Posts Tagged ‘stereo’

Mid-Side recording and Microphone Sensitivity

November 1, 2009

Small technical backgrounder this time.

When recording stereo it is essential to have equal gain settings for the microphones at the risk of the stereo image shifting to either left or right. More gain on the right mic channel will shift the image to the right and the other way around. Equal gain settings are especially important in XY, ORTF or any other spaced or coincident microphone positioning.

With mid-side recording the gain setting for the mid and side channels allow you control the stereo spread. More gain on the side channel compared to mid will widen the stereo image. More gain on the mid channel will narrow the image. Ultimately, no gain on the side channel will leave you with a centered mono signal from the mid mic.

Reading gain or trim pot settings is easy enough but you usually end up with differences in signal strength if you don’t take into account microphone sensitivity even with an equal gain setting on your recorder or mixer. Unless you have a matched pair, microphones have different sensitivity characteristics.

A typical MS setup is the Sennheiser MKH60/ MKH30 microphone pair.

The MKH60 has a sensitivity of 40mV/Pa, the MKH30 has 25mV/Pa. With equal gain you will end up with a much stronger mid signal and consequently a narrow stereo image.

The calculation to compensate for sensitivity is as follows:

±dB = 20 x log(MmV/SmV)

where

MmV = sensitivity of the mid mic in mV/Pa
SmV = sensitivity of the side mic in mV/Pa

In our example we need 4.1dB more gain on the side channel compared to the mid channel to get equal signal strength on both of the recorded channels.

This is easy to realize on mixers such as the Sound Devices 302. With MS stereo linking set trim pots allow you to set gain for the mid and side channels separately. The input 2 fader controls the overall gain for the pair.

Mid/Side + Mid recording

April 29, 2008

No need to introduce the Mid/Side technique developed by Alan Blumlein as a means of obtaining a stereo recording.

While the the technique allows for an omnidirectional microphone to be used for the mid channel one most often sees a microphone with cardioid pick-up pattern being used. This is also the case in my own recordings. For field recordings this usually is a shotgun mic like the Sennheiser MKH60.

Not all cardioids are created equal however. Their frequency response differs greatly and by design cardioids do not pick up low frequencies very well. The same holds true for the figure eight mic used for the side channel. The excellent MKH60 and MKH30 have a frequency response of 50Hz to 20kHz and 40Hz to 20kHz respectively. Some microphones show a more or less important roll-off on the low end.

The relative lack of low frequency information which is the result of this frequency response can seriously influence the recording quality. Recordings might appear particularly thin. In some situations, e.g recording of bird calls, this might not be a problem. In other situations the recording is just not going to be realistic. A typical example is the recording of high waves rolling onto the beach. There is lot of low frequency information in the sound of waves hitting the beach.

To solve this I thought of extending the classic MS technique by adding a low pass filtered third mic. This third mic preferably has an omnidirectional pick-up pattern as an omni captures the low end much better then do cardioids.M/S+M recording

As I said, the trick is to low pass filter the signal. This can be achieved in post production or in the field by inserting a low pass filter such as the Schoeps LP 40 U in the signal chain before recording.  There is of course a requirement to have three input channels available. One for the mid mic, one for the side mic and one for the low pass filtered omni mic. I call this third channel the +M channel because it needs to be panned center in the mix.

In the field I mount a DPA 4060 just below the mid-side pair but since only low frequency information is retained I would think that it can be postitioned anywhere close to the main coincident mid-side mount. At low frequencies the time delay resulting from the distance between the mics becomes irrelevant.

One needs to take care not to turn up de volume trim on the +M channel too high. Low frequencies eat up the available dynamic range very quickly. The whole idea is to get a natural sound character.