Radio Microphones and Comb Filtering

There’s always lots of discussion about which brand lav mics and wireless systems sound the best. And of course each system has its pluses and minuses but too often people think that buying the most expensive lav mic will solve all the problems and that simply isn’t the case.

An inexpensive lav placed in a good position will outperform a top of the line microphone placed in a poor position every time!

Classic example of poor mic placement…

Classic example of poor mic placement…

I have used the G3 kit ME2 microphones (notoriously not great sounding mics) to great effect in the past when they have been placed in a good position - one obviously clear of any rustle and not too far under the chin, nor too far away from the mouth.

But one element that I find that is so often overlooked is the effect of the voice bouncing off the chest. If you were to put you ear right up to someone’s chest it would sound as though the voice is also emanating from the chest. Some people even call this your chest voice. This isn’t entirely true however. The chest itself doesn’t actually create a noise. What’s happening is the voice is leaving the mouth and bouncing off the chest.

Now, why does this matter to a sound recordist? Well when placing a microphone we will sometimes go for mounting on the clothing for a number of reasons. And this can sound really great depending on the placing but sometimes it seems to randomly sound quite strange. I’ve heard other mixers call it “muffled” or “buried”. But what is actually happening is an effect called comb filtering.

Here is an example of comb filtering caused by two microphones placed too closer to each other.

Here is an example of comb filtering caused by two microphones placed too closer to each other.

Comb filtering is a phenomenon whereby a sound source gets delayed slightly, usually by two microphones at slightly different points in space being summed. This causes the signal to be slightly out of phase and hence creates dips and peaks in the spectrum - resembling a comb shaped frequency response. Most people will have heard this when summing a boom and lav track together to mono. As the two signals are at different distances to the talent’s voice they arrive at different times and therefore create this effect.

The other way this can occur is with a single point microphone but the source voice get’s effectively doubled due to a reflection… Like off a chest.

As you can see from this picture above the sound waves travel from the mouth and bounce off the chest then arrive delayed at the microphone.

As you can see from this picture above the sound waves travel from the mouth and bounce off the chest then arrive delayed at the microphone.

This can be clearly demonstrated by holding a lav microphone and talking into it while moving it closer and further away from your chest. You will hear different thinning out and muffling effects at different distances from your chest.

Some of my observations are as follows:

  • The majority of the change in frequency seems to be happening in the low to low mids. I believe this is due to the fact that lots of the high frequency energy will be dissapated by the chest as it bounces off (especially in males with hairy chests).

  • Therefore microphones with more low end appear to be affected more. This is somewhat subjective but I have found that due to the extended low end of most DPA capsules (compared to say a sanken Cos-11d) this effect is emphasized somewhat. This, I think, contributes to the fact that a DPA is often harder to find a good place for.

  • Women are often more prone to this than men. The reasons for this are twofold. For one, women usually have a larger space between the skin on the chest and their clothing. Secondly they also typically have smoother and less hairy skin.

So how does this help us? Well I’ve found since learning about this phenomenon that I’m much more likely to start with a microphone placement that is directly on the skin, rather than on clothing (especially if that clothing is hanging a distance of 3-4 inches away from the skin). Ultimately when setting lav placement you'll need to use your ears and adjust until it sounds good, but understanding why a microphone sounds muffled or incorrect is valuable in deciding what trick to try next.