Simple Symetrical Amplifier

[keantoken]

Lazy Cat, I have one sitting behind me, it is a delay line, not a bunch of inductors or capacitors. I have another one inside my currently functioning Tektronix 465B. The delay line is used with triggered ramp (my memory is foggy here) to allow the variable delayed trigger feature on these scopes.

[Esperado]

To measure the slewrate, you remove any filter before the input. Then you apply a square wave with a very fast transition time, at a very hight level, more than enough to bring the output of the amp to its max level. And you measure the transition time between 10% and 90% of the amplitude of the output signal.
Because the amp will not be able to follow the signal speed during the transitions, cr will not have time to send an inverted signal during those transitions, so the slew-rate is just a matter of the speed capacity of the amp in near open loop condition.
(Sorry for my poor english, i would be more comfortable to explain-it in French).

[mikelm]

Gold Star for explanation & English

[Esperado]

Originally Posted by mikelm

Gold Star for explanation & English

Not so sure, but thanks .

Once we know the slew rate of our amp, we will have to filter the input signal with a low pass filter is such a way that never any incoming signal will reach this slew-rate slope in order to avoid TIM.
As LAZY CAT suggest in a humoristic way, we are not in concern with delay, in the CR of an amp. It is a group delay.
Just to consider each pole (stage) as a low pass filter. Open loop bandwidth will be the one of the slower one ( if Nyquist allows). and phase shift an addition of all individual poles phase shifts.

Well, in a conventional 'voltage feedback' topology, the signal is applied to the non inverting stage of the differential input stage, while CR is applied to the inverting input.
So there is an additional pole in the CR path (the inverting stage) before the two signal are mixed together. And this pole add his slew-rate and bandwidth limitation, phase shift, noise, distortion and saturation risk to the CR signal.

In a so called "curent feed back", the CR is mixed with the original strait at the output of the input stage. Without any additional pole*. That makes all the difference.

* if the CR resistance is low enough that capacitances to ground at the end of this resistance (acting as a low pass filter) have no or very little influence at our used frequencies.

[Jay]

Originally Posted by Esperado

(Sorry for my poor english, i would be more comfortable to explain-it in French).

I believe you know your stuff very well because I like how you explain the electronics things. But you like to use CR instead of FB. I believe CR stands for Contre Reaction (?), where word by word it doesn't make sense

[Jay]

Originally Posted by Esperado

* if the CR resistance is low enough that capacitances to ground at the end of this resistance (acting as a low pass filter) have no or very little influence at our used frequencies.

How about capacitor quality issue? I don't know why, I don't like the sound of a big cap in the feedback loop. If it is high ESR problem, I think theoretically it doesn't matter? 1000uF is not enough for 1K/35R but 2200uF. And once I forgot to replace back the cap with voltage source and I saw horrible simulation result?

I have tried to use opamp to eliminate the cap but it seems that opamps do not work in this circuit?

[Lazy Cat]

PCB-s for my new CSA amplifier, soft start & utility PCB, plus rear panel utility PCB already assembled. CSA channels PCB-s will go into production this week after extensive check-ups made, so all together four PCB-s and two toroidal transformers as a separate parts form new CSA amplifier. Rear panel connections in a previous, sold SSA, simply took too much time for all the wiring necessary. New PCB will make things much easier now ...

[Esperado]

Originally Posted by Jay

I believe you know your stuff very well because I like how you explain the electronics things. But you like to use CR instead of FB. I believe CR stands for Contre Reaction (?), where word by word it doesn't make sense

Oh, so sorry. Thanks for the correction. Yes, In French "Contre Reaction" means exactly " Feed back". And years of habits...
But, it is a pity, the forum do not let-me edit my post anymore.
Please forgive-me.

[Lazy Cat]

Originally Posted by Esperado

In a so called "curent feed back", the CR is mixed with the original strait at the output of the input stage. Without any additional pole*. That makes all the difference.

In the past, before SSA, there was only LTP input diff I used regulary (cascoded, mirrored, you name it ...), but was really never pleased with the sound, so after SSA they stayed where they belongs - in the past, at least as far as I'm concerned.

[wahab]

Originally Posted by Jay

I believe CR stands for Contre Reaction (?), where word by word it doesn't make sense

Literaly translated , it s "counter reaction" , that is , not FB but NFB..