SERVOGRBL Breakout board for ESP32 DEVKITC (NodeMCU ESP32)

[TStar]

Hello everyone!

Some of you may be familiar with Bart Dring's incredible port of GRBL to the ESP32 https://github.com/bdring/Grbl_Esp32 and the excellent breakout boards available at his webstore. https://www.tindie.com/stores/33366583/

I was inspired by his work to design a simple breakout board which uses Arduino Pro Minis in place of stepper drivers to allow closed loop servo control of DC motors instead of steppers. This system builds on the fantastic Misan DCSERVO project by Miguel Sanchez https://github.com/misan/dcservo

I hope you will be interested to visit the repo for my SERVOGRBL breakout board at https://github.com/TSltd/SERVOGRBL

Anyone interested in testing the board can get in touch, once testing is complete I intend to make fully assembled units available for purchase, and I may make the full gerber files available as open source in due course.

Any thoughts or feedback on the project are very welcome. Thanks

Dan

[Doddy]

Please do me a favour and design your hardware to behave in a robust and dependable manner within the environment that you're advertising it. Don't think for a second that a RC filter will protect against spurious noise/triggers on sensor inputs; support 12-24V signalling; provide high current open drain drives for similar. Galvanic isolation where appropriate.

My point?, and my frustration is centred around many instances of people chasing problems with electronics that are not fit for purpose in a semi-industrial environment.

I've no problem with the use of microcontrollers here (some here do) and the ESP32 is a useful device compared to the AT Megas, but please research the real world problems people encounter with control systems with similar designs to those you link to,

[TStar]

Thank you for the advice, I have very little experience so far as I'm just a hobbyist, this is my first board to test the principle and will probably be in the testing stage for quite a while, hence why I'm inviting testers. Interference on encoder channels is certainly an issue I will look out for. The DCSERVO code uses rotary optical quadrature encoders but I believe a linear arrangement could also give good results in the manner of a DRO.
I understand the reluctance concerning microcontrollers particularly when dealing with stepper motors, but I am hopeful that a closed loop servo system will be quite a different story. Since it operates on a PID it should dynamically correct itself rather than suffering lost steps.

[TStar]

Update: made a Github Pages page which is a little bit easier to look at: tsltd.github.io

[TStar]

Galvanic isolation where appropriate.

Thanks again for this, I am revising the design to use optocouplers on all inputs as per the Woodpecker32 https://easyeda.com/andreasjakob/esp32grbl_copy and will add them to the encoder lines too

[Doddy]

If you want to take this IM please feel free to do so, this isn’t the place to iterate a design. Hint: consider how you’re going to support a differential drive from the encoders, if someone was sensible enough to choose such an encoder for a machine.