Hi Hank,
the OPA857 is no simple OPAmp, but a very special TIA:
The +input of internal OPAmp is biased at 1.83V. The feedback path of OPAmp makes that the -input of OPAmp sits at the same voltage. So, if no input current is flowing into the "IN" pin, the output "OUT" will also emit 1.83V. Output "OUTN" emits 1.83V as well.
Now, the special of OPA857: The chip only accepts input currents flowing INTO the "IN" pin. Because of this, the output "OUT" can only emit voltages equal and below 1.83V! Concretely spoken, the output "OUT" can only swing between 0.6V and 1.83V. That's exactly what the blue curve of figure 13 is showing.
Kai
Hi Hank,
I think the OPA857 is not fast enough to handle big signal currents, as Hasan already mentioned.
From the performance of a 10MHz signal with 20mA amplitude you would choose a 250R feedback resistance:
This will properly work with small signals, even at 100MHz:
But with a signal of 2mA amplitude problems start:
And with a 20mA signal the OPA857 collapses:
Kai
Hey Hank,
So with the specs provided, the OPA859 would probably be your best choice. I have attached a schematic below where you can see the recommended circuit, and see if it will meet your requirements. Because of the very fast pulses of your design, there will be a little bit of ringing at the output peaks, however, the OPA859 should be able to meet your bandwidth requirements. You can see this in the picture shown below, cursors a and b are set to 1V and -1V respectively:
If you would like to achieve a differential output, figure 57 of the OPA859 datasheet uses the following configuration:
You can find more details of this circuit on page 21 of the datasheet. Please let me know if there is anything else I can do to help.
Best,
Hasan Babiker
