I've been reading that opamp circuits with gains less than unity (between 0 and 1) can become unstable. (i.e. - simply making the input resistance larger than the feedback resistance doesn't cut it).
Why is this?
In addition, I've found a few circuits that claim to rectify any instability, but don't have any associated math to back it up.
Anyone have any good references that can settle this?
Less gain = more feedback.
If the op amp itself is not designed for that much feedback, it can become unstable. It has to do with the way yht wgain and phase of the op amp reduce at higher frquencies
Goggle key words:
op amp unity gain stable gain margin phase margin
The gain of an op-amp circuit is set by the gain of the op-amp itself and the gain of the feedback network. For lower amplifier gains the feedback network must have more gain to "hold back" the amplifier more. No op-amp has perfect, infinite gain, however. They don't even look like perfect integrators. So there will be some setting of the feedback network gain for which the op-amp circuit will go unstable.
Any good book on circuit design should clear this up.
The vast majority of opamps are "unity gain stable" which means they are stable with Rf = 0, ie as a voltage follower or as an inverter with any gain. A few opamps are undercompensated, and are clearly identified on their datasheets as "stable for gains > 5" or some such.
Extremely high impedance feedback networks can cause instability, but that's a different issue.
John
At low frequencies the output is 180 degress out of phase with the inverting input. Eventually at some higher frequency, the phase will shift enough to become positive feedback. To keep the amplifier stable, those higher frequencies are usually shunted or rolled off so that they do not appear at the output with enough level to cause oscillation. That's the purpose compensation. More gain requires less compensation. Unity gain requires more compensation. Less than unity gain requires more compensation than unity gain and is not usually specified in a spec sheet.
The vast majority of opamps are internally compensated to be stable at any gain, including the follower configuration which has 100% negative feedback.
John
Just to emphasise what John said above, a "unity gain stable" opamp when operated as an inverter is stable for _any_ gain, including gains below 1, all the way down to 0. An inverter with a gain of -1 requires an amplifier gain of 2. Google on "noise gain" and "opamp".
Regards Ian
Have something to add? Share your thoughts — no account required.
Ask the community — no account required