Picomite issues with MCP3002, dual channel ADC SPI

Author	Message
apalert Newbie Joined: 06/07/2023 Location: Australia Posts: 30	Posted: 09:31am 03 Feb 2025         TassyJim said   I forgot to mask off the higher bits in the 16bit code data0 = data0 AND &h3FF etc doing that will get rid of the obviously wrong readings. In the 8 bit version I should have masked of bits for each SPI read before adding them together. Jim, did as you suggested and you have nailed it! Results are amazing. Initially results were +/- 3. However I stabilised VDD on the VDC at 3 volts with an LM4040 and forced Pico into PWM mode while reading, now we get +/- 1 as per spec sheet. Was able to remove all pauses and bring clock back up to 1000000. Time to take 20 readings around 7.5m/s when not printing result of each loop. Hard to say which is better, 8 bit SPI or 16. Accuracy seems the same. 16 bit is a tiny bit faster. Thanks so much for your input.