Page 1 of 1

sensdiffmaster/sensdiffslave question

Posted: Wed Jan 16, 2013 8:38 am
by thomasal
Hi folks!

I'm calculating the solar radiation out of the temperature difference between outside and in a jar ("solar in jar").

This works rather fine, but ...


At night-times the temperature in the jar is lower as the outside temperature because the jar is located on the roof of the house, and the temperature sensor for outside temperature is located on ground level.

So at night times the difference of the temperatures gets negative and so the calculated solar radiation also gets negative.

Is there any possibility to use "sensdiffmaster 1 300" with an additional gawk-"skript"?
I want to use a "if-else" command. If difference is <0 => output 0, else gawk should output the difference.

Best regards,
Thomas

Re: sensdiffmaster/sensdiffslave question

Posted: Thu Jan 17, 2013 10:23 pm
by adrian_hudson
Not a reply I am afraid, just a question... What device do you run meteohub on please? sensdiffmaster/slave used to work for me but not since I moved to a DC01.

Thanks,
Adrian

Re: sensdiffmaster/sensdiffslave question

Posted: Thu Jan 17, 2013 10:30 pm
by thomasal
Oregon's WMR200 is running @ meteohub.

But the solar radiation would be just as well negative if a would use another weather station.
The temperature in the jar is at nighttime lower than the outdoor temperature.

Re: sensdiffmaster/sensdiffslave question

Posted: Sat Jan 19, 2013 6:22 pm
by adrian_hudson
Sorry, I wasn't offering a solution. I was just enquiring what hardware you run the meteohub software on since I have been having problems with sensdiffmaster/slave with meteohub running on a DC01.

Re: sensdiffmaster/sensdiffslave question

Posted: Wed Feb 06, 2013 8:22 pm
by admin
I finally got one of my DC01 back to live and can confirm the bug.
Bug is fixed now and will be resolved in the next update (4.9u).

Re: sensdiffmaster/sensdiffslave question

Posted: Fri Feb 08, 2013 8:04 am
by thomasal
Thanks a lot! :)

Re: sensdiffmaster/sensdiffslave question

Posted: Sun Apr 14, 2013 9:52 pm
by thomasal
well, i updated to 4.9u

"fixes virtual sensor bug on DC01"

Meteohub log:
2013.04.14 21:07:17 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -2.5<br>
2013.04.14 21:17:03 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -2.5<br>
2013.04.14 21:22:16 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -2.5<br>
2013.04.14 21:27:13 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -2.5<br>
2013.04.14 21:32:12 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -5.0<br>
2013.04.14 21:37:07 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -7.5<br>
2013.04.14 21:42:11 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -10.0<br>
2013.04.14 21:47:05 Wetterstation: weathernet: awekas-push: upload to awekas: Solarwert ung�ltig - -10.0<br>

So the calculated solar value is negative as at version 4.9t?
How this problem is fixed?

Thomas

Re: sensdiffmaster/sensdiffslave question

Posted: Sun Apr 14, 2013 10:45 pm
by admin
Fix is that sensdiffmaster/slave do run on DC01 as on the other platforms.
There has been a bug around that before.

When you subtract a larger number from a smaller one, the result will be negative.
That is not a bug to be solved, but how it works. Capping negatives to zero
will not make sense in other areas of application. Sorry.

Re: sensdiffmaster/sensdiffslave question

Posted: Wed Apr 17, 2013 6:20 pm
by thomasal
And this is not possible that only the solar radiation can't get negative?
A negative solar radiation doesn't make sense anyway.

BR
Thomas