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.

TMS320F28076: Unexpected ADC channel differences when migrated from F280049

Part Number: TMS320F28076


Tool/software:

Hi Experts,

My customer is migrating from F280049 to F28076, however, they are seeing a unexpected ADC channel difference when sampled from ADCA and ADCB .

Ax and Bx (Missing the exact channel number, will add this soon) samples same signal with different RC filter. However, A is higher than B by 500 ADC counts.

We've done the following test

1. Connect the two pin together, A was still approximately 100 ADC higher than B (difference reduced).

2. When the two hardware channels were swapped before the RC, the readings remained the same

3. Channel 0 of channel A and channel 12 of channel B are internally connected to the same pin. We attempted to read the same sample in both blocks and changed the trigger source. The results were consistent, with only a difference of approximately 10 ADC.

4. We attempted to change the S/H time between the two samples (these are dummy samples to aoivd channel cross talk) before channel B to 0x26 (320ns) and also to 0x26 on the channel under test, but there was no improvement.

5. We attempted to change the sampling time and trigger source, but there was no improvement.

6. When the same clean voltage (2.5V) was injected externally, the two samples read approximately 3200 ADC, a difference of approximately 30 ADC.

Any idea what's causing the difference?

  • Hi ,

    Couple of questions that might help to narrow down this further:

    • What Hardware is being used by customer to evaluate the TMS320F28076 device ? 
      • Just want to verify if ADC Reference is provided properly, also let me know the reference value 
    • Will need info on how are the inputs driven to the ADC (Buffer & RC circuit with values) 
    • Apparently in this device some of the channels are pulled down by an internal resistor (50Kohm)  [Refer Datasheet section 5.2.1] 
      • Can you confirm the ADC A & B channels here to verify if any channel chosen has this 50K ohm resistor PD 
      • That's the reason need to verify if the input is driven by high impedance or low impedance source (Input network)
    • There is one more info hope this is taken care ADC S/H W requirement is different for this Device ADC (min ACQPS ==>100ns) vs the F280049ADC (min ACQPS ==>75ns) 
      • Can you also share ACQPS value used for both channels to verify the same 

    BR,

    Nilesh

  • Yes, We use a ADCB1 pin and there is a 50kw internal pulldown on this pin. we will check for it. Thank you very much!

    ...

  • Why we add a inner pulldown 50k restor, and how about the parameter about the temperature drifting?

  • Hi ,

    I will get the details on the need for resistor and get back.

    Question :  On the Input network question,  is ADC input driven from a High impedence source (high value resistor) ?

    BR,

    Nilesh

  • Hi,

    The inner Pulldown resistor is for DAC actually so for All pins Mux-ed to DAC output this is present (3 pins). 

    The resistor value drift is about 1% across Temperature. 

    BR,

    Nilesh

  • thank you Khanna, it helps us.