The best way to polarize an LDC capsule for multipattern function

[skyy7]

Hello! I was researching how to best polarize an LDC and ran into several different methods.

I noticed that many Chinese mics use a mid-supply to polarize the backplate with 30V, one side with 60V, and another with 0V, then couple the AC signals together at the gate of the JFET (CAD GXL3000). While others used oscillators and rectifiers to generate a +60v and a -60v rail and polarized the membranes. I also ran into a circuit from @In76d (although abandoned) that only polarizes the backplate.

I imagine the solution from In76d would be the best as the membrane is not polarized and does not build up dust and also seems to be the simplest solution... What are the advantages of each solution?

Any advice would be appreciated. Thank you!

From @In76d at: https://groupdiy.com/threads/m4-8-7...tched-three-polar-patterns-single-6028.61926/


CAD GXL3000 from @Khron


From https://halfshavedyaks.xyz/pcb/BM700-polarisation.php

 

[Khron]

many Chinese mics use a mid-supply to polarize the backplate with 30V, one side with 60V, and another with 0V

"Many"? I've only encountered maybe three so far..?

But if you mean the CAD GXL3000 (and derivatives), the backplate is at 60v, front diaphragm at ground, and rear diaphragm is in fact at ground / 60v / 120v(!), for the three patterns.

Same goes for nearly all tube mic schematics i've seen (U67 among others)...

PS. Fixed voltages on all three elements only net you one single polar pattern

 

[OneRoomStudio]

There are a lot of threads about this already. It really depends on what power you have available for polarization. Most popular capsules (K67 derivatives, M7s, CK12s, K47s) have the best balance of sensitivity and headroom with something around a 50-60V differential between backplate and each diaphragm. So, ideally, you'd have at least that much voltage to start with (hint, P48 mics won't, unless there's a DC-DC converter).

The next consideration is that if you want to be able to make a figure-8 pattern, and you want to retain the ~50-60V backplate-to-diaphragm voltage differential, that means you need to either also generate -50/60V, or have 100-120V to start with. Most tube mics already have a B+ of around 120V, so it was easy to use 120V for one side, 0V for the other side, and ½ B+ (60V) for the backplate. Some solid state mics copied that idea using "mid-supply" on the backplate, even though the B+ was only 60V. That results in only a 30V backplate-to-diaphragm differential, which is 6dB less sensitive than a 60V differential.

For this reason, Neumann used a DC-DC circuit to convert 48V phantom to +/- 60V. This way, they still have a 60V differential on either side.

It's fine to polarize a capsule with less than 60V - you just lose sensitivity.

Another consideration is whether you want continuously-variable polar patterns or switched polar patterns. There are advantages to each. I'm a fan of "pure cardioid" (rear diaphragm disconnected) when possible, since the sensitivity and response are better. That requires a different setup than just supplying the same voltage to the rear diaphragm and backplate.

Lots of considerations. There's no "best" per se, just different methods depending on what else is going on in the mic and what your goals are.