Satellites... without the dish

I haven't posted in here in a long while, but I've had a really interesting last 2-3 years, and I wanted to try and get back to writing about things. So, I'm going to try and post some things over the next few weeks to fill in the gaps. Basically, for most of this time I've been looking to the skies doing satellite reception of all different types of signals. Mostly this is weather satellites for their imagery, but with the recent demise of the NOAA POES satellites and the loss of 3 great targets for downlink I decided I needed a new challenge.

As mentioned previously on this blog I have an interest in aviation data, which is why I currently downlink the ground-to-air side of ACARS from Immarsat satellites on L-band @ 1545 MHz. I'm pointed at the European Alphasat satellite at 25 deg East. Some people do this kind of thing with a dish, and so do I, but some people do this without a dish. This can be either with a patch antenna, or with a longer helix than normal - the longer the helix, the more gain you get. DerekSGC, who maintains an archive of useful helixes, has just the thing - an 8 turn helix that can be paired with a ground plane to make something that should have useful gain for receiving the geostationary Inmarsat signals with sufficient SNR. I'd been wanting to target a second satellite for ACARS, but I don't have (or want to mount) a second dish. So this dishless method, plus POES going, was the driver to give it a go.

And in case anyone is wondering, the cookies were indeed buttery and lovely.

I've made 4-5 of these things now, but I must say, I never get any better at it. Sometimes the wire is fed through, but despite everything being pre-measured the frequency is off. Sometimes that's because you need to push the wire through more so that it sits at the side of the holes towards the outside of the helix, rather than the inside, if that makes sense. Like, you don't pull the wire tight, you want it kinda loose and sitting on the outside of each hole. Push wire through until you can't push anymore through without the spiral deforming. And of course, there is matching to consider, but I just move the turns around a bit (pushing the first quarter/half turn down towards the ground plane works well) until the SWR gets as low as possible. Usually you can get around 2, 2.5:1 straight off the bat, but some matching strip work will start to bring it down under 2.0. It isn't too important unless you want it done right.

The test setup was an SDRPlay RSP1a, 1 m of RG58, a Nooelec Wideband LaNA, and the helix. I used SDR++ and Jaero, via the standard VB-Cable for passing audio. Only using 1 m of coax is quite an optimised setup, even with poor RG58, and it's worth remembering that my current system is a 60 cm oval dish, an RTL-SDR stick, and a great L-band Sawbird... but about 10 m of (admittedly lower loss) RG59. So I really was giving it the best possible chance of working. I also did this outside, for best line of sight, and hence best results.

And quite honestly I wasn't ready for how successful it was. 21 dB on a particular signal that I receive at 30 dB on the dish. And I only had a wideband LNA on it. With a proper Sawbird I'd probably gain a few more dB. Wow. Not only that but I managed to pull in the second satellite with ease as well. All 3 data rates of ACARS decoded fine. Voice was a BIT sketchy, but I think I could have made it work with slightly better LOS, filtering, and gain.

An interesting thing I did notice while testing indoors is that my window frame, which presumably has metal in it, is surprisingly good at coupling to an antenna and improving it... This window has also somehow improved HF reception in the past through coupling, and to see it work here was weird. It's pointing nowhere near the red dot, where the satellite is. Insane. The glass did stop a lot of the signal (first waterfall image), but just pointing at the frame sorted it all out (second waterfall image) and I had decodeable signals on the low and medium data rates, but not enoigh for the high data rate. Of course, having the window open was much better, almost as good as being outside and far from sources of noise.

While cool, this only being one side of the conversation is very limiting. The holy grail of satcom ACARS reception is, of course, the air-to-ground side of the link. Sadly there are various challenges to setting up a downlink for this side of ACARS. As documented by The Bald Geek in his excellent guides the kit for this is quite expensive, and it can require a large dish to make sure you pull in more of the signals (they range in signal power).

However, he also mentioned on his blog that it may be possible to obtain the 1.6 GHz uplink from the plane to the satellite - intercepting it on the way up, so to speak - and possibly decode some messages that way. I decided the best test of my new dishless helix was to attempt this (mostly because my actual dish setup has a Nooelec Sawbird IO filter on the end that didn't cover the uplink frequency range).

And so that's how I found myself in my back garden, aiming what looks like a raygun at planes. I had no idea what specific frequencies to look at, and it was very hard to look on a waterfall and accurately tune in to these fleeting signals. So, I decided to record a chunk of spectrum at various times between the expected range of 1646 and 1647 MHz and played them back with SDR++.

It took a while, but then I ended up seeing some bursts. But they were fleeting and rare, as mentioned. However, I saw something! A 1200 bps burst signal. It appeared to be a logon message. The hex of the from field on the message matched a plane that was... 23 miles away. What!? That's way further than expected...

That's really boring though. What we REALLY want are the actual ACARS messages from the cockpit. I thought the best chance for that was the 10500 bps messages. I wasn't sure if I'd have enough signal coming in for that, but a few minutes later, I saw a candidate... I played back the signal and Jaero gave me a full decode.

This decodes to what looks like a standard status report, including a position and routing/time information:

10:56:13 25-08-25 UTC AES:A9D286 GES:90 2 .N732AN ! H1 9 Boeing 777 323ER American Airlines Flight AA239
F51AAA0239#M2BPOSN55252W003279,TLA-24,105350,320,TOB-25,110944,ERAKA,M38,24217,1667A7E6

Very nice! I didn't see a full ADS-C position message, but I am sure if I kept it up I would. So, is this a viable thing to do? Maaaaaybe. You won't get any positions you don't already have on ADS-B. But arguably that isn't the point - as The Bald Geek frequently says, the main value in this stuff is the ACARS messages - finding out the why and not the where. Therefore, one of these pointed at an airport may well yield things like clearance requests, chatter about delays, aircraft issues etc. And this stuff may go out on satcom more than VHF - it varies a lot depending on company and how much they want to pay for services. But I AM considering this. I have a place I can aim something vaguely at Edinburgh Airport... Mind you, in the future I also want to do a full ADS-C feed on C-band anyway, so, we will have to see how that goes.

Anyway, dishless helixes. Very, very nice and great if you are challenged for space. I recommend trying them.