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.

Reading value from ads7828 is not stable??

Other Parts Discussed in Thread: ADS7828

Hi all,

I am using ads7828 i2c slave ic to read data, used channel 0 , 1. using single ended mode.

i tried with both internal reference, Vref = 0,Vdd = 5v and external reference Vref = 5v and Vdd = 5v. 

in both the connections, I am not getting the stable value for constant analog input 3.3V.

my circuit design is shown here. 

In address mode, i tried with(A0 ,A1 - gnd and A0 supply to A1 ground) , value not stable.

Can anyone help me to resolve this issue,..

  • Hello Pon,
    Please check the suggestions below and let me know your result:
    1. When your signal is not differential, the COM pin should be connencted to -IN(ground) and one Command Byte should be written to configure the inputs to Single-Ended shown in No. 11 page of datasheet.
    2. Also please notice that the internal reference defaults to off when ADS7828 power is on. The internal reference is only 2.5V. The minimal external reference is 50mV.
    3. Please use 2k or 10kohm as pull-up resistors on SDA and SCL lines.

    Thanks and regards
    Dale
  • Hi Dale Li

    Thanks For reply, i left my COM pin as No Connection. i use 1Kohm as pullup resistor. but other i2c slave devices DAC , Relay they are well working in 1kohm pullup resistor. 

    Initially i will try with COM to ground, then one by one. i will reply you after checking.

    Thanks

    Pon Alagappan

  • Yes Dale Li,

    Earlier i was facing heat issue and some times i got constant values and sometimes i got random values.

    Now it is bit improved.

    No heat issues, i did the changes (COM-GND) (2K pullup resistor)

    now, i did not see any random changes, But still values fluctuates around 30(+/- 15), is still high. i need more accuracy . in data sheet page no 9: it mentioned

    the internal noise of the converter typically contributes only 0.32LSB peak-topeak of potential error to the output code


    but for me it affects last 5 LSB bits. Please help me on this, i expect support from TI forum, finding solution as earlier as possible is mandatory for me

    I gave constant 1v in analog channel 0. Vref 5v, Vdd-5v. using single ended. 

    my results:

    837

    857

    838

    839

    835

    817

    842

    830

    842

    838

    839

    835