electronic lead screw

[Qtron]

Tormach makes several "personal" CNC machines. Several years ago, they replaced the Mach 3 controller with a proprietary controller based on Linux CNC and called PathPilot. My 770 came with Mach 3 controller and if I ran full bore up to the limit switch, it would overshoot and hit a hard stop.

With PathPilot, once you home the machine , you will not ever actually hit the limit switches (the home position is set about .050" shy of the limit switch). PathPilot also programs in a far limit, the default position based on the machine parameters but user programmable so the machine works to electronic limit positions.

You can download the manual here: https://www.tormach.com/support/wpdmpro/um10349-pcnc1100-manual-0916a-web/ Some people running non Tormach machines are using PathPilot. Since it is based on Linux CNC, it has to be open source. The communication is through a special board called a Mesa, IIRC.

I thought that Clough used a rather peculiar way of syncing the lead screw to the spindle. He explains his logic in his second video, around 26 min. in.

Wish u defined peculiar re syncing with the encoder!, re listened, makes sense, well, from a layman's p.o.v., but the floating point method is not necessarily more accurate than Jon's.. he said, 1 in 32000 of an inch. That was probably not your main point tho.

 

[Qtron]

As per the video someone posted previously. Disengage the leadscrew when you hit the relief. Stop the lathe. Back the tool out, reverse the lathe and start it up. When the thread dial comes back to the mark you where using, re-engage the halfnuts. Run the lathe back to the beginning of the thread and stop the lathe. Push the tool back in, and put your cut on. Lathe back into forward, start. Rinse and repeat.

if you don’t have thread dial, you’ll need to keep the half nuts engaged, and try and stop the lathe in the relief - but the rest of the procedure stays the same.

Hiya Greatoldone,
Did u prog Clough's board on a 32 bit machine, or 64 bit? And did u do it via the JTAG, or via the USB debug probe? (USB port).

 

[Qtron]

As per the video someone posted previously. Disengage the leadscrew when you hit the relief. Stop the lathe. Back the tool out, reverse the lathe and start it up. When the thread dial comes back to the mark you where using, re-engage the halfnuts. Run the lathe back to the beginning of the thread and stop the lathe. Push the tool back in, and put your cut on. Lathe back into forward, start. Rinse and repeat.

if you don’t have thread dial, you’ll need to keep the half nuts engaged, and try and stop the lathe in the relief - but the rest of the procedure stays the same.

Or u can simply, manually set the tool in the thread, by a good eye, wind back, set depth then restart lathe.. a pita tho, lol

 

[RJSakowski]

Using mobile ph ATM, this forum's protocols wont allow upload of a ph file, but anyway I note from your kindly provided link to Tormach user manual, they use 3 phase steppers! Must be an advantage with them, or are they simply a lower cost device?
I presume the 770 is not cheap!
Maybe Mach 4 has solved the 'head butting' issues of Mach 3 as u defined, RJ.
Will replay vid 2 re Clough, THANKS!, but largely the Q re Jon's approach, vs Cloughs remains unclarified.
Certainly the arduino is cheaper, easier to prog, (still no joy re mine not talking to the TI ide properly- I've asked Clough), and of course, the big appeal is Jon's touch screen, plus, and still the big unknown, the better torque achieved by fully utilizing the lathes gearbox (I don't mean change gears), an added complication of his ELS setup , but a worthwhile trade-off.

I bought my 770 eight years ago, almost to the day. We had just sold our startup company and it was a gift to myself. I wanted the capability that a CNC mill has and made up a package to include maximum machining options at the time, including 4th axis and the TTS tooling system and the package set me back over $14K.

As to why Tormach chose 3 phase steppers, I don't know. They are NEMA 34 motors. The rated torque for the x and y motors is 600 oz.-in. but the only time that I have lost steps has been on a hard crash. It was the reason that I decided to go with the 602 oz.-in. stepper for the ELS. They are driving 4mm pitch ball screws with angular contact thrust bearings so friction is reduced considerably (I believe that my total frictional load is around 50 oz.-in.) and they are driving with 80 volts vs. my 42 volts. My suspicion is that 3 phase offers an advantage over the two phase motor, particularly when microstepping. IIRC, microstepping is achieved by reducing the current to the windings by sine and cosine factors of the angle between the steps.

When I first read about Clough's strategy, I thought it a bit cumbersome to count steps from the point of the beginning and produce driver steps according to a given ratio of the encoder step count but I may have been over simplifying it. I guess the issue is the spindle rpm isn't a constant and can change with applied load but I am a long way from being an authority on these matters.

As to why Tormach went with the Linux CNC engine rather than moving to Mach 4, I believe that there were several factors at play. Mach 3 wasn't cutting it and they needed to move on. Mach 4 was still in development and I guess they liked what they saw in PathPilot. I really haven't used it much but Path Pilot is definitely superior to Mach 3. I am way behind on version updates and they have added a lot more features to the controller from the version that I am using.

Back to the ELS, I would like to see the ability to add user defined lead and thread pitch rather than the preset values. It would be nice to be able to set up some custom thread pitch if desired. Another nicety would be changing feed and pitch by just holding the buttons down rather than separate presses. However, the Clough controller offers greater thread capability than the geared option so I am still better off.

 

[RJSakowski]

Hiya Greatoldone,
Did u prog Clough's board on a 32 bit machine, or 64 bit? And did u do it via the JTAG, or via the USB debug probe? (USB port).

The TI IDE only works on a 64 bit machine as I found out when I tried loading it on my laptop.

 

[GreatOldOne]

Hiya Greatoldone,
Did u prog Clough's board on a 32 bit machine, or 64 bit? And did u do it via the JTAG, or via the USB debug probe? (USB port).

64 bit MacOS, via usb.

 

[kb58]

I'll preface this by saying that it may be my setup, but the Clough42 ELS hiccupped on me.

I had it set for a feed rate of 0.020" and had used it successfully several times. I put another part in and engaged the feed, then noticed that it was moving much faster. Somehow, it switched from Feed to Thread, and though I don't remember the actual thread number displayed, it wasn't as desired!

So, it "may" be EMI in my setup, so I'll keep an eye on it each time I start the lathe. If that's it, cables will be shielded and/or separated. It "may" also be the software, and as Clough42 said himself, this is beta code and not for anything important. So, yeah, as more people get these up and running, we need to keep an eye on things. And to be clear, I don't blame James for any of this, and very much appreciate him contributing such a cool tool to the community.

 

[Qtron]

I bought my 770 eight years ago, almost to the day. We had just sold our startup company and it was a gift to myself. I wanted the capability that a CNC mill has and made up a package to include maximum machining options at the time, including 4th axis and the TTS tooling system and the package set me back over $14K.

As to why Tormach chose 3 phase steppers, I don't know. They are NEMA 34 motors. The rated torque for the x and y motors is 600 oz.-in. but the only time that I have lost steps has been on a hard crash. It was the reason that I decided to go with the 602 oz.-in. stepper for the ELS. They are driving 4mm pitch ball screws with angular contact thrust bearings so friction is reduced considerably (I believe that my total frictional load is around 50 oz.-in.) and they are driving with 80 volts vs. my 42 volts. My suspicion is that 3 phase offers an advantage over the two phase motor, particularly when microstepping. IIRC, microstepping is achieved by reducing the current to the windings by sine and cosine factors of the angle between the steps.

When I first read about Clough's strategy, I thought it a bit cumbersome to count steps from the point of the beginning and produce driver steps according to a given ratio of the encoder step count but I may have been over simplifying it. I guess the issue is the spindle rpm isn't a constant and can change with applied load but I am a long way from being an authority on these matters.

As to why Tormach went with the Linux CNC engine rather than moving to Mach 4, I believe that there were several factors at play. Mach 3 wasn't cutting it and they needed to move on. Mach 4 was still in development and I guess they liked what they saw in PathPilot. I really haven't used it much but Path Pilot is definitely superior to Mach 3. I am way behind on version updates and they have added a lot more features to the controller from the version that I am using.

Back to the ELS, I would like to see the ability to add user defined lead and thread pitch rather than the preset values. It would be nice to be able to set up some custom thread pitch if desired. Another nicety would be changing feed and pitch by just holding the buttons down rather than separate presses. However, the Clough controller offers greater thread capability than the geared option so I am still better off.

Wow u must have a lot of customers for $14 k spend. Slick machine!
BTW, after relisten to vid 2 I suddenly realised he has no accel/decel built in. Not good.

 

[Qtron]

I'll preface this by saying that it may be my setup, but the Clough42 ELS hiccupped on me.

I had it set for a feed rate of 0.020" and had used it successfully several times. I put another part in and engaged the feed, then noticed that it was moving much faster. Somehow, it switched from Feed to Thread, and though I don't remember the actual thread number displayed, it wasn't as desired!

So, it "may" be EMI in my setup, so I'll keep an eye on it each time I start the lathe. If that's it, cables will be shielded and/or separated. It "may" also be the software, and as Clough42 said himself, this is beta code and not for anything important. So, yeah, as more people get these up and running, we need to keep an eye on things. And to be clear, I don't blame James for any of this, and very much appreciate him contributing such a cool tool to the community.

I'll preface this by saying that it may be my setup, but the Clough42 ELS hiccupped on me.

I had it set for a feed rate of 0.020" and had used it successfully several times. I put another part in and engaged the feed, then noticed that it was moving much faster. Somehow, it switched from Feed to Thread, and though I don't remember the actual thread number displayed, it wasn't as desired!

So, it "may" be EMI in my setup, so I'll keep an eye on it each time I start the lathe. If that's it, cables will be shielded and/or separated. It "may" also be the software, and as Clough42 said himself, this is beta code and not for anything important. So, yeah, as more people get these up and running, we need to keep an eye on things. And to be clear, I don't blame James for any of this, and very much appreciate him contributing such a cool tool to the community.

Thanks for the heads up KB58

 

[Qtron]

64 bit MacOS, via usb.

Maybe TI 32 bit version, which I fully updated is my problem then