Found this new board cools really compact and cool. Eager to hear your opinions on this
1 Like
Will you open a thread for every single esc you find?
…if it had CAN-bus instead of RS485 it would be interesting.
You also have to mount the controller directly to the motor for mag-sensor support, which is a limitation.
Sampling shunt resistors without pre-amplifyer is also not the best way. But it’s a trade off for being tiny.
1 Like
BTW: have you seen their “jumping squirrel” presentation? They use a pretty neat way to add a “gearing” between motor and leg by simply twisting a cord which then winds up.
1 Like
Hi that’s my project! Thanks for the feedback. I chose RS485 instead of CANbus because the microcontroller (STSPIN32F0A) doesn’t have CAN support and the RS485 transceiver can be tiny (THVD1420DRLR). That’s a good point though, it’s a bit of a hassle to use RS485 instead of existing devices for CAN, but if you want you can try my solution. The new version v3 has SPI solder pads for connecting an external encoder, so the motor doesn’t need to be mounted on the PCB anymore (though the solder pads are tiny and annoying to solder to, I might add an FFC connector in the next revision). As for the sampling shunt resistors, the specific microcontroller I’m using actually has amplifiers built in, so I can use the full ADC range. It’s a chip from ST meant for BLDC FOC and has built-in gate drivers too, so it’s super convenient for compact ESCs.
6 Likes
Hi @qwertpas,
after seeing your presentation video, I made my own “cord wrangler” actuator.
Have you seen my video?
Is that your design, I’ve never seen that before.
What type of cord do you use? It looks like rubber band…
1 Like
Just saw your video, I like the demo with such a large weight! The nice thing of the twisted string actuator (there are some papers about it, it’s not my original idea) is that the longer and thinner the string, the higher mechanical advantage there is. I’ve done some testing here: https://pintobotics.substack.com/i/137821874/twisted-string-experiment
It’s nonlinear and I found it is most consistent to get a string that is unbraided and doesn’t stretch much such as UHMPE bowstring
1 Like
Thanks for the link!
Would it be possible to add a strech-coefficient to your equation?
That way we could use cheap nylon fishing line
BTW: have you seen my video titled robot arm throws a ball?
It uses pretty similar energy storage as your squirrel, but is more compact.