tangent
Top Mall-Fi poster. The T in META42.
Formerly with Tangentsoft Parts Store
- Joined
- Sep 27, 2001
- Posts
- 5,969
- Likes
- 58
Quote:
Yes. It's called "peaking", and it means that op-amp has a tendency to oscillate at that peak frequency. The op-amp actually has some additional, unwanted gain here, which exacerbates any positive feedback in your circuit. I've been told that if this peak is more than 6 dB (?), oscillation is guaranteed. If you look at the gain vs. frequency graph of an uncompensated op-amp, you'll see a high peak like this if they show a G=1 curve, but no peak (or a very much flattened one) at the recommended minimum gain. Take a look at this graph in the AD8397 datasheet: there's a 3 dB peak at 40 MHz at G=1, but no peak at G=2. Guess what: this op-amp is cranky at gains less than 2, and if it oscillates, it's most likely to do it at 40 MHz.
This doesn't have much to do with slew rate -- not directly, anyway -- but it's one way op-amp manufacturers fudge the bandwidth spec on their op-amps. An op-amp with peaking looks faster on the front page of the datasheet than it is in practice. Some datasheets give bandwidth along with a gain flatness spec to take this into account. If you see a bandwidth spec of X MHz with bandwidth flatness +/- 1 dB (or whatever), the op-amp certainly has some peaking, so the bandwidth without taking peaking into account would be X * Y, where Y is > 0, and almost certainly < 2.
Quote:
You could estimate it from the rise time spec. But beware: this won't take into account overshoot.
| Originally Posted by Tomo Spectrums (frequency respose diagram) show a small peak right before high-end 3dB drop. Does this mean anything? |
Yes. It's called "peaking", and it means that op-amp has a tendency to oscillate at that peak frequency. The op-amp actually has some additional, unwanted gain here, which exacerbates any positive feedback in your circuit. I've been told that if this peak is more than 6 dB (?), oscillation is guaranteed. If you look at the gain vs. frequency graph of an uncompensated op-amp, you'll see a high peak like this if they show a G=1 curve, but no peak (or a very much flattened one) at the recommended minimum gain. Take a look at this graph in the AD8397 datasheet: there's a 3 dB peak at 40 MHz at G=1, but no peak at G=2. Guess what: this op-amp is cranky at gains less than 2, and if it oscillates, it's most likely to do it at 40 MHz.
This doesn't have much to do with slew rate -- not directly, anyway -- but it's one way op-amp manufacturers fudge the bandwidth spec on their op-amps. An op-amp with peaking looks faster on the front page of the datasheet than it is in practice. Some datasheets give bandwidth along with a gain flatness spec to take this into account. If you see a bandwidth spec of X MHz with bandwidth flatness +/- 1 dB (or whatever), the op-amp certainly has some peaking, so the bandwidth without taking peaking into account would be X * Y, where Y is > 0, and almost certainly < 2.
Quote:
| Do anyone know the slew rate of 74HCU04's? |
You could estimate it from the rise time spec. But beware: this won't take into account overshoot.
