• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 16 subscribers
  • Views 1526 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.

MUX506: Crosstalk between adjacent channels

Simon Lessard
Prodigy 40 points
Part Number: MUX506


Hi!

I'm using the MUX506 to select a photodiode from a photodiode array of 46 elements into a transimpedance amplifier. Since the non-inverting input of the transimpedance amplifier is to 0V, that makes 0V all across the net from the photodiode, input and output of the MUX up to the amplifier. Looking at the datasheet, I understand that the specification for crosstalk is a function of the input and output voltage of the MUX. Since it is 0V everywhere, is the crosstalk spec still relevant? 

My photodiodes have their cathode to 0V and their anode is directly connected to the MUX input. The mux output is directly connected to the amplifier. 

I observe that if I there is some signal on a specific photodiode but the output of the MUX is selected to read the adjacent photodiode, I can still get signal from that photodiode. In fact, that is true for the next 3-4 adjacent photodiodes, decreasing with the "distance" from the observed MUX input.

Is this normal? Is there a way to minimize that?

Thanks,

Simon

  • 0
    TI__Expert 3900 points

    Hi Simon,

    The output of your photodiode is a current source that is then passing through your tansimpedance amplifier to generate a voltage to be monitored in your system correct? Are you using this photodiode in a photovoltaic mode?

    Can you further explain your setup of the 46 element matrix? A diagram of how you have set up the photodiode array will be very helpful with the better understanding your setup to max sure we are on the same page. Once we have this diagram we can better understand how the 3-4 adjacent photodiodes impact the signal chain.

  • 0 in reply to Dakotah
    Prodigy 40 points

    Hi Dakotah,

    Thanks for your kind reply, I really appreciate. You are right, I'm using in photovoltaic mode.

    I attached my schematic below. This is an array of 46 photodiode. I have two outputs on my board. The goal is to address any of the photodiode on any channel so it is why you see that each PD is wired to two MUX input. CHANNEL1_pre is connected to an OP-AMP in a transimpedance configuration (non-inverting to 0V, inverting to 100K feedback).

    I ran extra tests and I found that the problem occur when I select, for example, PD1 on channel A (U6, mux input S16) and an adjacent element, let's say PD2 on channel B (U7, mux input S15). Channel A and B have their separate path using differents MUXes but even if PD2 is selected on channel B while PD1 is on channel A, these is a kind of "ghost connection" between them since I detect a change in the output voltage of channel A (after the transimpedence) when I select PD2, 3 and 4 on channel B... and I'm pretty sure is is not related to the optical signal since all diodes receive the same light signal at this time.

  • 0 in reply to Simon Lessard
    Prodigy 40 points

    UPDATE : Sounds like my photodiode array was completely saturated when I did the test. This was not related to the MUX itself finally. 

    If I don't saturate de diodes, I don't have this effect.

    Thanks for your help.
    Simon

  • 0 in reply to Simon Lessard
    TI__Expert 3900 points

    Hi Simon!

    Glad to hear that you found the root cause of the problem!

    Please let me know if you have any other questions!