AKG C12 A / B Version Build

[tubestation]

Hello Spence,
The Frontdiagramm goes go Ground, ( 0 Volt )
I think You can Test the Mic without the Reardiagramm (Cardioid  Mode. )
Lothar.

 

[Spencerleehorton]

The trouble I have for trying to test this with a capsule is that I only have a spare capsule that is centre terminated.
It also only has one side, so can I test just with polar pattern set to cardioid and only front capsule connected?

 

[Spencerleehorton]

The main problem I'm having now is when I try and connect a 18k to the anode and cathode the B+ is 206v.
I have checked the pin config on the socket for the nuvistor and the only thing I could point out is that the H+ and H- pins could be the wrong way round?
I thought I could have these pins either way?
I am puzzled though, when using load resistors 45ohm and 47k I get correct voltage, but when using 18k as tube I get wrong voltage?
I will double check continuity of socket pins just to make sure to eliminate any problem with the socket.

Any other ideas?

 

[Spencerleehorton]

I've put the chassis of the socket down to gnd, could this be the problem?

Regards

Spence.

 

[Spencerleehorton]

Well I spotted a mistake and have corrected that on the pcb, but have given it a good look now and I'm fairly confident that it is right now!
When I insert a 18k now in anode to cathode I'm nearly getting correct voltages although they are a tad high.

B+ = 126.7v
Grid = 15.7v
Plate = 114.9v
200meg/470k = 62.3v
560R/27k = 68.6v
15uf/63v = 70v!!!

How do you think I should proceed?

 

[Spencerleehorton]

Btw, when I insert a valve, B+ goes up to 206v still.

I can't see why b+ isn't drawing current when tube inserted?

Spence.

 

[Spencerleehorton]

Without load I get the following
B+ = 206v
Cathode = 0v
Anode = 206v
Grid = 0v
H+ = 32.4v


With tube in I get the following

B+ = 206.3v
Cathode = 0.292v
Anode = 206v
Grid = -0.180
H+ = 5.14v

Any ideas?

 

[Spencerleehorton]

Have left the first tube in there for about 10 minutes and cathode rises to about 25v and looks like it's going to keep on going, anode is about 172v and b+ is a little higher.
I also put a sine wave in on the capsule and scope confirms output on xlr.
Reluctant to put in one of the other tubes in case I damage it until I can get conformation that I have got this right with a few more tests.
Anyone got and ideas? Dodgy tube? Or something else? Is there a way to test?

Regards

Spence.

 

[Spencerleehorton]

I have just measured the psu and done some more calculations.
After bridge it is measuring 255v so I need to drop 135v, so 135 / 0.005 = 27000ohms.
So I need to change resistors to R1= 15k R2=10k R3 = 2.7k this would give me 120v at 5mA.
Seems the tube is only pulling 1mA at the moment?
So at the moment this is what's going on.
255v at bridge, 238v after 15k, 220v after next 15k, 212v after 6.8k, which makes a drop of 43v.
So 43 / 36800 = 0.001mA

I'm going to get this psu changed and then test again!!!

Spence.

 

[Spencerleehorton]

Good news!! Finally I have made a break through, it was a dodgy valve after all.
Once I had the courage to put in another tube the voltages all came together.
I've ended up with 22k, 22k, 2.2k on the B+ and I'm getting 122v on one mic and 119v on the other.
Cathode sits at 57.7v and on other one its 60v
Anode is 113.2v and on other it's 109.3v
Grid is 0v on both
H+ is 5.16v and on other it's 5.22v
At the 200meg/470k I have 60.2v and 58.4v
On first mic I changed 560R to 680R
Needs some fine tuning now to see if can get them both quite close.

One other point I could send a sine wave through the capsule front connections but not rear capsule? Is this normal?
I have ordered some capsules which are edge terminated and I'm still waiting on mains switch, pilot light and some 7 pin sockets.
At least I'm getting there now!!

Regards

Spence.

 

[Matt Nolan]

Looking good Spence.

Like the backplate voltage, you can't accurately measure the grid voltage actually at the grid. You should look at it at the other side of the 200Meg - between the 560 and the 27k. It should be a volt or two lower than the cathode voltage.

I wouldn't worry so much about replicating exact cathode voltages and whatnot in each mic. They will vary with the valves. Keep both cathode dividers the same so that you have the same feedback ratio in the follower - i.e. both 560 and 27k. Otherwise you may end up with a small difference in sensitivity of the mics.

You won't get a signal through from the rear diapragm connection without a capsule, or a dummy capacitor (80pF or so) in there.

Cheers,
Matt.

 

[Spencerleehorton]

Rather than 560R I've used 680R and cathode looks much better voltage, lower and as I've only got a 15uf/63v I'm happier about that!!!
Are you able to let me know what voltage figures I should be aiming for in all the appropriate areas please?
Or perhaps give me an example so I can work out what I need to know, not very knowledgable about this, I have done it before but a long time ago!!! Rusty!!

Regards

Spence.

 

[Matt Nolan]

So long as you use 680 in both mics, it will be fine.

You're trying to optimise the biasing of the valve for headroom. If your B+ is a little high but you bring the cathode back down to about half of the ideal B+ then that should be OK.

Essentially, the larger that small resistor is, the more negative the grid is with respect to the cathode, and so less quiescent current flows through the valve from anode to cathode. Less current in the cathode resistor (the 27k plus the 560 or 680) means less voltage develops across it.

Thinking about it further, it will help your cathode voltage situation using a larger resistor there, but also hinder your B+ situation. It may also lose you some headroom on the negative half-cycles. I would imagine that the AKG engineers optimised the quiescent current for SNR, running a little less may worsen your mics' SNRs.

Whether any of these effects are significant, I'm not sure. You can look at the headroom and where and how it saturates on an oscilloscope with an injected signal, and you can judge noise by ear or if you have good enough measurement gear. If anything seems totally out of whack then you could experiment with modifying some values to set up the valve slightly differently, or you could get some 75V tants and use the AKG schematic values as-is. If all is fine with your minor deviations and your cathode voltage is OK for your 63V tants, then just leave it be.

 

[Spencerleehorton]

So would it be better to have 120v B+ and 60v cathode, or 120v B+ and 50v cathode?

Regards

Spence.

 

[Matt Nolan]

If all is running as expected, the AKG schematic suggests 62.6V at the cathode, which is rather squeaky for your 63 volt rated tantalum output caps. If you run lower than this, you will reduce your headroom. Is it that big a deal to find higher voltage parts? Are they too big to fit in your case?

 

[Matt Nolan]

I forget which type, but the failure mode of certain tants is to go short circuit. That means that you possibly fry or magnetise your somewhat more expensive output transformers. I think 75V tantalums would be a good investment.

If you drop the cathode voltage to 50V, you lose about 2dB of headroom. Alternatively, you could switch that 27k for a 21k or 22k from ground followed by a 5k6 and tap your output through the tant at the top of the 22/21 which would preserve your headroom, but drop your sensitivity by about 2dB, and leave only 50V DC across the tant. Neither is particularly optimum.

 

[Spencerleehorton]

Yeah, I think your right, just invest in some 75v ones, these ones I have are a little large I've found some others that are much smaller.

Regards

Spence.

 

[Spencerleehorton]

Success at long last, have got one powered up and initial first results are ok, bit harsh on the treble side, polar patterns all seem to work. Output level through my mixes isnt all that great but through a neve 1290 sounds wonderful, put it through one of my LA-2A as well and it really comes alive.
Have only plonked it in the middle of the room and played some drums through it, tiny bit of bottom end boost at about 80hz.
I might experiment with changing the 15uf cap may try reducing it as advised by Matt down to 4.7uf and then try some higher values see what it does to the sound.
Need to give it a good test on all instruments before mods though.
Time to finish building the other one now!!

Thanks again all you guys, especially Matt Nolan for your patience, I know I ask some dumb questions but I am learning.

Regards

Spence,

 

[Spencerleehorton]

Couple of audio samples I've put up here:

http://www.mohawkstudios.co.uk/projects-felixstowe-recording-studio/

Would love some feedback.

Regards

Spence.

 

[Spencerleehorton]

Have to say im a bit shocked with how good this C12a is on vocals, put it up against the C12 and the ELAM 251 and for more modern tracks it kicks some serious arse in the mix.
Something i have noticed is the mic likes to be in the middle of the room and only used with pop shield, when used with reflection filter it sounds dull and lacks transients.
Just thought id comment on it.
My version 1 is 118v B+ and 61v cathode and H+ 5.05v and sounds wonderful.
my version 2  is 113v B+ cathode 56v and 5.3v H+ and doesn't sound anywhere near as good, so the changes are quite obvious.
need to get B+ up using 10k, 10k, 6.8k at the moment, will drop this a little to 8.2k, 8.2k, 4.7k and see what i get B+ wise
and need to drop H+ a little but dont have anywhere else to go with H+ as 100R pot full tweaked.
will i need to change R2 1k to 2K or R3 2.2k to 3.3K?

regards

Spence.