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.

TLV320ADC6140: Noise performance for less than 20Hz

Part Number: TLV320ADC6140

Tool/software:

Hi Ti

I'm using this impressive ADC to measure heart vibrations and therefore I have the inputs DC coupled to enable measurements down-to 2Hz.

Other channels on the second ADC has an accelerometer connected measuring DC as-well (gravity) thus the DC coupling altogether.

The noise is not spec'ed below 20Hz but the plot seems to be linear in the low region indicating a 1/f corner well below 20Hz. But this is guessing.

Do you have any specs for the frequency range from DC to 20Hz?

Kind regards

Jørgen

  • Hi,

    Let me have a look at this and update this thread with a follow up in a couple of days.

    Thanks and Regards,

    Lakshmi Narasimhan

  • Hi Lakshmi.

    Any news ?

    I have done some tests and found indications that I can expext an 1/f corner around 6Hz but still not exceeding -135dB at 1Hz...

    ....but I would like to have your specs.

    kind regards

    Jørgen

  • Hi Lakshmi

    still awaiting some news...

    Regards

  • Hi Jorgen,

    Sincere apologies for the delay.

    On the datasheet itself, we don't have a specification below 20Hz, as you have mentioned above.

    Is it possible to share the spectrum from your experiment, I can have a look at the same?

    Additionally, you have mentioned that the device inputs are DC-coupled. There is also a digital high-pass filter inside the device, whose default -3dB cutoff frequency is 12Hz, for a 48kHz sampling rate.

    This can be changed through the HPF_SEL register.

    Can you confirm if the HPF is kept at default setting in your tests?

    Additionally, the device does not support DC-coupled inputs with 2.5kOhm input impedance setting per the datasheet. Can you confirm if this is also accounted for in the device configuration?

    Thanks and Regards,

    Lakshmi Narasimhan

  • This is what I measured with HPF off and DC input resistance set to 10k.

    The input signal is around 1kHz and 1.8dB from full scale and as you see the noise floor

    (including the generator noise) is flat even at 1Hz with a level at -114dB ref FS not showing any sign of an 1/F corner.

    It all looks fine but still I would like to have your input as well.

  • Hi Jorgen, 

    When evaluating the ADC's performance for low frequency signals, i.e down to 2Hz range, I suggest configuring the digital HPF to all-pass (register 0x6Bh = 00) like Lakshmi mentioned. Then perform a test where the input signal sent to the device is a known low frequency (e.g. 5Hz) and compare the output plot to a 1kHz tone like the one shown above.

    The plots in our datasheet have parameters for 20Hz-20Khz since this is the audio band and most relevant to our customers.

  • Hi Daveon

    What I asked for was if you had any noise specs using the ADC below 20Hz. I have verified that the large-signal response down to less than 2Hz is OK.

    And as you see I can measure a noise level at 2Hz around -114dB ref FS and that includes both conditioning amplifiers and generator noise so that is quite OK. But as Lakshmi wrote that he would look around and come with an answer whether TI's R&D department could be more informative. I know for certain that this ADC is intended for audio applications but surely other customers will use this converter as I do down to DC.

    Kind regards

    Jørgen

  • Hi Jorgen,

    Unfortunately our validation team does not measure device characteristics or electrical specifications outside of the audio band.