My photo diode is here Characteristics are PW 10ns-150ns, PRF - 1Hz-1Khz, current output 10nA-100mA

i have selected a TIA for processing the above signal to an extent, by using LT spice simulator i was able to simulate the IC and its behavior and checked the performance to an extent.

but i want to emulate the photo diode behavior so that i can feed a signal similar to photo diode to my read out circuit

Problem statement:

in short i want to know how can i create a current source, in fact a pulsating current source with above characteristics,

What i want to try :

to solve the problem i have decided to go with a voltage source like this, what if i place a series resistor between TIA input and voltage source, by tweaking the series resistor wont i be able to change the current to the TIA ? am in the right direction ?? is there any alternate approach to create a current source having above characteristics ?


Any way due to opamp saturation at maximum the current can only go to vcc/rf which is 2.5V / 1K= 2.5mA

Please observe that the circuit shown is meant only to showcase what am i trying.


simulate this circuit – Schematic created using CircuitLab


I want to generate a minimum current of 30uA and maximum 2.5mA with above pulse characteristics, The voltage source is a positive polarity one, while i have negative polarity supply also, link to source

As raised by the answers, to clear more on the circuit, the circuit involves a protection diode at opamp input to surge extra current and also protect opamp input from voltage greater than 2V, the TIA is followed by a inverter then a high speed comparator which is a LVDS output having a 100 Ohm termination, that is it !

Something like this :


simulate this circuit

I know few short comings of the circuit, i want measure as much is possible from the circuit,please throw your comments more on creating current source of characteristics discussed in background, if not the above circuit i can experiment with some other, but my main intention in this question is to discuss and come to a conclusion on the best way to create a valid current source

  • 1
    \$\begingroup\$ You don't appear to be learning from your earlier questions. 100 mA through a 1 kohm resistor implies the opamp can produce 100 volt on its output whilst sourcing 100 mA. Do you really think the opamp chosen can do this? \$\endgroup\$ – Andy aka Mar 4 '17 at 11:21
  • 1
    \$\begingroup\$ Also, you won't get 100 mA from the photodiode in its linear operating region by a mile. Yes, if you do the math on the absolute max incident power (and effective receive area) it implies a few hundred mA but that assumes the responsivity remains at the normal value at these extremes and it won't. \$\endgroup\$ – Andy aka Mar 4 '17 at 11:31
  • \$\begingroup\$ I understand. The photo current is limited to 2.5mA. But iam more keen in knowing how can i create a current source \$\endgroup\$ – kakeh Mar 5 '17 at 7:46
  • \$\begingroup\$ slightly related noise (photodiode) question: How to estimate minimum detectable light pulse in a photodiode (5 ms, 44 kHz modulated)? any thoughts? \$\endgroup\$ – uhoh Jan 22 at 2:02

The opamp is ok. The circuit is not ok. You can burn the input of your opamp due its current mode nature and low impedance. 100mA photocurrent is not realistic. You will need hundreds of milliwatts light input and that would disintegrate the actual diode.

Find realistic photocurrent limits from the wanted detection requirements. Calculate the series resistor for your pulse gen output from that.

You really should keep in your mind that pulses this fast require proper transmission lines and matching. Your image ignores all that. This will never work without proper high speed pulse capable circuit design. A good opamp selection is only the beginning.

  • \$\begingroup\$ True its un realistic when i utter 100mA. My interest here more is how to create a current source \$\endgroup\$ – kakeh Mar 5 '17 at 7:48
  • \$\begingroup\$ @kakeh Your new edits have made much more clear what is wanted. Add still more: Polarity and which loads must be properly supplied. Note: A new answer must be written. This answer assumes that you have tried also to make a sense amp for the photodiodes. NOTE: If your diode give its photocurrent to already known load (a known resistor or a known current mode amp input), a passive 2...3 resistor network + a pulse generator should be an excellent simulation for the photocurrent pulses (=a known pulse easily generated in as high quality as your pulse generator is able) \$\endgroup\$ – user287001 Mar 5 '17 at 11:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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