Re: [weewx-user] 3.9.2 > 4.2.0 - A few minor things

[Tom Keffer]

The issues are not related.

1. maxSolarRad is a theoretical calculation, based on your latitude and the time of day. It is not measured.
2. The algorithm for calculating heat index was changed to the new NWS algorithm. I must admit, I don't like it either, because it can show values below sensible temperatures. I'm thinking of allowing the user to make a choice between algorithms.

-tk

On Fri, Nov 6, 2020 at 7:14 AM Jared <jez...@gmail.com> wrote:

Hi again,

First I want to say that you guys are awesome.  I took Thursday and Friday off from work and I figured I'd take care of this upgrade, which includes a move to a new system running CentOS 8.  I thought it might wind up taking most of this long weekend.  But I basically did it in a single day, including all my customizations.  Even switching from the rpm-based install to setup.py was not difficult.  You guys have done such a great job and I love working with this app.

As mentioned, the update has gone great.  So far I've come across two differences that I'm struggling to find an explanation for, and they may be related.

Since I'm still running weewxd manually right now, I'm getting to see the loop packets come in to the shell terminal.  I'm not sure why, but I'm seeing a value for maxSolarRad, and it's steadily increasing as the sun rises.  My Vantage Pro 2 does not have a solar radiation sensor.  This is the first weird thing.

The second weird thing is that I'm seeing Heat Index values that are lower than the current temperature.  Is this possible?  Since I was flipping back and forth between the 3.9.2 and 4.2.0 installs, I have it captured the phenomenon in images:

3.9.2 = https://i.imgur.com/BlMfgu9.png

4.2.0 = https://i.imgur.com/0awFxSd.png

Are these items related?  What am I missing?

Thanks

--
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 weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/7b61def8-38a8-4002-9384-ec16a14fbd49n%40googlegroups.com.

[Greg Troxel]

Tom Keffer <tke...@gmail.com> writes:

> 2. The algorithm for calculating heat index was changed to the new NWS
> algorithm <https://www.wpc.ncep.noaa.gov/html/heatindex_equation.shtml>.

> I must admit, I don't like it either, because it can show values below
> sensible temperatures. I'm thinking of allowing the user to make a choice
> between algorithms.

Sounds good to give people choice of course. It seems the tricky thing
is various hardware calculations vs various algorithms in software and
how the defaults change that. And the difficult UI thing is to have
unambiguous but not enormous labels for the N kinds of varying
country-defined Heat Index formulas. It seems that "Heat Index" is no
longer a single concept. But HI-NWS is icky as a label.

I don't understand the notion that Heat Index must be >= temperature.
It seems Heat Index value X means it feels like X and Y% humidity, for
some specfied Y (50?). So with lower humidity than Y%, it makes sense
that it doesn't feel as hot as the actual temperature.

I think we're used to this strict relationship with wind chill, but
there the reference wind is 0, so wind chill is necessarily
always <= temp.