Superscope EC3 mod to line level

[Chrisfromthepast]

I want to modify a superscope ec-3s microphone to have a hotter output: line level, but still unbalanced.
I’ve modified a couple of these mics to work with typical microphone preamps once by impedance balancing and another time by adding a transformer output. In both cases, I just left the 1.5v battery rather than trying to use phantom power. Great sounding capsule, but its all packaged up with a fet and source resistor. I have been leaving those little capsule assemblies intact.
This time, I want to use the stock 1/8” output cable, but have the output level be much much hotter.
Im ok with burning up the AA battery relatively quickly to supply the current necessary to get the amplification. Or if 3.7v is necessary , im open to a different battery

Stock circuit looks like this:

https://stereo2go.com/forums/attachments/ec-3s-1-jpg.17082/
Highish output impedance is OK in this case.

 

[Chrisfromthepast]

If Im simulating this correctly, the stock circuit outputs about dynamic mic level signal with a reasonable input signal.

 

[Chrisfromthepast]

And If Im understanding correcly, I can get more gain from an arrangement like this:
The issue is the Jfet is a black box at this point so the resistor values may be way off.

 

[Chrisfromthepast]

Should I try to test and figure out the jfet's parameters or just go for it?

 

[Chrisfromthepast]

Ok, no free lunch, I understand.
So Im guessing the simplest way to go about this would be to leave the fet bias alone, and come up with a bc550 gain stage, and a bc550 emitter follower to give the output a chance in the real world.
Im choosing that bjt because I have a hundred of them, Im open to any ideas that will fit inside the body of this mic.

 

[RuudNL]

Simple answer: there is no way to get a 'line level output' with a supply voltage of 1.5 V!

 

[Chrisfromthepast]

cool, thanks for hitting me with some sense.
should I stack a few 3.7v cells?

 

[RuudNL]

It is just simple math: say you want 'real' line level, like 0,775 V.
That means a peak-to-peak voltage of 2.18 V,
Since it won't be possible to obtain a rail-to-rail output voltage, you have to add at least 2 Volts, so a minimal supply voltage of 4.18 V. And even then it is doubtful if you would be able to get 0.775 V. without significant distortion.

 

[Chrisfromthepast]

Makes sense!
Did you see my second simulation with 3.7v?
That low source resistor isn’t going to work in real life right? Thanks for your patience Im just a sound guy with a multimeter…
So if I traded batteries for a23, I would have room to run a couple of wires next to the battery and 12v of potential to waste.
Should I voltage divide down to 1.5v, supply this to the capsule/fet assembly to avoid the hubris of redesigning the first stage?
And then make a 2 stage amplifier out of bjt’s? Or use a bipolar op amp?
Or would the 12v bias correctly with the stock circuit and get me most of the way there with a single gain stage?

 

[Chrisfromthepast]

If anyone has a sec to look over this design, I think I can shove all of this stuff into the body.
Id be happy to remove some components if someone with a red pen can help me simplify.

 

[Gus]

You could use a good rail to rail opamp
OR
Look for Gefell circuits for ideas

 

[Chrisfromthepast]

I’ve got some ADA4001 here. Would you do away with the original fet? I’ve never opened the lil capsule assembly, but its probably not too bad.

 

[abbey road d enfer]

The danger with high gain is that the risk of clipping is very high. That's why variable-gain exists, in order to adapt gain to the spl level. Alternatively, a compressor could be used.

 

[Chrisfromthepast]

I guess its worth getting into the specifics. I have this little toy made by critter & guitarri called the 5 moons. Its part 5 track recorder, part looper, and functions as a little scratch pad for ideas without using a DAW.
I’ve got a condenser mic going into a preamp and a compressor currently, and all of that stuff is just huge compared to this little toy.
I don’t touch the gain much, many of the melodica/percussion/vocal/toy piano parts are roughly the same spl range. I think I could get away with a high gain/low gain switch or just a trimpot hidden inside.

 

[Chrisfromthepast]

Ok, here’s another attempt, using a dual op amp package and lithium cell, no gain control yet.

 

[Khron]

That looks awfully overcomplicated for what it needs / wants to be.

A simple "opamp Alice" circuit should be just fine for this, but... Is that JFET part of the capsule or not? And for such a circuit, you'll want a very low-power opamp, ideally with rail-to-rail output at least, if not input as well.

I can see why you'd want the top opamp's non-inverting input biased at half-supply, but why would you power the JFET from the half-supply? And why "only" 500-ohm resistors in the rail divider? Opamps have input impedances in the megohms, at least. Unless you wanna burn off the energy in your lithium cell in half (?) the time...

 

[abbey road d enfer]

Ok, here’s another attempt, using a dual op amp package and lithium cell, no gain control yet.

The arrangement around the FET is not correct. The drain resistor should be much higher. Then you need to provide either an adjustment of the DC op point for the opamp or a way to DC-decouple it. As it is, it's very likely that the output is stucked to V+ or ground.
The 500 resistors should be 500k. 500 would draw 3.7 mA pernanently, which is not needed.
EDIT: Ah, Khron beat me to it!
And I maintain that using high gain with a low headroom circuit is simply wrong.

 

[Chrisfromthepast]

Thanks for your patience.
500K, divider, yeah, totally. The low power rail to rail op amp I have on hand is ADA4001. Since its a Jfet front end, I could couple to the capsule and dispense the silly jfet arrangement.
I was previously attempting to keep this capsule assembly together.

https://taperssection.com/index.php?topic=185994.0Since the original battery was 1.5V. I have, against all good sense, attempted to present 1.5(ish) volts to the original capsule/fet/300M resistor arrangement.
Nobody seems to like that idea, I'll give it up.
Im not married to any topology. the solution that Critter and Guitarri came up with has a high/low gain switch, and a trim pot, and is based on 3.7v battery. So, those can be the design constraints.
I'll redraw.

 

[Chrisfromthepast]

As a sidebar, how does the original circuit work without a source resistor. is it somehow the 1k5?

 

[RuudNL]

R2 is the source resistor, 3.3K.