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.

A/D low input impedance

RE: F28021 picilo processor    --- TMS320F28021PTT at 40MHz.

 

We are having some unexplained input impedance problem with ADC IN B3.

 

When there is no flash code, the input impedance to ADC IN B3 is very high like it should be.

When we load some dummy code, that just loops without reading the ADC, the input impedance is still very high.

 

When we read from the ADC IN B3, the input impedance drops to 9.73K, which causes extreme error.

 

All of the other ADC inputs have high impedance and read correctly.  There must be some problem we can’t understand with the set-up or operation.  Is there something different about this input ?  (IN B3 is supposed to be input only, not I/O like other A/D, is this correct ? )

Also the source to the ADCINB3 feeds to the ADCINA4 as a comparator, and the comparitor causes some, but not all of the load.

 

Thanks for any help you can give.

 

William Reed

  • William,

    You are correct ADCINB3 is same as the other non-IO ADC channels, and should behave the same way.  One item you mentioned

    "Also the source to the ADCINB3 feeds to the ADCINA4 as a comparator, and the comparitor causes some, but not all of the load. "

    That may be the answer, when the comparator is active there is ~100kOhm feedback resistor between the comparator output(either 0V or VDDA depending on the inputs) and ADCINA4.  If you have a relatively high input impedance on the voltage source this will cause changes to the resistor net and as such changes to the voltage on the pin.

    TI has provided a bit to turn off hysteresis for each comparator, which will in turn remove the feedback resistance from the path.  These bits are in the AdcRegs struct in the header files, and are also documented in the latest ADC UG on the web.

    Please give this a try and see if it solve the problem. 

    Another method would be to remove ADCINA4 from the circuit all together to make sure the above is 100% of the problem, not sure how easy that is depending on your board.

    Best,

    Matthew