In December 2024, the SF-HAB group flew a high-altitude balloon from the San Francisco Bay Area. The flight was a partial success, with the payload recovered the next day. Only several weeks later did I find out about this excellent Sondehub Grafana instance, hosted on AWS and directly connected to the Sondehub Amateur database.
In this post, I want to dive a bit deeper into the Horus Binary v2 telemetry that was transmitted from a reprogrammed RS41 radiosonde on this flight. This entire post is commentary and screenshots from this grafana dashboard, screenshotted here to reduce link rot. All times are in UTC on Dec 15th 2024.
Balloon Position Graphs
Due to the under-inflated balloon, we knew it was going to rise slower than predicted, and burst at a higher altitude. This balloon reached a maximum of 37,412 meters (~122.7k feet) at 21:09:57.
This is a graph of the ascent and descent rates for the whole flight. The blue line is what the GPS reports (in meters/sec, left axis), and the yellow line is calculated from sequential GPS altitude points.
The overall ascent/descent rate graph is kinda hard to read because of the large Y-axis scale. Excluding the time after balloon burst shows that the ascent rate was around 3.5 to 4 meters/sec. The overall ascent rate (from 744 to 37412 meters in 8590 seconds) was 4.26 m/s, which isn't that far off from the usual 5 m/s. One can see that towards the end of the flight the balloon rose faster than the beginning of the flight.
If you look at only the balloon descent rate, it free falls pretty quickly just after balloon burst, hitting -109 meters/sec (~244 MPH) just after burst when there isn't any air for drag.
The reported horizontal balloon reaches a maximum of 203 kph (~125 MPH) at max altitude just before burst. I didn't realize that the atmosphere was this active at such a high altitude. This data is directly from the GPS receiver.
This map shows the flight path of the balloon, as well as where the receiving stations are located.
RF Graphs
The first RF graph is signal-to-noise ratio (SNR). This is a measure of how strong the signal is at the receiver. Measured in dB, higher is better.
On this graph, the colors are a bit hard to differentiate. The left blue line is my car station using Horus-GUI (KF6ZEO-H1), and it was only active for the first few minutes of the flight. Then the red line is the receiver at my home using a Diamond X-50NA omnidirectional antenna. Towards the end of the flight, amateurs down in San Diego received the balloon, shown in light blue, green, and red. The last few packets in orange were received by the headless horusdemodlib station in my car.
This graph shows GPS satellites in use. More is better, but anything above 4 will give you an accurate location. Interestingly this shows that GPS kept lock on 13 satellites even after the balloon burst at 21:09:57.
Receiver reported payload transmit frequency is the reported frequency from the RTL-SDR dongles. This should be fairly consistent throughout the flight, as the TCXO on the dongles shouldn't change all that much over the course of the flight. I'm not sure what happened around 1945 UTC on KF6ZEO-H2, it seems like something reset the dongle.
Also, the sudden increase in frequency on the orange KF6ZEO-H3 line is due to doppler shift of payload starts falling at 2109 UTC.
The Per-Receiver Packet Count pie chart is pretty self explanatory, most of the packets were received by the station at my home.
Environmental Graphs
The environmental graphs for this flight are a bit boring, as there is only one temperature sensor on the RS41. I'm not quite sure how to interpret the temperature falling after balloon burst. Maybe the colder air rushing by the PCB is removing heat? But it's in an insulated case.
Battery voltage closely tracks with temperature.
