CF microphone

[Deepdark]

Quick one for you guys. I'm prototyping a mic based around a cathode follower (c37a or c60). I first looked for a 12ay7 (good choice?)

If i need around 60v polarisation, should i feed with 120v ht, which Will give me around 61v at cathode and 59v grid.  Is +/- 60v going from cathode to the transformer ok?

Looking at Rg, i mostly saw values around 100 to 150M. How should i determine this value?

By the way, i bias at 1ma / -1v.

 

[Matador]

Quick one for you guys. I'm prototyping a mic based around a cathode follower (c37a or c60). I first looked for a 12ay7 (good choice?)

Sure - what are you trying to do exactly?

If i need around 60v polarisation, should i feed with 120v ht, which Will give me around 61v at cathode and 59v grid.  Is +/- 60v going from cathode to the transformer ok?

Polarization is a different subsection from the tube biasing (in general). 

Are you working from a schematic?  You (generally) cannot send 60V DC through a transformer, and would need to cap couple it.

Looking at Rg, i mostly saw values around 100 to 150M. How should i determine this value?

By the way, i bias at 1ma / -1v.

You select Rg based on a balance between low-frequency roll-off and self noise of the resistance on the grid.  In general, you want the lowest resistance that will pass the amount of bass you are looking for from the mike.  The capsule C and the grid R form a high-pass filter.

For example, if you assume 70pF capsule capacitance, and you want to have a low-pass filter cutoff of 40Hz, you would need at least 60M for the grid resistance.  If you want more bass, make this larger:

100M -> 22Hz
250M -> 9Hz
500M -> 4.5Hz
1G -> 2.2Hz

My personal preference is 100M, as a) it's easy to find this value, b) it's inexpensive, and c) gives good bass response for many different capsule types.  Having a ~20Hz filter pole also reduces rumble/bass thumps from handling noise into your mike preamp as well.

Having said all that, lower values can be nice too.  If you want a vocal mike and are generally low-pass filtering everything below 100Hz anyways, you could lower the grid resistance down to about 20M and get a similar EQ curve "for free".  However once you build it, it will always be EQ'd this way.

No free lunch though - higher values have higher self noise (thermal noise) - in fact, jumping from 100M to 1G increases self-noise by 10dBv, all other things being equal.

 

[Deepdark]

Thanks matador. The schematic I'm based on are sony c37a and akg c60. Since I'm in a learning stage in tube design, i decided to start from scratch with a new tube (12ay7) and plan to use the rk7 capsule from microphone parts. Since there is no voltages references for both microphone, i cant compare my maths with them.

So, i start up with the known value of polarisation, around 60v.  Since i dont know what to expect at the cathode, i decided to work with a 120v ht, and bias it half way, so 60v, 1ma, -1v. It gave me Around 35vpk. My rk is 60k and rb is 1k. So, i got 1vdrop across rb and 60vdrop across rk. This is how i did it. I sure missel something since i cant find Any information about this circuit in microphone design. From cathode to the transformer, i put a cap, don't know yet the value. 4u7?

Are my maths ok? If not, may i know how to bias everything?

Thanks for your time

 

[Deepdark]

Here is the first draw. I'm not sure if my maths are ok. As I said, I start up with the idea that I need around 60V polarisation and then manage all the other values from that point. Don't know if I did it the right way. How do you bias the grid to get your 60v polarisation?

Since there is little information about this design in the microphone world, I struggle to understand exactly how it works and how to manage the right values.

For what I understand of this circuit: it's a cathode loaded circuit. The signal is put at the grid and the output is taken off at the cathode. The grid resistor RG is strapped between the load and the cathode bias resistor. The voltage across RK became the polarisation voltage. So, this is why I manage to get 60v there. Is it allright? But, what I try to understand is Do I need a ht of 60v cathode + 60v grid? Or the polarisation voltages came from the intersection of the 100M and the RB/RK, so my pol. voltages became 60v from the tap at the grid? In any case, I end-up with an output voltage of about 60v. isn't it?

Thanks for your help

 

[Deepdark]

Ok I did a little more research and found some piece of theory on internet. So I managed to redraw the load line and the circuit. Tel me if it looks ok. There is some point I'm still unsure such as how to evaluate Vin. Is there other important parameter I should look at??

Thanks again for helping me. I think there is no better Learning stage than designing it from scratch.