The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Tags: ,

MS stereo does not collapse to mono at LF Condenser Microphones
Old 2 weeks ago
  #1
Lives for gear
 
Yannick's Avatar
 

Thread Starter
MS stereo does not collapse to mono at LF

In my 40 m2 mastering room (which is ridiculously small for such test), I played back three test tones (26Hz, 50Hz, 75Hz) as can be seen on the screen shots.

I took a screen shot with the panner hard left, center and hard right.

You can see the results on the goniometer, and on the level meter of the MS plugin (L = M, R meter = S, if side is as loud as mid : this means hard pan)
(in the 26Hz test you can see the null of the MKH30 is a whopping -30 dB versus the mid mic)

In this test the mid mic was a MKH40.

At the same time you can see that LF rolloff of these directional mics at 26 Hz is non-existant in this small room. On the contrary, the 26 Hz measurements are louder than the 75 Hz measurement. (and no, I do not have an enormous LF boost on my monitoring).

Conclusion:

a. good directional mics give useful output well-below 30Hz. 10Hz I cannot test. Out of experience I know these MKH mics go at least to 20Hz.
b. stereo separation is good at least to 26 Hz. Probably in a big hall, with the right source, we could demonstrate it goes on until DC. If the output of the side mic falls off we could try compensating with a boost below 30Hz.

The main point is the side mic keeps its polar response at 26Hz. I could try testing later with a MKH800 (in fig8) as side mic to see what happens.
Attached Thumbnails
MS stereo does not collapse to mono at LF-2018-06-05-1-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-2-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-0-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-4-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-5-.jpg  

MS stereo does not collapse to mono at LF-2018-06-05-3-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-6-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-7-.jpg   MS stereo does not collapse to mono at LF-2018-06-05-8-.jpg  
Old 2 weeks ago
  #2
Lives for gear
 

Thanks for doing these tests Yannick, but I cannot discern the graph text or understand what plots are showing what. Can you show the de-encoded (ie L and R) output of hard left source and hard right source.
Old 2 weeks ago
  #3
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Sorry, somehow the screenshots are not really clear, and the order of the uploads is not ok.

Nr 0 is 26 hz left, nr 1 is 26 hz center, nr 2 right
Nr 3 to 5 are 50 hz
Nr 6 to 8 are 75 hz
Old 2 weeks ago
  #4
Lives for gear
Yannick, in your experience are there any directional mics other than MKH which have extended bass...as it seems to be generally received wisdom that directional mics lose out on the extended bass that pressure omnis have ?

If it's only the MKH that have this, is it part of the internal EQing that's inherent to the MKH RF way of operating, and absent in other makes of directional mics ?
Old 2 weeks ago
  #5
Gear Addict
 

Quote:
Originally Posted by Yannick View Post
I could try testing later with a MKH800 (in fig8) as side mic to see what happens.

To speculate, just for fun, I would presume the MKH800 won't hold its pattern quite as well at higher frequencies but will probably be unnoticeably close at these low frequencies.

With a single transducer fig-8, like the MKH-30 or (especially) a ribbon, the distance between each side is tiny (only as thick as the transducer), which means the signal reaches both sides at almost exactly the same time.

When the pattern is created by using two transducers, the distance between those transducers will be the determining factor in the time arrival difference between them, and the higher it is, the more the pattern will be imperfect at various frequencies, as the rear signal will not align precisely to 180-degrees but something else depending on frequency. In fact you should be able to predict exactly how the pattern will hold up just by running some math on the distance between diaphragms.

I would expect pattern control to be worse at *higher* frequencies because of this (it doesn't take much distance/time difference to be out-of-phase to a 10k signal vs a 100Hz signal). These low frequencies will still be roughly 180 (or 179... or 178...) degrees out of phase even with a bit of time difference.

So MS stereo using dual-diaphragm mics like the MKH800 might actually get a funky stereo image at the highest frequencies, not the lowest.

Plus there is the distance between the M and S mics to consider. I would again expect this distance to have more effect on higher frequencies than lower ones.

This is my (limited) understanding, anyway!
Old 2 weeks ago
  #6
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Old 2 weeks ago
  #7
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by studer58 View Post
Yannick, in your experience are there any directional mics other than MKH which have extended bass...as it seems to be generally received wisdom that directional mics lose out on the extended bass that pressure omnis have ?
My Royer SF series mics go very low as well. The Samar audio ribbons are linear to 20Hz (following another heated Gearslutz discussion a couple of years ago, I contacted Samar directly and got a clear response that their measurements are very honest, meaning at a decent distance from the source, so you do not need to record your orchestra from 6 inches distance ...)

As I just wrote in another thread, the octavas seem to be linear to 20 Hz (so price isn't the problem ...)

If I remember correctly my Pearl DS60 was also reasonable flat below 40 Hz. (a quad cardioid mic this).

It almost seems amazing that neither Schoeps nor Neumann seem to be able to do a small diaphragm cardioid with useful bass response.

Josephson CS4 cardioids are not flat below 100Hz, but have far less rolloff.
Old 2 weeks ago
  #8
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by dasbin View Post
So MS stereo using dual-diaphragm mics like the MKH800 might actually get a funky stereo image at the highest frequencies, not the lowest.

Plus there is the distance between the M and S mics to consider. I would again expect this distance to have more effect on higher frequencies than lower ones.

This is my (limited) understanding, anyway!
You are very right. I always take great care during positioning, choosing which source needs best HF alignment to the MS array.

In post, I shuffle around 1-2-3 samples with the M versus the S mic, to further improve imaging, and minimalize jumping around of some HF. This can have quite a dramatic stabilizing effect on mainly the hard left and hard right sources.
Old 2 weeks ago
  #9
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by dasbin View Post
To speculate, just for fun, I would presume the MKH800 won't hold its pattern quite as well at higher frequencies but will probably be unnoticeably close at these low frequencies.
I just did this test with a MKH800 on fig8 as side mic, MKH40 as mid mic. The lissajous graph follows my pans very nicely down to 19 Hz.

Meaning the MKH80/800/800 twin multipattern mics hold their fig8 response perfectly to below 20 Hz.

The MKH800 holds its fig8 pattern perfectly to 10KHz, at 16KHz there is a slight boost (+4 dB) at angles between 30 and 80 degrees, so you'll get slightly broader stereo at those angles if you use this mic as a side mic.

But still no mono bass !
Old 2 weeks ago
  #10
Lives for gear
 
jimjazzdad's Avatar
Yannick, thanks for posting this. I regularly record a youth orchestra with an MS pair. They set up 'North American fashion', with the celli, basses and low brass all in ranks from near-centre to quite far stage left. I have always found that the low frequencies predominate in the right channel (i.e. bass sounds emanate from stage left) and that the stereo image sounds accurate...to me anyway. Its nice to see confirmation that MS can produce real stereo imaging at all frequencies.

Last edited by jimjazzdad; 2 weeks ago at 03:32 PM..
Old 2 weeks ago
  #11
Lives for gear
Quote:
Originally Posted by dasbin View Post
Plus there is the distance between the M and S mics to consider. I would again expect this distance to have more effect on higher frequencies than lower ones.
I was thinking the same thing. I'd love a way to explore this. When you measure the distance between capsules, is it edge to edge, center to center, etc? There is that sound in MS that makes me think the funky stuff is pretty audible (even with MKH mics I have), as much as I do enjoy using it on occasion.
Old 2 weeks ago
  #12
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by leddy View Post
I was thinking the same thing. I'd love a way to explore this. When you measure the distance between capsules, is it edge to edge, center to center, etc? There is that sound in MS that makes me think the funky stuff is pretty audible (even with MKH mics I have), as much as I do enjoy using it on occasion.
I do not measure. Worst case the caps are 3.5 cm apart (the distance the piggy back puts them apart). That is 4.5 samples at 44.1KHz. When setting up an MS array, one of course has to take care the important part of the source is time-aligned to the source. This means when recording a solo performance, I do not have to delay anything. Sometimes, eg. a string quartet, it is about 1-3 samples (at the most) (I aim for the best compromise between middle players and edge players). That is for me the biggest advantage of 88.2K - the samples correspond to 3.85 mm only, instead of 7.7 mm. So I can lose more time fine tuning

As a side note, when Sennheiser was experimenting with the multipattern, they had a prototype with a MKH20 vertically in a MKH80 body, with a MKH30 capsule right underneath. The distance was minimal, and they wanted to test if the use of a "real" microphone + fig8 could yield superior results, vis-a-vis a double membrane capsule.

They abandoned in favour of the double membrane - which show something.

Alas, they could/should have developed a MKH800 twin with a MKH30 capsule mounted right beneath the double capsule. That would further reduce the 3.5 cm to a bit less than 2.5 cm and have the great advantage the main mic would be nearly invisible.

We will probably see that MKH800 triple stereo version when it is released alongside the MKH8030
Old 1 week ago
  #13
Lives for gear
 

Yannick, what are thee three graphical elements in your diagrams. There is an outer white boundary, a shaded orange centre and then a orange outline in the centre? Is this representing some sort of peak hold over the frequency sweep or something else?
Old 1 week ago
  #14
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Orange region: peak: current stereo image
White line: Peak-Hold: maximum value
White gradient: Energy: energy of stereo image

The screenshot I posted are right after I refreshed the goniometer.
You can also look at the level meters in the MS plugin.

Last edited by Yannick; 1 week ago at 07:00 AM.. Reason: Copy paste havok
Old 1 week ago
  #15
Lives for gear
 

Thanks for doing these tests Yannick, but I am uncertain of the conclusions.

Quote:
Originally Posted by Yannick View Post
Conclusion:

a. good directional mics give useful output well-below 30Hz. 10Hz I cannot test. Out of experience I know these MKH mics go at least to 20Hz.
This is known, especially the MKH series. But the cardioid will be close to an omni pattern, and the fig8 will still show its classic directivity, albeit at a lower output.

Quote:
b. stereo separation is good at least to 26 Hz.
All due to the fig 8 directivity maintaining some L/R discrimination after decoding, but its not a typical fig8. Cardioid probably not useful anymore at those frequencies.

Quote:
Probably in a big hall, with the right source, we could demonstrate it goes on until DC.
I doubt discrimination of L and R. Not sure of the point being made here.

Quote:
The main point is the side mic keeps its polar response at 26Hz.
Never in dispute. This is what is saving the array with these MKH. The 30 is an amazing mic.
Old 1 week ago
  #16
Lives for gear
Well done David
It’s taken you 10 years to admit it !
Old 1 week ago
  #17
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by Rolo 46 View Post
Well done David
It’s taken you 10 years to admit it !
And then still, he questions the conclusion. The thread title is confirmed, and is valid even if the cardioid would collapse to omni. If we boost the side mic gently below 30 Hz, the stereo angle stays constant easily, even if we get an omni mid pattern.

Stil, I remain unconvinced by David’s assumption that the other patterns cannot keep their pattern below a certain frequency.

In my experience it is quite clear that they do, does not matter if it is a MKH40 or MKH800 multipattern mic.
Old 1 week ago
  #18
Lives for gear
 

Quote:
Originally Posted by Rolo 46 View Post
Well done David
It’s taken you 10 years to admit it !
Thanks for following me closely for 10 years Roger. Respect.
Old 1 week ago
  #19
Lives for gear
 

Quote:
Originally Posted by Yannick View Post
And then still, he questions the conclusion.
Not only me.

Quote:
Stil, I remain unconvinced by David’s assumption that the other patterns cannot keep their pattern below a certain frequency.
Its not an assumption. Take a look at the MKH-40 plots.

Quote:
In my experience it is quite clear that they do, does not matter if it is a MKH40 or MKH800 multipattern mic.
Take a look at some other mics. The thread is about MS in general, I thought, or is it yet another Sennheiser fan post?
Old 1 week ago
  #20
Lives for gear
 

In a nutshell, the argument and observation is that at LF, the mid loses directivity and the fig 8 loses output. The Senn fig 8 have EQ boost at LF (verification?) to help mitigate against the second part of the issue.
Old 1 week ago
  #21
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by David Spearritt View Post
Its not an assumption. Take a look at the MKH-40 plots.
Take a look at some other mics. The thread is about MS in general, I thought, or is it yet another Sennheiser fan post?
Show me one MKH40 plot where LF becomes omnidirectional. I cannot find it...

I do not care if some cardioid mics lose their directivity at LF. Then they are flawed, and I would not use them.

There is a major problem in your logic. If all cardioid capsules become omni at very LF, that would mean they become pure pressure microphones.
Which in turn would mean there would be no LF rolloff at all.
See the problem with your assumption ? (which is quite widespread btw)

Also, let us not forget Sennheiser originally designed the entire MKH series with MS stereo capture in mind. In a way, this thread cannot be NOT about the MKH series.
Old 1 week ago
  #22
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by David Spearritt View Post
In a nutshell, the argument and observation is that at LF, the mid loses directivity and the fig 8 loses output. The Senn fig 8 have EQ boost at LF (verification?) to help mitigate against the second part of the issue.
What do you mean by LF ? Below 5 Hz ? Below 40 Hz ?

In my screenshots you can clearly see the separation is at least as good at 26Hz as 50/75 Hz.

At which point does "mono" subbass become an issue ? In the real world you are not going to find many playback rooms/hifi systems that have
a. meaningfull output below 30Hz
b. full range speakers or stereo subs
c. good enough (very low) bass treatment

It seems to me you are really hammering on one point (which is very probably not correct, or at least far less severe than you implicate), which in the end does not even matter.

The MS becomes omni myth originally was about sub 200 Hz frequencies if I recall...
Old 1 week ago
  #23
Lives for gear
 
Yannick's Avatar
 

Thread Starter
This is the only document where I could find some measurements down to 20Hz:

http://www.sanken-mic.com/upload/pdf/en/cu-41.pdf

On page 2 you can see that this Sanken mic performs worse then a MKH40 at 250Hz.
But, if you look at the FR plots, you can see it still has 10-12 dB less output from the back null then on-axis. Still far from an omni mic.

Now, if we take the example of a MKH40/30 MS setup, and assume the MKH40 becomes a sub-cardioid (-10dB attenuation at the back) at 20-30 Hz, then (at equal M and S levels) the recording angle changes from 106 degrees to 116 degrees at 20-30 Hz. This means you compress the stereo image a whopping 10% !

If you want to call that mono, OK.
Even if the cardioid becomes a true omni, we still have a stereo angle of 180 degrees. Now if we boost the S level only slightly, we go back towards our original 106 degrees rather quickly, as at S = +3 dB, the stereo angle is already narrower at 90 degrees !

So, if the rolloff of MKH40 and 30 is about equal at LF, we only need to boost the side level 1-2 dB below 50Hz to maintain the recording angle.

That is still assuming the polar response degrades as badly as David assumes !

EDIT: for reference: Educate me on old Senheiser mics?
specifically : https://sound-link.co.uk/docs/MKH%20Brochure%202004.pdf
towards the end of the document you can check the recording angles.

from the same document (MKH80) : • Lack of colouration in both the direct and dirruse sound fields due to the
accurate pick-up patterns which are virtually frequency-independent down to the lowest
frequencies, and the minimal interference caused by the microphone body and sound
inlet

Last edited by Yannick; 1 week ago at 10:25 AM..
Old 1 week ago
  #24
Lives for gear
 

Quote:
Originally Posted by Yannick View Post
Show me one MKH40 plot where LF becomes omnidirectional. I cannot find it...
Sorry, you are correct, I was looking at one of the 800 plots I think.

Quote:
There is a major problem in your logic. If all cardioid capsules become omni at very LF, that would mean they become pure pressure microphones.
Which in turn would mean there would be no LF rolloff at all.
See the problem with your assumption ? (which is quite widespread btw)
Yes, I agree with that logic.

Its good to see your measurements, although LF from loudspeakers in a small studio is not an ideal test in any way. I think the 90 deg off axis is the important angle to look at for directivity of stereo. I am uninterested in the rear response of the mid mic, if cardioid.

I will never argue that Blumlein doesn't maintain its directivity at LF, although with some mics there is no output.

I have been doing many similtaneous MS and spaced pair recordings in a very good acoustic and hear a more mono decay with MS than with the spaced pair. Back to the drawing board.
Old 5 days ago
  #25
Lives for gear
 
Plush's Avatar
Sorry, but that room is too small to let low freq. tones fully develop. Surely at least the bottom 1.5 octaves are absent from the test. So the real low frequencies are stunted here. They cannot be captured by the mic.

All cardioids have relatively poor response below 40 Hz.
Old 5 days ago
  #26
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Quote:
Originally Posted by Plush View Post
Sorry, but that room is too small to let low freq. tones fully develop.
My first axial mode is 23 Hz thank you.

Quote:
Originally Posted by Plush View Post
Surely at least the bottom 1.5 octaves are absent from the test. So the real low frequencies are stunted here. They cannot be captured by the mic.
All cardioids have relatively poor response below 40 Hz.
Not as shown by this test. As pointed out in my OP, nothing was boosted. You forget that - although the measurements as provided by Sennheiser show a rolloff below 40 Hz - the rolloff of these mics is not drastic.

Unfortunately I need to conclude you are calling me a fraud and a lyer on a public forum. It saddens me you have reached a new low, Hudson.

I reported your post 5 minutes ago.
Old 5 days ago
  #27
Lives for gear
 
Plush's Avatar
A 40 Hz. tone has a wavelength of around 28 ft. A 20 Hz. tone of over 56 feet to fully develop. So any tones played in a room of your dimensions cannot fully develop. They bounce around and the leading edges smear into each other.

Mine is a scientific complaint, not a personal one.
Old 5 days ago
  #28
Lives for gear
 

you don't need a room the size of the wavelength of the lowest frequency to record or play back very low sound - if it's 'bouncing back', you should work on your room acoustics/treatment...
Old 5 days ago
  #29
Lives for gear
 
Plush's Avatar
Of course it is bouncing back because the room is too small for the wavelength to properly reproduce.


If you are making a "test" you have to be able to reproduce that low frequency. The room cannot do it and the monitors cannot do it.

You can "play back" something but you cannot offer a legitimate test.

Why are organ low pedal notes played in a cathedral? Because there the low tones can be heard with maximum impact. To record those notes the mics have to be back towards the endpoint of the lowest frequency's travel.
Old 5 days ago
  #30
Lives for gear
 
Yannick's Avatar
 

Thread Starter
Plush, how big is your control room ? 1500 m3 ?
Are you seriously going to argue that a well-treated 120 m3 room cannot of full-range playback ?

Furthermore, you are missing the point entirely. If my MS array can show the same stereo separation at 26 Hz as at 75 Hz IN THIS ROOM, that more than proves that it works in a large hall.

How would you explain I do not get mono results at 26 Hz ? Which system in the known physics would apply ?

You know what you just did. You are not talking about small room acoustics, you called me out openly as a lyer and suggested I faked the results.

Now, this was just a quick test to demonstrate I could show something useful in bad conditions.

If you insist I could try to get the very same information out of some IRs I made in a 8000 m3 hall a decade ago. If I can find the source files.
But if I show the same result, what are you going to do ?

At the very least you will stop blabbering about my control room which is too small for your taste (this must be the 2nd or 3rd thread where you mention this).

I will ask one time: what is the cutoff frequency of your great English monitors, and what are the dimensions of your room ?
Can you produce a similar test in your mastering room, with a AB omni setup ?
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump