Just three more channels to scan and I’ll have them all done. So far, the vast majority of activity encountered is on very few channels.
I have a nifty little app called iFlare, and it reminds me of Iridium flares. Tonight I got the opportunity to see the -4 magnitude flare from Iridium 47!
With a waxing moon high in the south, it was a little hard to get good dark adaption. However, I was barely able to see +4 stars in the area (like the parallelogram of Lyra). The location for the maximum was a few degrees SE of Vega, and about 10 seconds before the maximum I caught the satellite N of Vega and headed toward the SE.
After the maximum, it was possible to follow the satellite for about 20 degrees, fading out before it got as far as Rasalhague. That 20 degrees was about 30 seconds or so, I think.
Row 1: channel frequency
Row 2: percentage of total messages received over the period (2111.28 minutes)
Row 3: absolute message count
Row 4: messages per minute based upon row 3 and the total period
Let the receiver run for the past few days on just the two above channels. 130.025 MHz is still a big winner in these here parts. Out of the 8 channels I’ve scanned over the past week or so, so far it’s 130.025, 131.550, then 129.125 MHz.
I’m trying to get a handle on the amount of traffic that there is on the various many ACARS channels that seem to be in use. In a previous post, I counted 20 channels and that information was from a few different sources.
So i pointed the output of acarsdec to a text file, and let it capture almost 82k messages over the 7.6 day period from 4 Jun to today. Due to the maximum instantaneous bandwidth limitations of acarsdec and the RTL-SDR dongle, I decided to focus on the following
As can be seen, the vast majority of ACARS messages on the 7 channels scanned over that 7.6 day period are on only 2 channels, 130.025 and 131.550. Row 1 is the channel frequency, row 2 the percentage of messages received on that channel, row 3 the absolute message count for that channel, and row 4 the average messages per minute on that channel.
Now I’ll change the channels and do another run to see if there is more activity elsewhere.
This Epson notebook, an MFJ TNC and a Yaesu FT209 handheld 2 meter radio set on 145.01 MHz allowed me to have data communications while motoring around the southwest. Back then, hams had assembled a tremendous network of mountaintop digital packet repeaters (digipeaters) that provided amazingly good coverage of California, Utah, Arizona, New Mexico and into Texas.
I don’t have a picture of it, but I’d built a floor-mounted stand for the Epson in my Rabbit diesel pickup, and a terminal program ran on the Epson talking to the TNC. I could send and receive short messages. The Epson had the tape-drive memory, which made it easy to write my log and store it on the tape.
The MFJ was pretty cool at the time, and could even do HF packet, though that was especially painful. The Rabbit had a FT757 HF transceiver as well, the antenna (not shown in the above image) was on a ball mount on the left rear side, about midway between the wheel well and the taillight.
It was my first foray into the world of digital, and especially digital mobile communications. From that point on, I always had a computer of some type in my trucks.
The stackup that I did inside the enclosure was fairly crude. The GPS receiver and its attached antenna are at the “top” of the stack, and stuck to the inside uppper surface of the enclosure using double-sticky foam tape. Immediately below that, the dual-band Wi-Fi board is component-side down. I’d removed both the SMA-RP female and the USB male board connectors to reduce height and length. The bottom of the stack is the USB hub, again with USB female sockets removed and component side down as there’s two electrolytic caps that stick up 8 mm or so.
I’ve used double-sticky foam tape for things like this over the years, and as long as it doesn’t absorb moisture, it’s quite RF-transparent and the antenna has no issues.
The Wi-Fi board has a ground-plane on top and bottom, but I wanted to reduce to a minimum any local fields from the components, so the board goes in component-side down to isolate it (maybe a bit) from the GPS antenna.
The GPS receiver itself is completely encased in a shield, and is slightly larger than the antenna, so there may be some added attenuation of spurious emissions from the Wi-Fi getting into the GPS.
The USB hub, at the very “bottom”, also has a top/bottom ground plane.
The Wi-Fi sniffing performance is as good as my permanent mobile setup, and the GPS gets excellent PDOP (<2.0) when there’s a decent field of view of the sky.
The apps used to grab the GPS data is 4river’s NMEA Monitor, a fine little program.
Before I left on a trip this past week, I was able to shoehorn all the components (GPS receiver w/integrated antenna, dual-band Wi-Fi module with external antenna input, 4-port USB hub, TTL to serial to USB adapters) into a single plastic case. It’s not yet weatherproof, but at least it’s splash- and rain-resistant.
The original cable on the USB hub was only a meter, so I grabbed a 2 m cable from the box, whacked off the end, and replaced the shorter cable.
I used double-sticky foam squares to create an electronics sandwich, with the GPS antenna at the “top” of the stack, the Wi-Fi below, and the USB hub at the bottom. The cable passes through a silicone-sealed hole in the case, and I removed the SMA-RP female from the Wi-Fi dongle and replaced it with a short RG178 cable and bulkhead SMA-RP connector that pokes through the top of the case.
As one can see, the box is a cheap one from Radio Shack or similar, it’s some kind of ABS. The box’s lid, which would usually be on the top, is now the bottom of the assembly. I silicone-glued 3 NdFeB rectangular magnets to the inside of the lid, and put 4 rubber-bumper feet to reduce any potential surface marring. Next, I took the thing out for a drive on a local freeway to see if it’d blow off. It did. I adjusted things a bit by getting rid of the rubber bumper feet, and replacing them with electrical tape on the outside of the lid for more of an anti-skid, compliant surface than the bare plastic alone. The next drive, and subsequent ones this week, proved that the widget is now fairly stable even at “high” highway speeds. I thought about painting it white to reduce heat absorption, but that would make it have higher visibility and I’d prefer to stay low profile.
In taking it for a drive or two around the neighborhood, it matches my mobile setup almost exactly in reception, and takes all of 30 seconds to deploy when getting a rental car.
I’m wondering if I can somehow add a temperature sensor inside and read it via the USB. But that’s not so important.
A few posts ago, I mentioned I’d set up acarsdec on a Raspberry Pi and a USB RTL-SDR dongle. Written by Thierry Leconte F4DWV, it’s a very nice lightweight ACARS decoder that puts a relatively small load on the RPi. A detailed writeup on installing and using acarsdec states that it can handle up to 4 receive channels and with a maximum frequency spread of 1 MHz.
I found that handling 5 channels doesn’t push the CPU load too high. Tonight I found that going beyond the 1 MHz separation barrier, at least a little bit, doesn’t cause any obvious issues either. I’ve even pushed it to 8 channels, that seems to be the absolute limit, while still getting under 80% CPU usage on my RPi 2 model B. However, the error rate becomes extremely high with many messages being lost.
There are a lot of frequencies assigned for ACARS use. While there are many sites that seem to show a number of frequencies, I’ve found that a combination of acarsd.org and radioreference.com make for what seems to be the most comprehensive list.
While this is a big list, it seems that what traffic there is is scattered over just a few channels. Here in Phoenix, I hear the vast majority of all ACARS messages on either 130.025 or 136.850 MHz. There’s some occasional stuff on 131.550, which is supposed to be the primary worldwide frequency, but it pales in comparison to the aforementioned pair. Some of the above channels are claimed to be airline-company specific, but to date I haven’t observed any evidence of that, even with over 2000 flights a day passing through my reception range.
Those two channels are way too far apart for acarsdec to decode them both in one instance. However, with two dongles I can listen to several frequencies in the low part of the band and also in the high part. I haven’t tried that yet, and I suspect it won’t work due to the CPU load with the current RPi.
There are other things that I can do first, which includes resurrecting my homebrew VHF turnstile antenna and putting an LNA/airband filter combo up on the roof right behind the antenna. That alone should help improve my reception since I’ve got about 20 m of coaxial cable between the antenna and the RTL-SDR dongle. As well, the dongle doesn’t have the greatest front end or sensitivity, so an LNA/filter can help with that as well as negating the cable loss.