[Xastir] SHOW OF HANDS REQUESTED: Davis Weather Station users

[Curt, WE7U]

On Thu, 19 Jul 2012, Tom Russo wrote:

> On Thu, Jul 19, 2012 at 01:34:14PM -0700, we recorded a bogon-computron collision of the <curt.we7u at gmail.com> flavor, containing:
>>
>> Well, rain I can do...  But Davis I cannot.  I have enough parts for two or three LaCrosse stations and one Dallas One-Wire station.
>
> No good.
>
>
> On the other hand, maybe if you can put one LaCrosse together and observe
> the Own Weather Data screen over the span of a day or two of rain to see if
> it reports data properly.

I can possibly put one together indoors and simulate rain into the sensor, then simulate more during a 2nd day.  I want to finish my bike project first though before I get distracted.  We're not having rain right now in the northwest either, but it's usually not long until we do.


> The issue is that Xastir needs a weather station to report "total rain" and
> uses the (monotonically increasing) value over time to compute 24-hour, 1-hour,
> and since-midnight values, which it then reports as an APRS WX report.
>
> The Davis code (and by extension, the LaCrosse code) reads 24-hour rain data
> from the weather station and stores it in wx_rain_total --- incorrectly, as
> far as I can tell.
>
> Later, Xastir uses "cycle_weather" and uses this wx_rain_total to compute
> 24-hour and other values, but since Davis is already storing it as 24-hour,
> it's going wrong because the value isn't a steadily increasing thing.
>
> Also, the Davis code, unlike other code, doesn't call "compute_rain" right
> after reading data, so some of the numbers are complete junk coming out
> of the parsing code.
>
> The patch I sent to Mike copied a compute_rain call out of other parts of the
> code, but I'm not convinced it's right yet --- especially since the Meteo
> code (which is used also by LaCrosse) pulls 24-hour data out of the wx station,
> not total rain.

I remember having to add the rain computation code because some weather stations didn't provide enough data, or data in the correct format, that we needed.  Perhaps I messed up the Davis code at that point.  The weather code is a complete mess, but was so before I started hacking into it.  It could use a complete re-write.

-- 
Curt, WE7U.        http://wetnet.net/~we7u
APRS:  Where it's at!        http://www.xastir.org