A LoRa Board Tracks Aircraft After Its Maker Bypasses the Radio Filter

A tiny filter was keeping aircraft out. The W12’s stock 902-928 MHz filter attenuates the 1090 MHz signals this project wants, so the longer-range build removes that component and bridges specific pads.

The project README is explicit: “firmware settings alone cannot bypass that filter.” Sometimes the setting is a soldering iron, which makes the undo button rather inconvenient.

The LR2021 captures the radio traffic while the ESP32-S3 validates and decodes aircraft messages and serves a map. Reception and the bundled US map work without a cloud aircraft feed. The reported 150-nautical-mile result came from the modified LF path with an indoor 915 MHz antenna; another placement may produce a very different map.

The README also describes an unmodified HF-input route with substantially reduced range in its tests. That alternative matters before anyone orders rework equipment: the project offers a way to explore the idea with a different compromise.

The board modification and the radio’s published limits describe different boundaries. Bypassing a filter changes which signal can reach the input. It does not rewrite the manufacturer’s receive-band specification or promise the same result on another board. A successful experiment can be interesting without becoming a replacement specification sheet.

That matters when judging the map. Seeing aircraft appear is a useful first result, but the display contains decisions about how to present a stream of intermittent messages. A clean-looking moving marker should not substitute for looking at reception diagnostics. The project’s separation of CRC-valid counts gives a builder something more concrete to compare than the prettiness of the animation.

The most useful home experiment would change one thing at a time. Keeping the receiver location fixed while comparing the documented paths would avoid crediting the filter for an improved view through a different window. Keeping the antenna fixed would answer another part of the same question. Moving everything and getting more planes makes an exciting screenshot with a rather ambiguous explanation.

The alternative input is also worth considering before anyone reaches for the soldering equipment. A reduced-range result may still meet the goal of watching nearby traffic, especially for a builder interested in the software rather than in maximizing reception distance. The decision follows the experiment being attempted and the willingness to modify that particular board.

For a project notebook, board revision, selected RF input, antenna and placement belong beside the message counts. Those details give the next builder a way to explain a different result instead of assuming a failed installation. The remaining comparison would show whether the same hardware continues producing valid messages under a repeatable test, rather than only displaying one impressive afternoon’s traffic.

The firmware only receives RF. The reported range belongs to the maker’s tested setup. A useful next comparison would hold location and antenna constant while changing the receive path, so the filter’s contribution can be measured separately from the view of the sky.