Need advice/explanations on some parts in the circuit ;)

In attachement is microphone circuit schematic which confuse me a little
Parts marked red.
What's the exact role of C15 - 1uF - ? Is it feedback only?
L5 - 270nH - am i right that it is for ground separation?
C7 and C9 looks to me like HPF - increasing C9 low end response should be more flat - right?

Does anyone see any LPF inside the circuit?
What for doubling 2.2nF RF caps?  Is it related with ground sepration (if am right about it)?

C15 - quoting from Douglas Self's "Audio Power Amplifier Design", the chapter / section about the emitter follower type II output stage (which, to me, looks similar to what's in your schematic):
"The shared driver emitter resistor Rd, with no output-rail connection, allows the drivers to reverse-bias the base – emitter junction of the output device being turned off. Assume that the output voltage is heading downwards through the crossover region; the current through Re1 has dropped to zero, but that through Re2 is increasing, giving a voltage drop across it, so Q4 base is caused to go more negative to get the output to the right voltage. This negative excursion is coupled to Q3 base through Rd, and with the values shown can reverse-bias it by up to ? 0.5 V, increasing to ? 1.6 V with a 4 Ω load. A speed-up capacitor Cs connected across Rd markedly improves this action, preventing the charge-suckout rate being limited by the
resistance of Rd."
http://m.eet.com/media/1064206/self_ch19_fig2bc.jpg (the example on the left)

L5 - some "extra" RF-protection/filtering, most likely
The common end of C7 & C9 seems to be some feedback node, judging by R25 coming straight from the output of that complementary stage. Not quite sure increasing C9 will yield any noticeable improvement - i see a 220k (R31) right after it, which would give a -3dB corner of 22Hz.

Don't quote me on this, but the extra pair of 2.2nF caps there might make the low-pass a 3rd order one (instead of a 2nd order LC-LPF).

What mic's this from?

And those two "paralellled" JFETs are, i assume, for manufacturing options? 2SK4041 seems to be some 4-pad "BGA"-ish package, and the 2SK3857 seems to be more like a SOT23.

output is a " diamond pair" http://www.tubecad.com/2010/05/blog0187.htm

Thanks Guys

Khron said:

L5 - some "extra" RF-protection/filtering, most likely

Click to expand...

Between ground only?
I know that inductor in ground path is for separation of grounds but not sure here

Khron said:

The common end of C7 & C9 seems to be some feedback node, judging by R25 coming straight from the output of that complementary stage. Not quite sure increasing C9 will yield any noticeable improvement - i see a 220k (R31) right after it, which would give a -3dB corner of 22Hz.

Click to expand...

Today i made simulation.
I connected input voltge source in many ways since i don't know how exactly negative feedbck is applied.
In all cases increasing C9 value changed low end response significant.
I found that 100nF in parallel with 33nF should be what i need

Khron said:

What mic's this from?

Click to expand...

Check my  last post in technical documents
I don't want to make visible name and brand in google searching for some reasons ;D

Khron said:

And those two "paralellled" JFETs are, i assume, for manufacturing options? 2SK4041 seems to be some 4-pad "BGA"-ish package, and the 2SK3857 seems to be more like a SOT23.

Click to expand...

I wouldn't be surprised if both are used.
One is for ECM type and another FOR IMPEDANCE CONVERTER OF ECM.
Unfortunately i can't verify that, because i have no acces to it.
All the parts around fets which have ** marking are mounted inside the capsule and there's no option to open it...

Gus said:

output is a " diamond pair" http://www.tubecad.com/2010/05/blog0187.htm

Click to expand...

This is what i needed!

BTW. polarisation voltage is -70V not -60V.
Basic idea is to mod it  a little, overall microphone sounds good but there's something unpleasant in it.
I can only exchange some parts from smd to tht.
Add mkp 100nF in parallel to the C9, change C15 for MKS , chnge C33 and C37 for silmic (if fit or any other good caps), remove C43 and C45 or replace it with fkp.

I did mentioned mod except RF caps and changed 33nF to 100nF (not 133nF in parallel).
Yes, it's better now
Now is more clear, with better low end control.
Overall it could be darker or have more  midrange but anyway i like it
Maybe next step will be change of two RF caps.

C15 is indeed an attempt to do what Self describes.  But it may not be appropriate for this circuit which is NOT a double emitter follower.  A capacitor here can charge up/down and result in too little/much Iq on overload.

I would test such a circuit until it clips at various frequencies and see what happens.

C7 & 9, R25, R27/31 and the O/P stage form a https://en.wikipedia.org/wiki/Sallen%E2%80%93Key_topology filter.  I have this in nearly all my commercial designs.  I'll claim Dip. Ing Wuttke copied my 1980s Calrec stick mikes for his 'new' CMC6 preamp  8)

By 22nF RF caps you mean C43 & 45, these with L1, 3 & 5 are to stop RF going INTO the mike.  They MUST be AT the XLR pins.  Even 1" wire between the pins and these makes them useless.  Best to have the PCB mounted on the pins so these can all be as short as possible though I would probably want to mount the 22n ceramics directly to the pins with the shortest possible leads.

And you want a DIRECT connection from p3 XLR to the mike body.

If all this is done, L5 is probably not needed.  But if the PCB is connected to the XLR by 1" of wire, the 'earth' lead is enough to be an aerial at mobile phone frequencies.

If you don't do this, the BBC and other broadcast organisations will curse you.  But don't worry.  There are several famous name German makers whose product suffer from this evil too  ;D  Their older stuff is OK

What mike is this?

As always thanks ricardo!!!

Overall is hard to work with this microphone, almost everything is truly tiny smd and there is no acces to the input of impedance converter.
XLR connector is soldered direct to the PCB, there is also additional shielding plate on the top of the connector.
Caps are 2.2nF and all truly close XLR.
I think i will leave whole topology as it is.
For now microphone sounds better and after writing in "Barstow U-47 Mod of Marshall MXL2001" i realized that i need to try headbasket mod. This could help with hi frequencies which are little bit annoying. It's not scratchy, harsh chinese k67 sound but could be better.

So i did comparison A/B test with headbasket and without it.
There's significant difference
I removed internal front mesh and now sounds much better at hi frequency range - smoother.
A little LPF or even band-stop filter could be great here.
Does anyone have idea how implement it in this circuit?
Even simple cap to ground could be great option.
If the circuit would be THT i would convert Sallen–Key topology for it but there's no option...
Rather i would like to add something than change existing circuit.
4kHz-10kHz is the region which interest me