Tuesday, November 24, 2009

A simple hydrophone

Here's a blast from my past: a piezo-electric hydrophone. This was my very first research gadget that I ever built. I put it together roughly 10 years ago for my senior research project during either my 2nd or 3rd senior year of undergrad.

The goal was to listen to predatory whelks feeding on mussels in order to estimate handling time. Because you can't really see the business end (the radula) of a whelk when it's feeding, the only good way to determine when a whelk is drilling is to listen for it.

Note the drill holes in the Mytilus edulis shells around the Nucella lapillus shown above.

The hydrophone is based on a piezoelectric noisemaker, commonly found in things like home smoke detectors and Hannah Montana singing greeting cards.

As sold, these discs make a noise when you apply a voltage across the central ceramic disc and the outer brass disc. But thanks to physics, you can do exactly the opposite as well, produce a voltage across the ceramic disc and brass disc by making noise. The hydrophone exploits this property of the piezoelectric noisemaker to turn it into a microphone.

I can no longer find the paper that I read that originally described this setup, but someone at the Underwater Acoustics Group of the Loughborough University of Technology put together this helpful pdf that describes the construction of the same unit: A cheap sensitive hydrophone for monitoring cetacean vocalisations.

The hydrophone essentially consists of one of these piezo noisemakers sandwiched between two pieces of plexiglass, with an o-ring to provide the watertight seal and make space for the wiring.

I used a long mono headphone extension cord, as shown in the top picture, for my wiring. The mono cord has two conductors in it. The central conductor gets soldered to the ceramic central disc (visible in the picture above), while the outer conductor gets soldered to the brass disc (not visible above, but see the pdf for a diagram).

The wire is passed through a drilled hole in one of the pieces of plexiglass, which is sealed up with some sort of silicone caulk. 2-ton epoxy might be good here as an alternative.

The other side of the piezo disc then needs to be epoxied to the 2nd piece of plexiglass. This provides a direct connection to the outside world so that sound is transmitted efficiently from the water to the disc. Spread some 2-ton epoxy between the disc and the plexiglass, and let it cure.

The two pieces of plexiglass then get sandwiched together with a thick o-ring (something like McMaster-Carr part number 2418T197) between them to keep water out. The bolts in each corner need to be tight enough to slightly compress the o-ring. In my naive youth I just went ahead and cranked things down until the plexiglass was ready to snap, but it always worked fine.
Once everything is put together, you should be able to just plug your cord into the microphone input of a computer and start recording sound. It may be necessary to crank up the amplification quite a bit due to the very low level (low voltage) signal coming from the hydrophone.

When I was actually recording whelks drilling, I found that it helped to use some 2-ton epoxy to glue one of the valves of the mussel directly to the outer face of the plexiglass, right on top of the piezo unit. When a whelk started drilling the mussel, this transmitted the drilling noise (more of a scraping noise really) directly to the hydrophone, helping it stand out from the background noise of water moving through the seawater system. This same system could be glued to a rock to listen to the grazing sounds of herbivorous gastropods. Just don't do anything lame like listen to whale songs. Study something interesting that doesn't have a backbone!