Analag 6AU6 tube mic

[pyjaman]

http://www.groupdiy.com/index.php?topic=22825
This attractive circuit seduced me for the idea of the partial cathode bypass and comparing with others (absolutely not saying those are less good) the fact that there is no coupling cap between capsule and grill...

In the way of building myself a tube mic, I was wondering several things:

- one thing that I don't clearly understand is how are determined the values of the three 0.1uf decoupling caps in the capsule polarisation network. many desing show lesser values in these spots (could be cool to save space in the mic body with yet-good-quality, smaller value caps).

- Same kind of question concerning the 250M resistor: how is its value determined? I can't source 250M resistors for less than about 25EUR/pc !

- I've got a peluso CEK12 capsule, while this circuit is based on the Neumann-ish CEK367, as anyone an opinion about CEK12 dropp-in (not considering, of course, the use of the plate-to-ground 500pf) with this circuit?

-while I can read almost everywhere that good suitable 6AU6 seem very hard to pick (I'm afraid I'm not rich enough to buy 100 to pick about 3 good spares...) I have a bunch of EF86 (Valvo,siemens,tfk,mullard,Philips...). could'nt we try, by tweaking Rp and mostly Rk, to try another tube in this topology (why not even a real triode as for example a 6n1p-ev wich I have a bunch of).
Or, has anyone an advice about any type/variant of tube that could favorably suit the circuit as it is ? (I will use  a 5:1 LL1538 tx at output - no money to change that either).

Every one is very welcome to give opinion or advice, including, of course, the designer himself !

Thanks,
Regards,
Laurent.

 

[pyjaman]

Bump...? Ouch!...

As anyone a short explaination about how are calulated those R and C that make polarization on the back diaphragme? Or a link?

Thank you.

Laurent.

 

[ioaudio]

look up:
ohm´s law
voltage divider
filter capacitor

 

[pyjaman]

Thank you. I really was suspecting that, and I know I can seem stupid, but I've started from nothing few years ago and learned everything I know by myself (and some on this forum, of course). So the basis of my knowledge is insecure and full off holes...

So if it's as simple, and if the polarization voltage comes from the PSU, (labelled "P") , Those 0.1uf-250M-0.1uf are simply a C-R-C filter depending on input resistance of the back diaphragme? so Why are those values are so different from one mic to another? Depending on the capsule type?

Feeling more and more stupid... Maybe that's a good thing.... :-\

 

[zebra50]

Here's another way of looking at it... don't worry about the exact values, I'm generalising...

We need to polarise the diaphragm and backplate with some voltage, but we don't need a lot of eletrons to do the job - we're not actually passing current, just creating a potential difference, so we can use a high value resistor - you'll often see around 100Meg to 1gig. But because so few electrons are around to do the job, the system is prone to interference and other crap picked up between the power supply and the capsule. So we need a capacitor before the big resistor to filter this out and stabilise the voltage. It needs to be big enough to do the job, and designers may err on the side of caution and over engineer. (For fun, try removing this  )

There may be a second capacitor which provides an AC path from the back capsule to ground (or to the amplifier grid).

Incidentally, you can use the capsule polarisation from the schematic above with either the G7 or Royer type amp circuits - it works just fine (I have tried it!).

Hope that helps a little.

Edit - and of course you can use a real triode - track down the UM57 circuit for some inspiration.

 

[Gus]

Real triodes? What do you mean?

Whats a VF14 in a 47? EF86 in a 67?

When 6au6s are wired as a triode they are really nice triodes.   Why do you think the  "DESIGNERS" at Schoeps and other companies used them? there were "true" triodes available

Try a different tube type this is DIY.  A ef86 might be close, be mindful of the G3 connection.  Triode mode is sometime G2 and G3 to anode and other times G3 to cathode and G2 to anode sometime tubes have g3 wired inside to cathode so look up your tubes.  Look at tube microphone circuit you can find on the web and search here.

 

[ioaudio]

sorry for my grouchy answer -  i should not post when i´m in a hurry...

 

[Sonic]

All greetings!
I used Russian analogue of a lamp 6au6 - 6j4p in such inclusion:
The grid 3 is connected to the cathode;
The grid 2 is connected to the anode;
Grid 1 without the dividing condenser on a cap.
Experimental method has defined, that for my caps enough the loading resistor of the first grid from 100 up to 200 MOhm. Filtering condensers of a polarizing voltage too I put 0.1 uf. The size of a polarizing voltage should be selected under a concrete cap (I normally work about 35 volt).
Actually, to sound it is possible to force any lamp, the main thing to pick up optimum parameters.
Yours faithfully, Alexander.

 

[pyjaman]

Zebra, Sonic, Gus, thank you for joining this post and complete the "hurry-ish" explaination Ioaudio gave .

Ioaudio: don't care about that, I felt a bit stupid but not hurt... That said, your "grouchy answer", was the way to go.

Gus: You surely understood what I meant by "real triode" (a 3 electrodes made tube) absolutely not saying that triode wired pentodes don't make (very) good triodes. Reading some threads about that, it seems that using "real" triode vs triode strapped pentode makes no "theorical" difference.
        Concerning the use of an EF86, some designers strapp G3 to cathode ( as in the G7 for ex.), others strapp it to plate ( Ioaudio's MK7... -seems a VERY attractive project ) .I think I'll try both if I go the EF86 way.

  So, if I resume : taking the example of this circuit, the 250M acts as a kind of current limiter wich provides Potential voltage at very low current to the back diaphragm, and the cap at its junction with "patern voltage" from PSU provides filtering and stabilization after cable lenght and so. could it be suitable to use for example a 100M resistor instead of 250 in this place (much easier to source) without too much current going to the diaphragm (with Peluso CEK12) ? Or 1G resistor (already have a few) without too much disturbance? What could be the effect on sound?
    concerning the "AC path" 0.1uf cap, tyed from junction of diaphragm with 250M res. ; In some design (Ioaudio's MK7, G7... to take the same examples as above) its value is more on the10nf range, but those "link" the AC with the other diaphragm, not only to ("virtual") ground. I guess kind of HPF here... Lower value could be tryed in this design?

Zebra: thanks for the UM57 idea, indeed some similarities in both designs...

Thanks to all of you. This forum rocks (and the guys on it too) !

regards,
Laurent.

 

[ioaudio]

sure, just use anything between 33M or 1G. 
regarding the value of the caps - shouldnt make too much difference.
about G3 - thats a good thing to try yourself, hook it up both ways and listen.

best,
-max

 

[zebra50]

could it be suitable to use for example a 100M resistor instead of 250 in this place (much easier to source) without too much current going to the diaphragm (with Peluso CEK12) ? Or 1G resistor (already have a few) without too much disturbance?

Yes! Either should work. It's DIY so use what you have! Same with the caps - 10n to 100n are common and not such a big leap! It is important to use really nice caps here though - usually polystyrene.

those "link" the AC with the other diaphragm, not only to ("virtual") ground.

Yes! Different designs achieve summing the signal from the two capsules in slightly different ways. One method (G7) looks like a parallel connection of the two capsule elements, with the backplate connected to the tube grid. The design above looks like series combination of the two elements - that's how it looks to me!

 

[ioaudio]

i just had a closer look -
we are talking about this schematics, or is there another updated one?
http://www.twin-x.com/groupdiy/albums/userpics/R8.JPG

the second membrane is has no AC-connecion to the grid as far as i can see?

 

[zebra50]

That's what I thought when I first saw it, so I built this section of the circuit to convince myself. It does work (running into G7 and Royer circuits).

I see the capsule as wired as two condensers in series - so the back capsule is AC connected to the grid through the front capsule! The circuit is completed by the 0.1uF to ground.

There may be another way of describing it!

 

[pyjaman]

Yes, I saw that it was unusual, just didn't figured how it was... But it works. How? Something else I don't really understand... Might be a new parameter in the choice of the caps values...

 

[ioaudio]

ah - now i see - the capsule´s inherend capacity is the AC-path, right?

 

[analag]

The cap values are not critical and can be anything between 0.01 and 0.1uF. A 6AH6 works great in that circuit.

analag 

 

[zebra50]

ah - now i see - the capsule´s inherend capacity is the AC-path, right?

THat's it! 

 

[pyjaman]

Ok, I understand too now !

Thanks again guys ! You gave me almost everything I need to go forward and make it !
Time to build and experiment now.

Rowan : thank you for joining and for the 6AH6 tip (I remember you kept it secret in your initial thread  ;D )

Regards,
Laurent.