Solidstate Idea

[analag]

This is an idea for a simple transformer input coupled mic pre (obviously the input caps would be disregarded) running off the phantom supply, it's still in it's infancy. I'm going to breadboard it then implement some kind of gain control, but before I do...all corrections are welcomed.

analag

 

[NewYorkDave]

Very ambitious, with all those nested feedback loops. Let's see, there's bootstrapping feedback to the emitter load of Q2, a negative loop around Q3 and M1, and an overall negative loop from M1 to Q1. Stability could be tricky, although I see there's only one R-C time constant inside the global NFB loop--which is good.

Can't comment further without doing an analysis, but my initial impression is that it's an interesting circuit.

 

[Larrchild]

It hard, gaan to bed, Rowan! Nice-Up, General.

Haffi check dem feedback loops, lilly bit phase shift at rf freq and you'll hava rasclotillator!

 

[JohnRoberts]

[quote author="analag"]

This is an idea for a simple transformer input coupled mic pre (obviously the input caps would be disregarded) running off the phantom supply, it's still in it's infancy. I'm going to breadboard it then implement some kind of gain control, but before I do...all corrections are welcomed.

analag[/quote]

As has been noted, consider some internal compensation to roll off gain before internal delay/phase shift turns negative feedback into positive. My first guess is small C across base-collector of Q3 but see what works.

JR

 

[analag]

[quote author="JohnRoberts"] Consider some internal compensation to roll off gain before internal delay/phase shift turns negative feedback into positive. My first guess is small C across base-collector of Q3 but see what works.

JR[/quote]

The spike in the 4-5 Mhz region was accentuated when a 100pF Miller capacitance was introduced into Q3, but placing the same capacitance across the base-emitter rolled off the spike considerably.

Hey Lar, if we ever meet each other I'll just speak to you in my native tongue. I see you got the hang of it.
I take it....rasclotillator means: rahsclaut oscillata

analag

 

[Guest]

Classical Schmitt trigger (replace R1 by capacitor to get asymmetrical multivibrator)

Nice grass, buddy! :thumb:

 

[pstamler]

What ratio transformer were you thinking of using? There's only about 13.4dB of gain in the electronics as shown.

Also...won't the 4.75k resistor in series with the first transistor's base degrade noise performance? Or will the tranny be high-enough ration that this won't matter?

Peace,
Paul

 

[analag]

Transformer ratio could be 1:2 for input and 1:1 for output. I want to try some M6 stuff for the input....just for listening and testing purpose. I think modern condenser microphones have almost line level output.

I revised the design, opting for increased headroom, better stability, linearity and so on.

analag

 

[Larrchild]

Can you better explain this circuit's use? Is it a remote mic pre?

 

[Guest]

[quote author="Larrchild"]Can you better explain this circuit's use? Is it a remote mic pre?[/quote]

It is pure creativity. Modern art.

Good grass? :thumb:

 

[Larrchild]

Wow a Jamaican with a Russian accent.



I asked him.


not you.

 

[Guest]

[quote author="Larrchild"]Wow a Jamaican with a Russian accent.



I asked him.


not you.[/quote]

He is busy.


Very busy.

 

[analag]

[quote author="Larrchild"]Can you better explain this circuit's use? Is it a remote mic pre?[/quote]

It's an idea I'm messing with. I haven't heard a solidstate pre I like yet, so here goes.

analag

 

[NewYorkDave]

Is the local negative feedback around Q1-Q2 intentional? If not, you'll want to AC-decouple Q1's base-biasing network.

Taking the signal off the junction of the two emitter resistors on Q2 results in additional loss; along with that local feedback loop, this is working at cross-purposes to your global feedback loop from Q4 to Q1. It also causes some added phase shift due to the higher source impedance looking back from the base of Q3. I'd take the output right off Q2's emitter and adjust the DC operating points of the circuit as necessary.

Slick trick: you could split Q1's collector resistor and cap-couple Q2's emitter to the junction of the two resistors, to bootstrap Q1's collector load and increase gain. Same deal with Q3-Q4.

I'd also consider AC-bypassing Q3's emitter resistor. All the voltage gain in this circuit is coming from Q1 and Q3 so there's not much available to throw away in nested feedback loops.

All of this assumes that the primary goal here is the amplifier's AC performance. :wink:

Discrete transistor design is not my specialty, so I invite PRR, John, Brad or anyone else to punch holes in my analysis. I don't mind.

 

[Svart]

Discrete transistor design is not my specialty

Not mine either.. but I did stay at a holiday inn express...

 

[analag]

Well I built the damn thing. I'm scoping it and drinking wine right now. It's a little different from the posted schemo, much better in actuality. Symmetrical sines, tight squares, nice gain ratio between stages.
I came across a very unusual output arrangement which I tried and it is as the author described....very good.

analag

 

[skipwave]

[quote author="analag"]I came across a very unusual output arrangement which I tried and it is as the author described....very good.[/quote]

Author's name? You gotta give us a bit more to go on, these little clues border on oblique.

I do think this a great initiative, we need more designs that run on a simple 48V rail.

 

[analag]

[quote author="skipwave"][quote author="analag"]I came across a very unusual output arrangement which I tried and it is as the author described....very good.[/quote]

Author's name? You gotta give us a bit more to go on, these little clues border on oblique.

I do think this a great initiative, we need more designs that run on a simple 48V rail.[/quote]

http://www.dself.dsl.pipex.com/ampins/discrete/cfp.htm

It's in this article, I just used a single supply and tweaked it a little.

analag

 

[skipwave]

[quote author="analag"]http://www.dself.dsl.pipex.com/ampins/discrete/cfp.htm

It's in this article, I just used a single supply and tweaked it a little.

analag[/quote]

Cool. I was giving that article some attention recently. I got started on it because of a thread here about a preamp for high output mics with a single Emitter Follower. The Sziklai or Complementary Feedback Pair was suggested to improve distortion.

So did you convert your EF into a CFP?

 

[analag]

[quote author="skipwave"]So did you convert your EF into a CFP?[/quote]

Yes I did, conducting very high current too.
I'm trying to see if I can make a solidstate pre that is as satisfying as my tube stuff.

analag