Any reason why this wouldn't work? (I didn't show the cascode loads for the input diff pair...too lazy to draw it).
I've built a few servos in the past, and generally make the resistor at the opamp output about 10x the size of the biasing resistor for the input transistor. In this case, that would be about 1M, which seems awfully large.
Wanted to do this for a while, as the Spec-2 does not have any adjustment mechanism for offset.
I've built a few servos in the past, and generally make the resistor at the opamp output about 10x the size of the biasing resistor for the input transistor. In this case, that would be about 1M, which seems awfully large.
Wanted to do this for a while, as the Spec-2 does not have any adjustment mechanism for offset.
An externally hosted image should be here but it was not working when we last tested it.
funny, I've just written about it one thread below. It may be problemous or not.
http://www.diyaudio.com/forums/showthread.php?postid=749282#post749282
basically this implementation of servo mechanism has total 180deg. of phase shift, because input C coupling works like a low pass for a servo output, which may lead to oscillations.
regards
http://www.diyaudio.com/forums/showthread.php?postid=749282#post749282
basically this implementation of servo mechanism has total 180deg. of phase shift, because input C coupling works like a low pass for a servo output, which may lead to oscillations.
regards
Adding a nother pole may indeed be a problem on some amps, the frequency must be signifficantly lower than the poles introduced by the input C if any and the feedback C, otherwise you get LF oscillation. In particular, the feedback C may be quite large and the low end cut-off very low (as in a couple of Hz at most) to minimize capacitor distortion, and the input can, of course, be DC coupled. You may end up with impractically high values of integrator C and R.
The resistor from the OPamp to the input depends on the range of output voltage from the OP, and that of course depends on the OP power supply. For all normal conditions, you don't want the OP output to get too close to it's power rails, to maintain proper servo action. If not, and if the OP is latchup-free, there should be no dramatic ill effect except that the servo may not be able to correct largeish offsets completely.
As an aside, I've seen solutions where the servo output voltage was sued as an excessive DC detector in a speaker protection scheme...
The resistor from the OPamp to the input depends on the range of output voltage from the OP, and that of course depends on the OP power supply. For all normal conditions, you don't want the OP output to get too close to it's power rails, to maintain proper servo action. If not, and if the OP is latchup-free, there should be no dramatic ill effect except that the servo may not be able to correct largeish offsets completely.
As an aside, I've seen solutions where the servo output voltage was sued as an excessive DC detector in a speaker protection scheme...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.