AKG genuine CK12 Brass capsule: building DIY mic with a single Jfet-transistor

Thread Starter

Schertler

Joined Dec 30, 2020
6
I have got discovered vintage AKG CK12 Brass capsule at home.

I would like to build a mic with this capsule and use a standard one-transistor Jfet circuit of ECM electret mics (common source either
source follower)...

Has anyone ever tried to wire such a large capsule with a single Jfet transistor ? Do you think this might work as a decent condenser mic ??

Thank you for the comments!
 

sparky 1

Joined Nov 3, 2018
543
An authentic build would try to recreate what Telefunken and others who were used in some high quality microphones.
What are the audio attributes of that component and what is the best way to bring out those effects? Which end use will it be used for?
Some microphone collectors might own a set of C414 each one having different characteristics some more valuable or less common.
I am not sure how to find it's frequency response. There are a lot of microphone tests that might be done or datasheet somewhere?
The scope of that project would take a considerable amount research and skill to replicate an authentic instrument to match what you have.

" The original capsule in the C414 EB was the "Brass" CK12, as used in the AKG C 12 and Telefunken ELA M 251 microphones. This complex capsule used many parts and was exceedingly difficult to assemble and tension/tune for any sort of consistency, and was later replaced by the less-desirable (but more cost-effective) "Nylon" CK12, which used a nylon ring to secure the diaphragm to the capsule's backplate. " The nylon variety was known for female vocals which is now done mostly with digital. The description describes inner workings as extremely delicate.

The AKG C414 EB w/ Brass CK12 sells used having price between $1,000.00 and $4800.00 USD which is the what it is worth to an artist or collector.
This market has been impacted by large supply of musical instruments in pawn shops and lower online consumer confidence.
 
Last edited:

Thread Starter

Schertler

Joined Dec 30, 2020
6
Hi All,
Thanks for your feedback!
Correct, CK12 capsule does require polarizing voltage which is set to 110V by AKG, but I do have 48V powering for this task, hope it should work well with 48V as I only need around 1-2mV output from the capsule....... I have a decent mic preamp with 1mV input!

The question is about the method of using 1-2 transistors with the capsule instead of standard huge internal electronics of the mic - tube and transformer. Actually, this approach is described by Robert Schulein: http://www.sdiy.org/oid/ldc/RSchulein.pdf

Hope the capsule can perform at its best with that basic 1-2 transistor combination....

LDC 34mm Capsule _ 2.jpg
LDC 34mm Capsule _ 5.jpg
LDC 34mm Capsule _ 8.jpg
LDC 34mm Capsule _ 11.jpg
 
Given the vintage nature of the capsule and you don't know if it actually works, they are extremely fragile- the project has a lot of risk.
A condensor mic requires high voltage for polarizing the diaphragm.
Since the AKG C12 (and C12a) is a vacuum tube condensor mic, and a look at the schematic shows it runs off 120VDC, they just used a resistor divider to get 60VDC.
It is typical to use a one-transistor Hartley oscillator to generate around 60VDC polarizing voltage, like the usual Schoeps JFET+two transistor circuit.

I would sell the capsule for the vintage part it is. You can buy new large capsules for DIY condensor mics and they are easy to get up and running. There is an entire craft devoted to making mics, housings, etc. I am working on a large capsule mic in the chinese BM800 housing, it's a popular project.

C12-schematic.jpg
 

Thread Starter

Schertler

Joined Dec 30, 2020
6
Well, polarization voltage is not the issue, if the capsule functions properly but shows any drawback due to 48V polarization, then I will just set it up to a maximum - 110V.

Housing and all the mounting parts are all made DIY, however they are fully reliable and look good too.... the mic will have an original seated design, its metal housing will seat on to small rubber buttons...... but this is an experiment, if the capsule is absolutely fine then I will build it into some standard DIY condenser mic body!

High-nominal resistor: do you think that there is a resistance limit of that resistor? I heard that the higher the resistor, the better! I
am thinking of putting up to 10G in the polarizing chain.... may this affect the sound quality ??
 

andrewmm

Joined Feb 25, 2011
1,468
WOW
Now that is a famous mic,
Just out of interest, you do realise those capsules, without the rest of the mic, sell for many hundreds of dollars US.

OK, I'm an AKG lover, D190 and D220 were the work horses when I was at a well known UK based radio and TV company,
The C414 which used the CK12 was used for the "expensive" artists..

back to OP point

I'd seriously suggest selling it, others in the audio world I am certain would crave it, I have goose bumps just typing this.

I would be sad, if your experiments damaged it,
so as forest Gump famously said,
"Thats all I have to say on the matter"
 
Thermal noise of a resistor goes up with its value, so a larger polarization resistor means more noise. It also means any leakage currents due to moisture or surfaces become a big issue. Old mics have teflon standoffs, pcb slots. I use 1GIG with JFET gate lead in the air. Best I can find are through-hole 1/4W TE RGP0207CHK1G0 or high voltage Murata parts.
You will find some science and a lot of superstition in the DIY microphone world.
The cheap chinese large capsules are copies of the U47 and others, they do OK at $20-$100 and it would be a good start. Hardest part is the mic saddle and housing. I'm doing a two board design to fit in the BM800, simple Schoeps. The stock BM800 electret condenser mics are terribly noisy and sound terrible, the circuit is a disaster so I'm ripping it up.
 

Thread Starter

Schertler

Joined Dec 30, 2020
6
The noise: if you simulate this circuit in LTSpice, you will see a totally different situation, the higher the resistor, the lower the noise and the wider the frequency response!!

I was told that a golden medium must be found with that resistor as increasing resistance decreases distortion threshold !!

Housing/mounting: when I complete with all, I will put some images and audio-samples of the mic, so if the capsule is OK then the DIY mic would be just fine......
 
Maybe post the LTSpice sim, it might be incorrect or it's showing the tradeoff of resistor noise verses the JFET noise.
JFETS from Linear Systems are renowned for lowest noise and distortion, as other manufacturers have pretty much stopped making anything decent for audio (non-switching) applications as nobody likes them in production due to their huge variance.
For electret condenser mics, 2SK596 JFET is the norm (very low IDSS) but has a built in resistor and protective clamp diode.
 

Thread Starter

Schertler

Joined Dec 30, 2020
6
Well, got them all......
The first version is 2SK596 but LTSpice does not list this Jfet in its library, so I put LSK389 there.
Second version goes with Russian KP303 transistor into a standard condenser mic input.
Just putting the figures in one place:

2SK596 with 300Meg resistor:
AC Analysis falls from 3.6dB to 4.2dB = 0.6dB
Noise: 3.45µV
THD: 0.1% (2SK596 should be far better with THD!)

2SK596 with 3G resistor:
AC Analysis falls from 3.592dB to 3.598dB = 0.002dB
Noise: 1.2µV
THD: 0.1% (2SK596 should be far better with THD!)


KP303 with 300Meg resistor:
AC Analysis falls from 1.8dB to 5.1dB = 3.3dB
Noise: 5.5µV
THD: 0.06%

KP303 with 3G resistor:
AC Analysis falls from 1.805dB to 1.850dB = 0.04dB
Noise: 2.1µV
THD: 0.1%
 

Attachments

Top