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.

DAC8750: Data converters forum

Part Number: DAC8750

My customer plan to use DAC8750 to implement an adjustable current output from 0 to 20mA. He needs to calculate the accuracy that DAC8750 can achieve for the following bands of output current:

up to 200uA
up to 2mA
up to 20mA

Temperature is from -25 to +90 degC.

What is the best possible way to do this calculation?

Thanks

Josef 

  • Josef,

    The static errors of a DAC are dominated by the offset, gain and linearity errors. The combined effect of all these errors is the TUE (total unadjusted error). For the full current range, the TUE specification can be taken from the datasheet using the -40C to 85C condition.

    For the other current bands, the best way to estimate the combined error  is to compute the effect of the individual sources of error for the current range using RSS method.

    k = (current band/ 20mA) * VREF error

    The offset, gain & INL errors specs are in page 6 of the DAC8750 datasheet. The specifications for the -40C to 85C range will be closest to your needs.

    Below are my results

    0-200 uA band: 0.103 %FSR

    0-2 mA band: 0.105 %FSR

    0-20 mA band (TUE): 0.16 %FSR

    Notice that the combined error is dominated by offset & INL errors for the lower current bands and this is expected. 

    Regards,

    Reza

  • Hi Reza,

    thanks for your answer.

    Are the given errors in the datasheet meant without the user calibration?

    So if we use the user-calibration-option explained on page 26 in the datasheet, we can decrease the gain and offset errors and therefore decrease the total error. Is this right?

    If i am right, which accuracies are then achievable? Ideally, only the INL error will remain, but I don´t think the offset and gain error can be completely calibrated out.

    Thanks,

    Andreas

  • That is right for both statements. The specifications in the Electrical Characteristics table of the datasheet are without user calibration enabled and the offset and gain errors can be calibrated out. 

    Actually with user calibration, the DAC offset and gain errors can truly be eliminated. There are two points to note however;

    1. The calibration is purely digital and cannot be used to extend range of current output. As an example, if code 0 has an output current of +200nA, the digital calibration cannot correct that output current to 0nA. 

    2. User calibration does not correct for the drift of the errors with temperature. If calibration is done at a single temperature, offset and gain errors are zeroed out only at that temperature. 

    To estimate the error after calibration, you can use the same expression as provided above but this time around, consider only the temperature drift component of the offset and gain errors. The drift component is estimated as the difference in the error at TA=25C and TA= -40C to 85C.

    Below are my estimates after calibration for -40C to +85C

    0 to 200uA band: 0.04 %FSR

    0 to 2mA band: 0.04 %FSR

    0 to 20mA: 0.08 %FSR

  • Thank you very much, that´s exactly what I wanted to know.

    Regards,

    Andreas