Receiving and Reporting Telemetry from Fox Satellites

Updated 3/21/19
We've now installed the EZ-Lindenblad antenna (the QST version with modifications - the AMSAT alternative wasn't available at build time) and found that it works quite well. Thanks to G0KLA's recommendation we've set the Auto-Start function on, which will use the Doppler correction to follow the currently-visible satellite rather than searching the entire passband for the strongest signal. This has significantly increased our packet capture rate.  We're now planning to install the 1.07 version of FIAB.

Updated 12/8/18
AMSAT has launched and maintained three "Fox"-type satellites that are easily available for ham use. They're in low-earth orbit and use standard 2 meter FM transmissions, and can occasionally be worked with an HT. A fourth satellite is scheduled to be launched on December 2 2018.

In addition to audio QSOs these satellites transmit telemetry information about various parameters related to the health of the satellites that are monitored by ground stations. This data is transmitted using a low audio frequency underneath the voice signals during a normal QSO using a "data under voice" (DUV) technology - so you can't hear it but it can be decoded. One of the telemetry screens for AO-85 looks like this:



Since the satellites orbit over all parts of the earth, it takes a network of stations to continually monitor the telemetry to give AMSAT a continuous stream of data. AMSAT distributes a free Fox telemetry program for Windows, Mac and Linux for users who want to download and view this telemetry. That program can optionally upload the telemetry to the AMSAT data warehouse, allowing AMSAT to record satellite performance as it flies around the earth.

Fox in a Box

To facilitate monitoring and reporting this data, AMSAT has developed a program for the Raspberry Pi computer connected to an SDR radio that provides a self-contained station for receiving and uploading the telemetry data. Called "Fox in a Box" (FIAB) it's pretty easy to implement and can be left alone to function (although we never really leave anything alone in our ham projects...). As far as I can tell the only source of this software is the AMSAT store where you can buy an SD card preprogrammed with the software. This card already has VNC installed and enabled, which is how we connect to the Pi that's running it.  The card also comes with instructions on installing the software.

Our Project

K2ZA and I decided to implement this at the GCARC clubhouse to allow us to contribute to the reporting network. From AMSAT I got the preprogrammed SD card containing the code, stuck it into a Pi and added John's Funcube SDR radio. Then I connected it to the 8x8 2 meter yagi antennas at the clubhouse through a tee-connector (we have about 4 radios connected to those antennas through tee connectors), logged into the Pi with a VNC connection and waited for a Fox satellite pass. Fortunately the Fox satellites were orbiting over NJ on a Saturday morning, which meant that John and I were both available to check the system out.

We all know that this rarely happens - but it worked the first time! As the satellite rose above the horizon the "Input" telemetry page started showing the "eye" pattern that indicates a decodable signal. The "Frames" counter at the bottom of the screen started counting up from zero and the values in the telemetry screen shown above moved from all zeros to their current values. Because of the excellent antenna location at the clubhouse we were able to collect telemetry from horizon to horizon throughout the pass. We had the same experience about an hour later with the next Fox pass. We modified a few parameters later, but it basically worked out of the box. Because there's a VPN connection to the clubhouse available I can monitor this from home by connecting to the VPN and running VNC to connect to the Pi.


How the software works

The software is really cool. First, it tracks the satellite itself without any external components. It automatically downloads the Keplerian elements from the AMSAT website and does the orbital calculations internally. Note in the upper right screen that AO-91 has a positive elevation value, which means that it's above the horizon. Note in the horizontal window at the bottom of the screen that the frequency band for all of the Fox satellites is shown, with the freq for each satellite delineated with purple lines and the strength of each signal within that band shown on the red line. This is a cool feature of using an SDR radio like the FunCube - it can scan the entire bandwidth for telemetry signals and lock onto the one that it finds. (I don't know if it only scans for satellites that are above the horizon - perhaps a knowledgeable reader of this blog can clarify that point.) So you can see the vertical purple bar over the peak in the red line in the AO-91 section of the band, which is the telemetry signal being received.

Reporting leadership

As of 12/1 we've downloaded 1100 telemetry frames that have been uploaded to the AMSAT data warehouse. We're far from being a leader in telemetry reporting, since the leader has more than 600k frames, but we're in the top 100 for the most recent 7 days as of 12/1/18. That's pretty cool too.

Next Steps

There are two upcoming steps for this project.  First, the Fox1-Cliff satellite is due to be launched tomorrow. We want to be ready to copy the telemetry for this satellite, but we can't set it up yet because the Keplerian orbital elements won't be known until the actual launch; consequently we won't know the time of crossing or antenna aiming until they're published by NASA. Hopefully we'll get them quickly and can be among the first to report them when the satellite passes over us. Telemetry information is always useful to AMSAT, but the first post-launch data is especially critical.

Next, we have to get this system working with an omnidirectional antenna. Unlike the SatNOGS Pi program that can control the rotators through a Hamlib TCP link, FIAB doesn't have any way of controlling the rotators. This means that we need to use an external program to point them at the Fox satellites during a pass. Currently we're programming SatNOGS to point at each Fox satellite pass, but that's a daily manual process that can't really continue (because I don't want to have to do it indefinitely). Until we have an automated way to do that (which we don't foresee in the near future) we need to get an omnidirectional antenna built that doesn't need to be aimed at the satellite. K2ZA has plans for a Lindenblad antenna and the May 2018 QST has an article describing a quadrafiliar antenna, either of which would work well. So we'll select one (probably the Lindenblad), order parts for it and do an antenna-building session at the clubhouse to get that assembled and installed. If anyone else is interested in building one of these antennas let me know, since if there's interest we could do a group antenna-build as a Tech Saturday project.

73 de Jon WB2MNF

Comments

  1. Hi Jon, Nice blog post. I realize it was a couple of months ago. Did you manage to get the antenna tracking in place? How about another PI running Gpredict?

    I can answer your question in the blog about about which spacecraft get tracked. In the mode you are running it is checking each sat band continually to look for a signal that is above the thresholds shown at the bottom of the window. If it exceeds those thresholds then it locks on and decodes it. It does not matter if your keps are old or inaccurate in that case.

    If you tell FoxTelem to start the decoder when spacecraft are up, then it will pause and start decoding a specific sat when it rises. Then the keps need to be more accurate.

    73
    Chris
    G0KLA

    ReplyDelete
    Replies
    1. Hi, Chris - thanks for the info on FIAB tracking. I'm assuming that the FIAB program will update the keps automatically which should keep them current. I switched the FIAB to tracking mode and it looks like we're copying more telemetry packets, which makes sense since it's not looking for packets on freqs where there are no satellites. We now have the FIAB connected to an EZ-Lindenblad (using the QST design which we built before seeing the AMSAT design) and it appears to work well. We can't use the yagis for FIAB until/unless someone adds Hamlib support that would let FIAB control the PST Rotator program like SatNOGS does (we do use the yagis for SatNOGS). But the Lindenblad seems to work pretty well, so we'll probably leave it that way and go on to something else.

      73 de Jon WB2MNF

      Delete
  2. Hello ,
    I've been tracking amsat FOX satellites using IC-9700 along with M2-yagi antenna/yaesu rotator , I input the audio into FOX-telem software at 48000Hz via serial port but have been unsuccessful at decoding any telemetry data.I only hear automated voice from AO-* satellites at a very good quality.

    I've decoded 5 - 7 frames of HuskySat-1 in fox-telem with BPSK mode but for some reason AO-* sats never show any telemetry when put in DUV mode or any other mode .I've tried setting USB IF-output from 10% to 100% and also AF - output but had no luck .

    Can you suggest me any new inputs or settings on ICOM-9700 or fox-telem which worked for you and i might be missing .

    ReplyDelete

Post a Comment

Popular posts from this blog

How We Built the GOES Reception Project

GOES Satellite Reception Project

Building the Raspberry Pi NOAA Satellite Station