#zephyr.
Posts on the Alpisto engineering blog filed under#zephyr.
- 01#flybeeper#fanet-vario#sun-vario
The note that never sounds: what a piezo vario actually plays
Small varios beep in a way nobody would call pleasant, and it is not because nobody cared. A piezo disc is a bell with one note of its own, and a vario plays a tune on it by feeding it square waves and letting the bell pick what it likes. I put one in a desk drawer with a microphone: at 1000 Hz the note that was asked for comes out 33 dB below its own third harmonic, loudness swings 16 dB across the tone table, and the obvious way to soften a beep turns out to make it sharper.
17 min read - 02#flybeeper#sun-vario#variometer
Fourteen metres that were never there: a soldered barometer's thermal drift
Everyone who builds these devices knows you must not solder near the barometer, and that the third rework kills it. Nobody says by how much. This one came off the bench at 470 Pa/K against a rated 0.5, inventing 14.5 metres of climb while sitting still. Thirty lines of firmware removed 95 % of that and every bench test agreed — until somebody picked the device up and the vario went the wrong way. The number behind the folklore, and why the fix helps an altimeter and provably cannot help a variometer.
24 min read - 03#flybeeper#fanet#ads-l
Teaching a LoRa vario to hear ADS-L, and finding the time to do it
An ADS-L receiver must know when each UTC second starts, and receivers normally get that from GNSS. My vario has no GNSS on purpose: its position comes from the pilot's phone. So I took the time from the phone too, over Bluetooth, and measured how accurate that is. It is accurate enough, with room to spare: the share of frames the device caught went from 15 % to between 50 and 100 %.
19 min read - 04#flybeeper#sun-vario#variometer
Measuring a vario with a USB tester: 20 seconds per mode, and 420 µA that were never mine
I spent eight hours per data point measuring a solar variometer through its own capacitor. Then I measured the same device directly with a USB tester and got a full consumption curve in twelve minutes — along with a 420 µA floor that belonged to the cable, not the device. What both methods agree on, where they don't, and how to read a WITRN tester from Linux.
14 min read - 05#flybeeper#sun-vario#variometer
A 10-farad vario, measured: where every microamp goes
Eight hours of discharge logs on a solar variometer running off a 10 F lithium-ion capacitor. The current isn't constant, the buzzer isn't the main consumer, the barometer is — and the runtime formula that falls out of it. Plus what the enclosure had to become once the electronics stopped changing.
18 min read