Trick for determining phase of two mics?

[cellardweller]

I've read the Aaron Carey "multi-mic recording" article backwards and forwards now. I'm not sure how it relates to my "setup" (dmp3/MR8), but as it's popular to say "that's another thread"!

Let me ask this:
If I place one mic close to the grill, obtain a constant noise (hum), bring that noise up to "0", then mute that mic.

Place a second mic equidistant as first, bring it up to "0"...

Reverse phase on second mic...


Okay, now that they are reversed in phase, can I just move the second mic until I find the spot where they "cancel each other out" ,that is, the place where the noise (hum) is at it's quietest? This being the location where the two mics are most "in phase" together?

Please and Thank You in advance!

 

[cellardweller]

Here is the article...

Aaron Carey article

For anyone interested.

 

[MadAudio]

I'm replying before reading the link you supplied, but this reminds me of a session I did in NY a few years ago at Sony. The engineer turned up the guitar amp until there was an audible hiss. He then began placing mics on it while listening to a headphone send from the main console. He started with one mic in the usual "starter" position. He then added a different mic, and began moving it around until the hiss sounded "right" (in phase, I guess) in his cans. In all there were 4 different mics on that amp, and no I can't remember any of them. There may have been a 414 on there. Anyway, it took for-freakin'-ever (which, considering Sony's hourly rate, is a BAD thing) and four years and three bands later I'm still not convinced it was worth the effort. There was not a stitch of phase cancellation that I could detect, though.

 

[baekgaard]

Okay, now that they are reversed in phase, can I just move the second mic until I find the spot where they "cancel each other out" ,that is, the place where the noise (hum) is at it's quietest? This being the location where the two mics are most "in phase" together?

I just read the article, and yes, the intention is to have the mics out of phase and find the point whereby you (when placing the 2nd mic) have as good cancellation as possible. This is similar to the technique you use to match and position mics for a co-incident stereo pair.

The reason it works is that it is much easier to find a "null" (i.e. where things cancel out) than it is to find a maximum, where the two would be in perfect phase. So: Revert the phase on one mic, find the spot where they cancel out one another, and restore the original phase, as you have now made sure they are in phase -- at least for the frequency you tested with...

This will probably work in most cases -- if you try to have the mics at roughly the same distance. If they are not equidistant, you can only find a cancellation point that will work for the frequency of the amp/guitar hum you're listening to.

Remember to unflip the 2nd mic afterwards, so that you record both mics in phase, btw... there are some important points in the article.


Did this help you at all? It's getting late here, and I'll stop working at the PC now,


-- Per.

 

[cellardweller]

Yes, that helped very much.

How complex this must get when searching for the cancellation point of multiple frequencies...

Thank you.

 

[mixsit]

Okay, now that they are reversed in phase, can I just move the second mic until I find the spot where they "cancel each other out" ,that is, the place where the noise (hum) is at it's quietest? This being the location where the two mics are most "in phase" together?

This is a bit of an odd reason to dual mic and it could work sort of, and you could find a spot where the 60 or 50hz gets nulled the most, but... There are always other frequencies that will combine in and out of phase to different amounts, never just one tone. As the distance between the two time paths diverge from equal, the combing starts in the highs and extends lower as the difference increases. By the time you get a null at 60, everything above that will be a roller coaster too. And this has nothing to do with a spot where the mic's are 'back in phase'. That happened back at the zero-delta point.

Wayne

 

[schulteboy]

i'm very much a novice. a friend of a friend recently recommended this trick. i was sure that it was pointless micromanaging, but read the article tonight and put on my headphones. for a moment, it thought this was all a big practical joke designed to destroy my ears, however i am very pleased with the results. i didn't find a spot that totally cancelled out the hum, but now i have the mics in the very best position. now that everything is back in phase, the cab sounds great with both mics (sm57 and E609). yesterday i wished that my recordings could sound as good as the amp. now i wish the amp could sound as good as the recording. and if anyone touches either mic stand....

 

[baekgaard]

This is a bit of an odd reason to dual mic and it could work sort of, and you could find a spot where the 60 or 50hz gets nulled the most, but... There are always other frequencies that will combine in and out of phase to different amounts, never just one tone.

When the distance from the sound source to each if the mics is exactly the same (i.e. they are "equidistant"), then all frequencies will in the theoretical case null out. This is of course not true in a reverbant room with a sound source that is not a single point, but for this approximation it may work.

As the distance between the two time paths diverge from equal, the combing starts in the highs and extends lower as the difference increases. By the time you get a null at 60, everything above that will be a roller coaster too. And this has nothing to do with a spot where the mic's are 'back in phase'. That happened back at the zero-delta point.

Exactly, as I also tried to point out above. However, if you are able to cancel out at 60 Hz, then the next null will be a full wavelength further away. At 60 Hz and at normal room temperature, with a sound speed of around 332 m/s, it is 5.5m away (17 ft), and I doubt you will accidently get the "wrong null" ;-)

So, if you place both at, say 1 m (3 ft) away, and are able to null out the lower frequencies, you're close to the "right" null, and would in the ideal case also have a null for all other frequencies. If you place one close to the amp, and the other 6 m away, then the null will be only for the frequency most dominant.


-- Per.

 

[drummersteve]

hi there
after reading this thread and the article, it has made the whole phase issue (which isnt always an issue) much clearer to me. just a couple of things i want to ask on the subject...
somebody mentioned if i have no phase flip swtich on my mixer, i can swap the connections of pin2 and pin3 on the XLR cable, hense giving me a phase switch. now, would i do this at both ends of the cable, or just one?
another question is a little more confusing...
the hum that you would create with your guitar would be of a particulat frequency. cancelling that hum out, then reversing phase would give you that frequency in perfect phase, correct? now that that particular frequency is in phase, not all will be, and because of this, particular frequencies which you want may be cancelled out! so, if i know which frequencies i want, how do i create a hum related to this frequency? i hope this makes sense!

thanks

steve

 

[mixsit]

Sorry Cellardweller, for a minute there I thought you were adapting it to get rid of (a) hum... I'm awake now, coffee's up.
wayne

 

[boingoman]

I read an interesting thread on Prosoundweb, speaking of phase.

This guy had a problem with his monitor feeding back onstage. It only fed back when he turned his guitar amp off. If it was on, no feed back.

Turns out his guitar mic was picking up noise from the amp. It was masking the same freq in his monitor/mic setup. When he turned his amp off, that freq got very hot in his vocal mic and fed back.

 

[baekgaard]

would i do this at both ends of the cable, or just one?

Just in one end!

another question is a little more confusing...
the hum that you would create with your guitar would be of a particulat frequency. cancelling that hum out, then reversing phase would give you that frequency in perfect phase, correct?

Correct, but if you read what I tried to explain above, you will see that if you get the "right" null, then it will work for all frequencies.

The trick is to make sure both mics are at the same distance from the source, which is what you're trying to get by moving it around.

You can place it at "another" null, but at low frequencies, the distance between nulls can be several meters, so the likelyhood of accidently placing the mic 6 meters away from the speaker is not that great if you're trying to get the same distance within, say, one meter from the speaker

now that that particular frequency is in phase, not all will be, and because of this, particular frequencies which you want may be cancelled out! so, if i know which frequencies i want, how do i create a hum related to this frequency? i hope this makes sense!

In the ideal case, you will cancel out "all" frequencies, provided the mics are at the same distance from the source. In practice, it may not be fully in phase due to other effects, but it should be reasonably close by this method.

It is probably a good idea, as stated in the article, that the "hum" has some high frequency contents also. This helps you achieve a better position, and a higher likelyhood that you get the best compromise.


Please note that if the sound source is a wider object than a single speaker, things gets much more difficult. You cannot do this realiably with an acoustic guitar, for instance.


-- Per.

 

[cellardweller]

I don't know how, but I managed to totally misinterpret "equidistant" Sometimes I make things so much harder than they are by not slowing down and digesting the text!

For some reason, I was superimposing the 3 to 1 rule for mic distance (talk about not awake yet). Keeping the mics equidistant as opposed to 3:1 ratio, wouldn't there be much less chance for cancellations at frequencies other than what your "hum" produces? And, were you able to, a constant high pitch would be preferable, as the mid/high freq are the most likely to have cancellations at 1-3 feet from the mic?



*EDIT* Sorry, slowing down a READING again:

This will probably work in most cases -- if you try to have the mics at roughly the same distance. If they are not equidistant, you can only find a cancellation point that will work for the frequency of the amp/guitar hum you're listening to.

I think I've found my answer.

Thanks guys.