Element 2 - Drum Machine, ambient

As mentioned in the last post, we'll be doing more than just plugging the direct signal of the drum machine into the tape recorder and having that be that. Most of our favorite sounding recordings balance between the direct sound from the instrument (or amplifier) and the ambient sound that the instrument makes in the room. This additional element adds a sense of spaciousness and complexity (which in our case will be a stereo, rather than just mono, signal) that is desirable.

To put this in less technical terms - by mixing the sound coming right out of the drum machine with some microphones set up in a room where a PA is playing the drum machine, we can blend the thing's detail and tone with a more physical and ... uh ... bigger sound. I'm not doing such a great job of explaining it, so we've got the sound clips below. While this is most important with the drum machine, we'll be making similar choices of direct vs. ambient with the bass, guitar and vocals.

One artificial way that people acheive this same goal is by "adding reverb" to the sound using an effects device. We've got one of these boxes but really never use it - the sound of an actual room is much more compelling, even if the control you have over what it sounds like is greater with an effects box.

There are a few different options to consider when pulling together this ambient sound. We'll take a look at a few - microphone configuration, ambient delay, signal path and mic placement. Relative level of direct to ambient will come later.

Microphone configuration


There are several different ways that you can take two (or more) microphones to create a stereo image. (Most microphones only provide one channel of output.) This is going to be one of the more technical parts that we'll cover in this journal, and there's a lot of background that we won't cover here. If you have any interest in understanding stereo theory, run, don't walk, to buy [ The New Stereo Soundbook ] today. It's worth whatever they're charging for it. So, the approaches considered are:

1. [ XY ] - Two cardioid mics, with the elements as close as possible. There is a range of generally accepted angles between them, we chose 110 degrees.

2. [ ORTF ] - Two cardioids, angled apart at 110 degrees with 17cm between the elements.

3. [ NOS ] - Two cardioids, angeld apart at 90 degrees with 30 cm between the elements.

4. [ Mid/Side ] - One cardioid (usually) and one bi-directional microphone. The signal from the bi-directional mic is split. The polarity of one of the two split signals is inversed and the two signals are then panned hard left (+ signal) and right (- signal). This is then mixed with the mid signal (centered); changing the relative values changes the perceived width of the stereo image.

5. [ Spaced Omni ] - Two omnidirectional microphones, spaced about 10 feet apart.

Approaches 1, 2 and 3 were all done using the same pair of beyer M160s (a hypercardioid, which gives a somewhat different signal than a strict cardiod), placed in the same place in the room and run though the same preamp. Approach 4 swapped out one of the M160 with an M130, an almost identical mic with a different pickup pattern. Approach 5 required a totally different pair of mics. We used the Earthwords TC30Ks with the big caveat that the mic itself sounds very different than the beyers (as will be obvious from the sound clip).

We did not do a Blumlein pair because we didn't have another bi-directional mic (same reason we skipped OSS - no Jecklin disc). We also didn't think that a binaural sound would be what we were looking for so we skipped that, too.


The five sound approaches are outlined in the clip below - each is blended at the same level with the direct sound straight from the drum machine. (It starts with just the direct drum machine for 4 measures.) This time I've included an audio note on the type of configuration with every switch and I got the computer to say it. Sounds more official, right?

Drum machine - ambient, 5-way comparison.

The most striking difference between these about these how the different phasing issues emphasize different tones. Without getting into the reasons why, you can notice that the snare drum, for example, sounds higher or lower from clip to clip. Stereo impression means that each ear hears the same thing differently. The type of difference from left to right depends on lots of factors - in this case the relative positioning of the microphones capturing each respective channel. The most important thing to remember is just that they will be different, although not always the same way. The other factors that affect microphone pickup - placement in the room being the most important - will change the tone of the sound even when it's the same microphones that are moved to a different place.

Second, it's too bad that I didn't have pair of omni microphones that sound similar to the beyer mics. The spaced omni clips do sound very distinct here - it's tough to separate how much of that is due to the configuration and how much is due to the very different microphone pickup response for these mics. In particuar, I notice that these seem to catch more of the deep LF sound.

The main point, though, is in how each of the arrangements presented the stereo information of the sound bouncing around the room. I'd rate them in the following order, starting with the most narrow to the widest.

NOS > ORTF > Spaced Omni > XY > Mid/Side

While the first four may switch around a little bit, the Mid/Side was far and away the widest image. This approach does allow one to make the image wider or narrower and I purposely set it for as wide as possible.


We're going with the Mid/Side approach. For drums especially, the wide stereo image creates a strong impression. Given the flexibilty that it gives us to even make the image not as dramatic (if we want), there doesn't seem to be sufficient advantage to any of the other approaches to try something else.

I was, however, struck that the Earthworks mic did a better job of capturing the lowest frequency sounds coming out of the subwoofer on the spaced omni approach. The very next thing we'll do is to do a 2-way the Mid/Side comparison, swapping between the hypercardiod beyer M160 (with a relatively pronounded low frequency drop-off) and one of the omni TC 30Ks for the "mid" microphone. After that, we'll introduce delaying the sound of the ambient microphones from the direct signal (for reasons we'll review next time).

Here's an overhead view of the studio (not realy to scale) of how we did these tests. The black #s refer to the choice of arrangement.

You need a pretty crazy camera angle to make this microphone stuff look even sort of interesting. But this picture looks almost exciting, right?

You better believe Justin reads this book over and over like crazy when it comes time to record. It is dynamite.


[ Archives ]


[ Overview ]

[ Basic Construction ]

[ Acoustics ]

[ HVAC ]

[ Electrical ]

[ The Grand Vision ]

[ Tool Reviews ]

[ Links and Resources ]