Welcome to the Sound On Sound Recording and Mixing podcast channel where you’ll find shows packed with Hints & Tips about getting the most out of the recording, mixing and mastering process.
More information and content can be found at www.soundonsound.com/podcasts | Facebook, Twitter and Instagram - @soundonsoundmag | YouTube - https://www.youtube.com/user/soundonsoundvideo
Mike Senior
Welcome to the Sound On Sound recording and mixing podcast channel.
I'm Mike Senior. If you ask me, acoustic guitar is one of the most challenging instruments to record. There are just so many physical components that contribute to the overall sound. So finding mic positions that catch an appropriate balance of all those things can be extremely tricky. Especially if you're micing very close to the instrument in search of a detailed, up front sound.
But on top of this, many performers will use several contrasting playing techniques within a single song. Or even take their lead from modern fingerstyle virtuosos by accompanying themselves with a bewildering array of percussive taps and scrapes alongside. What that all means in practice is that no matter how careful you are with your recording technique, there always seems to be some element of mixed down tweakery involved in getting decent end results.
So in this podcast episode, I'd like to explore some of the most common troubleshooting tasks I typically encounter when mixing acoustic guitars. And I'll also pass on some of my favourite techniques for tackling them. Let's start with the simplest problem first. Have a listen to this snippet of acoustic guitar recording.
So did you hear the simple problem there? Well, me neither. Because the whole point about this problem is that you don't typically hear it. What I'm referring to is useless subsonic rubbish that's happening right down at the bottom of the frequency range. Way below the point where most people's playback systems can actually reproduce it.
If you happen to be listening to this podcast on a system with a subwoofer, you might see the cone doing a bit of a cha cha cha halfway through. But even then, you're only going to really feel it rather than hearing it. And the bottom line is you just don't need those kinds of frequencies on an acoustic guitar part.
Furthermore, that useless low frequency energy could have an undesirable impact on the rest of your mix, reducing headroom in the master bus, for example, conflicting with the low end of your kick drum and bass guitar, or causing dynamics or saturation processes to respond in unmusical ways. Fortunately, this is a problem that is dead easy to fix using straightforward high pass filtering with negligible impact on the guitar sound itself.
For example, let me play that little acoustic guitar snippet again, and then repeat it with a 30 Hz high pass filter.
Whatever difference you might think you can hear between those two different files, it's really not going to matter at all within the context of any realistic mix. In fact, to be honest, this is something I do almost as a matter of course with acoustic guitars. Because there are so many ways in which unwanted low frequency information can get onto an acoustic guitar recording.
It's pretty common to be recording acoustic guitars with sensitive condenser mics, and using quite a lot of preamp gain because the instrument isn't that loud. So it's very easy to pick up things like traffic rumble, the player tapping their foot, or even just the wind turbulence generated by the player's strumming hand as it moves in the vicinity of the mic diaphragm.
And that's before you consider common playing techniques that effectively involve hitting the heel of the hand against the body of the guitar, as in this example.
Even in this case, where you can argue that the low frequencies are important to the rhythmic impact of the part. You can still high pass filter at 30Hz to get rid of useless subsonic information and not lose any of the subjective low end weight. Here's that example again, first without the high pass filtering, and then with it.
Another
common problem I find with acoustic guitar recordings is when some resonance in the room or in the instrument itself Causes certain notes to be significantly out of balance with others. Here's a real world example, where a strong 170 Hz resonance is making some of the lower notes louder than others.
Again though, once you realize what's going on, the fix is pretty simple. All you need to do is apply a narrow band EQ cut at that frequency, and it'll bring those notes back into balance. To demonstrate what I mean here. Let me play that acoustic guitar recording again, and then repeat it with that narrow band, 170 Hertz EQ cut.
Now, a trap that some people fall into when faced with imbalances between different notes like this. is that they try to deal with them using dynamics processing rather than EQ. But that doesn't address the basic tonal imbalance between those different notes. So all it does is change a balance inconsistency into a tonal inconsistency.
As well as adding in any real time gain change side effects of the compression itself. Let's compare, for example, the sound of my EQ fix with a version of that same recording where I've used compression to even out the balance inconsistencies instead. So first the EQ fix. And then the compression fix.
Not only do the offending notes feel a bit woolly in the compressed version, but the dynamics of the part in general feel less natural too. This kind of resonance issue also occurs with strummed parts, but it can sometimes be a bit trickier to key your ear into it when so many different notes are going at the same time.
Here's another real world example, which features four bars of strummed chords. And you can hear how the internal balance of the chord notes is being pushed out of kilter by a low frequency resonance during the second and third
chords.
Again, the resonance here is at about 170Hz, and a narrow EQ cut in that region immediately makes the guitar's tone feel more consistent across the whole progression. Like this.
Another very common technical difficulty that I encounter with acoustic guitar recordings is trying to reduce the level of over prominent mechanical noises. Pick noise, for instance, can easily be overemphasized by using bright sounding condenser mics close to the instrument. As in this example.
Now what works in your favor here is that the pick noise is much louder than the guitar harmonics in the high frequency region of the spectrum. So my favorite way of dealing with it is just to use a very fast multiband compressor. But only operating on the spectral region above about 5 kHz. The idea is to set the compressor's threshold so it just catches the pick noise with a high ratio, but otherwise leaves the high harmonics of the guitar tone itself untouched.
Let me show you how that sounds. First I'll play the unprocessed guitar part, and then I'll play it with a multiband compression.
The great thing about the multiband processing solution here is that we've been able to take the sharp edge off that pick noise without reducing the sense of drive and dynamics in the guitar part as a whole. And the same processing tactic also works well in situations where the pick noise is more dispersed rather than the extreme transient spikes you heard on that previous example.
Take this section of recording for instance.
And here it is with the same kind of multi-band compression processing to reduce the pick noise.
Now of course, how much you reduce the pick noise is a question of personal taste, and it'll naturally also depend on the mix context that the guitar has to fit within. But it's pretty easy to adjust the strength of this kind of processing, simply by changing the high frequency band's threshold and ratio settings.
A much trickier mechanical noise to deal with is fret noise, because it often overlaps the guitar notes, and isn't that easy to distinguish from the desired high frequency guitar harmonics just on the basis of level. So all your traditional threshold based dynamics processes aren't that useful, even in multiband form.
Personally, I've never found a satisfactory, set and forget way of dealing with fret noise. And I've only ever been able to get results I'm happy with by using region specific plugin processing to low pass filter each instance manually. For instance, here's a section of solo guitar line with prominent fret noise in it.
And then I'll play it again where I've used region specific low pass filtering to reduce the fret noise.
Now in isolation that might seem a little bit unnatural sounding, but within the context of a mix you don't really notice what's going on. You just find yourself concentrating more on the melody and less on the mechanical noises. Have a listen, I'll play that same comparison again, but this time within the context of a guitar duo mix.
So first without the fret noise reduction, and then with it.
There are of course some styles of guitar part where the mechanical noises are very much part of the rhythmic nature of the track. Like this for example.
The problem is that if you mic these kinds of parts from close up, as in this case, the mechanical noises tend to be captured by the mic louder than the sound of the instrument itself. But if you try and compress the percussive peaks to bring them down more into level with the sound of the guitar, you quickly find you're taking the edge off their very slender attack.
This is a textbook case for parallel compression. In other words, I create a parallel channel with a heavily compressed version of that previous guitar recording, designed to bring out as much of the guitar sustain as possible. So something a bit like this.
Now clearly that sounds unacceptably mashed on its own, but if you mix it in with the unprocessed signal, it increases the level of the guitar tone, without taking the edge off the percussion transients. So let's have a listen. First I'll play the unprocessed track, and then the track with the parallel compression.
So now we've explored some of my favourite methods for troubleshooting acoustic guitar recordings at Mixdown. They may not seem very inspiring on their own, but if you use them to get technical roadblocks out of the way early on in the mix process, you'll find yourself free to focus 100 percent on the more creative mixing tasks where the real magic happens.
That's all for now. Thanks for listening, and be sure to check out the show notes page for this episode, where you'll find further information including web links and details of all other episodes. You can also download a 24 bit WAV version of the show from there, if you'd like to hear the audio examples at higher resolution.
And just before you go, let me point you towards www. soundonsound. com slash podcasts, where you can explore what's playing on our other channels. I'm Mike Senior, and this is a Cambridge MT production for Sound On Sound magazine.