[rogerlike]

Hi, I've been looking at some solid-state DIY head-amps, and specifically the designs for the PPAv2 and the M^3. They both look very similar. The main difference that I can see is that the M^3 has a MOSFET discrete output buffer and the PPAv2 has a more complicated BJT discrete output buffer instead. They look like they have slightly different power supply sections for the buffers and op-amps too but, one thing that puzzles me is that the M^3 has a capacitor in the feedback loops, right in the signal path (where the PPAv2 has resistors instead, except for the ground channel), even though the AMB site talks about there being no capacitors in the M^3 signal path.

I am looking at the following schematics

http://www.amb.org/audio/mmm/mmm_sch.pdf
http://tangentsoft.net/audio/ppa/amp...schematic8.pdf

I'm just wondering if anybody can shed a little illumination on this for me, since i'm curious. Also, if anybody can comment on the qualitative effects on the sound produced by the differences in design I'd be interested.

Thanks!

 

[wolf18t]

Quote:

Originally Posted by rogerlike /img/forum/go_quote.gif ... one thing that puzzles me is that the M^3 has a capacitor in the feedback loops, right in the signal path (where the PPAv2 has resistors instead, except for the ground channel), even though the AMB site talks about there being no capacitors in the M^3 signal path.

I am no king of electronic circuit but, to me, the negative feedback loop is not in the signal path. It is used to correct "errors" and the gain of these amplifiers topology.

As for the capacitor in the feedback loop I suppose it is used to control the frequency response of the feedback loop or avoid oscillation.

 

[amb]

The negative feedback loop is in the signal path, but that capacitor (C1) is not a coupling capacitor. It's a compensation capacitor used to "tune" the ultrasonic behavior of the amplifier, and has no effect within the audio frequencies. As such, audio frequencies do not "pass through" it.

 

[rogerlike]

Quote:

Originally Posted by amb /img/forum/go_quote.gif The negative feedback loop is in the signal path, but that capacitor (C1) is not a coupling capacitor. It's a compensation capacitor used to "tune" the ultrasonic behavior of the amplifier, and has no effect within the audio frequencies. As such, audio frequencies do not "pass through" it.

Hi AMB, thanks for the reply. I'll have to think about what that means ^^

 

[tangent]

The impedance of a capacitor varies with frequency, decreasing as frequency goes up. (An ideal one does, anyway; we can ignore non-ideal behavior here.) But, there's a constant 11K ohm or so path past that capacitor in the feedback loop: R3 + R4. The frequency at which the impedance of an ideal 33pF capacitor is 11K is over 400 kHz. Down at the top of the audio band, its impedance is over 220K, far higher than the other path.

So no, it's not really in the audio path. Not even for bats.

 

[rogerlike]

Quote:

Originally Posted by tangent /img/forum/go_quote.gif The impedance of a capacitor varies with frequency, decreasing as frequency goes up. (An ideal one does, anyway; we can ignore non-ideal behavior here.) But, there's a constant 11K ohm or so path past that capacitor in the feedback loop: R3 + R4. The frequency at which the impedance of an ideal 33pF capacitor is 11K is over 400 kHz. Down at the top of the audio band, its impedance is over 220K, far higher than the other path. So no, it's not really in the audio path. Not even for bats.

Very interesting, thanks for the explanation