A mic that never was (a thought-experiment circuit for the RK-12 capsule)

[KrIVIUM2323]

Yes. I built the first version of it with BF245a in J1 and J2 position.
Here is a rap song with Rev 1 of that circuit.

Both the mic and mix sound great. The track is nice too. Who is the artist?
Off topic:
Analag, do you plan to share the different circuits you shown in your The Mic thread?

 

[analag]

Both the mic and mix sound great.
Off topic:
Analag, do you plan to share the different circuits you shown in your The Mic thread?

Thanks. I'll share them. In that song I'm using a DIY M1 type pre. Discrete opamp with double current mirror or Widlar CM and I did the bootstrapped CS for better performance than the diode biased thing. It's a modified 990 then into my optical comp. I did a total redesign on that one. It now uses 6N23p. I think it sounds better.

 

[analag]

That's my circuit mocked up with all the components that will be going on there. The other board will handle the transformer and the pair of 2.21K maybe some filtering. As of right now I am waiting on a capsule and a transformer. It's a long wait but I'm hanging in there!

 

[micolas]

... It's a modified 990 then into my optical comp. I did a total redesign on that one. It now uses 6N23p. I think it sounds better.

I put 6N23P-EV (6H23P-EB) NOS Soviet military grade in 2 microphones model Rode K2. It is an excellent tube!
Sounds great, much better than the stock tube.
Edit:Sorry for this off topic remark

 

[micolas]

Yes. I built the first version of it with BF245a in J1 and J2 position.
Here is a rap song with Rev 1 of that circuit. Mind you I mixing in
View attachment 144766
some grime to fit the track.

I quickly isolated the voice to hear it better, although it was processed.
It sounds very good
View attachment Analag Rap jFETs v1- voice.mp3

 

[mhelin]

I know and updated the attachment.

Is there any way to tell in schematic that L1 and L2 should be placed in close proximity with each other in certain way? Also can that circuit be simulated in simulator (LTSpice, TINA-TI, name it) any other way than using a transformer?

 

[micolas]

Is there any way to tell in schematic that L1 and L2 should be placed in close proximity with each other in certain way? Also can that circuit be simulated in simulator (LTSpice, TINA-TI, name it) any other way than using a transformer?

To maximize the voltage delivered by the oscillator (DC/DC) I put L1 and L2 very close physically, for maximum electromagnetic coupling. Attention The beginnings of the inductors must be respected.Otherwise the oscillator may not work.
There were cases when the marking with the color code was done wrongly by the manufacturer and the oscillator did not work. I have encountered this situation including with inductors purchased from Mouser.
I think that @RuudNL also encountered something like this.

 

[Khron]

Is there any way to tell in schematic that L1 and L2 should be placed in close proximity with each other in certain way? Also can that circuit be simulated in simulator (LTSpice, TINA-TI, name it) any other way than using a transformer?

I've seen all sorts of layouts in retail mics - some with the inductors parallel to each other and hot-glued in position; some with the inductors at 90deg to each other (minimizing inductive coupling).

The better coupling, the (possibly) higher voltage you can get at the output (for a given input), but "nothing is impossible", as it turns out...

 

[mhelin]

I've seen all sorts of layouts in retail mics - some with the inductors parallel to each other and hot-glued in position; some with the inductors at 90deg to each other (minimizing inductive coupling).

The better coupling, the (possibly) higher voltage you can get at the output (for a given input), but "nothing is impossible", as it turns out...

Yes, guess might be worth simulating. Found this post which tells about "SPICE statement K1 L1 L2 1" which makes a transformer from two coils, guess the last parameter in this case can be varied a lot:
https://groupdiy.com/threads/error-in-my-1073-simulation.89285/#post-1177658

 

[micolas]

To maximize the voltage delivered by the oscillator (DC/DC) I put L1 and L2 very close physically, for maximum electromagnetic coupling. Attention The beginnings of the inductors must be respected.Otherwise the oscillator may not work.
There were cases when the marking with the color code was done wrongly by the manufacturer and the oscillator did not work. I have encountered this situation including with inductors purchased from Mouser.
I think that @RuudNL also encountered something like this.

I marked the start of the inductors on my schematic. By very tight coupling between L1 and L2 (they physically touch) I obtained (at a DCDC supply with 8.2v) an output voltage after the voltage doubler with diodes a little over 110v.
It is possible that the hFE of the BJT also counts. The resistance was 2.2Mohm.
Out of curiosity, I applied this high polarization voltage to many capsules, none of them collapsed at rest, with no applied sound.

 

[RuudNL]

As @micolas already mentioned, I have experienced the problem with inductors not wound in the same 'direction', even from the same brand. My experience is that the closer the inductors are (in parallel) together, the higher the output voltage. In fact later I used a test setup to check if the combination of inductors was suitable for a DC converter. With 'good' inductors, the output voltage rises if you move the coils closer together. If the voltage gets lower when moving the coils together, it will be difficult to reacht the target voltage, or the oscillator won't start at all. I suppose that with the inductors placed at an angle of 90 degrees from each other this will be less critical, but I assume that the output voltage in this case will be lower than with the two coils in parallel.

​

 

[micolas]

I marked the start of the inductors on my schematic. By very tight coupling between L1 and L2 (they physically touch) I obtained (at a DCDC supply with 8.2v) an output voltage after the voltage doubler with diodes a little over 110v.
It is possible that the hFE of the BJT also counts. The resistance was 2.2Mohm.
Out of curiosity, I applied this high polarization voltage to many capsules, none of them collapsed at rest, with no applied sound.

In practice, I usually apply a maximum polarization voltage of 80v, if the input jFET has a high Idss, the circuit has low gain, and the headroom/spl is not important (quiet sources)

 

[micolas]

As @micolas already mentioned, I have experienced the problem with inductors not wound in the same 'direction', even from the same brand. My experience is that the closer the inductors are (in parallel) together, the higher the output voltage. In fact later I used a test setup to check if the combination of inductors was suitable for a DC converter. With 'good' inductors, the output voltage rises if you move the coils closer together. If the voltage gets lower when moving the coils together, it will be difficult to reacht the target voltage, or the oscillator won't start at all. I suppose that with the inductors placed at an angle of 90 degrees from each other this will be less critical, but I assume that the output voltage in this case will be lower than with the two coils in parallel.

​

This is exactly what I tried to describe with my vocabulary of a non-native English speaker

 

[micolas]

I think I mentioned before, the high voltage test champion for modern capsule polarization was @kingkorg with a Rode NT1kit Black capsule, taking the SNR to the extreme and getting the quietest microphone in the galaxy


That's the true DIY spirit!

 

[micolas]

Here you can see how tight the inductive coupling is between L1 and L2.
And the physical proximity between the other components favors oscillation and output voltage.
My favorite method,
"3D-point-to-point" is perfect for DCDC, in addition it allows miniaturization and use in the most varied shapes and sizes of microphone bodies. And it's very cheap, I don't need to build a PCB.