• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 46 subscribers
  • Views 1113 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS1298: ADS1298 daisy chain

Mary
Prodigy 120 points
Part Number: ADS1298
Other Parts Discussed in Thread: ADS1299

Hello

I need connect 4 ADS1298 to build a 32 channel system. But I have some problems. I wish you could help me with this questions.

1-Is it better to use bipolar or single supply? What are advantages and disadvantages of every method? In which applications, it's better to use single supply?

2-In page 58 of datasheet it said it's better to minimizing length of DOUT to minimize board-level signal delays . but in next sentences it said Placing delay circuits such as buffers between DOUT and DAISY_IN is another way to mitigate this challenge. One other option is to insert a D flip-flop between DOUT and DAISY_IN clocked on an inverted SCLK.

3-Is there any other PCB consideration about Daisy connected ADS129Xs?(Except points on page 58 of datasheet)

4- I want to get data from 32 differential channels (64 channels)  on 4mm*4mm array electrode. For this small electrodes and small distance between individual electrode,Is it necessary  to use GND electrode (RLD pins in ADS1298 and BIAS pins in ADS1299) to reduce common mode noise?

thanks very very much :)

  • 0
    TI__Guru 66446 points

    Hello Mary,

    Thank you for your post.

    1. ADC performance is the same for both supply configurations. The only advantage I can see to bipolar supplies is that GND can be used as your common-mode voltage. However, for biopotential applications, it is still beneficial to use the RLD amplifier to provide the mid-supply bias (rather than ground) in order to help reject common-mode noise.
    2. Minimizing digital signal trace length is always a good idea to preserve a clean layout and minimize communication issues. At the SPI speeds required for ADS1298 and ADS1299 applications, I do not think the D flip-flop technique will ever be necessary.
    3. Same recommendations would apply to single- and multi-device layouts. Power supply routing can be simplified through use of an internal plane layer.
    4. The RLD/BIAS electrode is still needed to establish the common-mode voltage on the body. Input signals from the electrodes must remain within the supply rails of the PGA amplifiers and not drift towards the rails. 50/60 Hz common-mode can also be reduced using this method.

    Regards,

  • 0 in reply to Ryan Andrews
    Prodigy 120 points

    Hi Ryan,

    Thanks for the answers, they are exactly what I need.

    But now I have another problem.

    1-You said I must use RLD/BIAS electrode, but actually, there's no space on array electrode for connecting to RLD/BIAS. And I don't want to record signal from body, so I can't connect It to other part of body.( I should  connect this array to a small part of excitable tissue). How can I resolve this issue?

    2-New question I forget to ask in my previous post: I know I should use terminating resistors in long digital tracks, near ADS129x, but what about multi device PCBs? Every device have it's own resistors,right?

  • 0 in reply to Mary
    TI__Guru 66446 points

    Hi Mary,

    It sounds like an interesting application. You're saying that you intended to excite some live tissue, and measure the response through the electrode array connected to the ADS1298? Those voltages, regardless of how small they may be, need to referenced to the ADS1298 analog supply. The amplifiers at the ADS1298 input expects signals to sit near a mid-supply common-mode. 

    If you cannot drive the body with a known DC common-mode voltage (or DC + AC for improved common-mode noise rejection), then typically the electrodes are AC-coupled on the PCB and biased with a pull-up resistor to RLDOUT. You can see examples of this in the ADS1298 application section as well as in the ECG online training videos linked here.

    I do not believe such terminating resistors will be necessary in your application. You may want to define a default state for some pins, such as /CS = 1 to disable the interface when you're not communicating with the device. Otherwise, termination resistors are generally needed for higher-speed applications where signal reflections can cause communication failures, or they may be needed to match the source impedance of external clock sources. You should not require termination resistors on the ADS1298 SPI signals.

    Best regards,

  • 0 in reply to Ryan Andrews
    Prodigy 120 points

    Hi Rayan
    Thanks for great answers, they are really helpful. And thank you foe shared video, I learnt a lot by watching them.

    But now I have new questions about RLD. It's nice of you if help me with this problems.


    1-You said I can pull channels to RLD with a resistor. Do you mean both of channels(like channel 8 in this picture) or just one(like channel 7);and if the answer is one, which channel(negative ir positive) should pull up?

    2-What is the best value for this pull up resistor?
    Should it be large or small?

    3-And in this structure, what should be Rrld, and Crld? Because now RLDOUT is directly connected to channels, and if gain be large, as it is good for Right Leg Drive feedback, PGAs may saturate.

  • 0 in reply to Mary
    TI__Guru 66446 points

    Hello Mary,

    The feedback components Crld || Rrld need to connect between RLDOUT and RLDINV.

    The RLDOUT should be connected to all channel inputs in order to set the common-mode voltage. A typical value of around 10 Mohms is usually appropriate as we want the input impedance of the circuit to remain large.

    Regards,

  • 0 in reply to Ryan Andrews
    Prodigy 120 points

    Thank you so much.

    Your answers are really helpful.

    :)