Only moved the needle on the project a little over the last few days, mostly cleaning out the lab to get it ready for another winter trapped indoors. Lots of sorting, moving, and getting the tools set up right. Under no circumstances should you watch Adam Savage on Tested, working in his space while doing this… it will cause all manner of organizational envy.
– Added a 6+1 pin header to the keypad instead of running it to the push-terminals. The 7th pin (the red line of the ribbon cable) is ground to the keypad.
– Added a JST and +5v wiring to the power supply. The JST is surface mount so it’ll be a kludge getting it on the main logic board. The +5v is for the motor driver board.
– Added the cabling for the auto/manual switch and LED. This needs a bit of explanation, so here goes:
- The intention is to pull the auto/man pin assigned on the MCP23017 to ground if set to manual. (switch closed). Ground to the switch centre pin, and return to the MCP GPIO pin.
- This should also light the LED, so this is also the LED cathode, allowing ~18mA to flow.
- In the switch open position the LED has +5v to it’s far side, (and flows thru a ~150Ω resistor), but no substantial ground path for any current flow (and the MCP has pull-ups set).
- The voltage at the MCP pin should be ~3v, so essentially this is in a pull-up condition to stop the MCP GPIO pin from floating and giving stray logic-low readings. In the event the MCP isn’t configured with pull-ups, this is still (just)within the per-pin tolerance of the MCP to sink up to 20mA.