ProfBoc75
commented
1 year ago Hi @Mikayex ,
I'm looking to your raw samples and let you know later my findings. I'm not surprise, this kind of weather station can be found with several options, with or without the Wifi for the display and with or without the UV/Lux on the weather station. I was already facing that situation with an older Optex Weather Station, 990018 with UV & Lux ( a rebrand from AOK ref 5056 ), it was very close the Holman ws5029 (without UV and Lux) already decoded by rtl_433 . I have rewritten a device driver for this older Optex/AOK , but I never shared it here, since the CRC / Checksum algorithm was never found, so lot of errors.
Mikayex
I've an Optex 990051, the little brother of the Optex 990040 (which is the Emax EM3390W6 under the hood). It is correctly recognized as an Emax weather station as the protocol is almost the same, everything is decoded correctly except UV and Lux sensor, as it doesn't have any! Also the internal reference, according to the documentation, is EM3551H-DCF.
Here is some decoded data:
And here is some raw data:
Before the integration of the Emax decoder, I was actually trying to decode my station (independently). Most of my findings matches more or less the W6 ones, except for the data after the rain bits :
4 UV: 0 LUX: 0101 G: 0b CONSTANT: 04050607080910111213141516 CHECKSUM: 89838a(my data)0 UV: 1 LUX: 0101 CONSTANT: 0405060708091011121314151617 CHECKSUM: 9dad9e(data on W6 notes)I found out that the
Gbyte correspond to wind gust*1.5. In my example, 0x0b/1.5 = 11/1.5 = 7.3333... and the station displayed 7.3 km/h gust speed. It worked for every test I made.The constant part is not the same between the 2 models (the W6 has one more byte, to compensate for the lack of
G?) Is it a good idea to differentiate the 2 models by that (in which case, I can open PR for that)Also, a bit unrelated but for documentation purposes, the battery low flag triggers when voltage <=3.7 V
I'll add some test samples later today
(cc @ProfBoc75 you may be interested ?)