Reply to Comment
When most people record audio in stereo, they usually go to the most common mic configurations, such as a matched pair of microphones arranged in an X/Y pattern, or possibly those same matched mics in an ORTF pattern for a wider stereo field. These methods are the most obvious because they closely replicate the way our ears hear. However, there are no hard and fast methods for capturing stereo sound. You may have heard of the Mid-Side (MS) microphone technique, but never tried it because it seemed too complex. The technique does require a bit more thought, but the advantages are well worth the effort.
|Mid-Side Pattern||X-Y Pattern|
Mid-Side is a coincident technique, meaning that both mics are placed as closely as possible to each other and the stereo image is created by differences in loudness rather than time delays, on which other stereo miking methods rely. Mid-Side uses a mid mic, which usually has a cardioid pattern, but can also be omni. The mid microphone is aimed at the source, while a bi-directional mic (figure of 8 pattern) is used to pick up sound from the sides. To create a stereo image, the signals must be decoded by either a mixer, or a dedicated decoding plug-in.
The MS miking technique theory was patented in 1934 by Alan Blumlein, but was not used widely until Danish State Radio engineer Holger Lauridsen did his stereo experiments. The technique has long been used in broadcasting because the resulting recording is mono compatible, which makes it a great method for studio, concert and location recording.
Flexibility: Typical stereo recording methods such as spaced pairs, ORTF or X/Y result in a fixed stereo recording with little adjustability options in post, and collapsing X/Y to mono can result in phase issues. With MS recording there is infinite adjustability in post, in both the width and depth of the stereo field.
Mono Compatibility: Because the signal from the side mic is copied to a second track and put 180º out of phase, then panned hard right and left with the mid-mic signal panned center, when all signals are panned center the side channels cancel each other out, leaving only the mono center signal.
Matched Pair Unnecessary: Because the stereo information comes from only one mic in a bi-directional, or “figure of 8” polar pattern, there is no need for the mid mic to be the same model or size.
Anywhere a stereo recording is appropriate, the MS technique will work. However, it’s important to have proper mic placement in any recording, because you can’t fix bad placement in post. Common places for MS technique are in the studio for drum overheads, piano, acoustic guitar or small vocal groups or string sections. MS is also great for broadcast audio, ambient or nature recordings, concerts, choirs or any kind of location recording.
As far as capturing your MS recording, you only need two mics and two tracks to record. One mic must be bi-directional (figure 8 polar pattern) and the other is most often directional, or cardioid pattern, although Alan Blumlein described using an omnidirectional transducer in his original patent. Omni will give you a wider-sounding recording with more low end. Feel free to experiment with super or hypercardioid, too!
If you want more simplicity in the form of an all-in-one recorder, the Zoom H2n features MS recording at the turn of a dial. The Tascam DR-680 portable recorder has added MS recording with firmware version 1.2, allowing you to record to tracks 1/2, 3/4 and 5/6, each as its own MS pair, and the DR-680 will decode the audio for proper playback. You can also find single microphones that are designed for MS recording such as the Sony ECM-957Pro or the Sanken CMS-7S.
There are also microphone preamps with MS circuitry built in, such as the Grace Design M201. To properly decode the recording you will need either a DAW (Digital Audio Workstation) such as ProTools or Cubase, an analog or digital mixer with phase reverse switches, or a dedicated mic preamp with a mid-side decoder built in.
Make sure the “side” mic is set to the figure of 8 polar pattern, and that it is placed so that the lobes of the pattern are facing either side of the sound source with what is normally the front side of the mic facing to the right. The “middle” mic will usually be in a cardioid pattern. Place the mid mic so that it is facing directly at your sound source. Get the two mics as close as possible without allowing them to touch. Then you just position the mic configuration as you would any other mic, of course paying attention to proximity and wall reflections, etc.
MS decoding requires what is called a “sum-and-difference matrix”. To get the sum you add one side signal to the mid signal, and to get the difference you subtract the other side signal from the mid signal by shifting its polarity 180º. It’s easier to do than it sounds. The most common way most people will be decoding the MS recording is with DAW (Digital Audio Workstation) software. If you already use a DAW for recording and mixing, nothing else is needed. You’ll use the DAW’s mixer to decode the MS recording by opening up a project with three tracks: one for the mid microphone, one for the side microphone and a third to put a duplicate of the side in. After you’ve recorded your source, whether within the DAW, or from a portable recorder, import the two mono files into the project and label them “Mid” and “Side” appropriately (one of your two Side tracks will still be empty at this point). Make sure the Side tracks feed into a stereo track or subgroup channel, allowing you to pan easily (you will have to create these tracks with your DAW software).
Here’s where the decoding magic happens. Duplicate the Side track with the recording in it, then reverse the phase of the original. Some apps have a phase reverse switch, and some allow you to do it with a simple process. Next, you pan the Side tracks hard left and hard right, and if possible, link their volume faders so you always have the same level. Start with the Side tracks turned all the way down, and play just the mid track. It sounds like mono. Then bring up the side tracks as the recording plays and you will notice the stereo image become apparent. If you want a wider stereo field, turn up the Side channels, or, if you want a narrower stereo field, bring the Side tracks down. Pan the subgroup of the Side tracks to change stereo placement of the instrument. If you mute the Mid track, and center each of the Side tracks, you’ll notice that they completely cancel each other out and you hear nothing. Return the pan of the Side tracks hard left and right, and experiment with the levels until you arrive at a mix that suits your taste. It’s that simple. Click here for illustration
Mid-Side Decoder Plug-ins: The setup is similar to the DAW example, but instead of duplicating the side track, you feed both mid and side tracks into a stereo channel and insert a Mid-Side decoder plug-in. A popular free decoder is Voxengo MSED. Its simple layout and controls allow you to quickly decode your MS recording without making duplicate tracks and putting one out of phase, which can get tedious if you have a lot of MS recorded tracks in your project. Some DAWs have an MS decoder built in
We’ve only touched on the basics of mid-side recording in this article. The best way to gain a more in-depth grasp of the concept is to break out some microphones and start experimenting. Once you get in there and do it once, you’ll realize the power of this simple technique.
Thanks for reading this B&H Tips & Solutions article. If you have any questions about these techniques, or if you want to share your own experience with them, we’d love to read about it in the Comments section, below.