interactiveinstitute / elvis

Educational tool for measuring energy usage with Fibaro Wall Plugs on Raspberry Pi
https://www.tii.se/projects/energy-in-a-box-at-komtek
GNU General Public License v3.0
3 stars 0 forks source link

Countdown continues when someone unplugs the entire plug with the load from the outlet. #49

Closed Anton04 closed 10 years ago

Anton04 commented 10 years ago

This could be improved by regularly checking that the plugs are reachable and send a 0.0W value when they are not.

sander commented 10 years ago

Done some small improvements, but we don’t get notified directly when a plug is unplugged. Right now the code waits for a timeout and then sets the power to 0 W.

Fixing this in a neater way seems to require more work that I can’t do these weeks. Note that even the Z-Wave.Me expert interface takes up to 30 seconds to recognise that a plug is disconnected, when asked for a sensor update. It doesn’t recognise this at all when doing nothing, or when just updating settings.

Implications for the user manual:

sander commented 10 years ago

@rebekahlolsen wrote:

We still have an issue with the visualization reporting energy use even after an appliance is turned off and unplugged. I test my hair dryer. After I turned it off, it continued to show energy draw in the visualization. I then unplugged the hairdryer while keeping the plugwise in the wall. It still continued to show energy use on the visualization.

I know that we can address this in the manual to the teachers, but how hard would it be to fix this? I assume the teachers that are testing it in the next few weeks will also see this as a problem. Rather than making them adjust how they use it, it would be worth it to fix these bugs.

Just to be sure, did the visualisation show progress when the hairdryer was turned off for more than a second, and the plug had not been unplugged yet? As long as the plug is connected, the visualisation should recognise each power change (including going to 0 W) within a few seconds.

The originally described issue (disconnecting the plug causes the power usage to remain constant) seems due to the way the Fibaro plugs work. I cannot find a technical solution for that. Maybe you can ask a Fibaro expert to look at the (public) code on this site. :-)