Zero Wind Chlls appearing with Build 188

[Gordon-Loomberah]

Hi Steve, I've just installed build 188 and shortly after restarting Cumulus I had a 0 dew point on the graph, which I corrected in the log file. After restarting I have noticed a couple of zero wind chills appear, (it should have been ~14C).
I zoomed in on the graph and noticed the graph was showing 0 wind chill, while the display said it was 14.
I'm not sure how that works...

thanks, Gordon

[Gordon-Loomberah]

I've just noticed on the graphs that I have a zero humidity point at the same time as the 2nd zero wind chill.

[steve]

It looks like your logger had a set of zero values in it. I've spotted a bug in the code I put in to validate that the humidity is greater than zero, but I'm not sure what I can do about wind chill and dew point, as zero is valid for those. In the monthly log file (sep09log.txt), what did the other values look like where the zeroes appear?

Are you using wind chill and dew point from the station, or calculated by Cumulus?

[Gordon-Loomberah]

It looks like your logger had a set of zero values in it. I've spotted a bug in the code I put in to validate that the humidity is greater than zero, but I'm not sure what I can do about wind chill and dew point, as zero is valid for those. In the monthly log file (sep09log.txt), what did the other values look like where the zeroes appear?

Maybe dont allow jumps of more than x between readings, at least for the DP, like with the temperature? Zero is valid for both, but a sudden drop of 14C isnt likely with the DP. I guess it is possible with wind chill, but still unlikely.

Gaps when I shut it down for installing latest build at 9:02, then when shut down for editing.

05/09/09,09:00,14.3,83,11.4,7.0,12.2,141,0.0,12.4,1015.9,31.6,15.5,61,0.0,14.2,14.3
05/09/09,09:01,14.3,83,11.4,5.9,12.2,139,0.0,0.0,1015.9,31.6,15.5,61,0.0,14.3,14.3
05/09/09,09:05,14.4,83,11.5,0.0,0.0,0,0.0,0.0,1016.0,31.6,15.4,61,0.0,14.4,14.4
05/09/09,09:06,14.4,83,11.5,0.0,0.0,0,0.0,0.0,1015.9,31.6,15.4,61,0.0,14.4,14.4
05/09/09,09:07,14.4,83,11.5,0.0,0.0,0,0.0,0.0,1015.9,31.6,15.4,61,0.0,0.0,14.4
05/09/09,09:08,14.4,83,11.5,0.0,0.0,0,0.0,0.0,1015.9,31.6,15.4,61,0.0,14.4,14.4
05/09/09,09:09,14.4,83,11.5,0.0,0.0,0,0.0,0.0,1015.9,31.6,15.5,61,0.0,14.4,14.4
05/09/09,09:10,14.4,83,11.5,0.0,0.0,0,0.0,0.0,1015.9,31.6,15.5,61,0.0,14.4,14.4
05/09/09,09:11,14.6,83,11.7,0.0,0.0,0,0.0,0.0,1016.0,31.6,15.5,61,0.0,14.4,14.6
05/09/09,09:13,14.7,0,11.7,0.0,0.0,0,0.0,0.0,1016.1,31.6,15.5,61,0.0,0.0,14.7
05/09/09,09:14,14.7,82,11.6,0.0,0.0,0,0.0,0.0,1016.1,31.6,15.5,61,0.0,14.7,14.7

Are you using wind chill and dew point from the station, or calculated by Cumulus?

Calculated by Cumulus.

thanks, Gordon

[steve]

I don't understand how Cumulus can have calculated a wind chill of zero when the temperature and wind speeds are OK. Those entries at 09:07 and 09:13 were presumably 'live' - i.e. not taken from the logger at startup?

[Gordon-Loomberah]

Yes I assume so, although I think I must have restarted again just before the 9:13 entry as 9:12 is missing. As I mentioned initially, I could see the correct value shown on the main window display, but the graph was simultaneously showing zero, which I thought was very strange. Are the graph values calculated independently?

thanks, Gordon

[steve]

Are the graph values calculated independently?

The graph values are the same, but they're snapshots at one-minute intervals (or at the logger interval, if taken from the logger), so they can differ after only a second or two.

I've uploaded build 850 with a fix for the zero humidity check; could you please try that and we'll take it from there?