• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 3 answers
  • Subscribers 16 subscribers
  • Views 368 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.

TS5A22364: Selecting an SPDT switch to multiplex very low amplitude EEG signals (<10 uV level) with negligent noise?

Raza Qazi
Prodigy 150 points
Part Number: TS5A22364

We are trying to use an SPDT switch to multiplex EEG channels, however previous switch we used referred input noise into the signal? We think it was partly because it wasn't compatible with negative signals. Now we choose TS5A22364DGSR with negative signalling ability, however, do you have any recommendation for the best SPDT switches for this application?

  • 0
    TI__Mastermind 23373 points

    Raza,

    I would guess that this was the problem as well. I think you picked a good part in the TS5A22364, honestly. Very low R-on, low THD, can handle negative supply rails. 
    Just note that the signal level is determined by the Vcc supply. The minimum analog voltage is Vcc - 5.5, so if you decide to run this at the max recommended supply, you technically lose your negative signal capacity. But this should work  

    Good luck!
    Rami Mooti

  • 0 in reply to Rami Mooti1
    Prodigy 150 points

    Thanks, we plan to power using 3.3V or 2.5V Vcc supply. Can you also recommend an SPST switch for similar application?

  • 0 in reply to Raza Qazi
    TI__Mastermind 23373 points

    Raza,

    Just to clarify because I realize I misspoke on the last post. I said 'can handle negative supply rails' but I meant 'can handle negative data signals'.

    Unfortunately, we won't have a similar SPST switch that has negative voltage carrying capabilities and On-Resistance quite that low. The TS12A4516 would be the best that offers negative signals carrying ability. 
    If you could somehow bias the signals and remove the negative voltage requirements, the TS5A3166 would be a good choice.

    Rami Mooti

  • 0 in reply to Rami Mooti1
    Prodigy 150 points

    Thanks Rami.

  • 0 in reply to Raza Qazi
    Prodigy 150 points

    With the TS5A22364, do we need to pull down the control signal with a high resistor to GND such as 10Mohms or there is not such requirement and it can be left floating?

  • 0 in reply to Raza Qazi
    TI__Mastermind 23373 points

    Raza,

    You shouldn't leave the IN control signals floating as we can't guarantee the state of the switch. It could stay high, low, or even switch back and forth.
    The need for a pull-up/down resistor will depend on the architecture of the device driving your logic. These switches don't have internal pullup or pulldown resistors so any logic can be passed but must be driven from another source. Depending on what that source is, you can determine if it's a pullup or down. 

    It you're simply wanting to hold this control signal permanently low though and don't intend on using the signal path at all, then it's still typically good practice to ground it as to not allow the pin to act as an antennae and absorb unwanted noise. 

    Rami

  • 0 in reply to Rami Mooti1
    Prodigy 150 points

    We plan to drive using an external uC, in that case is pullup or pulldown still required?

  • 0 in reply to Raza Qazi
    TI__Mastermind 23373 points

    Raza,

    Typically for uC I2C lines drives the lower signals for the 'LOW' state and will need an external pull up to give you the 'HIGH' state. So yes, pull-ups to the voltage level your working with will be required.

    Hope this helps,
    Rami

  • 0 in reply to Rami Mooti1
    Prodigy 150 points

    We are pulling it down to GND because we want NC connected by default. However, I had some more questions:
    1. We are using a 1Mohm resistor to pull down. With a 2.5V Vcc, can you suggested the best value for pull down resistor on the control signal?
    2. In test boards, we are using the larger footprint (VSSOP TS5A22364DGSR), however later we want to use the smallest switch footprint in the same family. Would it make any difference to switch to smaller footprint of same IC and any issues we should consider or test before switching to a different footprint?


  • 0 in reply to Raza Qazi
    TI__Mastermind 23373 points

    Raza,

    1) The Vcc level doesn't really directly impact the pull-down resistor. 1Mohm seems a bit high though. We have documentation on TI.com on how to choose the appropriate pull-down resistor. It's more dependent on the current and voltage being driven on that signal line.

    2) The footprints should have the same functionality but the only real issue (besides of course needing to solder on a smaller package onto your board) would be the thermal characteristics. Different packages are going to have different areas which means that energy(heat) is dissipated differently. 

    Hope this helps,
    Rami Mooti