Re: [Gumstix-users] Gumstix on-board ADCs

[Steve S. <sa...@gm...>]

On Fri, Oct 19, 2012 at 12:14 PM, jumpnowdev <sc...@ju...> wrote:
> I'm getting some bad measurements trying to use all 6 of the Gumstix ADCs at
> one time.
> It's a 3.2 kernel using the new madc driver accessible through sysfs.

I believe that this is an issue that has been discussed on the list
earlier. If so, it is related to interactions between the musb part of
the 4030 and the MADC part of the 4030.

These interactions are only on channels 3 through 6, and if you look
at you readings channels 2 and 7 look reasonable.

You might try repeating the test with something plugged into the musb
port.  I suspect you will get the desired results then.

There is a patch submitted to the list that google can probably find.

Steve

> The problem I am seeing is a signal applied to one ADC is affecting the
> other ADCs.
>
> I've gotten the same results with two different COMs (one Earth, one Fire)
> with two
> different expansion boards (Summit, Chestnut43) using two different
> breadboards.
>
> Here's the test:
>
> A signal applied to one ADC at a time.
> All other ADCs and AGND are connected to the signal ground.
> The signal is from a bench power supply.
>
> The first and last rows are with all ADCs connected to AGND.
>
> 0.5v DC signal
>
> root@overo:~/temp# ./polladc -d100 2 3 4 5 6 7
>
> (use ctrl-c to stop)
>
> ADC                2      3      4      5      6      7
> Read      158:    12    134    129    129    122     14
> Read      233:   535    139    149    131    124     17
> Read      343:    17    535    134    129    122     17
> Read      431:    14    134    540    207    171     17
> Read      526:    14    134    141    540    136      9
> Read      608:    12    134    134    122    525     12
> Read      705:     4    134    146    122    139    540
> Read      786:    14    149    134    124    127     21
>
>
> 1.2v DC signal
>
> root@overo:~/temp# ./polladc -d100 2 3 4 5 6 7
>
> (use ctrl-c to stop)
>
> ADC                2      3      4      5      6      7
> Read       76:    12    134    146    124    129     14
> Read      205:  1234    149    146    127    134     17
> Read      307:     7   1126    122    124    141     12
> Read      408:    17    207   1075    630    205     17
> Read      477:    19    131    464   1080    290     19
> Read      570:    14    127    109     78   1138     17
> Read      654:    19    136    131    122    134   1241
> Read      688:    19    122    122    127    134      9
>
>
> I've only ever used one of the ADCs at a time and never noticed this.
>
> Maybe someone else could verify this or explain what I'm doing wrong?
>
> The program I'm using can be found here
> https://github.com/scottellis/polladc
>
>
>
>
> --
> View this message in context: http://gumstix.8.n6.nabble.com/Gumstix-on-board-ADCs-tp4965762.html
> Sent from the Gumstix mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_sfd2d_oct
> _______________________________________________
> gumstix-users mailing list
> gum...@li...
> https://lists.sourceforge.net/lists/listinfo/gumstix-users