Hello again, dear reader.
If you've been following us on Facebook (and you really should be!) you'll already be privy to the fact that we reached another milestone this past week: MoNet's officially live, broadcasting data. As Jack said, this brings us one step closer to completion.
The next step - make sure it can manage multiple requests for a prolonged length of time. To give it an additional "production environment" feel, there wouldn't be any human intervention during this period either. So how did it go? I'm glad you asked:
That's right - zero failures!
While there's still more to be done, it's encouraging to see MoNet's capable of handling the workflow even now. You can probably tell there's a downside though.
We lost one of our NodeMCU units (the technical term for what happened is "fried,") something we talked about in last week's post.
This also brought some issues with the board design to light. MoNet's motor drivers need 5 volts, but one of them was only receiving 3.3 volts. As a result, motor direction changes weren't being picked up.
Time for a Revision!
To deal with this, the board design's been updated:
- The ESP8266 chip will handle the network...
- While the AtMega328 will handle driving the motors.
Since the ESP8266 itself couldn't drive the motors with everything else going on, both of these pieces are now running as Arduinos and the work's now divided equally between them.
As a bonus, the motors are running much smoother now that we're not using a WiFi MCU.
Well, this leaves just one more major piece of hardware to complete: the brush head. In a not-at-all surprising coincidence, the parts for it arrived just this week! However, with the Labor Day holiday coming up state-side and team travel in the near future, build session #2 won't be taking place for a while.
As soon as a date's finalized for it, we'll post on our Facebook/Twitter pages. Until then, talk to you next week!