Another newb question, sorry. How do some amps have both a high DF (>= 300) while simultaneously having a high power output (>= 100W). I understand that increasing NFB is a common technique to achieve higher DF, but that also limits the amplifier's gain. Unless there's a much more effective tactic, I guess my question is really How do some amps achieve the enormous amount of gain that enables them to also have such a high amount of NFB? I am just curious as to how, not so much the why, as there is already a lot of discussion as to how much DF is enough DF.
Cheap op amp chips can have 120dB gain or more at very low frequencies.
Then that bandwidth is shifted upward proportionately, as the gain is reduced by feedback.
Similar techniques are used for discrete amplifiers.
https://en.wikipedia.org/wiki/Gain–bandwidth_product
Then that bandwidth is shifted upward proportionately, as the gain is reduced by feedback.
Similar techniques are used for discrete amplifiers.
https://en.wikipedia.org/wiki/Gain–bandwidth_product
Standard 3-stage power amp tends to have from 80 to 120dB of open loop voltage gain at low frequencies - normally you only use 25 to 30dB of that once the feedback loop is closed, so even though the open loop gain drops at higher frequencies there's a lot of it to help linearize everything. This input stage typically converts voltage to current, the VAS converts current to voltage with lots of gain (think microamps in for dozens of volts out), and the output stage has lots of current gain (10000 ish?).
An emitter-follower VAS for instance might have a current gain of 10000, with a load resistance of 5k or so (at low frequency), so a transimpedance of 50Mohm. Given a standard input pair with emitter degeneration will have a transconductance of about 0.02S, combine them for a voltage gain of 10^6 (120dB)
An emitter-follower VAS for instance might have a current gain of 10000, with a load resistance of 5k or so (at low frequency), so a transimpedance of 50Mohm. Given a standard input pair with emitter degeneration will have a transconductance of about 0.02S, combine them for a voltage gain of 10^6 (120dB)
Last edited:
The emitter follower VAS has a high current gain, since most of the current signal from the first transistor goes into the second transistor's base. Thus a microamp level of input signal becomes milliamps at the VAS collector into the load impedance. The actual load the VAS sees is what converts current to voltage, and at higher frequencies the compensation capacitor forms a local feedback loop around the VAS that reduces open-loop gain by shunting current into the VAS input - this also serves to linearize the VAS at higher frequencies so long as the compensation capacitor is linear itself (i.e. is C0G/NP0)