128A3U ADC Puzzlin' Evidence

Go To Last Post
6 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have a product using the 192D3, which uses 12 of the 16 analog inputs of its *single* ADC. I built one up with one of my shiny new 128A3U samples, which as far as I can tell is supposed to have *two* separate 8-input adcs. I loaded my application *without any changes* onto the A3U device, and it works, with all 12 channels seeming to operate normally! Is there some sort of undocumented mux compatibility between devices going on here? My device should not be reading four of its analog channels without the second adc having been configured.

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The A3U parts have two seperate ADC converters, each containing four configurable channels. For each channel, you can select the various ADC pin inputs.

This means that the device is fully compatible with the old XMEGA devices - you just have a second copy of the ADC hardware in the device, giving a total of 8 configurable channels, and a maximum of two simultaneously sampled channels.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Dean,
So, is it true then, that the A3U has ADCA with a 16 input MUX connected to ports A and B (just like the D3), *plus* a second adc (ADCB) with 8 inputs connected to port B, and also each adc has four "channels" just like on an "A1"?

The manual and data sheets for the A3U are not clear on the mux structure.

Thanks!

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Where are you pulling the 16/8 MUX from?

Looking over the datasheets just now, it appears that both ADCs are entirely identical - they each have a MUX capable of selecting one of the 16 ADC input pins located on PORTA and PORTB (sourced from "31.2
Alternate Pin Functions" of the A3u specific datasheet). Each ADC has four configurable channels, and although only one physical conversion per ADC module can take place at the one time, but you can queue up all four to take place serially automatically for each ADC.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Nice! Looking at the pin function tables, the A3U is enhanced in many areas. Do know if the adc "factory programmed" gain and offset calibrations are functional? My next task will be testing the adc to see if the various defects are corrected.

Thanks!

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Haven't a clue - but supposedly the ADC has been substantially fixed in the new silicon. If it's in the datasheet, it probably works now.

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!