I didn’t find a technical description of the antenna on the SparkFun AS3935 Lightning Detector in the product description and tutorial.
Is the antenna on the AS3935 Lightning Detector directional?
The datasheet https://cdn.sparkfun.com/assets/learn_t … _EN_v2.pdf doesn’t specify, so I would assume that orienting the sensor to where it is “up” (pointing to the sky) might be the optimum position
The documents do mention the importance of ensuring you have the correct setting (default= indoor mode) and calibration, though!
Thanks. I will set up my own lightning generator to test directionality. The shape of the transducer suggests directionality.
It’s essentially just a coil of wire and shouldn’t be directional.
According to the Coilcraft MA5532-AE datasheet entitled RFID Transponder Coil https://www.coilcraft.com/en-us/product … hz/ma5532/, the transducer is described as “wound on a plastic base,” making it a multi-turn dipole. The overall coil diameter appears to be nor more than 3.5mm, which is much smaller than one half the wavelength (300m) of 500kHz (600m), making the MA5532-AE a short multi-turn loop dipole. The datasheet mentions that the MA5532-AE was developed for the austriamicrosystems AS3935 Franklin Lightning Sensor IC.
I conclude that the lightning detector antenna is directional. I expect the directional pattern to look something like this:
@dbpage: You are correct: a loop antenna is highly directional. For maximum response, the magnetic field lines of the propagating wave (which are perpendicular to the direction of travel) will be perpendicular to the plane of the loop antenna. In other words, for maximum signal the plane of the loop should be parallel to the direction from the source to the receiver.
I can’t tell from the illustrations of the Sparkfun board or the Coilcraft site what that plane is, though. If you can set up a spark gap to generate some “local lightning” it should not be hard to verify the directionality with respect to the PCB.
It would be easy enough to replace that tiny antenna with a much bigger 100 uH wire loop and the result should be more sensitive (sensitivity increases as the loop area).
You might find this paper on classification of VLF/LF lightning signals of interest: