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.

TLV320ADC3101: ADC3101 has very large DC bias

Part Number: TLV320ADC3101
Other Parts Discussed in Thread: ADS1282

Dear expert, 

I am using the ADC 3101 as the vibration data acquisition, but we found the final result have very strong DC bias, and I do not know why this happens, 

do you have any suggestion on how to improve the DC bias? can I remove the DC bias through software setting or just I need to re-design the PCB setting.

the attached is the signal plot when the vibration chip is in stand-still status without any vibration excitation.. it seems like there is a cyclic wave even the chip is not with vibration source. can you give me some advise on how to improve?

the sensor is with 40mv/g sensitivity

  • Dear teams

    I am looking forward to your support.

  • Hi,

    Generally audio ADCs are not trimmed for DC precision because most systems will have AC coupled inputs and use a digital high pass filter to remove any offsets from the ADC. There is a digital high pass filter integrated in the ADC3101 which I would recommend using if you are not. You may also want to verify with this low level oscillation that you have a stable circuit for your sensor. 

    Here is another post with some more information on the offset: 

    Best,

    Zak

  • Dear expert,

    thanks for your reply.

    can you pls tell where can I enable the Highpass filter? do  I just place a HPF in the flow by using the PPS software? or is there a register setting to enable the HPF function?

     "There is a digital high pass filter integrated in the ADC3101 which I would recommend using if you are not"

  • Hi,

    I would recommend you look through section 10.3.12.3 on page 34 of the datasheet which describes the user programmable filters. The previous section also describes the various processing blocks supported and can help you identify which ones you may want to use depending on your implementation.

    The ADC3101 GUI is a very useful tool for generating the biquad coefficients and can be run even without an EVM. Just make sure you run it as an administrator or it will not work correctly. This would let you easily create the filter response you are looking for and then export the code to program the coefficients necessary to generate the filter response.

    Best,

    Zak

  • Dear Expert,

    my application is the vibration measurement application, the low frequency like from 0.5~10 Hz is our interested frequency , but by using the digital HPF filter, my understanding is that it is difficult to design a digital filter from beginning from 0.5Hz. 

    so do you have any suggestions on how to reduce the DC bias through the PCB board design?  or is there any register setting on how to remove the DC bias?

  • Hi,

    In analog filters this can be difficult to realize due to the size of components involved, but I do not think you will have any issues filtering this with the biquad filters. I was able to generate an HPF cutoff of 0.5Hz without gain error or instability using the tool. I would recommend you use the ADC3101 GUI tool to generate the coefficients and test your desired filter implementation to see if it improves the performance in the way you are looking for and confirm the filter does not have any stability issues.

    Best,

    Zak

  • Dear expert,

    I try to use the ADC GUI for setting the HPF, but not work, because the frequency from 0.5 above is also decreased. but this is my interested area which with high accuracy

    is there any ADC chip that has high DC accuracy?

    in addition, can I use the PPS for setting the HPF? will it work better?

  • Hi,

    In the example you have attached it looks like your cuttoff frequency was set to 10Hz. The center frequency you choose for the filter is the -3dB point, so there will still be some attenuation at and around the cutoff frequency.

    Since you are entirely concerned with sub-audio frequencies, I don't think you should use an audio converter here. You may want to look at a device like the ADS1282 and/or post a new thread for precision ADC recommendations!

    Best,

    Zak