• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 3236 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.

Multiplex multiple photodiodes to one transimpedance amp

Matt Lang
Intellectual 540 points
Other Parts Discussed in Thread: OPA316, OPA170, OPA314, OPA320, OPA313, PGA281, TIPD102, MUX508

Hello,

I have an application that requires to sense the amount of light coming into multiple photodiodes for material recognition. To get a faster speed and to get the most linear output range i'm going to need to use a transimpedance amplifier with the diode reverse biased. Get a direct current to voltage conversion.

But putting a current to voltage amp on every photodiode is not practical due to cost and space. I was wondering what would be the best way to multiplex each photodiode to the same transimpedance amp? Sense the amp needs to directly measure the current through the diode I'm not sure how putting analog switches between the diode and the amp would affect how that works. 

I know the multiplexer would affect the bandwidth, noise, error, but that would be ok as long as the final system is within the bandwidth I need. I need to know if the whole principle of the current to voltage conversion still applies even when going through a mux of some kind.

Is the linearity of light energy in to voltage out going to be as good?

  • 0
    TI__Guru 62915 points

    Hi Matt,

    The main principles of operating as a transimpedance amplifier will still apply with a switch/MUX in series with the photodiode.  The main contributions the MUX / analog switches will add to the system are capacitances from each input/output of the switch to GND with a small series resistance between them.  The switch resistance won't be an issue because it will appear in series with the extremely high DC resistance of the photodiode.  The switch capacitance will appear in parallel with the diode capacitance (and may be substantially larger than the photodiode capacitance) which will increase the feedback capacitor value required to compensate the system.  An increased feedback capacitor will reduce the bandwidth of the system, and depending on your bandwidth requirements may require you to select a much higher bandwidth op amp than expected.  The basics of transimpedance amplifiers, bandwidth, and op amp selection are covered in the blog series linked at the bottom of this message.

  • 0 in reply to Collin Wells
    Intellectual 540 points

    Thank you Collin,

    Could you recommend some parts for me?

    I need to mux 6 photo diodes to a transimpedance amp. Then go into a programmable gain stage. Gain from maybe .5 to 10. Then out to an MCU ADC. Photo diodes will be receiving a good amount of light. This is not a very low light energy sensing application.

    Each photodiode has a matching LED emitter. Each LED needs to have individual current control for dimming (Not PWM). Calibration must be done, so if the solution has auto calibration features that would be great too. Multiple Transimpedance amps and adjustable feed back in one chip would be great too if that exists. Ok if it does not.

    My bandwidth is very low. No more that 50kHz.

    I see that TI has many different "AFE" Analog front end chips. Is there a solution for this type a data acquisition in a pre-made chip? Or something very close that can be used for this with other components? 

    Have any programmable constant current sources with many outputs? 

    Or just any general recommended parts or concepts that are made for this.

  • 0 in reply to Matt Lang
    TI__Guru 62915 points

    Hi Matt,

    If you have pretty large input currents then the input bias current of the op amp and leakage current of the MUX is not as critical as very low current applications.  Therefore most standard CMOS or JFET op amps will work fine for the transimpedance amplifier.  We have many options for both low voltage and high voltage op amps depending on your supply voltage requirements.  Consider the OPA320, OPA316, OPA313, OPA314 for low-voltage or the OPA170, OPA19x, OPA14x devices for a high voltage system.

    Consider the PGA281 for the PGA stage.

    Take a look at TIPD102 for a programmable current source circuit reference design.  I know there are many TI LED driver solutions out of the power management team, but I'm not sure if they would achieve the types of matching and accuracy that you will get from a circuit like the one shown in TIPD102.

    The MUX508 provides an 8:1 analog MUX function with pretty low (1nA) leakage currents and <10pF of switch capacitance.