Living in California, earthquakes are a part of life. Sooner or later, you'll feel one, and the first time can be pretty scary! If it's a small earthquake, which thankfully most are, it's actually hard to know for sure until you check the news later on. Sometimes a painting shakes on the wall, or a hanging lamp starts swinging, or you hear a rumbling sound - most small earthquakes are much more difficult to notice than you would expect.
In order to detect low magnitude earthquakes, we wanted to build a very sensitive Seismometer. This project was quite easy, mostly some carpentry, and the detector was a PocketLab sensor, using the magnetometer function. An iPad was used for data collection.
-wooden frame: one 8' 1x4 and one 8' 2x6
(I was able to get all this scrap lumber from 1 pallet)
- About a dozen assorted screws and metal straps
- Medium size magnet (about 0.1Kg)
- PocketLab sensor (www.thepocketlab.com)
- iPad or other smart phone for wireless connection to the sensor
Step 1: Making the Pendulum
Finding a design was easy, the very first google search produced a diagram straight from the USGS site. There are many other designs, but this one looked the easiest. It is basically a pendulum.
We happened to have an old pallet in the back yard from a shipment received long ago. We tore it apart and salvaged enough lumber to make a base, pole, and beam.
We bought about $15 worth of screws and metal straps at Home Depot to fasten the wood and brace it.
Step 2: Making the Detector
This was the fun and interesting part.
We had some luck at Home Depot, they have a large display of magnets. Most of them are for holding things or retrieving dropped metal objects.
We picked one that was relatively heavy and had a handle for attachment. It was rated as a "25lb magnet", meaning it could lift a 25lb ferrous object. We weighed it, and it was 0.091Kg.
For the detector, we decided to use the magnetometer function in the PocketLab. PocketLab has other sensors, the main other candidate being the accelerometer, but we couldn't figure out a way to use this as a recording device. The magnetometer seemed much more obvious.
We made a pendulum with the magnet as the mass, and a 70cm length of string. The equation for the period of a simple pendulum is:
T = 2π√(L/g) where
T is the period in seconds
L is the length of the string in meters or
g is the acceleration due to gravity (9.8 m/s²)
We get 1.68s, another way to look at this is 1/T which is the frequency f = 0.59Hz
The way the detector works is that if the magnet (mass) moves, the flux moves with it. The magnetometer is fastened to the base of the pendulum, so when the mass moves, it simply measures a change in flux.
In the graph, the magnetometer is measuring the flux change in 3 axes, and the resulting sine wave corresponds to the pendulum mass swinging back and forth. As a check, the graph shows a period of a bit less than 2 seconds, so this checks with the period calculation.
An interesting note, it turns out the magnetometer is very sensitive. Sometimes we couldn't even see the magnet moving, but still the magnetometer graph would show the flux changing.
Step 3: Collecting & Interpreting the Data
We put the seismometer on the concrete floor of the garage. It is very stable, and after a while, the graph would settle to the noise of the magnetometer, which was about +/- 3uT (micro Tesla).
We waited around a couple days, but didn't get a decent earthquake, so in the end we tested it by dropping something heavy on the concrete.
The real-time graph quickly jumped up to more than +/- 100uT. It was barely possible to see the magnet moving, but the periodic motion is clear in all 3 axes.
This is easy to see in the screen grab of Excel data.