Possible issue with greater than sensor value & decimals?

Hi!

I have water pipes in the ground bit too close to surface so I have electric heating in the pipes (just in case ) which is controlled by indigo and Fibaro switch. There is temp sensor also next to pipes to measure ground temperature by the pipes.

Last few days have been quite chilly (-20...-25 celsius) and ground temperature has been dropping close to zero (normally around +4 celsius). I have trigger to start heating when ground temp sensorValue becomes less than 0.5 celsius which triggered this morning when temp dropped to 0.4. But immediately when turned the heating on, the OFF trigger (sensorValue becomes greater than 0.5) got triggered too.

Well, I suspected that the sensor value has risen and was something like 0.51 but it only reports with .1 precision . So I changed OFF trigger to greater than 1 celsius but it didnt matter, it got triggered too. Then 1.1 and now it seems to work fine.

So it seems that there might be some rounding issue with device state triggers when dealing with sensor values and decimals. Or what do you think Matt & Jay?

[matt (support)]

I'm not sure what is going wrong here. This should work regardless of the decimal precision of the sensor. That is, there is no limitation on how small/large the change has to be.

Try creating a Trigger for when the device state has any change, then for the action execute an embedded script (like below) to log the sensor value to the Event Log. This might help you troubleshoot/debug what is going on.

Code: Select all

dev = indigo.devices[520564319] # change ID for your device here
indigo.server.log("sensor value is now: {}".format(dev.sensorValue))

Thanks Matt, I’ll debug value with that script.

This issue seemed pretty odd since I’m pretty sure I have similar trigger with decimals somewhere...

I also checked device history values from database and it didn't give any new ideas - all values saved to db are with one decimal.

But I leave the logging on for a while to catch data & the sequence of triggers.

Update on this issue.

I think this issue had nothing to do with decimals and/or rounding!

Yesterday we had similar situation - temperature has been between -10...-20 celsius and water pipe temp has dropped below 0.5 celsius (which is my trigger temp).

Same thing happened again: heating went on and couple seconds later it went off. Then I happened to check out pipe heaters power consumption which showed something like 2100W(!).

So I guess Fibaro double switch (FGS223) was overloaded and it turned itself off and caused chain reaction with my triggers.

What is interesting and might have to check couple things:
- heater cables are 2x10meters which both have power consumption 10W/m so 100 Watts each and they self-adjust their power needed
- if 2x100W cables causes consumption of >2000 Watts so Fibaros energy metering is way off or...
- heater cables are overheating which I dont think is the case since pipes felt warm, not hot or melting hot

So does anyone have similar experience/opinion about Fibaros energy reporting?

I might have falsely blamed Fibaro energy reporting . Couple rounds of google gave me explanation - those self adjusting heating cables with 10W per meter consumption have that nominal consumption at +10 Celsius! So true consumption varies by temperature quite a lot and found forum post which listed power consumption of almost similar cable (12m, 120W) at different temps like so:

Code: Select all

+3 Celsius: 340 W
+6: 240 W
+12: 200 W
+22: 140 W
+30: 90 W
+50: 19 W

So this matches quite well to Fibaros energy report which show currently (at +0.1 Celsius) about 550W.

So my setup with two cables connected to one channel of Fibaro switch probably overloaded the switch (6.5A per channel on resistive loads, roughly 1380W) when heating started.

[matt (support)]

Thanks for the follow-up post and description. That is quite a variance of consumption based on the temperature.

Thanks for the follow-up post and description. That is quite a variance of consumption based on the temperature.

Yeah but since those cables designed to prevent pipes from freezing, its understandable. But still, there should be max consumption mentioned somewhere...

And also sorry Matt&Jay that my first assumption was Indigos basic triggering conditions!