I must laugh. I thought.. "this guy is off his rocker" soldering this board... I will now confess you're the Soldering Jedi. My late father (an electrical engineer) would have absolutely loved this. Back when I was young (long, long ago in a galaxy far, far away) we had a General Elec-Trak E20 electric riding mower. Remarkable video.
This looks like a "bang bang" ESC. I'd love to see someone spin up a vector-sinewave ESC (requires very high speed switching 1MHz+) with a balancing reconstruction filter at the output. Like what they do for high efficiency large format VFDs, or Trinamic stepper drivers.
Thank you! The FOC is not possible, because the RP2040 doesn't have enough analog inputs for all phase currents, the input voltage, MOSFET temperature and motor temperature. Since the goal for this design was to keep it simple and cheap (so that we'll get it working soon), we decided not to measure the phase currents individually, but only the resulting current draw (for torque control).
@@ClemensElflein Sorry to bother you again: I want to use your xESC in a different project, but the motor will not have sensor feedback. Is it also a hardware limitation of the RP2040 that sensorless control is impossible? Or is it possible with the right software? Cheers!
I have no idea; The MOSFETs are rated up to 20A, but the pin headers will probably be the limiting factor here. I'd say about 5A continuous should be no problem. For the OpenMower project, we need currents of about 1A and this is easily doable.
Hi Clemmons, might as well get acquainted, bc I'm planning to go this project! I already started reading & following your stuff. I've got a different "microscope" but it'll work just fine. I have plenty of uC experience, tools, etc. I noticed you didn't use the hotplate, I thought you were going to use that for reflow. ( I saw an Adafruit video where she used one modified to do reflow of small 2-sided PCBs, I looked on their website, didn't see it, so I emailed them asking about it, I got a reply that was kinda snotty. That made me think the mods must be "widely available" on the 'net, but I haven't run across anything yet. Know anything about it?) I'm planning to do the Open Mower project, we have the perfect yard for it, very flat (also pretty small). Fence ,& house bounding the back yard, but nothing around the front--and we're also a corner lot. That's why I think it'd be perfect. I just learned about the project. I'm in the process of getting caught up with the Discord. I'll need to look for a decent but cheap mower to modify. I've been wanting to do this for years, but until I saw your GPS-RTK design, I didn't like any that I saw. Also what's slowed me down is the fact that, since we moved to this house, which is tiny, I don't have a work shop, or even a place to set up a card table! I managed to put a (less than) 2' square piece of plywood on top of a file cabinet for soldering, but the iron & other stuff takes up most of the space, lol. I recently hung some pegboard on the wall with small bins to hold parts, but it's nothing like the shop I built in the yard of our previous house: had all the little drawers of parts, & cabinets on the wall, 2 sir-down small-assembly benches, 1 stand-up heavy bench for wrestling & hammering on stuff (with vise & press), table saw, floor standing drill press, cabinets & shelves galore, etc. Unfortunately, had to leave it behind when we moved, too costly to move this far, and no place to put it in this yard. So, things go slow here. (Still have all the small tools, just not in one place!) Otherwise I would've probably already worked out doing this project myself by now, I'd like to think. Anyway, I hope you've got things together, by now. Like I say, I'm getting caught up on Discord, etc. Excellent work, by the way, dude!
Yes I know, I also do like FOC, but the focus for this design was to keep firmware implementation simple and cost low, since it will be only an intermediate solution until the semiconductor shortage is better. Sorry that the new design doesn't work for your use-case. If it's only start-up torque you're looking for you will need a sensored motor and the designs should be similar. With FOC the motor runs smoother, with less noise and torque ripple, but sensored startup with block commutation will also give good performance. If you want to discuss your project, you can also join the Discord, it's not only for OpenMower related topics.
@@ClemensElflein Thank you for your answer! When you say intermediate, does that mean you will update the design or have a totaly different design with an dedicated IC?
The new version is not able to run the VESC firmware, the STM32 version should support it, because it's just a VESC in another package. I haven't tried it yet, though.
You meant "now available"? I think there might be several reasons, one being the sponsor of the video... But another comment mentioned that it was a temporary design because of Pi board shortage.
@@_J_A_G_ no, I mean "not" available off the shelf. Because hardware that already exists on the market, developed by experts and ready for use is just a waste of time (for me) to delevop at home. There are thousands of ESCs available, regardless of market situation.
I must laugh. I thought.. "this guy is off his rocker" soldering this board... I will now confess you're the Soldering Jedi. My late father (an electrical engineer) would have absolutely loved this. Back when I was young (long, long ago in a galaxy far, far away) we had a General Elec-Trak E20 electric riding mower. Remarkable video.
Incredibly informative video with Nice design, well done, Clemens 👍👍
Nice job. Keep going open source ROCKS!
Excellent work on this latest video! The footage from the microscope is very cool.
Thank You
This looks like a "bang bang" ESC. I'd love to see someone spin up a vector-sinewave ESC (requires very high speed switching 1MHz+) with a balancing reconstruction filter at the output. Like what they do for high efficiency large format VFDs, or Trinamic stepper drivers.
I've used a small bake and grill kitchen oven with a thermometer to reflow though you don't have as much control over the reflow timing.
Awesome workstation setup! Plenty of space and all the tools
Impressive work! Keep it up 👌
@Clemens Elflein: dude! Such a cool project! Thanks for sharing, I'm loving it and hope to be able to build my own OpenMower soon. 👍🏻😎
Really nice project! I would be interested in a deeper explaination why FOC is not possible at all?
Thank you! The FOC is not possible, because the RP2040 doesn't have enough analog inputs for all phase currents, the input voltage, MOSFET temperature and motor temperature. Since the goal for this design was to keep it simple and cheap (so that we'll get it working soon), we decided not to measure the phase currents individually, but only the resulting current draw (for torque control).
@@ClemensElflein Sorry to bother you again: I want to use your xESC in a different project, but the motor will not have sensor feedback. Is it also a hardware limitation of the RP2040 that sensorless control is impossible? Or is it possible with the right software? Cheers!
Thanks! Looking forward to more vids.
Thank you for the work you are doing.
One year and no other update. Seems like terrible news to me. 😢
Hi Clemens, please do some follow-up videos. Many people are waiting.
So fascinating to see
Awesome !
Very cool, in every matter.
Its been a while since this video. How is it going.
Great Job. 🙂✌🏽
If your solder "roll" on the stencil, you will have better quality of soldering pst result, with a thin metal plate you can have better result
great channel
Is it possible to get jlpcb or similar to produce this card and the main board?
sure but it's only economical to do in bigger batches. the setup for one board is the same cost regardless if you get 1 made or 10000 made
How many amps is this ESC capable of?
I have no idea; The MOSFETs are rated up to 20A, but the pin headers will probably be the limiting factor here. I'd say about 5A continuous should be no problem. For the OpenMower project, we need currents of about 1A and this is easily doable.
This is amazing, would love to. Build my own, but without any program skills (yet) might be impossible
I love this.
Hallo Clemens
Gibt es auch ein CPL file dazu das man sich das löten sparen kann? :)
Fantastic!!
Extraordinaire
Hi Clemmons, might as well get acquainted, bc I'm planning to go this project! I already started reading & following your stuff.
I've got a different "microscope" but it'll work just fine. I have plenty of uC experience, tools, etc.
I noticed you didn't use the hotplate, I thought you were going to use that for reflow. ( I saw an Adafruit video where she used one modified to do reflow of small 2-sided PCBs, I looked on their website, didn't see it, so I emailed them asking about it, I got a reply that was kinda snotty. That made me think the mods must be "widely available" on the 'net, but I haven't run across anything yet. Know anything about it?)
I'm planning to do the Open Mower project, we have the perfect yard for it, very flat (also pretty small). Fence ,& house bounding the back yard, but nothing around the front--and we're also a corner lot. That's why I think it'd be perfect.
I just learned about the project. I'm in the process of getting caught up with the Discord. I'll need to look for a decent but cheap mower to modify. I've been wanting to do this for years, but until I saw your GPS-RTK design, I didn't like any that I saw. Also what's slowed me down is the fact that, since we moved to this house, which is tiny, I don't have a work shop, or even a place to set up a card table! I managed to put a (less than) 2' square piece of plywood on top of a file cabinet for soldering, but the iron & other stuff takes up most of the space, lol. I recently hung some pegboard on the wall with small bins to hold parts, but it's nothing like the shop I built in the yard of our previous house: had all the little drawers of parts, & cabinets on the wall, 2 sir-down small-assembly benches, 1 stand-up heavy bench for wrestling & hammering on stuff (with vise & press), table saw, floor standing drill press, cabinets & shelves galore, etc. Unfortunately, had to leave it behind when we moved, too costly to move this far, and no place to put it in this yard. So, things go slow here. (Still have all the small tools, just not in one place!)
Otherwise I would've probably already worked out doing this project myself by now, I'd like to think.
Anyway, I hope you've got things together, by now. Like I say, I'm getting caught up on Discord, etc.
Excellent work, by the way, dude!
Thank You! Welcome to the Project!
Oops! Watching again, I saw that you did use the hot plate, and not the reflow oven. My mistake.
Hi Clemens, any news?:/
I would like to use the xESC2040 for a different project, but i need FOC for it. Its a pitty you removed it :(
Just as an explanation: the Motors start turning much smoother with FOC then the 6 phase commutation
Yes I know, I also do like FOC, but the focus for this design was to keep firmware implementation simple and cost low, since it will be only an intermediate solution until the semiconductor shortage is better. Sorry that the new design doesn't work for your use-case.
If it's only start-up torque you're looking for you will need a sensored motor and the designs should be similar. With FOC the motor runs smoother, with less noise and torque ripple, but sensored startup with block commutation will also give good performance.
If you want to discuss your project, you can also join the Discord, it's not only for OpenMower related topics.
@@ClemensElflein Thank you for your answer! When you say intermediate, does that mean you will update the design or have a totaly different design with an dedicated IC?
Does it support silent HFI like available in the 6.0 firmware?
The new version is not able to run the VESC firmware, the STM32 version should support it, because it's just a VESC in another package. I haven't tried it yet, though.
hi Clemens. kannst du mal den schaltplan des esc als jpg oder pdf verlinken? ich hab schon bissl im github gesucht aber finde mich nicht zurecht.
ich hab KiCad installiert und konnte damit die schaltplane aus dem github angucken. vielen dank dass du uns alles open source angucken lasst.
Alles klar, freut mich, dass es geklappt hat!
Hadkoa!
I don't really get why you designing your own ESC. Why not focus on hardware which is not available off the shelf?
You meant "now available"? I think there might be several reasons, one being the sponsor of the video... But another comment mentioned that it was a temporary design because of Pi board shortage.
@@_J_A_G_ no, I mean "not" available off the shelf. Because hardware that already exists on the market, developed by experts and ready for use is just a waste of time (for me) to delevop at home. There are thousands of ESCs available, regardless of market situation.