Drum Mixing Woes

May 14, 2011 4:23 PM

I trying my hand at recording and mixing drums in a full kit, and I'm having trouble getting it to have that sort of punch that most drum recordings have.

I know that most big recordings will individually mic every piece of the kit while recording, and that might have something to do with it (all we used to record was a stereo pair above/in front of the kit, and a single mic in the kick), but I know there's also something to be said for using some compression.

The trouble is, as I've never done so before, I have no idea how to use compression effectively, and it's really not getting me anywhere near where I'd like it to go. Does anyone have any preferred compression VSTs, or presets/curves they like to use when mixing drums? I'm kind of at a loss here, and it's something I'd like to be able to do more often.
posted by askmeaboutLOOM (9 comments total) 6 users marked this as a favorite

if you're not going to mike the drums separately, then what you need to do is to duplicate the drum tracks in your DAW and have one track EQ'd for the kick and the other track (or more) EQ'd for everything else - that way, you can mix the drum track more appropriately, although it's not as good as having the various drums on separate tracks

basically, with the track you've separated for the kick, you're going to want to filter out anything above 200-300, except a narrow band in the 7-9k range for the "beater" sound - this is very important

with the everything else track, you'll want to filter out everything below 200-300

these are pretty loose guidelines - it may be that different songs need different settings

with separated tracks dedicated to various frequency ranges, you can choose different levels of compression for each range - and if you can do parallel compression, where the straight signal is mixed in with the compressed signal, that will probably work out better

it's pretty damned complicated - and you'll find that what works in one song won't work in the next

i know that mike placement is extremely important and a lot of bands use drum replacement technology like drumagog - but my drums are totally programmed or looped, so i don't have any experience with that sort of thing
posted by pyramid termite at 9:26 PM on May 14, 2011


You're not going to get a modern, punchy sound with overheads+kick but that doesn't mean you can't get a great sound.

There are a bunch of things to think about.

1. Compression

Use slow attack on anything you want to have punch. That way you get the transient and then squash the tail. EG getting a snare to crack, or a kick to have a good beater sound. If you use a fast attack you will squash the transient and lift the tail.

Use a slow release on anything where you want to lose the tail eg toms and a fast release on anything where you want to bring up the tail (eg cymbals). On overheads you will just have to play with it.

Watch out for pumping. If you set the attack too fast, every snare and kick hit will duck the rest of the kit. Might or might not be what you want.

2. EQ

Some EQ is counter-intuitive. EG on kick, it's the mids that will give you a punchy rock kick that will cut through a mix. You may in fact want to roll off some off the bass to clean it up.

3. Phase problems

This can be a biggie with the setup you are using. You have three mics and unless they are in phase for both the kick and the snare you will find it very hard to get a punchy sound. In order to do this the two overheads have to be equidistant from both the snare and the kick drum. The simplest way to achieve this is the so-called recorderman set up (google it).

In this set-up, you have one overhead pretty much directly over the snare, pointing down, and a second behind the drummer, pointing over their right shoulder toward the snare. These are both positioned exactly two drumsticks length from the center of the snare. Now you get a bit of string, a mic cable or something and measure the distance from the first mic to where the beater hits the kick. Now you have to move the second mic a bit so that it is the same distance from the kick while still being two drumsticks from the snare. Takes thirty seconds.

Check that this set-up works WITHOUT the kick drum mic. You should be basically there already. The kick gives you a bit more control. An SM57 or SM58 with the grill removed will work great on the snare if you want to add one more mic.

I strongly recommend this book, incidentally, if you want to get into the weeds.
posted by unSane at 9:43 PM on May 14, 2011 [1 favorite]


Oh, one more thing. A lot of modern rock drums are using limiting, as well as compression. The standard is the Waves L2 but most DAWs come with something vaguely similar. What this gives you is a really LOUD drumkit with most of the very sharp transients squished. It may or may not be what you want.

The way to try it out is put a limiter on the drum submix and then lower the threshold AND the gain simultaneously so that the perceived volume remains the same, until you start to hear some kind of distortion. Now back off a little bit, and raise the gain. The drums will suddenly be FUCKING loud, and you can play with the threshold to back off the effect until it's what you want.
posted by unSane at 9:49 PM on May 14, 2011


Very cool, thanks for the advice! The EQ is something I've been playing aruond with. The phasing sounds pretty well synced, as is. I've recently started toying with limiting, and I had a feeling I might be able to get at least close to what I want between it and the EQ/gain. I'll give it a go.

Thanks, again.
posted by askmeaboutLOOM at 10:46 PM on May 14, 2011


seconding PT's idea of parallel compression -- it's an absolute godsend

basically either duplicate the uncompressed drum track or send some of it to another buss, and compress the shit out of that one. Now you have two drum tracks -- one au naturel and one heavily compressed, and you balance them using the faders. What this does is preserve transients while giving you a beefy bedrock sound. It's sort of upward compression in some ways -- standard (downward) compression reduces the volume on the loud bits at the cost of the transients, whereas this approach essentially increases the volume on the soft bits and leaves the transients intact.
posted by unSane at 10:54 PM on May 14, 2011


After only about 20 minutes of fiddling with the limiting/gain/eq, there's already a very noticeable improvement. Sometimes I forget that, as drums are mostly noise waveforms anyway, a bit of clipping or distortion is just fine, and sometimes even desirable. Excellent, this makes me quite happy.
posted by askmeaboutLOOM at 11:23 PM on May 14, 2011


Guys, the compression plugin I'm using in Reaper (ReaComp) has separate dry/wet faders. Is mixing in the dry signal basically the same thing as parallel compression?
posted by vanar sena at 6:47 AM on June 18, 2011


Yep, that's exactly what it is.

The only thing you have to watch for in parallel compression is that some compressors reverse the phase of the signal. Studio comps are probably fine but guitar stompbox compressors are notorious for this - in fact the ones that *don't* flip the phase of the signal make a big deal of it. As a result if you try doing parallel on your guitar by using a splitter and feeding a dry and compressed signal to the amp it is most likely going to sound like shit.

I would imagine the ReaComp doesn't suffer from this but it's worth looking out for especially if you get weird filtering effects when trying to put together a parallel compression chain.
posted by unSane at 7:07 AM on June 18, 2011


Cool, thanks. Now I have to work out how the LPF and HPF built in to ReaComp fit in with all of this. I'm not hearing any phasing problems, but that could just be my n00b ears.
posted by vanar sena at 9:15 AM on June 18, 2011


« Older Joscho Stephan Trio   |   online music communities Newer »

You are not logged in, either login or create an account to post comments