Ooh very interesting I was thinking about doing this the other day but not sure what all I should be concerned about. Do you know feel a difference due to the difference?
Also, @Evwan welcome to the 4wd gang . Looks like it’s going to be a fun build.
Wow, what a sentence I have the duty cycle current limit start set so that I don’t reach the set maximum duty cycle on either ESC, therefore it feels just normal. I am not sure what would happen if I were to actually reach the maximum duty cycle though
It feels the same compared to just having lower currents set on the front (which is geared faster but limited down via the duty cycle limits). The speed limiting factor is always going to be the slower drivetrain though, so if you can it’s a better option to have the front and back geared for the same speed. And if you do this you absolutely must setup the duty cycle limits correctly otherwise the vescs may act up at the top speed of the slower drivetrain.
The only reason why I did this is because this board was only put together for testing our traction control over canbus university project, and I didn’t feel like buying any extra parts (not even things like motor pulleys and belts to change the gear ratio). This board is going back to 2wd right after the exam day because I really don’t like the extra weight of 4wd on a mountainboard.
It’s not going too well considering that January is the exam period but I think we’ve done more than enough to have a good grade. We have a half-assed prototype at the moment that sometimes helps a bit but is super quirky and overall pretty dangerous. The whole project will have to be redone from scratch to make something that reliably works, because the programming environment we were using from the beginning is too limited. But we will only have the time for this after the exams.
But two of my friends whom I am working with on the project both want to build 4wd boards over the first half of next year, and the next semester’s curriculum is control theory which is very closely related (arguably, probably even required for this project), so I am confident that we will make it work well sometime next year! I will make a thread for it once it’s ready. We also plan to try and have a go at implementing 4WD torque vectoring but I have no idea if we can actually pull that one off.
@glyphiks has shared before that traction control in vesc currently will lock up all your wheels if one of the wheels locks up for any reason. Imagine getting a rock stuck in your open gears for one wheel and having all them lock up at the same time. Bad situation goes to worse situation real quick. Would be dope if @Dinnye is working on something that is a different approach here, that wouldn’t suffer from this failure mode.
Battery is fully glued (this picture is a bit old)
I also spent a while working on the stomp pads. It’s been a wild journey, but I think I have them in a good spot. I printed out a first prototype, but realized that I screwed up the curvature somewhere, causing it to not conform to the shape of the deck.
I just finished a 3 hour stent of CAD work to get the second iteration designed, but i’m out of TPU. The print settings for the original was 10% infil, 3 walls, 3 top/bottom on 98a TPU, and it had the firmness of a stiff closed cell foam, which was nice to stand on. This version also has significantly more concave as the originals were a bit too flat.
This pad will be clamped onto the deck with the binding screws, so it’ll be easily removable too. I’m also thinking of replacing the big channel with grooves for individual wires, which should tidy up the wiring.
One thing that i wish i had done with my stomp pads was to add additional holes so that the pads could be secured independent of the bindings. This is still on my to-do list