FYI... I edited v0.5.0b4 driver and remarked-out the line "t_rainhour" from the default field map and this has stopped the erroneous reading. It's a temporary fix until b5 becomes available. Sorry for being impatient but I hated to see that value being reported to CWOP. Does anyone have ties to Ecowitt to correct their API?
On Sunday, June 12, 2022 at 4:35:31 AM UTC-5 Alan Stankevitz wrote: > I found this thread after investigating the hourly rain problem with my > CWOP readings. I attempted to download the b5 version using: > > wget -P /var/tmp > https://github.com/gjr80/weewx-gw1000/releases/download/v0.5.0b5/gw1000-0.5.0b5.tar.gz > > But I get a 404 error. > > Is version v0.5.0b5 available yet? > > Could I just do a calibration calculation in the weewx.config file to zero > out this reading? I hate seeing the false data being reported on CWOP since > our local NWS office uses my readings from CWOP for rain data. > > Thanks! > On Wednesday, June 1, 2022 at 7:30:28 AM UTC-5 Tomek T. wrote: > >> Thanks Gary. >> >> You mentioned WU earlier. >> I looked at the data of my station on the WU website and it seems to me >> that it matches the data on the Ecowitt dashoard. >> >> Tomek >> >> środa, 1 czerwca 2022 o 14:01:13 UTC+2 gjr80 napisał(a): >> >>> Tomek, >>> >>> The interceptor driver ignores the hour rain field from the GW device >>> and consequently the CWOP uploader falls back to obtaining (valid) hour >>> rain data from the database. Whereas the Ecowitt Gateway (GW1000) driver >>> passes the erroneous hour rain field through and it is used by the CWOP >>> uploader rather than obtaining hour rain data from the database. >>> >>> I couldn't get v0.5.0b5 out today but hoping for tomorrow. >>> >>> Gary >>> >>> On Wednesday, 1 June 2022 at 21:28:05 UTC+10 Tomek T. wrote: >>> >>>> Hi Gary, >>>> >>>> thank you for your reply. >>>> It's very interesting what you wrote. It looks like I will have to wait >>>> for your driver or firmware to be updated. >>>> >>>> Before, I used the weewx-interceptor >>>> <https://github.com/matthewwall/weewx-interceptor>. >>>> Today I discovered that the problem came after changing the driver. >>>> >>>> Before the driver change, the following CWOP frames registered in the >>>> APRS system: >>>> >>>> 2022-05-21 20:41:02.12 -- >>>> /007g017t053r000p021P021b10136h82L000.weewx-4.8.0-Interceptor >>>> >>>> 2022-05-21 20:46:07.838 -- >>>> /008g016t053r000p021P021b10137h82L000.weewx-4.8.0-Interceptor >>>> >>>> 2022-05-21 20:51:12.485 -- >>>> /008g016t053r000p021P021b10136h82L000.weewx-4.8.0-Interceptor >>>> >>>> 2022-05-21 20:56:17.218 -- >>>> /007g016t053r000p021P021b10136h82L000.weewx-4.8.0-Interceptor >>>> >>>> 2022-05-21 21:00:21.347 -- /007g015t053r000 >>>> p021P021b10135h82L000.weewx-4.8.0-Interceptor >>>> >>>> Subsequent frames, after changing the driver, already have an error >>>> (r039): >>>> >>>> 2022-05-21 21:20:22.214 -- >>>> /006g009t052r039p021P021b10135h83.weewx-4.8.0-GW1000 >>>> >>>> 2022-05-21 21:25:23.025 -- >>>> /008g015t052r039p021P021b10135h83.weewx-4.8.0-GW1000 >>>> >>>> 2022-05-21 21:30:21.938 -- >>>> /008g016t052r039p021P021b10135h83.weewx-4.8.0-GW1000 >>>> >>>> 2022-05-21 21:35:27.426 -- >>>> /008g016t052r039p021P021b10136h83.weewx-4.8.0-GW1000 >>>> >>>> 2022-05-21 21:40:16.266 -- /009g015t052r039 >>>> p021P021b10135h83.weewx-4.8.0-GW1000 >>>> >>>> Hope this helps in possibly eliminating the problem. >>>> >>>> Tomek >>>> >>>> >>>> środa, 1 czerwca 2022 o 00:43:51 UTC+2 gjr80 napisał(a): >>>> >>>>> One thing I forgot. You will not see any hour rain data in the >>>>> --get-all-rain-data output as the --get-all-rain-data output mimics >>>>> the 'More', 'Rain Totals' screen in the WSView Plus app. >>>>> >>>>> Gary >>>>> On Wednesday, 1 June 2022 at 08:40:37 UTC+10 gjr80 wrote: >>>>> >>>>>> Hi Tomek, >>>>>> >>>>>> Short answer is I believe it is a device API error. >>>>>> >>>>>> What I suspect is going on is that the Ecowitt gateway device API is >>>>>> returning 10.0mm for what the API documentation refers to as ' >>>>>> ITEM_RAINHOUR' (described as 'Rain hour (mm)'). The driver passes >>>>>> this value through to the WeeWX field hourRain, which in turn is >>>>>> used by the CWOP uploader (and I believe the WU uploader). >>>>>> >>>>>> Now that you have drawn my attention to it, I note the earlier post >>>>>> with seemingly incorrect (and identical to you) hour rain data. Ecowitt >>>>>> seems to have dropped the ball on hour rain totals; the API >>>>>> documentation >>>>>> includes reference to an hour rain field from the WS90 piezo rain fall >>>>>> sensor, but the current API does not support such a field (it supports >>>>>> all >>>>>> other documented piezo rain fields). Now it appears that there is an >>>>>> issue >>>>>> with (at least) WH40 hour rain data. >>>>>> >>>>>> So what to do? The WeeWX field hourRain is seldom used with its use >>>>>> in the WeeWX code base limited to (some of) the RESTful uploaders, and >>>>>> then >>>>>> those that use hourRain have a fallback for obtaining the data from >>>>>> the database if the field hourRain does not exist. I want to do a >>>>>> few more checks (I have no WH40 only a WS90) but at the moment I am >>>>>> inclined to release a b5 driver later today with the mapping from >>>>>> gateway >>>>>> device field t_rain_hour to WeeWX field hourRain removed. This will >>>>>> prevent the hour rain data from the gateway device from appearing in the >>>>>> driver loop packets (and ultimately WeeWX archive records) but still >>>>>> enable >>>>>> a user to add the mapping back via a field map extension if they have a >>>>>> station where the ITEM_RAINHOUR data is correct/valid. >>>>>> >>>>>> As an aside you are seeing t_rainhour (an internal driver field >>>>>> holding ITEM_RAINHOUR) in the --live-data output as the --live-data >>>>>> output defaults to displaying using the WeeWX Metric unit system and cm >>>>>> is >>>>>> the rain unit used in the WeeWX Metric unit system. hourRain appears >>>>>> as 10.0 when testing the driver as the packets emitted from the driver >>>>>> use >>>>>> the WeeWX MetricWx unit system which uses mm for rain. Finally, >>>>>> hourRain appears in inches in weewxd loop packets as weewxd loop >>>>>> packets use the unit system specified under [StdConvert] in >>>>>> weewx.conf which defaults to US which of course uses inches for >>>>>> rain. >>>>>> >>>>>> Gary >>>>>> On Wednesday, 1 June 2022 at 07:02:08 UTC+10 Tomek T. wrote: >>>>>> >>>>>>> ...and after a short rain as I wrote the above post: >>>>>>> >>>>>>> PYTHONPATH=/usr/share/weewx python -m user.gw1000 --get-all-rain-data >>>>>>> Using configuration file /etc/weewx/weewx.conf >>>>>>> Interrogating GW1100 at 192.168.x.x:45000 >>>>>>> >>>>>>> Traditional rain data: >>>>>>> Rain rate: 0.0mm/hr (0.0in/hr)) >>>>>>> Event rain: 0.7mm (0.0in) >>>>>>> Daily rain: 0.7mm (0.0in) >>>>>>> Weekly rain: 0.7mm (0.0in) >>>>>>> Monthly rain: 0.7mm (0.0in) >>>>>>> Yearly rain: 0.7mm (0.0in) >>>>>>> >>>>>>> [...] >>>>>>> >>>>>>> Tomek >>>>>>> >>>>>>> -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/b071a60c-2e68-45ac-8069-1d866f2a3ffen%40googlegroups.com.
