About Breezedude - Breezedude/breezedudeWindSensor GitHub Wiki

⁉️ Why all this?

A wind measurement station on the local mountain is a great thing for paragliders. It's just a shame that there isn't one there yet. Off-the-shelf measuring stations (Holfuy, etc.) are expensive to purchase and sometimes also involve ongoing costs for data transmission. DIY solutions such as GXAirCom, which use OGN/FANET for data transmission, provide a remedy. However, the technical implementation is up to the individual and therefore requires training and know-how. This inspired me to design a very simple but sufficiently good wind station.

📡 About FANET and OGN

The measuring station processes the sensor data and transmits it at fixed intervals (40 seconds) “into the air” - together with the station's GPS coordinates. Every receiver within range (FANET-enabled variometers, ground stations,...) receives the data and can display it. All ground stations that have received the data forward it to the OpenGliderNetwork via the Internet. From there, other services (e.g., Gliderradar or the Breezedude wind map) can access the data. The nice thing about this is that I don't necessarily have to operate my own FANET ground station with an internet connection nearby to operate a wind station. The OGN is already quite well developed and the range of simple ground stations is up to 20km and more with line of sight. The coverage can be displayed on Gliderradar by clicking on “Coverage”. Only in the event that there is no radio reception is it necessary to set up your own ground station. This is possible from around €30, see below.

🤨 OGN? FLARM, FANET? What?

A try to clearify what OGN and FANET actually are. OGN is a (voluntary) community project that was originally intended to receive FLARM and OGN tracker signals from gliders and upload them to the Internet. This enables live tracking and supports SAR searches. Since SDRs are used, many radio protocols and modulations that are on the same frequency band as FLARM (868 MHz) can be received and processed. For this reason, the software of OGN ground stations has been expanded so that these stations also support the open FANET protocol, among others.

Technically, all protocols can be operated with a LoRa module, which has given rise to projects such as GxAirCom and SoftRF. FLARM is a commercial product that is only intended for air-to-air communication. It is based on frequency shift keying (FSK) modulation and relies on pseudo encryption, which, according to the manufacturer, primarily provides legal protection against unauthorized decryption.

FANET, on the other hand, is license-free and supports not only the positioning of aircraft but also other data packets such as weather station data. FANET also operates on the 868 MHz band but uses LoRa's chirp spread spectrum (CSS) modulation technology.

The term “SkyNet” is also frequently used. This is nothing more than FANET, except that the received data is not publicly submitted to OGN, but only ends up on Burnair's server.

🛠 Technical FAQ

Why not using 4G?

4G, especially NB-IoT or LTE-M are nice for connected IoT devices.
But there are some downsides:

  • subscription cost
  • Complex setup
  • Fixed coverage
  • Hardware availability
  • huge coding effort to get it low power

Why not using a integrated module like RAK3172?

Some of them are quite nice but none did match perfectly. For configuration some user friendly interface without the need for a "programming adapter" is contemporary, likely USB or BLE. But also cost need to fit.

No USB, No BLE:

  • RAK3172 (STM32WLE5CC)
  • Wio-E5-LE (STM32WLE5JC)

📕 Breezedude history

In summer 2023 I have seen a bunch of GxAirCom weather stations when flying in Sillian. I checkt out the hardware and found an ESP32 not the right choice for a low-power device that does not need heavy computing power. As the Davis sensor used it 10x more expensive than the rest of the hardware and also known for breaking, I started looking for alternatives. I found the WS80 as a spare part for <100€. My plan was to grab the data entering the RF module to get the wind data. I gave it a try but soon found a much more convenient way: A Debug mode outputting human readable sensor data.

I started gathering requirements and build a prototype with an Adafruit Itsybitsy M0 Express, removing all parts not required.

December 2023 - First prototype setup

First breezedude at landing Hochries

March 2025 - After one year of on site testing I did a hardware update. The heater is useless, it would take a lot more power to keep the sensor working. With that removed the power requirement is a lot lower, so no need for a switching solar charger. A smaller battery is sufficient as well and with that a smaller case with a smaller solar panel is possible. Breezedude V2 was born.

April 2025 - As more people asked for the PCB (I was still hand soldering all the prototypes) I optimized the design to be manufactured in a economic way.