Project Logs

Collapse

• Finishing the project / fixing a noise problem

  AKA • 09/21/2015 at 01:02 • 0 comments

  Finally the 4-pole terminal blocks arrived from digikey and I was able to complete the first version of the project. Here's a photo of the installed terminal blocks.

  I found, to my dismay, that there is some pretty untenable noise on the stepper lines - both when powering up and just in general. I threw a large electrolytic capacitor on the barrel jack terminals that appears to have mitigated most of the problems. Here's the cap I installed - this greenwire hack worked fine, but I'll update the board files to accommodate a real decoupling cap soon.

  Three things prevent me from marking this project "done" and recommending that others try it:

  • There's still worrying noise on the stepper lines when the board powers up. Specifically, in the ten seconds that the Pro Trinket's bootloader pauses before initialization. If anyone had any advice about this, I'd be much obliged. (Should I remove the bootloader, which I don't want to do? Should I add pullup or pulldown resistors on the STEP and DIR lines to the EasyDriver?)
  • The EasyDriver gets uncomfortably hot during normal use. I don't think this is out of spec, but I think my design is poor in this regard - it places the back of my hand in close contact with the stepped board. I should design an enclosure to give the user's hand some space away from this!
  • The encoder PCB footprint needs to afford a better mechanical connection to the board. Right now, the encoder is just attached via the three terminals and not the two larger gripper tabs, and I worry I will eventually stress and break the encoder's tabs/

  Improvements aside, I'm happy to be able to move on to getting the rest of the K40 electronics working. Still a long way to go there, though!

• PCBs back from fab, testing OK

  AKA • 09/13/2015 at 02:47 • 0 comments

  The OSH Park boards came today, so I built one of them and did some initial testing.

  I tested the board this morning and all seems to be working properly. I'm still waiting for the motor screw terminals so it's not "done" yet, but all systems are working.

  In the repo, I updated the firmware to make MS1 and MS2 (microstepping settings) addressable by the Trinket. I'll add further firmware updates as they're made, but there probably won't be any large changes.

  I'm thinking of running one more rev of the board to add mounting holes for an enclosure, and a better footprint for the rotary encoder, whose fit is not great at the moment.

• Eagle boards almost ready to be sent out

  AKA • 08/31/2015 at 15:22 • 0 comments

  I spent some of last night laying out the first rev of the board for this project - you can check it out now in the Github repo if you like.

  Even if the board comes back and is functional, I will probably do a second run with more of an eye to the ergonomics - I need to figure out exactly how I'd like to use this, how it should feel in my hand, and whether or not to add extra buttons (ie, toggle the stepper's "enable" pin to allow it to be manually moved, allow inputs for limit switches, and provide an optional header for more common OLED screens)

  I'm not looking to make this a very generic controller - it's pretty specifically a one-channel jog for CNC devices. That said, let me know if I'm missing a crucial feature, it'd be nice ot hear what you think!

View all 3 project logs