New: A Direct-Coupled Input-Capacitorless Active Mic Preamp

[mediatechnology]

JR - That circuit is brilliant based on its simplicity.


Symetrix 528E MIc Preamp with input-capacitroless-front end.

By using a PNP front-end the common mode range includes ground intrinsically.
I like it. So obvious but not.

[JR.]

In my old late night doodling on this subject I used pnp (737s) but not like this.

Somewhere I have the old high voltage capacitors I bought to breadboard one (decades ago), but never did... similar +60V rail

BTW the NF does not wrap around the input devices so this is a variant on open loop soundcraft preamp. Not uber-linear performance.

note: they could cap couple some NF up to the device emitters and make this right, or not. They have both polarities available in the dual op amp.

JR

[mediatechnology]

In my old late night doodling on this subject I used pnp (737s) but not like this.

Somewhere I have the old high voltage capacitors I bought to breadboard one (decades ago), but never did... similar +60V rail

BTW the NF does not wrap around the input devices so this is a variant on open loop soundcraft preamp. Not uber-linear performance.

note: they could cap couple some NF up to the device emitters and make this right, or not. They have both polarities available in the dual op amp.

JR

It suffers from the same lack of input stage feedback as Doug's Yamaha preamp.

But the input-capacitorless technique looks solid.

Looks like a job for the ZTX951.

[JR.]

In my old late night doodling on this subject I used pnp (737s) but not like this.

Somewhere I have the old high voltage capacitors I bought to breadboard one (decades ago), but never did... similar +60V rail

BTW the NF does not wrap around the input devices so this is a variant on open loop soundcraft preamp. Not uber-linear performance.

note: they could cap couple some NF up to the device emitters and make this right, or not. They have both polarities available in the dual op amp.

JR

It suffers from the same lack of input stage feedback as Doug's Yamaha preamp.

But the input-capacitorless technique looks solid.

Looks like a job for the ZTX951.

I'd have to look at (find) my old notebooks, but imagine two current sources instead of one, and decent sized electrolytic caps connecting to the high sides of R10 and R11. The caps would also provide a first order filtering of noise from the current sources.

Further those electrolytic caps could be inside the feedback loop if actual outputs are taken from junction of caps and now 2k feedback Rs.

Years ago I hypothesized my 3 transistor long tail pair to level shift and convert the two floating legs to a single ended ground referenced output, while a serious box might want the dual output legs again.


Again I never melted solder on this but my mental review looks good. Note: It looks like I did keep a balanced output, but level shifted and ground referenced. Damn that was over 12 years ago.

Of course there are other details to get sorted... the schematic I posted the other day is for a broadcast processor so a little dirt might have been just right for it.

JR

PS: I still like the idea of floating an A/D convertor up with the front end, but nah...

[billshurv]

Floating AD sounds good. I would buy one just to branded a heretic.

[JR.]

Floating AD sounds good. I would buy one just to branded a heretic.

The problem with this is which A/D to use... the A/D I would have used 20 years ago would be a joke today.

JR

[mediatechnology]

The Symetrix design lends itself well to a simple AC-coupled Rgain without overall NFB.

I was thinking about the Symetrix circuit and how one would add feedback to it.
Since the op amp output isn't floating up to Ve, the (current) feedback would have to be AC-coupled into the emitters.
Keeping the resulting 2 pole response from peaking might be difficult.

@JR
I do remember that drawing.
There sure are a lot of transistors in that level-shifting circuit.*

(Statement coming from a guy chastised by the engineering manager for using an 8 cent multi-sourced part in production for 30+ years. TL431 cough.)

[emrr]

Nice find!

[JR.]

The Symetrix design lends itself well to a simple AC-coupled Rgain without overall NFB.

I don't like the cap coupled gain either, while I have used it more times than not...

I was thinking about the Symetrix circuit and how one would add feedback to it.
Since the op amp output isn't floating up to Ve, the (current) feedback would have to be AC-coupled into the emitters.
Keeping the resulting 2 pole response from peaking might be difficult.

That's what I was trying to describe a couple posts ago.

If you take the actual output from the top of the caps there would not be a LF bump there, just down at the op amp.

@JR
I do remember that drawing.
There sure are a lot of transistors in that level-shifting circuit.*

no more IMO than needed to perform the task well... the low impedance buffer stages could be removed and replaced with op amps but that would involve far more devices. The all discrete nature of this might appeal so some market segments, who buy with their eyes.

I have looked at replacing my 3-put circuit with real op amps and it would take at least two, and they would need to be biased up at +24V so marginal for 36v IC process. The bias point could be lowered with lower final output level perhaps requiring make up gain. A third opamp to single end the output could add gain (and noise), but then we would likely want a 2 legged output so even more additional ICs.

(Statement coming from a guy chastised by the engineering manager for using an 8 cent multi-sourced part in production for 30+ years. TL431 cough.)

I should not have to defend my design cred (and for the record I have been more than an engineering manager, that was just one of my several positions along the way), but I am glad you are so easily amused. If only 30 years old that part was not even in production, when I started working at Peavey . It would be far from a cheap, old, trusted part to look into alternate applications for back when I was most active in executing new circuit designs. FWIW I also had to deal with a number of specialty ICs that went EOL (like dedicated meter ICs) during my time in engineering. I'm sure they all seemed like good ideas at the time.

I am not opposed to embracing new technology and used my share of new components in my drum tuner designs (like class D chips, LDO cmos regulators, and LED drivers). Back at Peavey there was a pretty high threshold to bring in new parts without some compelling advantage. Over 15 years I brought in only a handful (like a PTC fuse for my trick boosted rail power amp to power limit the boost stage) but not many. I already had a rich inventory of components to pull from. No need to bring in a low noise 737 when there was already a 786 in the system.

As I have stated many times, use whatever floats your boat, you don't need (or apparently value) my blessing.

JR

[mediatechnology]

The earliest reference I found to the TL431 was its predecessor the TL430 in 1978. So about 41 years.
By about 1980 the TL431 was found in the Apple II power supply.

JR I do value your input. Until 2018 I had overlooked the TL431.
Glad I tried it.
In the current shunt monitor it rocks.
I think its just better to try something and piss on it if it doesn't work than just piss on it and never know.

I like the floating A/D if no analog out is needed.
Since there are two data paths that need to cross the galvanic barrier, audio and control, it would be cool if a single UART that embeds control into audio could be used so that only one bidirectional opto-coupler or data transformer is needed.
Would DANTE lend itself to that?