Since you seem fairly knowledgeable about this subject, perhaps you could answer a related question I've been wondering: Would an effective low-cost way of measuring low frequencies (let's say ~20hz) be to use a subwoofer speaker (used as a microphone) hooked up to something that would amplify and recording of the signal? My reasoning is that they're naturally designed to work with low frequencies to begin with (more so than a typical microphone anyway). I would also reason that you could also do a frequency response test over a range of known frequencies and power levels for the specific subwoofer to begin with, and then write a specific software filter to normalize whatever you record from it to get the 'true' sound. Does that make sense?

Edit: Also, note that I mean doing hardware amplification of the electrical signal in the wire coming directly from the motion of the magnet of the speaker.