• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 52 subscribers
  • Views 2084 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

OPA 657 as transimpedance

Anas Ansari
Prodigy 10 points

Other Parts Discussed in Thread: OPA657, TINA-TI, OPA847, OPA659

Hello Everyone...

I want to use OPA657 as transimpedance amplifier for data rates upto 100 Mbps.

My photodiode is OPF430 which has capacitance of 2pF when biased negative (from datasheet).

the figure given below is the TIA using OPA657 where Cf = 10uF

 

i want to ask, is this circuit going to work at 100 Mbps ?

 

Thanks

i have tried to use it but couldnt find result. can anybody here p

  • 0
    Genius 5430 points

    Anas;

    With a 10uF feedback capacitor it sure won't work at 100 MHz.

    I assume that is a mis-type but since you've drawn your circuit in TINA-TI, why not do a simulation with the correct Cf and find out?

  • 0 in reply to Neil Albaugh
    TI__Expert 4115 points

    Anas,

    Mbps implies some type of data stream with squarewaves. It will be hard to get that with op amps. 100MHz sine wave is achievable, but will take some tweaking to get a response you like. I did the basic calculations of bandwidth per the equations in the applications section of the OPA657 data sheet, and get about 87MHz of BW. Given the low TIA gain (4.7k), the OPA847 is probably a better choice with 150MHz of BW. The problem with either is the (theoretical) ideal CFcap is fractions of pF, and board parasitic may be larger; thus tweaking with prototype will be needed. I attach Tina file with example using OPA847.

    Note: TIA design formulas for calculation of bandwidth, compensation, etc is available in in the applications section of the OPA657 data sheet and other sources can be found on the web like http://www.ti.com/lit/an/sboa122/sboa122.pdf 

     

    8420.100MHz TIA with OPA847.TSC

     

  • 0 in reply to Jim Karki
    Genius 5430 points

    Jim; Is there some reason that the OPA847 input capacitance is not included in the OPA847 model?

    If the customer's requirement is for 100 megabits per second, he will need only 50MHz BW if it is an NRZ code.

  • 0
    TI__Expert 5400 points

    Muhammad,

    With regard to the offset voltage of the OPA847 simulation, the input resistances are matched as well as can be expected (4.7kohm on each terminal), but given the tolerance of the resistors and a slight mismatch of the capacitive components, there can still be slight offset at the output, though well within the offset specifications of the device.  The offset also appears more significant in this simulation since the output signal is so small (~500uV).

    One way to nullify this offset would be to replace R3 with a potentiometer, and tune to match precisely the impedance at both input terminals.  Another possibility in your square wave application would be to place a DC blocking capacitor between the output of the transimpedance stage and the input of the comparator stage.  Note that this would need to be followed by a resistive divider network after the capacitor to maintain stability of the comparator.

  • 0 in reply to Bart Stiller
    Prodigy 45 points

    Hello Bart,

    Thank you for the reply.

    Bart Stiller said:

     Another possibility in your square wave application would be to place a DC blocking capacitor between the output of the transimpedance stage and the input of the comparator stage.  Note that this would need to be followed by a resistive divider network after the capacitor to maintain stability of the comparator.

    Can you give me an example of a resistive network that i would need.

    Also,

    When I try to build these circuits, on breadboards and test boards(not pcb), i am getting a lot of noise and not the desired response.

    Do I have to build PCBs for better responses?

    I have tried circuits with OPA657, OPA659  and OPA847, and generally all of them have so much noise that the signal is almost lost.

    What should I do?

    Best Regards,

    Muhammad Talha

  • 0 in reply to Muhammad Talha
    TI__Expert 5400 points

    3343.upload_002[1].TSC

    In transimpedance applications, the signals are typically so small that successful implementation on a bread board would be quite difficult.  It's difficult to predict how noisy a bread board circuit will be with so many variables, but certainly PCB use is recommended.  Companies like PCB Express offer free layout tools and low cost board manufacturing that may be useful for you --

  • 0 in reply to Bart Stiller
    TI__Expert 5400 points

    Muahmmad,

    If you have further questions, please post in a new thread -- Thanks.