The EMP sensor is essentially a radio receiver that is most sensitive in the VLF portion of the spectrum. The energy released from a lightning discharge is greatest in this portion of the spectrum. A simple low pass filter followed by a darlington-pair of transistors makes up the receiver. Nominally, it takes a pulse that will produce a Vin >= 1.4V to turn on the output transistor. This is due to the two forward Vbe drops of 0.7V on each of the transistors used in the darlington-pair. Not only does this improve the gain of the circuit, but also serves to offer a large degree of immunity to man-made noise and interference (i.e., under normal circumstances only a lightning discharge is capable firing the output).
Since the receiver is very similar to the old “crystal” radios, an efficient ground plane is required for operation. In fact, the effectiveness of the ground plane will be the greatest factor in determining the sensitivity of the receiver. The second most important factor is height. The higher the sensor (i.e., the antenna), the greater the sensitivity.
From the above discussion, it can be concluded that sensitivity is dependent on height above ground and the presence of an adequate ground plane. A height of 20 to 30 feet seems to work best. An effective counterpoise, or ground, can be made using ground rods. However experience has proven that the best performance of the sensor is when the steel frame of a large building is used as the counterpoise.
Note! - If you are using the X1W-4 EMP Sensor kit, refer to that manual for installation instructions. The X1W-4 solves many of the installation issues dealing with the coax/wire connection. The following discussion is intended for installing the LSU-30 sensor from Stormwise.
Since the sensor is essentially a radio receiver, the most important part of the installation is the ground system. The sensor must be mounted on a metal mast of at least 10' in length or the mast assembly must be adequately bonded to a grounded metal structure (tower or the building's steel frame). The easiest method of mounting is to use two stainless steel hose clamps on the lower (wider) body of the LSU-30 sensor. The clamps will conform themselves to the sensor and the pipe to which it is attached and will hold the sensor in place without any problems.
Use RG-58 or RG-59 coax to connect the sensor to the X1W-1 inside board
(actually most any coax will work here). The red lead from the LSU-30 is
connected to the inner conductor on the coax while the black lead is connected
to the shield. An additional short piece of flexible wire is also connected to
the black/shield connection. This lead is attached to the mast just below the
point where the sensor is attached. The point of attachment can be another hose
clamp, but the use of a ground solder terminal screwed securely to the pipe is
The entire outside connection should be carefully sealed with coax seal
and heat-shrink tubing. A loop should be used to help prevent water from
entering the coax. While the weather seal should be adequate, the added
insurance not only makes sense, but provides a strain-relief for the cable
as well. See the pictorial for details of this and the mounting technique.
Inside, attach an RCA type phono plug to the end of the coax. If you can find them, the type that is made for coaxial cable is preferred.
To test the sensor, use a fireplace/grill lighter. Pull the trigger with the end close to the midpoint of the sensor. The spark produced by igniter should be adquate to fire the sensor. Be careful not to melt the PVC housing! Actually, a grill light with a spent cartridge would make a good choice for the test instrument.
Example Installion 1:
Example Installion 2:
Example Installion 3:
Interpreting the Data
Sensor response will vary greatly from one location to another. It is not unusual, especially during the summer months, for the sensor to hear several strikes every hour. The higher the sensor is and the more efficient the ground is, the more strikes the sensor will detect.
So how do you interpret the data? There are no hard and fast numbers. At the EVVWXN location, if the sensor trips, there is an active storm within about 30 miles of the QTH. However, at OWBWXN, there is almost constant activity. At NEWWXN, the threshold for when a storm is in the immediate area is around 200 hits in a five minute period. If you graph the strike data, it will help put things into perspective. Looking at numbers on a page can be misleading. If it helps, imagine listening to a ball game on AM during a summer evening. You may hear static crashes from time-to-time faintly in the background. As the storm draws closer, the amplitude and frequency of the static crashes increases. It's the same here. Over a period of time, you will begin to get a feel for how your sensor performs.
Neither of the following has been tried, but should work.
Although untested, there is a homebrew version simlar to the Stormwise sensors available on the web at http://homepage.mac.com/tbitson/weather/bslam/. If you elect to use this circuit, be absolutely certain you do not include R4 and the 9V battery. This has already been allowed for in the X1W interface. Diode D2 is not required, but won't hurt anything to leave it in.
Bob Radmore, N2PWP, in the April 2002 issue of QST, also has a lightning detector circuit. This too can be modified to work with the X1W-1 by removing the indicator lamp and connecting a short coax with the inner conductor connected to the resulting open collector of the output transistor and ground. See this on the web at http://www.techlib.com/electronics/lightning.html.