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.

Question on Video: Developing the SAR Reference input model

Other Parts Discussed in Thread: OPA320

Hi,

I have been watching the video series and have a question on the SAR reference input model video. 

https://www.ti.com/video/series/precision-labs/ti-precision-labs-analog-to-digital-converters-adcs.html

From this video I have developed a model for the F28388D MCU but am uncertain on the results. 

A screen shot of the video is shown below. It indicates that if the error reads 0 then everything is fine. 

When I run the simulation VrefError corresponds to Vref error. I can see Vref charging when the switch is closed but it does not charge all the way to its final value. When the switch is open Vref jumps up to it expected value so its error is near zero like the video indicates. However the voltage sampled on the node Cref in the slide (VchError below) is -80 mV below expected value. The video does not say anything about what Cref should be at just Vref. Will the measurement have -80 mV of error?

Thanks, 

Jennifer

  • Hello Jennifer,

    Yes, at full scale input, the error will be approximately 80mV, since this is the final error voltage on the reference sampling capacitor.  Typically, the external capacitance on the reference input pin will hold the voltage level and not jump like your simulation results show.  I am guessing the immediate jump in voltage is a result of too little external reference capacitance, but it could also be due to the specific reference that you are using.

    Please share your simulation file and I can provide a more detailed answer.  The internal reference sampling capacitor at the end of the reference acquisition phase should be within 1LSB of the steady state reference voltage.  If you are using a 2.5V reference, and 12b single-ended configuration, then 1LSB=2.5/2^12=610uV.

    Regards,
    Keith Nicholas
    Precision ADC Applications

  • We are using this in a 16-bit application. I didn't save this specific file but will attach a file that shows the same characteristic. I grabbed the values from the F28388 MCU datasheet.

     F2838x_Ext_16BSE_Motor_opa2320_refdriver_80mV.TSC

  • Hello Jennifer,

    Attached is an updated simulation file. 

    F2838x_Ext_16BSE_Motor_opa2320_refdriver_80mV_rev1.TSC

    I noticed that the switch resistance was too high for the internal reference sampling capacitor to settle.  You can simulate this by driving the reference input circuit with an ideal 2.5V source (no OPA320 buffer).  Typically, the reference capacitance will be 1/4 of the input sampling capacitance, which you have taken into account.  However, since the conversion process runs at a much higher frequency than the overall sampling rate (50MHz in your configuration), the internal switches for the reference input will be increased in size to reduce the total resistance.  If not, then the device cannot support the desired resolution and maximum throughput rate.  As an estimate, I have assumed that the switch resistance will also be 1/4 of the input switch resistance.

    Below are the results with the switch resistance set to 106.25Ohm and the reference cap set to 8.125pF.  Total error in this case is 12.8uV, which is about 1/3LSB.  Typically, you want the reference voltage droop to be no higher than 1LSB, so if these values are correct, then this should work well.

    Regards,
    Keith