5-pattern mic circuit

[Khron]

R37 and R38 MUST have a common end where the traces overlap, otherwise the powering makes no sense. And if you rotate Q10 by 90deg clockwise, that along with R32, R33, D8 and C16 will form a (more easily-distinguishable) voltage regulator.

Those RLC filters to ground, on the output, look really funky. They sure must've taken their time to notch out some specific frequencies(?)...

Either way, i took the liberty to re-draw the bulk of the schematic into a form that makes more sense / resembles quite conventional topologies.

On that note, are you sure there's no resistor to ground, from the base of Q8? That and Q7 "above" it form a constant current sink, same as the one made up by Q2 / Q6 / R23 / R8.

 

[abbey road d enfer]

I have doubts about the resistor values associated with Q9. With the values shown, this stage will clip 16dB earlier than the mic's output, which would ruin the High Pass filter and distort the output enormously. Since the head amp has 6 dB gain, this NFB path should in fact attenuate by a similar amount.

 

[dhammer]

Khron - thanks for this!  This weekend, I'll redraw with yours as a template. 

I checked values of the parts you guys questioned: 

I did leave off a resistor (R27) that ties the base of Q8 to ground.

Checking the RLC filters – the color code on L5 is tough to be sure about but it could be 10mH instead of 1mH.  The rest of the values are correct.  Are these high-cuts on the outputs?  I see something like this on the 414 TLII but with very different values (and matching values on both legs).

I double-checked the values & wiring of R28, 29, 30 & 31 & they are correct on the drawing.

 

[Khron]

I don't think i've seen the schematic for the TL II.

 

[dhammer]

414 B-ULS/TLII service manual attached

 

[Khron]

That's for a TL indeed, but not for the II

That being said, those inductors are very clearly in series with the output leads / pins in the AKG's (and many other mics), and not as part of RLC filters going to ground.

R39 / C24 / L? do actually have the same values as R35 / C17 / L3, and the arrangement does indeed look like an attempt to provide a similar AC impedance on both XLR pins 2 and 3.

But the R43 / L5 / C27 really throws a spanner into the works... Are you sure those aren't between the XLR pins 2 and 3 (instead of 2 and 1)?

 

[dhammer]

Ah – I think the only difference between TL & TLII was a capsule redesign.  The parts list in this manual is headed "TLII".

I double-checked  R43/L5/C27 it's drawn correctly with C27 negative pole connected to ground. 

Oh also, from your previous post – R37 & 38s pads are tied together where the trace from R32 comes in.

 

[Khron]

I don't suppose you could attach a photo or two of the circuit board, could you?

 

[abbey road d enfer]

But the R43 / L5 / C27 really throws a spanner into the works... Are you sure those aren't between the XLR pins 2 and 3 (instead of 2 and 1)?

This circuit provides a few dB attenuation centered at about 5kHz. That may be done as an attempt to diminish harshness.

 

[Khron]

A bit of a "brute force" approach - surely one could've done something similar within the feedback network, i think...

 

[ln76d]

That's an interesting statement, considering the original Schoeps schematic also has those caps at 22nF. Silly germans didn't know what they were doing, hmm? ;D

Don't try to be smartass, rather focus much more on details, then you will know what you are writing about

First of all - we both can't speak about intentions of designers from 50's, 60's or 70's etc.
Please remember that at these days all designs were made for broadcasting which had always own rules.
Other thing is that "Silly germans" used RF LC filter at the output 50uH/22nF, which have different curve than most of various RF RC filters used in chinese microphones. Don't even compare schoeps cmc microphone to similar topology used in many chinese microphones - leftover after screwed up Rode NT1 design from 90's. These are completely  different microphones.
Another thing is that nasty effect is much more audible exactly with chinese belowed capacitor in series input topology than with "pure cardioid". By occasion - why that "Silly germans" bothered to make pure cardioid with sdc? Easier would be made capsule permanently grounded?
Lets look on other "Silly germans" - in Berlin was used 47uH/330pF RF filter - (excluding other factors and make oversimplification since it is different topology) are these values not enough?
From the other hand lets look on other Silly Austrians/Germans - again we have 22nF but now is 250uH - and it's worse.
What's funny C414E1 have these values inside microphone, but doubled RF filter in S42 box is just 250uH/5.6nF and the curve looks much better for it.
Why some use pretty high values - i don't know - there can be many factors - but still when something isn't needed - don't use it, if lower values have same purpose i will go that way.

 

[ln76d]

I would disagree somewhat there. Capacitors in series introduce distortion when their value is small enough for a significant voltage to develop across it. In this example, the cap is about 15 times larger than the capsule and loaded with a high impedance so no significant voltage develops there.

Agreed, but my question was regarding the head amp topology, not the bias. You mentioned input topology, which IMO concerns the head amp, not the bias circuitry. But ok, matter of language.

Rather i'm about adding few factors to eachother - different impedance (and no - double 2Gohm resistors will not do the same work), no ground at the capsule at all (then headbasket acts like Faraday cage), and that capacitor - sometimes even change of type makes audible difference same as a value of capacitance. Overall each topology acts different - small changes of sensitivity, different polarisation point (like for example front diaphragm in C414 topology) which also affect diaphragm tension, different DC filtering points etc. Don't mind extra few volts  from the circuit on front diaphragm (or backplate in C414 type of connection).
This isn't easy to explain in theory, especially that not much theory exists in that topic, especially concerning on properties of condenser transducer.  Tests are best option here

Agreed. In a single-FET head amp, 2SK170 is not a good option. Less a problem in a source-follower configuration, though.

2N3819: 6nV/sqrtHz, 2SK170: 0.95 nV/sqrtHz.

Bootstrap tends to make the difference less noticeable. Wet finger says capacitance is halved, which is still 15pF. But OTOH, in a common-source topology, Miller effect adds about 20pF to a 2N3819.

Multum thanks for explanations!
Where did you get input noise values??

Another thing is that change need differnet biasing, different values of resistance and definately both options sounds different - i prefer definately 2n3819 (it's not about that Neumann used it - it's just pretty good FET for application).

 

[abbey road d enfer]

Rather i'm about adding few factors to eachother - different impedance (and no - double 2Gohm resistors will not do the same work), no ground at the capsule at all (then headbasket acts like Faraday cage), and that capacitor - sometimes even change of type makes audible difference same as a value of capacitance. Overall each topology acts different - small changes of sensitivity, different polarisation point (like for example front diaphragm in C414 topology) which also affect diaphragm tension, different DC filtering points etc. Don't mind extra few volts  from the circuit on front diaphragm (or backplate in C414 type of connection).
This isn't easy to explain in theory, especially that not much theory exists in that topic, especially concerning on properties of condenser transducer.  Tests are best option here

Using two Gig resistors instead of one introduces an additional pole to the response, so it's gotta change something in the sound; the effects of varying bias voltage are very well known and understood. Different arrangements of stray capacitance when changing from grounded capsule to grounded diaphragm are also well understood. Effects of capacitor type are also well known and quantized. Any difference in response can be explained scientifically. What is harder to quantify is which alternative will sound best...

 

[abbey road d enfer]

Silly germans didn't know what they were doing, hmm? ;D

I think your humour was lost in transmission...

 

[abbey road d enfer]

Where did you get input noise values??

http://www.futurlec.com/Transistors/2SK170.shtml
http://www2.eng.cam.ac.uk/~dmh/ptialcd/jfet/2N3819.pdf

 

[abbey road d enfer]

With both types used in 797 designs 2sk170 and 2n5457 difference is really significant when change it to 2n3819 (also other FETs - but this is one which i prefer most).

Typical case where a component is used in an inadequate configuration. It doesn't result in 2SK170 being bad and 2n3819 being good. In the discussed topology (opamp with differential input) 2SK170 may have some advantages due to higher Gm.

 

[dhammer]

Attached is latest rev of the drawing – thanks Khron for huge help with this.  I hope I have it this time.

I'll put pix of the boards on a followup message.

 

[dhammer]

Here is a photo of the boards – the solder sides are flipped so they line up to the parts on the component-side views.

 

[ln76d]

What is harder to quantify is which alternative will sound best...

There's no rule - because it also depends on many factors
Usual M49 type of ine optput (cap between diaphragms) is safe option, sounds really good, worth to think about little bit higher polarisation voltage. I really recommend to find best possible cap value - in most of designs value can start from 220pF to 1nF at the end.  If from any reason (because there's a lot of factors on the way) capsule sounds too bright (even due to headbasket resonance or other factor) then C414 topology with signal taken from the backplate and symetrical polarisation of both diaphragms should work better. Not without a reason lower values of resistors and capacitors for DC were used Going higher with the values, only change for worse.

Typical case where a component is used in an inadequate configuration. It doesn't result in 2SK170 being bad and 2n3819 being good. In the discussed topology (opamp with differential input) 2SK170 may have some advantages due to higher Gm.

Yeap, that's why am writng about tests all the time For me something can be better for other maybe not

http://www.futurlec.com/Transistors/2SK170.shtml
http://www2.eng.cam.ac.uk/~dmh/ptialcd/jfet/2N3819.pdf

You have definately better google
I only found on-semi datasheets for 2n3819 on several sites
Thanks!

//EDIT//

I forgot to mention, fulll multipattern in C414 topology needs negative voltage for supercardioid-figure of eight setups.

 

[dhammer]

U89 circuit attached for comparison