this might be a little silly question, but please advise me. I am trying to use two photodiodes (one for the absorbance measurement and one for the source of the light as a reference) and the voltage from the transimpedance amplifier from each photodiode connected output pin of OpAmps will be connected to the differential amplifier input to improve the SNR.

Here is my question. Let's say the output voltage from the 1st PD is V1 and the voltage output from the 2nd PD is V2. VD is V2 - V1. I want to measure VD and V2 simultaneously. So VD is the voltage difference between V2 - V1 and V2 is simply single end measurement (relative to the ground reference).

If I use two ADC (microcontroller) channels to measure those two voltages, can I connect 1) V1 - V2 as inputs to one differential amplifier and 2) V1 - ground as inputs to the other differential amplifier?

Say figure below is a dual differential ADC IC from Analogdevices.

Can I use that IC to measure those voltages without any problems? Thank you in advance for your advice.

enter image description here

  • \$\begingroup\$ A schematic would help, but I think I understand what you are trying to say. ADCs typically have enormous input impedance, so much so that it's as if it wasn't even connected to your circuit. As a result, connecting the same voltage to two different ADC channels (in this case, I think that's what you are doing with V2) should be okay :) \$\endgroup\$
    – DSWG
    Jul 15, 2018 at 0:39
  • \$\begingroup\$ If you replace the source with an equivalent R and get zero Vdc and -60dBc AC then you can subtract them anywhere. But that depends on your layout, cables, noise and offset voltage. \$\endgroup\$ Jul 15, 2018 at 1:31

1 Answer 1


I am not sure if I completely follow you, but you seem to already took the difference between V1 and V2 from your transimpedance amplifiers with a differential OPA. So you can use a single-ended single-channel ADC. And since your output will be low-impedance (hopefully 10-15 Ohms), you should be fine with any kind of ADC, even embedded into a cheap MCU.

If you don't use the differential OPA, you can use the differential ADC instead, like LTC2341, as the one you selected. But you will need only half of this circuit so the LTC2341 is overkill.

If you want to convert your V1 and V2 independently using some sort of shielded differential signaling over long cable, the LTC2341 is perfect for this, it samples both channels simultaneously. But keep in mind that the input to LTC2341 is a bare sample and hold unit, which has a fairly low equivalent impedance,

enter image description here

which usually requires a good low-impedance source/buffer. I assume that your transimpedance amplifiers do have a low-impedance output, so you should be fine.

On a last note, "can I use that IC", you need to realize that you can't just "use" it. You will need to design a PCB for 8x8 QFN, likely in 4 layers at least, provide two clean power rails, have some connectors, and a LVDS or SPI channel to something else like another MCU to collect the data. This would be a lot of work and require some qualification. I think you will be better off with some off-the-shelf DAS like the one of the DATAQ list.

  • \$\begingroup\$ I thought I left the comment below, but I did not. I would like to say thank you for your advice. It really helped me understand it. \$\endgroup\$
    – Tiger Balm
    Sep 18, 2018 at 4:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.