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.

ADS1299: Internal PGA architecture

Part Number: ADS1299

Hello!

I have run an interchannel crosstalk test for ADS1299 and I have found out that channels 3 and 4 and channels 5 and 6 interferes more with each other.

The method I used was to apply a sine wave on one of the channels and analyse the FFT of the output of the other channels. I would like to know whether there is a relationship between the mentioned channels regarding their architecture inside the chip. Are the corresponding PGA next to each other? Is there a layout figure that shows the placement of the channels (PGA and connections) inside the chip?

Best regards

Mateus

  • Hi Mateus,

    Highly appreciate for using TI's products.

    This ticket has been received/viewed, and I will get back to you between 3~5 days.

    Appreciate.

  • Hi,

    May I ask some more questions to collect more detail info?

    Which exact ADS1299-X are you using? 4, 6, or 8 channels?

    Would you provide more info/detail, if possible videos, pictures or drawings of the crosstalk test?

    What was the magnitude and frequency of sine wave you were generating? And, what was the gain and data rate setting you selected?

    Do you have an eval kit/board?

    In the meantime, let me circle back to our engineers team to understand whether there is any significant difference in the internal PFA architecture or location/placement.

    Appreciate.

  • Hi,

    I am using ADS1299, with 8 channels.

    About the crosstalk test, I use a wave generator, set a sine wave with frequency of 10 Hz and magnitude of 2 V. The gain was kept in 1 V/V and the data rate was 2000 sps. I am using a custom board that was designed based on the ADS1299 eval board. The figure below shows a drawing of the scheme that I used for the test:

    I applied a 1 Vdc so that the other channels did not float. All the negative inputs are tied to ground. After collecting 10 s of data, I analysed the FFT spectrum to calculate the crosstalk. The following two figures show the results:

    - Input signal spectrum (channel 1)

    - Interference spectrum (channel 2)

    Obs: I ignored the 60 Hz noise.

    Then I measured the magnitude of the highest component of the interference spectrum and calculated the crosstalk as following:

    crosstalk (dB) = 20.log10(V_interference/V_input)

    where V_interference is the magnitude of the greater component of the interference (in this case the second one (20 Hz)) and V_input is the magnitude of the fundamental component of the input signal. I measured the interference for all channels and changed the input each time. Then I calculated the crosstalk for every channel. The figure below shows the results:


    Channels 2 and 8 have a problem in my board, so I kept them shorted and did not use them.

    I hope this information is enough for now, but I will provide more of it if needed.

    Thank you!

  • Hi,

    How/what was the signal at the inputs if you use oscilloscope probe to check?

    Could you do the same/similar test on the Eval board/kit after check the following -

    Please first check whether you setup the chip correctly with the following table in the datasheet and the acceptable input voltage range -

    Do you want to try connect unused analog inputs directly to AVDD?

    May I ask what is your INxP - INxN?

    Also, would you try signal with lower amplitude, e.g. under 5mV or 1mV as ECG, EEG signals are usually in this range of scale.

    Thanks

  • Ok, I will try these and answer back when I have the results.

    I am satisfied with the results of the crosstalk that I have obtained, since most of them are within the specified value on the datasheet (-110 dB). I was just curious about the channels 3 and 4 and channels 5 and 6, which seemed to interfere more with each other. That is why I asked about the internal layout of the channels to verify whether the distribution of them could cause this effect.

    Thank you!

  • Hi,

    Just talked to our engineers. They say the board layout is important to affect the interference and crosstalk, would you please refer to page 72 to 73 in the datasheet -

    https://www.ti.com/lit/ds/symlink/ads1299.pdf?ts=1649431886005&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FADS1299

    Thanks,

    Chien