We took apart and examined one of the Guralp seismometers this afternoon. For the most part we think we understand how it works. The horizontal sensors are a little more confusing, since we didn't end up finding the moving masses. The vertical sensor is a flat rectangle, hinged at one edge. There are capacitive sensors above and below the rectangle. The hinged end is connected to a leaf spring.
The PCBs are packed full of old-school 80's components. We probably need an actual schematic to figure out where the preamp circuit is, which is what we'd want to think about fitzing with, if we were to try to improve the noise of the seismometer. For now, we put it all back together, and back out on the granite slab.
There was a wee bit of confusion when putting the N/S marker-spikes back on as to where they should go. The solution is that the handle of the seismometer is aligned with the North/South axis, so the spikes should be aligned with the handle. The lid of the seismometer is uniquely aligned to the stuff inside by the ribbon cable connector, as well as the holes in the lid for accessing the centering potentiometers. So, align the lid to the pots, and then align the spikes to the handle.
Photos are on Picasa.