1. Home
  2. Electronics Design
  3. RFI versus stability, in opamp circuit

RFI versus stability, in opamp circuit

Consider a simple opamp circuit, with feedback resistor from output to inverting input, and a capacitor in parallel with that resistor to reduce gain at high frequencies so as to avoid oscillation.

That capacitor also serves as a path for out-of-band RF coming in through the output line to travel to the inverting input, where it will perhaps be rectified and demodulated, creating in-band noise or distortion on the output.

One way around that would be to put an RF filter on the output line. But I think to do that I either need inductors big enough to handle whatever current I'm supplying or series resistance (which would increase the output impedance of my amp). Is there another common approach? E.g., make the feedback network be a bandpass rather than hipass filter?

Thanks! -walter

Reply to
Walter Harley
Loading thread data ...

when you put the 15-100p cap directly from O/P-pin to inver. input and go out via 100 or 220R and put the feedback resistor from there you have not only avoided your scenario, but also isolated the cable capacitance/inductance from the O/P, actually that is what everybody does in stages which go off board.

--
ciao Ban
Apricale, Italy
Reply to
Ban

Thanks, Ban. Do you mean to take the feedback from the opamp side of the

100R or from the output side? It would seem that if it is the opamp side, then 100R has been added to the output impedance (which is bad if I want low output impedance); if it is the output side, then I still have an RF short from output to inverting input.

In audio power amps (which must have low output impedance) I usually see an L in parallel with an R between the output devices and the speaker terminals. I could do that here, too. But I was wondering if there was a way to manage it without compromising output impedance and without requiring an inductor capable of handling the output current.

Thanks! -walter

Reply to
Walter Harley

"Walter Harley"

** Are you concerned with op-amps or power amps ???

An op-amp has only miniscule output current.

However, most will not drive a capacitive load or arbitrary length of shielded cable without risk of oscillation unless there is an output resistor or inductor in series.

....... Phil

Reply to
Phil Allison

___ .--------|___|-. | || 10k | o-||---. | | |15p | | | | | ___ | |\\ | | o-|___|-o-|-\\ | ___ | || 10k | >-o-|___|-o--||--o GND-|+/ 220 || 10u |/ (created by AACircuit v1.28 beta 10/06/04

formatting link

From the output you do not need any inductor, and in fact I have never seen one there. The alleged RF gets shorted to gnd by the input capacity of the opamp, which is as high as the small compensation cap, this also is the case with the O/P, which has a higher capacity. If something in the lower MHz range enters there it will be partially cancelled by the inverting action of the opamp. The output current is at the most a few milliamps, usual load is 10k, the

220R will only reduce the output swing, which anyway has enough headroom, if done properly. 
--
ciao Ban
Apricale, Italy
Reply to
Ban

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.