AD9643

Hi,

i am working on AD9643 ADC, i am not able to do the data output from this fro different input voltages as per the table11 given in datasheet.

i used a formula

DATAin decimal=(applied input volatge*2^14)/1.75

am getting

vin                 data            binary

-0.875            E000          10 0000 0000 0000  // as per table 11 it is 00 0000 0000 0000

can any one help me in this please.

waiting for reply soon

Thank you

Pushpa

  • 0
    •  Analog Employees 
    on Jan 11, 2013 8:58 PM

    Hi,

    The output code format of the AD9643 is either offset binary or two's complement.  In offset binary, the lowest value is set to all 0's, hence the value of -0.875V being equal to 00 0000 0000 0000 for offset binary.  For two's complement the value is simply derived by inverting the MSB of the offset binary, which is 10 0000 0000 0000.  The MSB in this case functions as a sign bit with the '1' indicating that the value is negative.  I hope this helps out.

    Regards,

    Jonathan

  • Hi ,

    thank you for your reply,

    can you please reply the DATA output for 0.875,-0.875,0 and also what formula you are using for DATA out?

    Regards

    Pushpa

  • 0
    •  Analog Employees 
    on Jan 17, 2013 11:17 PM

    Hi Pushpa,

    As shown on page 27 of the AD9643 datasheet, the output codes for the voltage levels of the converter are as follows:

    Input Voltage     Offset Binary                         Twos Complement

    -0.875V             00 0000 0000 0000 0000        10 0000 0000 0000 0000

    0V                    10 0000 0000 0000 0000        00 0000 0000 0000 0000

    0.875V              11 1111 1111 1111 1111        01 1111 1111 1111 1111

    There are 16384 codes that represent voltage levels for a 14 bit converter.  The lowest value (-0.875V) is 0 and the highest value (0.875V) is 16383 and the middle value (0V) is 8192.  When you convert these decimal numbers to binary you get 00 0000 0000 0000 0000, 11 1111 1111 1111 1111, and 10 0000 0000 0000 0000, respectively.  Twos complement is derived in a similar manner, it just has the sign bit.  In this case you use a scale from -8192 to +8191. 

    Regards,

    Jonathan