- Joined
- Mar 20, 2014
- Messages
- 418
He’s up to video 6, where he’s been doing test cuts with the prototype.
-
Welcome back Guest! Did you know you can mentor other members here at H-M? If not, please check out our Relaunch of Hobby Machinist Mentoring Program!
electronic lead screw
- Thread starter greenail
- Start date
He’s up to video 6, where he’s been doing test cuts with the prototype.
My divorce got in the way of my attempt to implement an ELS but I recently saw a deal on mini lathe and ended up starting a CNC conversion with centroid which cut it's first chips last night!
I've watched clough42's videos. I wish he had picked a more common and cheaper controller. An esp32 is around $4 and an stm32 is about $1.5. Both should be able to do the work. The nice thing about the esp32 is that it has 2 cores and you can have one core handle monitoring the button inputs and the display while the other stays focused on spindle sync. I also wonder if he is going to power the cross slide or if he is just going to leave it setup to sync the screw. Adding cross slide feeding and feeding back to the start position saves a ton of cranking.
My attempts used an esp32 which has plenty of processing power to synch the spindle. I also attempted to set my implementation up to allow you to cut both ways but the backlash in my lead screw made this not possible. I may dig back into it at some point but now that I have centroid to learn it may be a while.
If you implement an ELS you are really about 5 steps away from a full CNC implementation. The thing that pushed me towards ELS was keeping manual control. The ideal world has the full CNC conversion but allows for a really good manual control scheme. I was thinking that some force feedback control wheels would be a very nice way to manually drive a converted lathe and maybe is a better project overall to work on. I wish grbl's jog interface didn't suck so much, there are a few forks that implement spindle sync and bart dring's esp32 port is very nice.
I've watched clough42's videos. I wish he had picked a more common and cheaper controller. An esp32 is around $4 and an stm32 is about $1.5. Both should be able to do the work. The nice thing about the esp32 is that it has 2 cores and you can have one core handle monitoring the button inputs and the display while the other stays focused on spindle sync. I also wonder if he is going to power the cross slide or if he is just going to leave it setup to sync the screw. Adding cross slide feeding and feeding back to the start position saves a ton of cranking.
My attempts used an esp32 which has plenty of processing power to synch the spindle. I also attempted to set my implementation up to allow you to cut both ways but the backlash in my lead screw made this not possible. I may dig back into it at some point but now that I have centroid to learn it may be a while.
If you implement an ELS you are really about 5 steps away from a full CNC implementation. The thing that pushed me towards ELS was keeping manual control. The ideal world has the full CNC conversion but allows for a really good manual control scheme. I was thinking that some force feedback control wheels would be a very nice way to manually drive a converted lathe and maybe is a better project overall to work on. I wish grbl's jog interface didn't suck so much, there are a few forks that implement spindle sync and bart dring's esp32 port is very nice.
- Joined
- Jan 10, 2019
- Messages
- 1,220
The controller board is maybe another $20, roughly 15% of the total. Does that make it a no-go?
Last edited:
answer a: I have a few spares of the other MCU's so it would be easy to test if he chose something I already had on hand.
answer b: I think you could get away with a non-hybrid stepper at a much cheaper cost. the hybrid will give a much better torque curve and that really helps if you are trying to thread at a high speed.
answer c: at $250 you are pretty close to a centroid or linux cnc setup. The utility of a full CNC lathe setup is questionable for the hobbyist.
answer d: I think engineering the costs down as low as possible would make it more approachable to small lathe owners and the more prevalent the MCU is the more people could contribute. I also think a fully canned threading cycle like the Russian ELS has would be fairly easy to do and would be worth the effort.
All in all I wish the CNC community was a bit more open source oriented in general. RepRap did a ton for 3d printing and I think there is still tons of room for the CNC hobbyist community.
answer b: I think you could get away with a non-hybrid stepper at a much cheaper cost. the hybrid will give a much better torque curve and that really helps if you are trying to thread at a high speed.
answer c: at $250 you are pretty close to a centroid or linux cnc setup. The utility of a full CNC lathe setup is questionable for the hobbyist.
answer d: I think engineering the costs down as low as possible would make it more approachable to small lathe owners and the more prevalent the MCU is the more people could contribute. I also think a fully canned threading cycle like the Russian ELS has would be fairly easy to do and would be worth the effort.
All in all I wish the CNC community was a bit more open source oriented in general. RepRap did a ton for 3d printing and I think there is still tons of room for the CNC hobbyist community.
I'm actually waiting for parts to give it a go on my PM1127. I ordered a larger stepper, not the hybrid. He mentioned testing both successfully, and I don't need the higher speeds. I suspect the stepper is overkill, but it wasn't that much more expensive than the smaller ones. I ordered a couple of belt and gear options, and the lathe has a gearbox for rough adjustment.
While I have ESPs and other micros around, the cost of the MCU wasn't an issue. I know that it works and it looks like it has plenty of headroom. The MCU has a free IDE that is cross platform. Not quite as easy to pick up as Arduino, but it doesn't look difficult to work with.
I know it's not much further to go full CNC, but I've been wanting to do at least the feeds this way for a while, and threading needs almost the same setup. That's enough for me, having a lathe has taught me that I always want a manual lathe around. CNC might be interesting on a second lathe in the future if I use my CNC router enough.
While I have ESPs and other micros around, the cost of the MCU wasn't an issue. I know that it works and it looks like it has plenty of headroom. The MCU has a free IDE that is cross platform. Not quite as easy to pick up as Arduino, but it doesn't look difficult to work with.
I know it's not much further to go full CNC, but I've been wanting to do at least the feeds this way for a while, and threading needs almost the same setup. That's enough for me, having a lathe has taught me that I always want a manual lathe around. CNC might be interesting on a second lathe in the future if I use my CNC router enough.
- Joined
- Mar 20, 2014
- Messages
- 418
Me too. They all seem to be on the same slow boat from China though. Probably not the launchpad though. That’s coming from Digi-Key “UK”, which probably means the US but without the import duties...
I got parts in, now I need time. I have a couple of things in the way before I can start installing things on the lathe. I did wire up everything to bench test and it looks good.
I had some issues with the stepper setup I initially got. I found that they have a resonance point that causes what looks like missed steps. I think it would work with a digital controller, but I just got the same hybrid servo he used.
Looking at the GT2 pulleys, it looks like the center is pressed in. Kind of annoying, as I need to cut a keyway in one. I think I'll also need to cut off the protruding bit with the set screws to fit the change gear axles on the lathe. I guess I could also press it out and replace it with my own.
I have a couple of things in the way before I can start installing things on the lathe. I did wire up everything to bench test and it looks good. I had some issues with the stepper setup I initially got. I found that they have a resonance point that causes what looks like missed steps. I think it would work with a digital controller, but I just got the same hybrid servo he used.
Looking at the GT2 pulleys, it looks like the center is pressed in. Kind of annoying, as I need to cut a keyway in one. I think I'll also need to cut off the protruding bit with the set screws to fit the change gear axles on the lathe. I guess I could also press it out and replace it with my own.
- Joined
- Mar 20, 2014
- Messages
- 418
Let the games commence!
- Joined
- Jun 16, 2016
- Messages
- 221
I’ll be interested to see how the Launchpad works for you. I used an Arduino Mega and pretty much the same parts you have, and have it working quite well but keep thinking up “one more” improvement. I suppose I should start participating in this thread and put up what I’ve done.
- Joined
- Mar 20, 2014
- Messages
- 418
For those of you attempting the Clough42 build, these might be of help. My LED&KEY module had different hole spacing to his, so 8 ended up remixing his 3d printed parts to fit my board.
www.thingiverse.com
Modified Clough42 ELS Panel by GreatOldOne
Remixed lid and spacer for the ELS project, as the LED&KEY TM1638 module I purchased from eBay had different hole spacing. Also the LEDs along the top are too close to the 7 segments to allow the top mounting holes to align with the spacer.These parts fit this...
www.thingiverse.com