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.

ADS4249: Technical Inquiry about ADS4249

Part Number: ADS4249
Other Parts Discussed in Thread: LMH6554

Tool/software:

HI,

        We use the LMH6554 to drive the ADC as Figure 1,the ADC is ADS4249.The LMH6554 converts the 10 MHz(1V VPP) single-ended signal into a differential signal and outputs it to the ADS4249.The sampling frequency of the ADS4249 is 250 MHz.When we used an oscilloscope to measure the waveform at point B, we observed a 250 MHz high-frequency signal superimposed on the 10 MHz output signal as Figure 2.When measuring point A with an oscilloscope, this phenomenon was not observed.When the 22nH inductor is replaced with a 0-ohm resistor, this phenomenon also disappears.The 250 MHz high-frequency signal appears to be related to the ADC sampling and the 22nH inductor.Please help us analyze why ADC sampling interferes with the measuered signals.

                                                              Figure 1

                                                                          Figue 2

  • Hi,

    The ADC is a switched capacitor design and the analog input pins of the ADC are directly connected to the internal sampling switch.

    What you are seeing on the oscope is the charge injection of the ADC's sampling capacitor opening and closing. This is also known as kickback.

    The ADC's kickback settles in time in order to take a valid sample.

    If you want to understand this more, please see this article. Read pages 1-3.

    Regards,

    Rob

    www.analog.com/.../an-742.pdf

  • Hi

        Due to the differential input the kickback does not affect the ADC sample precision.Is this right?

        Reguards

        Aaron

  • Hi Aaron,

    This is correct.

    Regards,

    Rob

  • Hi Rob

        The tipical output voltage value of VCM(ADS4249 PIN23) is 0.95V,but the VCM output voltage is 0.978v on our pcb board.Is this voltage of 0.978v normal?What is the output voltage range of the VCM?Will the diffierence between the actual output voltage(0.978V) of the VCM and its theoretical value(0.95V) affect the ADC measurement results?

    Reguards

    Aaron

        

         

  • Hi Rob

    We use the ADS4249 to acquire signals and encountered some issues during the process of configuring its registers.

    1:The ADC was configured in High-performance mode, High-frequency mode, and High-speed mode (with the following register settings: write 03h to register 03h, write 01h to register 4Ah,write 01h to register 58h, write 40h to register 02h, write 18h to register D5h,write 0Ch to regiser D7h, write 20h to register DBh). We measured the voltage of  Vcm is 0.903V (not the expected 0.87V) . When applying a 0.5V VPP(±0.25V) sine wave input, the maximum value ADC output is 2095 and the minimum value ADC output is -1945.

    2:We attempted offset correction by writing 2Ch to register CFh and 20h to register 3Dh. After 9 seconds, ACh was written to regiser CFh.
    The measured Vcm remained at 0.903V (not the expected 0.87V).With a 0.5V VPP(±0.25V)  sine wave input, the maximum value ADC output is 2095 and the minimum value ADC output is -1945.

    3:We attempted to modify the channel gain by writing C0h to register 25h and C0h to register 2Bh.When applying a ±0.25V sine wave input, no observable change was detected in the ADC output values.

    4:We attempted to adjust the offset pedestal by writing 80h to register BFh and 80h to register C1h.With a ±0.25V sine wave input applied, no measurable change was observed in the ADC output values.

    Please assist us in analyzing the following issues:

    1:Under the configured mode, the voltage of the VCM should be 0.87V. Why is it measured 0.903V on the circuit ?

    2:For a ±0.25V input, the expected output should be ±2000. Why are we observing larger positive values (+2085) and smaller negative values (-1945)?Shouldn't the offset correction function adjust the range from 2095/-1945 to ±2000? Why is it ineffective?

    3:After setting a 6dB gain, the full-scale range should become 1Vpp. However, experimental results shows no impact on the  ±0.25V sine wave input. Does gain adjustment only modify the input voltage limitation range?

    4:Why does adjusting the offset pedestal also fail to affect the captured output values? Specifically, when set to midcode-32, the output should decrease by exactly 32 codes—why is there no observable change?

            

  • Hi Aaron,

    Yes, a higher VCM can affect performance. Can you measure the DC voltage on both of the analog input pins, nearest to the ADC's analog inputs?

    What are those two voltages, they should be the same.

    Regards,

    Rob

  • Hi,

    It might be best for you to send over your schematics so I can review them first and make sure everything looks good from that standpoint first.

    Please advise.

    Thanks,

    Rob

  • Hi Rob,

        I will provide you the  schematics two days later.Tomorrow and the day after tomorrow are weekend.We powered the ADS4249 through 1.8v power supply.but we use 2.5v I/O to communicate with ADS4249 through SDATA SCLK SDATA and RESET.Can the 2.5V I/O affect the vcm.

    Reguards

    Aaron

  • Ok, thanks Aaron, send over the schematics when ready and I will look them over.

    Thanks,

    Rob

  • Hi Rob,

        In the E2T forum,I can't find the way to add an attachment,SO I have send over the schematics throuth email(noreply-e2e@mail.ti.com) ,please check the email.

        Thanks

         Aaron

  • Hi Aaron,

    I will send you an email and you can send it over.

    I will close this post and we can work this thru email.

    Please be on the lookout for my email.

    Thanks,

    Rob