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.

ADC on 28335

Hi,

I'm using TI's experiment kit for 28335.

I connected a 2V DC source to ADC0 + GND and then ran the sample code: adc_seqmode_test

After few iterations I checked the array:  SampleTable

According to the manual: Digital = 4096 x Analog / 3 

This is becuase the signal ADCLO is connected to GND in this kit.

So why am I getting differences between the expected and actual digital value ?

The expected value is: 4096 x 2/3 = ~2730. But I'm getting ~2800 and more.

Should I use the value of OFFTRIM register in my calculations ?

Thanks.

  • Keep in mind that there will be accuracy error per the datasheet to any conversion, however, the most this should be at this code is ~35 codes of error.  Is there another source of offset error in the system, if so you can use OFFTRIM to correct this out. 

    I would check to see if increasing the ACQ_PS improves the code error, in which case the input voltage to the ADC is not driven properly to support the min S/H window of the device.  

     

    Best,

    Matthew 

  • Hi Matthew,

    You wrote "the most is ~35 codes of error".

    Do you mean that the maximum difference between the actual and expected digital value is 35 ?

    Let's say that I found that the above difference is X. Will this X be the same for all conversions ?

    Shouldn't the calibration code supplied on the DSP's flash handle this error ?

    Thanks.

     

     

  • Accuracy of ADC reading depends on many factors.

    - Sample/Hold time

    -Stability of DC input (if its giving 2.01V instead of 2V, then the could will varying.)

    -Reference voltage.

    -Stability of reference voltage