Page 1 of 1

Update 4.6o (experimental, no auto update)

Posted: Sat Aug 21, 2010 9:02 pm
by admin
Update to Meteohub V4.6o (NSLU2, x86, SheevaPlug) brings these features:
  • Fixes a bug that prevents from sending emails on alarm conditions
  • Adds support for pure leaf wetness sensors on Vantage
  • Adds support for virtual fail-over sensors and virtual sensors that computes difference of other sensors' values. Format to hand-over parameter to gawk scripts has changed slightly since 4.6n. Meteohub manual explains how to make use of failover and difference sensors in section 2.5 and appendix P.
Link: http://www.meteohub.de/files/update-v4.6o.upd

Warning, this update contains major code changes for various Meteohub components. Omit this update when you are focusing on a maximum stable production system. However, virtual sensors might be something exiting to explore for early adopters taking the risk of discovering bugs. Please install this update via manual file-based method. Downgrading to 4.6m or earlier will set your configuration to standard settings, as settings file has changed in format and size and is not backward compatible.

Re: Update 4.6o (experimental, no auto update)

Posted: Sat Aug 21, 2010 9:10 pm
by admin
Some info about the new virtual sensor types (taken from the manual)

Failover
When designing a failover mechanism, you need to define a master (which provides data to the virtual sensor during normal operation) and a slave, which provides data instead of the master sensor, when the master has not been sending data for a certain amount of time. To make this definition your have to make two entries in virtual sensor section for the new ID.

Code: Select all

New ID    Name          Trigger      Conversion
th9       failover-th   th0          failovermaster
th9                     th2          failoverslave 300
Example above defines "th9" to receive data from primary sensor "th0". When "th0" has not sent data for more than 300 seconds and "th2" provides fresh data, then data from "th2" is used to feed virtual sensor "th9".

Sensor Difference
When designing a sensor difference evaluation mechanism, you need to define a master (which provides the reference data) and a slave, which provides data to be compared. The result represented in the virtual sensor is the subtraction of the slave sensor from the master sensor. Which part of data provided by the slave sensor gets subtracted from which part of data from the master sensor is controlled by a position argument. A second parameter is the time frame how long a recorded value is regarded valid. When the time frame has passed no differences are computed.

Code: Select all

New ID     Name       Trigger     Conversion
t9         tempdiff   thb0        sensdiffmaster 1 300
t9                    th0         sensdiffslave 1 300
Example above defines "t9" to represent the temperature difference between indoor ("thb0") and outdoor ("th0"). Difference is only computed when sensor data is not older than 300 seconds. Values to be subtracted is parameter 1 from "thb0" sensor (first parameter on THB sensors is temperature) and parameter 1 from "th0" sensor (first parameter on TH sensors is temperature).

Re: Update 4.6o (experimental, no auto update)

Posted: Sat Aug 21, 2010 9:16 pm
by admin
A virtual sensor definition that does a computation of temperature difference between outdoor temp and windchill could look like...

Code: Select all

New ID     Name        Trigger    Conversion
t8         chilldiff   th0        sensdiffmaster 1 300
t8                     wind0      sensdiffslave 4 300
As wind chill temperature is parameter #4 in WIND data sets (as lined out appendix P) "sensdiffslave" has to address position 4.

Re: Update 4.6o (experimental, no auto update)

Posted: Sun Aug 22, 2010 11:37 pm
by wfpost
Since updating to 4.6o, a new sensor has appeared with my VP2 named Leaf-4
I do not have a leaf sensor. Therefore, I guess, this is a bug?


Image

Re: Update 4.6o (experimental, no auto update)

Posted: Mon Aug 23, 2010 8:12 pm
by admin
When I understand Vantage right, there can be only a maximum of 2 leaf wetness sensors. I will limit that to 2 in the future.