Starting this day, I'm calling this project finished and ready for #The Square Inch Project
I believe it met the goals I set up in my first thoughts and after having problems with hardware and totally failing with first ADC and hardware redesign I managed to finish it into satisfactory shape, to point out a few project logs from total 19, not including this one.
Here is a few things I'd consider when doing micro progmeter v2:
- fix leakage of analog switches for resistance measurement.
Leakage of BSS84 transistors causes non-zero voltage (~ 0,01V) on voltage measurement range with measurements probes floating. Once shorted, it shows 0,000V, as expected. This is more like cosmetic problem, IMHO - but one could hopefully cure it by connecting common point of R6 and R7 (refer to schematics on github) not into analog 3,3V node, but into D electrode of another PFET switch with BSS84. This point should be grounded via appropriate resistor, too high value would cause poor leakage masking, too low value would cause too high current draw. Something like 1-50kOhm (to be tested) should do the job. The third transistor should be kept closed (G on high potential) when performing voltage measurements.
- find better MCU, this one is really FLASH starving.
Though PIC16F1829 or 1509 are my favourites when it comes to small projects, I seriously underestimated how much FLASH hungry the floating point routines are. There are options like STM32F031, but I like Microchip parts and this project was aimed at pointing out other than the most mainstream paths in field of microcontroller based designs, so I'd stick with this manufacturer.
- smart watches?
Honestly, I hate the hype around smart watches. Smart watches are here for a long time, even the DIY versions, so it is basically nothing new. I was hanging out with my friends and showing them this project, somebody pointed out its remote similarity with watches
DMM watches, pretty pointless, but cool.