Using a RTL-SDR Dongle to Measure FM Deviation
I’ve very successfully used the Bessel Null / Modulation Index method using a rather crude test setup.
First I made a audio patch cable with a PTT switch to go from my laptop to my handheld.
Then I plug in my RTL-SDR dongle and open GQRX spectrum analyser and tune it to the repeater input frequency and zoomed in so I could only see about 8 – 10 KHz of bandwidth at the RX frequecy.
Then we need to figure out the audio tone frequency that would give us 4.5 KHz deviation using the First Bessel Null. There’s a few good pages on the web that get into the math and theory behind these calculations.
Here’s how you determine what modulation frequency you need.
Deviation / 2.4 = the Modulation frequency
2.4 is the mid index of the first Bessel null.
So 4.5 KHz deviation / 2.4 = 1.875 KHz modulation Frequency. (This is the number we want)
At this point I use Audacity to make a 30 sound file with a 1875 Hz sine wave tone.
I open the sound file in a music player on my laptop with the volume set at 0% and set it to loop the file continuously.
Now with the gqrx window open, I turn on the ptt switch on the patch cord and confirmed the signal was being received by gqrx. It should be a narrow spike at this time. (Probably doesn’t need to be said but make sure the antenna on the handheld is properly terminated into a dummy load or an antenna.)
I slow increase the volume until the centre peak of the received signal dipped to it’s lowest point compared with the first side bands.
(Spend a little time going back and forth until you get precisely the lowest point.) Now the handheld is calibrated at the correct deviation.
Next step is the repeater.
On the controller, while still TXing with the hand held set the rx level as per the controller manual which usually is 1v P to P which is .35v rms.
Now switch the frequency in GQRX to the repeater output and adjust the TX audio pot until I get the lowest point of the null just like when I calibrated the hand held.
I’ve since had a chance to verify the deviation with a service monitor and with this method I was able to achieve 4.6 KHz deviation which is better than most of the repeater users radios.
Instead of gqrx any other SDR spectrum analysers could be used. In Windows I think SDR# is one of the better ones.
When I get a chance I’ll take some photos and screen shoots to up date the post with.
Leave a Reply