MS mics placement
- Thread starter Teejay41
- Start date
Help Support GroupDIY Audio Forum:
Would you mind a brief, schematic recap? Thank you.
BertBert37
Member
OK, I will take the pain to repeat my setup with a SDC Omni condenser and a very close placed (prototype) ribbon and copy the oscilloscope patterns for 250, 1000, and 3000 Hz to make sure that I am not at the edge of the transmission range. (The ribbons reproduce up to 30 kHz). However this will take some time because I have more important tasks on my desk.
So far only some theory:
Assume to signals of the same amplitude and frequency that undergo the MS procedure to form L and R:
For equal phase: M+S ---> + 6 dB, phase 0, M-S ----> - inf. dB, phase undefined
For 90 deg. phase: M+S ---> + 3 dB, phase +45 deg., M-S: ---> + 3 dB, phase -45 deg.
---> no processing ocurred !
PS: One point means coincident mics !
So far only some theory:
Assume to signals of the same amplitude and frequency that undergo the MS procedure to form L and R:
For equal phase: M+S ---> + 6 dB, phase 0, M-S ----> - inf. dB, phase undefined
For 90 deg. phase: M+S ---> + 3 dB, phase +45 deg., M-S: ---> + 3 dB, phase -45 deg.
---> no processing ocurred !
PS: One point means coincident mics !
I dpon't get this. What is "phase 0"? Did you mean 180°?
This is debatable, since all along the mutiple possibilities for M+/-kS where M and S are colinear, phase is 0, so mathematically, the probablity for phase being different than zero at a singular point is feeble. Now, in practice, M and S contain noise, which would result in undetermined phase correlation. Different POV result in different conclusions.
I agree. The angle between M+S and M-S is still 90°.
?
BertBert37
Member
Phase is always the angle between 2 vectors. These are the 2 signals I had defined. But the level of discussion has now reached a point that I quit !
"I quit" is the ultimate resource for one that does not manage to prove their point, often because it's unprovable.
The only thing that matters is that the result sounds amazing 
Please keep the conversation going with your experiences of recording organs and classical music, it was a lovely conversation you guys had! Don't hesitate to post recordings you've done as well!
Please keep the conversation going with your experiences of recording organs and classical music, it was a lovely conversation you guys had! Don't hesitate to post recordings you've done as well!
BertBert37
Member
I respect a moderators freedom to make personal offences due to his large number of messages as long as it is underlined by technical knowledge.
Therefore I make a very last try – starting by GabrieleP's statement that the diaphragm is moved by air pressure. Most evident it is true – but it does not matter. The point is how this mechanical movement is transformed into an electrical signal, which is the basic task of a microphone.
Assuming that very same diaphragm is part of a condenser microphone in one case and part of a dynamic microphone, such as moving coil, or ribbon in the other case. Although it is difficult to achieve this very same structure in practice, let's take it as a theoretical experiment:
In the condenser case, the resulting signal follows amplitude (elongation) by change of capacitance in the capsule.
In the dynamic microphone case (moving coil or ribbon) the resulting signal follows the velocity of the movement by law of induction. Thus if the exciting sound wave is of harmonic nature (sin or cos) the elongation is the result of integration from velocity, and apart from a constant, is cos or sin, and exhibits a phase shift of 90 deg. So far, that is basic math and proves the fundamental incompatibility for a combination of condensers and dynamic microphones in a MS-configuration, which by its nature is phase sensitive. Thus the problem is simply a matter of the conversion method, and not at all by any effects on the acoustical side.
How far this basic incompatibility can be overcome by structural manipulations in design of different diaphragms for the 2 types is a matter of arguing and mostly lacks a stringent base for proof.
PS: The citation of Molke „German reference" from Manfred Hibbing (microphone developer at Sennheiser) concentrates most on proximity effect vs far field and spherical wave fronts. All other points also are limited to the acoustical part and therefore are outside the basic MS discussion. Other than most forum members I am able to read and understand german as it is my native language.
Therefore I make a very last try – starting by GabrieleP's statement that the diaphragm is moved by air pressure. Most evident it is true – but it does not matter. The point is how this mechanical movement is transformed into an electrical signal, which is the basic task of a microphone.
Assuming that very same diaphragm is part of a condenser microphone in one case and part of a dynamic microphone, such as moving coil, or ribbon in the other case. Although it is difficult to achieve this very same structure in practice, let's take it as a theoretical experiment:
In the condenser case, the resulting signal follows amplitude (elongation) by change of capacitance in the capsule.
In the dynamic microphone case (moving coil or ribbon) the resulting signal follows the velocity of the movement by law of induction. Thus if the exciting sound wave is of harmonic nature (sin or cos) the elongation is the result of integration from velocity, and apart from a constant, is cos or sin, and exhibits a phase shift of 90 deg. So far, that is basic math and proves the fundamental incompatibility for a combination of condensers and dynamic microphones in a MS-configuration, which by its nature is phase sensitive. Thus the problem is simply a matter of the conversion method, and not at all by any effects on the acoustical side.
How far this basic incompatibility can be overcome by structural manipulations in design of different diaphragms for the 2 types is a matter of arguing and mostly lacks a stringent base for proof.
PS: The citation of Molke „German reference" from Manfred Hibbing (microphone developer at Sennheiser) concentrates most on proximity effect vs far field and spherical wave fronts. All other points also are limited to the acoustical part and therefore are outside the basic MS discussion. Other than most forum members I am able to read and understand german as it is my native language.
BertBert37
Member
I respect a moderators freedom to make personal offences due to his large number of messages as long as it is underlined by technical knowledge.
Therefore I make a very last try – starting by GabrieleP's statement that the diaphragm is moved by air pressure. Most evident it is true – but it does not matter. The point is how this mechanical movement is transformed into an electrical signal, which is the basic task of a microphone.
Assuming that very same diaphragm is part of a condenser microphone in one case and part of a dynamic microphone, such as moving coil, or ribbon in the other case. Although it is difficult to achieve this very same structure in practice, let's take it as a theoretical experiment:
In the condenser case, the resulting signal follows amplitude (elongation) by change of capacitance in the capsule.
In the dynamic microphone case (moving coil or ribbon) the resulting signal follows the velocity of the movement by law of induction. Thus if the exciting sound wave is of harmonic nature (sin or cos) the elongation is the result of integration from velocity, and apart from a constant, is cos or sin, and exhibits a phase shift of 90 deg. So far, that is basic math and proves the fundamental incompatibility for a combination of condensers and dynamic microphones in a MS-configuration, which by its nature is phase sensitive. Thus the problem is simply a matter of the conversion method, and not at all by any effects on the acoustical side.
How far this basic incompatibility can be overcome by structural manipulations in design of different diaphragms for the 2 types is a matter of arguing and mostly lacks a stringent base for proof.
Therefore I make a very last try – starting by GabrieleP's statement that the diaphragm is moved by air pressure. Most evident it is true – but it does not matter. The point is how this mechanical movement is transformed into an electrical signal, which is the basic task of a microphone.
Assuming that very same diaphragm is part of a condenser microphone in one case and part of a dynamic microphone, such as moving coil, or ribbon in the other case. Although it is difficult to achieve this very same structure in practice, let's take it as a theoretical experiment:
In the condenser case, the resulting signal follows amplitude (elongation) by change of capacitance in the capsule.
In the dynamic microphone case (moving coil or ribbon) the resulting signal follows the velocity of the movement by law of induction. Thus if the exciting sound wave is of harmonic nature (sin or cos) the elongation is the result of integration from velocity, and apart from a constant, is cos or sin, and exhibits a phase shift of 90 deg. So far, that is basic math and proves the fundamental incompatibility for a combination of condensers and dynamic microphones in a MS-configuration, which by its nature is phase sensitive. Thus the problem is simply a matter of the conversion method, and not at all by any effects on the acoustical side.
How far this basic incompatibility can be overcome by structural manipulations in design of different diaphragms for the 2 types is a matter of arguing and mostly lacks a stringent base for proof.
PS: The citation of Molke „German reference" from Manfred Hibbing (microphone developer at Sennheiser) concentrates most on proximity effect vs far field and spherical wave fronts. All other points also are limited to the acoustical part and therefore are outside the basic MS discussion. Other than most forum members I am able to read and understand german as it is my native language.
Thank you for not leaving the discussion, even though it may have become a bit lengthy.
Just out of curiosity, I quickly put a ribbon and small omni condenser near(ish) coincident 0.5 m in front of a speaker. This is just a digital scope (and somehow I managed to get by without ever using one before, so bear with me).
So far, I agree. What do you think happens to the output phase after you choose some mass and tension of the diaphragm and put your fully finished microphone in a sound field?
Just out of curiosity, I quickly put a ribbon and small omni condenser near(ish) coincident 0.5 m in front of a speaker. This is just a digital scope (and somehow I managed to get by without ever using one before, so bear with me).
Attachments
Look pretty close phase wise to me. Maybe the phase lead in the ribbon is offset by its phase lag in the ribbon’s transformer?
I agree.
Everything is in pages 26 & 27 of the Boré/Peus "Microphones" book. They clearly state that in order to have a flat response, dynamic mics must be heavily damped, which applies mechanically a -90° phase shift that counteracts the +90° phase-shift due to transduction.
Basic math that neglects the mechanical effects is not science.
So far, your point has not been proved efficiently.
You are confusing phase response due to transduction and phase response due to acoustical paths.
Not a matter of "arguing". The fact that a dynami mic has a relatively flat response just shows that the simplistic approach of considering only dPhi/dt is incomplete.
I already agreed with that.
I'm far from being a native German speaking individual, but I'm quite able at reading a Google translation, as I'm sure many other members are.
Is the green trace the condenser?
Yes, if I remember correctly. Since the question was relative phase, I didn’t take any notes. And I wouldn’t over-interpret these plots; the quick setup was far from perfect. Actually, as far as I understand, any deviation from constant amplitude response may give you different phase response as well. You don’t need any transformer for that.
This kinda reminds me of this discussion:
https://groupdiy.com/threads/why-you-should-never-use-multi-pattern-mics.81157/
There are two schools of thoughts. One - if it sounds good, it's good, and then you can throw away all I stated in the thread above, and what Bert is saying.
Then, you can go by rigorous measurements, empiricals, theory, and try to figure out what exactly is going on.
Bert's claim is way beyond my current knowledge, or i just don't get what he is claiming.
One thing i know for sure is that ribbon/omni M/S setup is extremely complex, it will depend on frequency, angle of incidence, distance from the source, acoustics of the room, how the omni is placed. End adress omnis are not ideally omni-directional. They are still slightly directional, especially at close proximity. There will be diference depending on how you position the omni.
Ideally, you need identical frequency response for omni and F8. Otherwise the whole thing falls apart, and you get discrepancies throughout the frequency range.
https://groupdiy.com/threads/why-you-should-never-use-multi-pattern-mics.81157/
There are two schools of thoughts. One - if it sounds good, it's good, and then you can throw away all I stated in the thread above, and what Bert is saying.
Then, you can go by rigorous measurements, empiricals, theory, and try to figure out what exactly is going on.
Bert's claim is way beyond my current knowledge, or i just don't get what he is claiming.
One thing i know for sure is that ribbon/omni M/S setup is extremely complex, it will depend on frequency, angle of incidence, distance from the source, acoustics of the room, how the omni is placed. End adress omnis are not ideally omni-directional. They are still slightly directional, especially at close proximity. There will be diference depending on how you position the omni.
Ideally, you need identical frequency response for omni and F8. Otherwise the whole thing falls apart, and you get discrepancies throughout the frequency range.
When I set up mics for drums I use a series of impulses like rimshot on the snare or stick hits, I record that into the DAW and look at the time difference between the snare mic and the other mics. After recording I can adjust the track position of close mics to coincide (or use a variable delay on the early tracks to coincide with the latest close mic) using the impulse at the beginning of the recording as a guide. It’s done to achieve the best sound - no math involved.
Possibly a good test for mic phase comparison for phase might be to line them all up - ribbon, dynamic, condenser cardioid and omni and point them all, at the same distance, from an impulse source - something like a ruler slap on a table top - recording this on separate tracks and play them back through a twin trace scope - two at a time having the earliest as a guide on trace 1 as trigger.
Edit: you need a series of impulses for the scope to work, but you can also just look at the expanded waveforms in the DAW with time as the scale rather than bpm to note the difference.
Possibly a good test for mic phase comparison for phase might be to line them all up - ribbon, dynamic, condenser cardioid and omni and point them all, at the same distance, from an impulse source - something like a ruler slap on a table top - recording this on separate tracks and play them back through a twin trace scope - two at a time having the earliest as a guide on trace 1 as trigger.
Edit: you need a series of impulses for the scope to work, but you can also just look at the expanded waveforms in the DAW with time as the scale rather than bpm to note the difference.
Last edited:
moamps
Well-known member
Thank you very much for your effort. Could you repeat the measurement at a frequency of 100Hz?
moamps
Well-known member
That's nothing new here lately. Now JR will jump in and close the thread. Before that, he will tell some of his technical experiences from 30-40 years ago.
moamps
Well-known member
Same here. In the end I stuck with using the Neumann USM69 for the M/S. The last thing I tried for stereo recording and what I really liked was the B&K omni microphones in the Jecklin disc configuration.
