• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 80 subscribers
  • Views 777 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Inconsistency in MSP430x22xx User's Guide re: ADC10 SREFx ?

Peddler
Intellectual 365 points

Hi,

I'm making some voltage measurements with ADC10 and have run into something puzzling in the MSP430x22xx guide Rev F (slau144f.pdf).  On pg 20-25 it says the SREFx bits select Vcc as VR+ for 000 and 100.  However, on pg 20-3 the schematic indicates that the reference is actually AVcc.  AVcc is a different supply voltage and different pin than Vcc.  It's late and maybe I'm not thinking clearly but am I missing something here?  Anybody know which is correct?

 

  • 0
    Intellectual 345 points

    Hi Chris,

         When the ADC portion of the userguide talks about Vcc it means AVcc.

         Many of the MSP430 device/package combinations have both an analog Vcc (AVcc) and digital Vcc (DVcc), although some of the smaller packages do not (in this case AVcc=DVcc=Vcc).  Comparing ADC performance between devices that do/do not (such as the F2013 QFN vs DIP package options) shows that the ADC usually behaves a tad better when it has a supply (AVcc) disjoint from the CPU (DVcc).

         Even better, if you can stand the reduced input range, the on chip reference is likely a quieter choice than using your supply rails as a reference, although there are plenty of good reasons for using them.

          The datasheet should be corrected to refer to AVcc.

    -memoryleak

  • 0 in reply to memoryleak
    Guru 227245 points

    memoryleak said:
    When the ADC portion of the userguide talks about Vcc it means AVcc

    Not exactly. IIRC, the VCC/2 internal channel input is indeed taken from DVCC - if there is a separate AVCC and DVCC input at all.

    Anyway, AVCC should not differ from DVCC (the maximum allowed difference is a 200mV I think due to clamp diodes). Both are intended to be from the very same voltage regulator. Just the routing should be separate, so voltage drops over the DVCC supply wire caused by current changes on DVCC (due to core current consumption or other peripherals) won't influence AVCC. This is for both, AGND and AVCC. Both should be routed separately to DVCC and DGND.
    A series resistor (~10Ohm) right besides the AVCC input and a small capacitor are a fair substitute for a separate wiring if you don't have large current changes due to peripherals.

  • 0 in reply to Jens-Michael Gross
    Intellectual 365 points

    Thanks guys, that clears things up.  I noticed the datasheet has an errata document but apparently not the UG.  Anyway, yes, I'm using similar routing to what Jens described.  In my case, it's about potential noise issues so I needed to understand/verify the internal design.  Thank you!

     

**Attention** This is a public forum