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.

LMH6551: Validity in using a fully differential amplifier to amplify a photodiode signal

Part Number: LMH6551


I am trying to design a highly sensitive transimpedance amplifier (TIA) system to amplify very low modulated current levels of a photodiode (order of ~10 nA peak-to-peak). This is my first time simulating/designing an analog circuit and it has been a rewarding challenge so far.    

I recently went through the TI "Precision labs series: Op amps" video series to educate myself in the basics of analog design. One of the units within the series covers the fully differential amplifier (FDA). Upon seeing the videos and how FDAs excel at removing common mode noise between the inputs, I thought it has potential for being used as a highly sensitive TIA. What I originally envisioned was the following circuit for a FDA-TIA system:

Where the current source IG2, capacitor C1, and resistor R1 is equivalent photodiode model for my system. Is there any inherent issues in using an FDA like a TIA? (Note: I simply used one of the macro models available on TINA TI for illustration purposes... I have not done any simulations as I am still learning how to properly simulate an FDA with TINA TI.)

I scoured online to see if anyone has tried using a FDA directly coupled to a photodiode... but nothing came up. The closest thing I could find was from literature published by Zhang et al. (2019) where they convert the photodiode into a voltage signal first using TIAs and then couple the output voltages to a FDA as shown below. Is it a rule of thumb to only be able to use a voltage source/input for a FDA?

I also have a question regarding actually measuring the differential output (ΔVout) of the FDA to a single input instrument such as a lock-in amplifier. If I wanted to measure ΔVout using a single input device, would I have to couple the FDA outputs to a single ended op-amp in a differential configuration first, then measure the single ended output? Measuring ΔVout with a multichannel oscilloscope seems quite easy as I can plot the difference of Vout- (input channel 1 on oscilloscope) and Vout+ (input channel 2 on oscilloscope) using a basic difference math operator on the oscilloscope. TIA_VS_FDA.TSC

  • Hi Camille,

    This is an interesting idea however I don't think it would work in real life when trying to implement as well as trying to establish the behavior of this circuit as it gets more complicated than a regular TIA circuit. When trying to implement an FDA with a TIA circuit we always recommend a TIA followed by an FDA. The video found here highlights idea of a TIA circuit and why it is implemented how it is. Trying to implement a TIA stage with an FDA has hardware configuration issues as the current source for the TIA is a photodiode that requires a certain biasing configuration and would not be connected like the design shows in TINA although it does result in some kind of output when simulating the circuit using an ideal current source. As for your last question, a common circuit component that is used to convert a differential output to a single ended signal is a transformer and it is primarily used for measuring into our single ended lab equipment. Figure 35 in the datasheet shows this configuration to convert the fully differential output to a single ended signal.

    Best Regards,


  • Hi Ignacio,

    Thank you for your thorough response Slight smile

    You make a good point with the biasing of the photodiode being an issue as the capacitance (C1) and resistor (R1) would change as a voltage would build up on the anode of the photodiode for the FDA configuration. 

    The link to the video is an awesome resource for me, thanks! I will proceed with sticking to a classic TIA topology and then potentially follow with a FDA. 

    I will also look into the transformer you suggested as well from the datasheet. 

    This resolves/answers my questions!