[squarewave]

To avoid the effect of the capsule coupling capacitor.
(Many people think a coupling capacitor has an audible effect on the sound quality.)

I'm not even sure that makes sense. They just moved the cap to the other side of the capsule so any distortion in that cap is still reflected as a difference across the gate and drain of the JFET.

[RuudNL]

The capacitor at 'the other side of the capsule' has to be there anyway, to obtain an AC ground for the polarisation voltage.
So it is still one capacitor less...

[squarewave]

The capacitor at 'the other side of the capsule' has to be there anyway, to obtain an AC ground for the polarisation voltage.
So it is still one capacitor less...

That's not what I see. Although they could use one less capacitor they don't. They added the AC ground cap to make what is also a 3rd filter stage. And 1G/100n is a much lower RC than the two previous 10M/100n. So yes, they removed the 470p but added a 100n.

I can only guess this was done to be more like a U47 circuit with how it applies polarization to the backplate AC coupled to ground and then grounds the diaphragm where it also takes signal. But in this case you might as well drop one of the 10M/100n stages and make the 1G/100n more like the 100M/10n of the U47. Otherwise it will take a loooong time to charge (and what is the leakage current of a 100n film cap?).

I think I'll use the stock KM84 circuit but with the K47 capsule. Just swap it in. No changes in capacitors. No changes in polarization. Does anyone see an issue with that? The capacitance is quite different (36p for K84 vs 72p for K47) but does that matter? Sorry to tear down someone's project but I just want to make sure I understand this correctly because these parts are expensive. Any and all criticism is welcome.

[RuudNL]

Yes, the 0.1 uF/1G.ohm would give a ridiculous long time to charge. (Over 1 minute...)
So you could omit those components.
But for audio, the effect would stay the same; it doesn't matter if you have the one 0.1 uF in the circuit 'at the other side of the capsule', or the other 0.1 uF.

[squarewave]

I'm building a KM-84 mic with a Peluso PK-47 capsule but it's not obvious to me which side is the "front" or how to wire this precisely. The KM-84 schematic doesn't make it obvious either and I have never worked on mics so I want to make sure I understand this perfectly.

The PK-47 capsule has a wire attached to each diaphragm and screw holes for the mount that are slightly offset to one face. Included are 3 small screws that appear to be for mounting. There is no wire attached to the ring. I also have the correct mount which also has holes offset to one face.

Just to be crystal clear, the capsule is mounted so that the offset holes complement each other such that the capsule is centered correct?

Which side is the "front"? Do I want the front diaphragm wire to be the input and thus gets the bias? And then attach a wire to the ring which is the "back"? And then the other wire is simply left unconnected? There is no eyelet or wire for the ring. Why? I can salvage this from the donor mic but I am confused as to why it was not included.

[ln76d]

You should ask manufacturer why he didn't add third wire for capsule.
https://io.groupdiy.com/file/gdiy-2017/u/39511/58d1402a088e3.pdf
Use this schematic, front diaphragm should be connected to the R1/C1/Q1 node, backplate ("ring" i guess) to R2/C11.
If you are buildng cardioid only mike, then the rear diaphragm left unconnected (isolate wire).

[squarewave]

How do the Neumann mic capsule schematic symbols work?

In this graphic which network is connected to the the backplane and which is connected to the input of the mic circuit?



Is the thick rectangle the backplate/mounting ring and the thinner bar a diaphragm?

So in this particular mic (KM84) the diaphragm is grounded and the signal is taken from the backplate as well as charged with bias?

[ln76d]

Is the thick rectangle the backplate/mounting ring and the thinner bar a diaphragm?

Yes

So in this particular mic (KM84) the diaphragm is grounded and the signal is taken from the backplate as well as charged with bias?

Yes, like in most of SDC microphones (schoeps was one of exceptions), where due to capsule design, diaphragm have permanent connection to the microphone body which is grounded.
In case of LDC, where the capsule don't have to be connected to the ground better is to use connection with dipahragm connected to the circuit input and polarize backplate. This allow to remove input capacitor. Both changes are improvement.

[squarewave]

What schematic is this from? Clearly it's a Neumann schematic but which mic is this?

[ln76d]

I just redrew  your pic in ms paint

[Recording Engineer]

You know, this whole thread has got me thinking... Is there still not yet a cheap Chinese mic out there (tube, FET, transformer, transformerless, or otherwise) that at least has a circuit and parts that most would agree doesn't need modification (or at least not more than 1 or 2 stupid-simple changes); capsule, transformer, tube,  or body swapping aside?

[ln76d]

You know, this whole thread has got me thinking... Is there still not yet a cheap Chinese mic out there (tube, FET, transformer, transformerless, or otherwise) that at least has a circuit and parts that most would agree doesn't need modification (or at least not more than 1 or 2 stupid-simple changes); capsule, transformer, tube,  or body swapping aside?

Still not

Akg perception could be good candidate (this time it's U87Ai mixed with C414 PNP follower) but aslo screwed up and need modification to sounds pretty good.

Current Rode NT1 (black) is pretty decent (first NT1 which doesn't suck!!!) and there's nothing to change inside.
What's funny is that it have all the "mods" which i was doing for years in NT1a - coincidence?
Still you need to like it sound, because still it's very specific but much better than all the previous versions.

[Recording Engineer]

Wow! You'd think after however many years and however many renditions of the same things over and over again, I would have thought at one of the Chinese manufacturers or importers would have made at least one!