I think it was discussed several times already, and again and again, that you absolutely must calibrate your sticks before running anything related to this project. You'd shoot yourselves in the foot applying a fixed offset that works in a single setup that uses no calibration at all.
Please also note that the original rtldavis repo has a -26 kHz average error on the frequencies, probably because the author also forgot calibrating. Try to use the well-known ones from the RFM69 sources: https://docs.google.com/spreadsheets/d/1SyNKF-ynoYvKEiOfr7CcZSvaKDU49rJLbNvfN0avzbQ/edit#gid=1265405143 - if you calibrate and use the calibration ppm, you can be in for a surprise again, just because of the offset error in the original source. I'll try the EU frequencies again after the many small fixes Luc made... On Monday, March 25, 2019 at 11:28:58 PM UTC+1, rich T wrote: > > Luc > > The fc offset does not work for me. Wondering since each dongle ppm is > most likely different the fc number will vary. Let run kalibrate on the > dongle and figure my offset. > > Rich > > pi@raspberrypi:~/work/src/github.com/lheijst/rtldavis $ > $GOPATH/bin/rtldavis -maxmissed 4 -tf US -fc 71000 > 18:18:54.206814 rtldavis.go VERSION=0.10 > 18:18:54.207304 tr=1 fc=71000 ex=0 maxmissed=4 actChan=[0] maxChan=1 > 18:18:54.208089 BitRate: 19200 > 18:18:54.208139 SymbolLength: 14 > 18:18:54.208177 SampleRate: 268800 > 18:18:54.208212 Preamble: 1100101110001001 > 18:18:54.208246 PreambleSymbols: 16 > 18:18:54.208280 PreambleLength: 224 > 18:18:54.208313 PacketSymbols: 80 > 18:18:54.208348 PacketLength: 1120 > 18:18:54.208382 BlockSize: 512 > 18:18:54.208416 BufferLength: 2048 > Found Rafael Micro R820T tuner > 18:18:54.624193 {ChannelIdx:0 ChannelFreq:902348338 FreqError:0} > Exact sample rate is: 268800.001367 Hz > 18:18:54.806291 Init channels: wait max 133 seconds for a message of each > transmitter > 18:19:17.794307 TRANSMITTER 0 SEEN > 18:19:17.794747 Hop: {ChannelIdx:19 ChannelFreq:911881597 FreqError:46} > 18:19:20.421401 ID:0 packet missed (1), missed per freq: [0 0 0 0 0 0 0 0 > 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 > 0 0 0 0 0] > 18:19:20.421613 Hop: {ChannelIdx:41 ChannelFreq:922920108 FreqError:46} > 18:19:22.983341 ID:0 packet missed (2), missed per freq: [0 0 0 0 0 0 0 0 > 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 > 0 0 0 0 0] > 18:19:22.983559 Hop: {ChannelIdx:25 ChannelFreq:914892100 FreqError:46} > 18:19:25.545226 ID:0 packet missed (3), missed per freq: [0 0 0 0 0 0 0 0 > 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 > 0 0 0 0 0] > 18:19:25.545491 Hop: {ChannelIdx:8 ChannelFreq:906362342 FreqError:46} > 18:19:28.109203 ID:0 packet missed (4), missed per freq: [0 0 0 0 0 0 0 0 > 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 > 0 0 0 0 0] > 18:19:28.109582 Hop: {ChannelIdx:47 ChannelFreq:925930611 FreqError:46} > 18:19:30.671166 ID:0 packet missed (5), missed per freq: [0 0 0 0 0 0 0 0 > 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 > 0 1 0 0 0] > 18:19:30.671453 Init channels: wait max 133 seconds for a message of each > transmitter > 18:19:30.671644 Hop: {ChannelIdx:0 ChannelFreq:902348338 FreqError:46} > > > On Monday, March 25, 2019 at 12:03:16 PM UTC-4, [email protected] wrote: >> >> Hi Paul, >> >> Good news! >> >> I'm curious if the offset of 71000 Hz also works for Rich. >> If so, we simply add (the minus 71000) to all frequencies in our US list >> and we are done! >> >> Luc >> >> PS. >> The best way is to sort out why we need to use these >> trial-and-error-frequencies and not the common known and used frequencies >> of other projects. >> So far I didn't hear from the original author and I'm still trying to >> understand the ins and outs of the demodulate process. >> I wish I could find a useful article about FHSS and FSK decoding of Davis >> weather equipment signals... >> >
