Viewing Lightning RF Bursts with an RTL-SDR
Lightning produces fairly wideband bursts of RF energy, especially down in the VLF to HF frequencies. Detecting these bursts with custom radio hardware is how lightning detection websites such as blitzortung.org work.
It is also possible to detect lightning using an RTL-SDR that can tune to to HF and lower, such as the RTL-SDR V3, or an RTL-SDR with an upconverter. Over on his blog Kenn Ranous (KA0SBL) has uploaded a short post showing what lightning bursts look like on an RTL-SDR waterfall. He uses an RTL-SDR V3 to tune down to the LF – MW frequency bands and looks for wideband pulses of noise which indicate lightning.
It would be interesting to see if this type of detection could be automated with DSP so that a similar service to Blitzortung.org could be created.
HAMSCI is working on a project for a Space Weather Station. Lightning detection will be incorporated. The exact SDR used is undefined so far. ELF/VLH, and HF will be utilized initially to identify the appropriate waveforms, and then some form of DSP pattern matching will be undertaken to speed up identification. Using GPS timing and geolocation, Time of Arrival (delay) trilateration is the most probable geolocating system. Eventually, incorporation of the VHF/UHF/SHF components of cloud-cloud and intracloud flashes will be incorporated.
I happened to see this article a few hours before storms rolled into my area. Using Airspy with a Spyverter in GQRX: http://imgur.com/a/5ePGx
And what did you use for an antenna?
Always enjoy these interesting applications for the SDR-dongle, which due to the use of spectral displays and waterfalls makes it a powerful tool for the tinkerer as well as the listener. Nice job Kenn.
As an aside, DB Gain is a great pen name hi hi.
Thanks! We had a storm early this morning with a lot of cloud to cloud lightning. I used CubicSDR and turned the waterfall spped to max 1024 lines per second so I could take a closer look at the time domain.
In that image, the waterfall is ~1 second top to bottom.
I did a science fair project on lightning detection as a weather forecasting tool in high school.
My method then was to tape a few minutes of AM broadcast band and then count the number of pops I heard. I would then compare it to current and future weather conditions.
With an SDR and modern software this could probably be automated, with greater precision.
Can please someone dirct me to a site that explains how to use the HTML tags of this site?
Did you put angle brackets (“”) around the tags?
Let’s see if I can get a blockquote to work:
Guess not. Something is eating the angle brackets.
Why not using cite?
blockquote cite=” I’m using an RTL-SDR version 3 dongle and GQRX in Linux, but any SDR and spectrum analyser software that can tune VLF should do it”
– Some questions, hopefully someone read the answers in other post from hi, so I can understand the signal strength involved that he received
? Does he use Direct Sampling HF Mode? If yes how sensitive is the Dongle below 600 kHz?
blockquote cite=”so I chose to monitor 0 to 600kHz. The antenna is a 1/2 wave trapped dipole for HF but any bit of wire at least a few meters long will work.”
?Has anyone read in another post by him what antenna dimensions he uses?
At 600 kHz the wavelength is already 500m, increasing the lower the frequency the wavelength increases to unhandelbar dimensions unless you are military.
SW is for HAM’s about 160 m (~1.8 MHz) to 10 m (~29 MHz), so it’s quite a difference if his 1/2 wave dipole is 80 m or just 5 m.
While its nice to see the spectra, I wonder if the detection would be not easier using an active loop, followed by an IF amplifiier/limiter from a NBFM demodulator (e.g. NE-604 or similar) and a diode detector.
Then there is href=”lightning protection” title=”https://www.rtl-sdr.com/forum/viewtopic.php?f=6&t=1871#p4945″ to consider, as I wrote just a short time ago in reply to a question.
I had a 70ft wire that fed a transmatch via a balun and twinlead for a while. One day I noticed after hearing a zap that seemed to be coming from the transmatch that the cross needles of the swr meter in the transmatch were slowly rising to their stops and then I’d hear a pop and they’d drop back to zero only to slowly build again. Clear sky, dry, nice day. The closest tstorm was 100 miles away. The storm was charging the wire and hence the swr circuitry until the tuning cap arched over. It was a fairly reliable method to detect tstorms but dusty wind and snow did the same thing.
Long wires with insulation and wind blowing across them can lead to significant static charges. I suggest an RF choke or a high value resistor (say 100 k ohm) shunted to your station ground. This will discharge the static on your antenna.