P
Peter B
Guest
When a mid side recording is decoded manually, does it matter whether the original side track or the duplicate side track is inverted, and does it matter which is panned right and which is panned left?
John Willett
Circle Sound Services
Yes it does matter.
The +ve lobe of the fig-8 should face left.
Mid mic. into channel 1 - panned centre
Side mic. into channel 2 - panned hard left
Side mic. polarity-reversed into channel 3 - panned hard right.
The trick on getting the side mic. correct is to pan centre and first set the level of the in-phase on channel 2. Then - set the main fader of channel 3 (the phase-reversed) to the same level and adjust the input gain until the mic. fully cancels and you hear nothing. Then pan channel 2 hard left and channel 3 hard right.
I hoipe this helps.
P
Peter B
Guest
Thanks for a great reply John, I was confused about it before. Is that facing left looking at the mikes or left from behind the mikes?. I'm really going to show my newbie status now and a ask how I know which side of the figure 8 is the positive side. I'm using a multi pattern mike, omni, cardi, fig 8. Is the positive side of the figure 8 the same as the 'front' when switched to cadioid pattern? Great advice on setting levels too, much appreciate your help.
John Willett
Circle Sound Services
The +ve lobe is the front of a fig-8.
In MS it faces the left when looking at the stage.
P
Peter B
Guest
Great, thanks very much. I set up on a budget about 6 months ago to record my saxophone quartet at home and gradually built up a basic recording facility working through a Lexicon interface into Cubase on my PC. I very quickly realised that there's an awful lot to learn, even for a very basic home recordist like me, so forums like this and experienced contributors like yourself are a lifeline for us novices.
I could post dozens of questions here, but I do try to work things out for myself first. The trouble with that is that there are occasions when I simply don't know whether I'm doing it right or not. In this case for instance I had the figure 8 mic the wrong way round, positive facing right, and I was panning the duplicated, inverted side track left, not right. So there are times when a question here is necessary to sort something out.
Thanks again for your help.
I could post dozens of questions here, but I do try to work things out for myself first. The trouble with that is that there are occasions when I simply don't know whether I'm doing it right or not. In this case for instance I had the figure 8 mic the wrong way round, positive facing right, and I was panning the duplicated, inverted side track left, not right. So there are times when a question here is necessary to sort something out.
Thanks again for your help.
S
sjjohnston
New member
It doesn't matter which is inverted - I mean, they're identical, right?
It somewhat matters how they're panned, but the world won't end if you get it wrong. The non-inverted track should be panned to whatever side the front of the mic was facing; the inverted the opposite way. If you (wisely) always point the figure-8 mic to the left, you'll always know which way to do it.
If you do it backwards, all that will happen is that you'll reverse the stereo image. It'll still sound about like it would've for a person standing where the mics were placed, but as if he had turned his back to the musicians (or - same thing - the musicians had reversed places, if the room is symmetrical). It would sound notably "wrong" in some situations: most obviously in a recording of a symphony orchestra, where listeners would go "huh?!" if, say, the cellos (celli to the afficiandi) were suddenly on the left. In other situations, there's no left-right standard. Even where there somewhat is (drum sets), people release recordings with them panned both ways (drummer's perspective or audience's).
EDITED TO ADD:
It also can matter (quite a bit, possibly) if you've got more mics than just the mid and side in the mix. If you, for some reason, have two stereo images turned backwards from each other (possible with, say, drum overheads and a room mic) weirdness can result.
It somewhat matters how they're panned, but the world won't end if you get it wrong. The non-inverted track should be panned to whatever side the front of the mic was facing; the inverted the opposite way. If you (wisely) always point the figure-8 mic to the left, you'll always know which way to do it.
If you do it backwards, all that will happen is that you'll reverse the stereo image. It'll still sound about like it would've for a person standing where the mics were placed, but as if he had turned his back to the musicians (or - same thing - the musicians had reversed places, if the room is symmetrical). It would sound notably "wrong" in some situations: most obviously in a recording of a symphony orchestra, where listeners would go "huh?!" if, say, the cellos (celli to the afficiandi) were suddenly on the left. In other situations, there's no left-right standard. Even where there somewhat is (drum sets), people release recordings with them panned both ways (drummer's perspective or audience's).
EDITED TO ADD:
It also can matter (quite a bit, possibly) if you've got more mics than just the mid and side in the mix. If you, for some reason, have two stereo images turned backwards from each other (possible with, say, drum overheads and a room mic) weirdness can result.
Last edited:
P
Peter B
Guest
Ah, right, so if that's the general case then more by luck than judgement I was OK with the positive side facing right and panning it right and panning the inverted track left. I thought my stereo image was OK.
I have to say that I've tried to get my head round how this actually works or rather why this actually works, but I don't know enough about how a dual diaphragm cardioid works. I think I can see why it works for a ribbon figure 8 because it works on pressure difference side to side, no pressure difference no output signal, but a dual diaphragm cardioid doesn't work like that does it.
So I've given up trying to work out why this mic technique works but one thing's for sure, it certainly does work and very well in my case for a seated sax quartet. I guess it's the engineer in me, though I'm long retired now, that just makes me want to know how/why things work.
Thanks again for your help.
S
sjjohnston
New member
I'm not an engineer, but I sometimes like to figure out how/why things work also.
On mic polar patterns: I once made the effort to figure them out, using an Excel spreadsheet.
- The omni is easy to draw, obviously. Physically, it's the pattern you get if the capsule is just a sealed chamber with the diaphgram over an opening: the diaphragm moves in response to the momentary air pressure outside of the opening (as the air is compacted or rarified slightly), and doesn't know or care where the pressure wave came from.
- A ribbon figure-8 diaphragm wiggles back and forth as the air moves back and forth to create the high/low pressure waves. If the sound wave comes directly from the side of the ribbon, it slides right past the diaphram and does nothing; if it comes directly from in front or back of the ribbon, it pushes the diaphragm back and forth to the maximum extent; if it comes from an angle, you can figure out the effect with the basic trigonometric functions.
- If you just add together the formula for an omni pattern and a figure 8, you get a perfect cardiod. There's a complete null at the back, where an omni and a figure 8 both have the same amplitude, but in opposite directions. Most real mics don't actually add up two signals, of course, but use a capsule design that combines both effects in a single diaphgarm.
- You can get a super-cardiod just by adding relatively more figure 8 and relatively less omni.
If you take the polar plot of a cardiod, and add a figure 8 turned sideways, you get a pattern with two separate lobes (each of which looks like a hypercardiod) each pointing at an angle to either side of the front of the cardiod.
Knowing the relationship between a cardiod pattern and a figure 8 pattern, you can figure out that adding together the output from two cardiod capsules pointing in opposite directions, with one inverted, creates a figure 8 pattern.
Where "F" and "R" indicate the direction a mic is facing (one being opposite of the other), and a negative sign indicates you've inverted a signal:
figF = -figR (because of how the mic works)
figF + omni = cardF (see above)
So:
cardF - cardR = (figF + omni) - (figR + omni) = figF + omni + figF - omni = 2 figF
On mic polar patterns: I once made the effort to figure them out, using an Excel spreadsheet.
- The omni is easy to draw, obviously. Physically, it's the pattern you get if the capsule is just a sealed chamber with the diaphgram over an opening: the diaphragm moves in response to the momentary air pressure outside of the opening (as the air is compacted or rarified slightly), and doesn't know or care where the pressure wave came from.
- A ribbon figure-8 diaphragm wiggles back and forth as the air moves back and forth to create the high/low pressure waves. If the sound wave comes directly from the side of the ribbon, it slides right past the diaphram and does nothing; if it comes directly from in front or back of the ribbon, it pushes the diaphragm back and forth to the maximum extent; if it comes from an angle, you can figure out the effect with the basic trigonometric functions.
- If you just add together the formula for an omni pattern and a figure 8, you get a perfect cardiod. There's a complete null at the back, where an omni and a figure 8 both have the same amplitude, but in opposite directions. Most real mics don't actually add up two signals, of course, but use a capsule design that combines both effects in a single diaphgarm.
- You can get a super-cardiod just by adding relatively more figure 8 and relatively less omni.
If you take the polar plot of a cardiod, and add a figure 8 turned sideways, you get a pattern with two separate lobes (each of which looks like a hypercardiod) each pointing at an angle to either side of the front of the cardiod.
Knowing the relationship between a cardiod pattern and a figure 8 pattern, you can figure out that adding together the output from two cardiod capsules pointing in opposite directions, with one inverted, creates a figure 8 pattern.
Where "F" and "R" indicate the direction a mic is facing (one being opposite of the other), and a negative sign indicates you've inverted a signal:
figF = -figR (because of how the mic works)
figF + omni = cardF (see above)
So:
cardF - cardR = (figF + omni) - (figR + omni) = figF + omni + figF - omni = 2 figF
Last edited:
P
Peter B
Guest
OK thanks for that. I like your algebra analogies, right up an engineer's street, and they make perfect sense.
This is all a bit of a mind bender though isn't it. I know I should just accept it at face value.......do this, do that and this is the result...........but I just can't stop myself wondering why - old disciplines die hard!!
I've read that the basics principles of a dual diaphragm condenser mic are:
Both diaphragms on and in phase = omni.
One diaphragm switched off = cardioid.
Both diahpragms on with one out of phase = bidirectional.
This pretty much matches your analysis and kind of settles my mind a bit as to how it's all put together inside the mike case.
Thanks again.
This is all a bit of a mind bender though isn't it. I know I should just accept it at face value.......do this, do that and this is the result...........but I just can't stop myself wondering why - old disciplines die hard!!
I've read that the basics principles of a dual diaphragm condenser mic are:
Both diaphragms on and in phase = omni.
One diaphragm switched off = cardioid.
Both diahpragms on with one out of phase = bidirectional.
This pretty much matches your analysis and kind of settles my mind a bit as to how it's all put together inside the mike case.
Thanks again.
John Willett
Circle Sound Services
If you set up the MS rig with the mic. pointing right you do the polarity-reverse to the left channel instead.
cobaltaudio
New member
Actually, you're pretty close; switchable pattern mics do work on exactly this principle. The only thing is that it's not an omni and a fig 8.
It's two cardioid diaphragms literally back-to-back. Well researched Peter B.
Use one to get cardioid
add together to get omni
subtract one from the other to get fig 8. (well... if you're going to be a pedant, and of course I am... yes you phase invert one and then add)
You can of course add/subtract the second capsule in different amounts to get a wide/hyper cardioid respectively.
Easy huh?