Meteor detection using Radar

By RGP

Wednesday, May 02, 2012

The Fun Cube Dongle, or FCD


FCD for Meteor Radar

The FCD is fairly easy to use for Radar meteor detection. It is small, sensitive and, using a host of free software, easy to use. With only the addition of an easily made antenna I managed to detect meteors using a professional satellite Radar in France called GRAVES.

In this short article I will detail all the steps I took to get FCD up and running and hopefully you will not then make the same mistakes I did and get there much quicker.

The first thing I would say is there are a lot of other good articles on the subject (see below for links to the ones I found). These are all very good and discuss a variety of issues you will encounter. I think there is one point that I didn’t appreciate until I finally started to see meteors is that the traces they leave are generally very small and quite difficult to see, until things all come together just right.

The steps I took in setting up my Radar detection system are as follows:

Attach the FCD

Attach the FCD to the PC using an active USB extension cable, in my case I have a 15M cable. This gets the FCD well away from the PC which definitely reduces the noise in the system for me. If I have the FCD near the PC I get bands on the display from noise coming out of the PC

Install the FCD software

1) Install the application for controlling the dongle itself FCHID - found here: http://www.funcubedongle.com/?page_id=313
2)  Install SpectraVue or SDRRadio (SDRConsole) or similar to hear and see radio - setting up of this software will follow

Firmware upgrade

Update to the latest FCD firmware - this is essential as the firmware that ships in the FCD is not designed for normal use. It is purely for use during its manufacture, although I think the latest FCDs are now coming with the latest firmware preloaded. I suggest you just check.

The antenna

The type and construction of the antenna is a very personal choice. The good thing about the GRAVES Radar is that it is just about in the 2M radio ham band, which means there are lots of designs to go for. The one I opted for seemed to have good performance and was relatively easy to build. It can also be purchased quite reasonably as both a kit and ready built. It took me a day to make and about £10 for the material along with some of it from the scrap bin. The kit is £45 and if you want my advice on this I would say just buy the kit! It is available from here: http://www.powabeamantennas.co.uk/cqm5uc/index.html as are the instructions to make your own. Your choice.

I attached the antenna to the FCD using a fairly short 2M length of RG59 (from Maplin) as it isn’t super low loss cable. The good news is that the FCD sits on the mast in a waterproof box with the 15M USB extension running into my PC. This avoids having to buy expensive super low loss RF cable. At this length the RG58 should only lose less than 1dB of the signal.

The mast I put it on for testing is an inexpensive folding light/flash stand avaialable from places such as CPC here: http://cpc.farnell.com/ultimax/umtrl24/tripod-2-4m-thumbscrew-20kg/dp/ST01376 It gets the antenna about 2.5m off the ground and is quick and easy to erect. If it is windless it will free stand but guys would be necessary if there was a breeze.

Calibrating the FCD

I initially didn’t bother with this step and just got frustrated. Don’t make the same mistake. You need to make sure that the frequency you set the FCD to is the frequency you are listening to. As the meteor traces appear in a very specific and small frequency range if you omit doing this calibration you will probably not see the little blighters!

To calibrate the FCD tuning you must find a good local transmitter that has (or is likely to have) a definite and accurate frequency. In my case this is the local airport which has published frequencies for the tower etc. I chose the tower as I am sure this will be pretty accurate and I would imagine it is fairly frequently checked and maintained on the published frequency for obvious reasons.

It is also better to pick an AM transmitter (such as airband). At first I tried FM broadcast stations but I found they were too broad and wandering to see if your FCD is set accurately.  While listening to the station adjust the frequency calibration in FCHID (set in PPM) until the station is right on the advertised frequency. Mine ended up been -23PPM  out (confusingly the actual value in the dialog of FCHID is 999977 but this can vary between applications so watch out!). After which I then tuned to a number of other known stations and this helped verify that the FCD was now pretty much spot on. Interestingly the aircraft themselves often seem a way off, however this may be due to them moving (doppler shift?) or just poor maintenance! All the tuning work I performed in SDR-Radio (http://www.sdr-radio.com/) as this has a nice, easy to use interface with excellent de-modulation of the AM signal. Other people have had a lot of luck with SpectraVue and SDR# (to name but a few). Although the main detection work is done in Spectrum Lab it does not have a great (nor easy to use) demodulator.

Using Spectrum Lab with the FCD

Spectrum Lab is an amazing piece of software. I would go as far as to call it an audio/waveform analysis operating system. The downside of this is that it is complex and takes quite a lot of learning (that’s if you ever do learn this product in its entirety?). The good news is that you can save all the settings into a configuration file called a CNF file. I have saved my working file here so you can download it and you will then hopefully have something that is very close to working for you. What you will need to change for your own system is the following.

Make sure you have your PPM values correct in FCHID and set the FCD frequency to 149049kHz. (The GRAVES tx is on 149.050mHz, however you need the 1kHz offset to enable you to see/hear the pings as a side-band signal). Now change the view of your waterfall in Spectrum Lab to look at frequencies around your tuning point of roughly 1kHz below and 2kHz above. Ideally your meteor hits should be occurring around 1kHz above the centre frequency (the one you tuned on the FCD - 149049) but there are always some slight mismatches and this may be a little different.

Be patient and look carefully. You are looking for a small, dense blob. It may have a slightly descending frequency; a line running into it but this will probably be in the order of about 200-300Hz tops... something like this.

<Piccie of M-hit>

If you are lucky enough to see them then zoom in on that frequency using the mouse to draw over the feature and then right-mouse clicking and choosing zoom in on selected area from the menu. Be patient and wait until you have seen some more...then hopefully you will be on your way.

If you don't see anything then zoom out a little bit and look again. Remember these chaps are only small (about a second or so long and over about 200-300Hz) unless you are in a storm and/or there are fireballs. Mostly the sporadics, and the general storm meteors, are just little chaps and are tricky to see.

Once you have spotted them appearing then you will wonder how you ever missed them and then you can be sure you have everything working in harmony.

Things I want to do to improve my detections

Increase gain with an LNA
Reject blocking from adjacent stations using a bandpass filter
Elevating the antenna some more
Getting away from the RF noise of the city

meteoblue.com

 

Click to join yorkastronomy

Click to join yorkastronomy