An Automated System for Detecting Meteors


By Rick Boozer



Two meteoric events that were visually confirmed.


One of the pieces of astronomical lore I have heard for many years is that an FM radio can be used to detect meteors.   People have claimed that, having tuned a radio to the frequency of a far off station, they could then detect meteors whenever the station’s signal reflected off of a meteor’s trail.  What would happen at such times is that the normally faint or even undetected station would suddenly blare out from the radio’s speaker.  In the following article I shall discuss my own observations of this phenomenon.  Ultimately, these researches led me to use my expertise as a professional software engineer to develop a system that automatically detects and records meteors entering the Earth’s atmosphere using a Windows™ based PC.  In addition, I shall tell you everything that you need to know to construct such a system yourself should you wish to do so.  Though construction of your own system will not require an advanced knowledge of electronics, the project will require the use of a soldering iron.


Before you proceed to the rest of this article, you should know that this online text consists of the story of the trial and error steps of the research involved in developing the automated meteor detector along with some tips for operating the detector.  The former developmental information is included for those readers who are interested in the research and development process, while the latter operational text is for those who actually wish to use a detector after they finish building it.  Readers who want to actually make their own detector can find instructions for downloading both the operating software and hardware construction instructions at the end of this article.  I mainly created the system as an intellectual challenge; therefore, I have only used it enough to verify that it does indeed detect meteors.  I leave it to others to use it extensively and determine its degree of usefulness.  Having made that point, here is the story about the underlying principles and the development of my automated meteor detector.


Lower frequency radio waves (such as those of the AM band and international shortwave) may be deflected downward by a special layer of charged particles within the ionosphere many miles above the Earth’s surface.  At night the above-mentioned layer is much higher than it is in the daytime, so that the waves bouncing off of it can go farther than they usually do.   Thus, during the evenings of my youth in South Carolina, I could listen to station WLS in Chicago.   Ham operators can make themselves be heard thousands of miles away for the same reason.


Starkly contrasting the behavior of low frequency signals, microwaves pass right through the ionospheric layer into outer space.   Transmissions in the television and FM radio bands belong to this class of electromagnetic radiation.  In fact, commercial FM frequencies lie between channels 7 and 8 in the television band.   Because such radio waves cannot reach a receiver that is over the horizon from the point of view of the originating transmitter, they are known as “line-of-sight” transmissions.


A Less Than Successful Start

My first investigative effort, involving just an ordinary battery powered FM portable radio, was less than spectacularly successful.  While observing the Geminid meteor shower a few years ago, I tuned the radio to a very faint and weakly received station in the manner that was recommended by others who claimed success with radio meteor detection.  The signal was so faint that, though I could tell that someone was talking or singing, the words were muffled and unintelligible.  Such a station, theoretically, would have just the tip of its transmitter tower above my horizon so that my radio would only receive a small fraction of the signal it would have gotten had the curve of the Earth not been in the way.  When a meteor appeared, some of the radio waves that would have gone into space were supposed to bounce off of the trail of ionized gas and metal particles generated by the intense heat of the meteor’s atmospheric entry.  At that time the voices and/or music in the radio broadcast were supposed to be heard much louder.   This result did not occur even once during the four hours during which some friends and I conducted observations, though we counted over three hundred meteors at a good dark sky site.  Not even the bright bolides brought a signal increase.


Later experimentation with the radio I used that night would reveal the reason for my initial lack of success.  It turned out that my portable was not tuned to a station that was over the horizon, even though the signal was coming in weakly.  Not being the strongest of receivers, the portable radio would only pick up some local stations well if its telescoping antenna was angled and turned just right.  After moving the antenna around to an appropriate position, the station that I had thought was distant came in loud and clear with constant signal strength. Hence, it was a lack of antenna adjustment that was causing the station not to come in well rather than its distance!


Detection At Last

I figured the answer to the antenna problem was to either get a more sensitive receiver or make the receiver that I already had more sensitive.  Deciding that the easiest way to implement the latter solution was to make the antenna longer, I bought a roll of picture hanging wire (eighty-eight cents at Wal-Mart!) and an alligator clip from Radio Shack.  I connected the alligator clip to a 20 foot length of the wire, then stretched the wire into an L shape about seven feet above the ground.  Finally, I attached the wire to the end of the radio’s fully extended antenna with the jaws of the alligator clip.  The idea behind the L shape was to insure that I could get a strong signal in any direction from either a local station or a meteor-enhanced over-the-horizon station.


Sure enough, local stations that before had only come in strong when the antenna was turned a particular way, now came loud and clear with the wire attached – with no adjustment whatsoever!   The question to be settled now was: would it pick up a meteor-enhanced station below my horizon?  Not being under any kind of deadline to find the answer, I chose a leisurely course of action.  During the hour to two hours per week that I spent reading, I would NOT tune the radio to the frequency of a faintly received station, but to an area of the dial where I normally heard nothing at all.   It was my hope that the long wire antenna had made the radio sensitive enough that a station whose transmitter tower was completely below the horizon might come in strong during a meteoric event.  I patiently waited for such an enhanced signal that would be loud enough to divert my attention from my reading.


I tried several “blank” areas on the dial on several different days before I found a frequency where, on occasion, a radio station would suddenly come in loud and clear and then slowly fade out over a period of several seconds.  But, I then noticed that I had even better luck than I realized.  There were two distant radio stations on that frequency: one to the north and one to the south!  For some reason there were times when the one to the north would come in stronger, while at other times the one to the south would be the louder.  If the sudden reception of these stations was indeed due to a meteor, there seemed to be only one explanation for this state of affairs.  When the meteor appeared in the northern part of the sky, the path from the northern station (up to the meteor trail and down to my receiver) was shorter than the path from the southern station (up to the meteor trail and down to my receiver).   Of course, when the meteor was in the southern sky, the southerly station was louder due to a correspondingly shorter signal path.


Making It More Convenient

While I had made significant progress improving the sensitivity of the radio, it was hardly a convenient way to detect meteors since I would have to constantly be close enough to the radio to hear the station suddenly come in.  What if I came up with a way for my computer to listen to the radio?  That way signal strength increases could be automatically detected and recorded 24 hours a day for me to analyze later at my convenience.   Instead of having the radio output its amplified signal to its speaker as it normally would, I would output the signal to an analog-to-digital port on my PC.  Some custom written software would measure the change in signal strength.  The stronger the signal, the higher the voltage the A/D port would see.  The next question I asked myself was, “Can any of the already existing peripheral ports on my PC be used for this task?”


Well, it just so happens that there is a nearly perfect interface on most PCs for this purpose.  The game port measures variations in electrical voltage caused by changes in electrical resistance when potentiometers are turned as the user moves a joystick.  Instead of measuring voltage changes from a joystick, I would read voltage differences from a radio hooked up to the joystick port!  However, even when the radio is not receiving a strong signal, the voltage it puts out is still higher than the highest voltage that the game port can measure.  Having a potentiometer in the circuit can allow the user to “tune” the output voltage of the radio down to a range that the computer can measure.  The simplest way to do that is just to open the case of a joystick and connect the output of the radio (in series) with one of the joystick’s potentiometers.   I give you directions on how to connect the radio to the joystick in the built-in help text of the aforementioned custom written software.   Because I have both full hardware assembly directions and instructions for using each program in the software’s built-in help, I am not including those instructions in this article.   Please note that the user of a laptop computer equipped with a USB port instead of a game port will need to make the additional purchase of an “analog game port to USB adapter”.



The assembled AMOS hardware: GIA, FM radio, and optional amplifier

Assembly and Use of the Automated System

I bought a cheap joystick and made it into an attractive peripheral device by merely mounting its innards in a black plastic box, putting knobs on each potentiometer and running speaker wires from each potentiometer to a mini-plug; thereby, allowing easy connection to the radio through its earphone jack.  There is a right way and a wrong way to connect the speaker wires, as you will see explained in the software’s built-in help.   I dubbed the repackaged potentiometers and wiring GIA, for “Game Interface Adapter”.


The software actually consists of three separate programs: an analog-to-digital port calibrator, meteor detecting/data archiving software, and archive retrieval software.  The first program is used when you are adjusting the voltage of the radio signal.  Voltage adjustment is made through turning the knob on the appropriate potentiometer until the program indicates that the correct voltage is obtained.  The second program performs actual meteor detection by plotting voltage variations in real time while archiving the data to your hard drive along with time stamps indicating when the observations were made.  A whole day’s worth of data occupies about 350K bytes of disk space.  The third program gives the user a convenient way of browsing through and examining the stored plotted data in the archive.  I also came up with a mathematical algorithm that will eliminate some of the unrelated noise in any plot that the user chooses to retrieve.  This feature may be invoked with the click of a button.


After connecting all of the hardware together, it was time to put the software to the test.  After proper calibration, I started the detection and archiving program one morning.  I had added an extra feed from the radio to a small amplifier so that I could hear what the computer was picking up whenever I was nearby.  Please note that the amplifier is an unnecessary piece of equipment, but it allowed me to spot how the graph of a lightning strike or electrical interference differed from a suspected meteoric event.  Thus later on when I was looking at archival data acquired during the hours when I wasn’t around, it was easy for me to spot events that should be rejected.


The weather was cloudy on the first day that I used the program.  I stayed next to the system for a few hours to learn the difference between the graphs of spuriously occurring interference and the graphs of times with loud radio station reception.  There was a BIG difference.  The loud reception events graphed out with much more gradual build-ups and decayed even more slowly when compared to the rejected events.  Having satisfied myself that I could now tell what was unwanted interference and what was not, I left the system to run on its own.


Proving That the System Actually Detects Meteors

The next morning I looked over the data that the system had accumulated.  During the day it had logged a significant event about every one or two hours.  However, there was a significant increase in suspected meteor activity in the couple of hours before and after sunrise.  Such an increase greatly fueled my suspicion that meteors were indeed the cause of the activity.  Sunrise was the time when my part of the Earth was facing in the direction of the Earth’s motion around the sun; thus, my side of the Earth would be going headlong into any debris that lay in our planet’s path.  That was good circumstantial evidence that I was detecting meteoric events, but hardly clinching.


The next star party of the Foothills Astronomical Society that took place near Landrum, SC was to provide the final proof.  The location was only about fifteen miles from my home.  I started the detection program that afternoon before I left for the observing area.  While we were setting up our telescopes I asked the other members of our club to let me know if they spotted any meteors during the course of the night so that I could write down the time of each such event.  Indeed, several bright meteors were noted and duly logged.


When looking at the archived data the next day, there were indeed intense radio signal peaks at the times I had noted the night before!  The people who had spotted the meteors got a big kick out of the graphs I e-mailed to them of their recorded sightings.  Inspired by my success, I thereupon christened my creation AMOS, for “Automated Meteor Observing System”.  The really great thing about this method is that it should detect daytime meteors as easily as it detects nighttime meteors.


Try It Yourself!

I don’t wish to keep the fun to myself, so I am offering the software and the details for constructing and using a complete AMOS system at no charge to visitors of the Singularity Scientific website.  All of this is available for you to download by clicking the links shown at the bottom of this webpage.   Please note that you must install the software on your computer and run it to obtain instructions on constructing and using the hardware.


            Keeping the system running in an optimum manner may require some occasional tweaking.  In many lower priced radios and practically all older radios, the tuner is typically a variable capacitor consisting of air-spaced metal plates.  Charge leak from these plates may cause “station drift”.  I have found that the 30 year old radio I have been using requires adjusting about once every twenty-four hours.  What usually happens is that a local station will start to be faintly heard at my chosen dial setting while the tuner drifts away from the frequency of my chosen “over-the-horizon” station.  The long L-shaped antenna (see previous text for details) makes the radio so sensitive that my chosen station lies in an almost nonexistent gap on the tuning dial between the offending local station and another local station.  All I have to do is turn the dial an almost imperceptible amount until I no longer hear the offending station.  Radios with digital tuners should not exhibit station drift and thus will not require this extra adjustment.


            Having given you that final caveat, I hope that you enjoy experimenting with AMOS as much as I have!

Free Downloads: 1) Calibrator, 2) Meteor Detector, 3) Data Displayer

Like this project?  Want to keep more such original projects coming?   This website and the rest of  the other services that I do for the public cost money.   Please donate whatever you can spare.