced pulley that was next to come off.

Attachment 208706 Attachment 208708

So, I turned down a little shaft plug with a center detent, and used a gear puller and it came right off. That left the shaft key and a backing spacer behind.

Attachment 208710 Attachment 208712 Attachment 208714

Once that pulley was off, I was able to remove the motor sub-casting from the main spindle casting. The spindle was free to come out at that point, just by undoing the bolts around the diameter of the spindle nose. Given the value of the spindle alone, I decided it would be wise to protect it. I wrapped it in some oil soaked rags which made it a nice slip fit inside of a piece of PVC pipe. That should protect the exposed drawbar end nicely.

Attachment 208716 Attachment 208718 Attachment 208720

With everything stripped down, it was refreshing to finally see some machined surfaces. In the second picture below, I'm starting to puzzle out how I'm going to support the casting on the table of the mill to do some machining of the bolting flange. Unfortunately I don't have any machinists jacks, which would be a big help. Fitting it on my table is also a challenge - I'd rather have the bolting flange in parallel with my X axis, which means that heavy end with the bore will be overhanging the table. This operation may be better left to when I have access to a larger manual machine to do the Z slide mods.

Attachment 208722 Attachment 208724


As to why I want to machine that bolt bearing surface, take a look at this mess. I scraped at one of the holes a bit with an X-acto blade. Somebody loves bondo, but it's not me. The second pic is a detail of the half-hearted bondo fillet around the mating surface for the spindle cartridge.

Attachment 208726 Attachment 208728

Well, enough for tonight. Here's a parting shot with lots of little parts collecting from the teardown.

Attachment 208730

[WOTDesigns]

To hold it I would just bolt a 246 block vertical and clamp the casting to it

Brian
WOT Designs

[TangentAudio]

I like that idea, though I don't (yet) own any 246 blocks.

The casting does weigh a lot, so I think it would still need support at the far end. But it does solve the problem of getting the casting aligned vertically nicely.

[WOTDesigns]

The far end is what machinist Jacks are made for =)

Brian
WOT Designs

[TangentAudio]

My VFD showed up today, so I got it up and running with the Tormach motor. No real mystery to wiring up the basics: 120VAC single phase in, 3 phase wires to the motor, and some earth grounds. Had to set a few basic parameters to get the motor spinning the first time. Took some further tuning to speed up the accel/decel, and enable braking on stop.

Attachment 208886 Attachment 208888

Some research turned up that LinuxCNC does have some rudimentary Modbus RTU support, so I figured I'd mess around with that (being the software/electronics geek that I am). It took some doing to get RS485 working properly on the embedded motherboard that I use in my control. There's a combination of physical jumpers, BIOS settings, and non-standard cabling to get it running properly.

For early debug I'm using a utility called modio that lets you issue Modbus commands and query registers. It took more work to get my first basic query to work, and then a bunch more effort to figure out the basics of how the Control Techniques VFD uses its registers. I wrote some code and modified the modio utility to make this easier.

Attachment 208890 Attachment 208892

The upshot of all of today's effort is that I can run the motor forward, reverse, and set the speed all under computer control.

[TangentAudio]

Didn't have time to get much done today, but I did order a few things so I can make progress. Picked up a set of 246 blocks and some small machinists jacks, to work on the main casting. Also picked up a granite square that I can use for taking a few different squareness measurements on the machine.

[Fenza CNC]

Someone's been busy :cheers: just out of curiosity what brand is your granite square and where did you get it from?

[TangentAudio]

Just a cheap one from Shars. Can't speak to the quality until it gets delivered.

shars.com - 10quot x 6quot x 1quot Precision Granite Square

[TangentAudio]

After some thinking, it may make more sense to modify the Tormach casting rather than the Z slide on the X3. I took some measurements and did some sketching last night and it looks like I could slot out the holes on the Tormach head to line up with the X3 Z slide holes. There's plenty of material on the head mounting flanges to do that. The Tormach holes are not directly centered (left-right) in the flanges, but instead offset towards the outside edges. The new holes need to move inboard 8mm on each side. Looks like there's plenty of room to do that, and still allow the backing washers for the bolts to sit flat on the flanges.

I could make some plugs out of cast iron rod, and either press fit or slip fit + red Loctite them into the Tormach holes. Once in place then I could machine the new mounting holes. Maybe this is overkill, but on the whole I think it's less effort than my other approach.

I also like this plan a little better because it means I can do all of the work on my own machine. My friend's manual Rockwell mill that I was planning to use won't be available for a while. Another bonus is that I wouldn't be making any permanent changes to the X3 and could go back to the original head easily.

[Fenza CNC]

I know this isn't ideal and I'm sure you've ruled it out already but what about a cast iron plate to sandwich between the head and the Z plate, then offset the head vertically.

I think it'll be best if you don't try to change the head casting too much, at least then if all goes wrong you can still sell it.

[TangentAudio]

My first idea was a plate in between the new head and the Z slide, but it will further offset the spindle from the column. The Tormach head already has the spindle further from the column than the X3 head, so I don't want to make that any worse. I'm not sure I understand what you mean by a vertical offset, though.

[TangentAudio]

I caught the nasty cold going around the office, but I'm trying not to let it totally kill my productivity. So, I forced myself to get a little bit done down in the shop tonight.

Started with some cleanup of the mating surface. It was smeared with bondo, paint, and who knows what else. Left is before, the right two are after ten minutes with a wire brush and some lacquer thinner. Nothing like some lacquer thinner to clear the sinuses and destroy some brain cells.

Attachment 209304 Attachment 209306 Attachment 209308

With the back mating surface cleaned up, I got it up on the table, squared it up, and faced it. This was like machining Easter candy. Machine off the crunchy outer candy layer to get into the nougat cast iron center. The surface was an uneven mess under all that paint, and even after facing off quite a bit (over 0.100"), I still wasn't down to metal in that front corner. I didn't want to take off any more material, though.

Attachment 209310 Attachment 209312 Attachment 209314

Once that side was faced off, I was able to flip the casting over. This is the position that will allow me to mill the new slots that match the X3 hole pattern. There's not a lot to clamp on except the spindle bore, so I sliced up some PVC conduit to make some soft "jaws" for the step clamps. Not sure if this will be rigid enough for the machining, but I plan to take pretty light cuts with a small end mill to do the slotting.

Attachment 209316 Attachment 209318

The fly cut surface isn't exactly smooth, there are 1-2 thou ridges from the cutter all over the surface. It's not as flat as I hoped, but for the most part it's within a few thou across much of the surface. I might spend a little time trying to shim it up and get it more level for the final setup before I do the slotting and other operations on this side.

Attachment 209320

Got a few other items ordered today too... Ordered the drawbar from Tormach (the shorter one for manual use, though eventually I will probably get the power drawbar one). Also some random stuff from McMaster: a chunk of 1215 steel rod for plug material, a brass shim stock assortment (for shimming up workpieces for machining and for tramming), and some shorter Class 12.9 M10 cap screws for mounting the head.

[timewarp]

Hi, i am really interested in this thread, love the work you did on the controller. Can you explain your reasoning for undertaking this project? I would have thought a belt drive conversion and maybe bearing upgrade would have given you pretty good results as well? I can't help thinking that when you have finished this upgrade you are going to want to upgrade the column and then the table! I really hope this is not coming across in a negative way!

[TangentAudio]

Thanks, glad you enjoyed the controller work.

Part of the motivation was because I hadn't seen anyone else do it, and it seemed like a fun and realistic project to tackle. It also seemed like a pretty quick way to get to a 10,000RPM spindle, manual spindle lock, belt drive, and motor upgrade all at once. It is definitely not the cheapest way to do this, and like you say it may just expose the rest of the limitations of the X3 pretty quickly. I do have some follow-on projects in mind to make some improvements, though.

You could upgrade the spindle bearings and do a belt drive conversion and replace the motor, but I'm not sure what the practical upper limit would be in terms of speed with a bearing replacement. I've heard of people getting 5000-6000RPM out of an X3 spindle, but I don't think I've heard of anyone getting higher than that. The spindle cartridge in the Tormach seems to be better made, has larger bearings, and was designed, manufactured and balanced to run at 10,000RPM so there's no mystery as to whether it will work. Working on the spindle internals is definitely getting out of my comfort zone, but there's no doubt I'd learn a lot doing that project too.

Another benefit is that I should be able to use some Tormach accessories. If I was willing to spend the money, their power drawbar kit would bolt right on for example. Also, their 3X spindle speeder and other stuff that mounts to the spindle nose would fit the machine.

The short story is that I'm just doing it for fun and because I couldn't let go of the idea once I got it in my head. It's just how my brain works sometimes.

[timewarp]

Those sound like compelling reasons. There are some other really good mills around that have used parts from different sources. The ork mill comes to mind.