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.

Transimpedance Amplifier using OPA2354AQ1

Other Parts Discussed in Thread: OPA2354

I had some questions regarding T.I. part# OPA2354AQ1.  The company I work for wants to develop an Application Note for a Single Power Supply Transimpedance Amplifier which uses one of our company's optocouplers along with the OPA2354AQ1.  When testing the circuit (schematic below) with a voltage pulse coming from the optocoupler (only photodiode is shown on schematic), the OPA2354AQ1 outputs a non-biased pulse signal very similar to the input when its output is expected to be a 2.5V biased inverted pulse.  And the strangest thing, when the VCC is removed from Op-Amp it still outputs the non-biased pulse, like it is being shorted.  I am testing on a breadboard, I used a DIP Adapter purchased from T.I. for the OpAmp.  I thought maybe I shorted some of the pins together when I soldered but so I checked resistances between the pins and that does not seem to be the case.  I am looking for any advice as we are set on creating this application note using T.I. part# OPA2354-Q1

  • Hello Joshua,

    That is very strange.

    What happens when the photodiode is removed from the circuit...where does the output of the amplifier settle to?

    What is the quiescent current being consumed?

    I don't understand your comment on removing Vcc...can you please provide a scope shot of the output with and without Vcc applied.

    What is the 2nd channel of the amplifier doing? I was wondering if it is railed out and is causing the 1st channel to fail.

    Can you please provide some good pictures of the top and bottom side of the DIP package and a picture of the breadboard so I can give it another look to see I spot any issues?

    -Samir

  • Hi Samir,

    Thanks for response, and sorry for the delay.  It's my first day back in the lab today.  As for your questions:

    What happens when the photodiode is removed from the circuit...where does the output of the amplifier settle to?

    It settles at zero (ground) if I just feed a pulse into op amp, even though it is still 2.5V biased.

    What is the quiescent current being consumed?

    I assume you are referring to leakage current from photodiode which is around 500nA when Vcc is 5V.   If this is not what you are referring to please let me know.   

    I don't understand your comment on removing Vcc...can you please provide a scope shot of the output with and without Vcc applied.

    Sure here is the picture of both.  Essentially I leave the scope hooked up and just remove the Vcc wire for the picture “without Vcc”.  

    With Vcc:

    Without Vcc:


    What is the 2nd channel of the amplifier doing? I was wondering if it is railed out and is causing the 1st channel to fail.

    There is no signal coming from other channel.  I checked all inputs and output for that second channel with the scope.  

    Can you please provide some good pictures of the top and bottom side of the DIP package and a picture of the breadboard so I can give it another look to see I spot any issues?

    Thanks again for your help Samir.

  • Hello Joshua,

     Couple of comments and observations:

    1. With a high speed amplifier like the OPA2354, using a device in a socket vs. soldered down on a PCB does pose some risks with regards to stability. Refer to pahe 15 of the datasheet for additional information. Also, you do not seem to have any bypass caps. close to the power supply pins. This is also not good. I would recommend purchasing an evaluation module from here to check basic testing and functionality:

    www.ti.com/.../DEM-OPA-MSOP-2A;tisearch=Search-EN-Everything#Technical Documents

    2. It looks like the 2nd channel of the amplifier is just left floating. This is not good. I would recommend tying pin 5 to 2.5V and shorting pins 6 and 7 in a unity gain buffer config.

    3. Quiescent current is the current consumed by the amplifier when it is unloaded. In the case of the OPA2354 this will be around 4.9mA * 2=9.8mA. See page 5 of the datasheet for this spec. You will see more current than this in your case since you have the two 100Ohm resistors to GND. (The blue resistors on the breadboard)

    4. Removing Vcc makes no sense as to why the amplifier output is moving. An amplifier requires power supplies to work. This tells me that the amplifier is getting power at its pin 8 from somewhere else. Are you just disconnecting the orange wire from pin 8 and seeing the output move?

    5. When you disconnect the photodiode from the amplifier terminal the output should be at 2.5V. The amplifier is simply in a unity gain buffer config here. Please confirm if this is true.

    -Samir

  • Hi Samir,

    Thanks for the tips. I will take into consideration the evaluation board you recommended, we just really wanted to use sockets so we could easily swap out components like resistors for different simulations.

    I added the caps like you recommended and tied down Pin 5 to 2.5V and shorted Pin 6 and 7 together, but I am still getting the same non biased output as before.

    As for the Vcc, yes I just removed the orange wire from Pin 8 (originally by mistake) and noticed that somehow the signal still comes out of the op amp output which is boggling to me. I will try to find where the mystery power source is coming from.

    And lastly, when I remove the photodiode the signal goes away and just 0V (GND) and not the 2.5V DC bias output I was expecting. Any other recommendations and thank you so much for you help already?

    -Josh
  • Also is there any possible way to get a sample of that evaluation board you recommended?
  • You can order the EVM from the link below. Click on the "Buy from TI" button:

    http://www.ti.com/tool/dem-opa-so-2a?keyMatch=dem-opa-msop-2a&tisearch=Search-EN-Everything

    Thanks,

    Samir