OPA857: OPA857 schematic review plz..

Hello Adrinil,

  Thanks for reaching out. We will be reviewing your questions and feedback and get back to you by mid this week.

  Quick question that would also help with the review, what is your PD's internal capacitance? 

Thank you,
Sima

Hello Adrinil,

  I apologize I missed the reply to your thread. Thanks for reaching out again!

  I placed my feedback on a word document attached, if you are unable to open it I also copied it below.

Photodiode:

0.6pF to 0.75pF when VR = 5V

Rise time = 0.3ns à bandwidth of 0.35/(0.3ns) = 1.2GHz

Max current of 10mA

Typical current of 240uA

My Questions:

1. Test_IN is connect to VS+ for normal operation?
   1. Double checking since I can not tell from schematic
2. CTRL always connected to gnd, 5kOhm transimpedance gain à closed-loop bandwidth of 130MHz when input PD capacitance is 1.5pF (This value is selected because the device is expected to be driven by a photodiode with biasing high enough to include the photodiode capacitance contribution between approximately 0.5 pF and 0.7 pF, leaving between 0.8 pF to 1 pF for the external parasitic.)
   1. It might be best to change to a different amplifier if you do not need the differential output of the TIA since not using the CTRL feature for switching gains (I recommend the OPA858 or OPA855 due to the high calculated closed-loop bandwidth needs of the chosen PD)

Questions Answered:

1. Is the output of OPA857 a positive voltage? (just wanted to confirm)
   1. Device is positive voltage, but will be a pulse signal that swings from 1.83V (Set DC bias internal to device) down towards 0.6V
   2. Bias the photodiode with the cathode connected to a positive supply, and the anode connected to IN pin of the OPA857. These connections make sure that the photodiode sources an output current that results in the OPA857 output swinging down below the reference voltage = (5 / 9) × 3.3 V = 1.83 V (In schematic, not sure if PD+ refers to cathode or anode, but looking at the PD datasheet you linked earlier, VR = 5V positive voltage which would work for this device)
2. Can we connect resistor and cap like this from IN to OUT?
   1. Yes, you can this will have 732 Ohm external feedback resistor in parallel to internal resistors. However, I am unsure how well this will work with this device with a feedback resistor that low. As shown in section 8.2.2 page 21 in the datasheet, adding this can increase your bandwidth, but I think your chosen capacitance values might be too high and cause even lower closed-loop bandwidth.
   2. An external feedback resistor, RF, is added in parallel to the internal transimpedance gain resistors of the OPA857. This resistor has the effect of reducing the overall transimpedance gain, but with increased bandwidth. (732 Ohms and parallel capacitors which seem too high, will reduce bandwidth)
3. In datasheet schematic, there's a 25ohm resistor after the op-amp. Does that affect the gain? Or is it there to impedance match for RF?
   1. This is a really good question and point out. Yes, this will affect your gain when you load the OPA857 since it will create a voltage divider. You would need to take these resistors in consideration for your total load and impedance match if output to lengthy cables. do not need to impedance match if your traces/length is very small to the load: the EVM shows impedance matching + loading device of differential of 500Ohms. For a 500-Ω load, including the GND alternatives resulting from the internal 25-Ω series resistor on each output, the resulting gain is 4.5 kΩ for 5kOhm internal selection.
   2. 

OPA857_Adrinil.docx

Best Regards,
Sima