Electrical Audio How-To: Time Alignment of Multiple Sound Sources

hi i'm steve albini and today i'm going to talk about time aligning multiple signals from a single audio source a common situation in the studio is that you have a microphone on an amplifier but also are recording a direct signal from an instrument a bass guitar or a keyboard or a synthesizer for example this causes an arrival time difference between the direct signal and the signal on the microphone when to the recording medium one of the reasons for that is that there is an acoustic delay inherent in the microphone recording the signal from the speaker got a little diagram here that i will show you sound travels very pro approximately one millisecond per foot that's a very an extremely rough approximation uh and it varies with humidity temperature altitude local air pressure a lot of things but as a rule of thumb approximately one millisecond per foot so if you have a sound source coming out of a loudspeaker and there's any physical distance between the loudspeaker and the microphone even a matter of inches then there will be an acoustic delay in the arrival of that sound at the microphone if that signal is also being recorded as a direct signal then there isn't then there is an imperceptible or essentially zero delay for the signal that's being recorded off of the direct signal so the time difference between those two creates a phase difference in the two signals if you were to try to combine them and the phase difference can cause cancellation of some frequencies a thinning out of the bass sound and a comb filtering effect which can be unpleasant only when i can move okay i want to show you a technique you can use to compensate for that arrival time difference by delaying the direct signal the sound source for this example is going to be this synthesizer this synthesizer is going through a direct box directly to the desk it's also going through an amplifier and that amplifier has a microphone on it which is also plugged into the desk the direct signal is getting to the desk from the synthesizer almost immediately not quite speed of light but very close to the speed of light a significant fraction of the speed of light but that signal's getting to this amplifier and it has to make its way through a pretty elaborate system before it finally gets to this microphone it has to go through the input circuitry of this amplifier which has a time constant and then the pre-amplifier has to get the signal to the power amplifier section which then has to modulate the power supply through the output output transformer which then needs to couple the primary and secondary windings to the output cable which goes to the loudspeaker which then gets its voice coil energized and then that [Laughter] it's a wonder it works at all every system is different every amplifier is different and every physical installation is different so this cumulative delay of all of these time constants can be anything from less than a millisecond to several milliseconds and if there are effects pedals in line those effects can also affect the latency the arrival time difference is audible but i can also visually show you the arrival time difference on the oscilloscope by tracking the two waveforms against each other when i play notes on the synthesizer you can see them displayed on the oscilloscope the bottom trace on the oscilloscope is the direct signal from the synthesizer and the top trace on the oscilloscope is the microphone signal on the amplifier when i play these notes you'll see that the waveforms don't line up that is the peaks and valleys will be in different locations on the timeline [Music] and in extreme examples they can be almost completely out of phase with each other well that sucked so i'd like to correct the discrepancy between those two signals by inserting a delay into the direct signal path now the delays that you're going to use for this kind of correction are going to be very short sometimes less than one millisecond and it's important to have adjustability in the microsecond range i recommend using a an analog delay that's a studio quality analog delay or a digital device that allows adjustment in the microsecond range here at electrical audio we have some delays that were made for us by the eventide company that are adjustable in the single microsecond range so you'll need to use a delay of this type or an analog delay or what's called an all pass delay all pass filter so this is the signal with the delay bypassed i'm going to engage the delay and i'm going to scroll it until the waveforms come into very good alignment [Music] you should be able to see the waveforms moving one against the other here [Music] i'm not satisfied with that i feel like the waveforms are all in very good alignment and the sound quality is very good as well so the total alignment delay that we needed to use to compensate for the acoustic delay between the microphone and the loudspeaker and the mechanical system of the loudspeaker and its compliance ended up being 774 microseconds that's a significant delay but the precision of the delay allowed me to get the waveforms into very good alignment the audible effect of this is that the low frequencies are better supported because there's less phase cancellation but you should also hear a cleaner less comb filtered effect on the high frequencies as well another common scenario is to have two microphones on a single sound source say a guitar cabinet you might have two microphones with different sound qualities or you might want to create a stereo image from having using two microphones that respond slightly differently those two microphones no matter how carefully you position them are going to have some slight difference in arrival time between the loudspeaker and the microphone and that arrival time difference can cause phase cancellation which can muddy the sound quality i have a oscilloscope set up here in what's called the lissajou pattern and the lissajou pattern is a display that shows you the phase relationship of two signals in this case i have one microphone on the vertical axis and another one on the horizontal axis and this dot will trace a squiggle that shows the phase relationship of the two microphones so let's go downstairs and i will set these microphones up on the cabinet so these are the microphones that i've got up on the amplifier this is an old rca ribbon microphone which was had its cable replaced by our technician greg norman and when he did that he ensured that it followed the pin too hot convention of modern american microphones that is a positive air movement in front of that ribbon will create a positive voltage on pin two of the output connector that's standard for almost all microphones but there are a few oddball microphones and a few very old microphones where that's not true so it's always worthwhile to check to see if the microphones you're using observe the pin 2 pin 3 convention this microphone is a russian condenser microphone and i suspect that it's pin 3 hot and not pinned too hot but we'll confirm that when we listen to it upstairs playing guitar for us is john san paulo america's top liked guy he has a up knee so he can't leave let's go see if he can play guitar alright so let's take a listen to those microphones and see if we have any problems john can you play for me for a moment [Music] okay i can hear that there's a significant phase difference between the two microphones and it's so extreme i'm pretty sure one of the microphones is of opposite polarity than the other you can see that this ellipse is quite wide and it's actually working in an anti-phase direction that is in the direction opposite of what it would be if it were a mono signal so i'm going to try flipping the polarity of the microphone that i think is the culprit and see if it brings it more into alignment [Music] that sounds better to me and i can see verified on the oscilloscope that i have brought the two signals more into alignment so my hunch about the polarity of that microphone was correct so from looking at the display i can't tell which of the microphones is too close and which is too far but by listening to the two microphones in stereo i can hear that the image is pulling very slightly toward the side of the condenser microphone that means my ears can detect an earlier arrival of that signal and that means that the condenser microphone relative to the ribbon microphone is too close thanks john let's go down and see if we can correct that i could tell from listening that this microphone was a little bit farther away from the cabinet than this microphone so i'm going to move this one in a little bit it wasn't a drastic difference but it was probably enough that i could measure the difference so i'm going to use my hands to see if i can estimate how much i need to move the microphone and it appears that this microphone needs to move about an inch inch and a half closer to the speaker all right let's go see what it sounds like let's see how it sounds with that adjustment john can you play the guitar for me for a moment [Music] so that sounds better to me and i can see from the list as you display that the ellipse has narrowed considerably which means the microphones are more closely in alignment there are however still some curlicues that indicate that there is some phase cancellation going on at certain frequencies and when i listen in stereo it sounds to my ear like the ribbon microphone is now slightly closer than the condenser microphone so i want to make it another small adjustment i think we can do better than that i could tell from listening that this microphone needs to be a tiny bit closer but not very much so rather than try to move the stand i'm just going to slide it in the mic clip a fraction of an inch that should be enough let's go see so let's see what that adjustment sounds like john can you play the guitar for me please so you can see here that the two signals are in very close alignment making a very narrow ellipse but there's still some difference between the two microphones that'll allow us to hear the stereo image i was describing you should never be afraid to experiment with things like trying multiple microphones diagnostic tools like the oscilloscope and your own ears should guide you in moving and correcting placement errors

Channel: Electrical Audio

Views: 193,248

Rating: 4.9567652 out of 5

Keywords: Albini, Steve Albini, Engineer, engineering, audio engineer, production, sound, audio, Electrical Audio, analog, tape, Eventide, phase, studio, recording, polarity

Id: c52AaUmEz5c

Length: 12min 11sec (731 seconds)

Published: Fri Jun 21 2019