We had an earthquake in San Diego at about 11:55 pm on October 30th, 2001. It was a 5.1 magnitude, and the epicenter was in Anza, CA. I felt it quite a bit. The water in my glass visibly moved, and my computer monitor shook noticeably. It went on for quite a while, but I can't guess exactly how long it was. According to some seismograms I found, it lasted for over a minute. It was fairly disturbing, but also intersting at the same time. However, I probably would have felt better if I had been on the ground floor of my house.
I decided to save the seismograms I found, since they update automatically. If I didn't save them then the "snapshot" would be lost. They are from University of Arizona's seismometers. I added an explanation of what each graph is.
This displays the broadband vertical data. This is basically a "recording" of all the frequencies "heard" by the seismometers. The typical range for earthquakes is 50 Hz to 0.001 Hz. The frequencies are measured by their period, with one second being 1
Hz.
This displays the long-period data which will clearly display large (magnitude greater than 5.5) earthquakes. The long period waves are those with periods greater than 10 seconds -- the powerful ones. This graph will show the really big quakes very well since all the short period (or high frequency) stuff is filtered out. You don't want to be around when there's a big squiggle being recorded
on this graph...
This displays the high-frequency (short-period) data. This will also record mining activity, underground nuclear testing, and the like very well.
That's all I could find from UofA. If I still lived near the Gould-Simpson building, I'd be able to go in tomorrow morning and grab a real paper seismogram. I used to have one from the Northridge quake and one which showed a Soviet underground nuclear test taped to my fridge.