All my troubles with early seismometers and accelerometers can be explained with this figure.
To the left is the full dynamic range of earthquakes. You really need to record that, if you want to get anywhere new with earthquakes. My shitty 8 bit (something like 80 db) accelerometers could only record the very top of that range. My old 'velocity coil' seismometers could only get the very bottom. As well as the dynamic range, we have the frequency range, and again, these instruments were very limited.
That left the middle totally uncovered! The coil seismometers only measured some jiggly-piggly form of velocity, and could never be related to the accelerometers, which measured low-frequency acceleration as defined on the great soil basins of California.
This was a mess. As an example, consider the 1988 Saguenay earthquake. This triggered tons of standard clockwork accelerometers that were placed willy-nilly in buildings. The results were all over the map, with something like a factor of 10 between the highest and lowest at equivalent distances. Totally useless.
-next, one day, when I feel like it, but nobody really likes this series, so I may fade out....