Hi group,

I am currently using the ezdsp f2812 and trying to study the ADC 
module. A couple of questions need someone input and help. Firstly I 
connected the VREFL0 to AGND which is 0v. I ran one of TI examples 
named AdcSeqModTest. This example is very simple and it works OK, 
but the reading from ADCRESULT0 is not what I expected. I measured 
the pins ADCREFM = 0.971V and ADCREFP = 1.941V. Also out of the 
running if I left the ADCINA0 pin floating the ADCRESULT0 is around 
18736 & if I grounded the ADCINA0 pin, the ADCRESULT0 is around 976. 
I know this not right but I don't know how to fix and to read a 
reasonable value.

Thanks,

Wen-Ruey

You can post a message or access and search the archives of this group on DSPRelated.com:
http://www.dsprelated.com/groups/c28x/1.php

_____________________________________

 

Hi Wen,

I do not totally understand your problem but I would
share some thoughts about the ADC reference. You've
mentioned that the ADCREFP is 1.941V and ADCREFM is
0.971V. The difference voltage is the bandgap voltage
which comes to about 0.97V. Ideally (as per the
datasheet), this should be 1.0V which when multiplied
by 3 gives the ADC full scale reference of 3.0V. Since
you do not have 1.0V as the band-gap voltage, your
reference is not 3.0V but it is 0.97X3 = 2.91V only.

The implication of this will be in the ADC reading.
Assume that we are giving a absolute voltage of 1.5V
to the ADC input. If the band-gap reference was
correct value of 3.0V, then the ADCRESULT register
will have the following value

(1.5/3.0) * 4095 = 2047 = 7FFh. (The actual register
will have the value 7FF0h because it is right
justified)

But in your case since the internal voltage reference
is only 2.91V, the reading of the same 1.5V will be

(1.5/2.91) * 4095 = 2111 = 83Eh. (The actual register
will have the value 83E0h because it is right
justified)

Maybe your difference in the band-gap reference is
causing this discrepancy. You might have to check
that.

I do not know if I have addressed your problem
correctly but this was one issue that I faced with ADC
readings & some compensation in the ADC readings
helped me to overcome this. Thought it might be useful
for you and some others in the group who might be
facing a similar problem.

Thanks,
Vishal.
--- wrhwang44720 <w...@timken.com> wrote:

> Hi group,
> 
>  
> 
> I am currently using the ezdsp f2812 and trying to
> study the ADC 
> module. A couple of questions need someone input and
> help. Firstly I 
> connected the VREFL0 to AGND which is 0v. I ran one
> of TI examples 
> named AdcSeqModTest. This example is very simple and
> it works OK, 
> but the reading from ADCRESULT0 is not what I
> expected. I measured 
> the pins ADCREFM = 0.971V and ADCREFP = 1.941V. Also
> out of the 
> running if I left the ADCINA0 pin floating the
> ADCRESULT0 is around 
> 18736 & if I grounded the ADCINA0 pin, the
> ADCRESULT0 is around 976. 
> I know this not right but I don't know how to fix
> and to read a 
> reasonable value.
> 
>  
> 
> Thanks,
> 
>  
> 
> Wen-Ruey

You can post a message or access and search the archives of this group on DSPRelated.com:
http://www.dsprelated.com/groups/c28x/1.php

_____________________________________