Getting a big‑sounding drum recording in a small space is never easy — but it can be done...
Capturing a big rock drum sound is never easy, but it's even harder in a small space, and they don't get much tinier than the basement room in which I was invited to help young riff‑rock band Magician's Nephew record their drummer, Noah. With a floor area of about 10x12 feet, when the drums and mics were set up it was a challenge just getting from one side to the other. The live and control rooms were less than six feet tall below the joists and connected by a four-feet high passageway, making it even more 'cosy' given my height of six foot four!
The kit was fairly conventional, with kick, snare, three toms, hi‑hat, ride, and three crash cymbals. The band's normal tracking setup was based around a set of Audix dynamic mics (an i5 for snare, a D6 for kick, a D4 for floor tom, and D2s for the rack toms), an AKG C451 small‑diaphragm condenser for the hi‑hat, and an overhead pair of Rode NT2A large‑diaphragm condensers. Despite the respectable mic line‑up and having spent considerable time and effort experimenting with mic positions, the band weren't entirely satisfied with their recordings, which were slightly hollow and abrasive. They'd even tried replacing the Rode mics with a pair of Shure SM57 dynamics in front of the kit, and although these did a creditable job of representing the kit's balance, they also lent the timbre an unwelcome indie trashiness that was at odds with the band's classic rock influences.
I asked them to send me one of their latest raw multitracks, so I could try to diagnose the malaise before the session. As expected, given my own experience of the NT2A, a harshness to the upper spectrum of their overhead pickup seemed partly responsible, but it seemed to have been exacerbated by pointing the mics at the cymbals from above, emphasising the brightest cymbal perspective with the mic's on‑axis HF boost. I also noticed that the snare and tom close mics seemed to have been placed very close. An understandable tactic to minimise the small‑room acoustic signature, this had ejected the baby with the bathwater: the mics highlighted mostly head resonances at that distance, without picking up much natural tone, and the dulled off‑axis spill in the overheads did little to supplement that. Also, since the kick's resonant head had no hole, it had been recorded from the outside, so it didn't deliver the tight, aggressive sound that's usually associated with mainstream rock.
With all this in mind, I prioritised capturing a fuller, more representative drum‑kit timbre, rather than strive too hard to downplay the character of the recording room. It's far easier to fake the impression of a larger room using reverb at mixdown than to bolster a fundamentally anaemic kit tone!
The easiest way to get a full‑sounding kit is to use a main stereo pair to capture a fairly well‑balanced presentation of the instrument as a whole, and then use the less natural‑sounding close mics only for presence and definition. But you stand or fall on the quality of that main stereo pair, so I wanted to set that up at the outset. I'd brought some of my mics along but was keen to make the best of the band's mics. That way, the band could apply any techniques they learnt on the session to their future work, without shelling out for new gear. So there were only two condenser pairs available: the Rode NT2A large‑diaphragm condensers, or the Audix ADX51 small‑diaphragm condensers.
Given the band's difficulties with the NT2As, you might have expected me to choose the ADX51s, but I stuck with the Rodes for two reasons. Firstly, prior knowledge of the ADX51s inclined me against them; I've always found them rather cold and hard‑sounding and a little bass light, as though they're voiced more as cymbal close mics than for overall pickup. Secondly, I knew the multi‑pattern NT2A's omnidirectional polar pattern would sound much smoother than the cardioid pattern the band had instinctively used — it really is a tremendous difference, almost like switching to another mic! Besides, I figured that the wider pickup and flatter off‑axis response of the omni pattern could help me achieve a more holistic impression of the kit.
Once I'd switched the polar pattern and engaged the 10dB pad to protect against overloads, I listened around the kit while Noah played, and decided to start with one of my tried‑and‑trusted setups. I placed the NT2As over the drummer, spaced about a foot or so apart, and angled the capsules slightly inward, to give the snare the benefit of their on‑axis high‑frequency boost. This delivered the kind of smoother, fuller sound I'd hoped for, but there was a fly in the ointment: the toms were dominating over the cymbals.
To mitigate that, my next step was to move the mics more towards the cymbals, but while this made the metalwork more present (at the expense of some loss of snare brightness), the toms seemed to dominate the balance more than ever!
I began to wonder if I'd overlooked the toms being too loud acoustically, and when I returned to the live room that seemed to be the case, pretty much no matter where I listened from. I tried the floor tom in different locations to see if that made a difference, but it didn't seem to have a huge impact. Finally, though, I realised I'd fallen into the trap of always listening at the same height (hunched over slightly, with my head grazing the ceiling!). The moment I tried stooping even more, I discovered that a strong floor‑to‑ceiling resonance in the room meant ceiling‑height mic positions really overemphasised the toms' low end.
With the mics down at roughly three feet above the floor, the effects of this resonance were reduced and the toms started coming across far more appropriately. The floor tom was still a bit loud due to its increased proximity to one of the mics, and I was also getting too much stick noise from the ride, so I tried some smaller height and angle tweaks. They didn't really deliver substantial improvements, so I decided to try moving the mics to the other side of the kit, which was further from the floor tom and where the ride cymbal would help shield them from the highest frequencies of its stick noise.
Listening in the live room again, I placed the mics a little to one side of the kit's centre line, very close to the position the band had chosen for their SM57s. Suddenly we were making real progress, with the balance between the cymabls, snare, and toms finally seeming to slot sensibly into place. Given some remaining concerns that the lower mid‑range of the toms was a little congested, I moved the mics again, this time a little closer to the centre line and a few inches further from the floor. We'd arrived at something pretty workable, but the band asked for a slightly wider stereo image (I'm naturally a bit conservative in that respect), so we nudged the mics maybe a foot wider apart to end up with the sound we agreed on — the eighth position we'd tried!
Although our main mic pair started in an overhead position we'd moved them further away from the kit, so I knew I'd need close mics on all the drums and cymbals for presence and balance control in the mix. Starting with the cymbals, as I guessed their mics would pick up the most spill and interact with the main mic pair most strongly, I was again reluctant to turn to the ADX51s, so used my pair of Oktava MK012 omnis, which I knew had a smoother tone and wider pickup angle. (And at about £350$450 per pair, they wouldn't break the bank if the band wanted to invest.) Putting these a couple of feet above the two groups of cymbals gave the brass‑heavy balance you'd expect, but with the same over‑prominent low‑end tom sound we'd experienced with the NT2As at a similar height. Fortunately, we only needed those mics for their cymbal pickup, so I was easily able to solve this with a little high‑pass filtering.
For the kick, I suggested that we try completely removing the resonant head for a tighter sound, and the results from the Audix D6 spoke for themselves, with a much more solid sound than the band bad achieved before, with a beefier low end and a more clearly defined beater transient. Backing the snare mic off a few inches brought another serious improvement, clarifying the instrument's low mid‑range, weakening the batter head's pitched resonances, and generally capturing more upper spectrum energy. Granted, there was a little more hi‑hat spill, but in practice that's not something to worry about so long as the snare doesn't need masses of EQ to brighten its timbre. The bottom line is that I'd rather have an appealing snare with a bit of hi‑hat spill than a spill‑free track that sounds like someone tenderising a hubcap!
There wasn't much scope to tweak the toms' sound, as the clip‑on mic mounts the band were using provided less leeway for adjustment. But I did swap the floor tom's D4 close mic for a Shure SM57. It wasn't that the D4 sounded bad, but rather that the instrument's low end was already well represented through the other mics, and I figured that a leaner close‑mic sound might supplement that more usefully. In hindsight, I was overcompensating. At mixdown, I couldn't really boost the floor tom's low end with EQ without clouding the snare and kick spill, whereas I could have freely cut low end from the original mic's bassier signal if necessary. (I could have gated out spill on the tom mics, but I find that those spill components often enhance the complexity and cohesion of a kit.)
The hi‑hat mic was ready to roll in seconds, partly because we seemed to get a reasonable sound right away, but also because I guessed from the hi‑hat level in the kit balance that I'd need very little from that mic at mixdown; why spend hours tweaking a sound that'll barely be heard? I added one extra mic to the band's setup, though: something I call a 'kicksnare' mic. This mic is placed in the small space at the front of the kit between the kick and snare, pointing roughly at the drum throne. This catches a good dose of snare‑wire rattle (remember we'd set up no under‑snare mic), lots of nice shell tone from both drums, and a certain amount of kick 'rattle' from the hardware that can add rock credentials to any record — especially if you slap it around a bit with compression! Plenty of engineers have their own variation on this and mic choices differ widely, but most of the time I use a large‑diaphragm condenser. As luck would have it, the band actually had a Neumann TLM103, and mixing that in gave a real lift to the overall drum sound, lending both kick and snare more attitude and significantly thickening the snare sustain.
Personally, the only way I feel I can reliably judge whether each new drum mic sounds right is by building a mix of the kit as I go. This means experimenting with faders, pan pots and polarity switches, and maybe applying high‑pass filtering where necessary (in this case, to remove that tom 'boom' on the Oktava cymbal mics). But all these mix settings were only on the monitor channels in the band's Logic‑based DAW, so the mic signals were recorded with no processing. That was a useful safety measure, given that I wasn't familiar with the monitoring in the band's control room. You're giving a hostage to fortune if you make irreversible processing decisions without being totally confident in what you hear!
Fortunately, in my own mix room our tracking decisions held up pretty well, and it didn't take much processing to get a decent sound. I'd slightly undercooked the kick's low end, which needed a little EQ boost, and the snare and cymbal mics benefited from a few decibels of HF shelving lift when set against the band's overdriven guitars, but otherwise my mix's drum EQ amounted to a handful of high‑pass filters for low‑end clarity and the odd low mid‑range notch to combat some residual tom resonances.
Working in a purpose‑built studio with high‑end gear is tremendously forgiving of mic technique, and it's easy to get complacent — you can often get surprisingly good results under such circumstances just setting up your mics 'by eye'. When recording in project‑studio conditions, on the other hand, you typically have to work much harder with your mics, because the acoustic anomalies of domestic rooms and the colorations of budget equipment stack the odds very heavily against you.
I'd estimate that I have to swap out or move at least 80 percent of my initial mic positions when working in project‑studios, and I think a lot of younger engineers give up on this way too early — especially if they've developed their recording skills in the expensively designed and equipped studios of a top music‑tech college course.
As I've demonstrated in this column, there are times when you simply have to grit your teeth and treat the miking process like a war of attrition. Sure, the inconvenience of working through eight positions for one pair of drum mics tested the patience of everyone. But the alternative would have been a much bigger burn — a sub‑par drum sound!
Magician's Nephew are a young rock power trio from North London, comprising Josh Lima (lead vocals, guitar), his brother Noah Lima (drums, alto sax), and Euan Campbell (bass, keyboards). Despite a combined age of just 40, they've been gigging local festivals for years, have already released their third album, and have recently appeared on BBC Introducing and The Voice Kids UK. The song featured this month is called 'Get Out Of Bed', from their album Today's News For Tomorrow's World, and you can hear their latest single 'Symphony Of Silence' (also recorded and mixed by Mike) on their website now.
For more information about this session, you can check out a special video I've posted at www.cambridge-mt.com/youtube.htm.
You can also find the raw multitracks for this whole song at www.cambridge-mt.com/ms-mtk.htm#MagiciansNephew if you'd like to try mixing these drum recordings in context.
In addition, I've posted plenty of detailed audio examples on the SOS site at www.soundonsound.com/techniques/session-notes-media-0119, illustrating the different mic techniques I used. The filenames of these audio examples should be fairly self-explanatory, but you will find some additional notes describing exactly what you're hearing in each case.
- The ZIP file in the righthand sidebar contains these notes, plus hi-res WAVs and MP3s.
- Click below to download the Audio ZIP file that accompanies this article.
Download | 68 MB