Topic: Omnis XY

[deadheadcorey]

has anyone ever ran omnis xy

good or bad idea

[Gutbucket]

I've heard of doing that as a close vocal mic setup, may add a bit of dimension to a group singing around one mic.

For concert recording, nah, I vote bad idea.

[John Willett]

XY omnis will give you mono - basically.

You will get a little directivity at higher frequencies as the body size of the microphone starts to give some directivity.

I would never use XY omnis.

I *have* used 20cm spaced omnis on a piano, which works very well, and have also used this narrow spaced omnis on a Jecklin or Schneider Disk, which also works very well.

I *have* heard a recording that someone made with XY omnis by mistake and it was a mono sound with a bit of air (due to the increased directivity at high frequencies).

Not a good idea, really.

[Gutbucket]

"a mono sound with a bit of air (due to the increased directivity at high frequencies)."

That's a good summary.  I'd categorize it as possibly useful as an advanced spot mic technique.

[yug du nord]

close spacing with a jecklin.
spaced wide.
healy method.
all of these would be better than mono...  unless you want mono.

play around and try em all at some point!!

[Gutbucket]

^^ Yes-
A deeper question is "how much mono would you like?"  Or rather, "how low down do you want the stereo difference information to go?"  That is directly determined by the spacing between microphones for the entire frequency range, and is the only mechanism responsible for it at frequencies below about 1kHz regardless of the presence a baffle or pointing the microphones so that they have an angle between them.  A Jecklin disk baffle begins being effective somewhere around that frequency and higher, at lower frequencies the baffle effectively becomes 'invisible' to sound.  Angling the microphones apart from each other is only going to effect frequencies significantly higher than where a baffle starts to become effective. 

Perfectly coincident, parallel omnis with no angle between them are monophonic at all audible frequencies.  As the microphones are moved farther apart, stereo difference information begins to manifest above a decreasing point in the frequency scale, in direct relation to the spacing.  At the very lowest frequencies, even very widely spaced omnis are still effectively monophonic.  Near-spaced omni techniques like Healy and Jecklin are monophonic up to much a higher frequency.  Jecklin is effectively monophonic up to somewhere in the midrange.  Compared to non-baffled parallel omnis with an otherwise identical spacing, Jecklin baffled omnis will produce greater stereo difference information at all frequencies above that point.  Healy, with the same spacing is going to be monophonic to a higher frequency than Jecklin (but may sound 'airier' way up top).

There are other things going on too with changes in spacing, but this aspect is a basic and important one.

Where is the appropriate mono to stereo frequency break point?  It all depends.  For a recording of an orchestra or an audience recording of an amplified band, I prefer to have stereo difference information down to a lower frequency.  As long as the other aspects are kept well balanced, that sounds bigger, wider, open, more natural and just better to me than a narrow spacing.  But for something like a solo piano recording, the technique John mentions above using  20cm spaced omnis is going to be far more appropriate and natural sounding.  For a close shared vocal mic, where the intent is to place the vocals solidly in the center of the playback image, a mostly all monophonic recording with perhaps a bit of added air and space up top from using x/y omnis could be entirely appropriate.

[DSatz]

Localizability of sound sources is one thing (it depends mainly on cues in the midrange to upper midrange); the sense of space in a recording is something else (it depends mainly on difference information between the channels in the lower midrange and bass). Stereo recordings can offer both of these positive qualities at the same time if the circumstances allow and appropriate recording methods are followed. But coincident omnis can only ever pick up reverberant sound in mono, so they're out of the running as far as the feeling of spaciousness is concerned.

--I "co-recorded" a Mozart string quartet once, maybe 25 years ago, with someone who used a pair of Crown PZMs--a type of boundary layer microphone that was promoted as almost a "wonder drug" at one time--truly the most aggressive marketing campaign for any type of microphone that I've ever experienced. These microphones were based on miniature electret pressure transducers, which are omnidirectional when placed in free space. In a Crown PZM, the transducer was built into a short metal arm that extended over a roughly rectangle-shaped paddle; the transducer's membrane faced the paddle across a small air gap, if you can imagine that. (For now I'll refrain from mocking the technical shortcomings of this arrangement, or the wholly inappropriate usage suggestions that Crown included in its sales literature.)

The person set up the PZMs on their edges, i.e. standing the "paddles" up sideways on the floor in front of the quartet, with the front ends of the paddles touching one another and maybe a 60-degree angle between their backplanes. This separated the transducer elements by a couple of inches while the paddles themselves, thin as they were, partly isolated the microphones from one another down into the upper midrange. The setup was fairly close to the feet of the musicians.

Afterward, the other guy let me hear part of his recording over his headphones. The reverberation balance was pleasant, and the stereo effect was greater than I expected, and definitely listenable. I didn't get to hear the recording over loudspeakers, but I imagine it would have some virtues as well as some real defects, and that one could choose to pay attention to either or both (or maybe just listen to a well-performed Mozart quartet).

--best regards

[Bruce Watson]

has anyone ever run omnis xy

good or bad idea

It's not so much whether it's good or bad idea, it's just sorta pointless.

The problem here is that XY (or any co-incident technique) depends on loudness differences between the two mics to make the stereo image. The reason this is typically done with cardioid mics is because directional mics are "loudest" where you point them, and they have a more or less predictable fall off as your sound source goes farther and farther off axis of the microphone. So one mic hears the left side louder than the right side (because it's pointed at the left side), and the other hears the right side louder than the left side. These differences in loudness are the foundation of the stereo image.

Omnis, by their nature, are nearly directionless. That is, they hear everything, all around them, the same. As your sound source moves farther and farther off axis of the omni microphone, it still sounds just as loud as it did when it was on axis. Two omnis used in XY therefore don't have much difference in loudness, and therefore don't have much of a foundation for a stereo image.

[yates7592]

How about ORTF with omni's and DPA APE's?

[John Willett]

How about ORTF with omni's and DPA APE's?

ORTF is cardioids at 17cm capsule spacing at an angle of 110°  -  anything else is not ORTF !

Having said this - omnis at about a 20cm spacing does work very well for grand piano recording.

[DSatz]

John W., I agree, and you can add to the ORTF definition: ... with small, single-diaphragm, axially-addressed condenser microphones. (This from one of the retired gentleman at Radio France who invented the system.)

--best regards

[Bruce Watson]

How about ORTF with omni's and DPA APE's?

While you can put omnis in an ORTF configuration, you don't get an ORTF representation out. The sound you get out is (17cm) spaced omnis. That is, all the information that makes up the stereo sound field is from signal arrival delays (phase differences), and very little of it is from loudness differences between the microphones.

With cardioids in ORTF/NOS, you get more or less 50/50 loudness variation and phase variation.

With cardioids in XY, you get all your stereo field information from loudness differences, and none from phase differences.

[MIQ]

Hi Bruce,

I think Yates is asking more about the directionality provided by the APEs and if they would provide enough directionality to create a satisfying stereo image.  When we've discussed APEs in the past here, polar plots and on axis freq responses have been provided giving good clues about how much directionality is provided at what frequencies.  If I recall correctly the common APE sizes (30 to 50mm) create some directionality and a response boost.  The freq boost starts around 1-2kHz but the polar plots don't show much directionality until higher frequencies 5-10 kHz.  I haven't played with the DPA APEs but from the info provided I'd imagine the spaciousness would occur only at the highest frequencies in addition to the directionality created by the arrival time differences created by the mic separation.

I would also like to mention that calling the arrival time differences provided by spaced microphones a "phase" difference is not quite right.  The brain is using interaural arrival time differences (along with level differences if they exist) between ears to determine the sound source direction.  In fact if you play continuous sine waves with a phase shift between the two signals (left/right) you will not get the imaging you would like.  Or more commonly, you could have the same signal in both channels but the polarty of one flipped so the "phase" relationship is 180 degrees different but the image will be the "out of phase" hole in the middle you get when you accidentally hook up one speaker's + and - terminals backward.  That is a lot different than signals with a difference in arrival time.  Not trying to bust your cojones just wanted to clarify that phase and delay are not quite the same things, especially when the signals are not pure sine waves.

Miq

[dabbler]

I've did X-Y omnis (DAB > DA-120F) once for stack taping with them mounted at the top of my head.  I figured it'd get me closer to the tweeters of the stack I was recording, but yeah, the result was mono and painful to listen to on headphones.

I was stack taping from the right and angled left facing the stage, so both mics captured the excited crowd sounds from behind me the same.  If I were to record from that position again, I probably would've tried the left channel at the top of my head and the right in my normal position above my ear.

I'll get around to uploading the recording to LMA at some point.  I'm thinking
I'll fake a stereo image by delaying the mids/highs in one channel by 30ms or so?  I need to read up on faking stereo...