PDF Version of Guy and Jeff's Paper
by Brad Olson
Why Use A Dual-Diaphragm Mic?
Chicago Section AES member Guy Torio from Shure Incorporated spoke on Monday November 20th. He talked about research he has done in the area of dual diaphragm microphones, similar to that which he presented at the national AES convention last September (AES preprint #5179).
The concept of dual diaphragm mics is not new. However, their work, Shure's Guy Torio and co-worker Jeff Segota attempted to develop intuitive lumped-parameter circuit models to explain the operation of such microphones. After coming up with dual diaphragm mic models, they compared modeled results to measurements of actual prototypes, in order to confirm the viability of their models. Models and data for single diaphragm capsules of similar topology were also evaluated for comparison.
The single diaphragm capsule is modeled as a simple R-C circuit. The circuit elements are voltage sources denoting the front and rear pressure into the microphone, capacitive elements representing the diaphragm and acoustic cavity compliances, and a resistor which represents the acoustic resistance (ie. delay) of the rear port.
The gain and directional response of such a model can then be seen as the sum of two first order filters, one high-pass, and one low-pass, with the same corner frequency. For the purposes of analysis, it is very easy to look at two limiting cases, one at high frequencies, and another at low frequencies. At high frequencies the single diaphragm unit has a cardioid response, independent of source-to-mic distance. At low frequencies, the unit has a proximity effect, making gain inversely proportional to source distance. The polar response shape becomes bi-directional or "figure-eight" in close proximity, while it remains cardioid at greater distances.
In contrast, the dual diaphragm mic is modeled as a similar R-C network, but with two extra capacitors to represent the second diaphragm compliance and the second acoustic volume now present (note that the acoustic resistance is placed between the two cavities in this model). For simplicity, it is assumed that the extra back compliances are of the same values as corresponding front compliances.
Once again, the response is analyzed as the sum of first order low and high pass filters with a common corner frequency. However, one effect of the second diaphragm and cavity are to scale the corner up by one octave over the single diaphragm case. The new system still a cardioid response at high frequencies, but at low frequencies, interesting changes have occurred. The gain and directionality changes with distance are more complex. The modeled response now becomes omni directional at great distances, bi-directional at small distances, and cardioid in between.
Guy presented data from a real prototype to verify these models. He noted that in order to avoid introducing additional variables, he made the single diaphragm mic from a dual diaphragm mic with the second diaphragm cut out. This was supposed to remove the extra compliance elements without introducing any new parameters.
The intermediate distance, where the dual diaphragm mic acts as a cardioid is given by the product of the front-to-back delay (in distance) and the diaphragm compliance, divided by twice the cavity compliance value. In the example presented by Guy, the compliance ratio is about 31, and the delay is 3.2 cm, so the cardioid response occurs in the vicinity of 0.5 m. Thus, when he measured the dual and single diaphragm systems at three distances, he chose a "far-field" distance of 8 feet, an "intermediate" distance of 2 feet, and a "near-field" distance of 6 inches.
Polar curves and response plots generally concurred with the models. The dual-diaphragm units had a slightly supercardioid response at low frequencies in the near field, rather than a true figure eight response. However, this was also predicted by the model. Despite the fact that curves were taken at various distances to verify the models, Guy pointed out that the microphone is really intended for use in near field situations, and those are the effects to focus on.
To conclude, Guy presented an excellent demonstration consisting of recordings made with both single and dual diaphragm microphones in the same situations. The results appeared clear to most listeners. A slight increase in the low frequency response around a few hundred hertz was heard when the dual diaphragm mic was used, and "popping sounds" were less likely. This made it clear that the modeled differences existed, and that the dual diaphragm mic would be considered an improvement when used in the proper application. The best application is likely to be a near-field vocal recording situation.
The Chicago section thanks Guy for his informative presentation.
