MS mics placement

[abbey road d enfer]

Maybe, that’s a language problem, but all the references (mostly in German) I have here consider also ribbon microphones to be pressure gradient transducers.

Of course, what makes the diaphragm moves is the force resulting from air particles hitting it. However you can't deny that

Velocity microphones are certainly possible and apparently have no membrane.

So-called "velocity" mics don't measure velocity directly, so they can't actually qualify as "native" velocity sensing. It would require identifying a particular air molecule or group of, and following its movements.
OTOH, dynamic and condenser mics do not measure pressure directly either; they meaure th emovement of a diaphragm submitted to such pressure.
Actually, in metrology, teh only measurements that can qualify as "direct" are length measurements; all the others evaluate a direct consequence of the property to be measured.

 

[molke]

By now, I’ve learned that the term “velocity microphone” is common in English. What irritated me a bit was the dichotomy “pressure vs. velocity”, where (e.g. according to Table 1 in the Neumann booklet) it maybe should have been either “velocity vs. excursion” or “pressure vs. pressure gradient”. In this case, the latter is presumably more relevant.

 

[GabrieleP]

First because you think taht combining two signals thar 90° apart results in a "wrong" signal. Then for mixing acoustic phase issues, due to different acoustic paths with phase-shift related to transducers.

Not true. If the diaphragm was left with no damping, that may be partially true, but in order to achiebve a flat-ish response, it must be damped, which results in mechanical phase-shift that compensates whatever phase-shift due to the transduction process.


Care to explain?

Quoting bert37: "Therefore a MS mix of ribbon and condenser will not work, as there is no correct sum and difference processing"...

This is 100% true, being MS part of the coincident family of stereophonic techniques and as such relying ONLY on phase differences between the two signals.

Quoting Handbook for Sound Engineers, chapter 6, page 505: "The electrical waveform output from the moving-coil microphone does not follow the phase of the acoustic waveform because at maximum pressure the diaphragm is at rest (no velocity). Further, the diaphragm and its attached coil reach maximum velocity, hence maximum electrical amplitude at point c on the acoustic waveform (i.e. when pressure is at null). This is of no consequence unless another microphone is being used along with the moving-coil microphone where the other microphone does not see the same 90° displacement. Due to this phase displacement, condenser microphones should not be mixed with moving-coil or ribbon microphones when micing the same source at the same distance”.

Reading "ASSertion" was disappointing. Maybe a typo...

 

[abbey road d enfer]

Quoting bert37: "Therefore a MS mix of ribbon and condenser will not work, as there is no correct sum and difference processing"...

Not true. Signals with constant 90° difference mix rather well.

This is 100% true, being MS part of the coincident family of stereophonic techniques and as such relying ONLY on phase differences between the two signals.

Deep misunderstanding here.
The stereophonic effect relies on phase differences due to different acoustic paths. Shifting phase of one of the M-S signals by 90° constant does not create any problems because it shifts the resulting L-R signals by the same 45°, keeping teh phase relationship between L and R identical..

Quoting Handbook for Sound Engineers, chapter 6, page 505: "The electrical waveform output from the moving-coil microphone does not follow the phase of the acoustic waveform because at maximum pressure the diaphragm is at rest (no velocity). Further, the diaphragm and its attached coil reach maximum velocity, hence maximum electrical amplitude at point c on the acoustic waveform (i.e. when pressure is at null).

The author does not say what type of moving coil mic he refers to.
Instant pressure is 90° apart from pressure-gradient, since the latter is the derivative of the former. So an omni mic is different than a cardioid.

This is of no consequence unless another microphone is being used along with the moving-coil microphone where the other microphone does not see the same 90° displacement.

What is the consequence? Signals shifted 90° are in the same quadrant, so do not combine destructively. Actually being of constant 90° difference, they combine rather well at 45°.

Due to this phase displacement, condenser microphones should not be mixed with moving-coil or ribbon microphones when micing the same source at the same distance”.

This conclusion is unsupported in practice.

 

[k brown]

Never knew Angels dancing on the head of a pin could make so much bloody noise.

 

[MicUlli]

Hmm, perhaps a little bit physics again?
All mic types (condenser, dynamic, ribbon) are minimum phase transducers at least in the medium frequency range (lets say 200..8000 Hz).
In a minimum phase system frequency response and phase response are strongly linked together. Constant frequency response -> 0° phase, -6dB/oct FR -> -90° phase, +6dB/oct FR -> +90° phase.
All microphones with considerably flat frequency response will therefore have THE SAME phase (or -180° if you exchange the connection wires) !!!
You can put it all together by yourself reading the "Neumann booklet" carefully.
BR MicUlli

 

[Bert37]

Hmm, perhaps a little bit physics again?
All mic types (condenser, dynamic, ribbon) are minimum phase transducers at least in the medium frequency range (lets say 200..8000 Hz).
In a minimum phase system frequency response and phase response are strongly linked together. Constant frequency response -> 0° phase, -6dB/oct FR -> -90° phase, +6dB/oct FR -> +90° phase.
All microphones with considerably flat frequency response will therefore have THE SAME phase (or -180° if you exchange the connection wires) !!!
You can put it all together by yourself reading the "Neumann booklet" carefully.
BR MicUlli

NO !

 

[GabrieleP]

Not true. Signals with constant 90° difference mix rather well.

Deep misunderstanding here.
The stereophonic effect relies on phase differences due to different acoustic paths. Shifting phase of one of the M-S signals by 90° constant does not create any problems because it shifts the resulting L-R signals by the same 45°, keeping teh phase relationship between L and R identical..

OK, I understand negotiating on the MS rules is not a problem for you. Then I guess you would easily allow a Side mic set up at distance from the Mid... that's fine. Not my personal approach.

The author does not say what type of moving coil mic he refers to.
Instant pressure is 90° apart from pressure-gradient, since the latter is the derivative of the former. So an omni mic is different than a cardioid.

The author does not say because he means all moving coils. Question: where would I find info on the 90° shift between instant and gradient pressure?

What is the consequence? Signals shifted 90° are in the same quadrant, so do not combine destructively. Actually being of constant 90° difference, they combine rather well at 45°.

See above

This conclusion is unsupported in practice.

This same conclusion is very supported in my practice. Of course "[...] should not be mixed [...]" is a little bit too much, but what the author is saying was quite revelatory and helpful for me and my work.

 

[Bert37]

NO !

Minimum phase does not mean that all kinds belonging to this group have the same (absolute) phase with respect to stimulation. For most networks or transducers a change in amplitude vs frequency goes along with a change of phase. As long as the phase is the minimum related to the amplitude change the pertinent circuit (or transducer) is called minimum phase. If there is excess phase (such as in an allpass filter) the pertinent device is non minimum phase. Signal theory allows for calculation of one to the other (in the minimum phase case) but is much to complicated to explain here.​

 

[abbey road d enfer]

OK, I understand negotiating on the MS rules is not a problem for you.

I don't negociate anything, just using basic math.

Then I guess you would easily allow a Side mic set up at distance from the Mid... that's fine.

So you guess wrong. Distance introduces a variable phase shift, and I'm very much aware of the consequences.

The author does not say because he means all moving coils.

The principle of transduction does not change what the mic "reads"

Question: where would I find info on the 90° shift between instant and gradient pressure?

Check post #96.

This same conclusion is very supported in my practice. Of course "[...] should not be mixed [...]" is a little bit too much,

There are many other factors at play, like different frequency response, impossibility to locate the two mics in the same point.

but what the author is saying was quite revelatory and helpful for me and my work.

Unfortunately, the basis for his reject is based on false assumptions.

 

[abbey road d enfer]

All mic types (condenser, dynamic, ribbon) are minimum phase transducers at least in the medium frequency range (lets say 200..8000 Hz).
In a minimum phase system frequency response and phase response are strongly linked together. Constant frequency response -> 0° phase, -6dB/oct FR -> -90° phase, +6dB/oct FR -> +90° phase.

I'm not so sure.
Actually MP does not imply that the reference phase, i.e. phase at 0Hz is 0°.

All microphones with considerably flat frequency response will therefore have THE SAME phase (or -180° if you exchange the connection wires) !!!

Phase-shift is certainly similar, but absolute phase may be starting at 0° for one, at 90° for the other, or any value dictated by physics.

 

[MicUlli]

My last try:
Consider a sound source producing a signal A = A0 sin (wt). A sound receiver is some distance away and gets the signal after the traveling time t0. So the received sound is k A0 sin (w(t + t0)) with k as a constant amplification factor. sin (w(t +t0)) may also expressed as sin (wt + phi). The claim is that phi is constant over radiant frequency w. This is only solvable if t0 = phi/w. The physical implication would be that t0 is high at low frequencies and low at high frequencies what leads to the conclusion that the sound source needs to change its distance frequency dependant. This is of course never the case and physically nonsense.
By the way: t0 can also be named group delay...
BR MicUlli

 

[abbey road d enfer]

My last try:

I don't know who you're addressiing this to, because it seems we agree on some points and not on others.

Consider a sound source producing a signal A = A0 sin (wt). A sound receiver is some distance away and gets the signal after the traveling time t0. So the received sound is k A0 sin (w(t + t0)) with k as a constant amplification factor. sin (w(t +t0)) may also expressed as sin (wt + phi). "This is only solvable if t0 = phi/w. The physical implication would be that t0 is high at low frequencies and low at high frequencies what leads to the conclusion that the sound source needs to change its distance frequency dependant.

I think you're mistaken.
I agree with t0 = phi/w, but the physical implication is that phi is very high at HF and low at LF. Of course t0 is constant.
1ms of delay is 2.pi at 1kHz, 20.pi at 10kHz, and only 0.2pi at 100 Hz.
We know that a fixed delay results in a phase angle that increases with frequency. Among other consequences is the well-known "comb effect".

Now, regarding "The claim is that phi is constant over radiant frequency w.", it pertains to the phase-shift between velocity and pressure, which is similar to the well-known phase-shift between current and voltage in a reactive component.
And, in reference to this thread's title, it has nothing to do with placement.

 

[Bert37]

Reentering the discussion I try to be very concise:

1. I have opened a can of worms (a big one !). I think that is the only point of no discussion !
   
2. My statement that condensers (omni for M) and ribbons (figure 8 for S) do not match has been confirmed by citation of reference as well as by my own measurements. It was not my invention at all.
   
3. I have never referenced „pressure gradient“, phase lag in ribbon transformers, lag in voice coils, Blumlein array, minimum phase relations or other loosely used expressions that have no correlation with the problem, but are nonsense or even BS.

Since I dislike one point miking anyway except for video, all the gossip is of no consequence for me.

 

[abbey road d enfer]

Reentering the discussion I try to be very concise:

1. I have opened a can of worms (a big one !). I think that is the only point of no discussion !

These cans o'worms often result in explanations that help demistifying, but as equally often turn into hopeless confrontations of preconceived ideas. I tend to favour the former.

My statement that condensers (omni for M) and ribbons (figure 8 for S) do not match has been confirmed by citation of reference as well as by my own measurements. It was not my invention at all.

In what respect do they "not match"? There is no physical explanation as to why two signals with a constant 90° shift result in an improper result. On teh contrary.

I have never referenced „pressure gradient“, phase lag in ribbon transformers, lag in voice coils, Blumlein array, minimum phase relations or other loosely used expressions that have no correlation with the problem, but are nonsense or even BS.

That is very clear.

Since I dislike one point miking anyway except for video, all the gossip is of no consequence for me.

I don't understand that. The subject is two-mic micing.

 

[molke]

1. My statement that condensers (omni for M) and ribbons (figure 8 for S) do not match has been confirmed by citation of reference as well as by my own measurements. It was not my invention at all.
2. I have never referenced „pressure gradient“, phase lag in ribbon transformers, lag in voice coils, Blumlein array, minimum phase relations or other loosely used expressions that have no correlation with the problem, but are nonsense or even BS.

Can we see those measurements? According to your claim, then a mix of either “pressure velocity” (dynamic omni) microphones and ribbons or omni and figure 8 condensers (Schoeps MK2/MK8, Neumann KM120/130, Sennheiser MKH20/30) is no problem?

To add to the “nonsense or even BS”, here’s a Google translation of a paragraph concerning explicitly phase in ribbon microphones in a German reference from Manfred Hibbing (microphone developer at Sennheiser):

The phase balance for sine waves is as follows. The pressure gradient is 90° ahead of the pressure (differentiation). Since the dynamic converter does not evaluate the acceleration but the speed (integration), the phase is rotated back by 90°. Overall, the output voltage of the ribbon microphone is in phase with the sound pressure, although it works as a so-called pressure gradient microphone. Basically, all microphones, regardless of their polar pattern, are always in phase with the sound pressure if their frequency responses are flat. This means that these microphones are also in phase with each other.

You can’t have a ribbon microphone (with a flat frequency response) without pressure gradient, ribbon velocity, and low resonance frequency together.

That is very clear.

Not to me, and apparently not to others.

 

[abbey road d enfer]

Can we see those measurements? According to your claim, then a mix of either “pressure velocity” (dynamic omni) microphones and ribbons or omni and figure 8 condensers (Schoeps MK2/MK8, Neumann KM120/130, Sennheiser MKH20/30) is no problem?

I don't see why it would be a problem...

To add to the “nonsense or even BS”, here’s a Google translation of a paragraph concerning explicitly phase in ribbon microphones in a German reference from Manfred Hibbing (microphone developer at Sennheiser):

As far as I can tell, this paper relates mainly to the operation principle of directional mics. I don't see that it proves an incompatibility between omni and directional mics. And I don't see anything about ribbon mics. Seems to me the text in English is not drawn from the reference document.

 

[molke]

I don't see why it would be a problem...

I don’t either.

I don't see that it proves an incompatibility between omni and directional mics. And I don't see anything about ribbon mics.

To be clear, I see no incompatibility—neither between omni and directional nor between ribbons and condensers.

Seems to me the text in English is not drawn from the reference document.

Page 2, penultimate paragraph. The English translation was just a quick Google translation. If you can read German, the original is probably better.

 

[abbey road d enfer]

Page 2, penultimate paragraph. The English translation was just a quick Google translation. If you can read German, the original is probably better.

OK, I can see it now.
I don't read German, but Google Translate does rather well.

 

[GabrieleP]

First because you think taht combining two signals thar 90° apart results in a "wrong" signal. Then for mixing acoustic phase issues, due to different acoustic paths with phase-shift related to transducers.

OK, got that.