U67 de-emphasis network

[OPR]

How efficient is a condenser capsule at energy conversion, in either direction?

My understanding is that condenser capsules put out a very low power signal, lower than ribbons', which are lower than typical dynamics; that's why condensers need some kind of head amp (in the mic) to put out enough current to drive a preamp.

That suggests to me that they're a lot less efficient than dynamic mic capsules, however efficient those are, so not very efficient.

That in turns suggests to me that the feedback-into-the-mic-connection trick is doing a lot less to oppose the diaphragm's motion than to electrically cancel its LF electrical output at the input of the head amp.

let me try and explain it this way.
In order for the capsule to pass the NFB audio in the form of AC voltages the diaphragm must move.

And in order to create the required cancellation, around 5 6 db in the LF it would need around half of the input voltage to bring the low end back to flat.
try it in a DAW canceling audio against it's self and it might make more sense to you can roughly get an idea of the ratio blend required for it to function

 

[kingkorg]

In order for the capsule to pass the NFB audio in the form of AC voltages the diaphragm must move.

Maybe this is the part where we diverge? The diaphragm doesn't have to move. If this were true the NFB wouldn't work if the capsule was replaced with a plain capacitor.

 

[OPR]

Maybe this is the part where we diverge? The diaphragm doesn't have to move. If this were true the NFB wouldn't work if the capsule was replaced with a plain capacitor.

And neither would the mic...

Well then ask your self the question... is a microphone capsule a plain capacitor?

 

[Paul W]

And neither would the mic...

Well then ask your self the question... is a microphone capsule a plain capacitor?

Mostly. It is a capacitor and an inefficient transducer. The feedback voltage will serve to oppose the LF motion of the diaphragm, but only a little. Most of the diaphragm motion will still happen, and the signal the capsule puts out will be further high-pass filtered electrically.

Most of the high-pass filtering effect will happen whether or not the capsule acts as an inverse transducer and resists the LF motion, because you're feeding an out-of-phase LF signal from the amplifier output back to the amplifier input.

 

[OPR]

Mostly. It is a capacitor and an inefficient transducer. The feedback voltage will serve to oppose the LF motion of the diaphragm, but only a little. Most of the diaphragm motion will still happen, and the signal the capsule puts out will be further high-pass filtered electrically.

Most of the high-pass filtering effect will happen whether or not the capsule acts as an inverse transducer and resists the LF motion, because you're feeding an out-of-phase LF signal from the amplifier output back to the amplifier input.

Well it sounds like you mostly agree with me.
At risk of repeating my self...
Take a look at the curves. If the NFB was to defy the laws of physics and have no impact on diaphragm motion. then why would the low frequency
NFB effect the frequency response all the way up to 10K ,and possibly beyond, as this is where the HF NBF kicks in, which is still intact in this measurement. Could it be the result of extra diaphragm damping due to the LF NFB?... possibly.

 

[Paul W]

Well it sounds like you mostly agree with me.
At risk of repeating my self...
Take a look at the curves. If the NFB was to defy the laws of physics and have no impact on diaphragm motion. then why would the low frequency
NFB effect the frequency response all the way up to 10K ? could it be the result of extra diaphragm damping due to the LF NFB? possibly.

I don't see the connection. As I understand things, there are two separate questions.

There's circuitry that low-passes and 180 degree phase shifts the output of the amplifier, and feeds that out-of-phase LF back to the input of the amplifier, but also to the capsule which is connected there as well.

One question is how much of that high-pass filtering is accomplished by the feedback voltage applied to the capsule opposing LF diaphragm motion, and how much is done by simple electrical cancellation at the input of the amplifier.

It seems like a whole different question why either effect extends way up the spectrum. I would guess it's because the low-pass filter at the output of the amplifier doesn't have a steep slope across the middle frequencies, so the feedback (to both the capsule and the amplifier input) has middle frequencies in it, as well as LF.

If that weren't true, I don't know why either high pass filtering effect (by capsule backpressure OR by electrical cancellation) would extend up that high in the spectrum.

And I'm curious why the slope is so gradual across the middle frequencies. Is it designed that way as a gradual "tilt" EQ, rather than a normal high-pass filter?

 

[jBam]

Some very interesting discussion and reading in this thread since I last checked in!... It's an interesting concept / question: is it different to modulate the capsule vs. the signal-at-the-amp (or anywhere after).

The way I see it simplified is without thinking about a filter at all... just a raw feedback of the signal... e.g.

If you had a raw 1kHz test tone being recieved by a mic capsule (as in, out of a speaker, into the mic capsule), and you bypass the filter and:

a) feedback a small portion of the recieved signal to the amp input's signal (but inverse phase) then you're effectively reducing the level of the signal.
b) feedback a small portion of the recieved signal to the capsule polarisation to modulate it against the incoming signal, then, for me, that's the question: "is it the same outcome as a) above"?

a) is solid.

b)... well it's hard to grapple (to me, with limited knowledge but keen interest). My mind starts to play tricks on me, as I wonder how this feedback would work... this system would feedback the recieved voltage, which instantly lowers the signal level recieved at the diaphram by reducing the polarisation voltage, but that lowered level is the levels that's then actually recieved... which is fed back to lower the sensitivity which lowers the signal, and on and on.... I don't understand how this would be stable. But I don't understand a lot of things lol... and it has to be stable, because even with the ACTUAL filter used, this is what's going on (right?)... I imagine feedback to the capsule has a sort of logarhythmic reaction, where the effect of feedback lowering the level of the signal diminishes over and over until it's not "really" diminishing anymore and is effectively "stable", which is all instantaneous and what we end up hearing (as opposed to the signal instantly turning itself off due to constantly reducing it's own signal in e.g. a linear fashion).

So - my understanding above is possibly roughly correct, or just flat out wrong ... Eitherway, the question for me remains: "Is the result of this feedback in option b) the same outcome as option a) above"?

For feedback to the capsule to be identical to feedback at any point after that, it would entirely depend on the capsule having a linear reponse to voltage changes in the same way that it does with option a) above... a consistent sensitivity change for varying polarisation voltages (let's say "in normal operating conditions"). That's either how it is, or it's not... as a result, I see two options for what's actually happening:

1. having the polarisation voltage modulated by the recieved signal slightly pushes and pulls the diaphram based on +/- incoming signals in a manner that restricts it's movement resulting in a change in signal that is the same amount as e.g. lowering the signal at any point after the capsule... i.e. a linear response.
2. OR it is not a linear response, and the signal becomes distorted to some degree... i.e. it is NOT identical to doing this after the capsule due to non-linear ways polarisation voltages effect the recieved signal.

Man... but it kind of has to be Option 1... How else could a microphone with feedback networks recieve any filtered signal accurately with low distortion if it wasn't...

If it's Option 1, then the key benefits I see would be in diaphram protection / longievity... it physically restricts signals at the diaphram... sure not a huge amount physically, but certainly an audible / electronically signficant amount... and if those signals are areas that might cause damage like big plosives and such, then that has a benifit right?... HF filtering...? well less critical from that perspective... but if you're already filtering to the polarisation voltage, why not keep at it!?

 

[OPR]

I don't see the connection. As I understand things, there are two separate questions.

There's circuitry that low-passes and 180 degree phase shifts the output of the amplifier, and feeds that out-of-phase LF back to the input of the amplifier, but also to the capsule which is connected there as well.

The the LF NFB connects to the capsule back plate and not directly to the amp so the capsule is part of the signal path.

One question is how much of that high-pass filtering is accomplished by the feedback voltage applied to the capsule opposing LF diaphragm motion, and how much is done by simple electrical cancellation at the input of the amplifier.

Again The the LF NFB connects to the capsule back plate and not directly to the amp so the capsule is part of the signal path. there is no direct path between the LF NFB and the amp.

It seems like a whole different question why either effect extends way up the spectrum. I would guess it's because the low-pass filter at the output of the amplifier doesn't have a steep slope across the middle frequencies, so the feedback (to both the capsule and the amplifier input) has middle frequencies in it, as well as LF.

Here are the details for the 67 amp FR. The1500 should be 15k for the the HF roll off the lows are at 40HZ so dont extend that far up.

Regardless....
in simplest terms we have an AC voltage being fed in to a capsule if you don't think the capsule will wiggle from this. I don know...
It's a bit 2+2 =5

 

[Khron]

In order for the capsule to pass the NFB audio in the form of AC voltages the diaphragm must move.

Maybe this is the part where we diverge? The diaphragm doesn't have to move. If this were true the NFB wouldn't work if the capsule was replaced with a plain capacitor.

is a microphone capsule a plain capacitor?

Mostly. It is a capacitor and an inefficient transducer. The feedback voltage will serve to oppose the LF motion of the diaphragm, but only a little. Most of the diaphragm motion will still happen, and the signal the capsule puts out will be further high-pass filtered electrically.

One question is how much of that high-pass filtering is accomplished by the feedback voltage applied to the capsule opposing LF diaphragm motion, and how much is done by simple electrical cancellation at the input of the amplifier.

in simplest terms we have an AC voltage being fed in to a capsule if you don't think the capsule will wiggle from this. I don know...

I can't help but side with the majority here, and "still" doubt that the feedback signal electromechanically affects the diaphragm motion quite THAT much, as opposed to the purely electrical effect (using the capsule as "just" a capacitor to pass said AC feedback signal while mixing it with the signal transduced from the air moving the diaphragm).

Sure, the feedback signal may well wiggle the diaphragm, but the extent of that is, in my humble opinion, highly debatable. I mean @OPR - how have you determined how much of the effect is electromechanical (feedback "counter-wiggling" the diaphragm) versus PURELY electrical?

There's some correlation vs. causation thing going on here, methinks...

Just thought of this - instead of feeding the feedback signal *through* the capsule itself, one could potentially feed that signal back through a similar-value capacitor back into the amplifier input, and verify the electrical-only effects, could one not?

 

[OPR]

I can't help but side with the majority here, and "still" doubt that the feedback signal electromechanically affects the diaphragm motion quite THAT much, as opposed to the purely electrical effect (using the capsule as "just" a capacitor to pass said AC feedback signal while mixing it with the signal transduced from the air moving the diaphragm).

Sure, the feedback signal may well wiggle the diaphragm, but the extent of that is, in my humble opinion, highly debatable. I mean @OPR - how have you determined how much of the effect is electromechanical (feedback "counter-wiggling" the diaphragm) versus PURELY electrical?

There's some correlation vs. causation thing going on here, methinks...

Just thought of this - instead of feeding the feedback signal *through* the capsule itself, one could potentially feed that signal back through a similar-value capacitor back into the amplifier input, and verify the electrical-only effects, could one not?

.I've provided measurements with and without the S2 Feedback and the difference was visible way up to 10. why?

.I've provided audio and the sound was obviously different. with s2 and without. why?

I've explained in as simple terms as I can.

let's leave it here and say more testing need to be done.

 

[Khron]

let's leave it here and say more testing need to be done.

I guess you're refusing (?) to acknowledge / accept that the effects of said LF feedback may very well most likely happen due to electrical effects, at least for the most part. Noone said the diaphragm motion is not affected; quite the contrary; we're only questioning the EXTENT to which that contributes to the resulting electrical signal, as opposed to the electronic feedback alone.

 

[ZHOUJUN]

The capsule has a "diaphragm" that receives sound waves and is also a vibrating element. When the capsule is working, it first receives sound waves, and the capsule's "diaphragm" vibrates under the drive of the sound waves, and then the vibration of the "diaphragm" completes the conversion of "sound" into "electricity". Therefore, the state in which the capsule receives sound waves and the control state of the diaphragm vibration are directly related to the directional characteristics and frequency response characteristics of the capsule expected by the designer.

The diaphragm vibrates under the action of external force，These can all be calculated.

 

[OPR]

I guess you're refusing (?) to acknowledge / accept that the effects of said LF feedback may very well most likely happen due to electrical effects, at least for the most part. Noone said the diaphragm motion is not affected; quite the contrary; we're only questioning the EXTENT to which that contributes to the resulting electrical signal, as opposed to the electronic feedback alone.

I'm not refusing anything... I'm happy to learn if you have something to teach us??
So, you do agree that it has some effect though very little yes? well that's a start, a little goes a long way in a delicate instrument like a mic capsule.

An explanation of the thinking that leads to your humble opinion would be enlightening? I'm all ears.

 

[soliloqueen]

i agree with opr about the principle but the biggest question i'd have is whether the tube can handle the voltage of a plosive or if it'd pass a clipped feedback back to the capsule

 

[jBam]

Ohhh... wait... I get it now ... I know I'm not particularly part of the deeper discussion here... but I'll rant off like I do anyway...

Took me a while to wrap my head around the "capsule is a capacitor" thing... I never really understood that until now. It's a strange capacitor indeed though, as it varies it's actual capacitance/charge/voltage simultaneously... Pretty wild.

So the backplate gets a hefty charge - say 60V... this charge phsyically draws the flexible diaphrams toward it to a fixed point due to electrostatic charge... more charge voltage in the backplate? it's drawn closer; less - it's drawn less close (i.e. further away from the backplate). At e.g. a fixed backplate charge, the diaphrams just sit there at a fixed point until air moves it.

When a sound pushes the diaphram closer, it's physically closer to the backplate so more electrons are pulled into the diaphram (i.e. votage increases in the diaphram and it's wired terminal/circuit etc); when air pulls the diaphram away, it's physically further from the backplate so less electrons are pulled into ther system and the voltage drops. i.e. you get a voltage curve that represents the sound that moved the diaphram back and forth. This is all the basics, but I'm glad I have my head around it (or think I do).

With a feedback system sending the a filtered signal to the backplate, the actual backplate voltage is changing. So two things are occuring simultaneously. The backplate voltage changes, so yes, the diaphram will physically move (or want to) toward or away from the backplate. It has to. But also the actual voltage which is pulling the diaphragm toward the backplate is changing too, so that also MUST effect the amount of electrons being pushed and pulled in the diaphragm too. i.e. it can't be pulled closer to the backplate with a higher voltage and NOT also have more charge effect from the additional V at the backplate... it is pulled closer AND has more charge so it's two things working together concurrently.

There's a lot going on, and that's without even thinking about the diaphragm shape and tension and other physical matters effecting the whole system.

The whole idea of both physical (diaphragm charge/movement) and electrical (capacitance) occuring has regularly been mentioned above too... it just took me a while to understand how and why.

As for "how much effect from feedback is physical and how much is electrical...?"... Well, I've spent an uneccessary amount of time thinking about this lol... and there's so many variables I can't imagine having a solid feel for it until (if) @soliloqueen or someone else gets a lazer measure out and compares diaphragm movement for a given V change at the backplate vs. a given V recieved at the diaphragm ... For now, I'm comfortable in knowing that both are occuring to some degree.

 

[OPR]

i agree with opr about the principle but the biggest question i'd have is whether the tube can handle the voltage of a plosive or if it'd pass a clipped feedback back to the capsule

I guess the feedback counteracts it before it can cause issues. It's like a time machine traveling back to to future!
Lol! Pretty ingenious design. It's amazing how close you can get to the capsule without making it pop.
I know some screamers used them.
Kurt Cobain (Nevermind )Chris Cornel (king animal )as well as well as Jazz singers getting up close and intimate. Madeline Peroux used it on (Half the perfect world) album I believe.

 

[GeorgeToledo]

Neumann has classically described the circuit as a *proximity effect* elimination, not a high pass.

I find it strange to declare things that are in fact happening do not matter.

 

[GeorgeToledo]

Took me a while to wrap my head around the "capsule is a capacitor" thing... I never really understood that until now.

They used to commonly be called capacitor microphones.

 

[ccaudle]

They used to commonly be called capacitor microphones.

And the antiquated term for a capacitor is condenser (I think because it "condenses" electrical charge), which is why the name condenser microphone became common.

I get it now

Almost. Sorry if this comes across as unnecessarily pedantic, but there are a few misconceptions that hopefully can clarify the actual operating mechanism.

it varies it's actual capacitance/charge/voltage simultaneously...

For a condenser microphone to work properly, the charge must be nearly constant. That is why such high resistances are needed for supplying the polarization voltage and for the input to the buffer amplifier. The equation describing the relationship between charge and voltage of a capacitor is:
Q=CV
where Q is the charge, C is the capacitance, and V is the voltage. You can re-arrange that to V=Q/C, and if Q is held constant, when C changes, that causes V to change.
The capacitance is defined by the distance between the conductive plates (in the microphone case the backplate and the conductive diaphragm), and the dielectric constant (which in the case of a microphone is air between the backplate and diaphragm, so effectively dielectric constant of 1).
The movement of air causes the distance between the diaphragm and backplate to vary, which varies the capacitance.

If the charge were allowed to vary as well, when the capacitance changed, the voltage would not be predictable based on just the capacitance change.

So the backplate gets a hefty charge - say 60V...

60V is a voltage, which as you can see above is one of the three terms in the equation, and is not the charge term.
Q=CV, so for a typical large diaphragm capsule would be around Q=60pF*60V->Q=3.6x10^-9 coulombs, or 3.6nC.

When a sound pushes the diaphram closer, it's physically closer to the backplate so more electrons are pulled into the diaphram (i.e. votage increases in the diaphram and it's wired terminal/circuit etc);

Nope, when a sound pushes the diaphragm closer, the amount of electrons in the capacitance stay the same, and because V=Q/C, the closer spacing causes C to become larger, which for a fixed Q causes V to become lower.
If more electrons were pulled into the capacitor, then C and Q would change until V stayed the same, which is usually not what you want with a typical voltage based buffer amp. I suppose you could try to make a really sensitive current-to-voltage amplifier (aka a transimpedance amplifier), and amplify charge going in and out of the capsule with a fixed voltage, but I think the noise constraints of that type of design would be difficult. I have never seen a microphone built that way, even though it would in theory be possible to have a fixed voltage and varying charge.

With a feedback system sending the a filtered signal to the backplate, the actual backplate voltage is changing.

Yes. Typical designs have either the backplate voltage fixed, and the signal taken from the diaphragm, or the other way, with the diaphragm voltage fixed and the signal taken from the backplate (with some switchable pattern dual-diaphragm mics, for example).
When the backplate voltage is not fixed, you still have the change in voltage across the capsule, but usually when a voltage is described with no qualifiers, there is an implicit assumption that the voltage is referenced to the circuit common/0V/"gnd" node. That is short-hand for output=capsule voltage - 0V.
In the case of feedback sent to the capsule it becomes output = capsule voltage - feedback voltage. So the voltage across the capsule is (mostly) the same, but complicated by the fact that the feedback voltage could be changing the electrostatic attraction between the diaphragm and backplate and so modifying how much the capsule capacitance varies for a particular sound pressure.

I have just been following this thread casually, but I have not seen a transfer function analysis yet that would show how the capsule voltage varies with the feedback. If when the feedback voltage increases the backplate voltage, and that causes a corresponding increase in the diaphragm voltage (because the high input impedance of the buffer amp is essentially allowing the diaphragm voltage to float), then that might result in an increase in the voltage at the tube grid even though the voltage between the backplate and diaphragm did not change. Only the voltage between the backplate and diaphragm controls the electrostatic attraction. If the entire capsule is "floating" on the feedback voltage then the feedback will change the voltage which actually gets to the buffer amp, but will not change the mechanical properties of the capsule at all. Based on the impedance differences between the drive to the backplate and the grid resistor of the buffer amp that could be the case.

 

[jBam]

Almost. Sorry if this comes across as unnecessarily pedantic, but there are a few misconceptions that hopefully can clarify the actual operating mechanism.

No!... Not at all ... in fact, it's greatly appreciated! It'd be a shame to roughly understand something only to have incorrect assumptions left floating out in the ether.

Your whole response is awesome. It's the beauty of being undebatably incorrect in a freindly forum - you gain a better understanding of something.

To be honest, it's really cool and interesting stuff. I started typing in this thread under some loose guise of understanding and designing feedback ideas into a mic and understanding why or if filtering in the mic is important vs after...

I've instead thrown myself into a deeper understanding of the fundamental design principles (which is of course far more useful and fun).