Spending years in bands self producing and releasing material, it was always clear to me that one part of the process stuck out and that’s drums. Recording vocals, bass, keyboards, guitars, horns, strings etc can be done with one or two mics in a fairly ordinary space and with an entry level soundcard producing very passable results.
While it’s not impossible to record drums with one or two microphones, doing this leads you down a certain pathway sonically and it doesn’t allow the separation of each mic allowing you to rebalance the kit as you see fit.
You can think of the drum kit like several individual instruments rather than one whole one; a kick, snare, hi-hat, number of tom toms, ride and crash cymbal. I know of engineers that go into a huge amount of detail micing each element up but what I want to discuss in this article is a middle ground; recording drums on a budget.
Recently a good friend of mine and top drummer (Stuart Pringle) acquired a space to set up his drums and record them, so naturally I jumped at the chance to get some recording done. We spent a few hours recording loops of about 4 – 16 bars in duration at all different tempos. Interestingly Stu is a big fan of Ableton Live too, a DAW that I think is often overlooked in a recording environment for one reason or another.
We did all of this on a budget of virtually zero, pooling together our own microphone collections (and borrowing a stereo pair) with no fancy 500-series mic preamps or other hardware and humble soundcard choices. That’s not to say it hasn’t taken a while to get to this stage, but it is possible to do without going into the studio. This is a sort of update on this tutorial I did back in 2015.
Here’s a video quickly explaining what we’re going to be talking about, but if you want a little more detail, do keep reading:
Sadly as this recording was done around a year ago, I can’t remember the specifics of the recording in every detail. I do remember using at least 1x AKG D112 on the kick and likely a pair of SM57s on the snare (top and bottom).
As for the rest it would be a bit of guessing game, but if anything comes back to me I’ll update this article. What I do remember though is there wasn’t anything expensive going on and most of the mics used could be picked up for under £150 or so.
The soundcard we used was a Focusrite Scarlett 18i20, which offered us eight mic pres. If we needed more it would have been possible to add another portable soundcard as an aggregate device (read about how to do that in OS X here).
I favour close micing the kick with two mics – one on the beater and one slightly outside the drum to pick up some more bottom end. Two mics on the snare – top and bottom, a close mic on the hi-hats and a pair of stereo overheads.
If we had more inputs and mics on hand (and perhaps a nicer room to record in) we could have close mic’d the toms and added a room mic(s) but this gave us enough control.
Before we move on, let’s hear how the completely unprocessed drums sound:
In Ableton, I’d set the default warp algorithm to Repitch – this makes audio behave like an analog medium – if you speed the tempo up it increases in pitch and vice versa for slowing down. I generally find this protects transients a little better than any of the other dedicated warping modes but as we’re not going to need to do too much warping today it doesn’t really matter which mode you use.
If I do need to tighten a hit up I’ll switch to complex mode then render the loop in place. A really neat feature of Ableton is the multi-track warping, something definitely worth familiarising yourself with:
Let’s start off with the kick – one of the backbones of the most tracks in nearly any style of music. As mentioned above we had two mics on the kick, one for the beater and transient sound, the other placed about a little further back.
Firstly I’m going to group the two tracks together (shift select them hit cmd + G) – this allows us to process them both individually and as a group. Let’s start with the close mic:
First off I’m going to gate this use Live’s built-in one. I’m trying my best to isolate the kick, discarding as much snare as possible. I’ve used a quick attack and set a conservative hold and release times. Getting the threshold right is the most important thing.
I can hear a bit of a bump just above the fundamental of the drum that sounds a bit grim, so I’ve dipped that out a bump at 140 Hz using FabFilter’s Pro-Q 2 and added a light low-cut filter.
I’ve been getting into Wave’s Pultec emulations recently, and found this tutorial on using them rather helpful. I started out emulating what the author did to their kick in the tutorial and fiddled until it suited my needs.
Finishing off the kick I added one of my favourite cheap plug-ins Fielding DSP’s Reviver (adding in third order harmonics) and some transient design courtesy of the Waves Trans-X Multi, tweaking the Aron Kick 1 preset.
Moving on to the other mic, we have a rounder tone to this, capturing more bottom end (exactly what we were after) but a little more spill from the snare.
If there’s a negative phase relationship, when the plug-in is disabled it should sound thinner. Luckily my mics are in phase so I can keep it disabled.
We’ll need to do this with other mics too, normally if there’s more than one mic on a source it’s something you’ll need to check for. Some channel strip emulations have a phase/polarity invert button, so look out for that.
I’ve added Live’s Gate to isolate the source as best possible. I found this harder than the above example but it was still a marked improvement. I’ve used longer hold and release times to allow more sub to come through (the subbier part of a kick is at the end).
Next I opted for a post production tool; the Waves LoAir. I personally think this sounds better than their alternatives (Renaissance Bass). I’ve not done much – just tweaked the range. Ensure the Direct is turned right up to let the original signal through. I finished off the mic with some Live EQ to remove unwanted top-end and dip a little around that troublesome 140 Hz again.
It might sound a little muffled but that’s because our presence will come from the close mic and overheads.
Next I’ve done a favourite trick of mine. In Logic this is a little simpler as it has a test oscillator bundled but in Live I’ve resorted to using Operator. Draw a MIDI clip and add a note that has a duration of your loop. Set Oscillator A’s tuning to Fixed and tune it to the octave below your kick’s fundamental. Mine turns out to be around 50 Hz, give or take.
Now add a gate and enable the sidechain analysis. Take the Audio From your Kick In mic, enable the EQ and high-pass at wherever feels good to you. Now we can trigger the gate whenever our kick hits and tweak the hold and release times of the gate to give us a nice low octave with whatever duration suits us. Using the EQ in the sidechain allows us to get a snappier sidechain signal.
I’ve routed this Operator to the same group as the Kick In and Kick Out mics. I’ve used a little of Eventide’s UltraChannel on the group as it has a nice EQ and excellent O-Pressor compression. I’ve dipped a little around 500 Hz, boosted below 100 Hz and at 6 kHz. Here’s how the group sounds with all three channels processed:
The snare also has two mics, a top mic to capture the ‘crack’ and a bottom one to pick up the actual snare rattle itself, a more sizzly , high-end sound. Both have been grouped. Let’s hear the top mic in isolation:
After trialling a few different channel strips I settled on the Waves Kramer HLS which I am reliably informed is modelled on his Olympic Studios Helios desk. I added in a bit of 120, 700 and 10,000 Hz. The made the snare pop a little more and added some more bite to it.
Again I reached for Fielding DSP Reviver and some compression from the Focusrite Compressor. I have to admit I was surprised how good this sounded, not bad for another freebie. I finished off with some more Pultec EQ to add in some body.
Checking my bottom snare mic was in phase (it was), all I did to this was add a high-pass at 370 Hz filter.
I created another Operator and did (basically) the same trick as what I did with the kick, only this time opting for the white noise wave, not a sine. This doesn’t require tuning but a heft amount of boxing in with EQ. This was again gated using the sidechain analysis and I added a little of Live’s Reverb plug-in to add a bit of roomy-ness to it.
On the snare group I added some API compression in parallel using the Audio Effects Rack in Live. Be careful about the volumes after doing this. I also added a tiny bit of sidechain compression triggered by the kick. This helped leave a little more room for the kick in the mix, even though the kick and snare rarely hit at the same time, the snare mics picked up a fair bit of kick which was difficult to gate or filter out.
I have to confess I’m not normally a big fan of using hi hat mics and have be known to discard them, but keeping them around for this is not only helpful to discuss some ways to get them sounding better but also on this occasion I didn’t mind the result too much. Here’s how they sounded out the soundcard:
I found a bit of an unwanted lump at 150 Hz so removed that with some Pro-Q 2 and tried to tame the barked hat with a dip at 8.5 kHz. then added a high-pass at 70 Hz.
Normally I would reach for a De-Esser but oddly Live doesn’t have one so I used the FabFilter Pro-DS, although I couldn’t really get the result I was after so I ditched the idea (no shame in this).
What I did opt for though was an excellent (and cheap) emulation of the Empirical Labs Distressor, the SOR8 by Pensadia. I find you can hit this quite hard and so long as you leave the attack slightly longer it retains nice detail while brining out some nice frequencies.
On to the overhead mics. I have two tracks which I grouped together same as the kick and snare. I also panned these two apart. Some people prefer to pan just a little but I like hard-panning them (no doubt someone will tell me why I’m wrong…). Again ensure you check the phase relationships between these two mics.
On the group I used some SSL Channel strip (E-Series) to shave off below 150 Hz and did a little boost at 300 Hz. I wanted to add a little compression to clamp down on the kick and snare transients, leaving the close mics to cover that ground. I like the vladg/sound Molot for this as it’s super fast… and it’s free! The difference should be subtle, as at this stage I don’t want to drastically change the character of the kit, especially as the overheads are the “truest” representation of the overall sound.
Master Bus and Reverb
Here’s how all our drum tracks sound together. Not bad, but work can still be done to get them sounding a little more original.
I’ve created a return track (alt + cmd + T) and added the excellent Max4Live Convolution Pro to it (if you’re unsure about what convolution reverb is, have a read of this). I’m sending some kick, snare and overheads to it.
I’ve used the Swede Plate 1.0s (named after Bruce Swedien) increasing the decay and size a little and rolling off some bottom and top in the EQ. Subtle, but that’s what we’re after.
Here’s just the reverb solo’d (if that’s of interest to you):
One downside to Ableton is that you can’t nest groups within groups. To get around this I’ve routed my kick, snare and overhead group (as well as the hi-hat track) to a new audio track. Ensure you set the monitor to ‘In’ otherwise you want hear the resulting audio. Now we can further process our tracks together.
I’m going to use two of my favourite new plug-ins to totally transform my drums from drab to awesome in not many steps at all. The first is the excellent Vulf Compressor from Goodhertz.
This is modelled on the seminal and sought-after Boss SP-303 “Dr. Sample”‘s vinyl simulator mode, supposedly a favourite of J Dilla, Fourtet and Madlib amongst others. It’s a compressor but it does so much more. Watch the below video to get an idea or watch this slightly longer explanation from Jack Stratton, drummer and brain behind Vulpeck, who conceived of the plug-in with Goodhertz.
It doesn’t take much fiddling to get some awesome turbo-charged 70s breakbeat sounds from this. Here are the settings I’ve used:
The WOW seems to add a stereo flange that I don’t really want in this instance, the LOFI is a noise/sample rate and bit depth reduction circuit (to my ears anyway), and the rest is a fairly simple compressor but it really does sound great.
Added to that is a new favourite of mine, Boost by Sample Magic. Not only is this in the right price bracket for me but I dig simplicity. There are just four controls, Compressor, Colour, Stereo and Limiter.
And that’s kind of it! We’ve taken some fairly ordinary sounding drums and made them into something special. I’m not going to pretend the last two steps didn’t make a huge difference but carefully treating each mic has given us a better canvas to work from.
There are some basic things you can do in any DAW (nice plug-ins or not) that will drastically clean up most any recording. Gating is key to isolating what you want to be dealing with, but don’t go crazy, you can only work with the recording and getting separation at the tracking phase is preferable.
Careful EQing can further help remove unwanted frequencies or ones that are building up when several mics are in the mix. Checking for phase problems is key to getting two mics on the same source (snare, kick etc) working in harmony. Compression narrows the dynamic range of a mic meaning you can either tame wild hits or bring up the average level of the mic as a whole.
Once we’ve got the mics sounding good, that’s when you can be creative and process things to taste. Hope this has been of some benefit.
Here are some of the other loops from the session with the above processing:
And because we’re using repitch algorithm, here are some of the beats not at their original tempo:
Originally recorded at 92 and pitched up to 131 bpm:
Originally recorded at 170 and pitched up to 186 bpm:
And lastly this was originally recorded at 95 and pitched down to 90 bpm:
Also I’ll be giving away some of the drum recordings from this session in the near future so keep your eyes peeled for that. Enjoy!