• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 1052 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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: Schematic and more questions

Brian Santillan
Prodigy 30 points
Part Number: ADS1299

Hello,

We have a schematic version that actually works pretty good. Nevertheless, we have some questions to continue with our investigation.

1) Input filter. In the datasheet (page 68) says that the filter capacitor in the input stage (Cfilt) should be connected as differential to obtain better common mode rejection. But we have noticed that in the EEG Front-End Performance Demonstration Kit schematic does not have the capacitor in differential mode. It is connected to ground. Why do you have made this choice?

2) Bias filter. The bias filter has aproximately 50hz of frequency cut off in the EEG Front-End Performance Demonstration Kit and 150Hz is recommended in the ADS1299 datasheet. Why the change?

3) Bias. As example, suppose that we have 1 electrode to take measures and 1 referential electrode. And we are using one more electrode connected to GND. We have a measurement that is fine. But we are not using the Bias electrode. How should be the measurement done correctly? We tryied using the Bias Electrode instead of the GND electrode, but we cant get the signal.
We have set CONFIG3 = ECh and BiasSens = 01h. Is there something more that we have to set?

4) Active electrodes. We know that in datasheet (page 68) says that the ADS1299 it does not need active electrodes. The questions are, is not recommended? Is not needed to measure some basic signals? Or you dont know or its a matter of further investigation if using an active electrode (buffer or preamplifier) could obtain a better signal.

Well, thank you for reading.
Regards.
Santillan Brian.-

  • 0
    TI__Mastermind 47625 points
    Hello Brian,

    Welcome to the forum and thanks for your post! You have very good questions.

    1) The capacitor being tied in a single-ended manner was what we thought was best at the time. We have learned otherwise since then and have changed the datasheet to reflect this, supporting a differential cap.
    2) 50Hz is ideal for EEG applications, but what we found was that customers were using this device for ECG as well. Since ECG signals require slightly more bandwidth (~150Hz) and EEG measurements are still acceptable with this filter - it made sense to support 150Hz as the general recommendation.
    3) Are you trying to do single ended or differential measurements? If you are writing ECh to CONFIG3 then it makes sense that the measurement using the BIAS electrode will not work since the BIAS buffer is powered down. I would recommend using the internal BIASREF signal, turning on the buffer, sense, and connect the bias. Are you writing 01h to the positive or negative node of channel 1?
    4) Since the ADS1299 has internal PGAs, it is not necessary to have an additional gain stage in the electrode. Active electrodes are fine as long as the signal does not exceed the input limitations of the ADC. It is possible that the gain stage in the active electrode has better noise performance than the internal PGA, but this has not been explored as it is out of our domain.
  • 0 in reply to Alexander Smith
    Prodigy 30 points

    Hello Mr. Smith, thanks for helping us.

    1) Differential cap then, perfect!

    2) Got it, thank you.

    3) Single ended. But ECh is 1110 1100b. Looking the CONFIG3 register, Bias Buffer power should be 1 (on), using this configuration right? Or I did not understand something? Or do I need to activate this BIAS_LOFF_SENS in CONFIG3 register too?
    I am writing 01h to the positive node of channel 1 using the register BIAS_SENSP.

    4) Great.


    Your answers were very helpful!
    Thank you so much!
    Santillan Brian.-

  • 0 in reply to Brian Santillan
    TI__Mastermind 47625 points
    Hi Brian,

    After some thought I would like to revise my comments for 2) & 3).

    2) The filter is actually poorly designed in the user guide and the cutoff frequency should be increased to at least include 50/60hz in order to cancel the mains noise. This is not dependent on what type of application is being used.

    3) Bias is typically the average of multiple channels of data. Since you are only using one channel, it is not possible to compute an average and you end up injecting your inverted signal back onto the patient. Try connecting additional channels and see if it helps!
  • 0 in reply to Alexander Smith
    Prodigy 30 points

    Hi!

    2) Yes, I thought about that and I have already change the cutoff filter to 150Hz aprox., I think it is enough.

    3) Oh, thats the point. Ok, I will do some other tests.

    Thank you so much Alexander!
    Regards,

    Santillan Brian.-

  • 0 in reply to Brian Santillan
    TI__Mastermind 47625 points
    Hi Brian,

    Happy to help! Let me know how your testing goes. If you have unrelated questions please start another thread, otherwise just respond to this one!

    Good luck!