Stephan Boettcher wrote:
And since I do not have much experience with GHz class opamps, I tend to put the parts rather too close than too far apart. I do not want anoscillator there.
Some recent work at my day job involved 400 MHz opamps. These things require careful layout to keep from oscillating, just as you say. The vendor (TI, I think) recommends removing copper on all layers around the input pins. Trust me, this works. One board accidentally removed the cutouts, and the opamp sang like a soprano. Other stuff that is pretty good practice is to place a series 10-Ohm resistor on the output of the amp. You can always zero it out, but it gives you the chance to decouple the output from the input - especially if you are running unity gain. Watch that phase margin!
I suggest to follow the current path, from opamp vcc/vee to load, and back to opamp. Be sure the return path is not through the supply, possibly off the card.And since we want to sample with 14-bit resolution, I do not want too big antennas on the input circuits either, to pick up digital noise from the back end.
_______________________________________________ geda-user mailing list geda-user@xxxxxxxxxxxxxx http://www.seul.org/cgi-bin/mailman/listinfo/geda-user