60's Tube Condenser Microphone

[whoops]

Hello everyone,this is not a guitar amplifier it's a Tube microphone.
I hope it's fine to post here as it's a simple circuit and your knowledge is invaluable.This a 60's Tube condenser microphone it uses ECC83 (12AX7), it was made in East Germany in the 60's by Joachim Wetzel.He produced this kind of handemade looking microphones.



There's not much info on the web on these microphones, so I had to trace the circuit myself and draw a schematic from the PSU and from the Mic.The Power transformer was shot (dont remember now if it was the primaries or the secondaries), it had one shot selinium diode.
The power transformer was rewinded, the selinum diode was replaced for 1N4007, all the electrolytic caps were replaced.All the remaining components and connections were measured and measure fine.A 3 prong IEC connector was installed with the Earth lug connected to the Chassis.
here is the PSU schemactic:


here is the mic circuit schematic:


These microphones although if even looking similar were produced with diferent components and revisions, so the other schematic I found on the web is slightly different than my mic. Probably the diferencies in circuits are improved revisions for better performance or due to component availability, so there's no definite schematic.
I would like to ask you before I plug the mic and PSU to test it out, if by looking to the schematics you can see something strange, or have any advices for improvments over the old stock circuit.
Thank you so much

[whoops]

Here is the other schematic I found on the web, there's somo differencies to my circuit, the 6.3v on the heaters might be a mistake :

[whoops]

Voltages with the Mic connected to PSU are:
B+ 218V (after settling down)
Heater 12.8V

[jjasilli]

Some questions.


Heater supply
Schematics say 6.3V; but you say 12.8V???
Might be less noisy to ground reference the heater supply near the tranny; and float the rear of the diode bridge.  Maybe also reference the heaters to a +voltage.


Power Supply
Why is the B+ only half-wave rectified; and with so little filtering?
The 50n or 22n series cap at the end of the B+ supply should probably be a shunt cap; or it will block DC. 

[PRR]

> Might be less noisy to ground reference the heater supply near the tranny

No; reference close to the tube.

> Why is the B+ only half-wave rectified

Because the current demand is VERY small, and because that diode was *expensive*. It's fine.

> and with so little filtering?

Looks like adequate filtering for the very small demand.

> The 50n or 22n series cap at the end of the B+ supply should probably be a shunt cap; or it will block DC.

That goes to the Audio Output! Power and signal run on the one wire. DC comes in from left. Audio goes out to right.

[whoops]

Some questions.


Heater supply
Schematics say 6.3V; but you say 12.8V???
Might be less noisy to ground reference the heater supply near the tranny; and float the rear of the diode bridge.  Maybe also reference the heaters to a +voltage.

Hi, thanks for your reply

The schematic I drew doesn't say 6.3v, the xaudia schematic does, that might be a mistake.The Heaters in the tube are wired for 12V, so I think the 12.8V are fine

[whoops]

Power Supply
Why is the B+ only half-wave rectified; and with so little filtering?
The 50n or 22n series cap at the end of the B+ supply should probably be a shunt cap; or it will block DC.

I think the use of 8uf electrolytic capacitors were probably for cost and size reasons in the 60s. And like PRR said the current demand is small on the B+ rail.
Anyways I re-caped the unit and I used 22uf for the first cap and 22uf instead of the 2x 8uf caps.I used those values because it was what I had in stock for that voltage rating.

[jjasilli]

Is it working?  How does it sound?


I checked out some websites and most reviews like the sound, but thought it was a little bit noisy.  That's why I asked about nose related questions.

[whoops]

Yes, I plugged it in and it seens to work properly.
The sound seems fine, but maybe a bit too much high end, I have to compare A and B with another large diaphgram condenser to know for sure.
There's some noise yes, more like buzz, and not so much hum,I think that might be improved with replacing the cable (PSU to Mic) with a Shielded cable, as the stock cable has 4 conductors but no shielding.Also as the PSU box (that also receives and outputs the audio signal) is mainly plastica box with a metal front plate, I was thinking of using copper foil to shield the inside of the box.Maybe that improves the noise.
Another thing that was suggested to me , was taking the Audio output signal from Pin3 (the second 12AX7 stage Cathode), to have a lower impedance, and then use a transformer to balance the signal.The instead of having and unbalanced Hi-z signal I would have a Low-Z balanced signal.
What do you guys think of these ideas?
Thanks

[jjasilli]

I agree with shielded cable.  Even the the DC filament supply is prone to pick up ambient noise - it lacks the noise cancelling of a balanced AC supply.  So a long run of unshielded cable can act as an antenna in both the signal and power wires.


The mic's output is reportedly unbalanced Hi-Z.  Some reviewers prefer to add a mic tranny to convert to balanced Lo-Z.  Presumably, this would reduce noise.

[whoops]

Thanks I will buy shielded wire with 4 conductors and install it.
I would like to add an output transformer,
I just don't know what transformer would be suitable for this aplication

[jjasilli]

Try Jensen transformers website.  Resources. Schemtics.  Might help.

[DummyLoad]

cinemag


http://cinemag.biz/mic_output/mic_output.php   
 
10K-15K to 600R or 200R or 100R - your choice - you'll need to chose the power level for the application. you don't need CT since you're not running phantom power.

--pete

[PRR]

> more like buzz, and not so much hum,I think that might be improved with...

Move it around.

If it is inside the box, it won't change. If it is the cord, cord location relative to other gear will change the noise. If it is the plastic box, box location re: other gear will change the noise.

[whoops]

cinemag


http://cinemag.biz/mic_output/mic_output.php   
 
10K-15K to 600R or 200R or 100R - your choice - you'll need to chose the power level for the application. you don't need CT since you're not running phantom power.

--pete

thank you so much, Cinemag has a really good offer, I will go with one of their transformers.

At the moment the output seems to be unbalanced and 22K output impedance,do you guys think I should take the Audio output for the transformer from the second cathode?

[whoops]

> more like buzz, and not so much hum,I think that might be improved with...

Move it around.

If it is inside the box, it won't change. If it is the cord, cord location relative to other gear will change the noise. If it is the plastic box, box location re: other gear will change the noise.

Thanks PRR,I will do as you suggested.
How about the differencies between my circuit and Xaudia Schematic?
In my circuit the 22.5M resistor goes to ground, in Xaudia it goes to the cathode of the first tube stageAlso there's a Cap to ground on the second stage cathode (Pin3) that I dont have in my mic.
Should I include any of these in my circuit?

[jjasilli]

Actually it's not clear to me from your schematics where the audio output is, Except for your Reply # 14.  I'm not sure that the 2nd stage cathode tap is really Lo Z.  It's lower than the plate.  But I think it's still about 33K.  It would be Lo-Z if it were a cathode follower, but it isn't.  Also, it's not clear what the cathode voltage is.

[whoops]

Actually it's not clear to me from your schematics where the audio output is, Except for your Reply # 14.  I'm not sure that the 2nd stage cathode tap is really Lo Z.  It's lower than the plate.  But I think it's still about 33K.  It would be Lo-Z if it were a cathode follower, but it isn't.  Also, it's not clear what the cathode voltage is.

Please have a look again.Besides reply 14 , the audio output is on the B+.PRR also explained that before when replying to your question:

> The 50n or 22n series cap at the end of the B+ supply should probably be a shunt cap; or it will block DC.

That goes to the Audio Output! Power and signal run on the one wire. DC comes in from left. Audio goes out to right.

it's also labelled in my schematic

It would be Lo-Z if it were a cathode follower, but it isn't.

It seems like a Cathode follower to me

[jjasilli]

1.  Audio output cannot possibly be coming from the B+ rail.  Where is Din pin5?  At the very least, the schematic is unclear.


2.  The second tube stage is above my pay grade.  It appears to be a 2nd gain stage, cascaded from the 1st gain stage; it is resistive coupled with a coupling (blocking) cap between the plate of the prior stage and the signal grid of the 2nd stage.  Maybe it is an AC coupled cathode follower.  In either case, if you tap signal off this cathode: i) I don't think you can use a cathode bypass cap, without splitting the cathode resistance and tapping in the right place; and ii) I don't think that this cathode's output is Lo-Z, i.e 600 Ohms or less.

Using schoolyard logic, if the cathode is the usual signal output for this mic, then we already know it's Hi-Z.

[whoops]

1.  Audio output cannot possibly be coming from the B+ rail.  Where is Din pin5?  At the very least, the schematic is unclear.

Letter "B" on the PSU schematic connects to letter "B" on the microphone schematic.
That takes the B+ DC supply to the mic, and brings the Audio (AC) Back to the PSU. The 22nf cap blocks DC so only the Audio (AC) goes to the Output connector.


I'm really really sorry but I can't see how it can be clearer that this, or clearer than what PRR explained


Thanks

[jjasilli]

Ok, thanks, now I understand the B+ connection, and audio output.  The original design seems to be calling for 2 cascading gain stages.  This suggests that the voltage output of the condenser is very low.  Maybe near the noise floor of the tube.  This in turn suggests that noise is inherent to the design; unless you go radical, and use SS as a 1st gain stage to lower the noise floor.


Still don't know if tapping off the 2nd stage cathode will get you the right impedance and voltage.  Since Dummyload knows what mic tranny to use, that's what I would do.

[sluckey]

There is obviously something missing in the schematics!

[PRR]

It all looks clear to me. 30 years later I re-invented this trick of sending power down and signal back on the same wire. For the low levels involved it works great. (I was getting 7V rms clean, from SS and a 30V supply.)

I have no idea why you think you need to change stuff. Both original variations work with very little difference. Taking signal from second cathode looks like a loser to me: less signal and more wiring out of the microphone. The stock output can probably go through short wire to an INST input on your mixer/interface, same as a guitar. A DI or transformer can pull a longer line, but how big is your studio?

Fix any broken pieces, use it.

[PRR]

OK, here's a re-draw for understanding.

Cascade 12AX7. Unusual: RP for V1 goes to V2 Plate. This is NFB and can be found in other schemes. To work well, it either needs a beefy V2 or very small signals. Here V2 runs over 2X the current of V1, and signals are small. The closed loop gain is around 120, compare to ~~2500 for cascade 12AX7, so there is Significant NFB. The output impedance is a low 6K, but it will not *drive* 6K without high THD, keep loading 50K or more; 20K if color is tolerable.

Taking the "22.5M" bias divider to common or to V1 cathode is moot. Perhaps bias is insignificantly more stable to-cathode, but the difference must be tiny. I would suspect it was just more convenient.

My simulation says the loaded voltage before the 45K in the power supply is near 70V, quite low. However the "-2.6V" on a grid is dubious, so I dunno how far to trust these notes.

[whoops]

The stock output can probably go through short wire to an INST input on your mixer/interface, same as a guitar. A DI or transformer can pull a longer line, but how big is your studio?

Thanks for all your help PRR.
I work in a lot of diferent studios and places, most of the times you just have tie lines in the Live room, I can use DI's for sure but its much more convenient to me and for my uses to have already a Balanced Low-Z Microphone signal output, so I will add a transformer and an XLR output, that extra cost is fine by me.


Taking the output signal from the cathode of the second half was a suggestion from Ian Bell a fellow member from GroupDiy.

Here is what Ian said:
"I don't understand why the audio out is taken from the B+ terminal. This is an attenuated version of the signal at the anode of the first half of the ECC83 and it looks like the second half of the ECC83 is not used at all although it is wired as a cathode follower. Perhaps the output was mean to be taken from the cathode of the second half of the ECC83 via a capacitor and/or transformer?Cheers
Ian"

"I recommend you get it working with the Hi-z output first. The you could try connecting  the output cap to the second cathode which will give you a low impedance output."

here is the thread at GroupDIY:
https://groupdiy.com/index.php?topic=28914.0


Thanks PRR

[jjasilli]

I think Ian had the same confusion I did.  It seems that the 50K R on the B+ rail is in series with the 250K R to the first stage plate for a total of more than 300K anode resistance to the 1st stage. That same 50K R is also serving as the plate resistor to the second stage.  Because it's serving two anodes, its effective resistance is higher than 50K -- somewhere between 50K -to- 100K, I think, for the 2nd stage.  The 22nF coupling cap > audio output, is located between the 2nd stage plate and its 50K plate resistor.  Everything is as it should be for consecutive gain stages, though it's hard to see that as drawn.

[Toxophilite]

Out of curiosity have you tried any other of the 12A type tube variants in the microphone. Particularly a 12AT7 or a 12AY7.

[whoops]

Out of curiosity have you tried any other of the 12A type tube variants in the microphone. Particularly a 12AT7 or a 12AY7.

Hi Toxophilite,no I haven't changed the stock circuit.
I had a failled power transformer and 1 selenium diode.So the transformer was rewinded, and the diode replaced.Besisdes that I just did a re-cap of the electrolytic capacitors, some of them I installed an higher value than original because it was what I had in stock for those voltage ratings (8uf for 22uf on the filtering, and 22uf instead of 8uf+8uf)
I just want to be able to redude the noite a little bit, probably shielding cable will help,and also have a balanced low-z mic output instead of the unbalanced Hi-Z ouput.
thanks