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

[micolas]

If you swap BOTH pairs of wires, it doesn't change anything.

You are absolutely right, the circuit works in both versions.
But...
In many years of working with transformers I have noticed some rather subtle details.
If I connect the beginning of the primary coil to the hot part of the audio signal, and the end of the primary coil to GND, the coil will behave as a kind of electromagnetic screen (especially the last layers of the winding) and we will have less hum. The secondary having low impedance will be less sensitive. That's why I chose the second variant of connecting the transformer.
Maybe that's why in many schemes we see the beginning of the coil marked, connected to the hot part of the signal.

 

[micolas]

OK, this is how I understand it so far, hopefully it's correct now:

View attachment 144758

Put C6=22uF though.
nte10-3 sounds a bit thin, has small magnetic core
I even increased C11 to 1.5...2.2uF.

 

[Jakub W]

Thanks a lot @micolas
That's the DCDC circuit I am most familiar with and would like to use:

 

[micolas]

View attachment 144756

I want to use a UTM0547 as OPT. I have used other ratios with a similat design in the past. J2 can be even be an BJT, but I'm just using parts I have laying around. Disregard R17, I forgot to pull it out, lol.

Many are of the opinion that J2 would be better off being a BJT.
But it's everyone's choice.
Is the schematic your design?

 

[Khron]

That's the DCDC circuit I am most familiar with and would like to use:

Are you 100% sure that schematic is drawn correctly?

 

[micolas]

Thanks a lot @micolas
That's the DCDC circuit I am most familiar with and would like to use:
View attachment 144762

DC/DC consumes very little current. C1=330uF is huge. 47uF is enough.
You will need to feed the DC/DC through a resistor so that you give the Zener 9.1 just a few extra volts to effectively stabilize, not "stealing" more current from the audio circuit.

 

[micolas]

Are you 100% sure that schematic is drawn correctly?

@Jakub W
Here is a proven scheme that I use in most situations.
At such low currents, I don't even use a zener diode, which can be noisy. The electrolytic capacitor behaves as a pseudo current reservoir.

 

[Drosselmeier]

Thanks a lot @micolas
That's the DCDC circuit I am most familiar with and would like to use:
View attachment 144762

To reduce the current consumption I've put a 1.6 meg resistor R9 for the 2n5551 transistor.
I selected it based on min current and max output voltage.

 

[micolas]

To reduce the current consumption I put a 1.6 meg resistor R9 for the 2n5551 transistor.
I selected it based on min current and max output voltage.

I often used between 1.2 ...1.5Mohm and obtained the higher polarization voltage, 90...100v
In addition, I made a tighter electromagnetic coupling between the inductors.

 

[analag]

Is the schematic your design?

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 Make It.mp3
some grime to fit the track.

 

[micolas]

Yes. I built the first version of it with BF245a in J1 and J2 position.

I like the design, and I thought about something like that. Many years ago I was building guitar pedals with multiple jFETs, they behave closer to a tube than a BJT.
I'm really following @thor.zmt 's projects with MosFets very carefully. They have great potential.

 

[Jakub W]

@Jakub W
Here is a proven scheme that I use in most situations.
At such low currents, I don't even use a zener diode, which can be noisy. The electrolytic capacitor behaves as a pseudo current reservoir.

Thanks a lot guys, I thing I got the final sketch for the PCB design fo the circuit, IF I understood and drafted everything correctly. Any further hints will be greatly appreciated.

 

[micolas]

Thanks a lot guys, I thing I got the final sketch for the PCB design fo the circuit, IF I understood and drafted everything correctly. Any further hints will be greatly appreciated.

Between +C3 (Audio) and +C1 (DC/DC) you must insert a resistor of several tens of Kohm. To begin with, you put a semi-adjustable R* of 47Kohm set to the maximum value.
Connect the DMM to the DCDC output, [+D3/C13] and decrease the value of R* until you measure approximately 62V...64v. Less will fit on the capsule. At the end, R* will be measured and replaced with a resistor of the same value

 

[Jakub W]

Between +C3 (Audio) and +C1 (DC/DC) you must insert a resistor of several tens of Kohm. To begin with, you put a semi-adjustable R* of 47Kohm set to the maximum value.
Connect the DMM to the DCDC output, [+D3/C13] and decrease the value of R* until you measure approximately 62V...64v. Less will fit on the capsule. At the end, R* will be measured and replaced with a resistor of the same value

Thanks again. So this seems to be the final sketch? Self doubt is growing in my veins .

Edit. Corrected thanks to @Khron

 

[ricardo]

The other day I was thinking about the design of the ELA M251, and how it differed from the C12 in ways that were intended to make it more Neumann-like ... more EVIL heresies ...

evil, Evil, EVIL

Us mike makers that come from da Holy AKG line spurn such EVIL Heresies.

Bernhard Weingartner would be turning in his grave if he wasn't still around

 

[Khron]

IF I understood and drafted everything correctly

How are BOTH inductors in the oscillator now connected to ground?

Please recreate that circuit in LTspice, do a transient simulation and see if and how it works...

 

[OneRoomStudio]

evil, Evil, EVIL

Us mike makers that come from da Holy AKG line spurn such EVIL Heresies.

Bernhard Weingartner would be turning in his grave if he wasn't still around

In fairness, it was really Telefunken that wanted it to be "Neumann-like." The SM204/23 was already most of what became the ELA M251 E, so AKG had already done most of the design work before Telefunken ever came knocking.

AKG can remain Holy

 

[micolas]

@Jakub W

R3 (audio): a semi-adjustable resistor R** also of 47Kohm set to the maximum value is initially used. With Zener 33v NOT CONNECTED, adjust R** so that at +C4 you measure 36...37v.
Now connect the zener diode and you should measure at +C4 around 33v (+/- depending on the zener specs in the datasheet)
At +C5 you should measure about 29v +/-
First adjust the Bias (from TP1 25Kohm) to measure at the jFET Drain about 10.5 ... 11.4v. The microphone must work.
(Later you will use a more appropriate method for Bias)
During these adjustments, fine-tune R*, R** and TP1 to obtain the previously mentioned voltages.

 

[micolas]

How are BOTH inductors in the oscillator now connected to ground?

Please recreate that circuit in LTspice, do a transient simulation and see if and how it works...

@Jakub W

@Khron has sharp eyes

 

[Jakub W]

@Jakub W

@Khron has sharp eyes

I know and updated the attachment.