Consumption with switch off

[charliehorse]

Hi Folks
I have a problem with consumption of power.
I have an Owl power monitor. I have three double tube light fittings with LED tubes and starters.
I noticed the power consumption with the lights on was a little on high side registering
around 500 watts for six small tubes.#
I disconnected all the light units with the aim of connecting them up one by one to trace the culprit.
To my surprise, after disconnecting the lot, I still have around 350 watts being used without any ligths on.
If I put the switch to the off position, it scrubs 350 watts off the Owl monitor.
Put it to the on position and it adds 350 watts but it does not have anything that I know of connected.
There are two unused loose wires on the same ceiling but nothing connected.
Could there be a live touching maybe a wall and it is leaking current?
Surely this would trip the leak to earth breaker.
Unfortunatly all the wiring is above the ceiling panels
Any ideas would be great
Charlie

 

[James]

this is most likely to be an unknown load attached.
if you can see where the cables are heading look for anything heading out of the room.
outside lights, office next door, some growing operation in the loft that you dont know about??

 

[Lucien Nunes]

350 watts cannot simply go missing. It's either a lot of light or something unrelated powered via the switch. It's most unlikely to be a wiring fault, as that would be dissipating the power as heat and the fault would get blisteringly hot in seconds and burn out. I have an idea though that the 350W isn't really being consumed at all. You mention starters, suggesting that the fixtures previously had fluorescent tubes in rather than LEDs (LEDs don't have starters). If so, is it possible that the fluorescent control gear, including the power factor correction capacitors, is still fitted to the fixtures?

The relevance is that (AFAIK) the Owl monitor ignores the power factor of the load it is measuring; it senses only the current and calculates the power / energy consumption based on an average figure for typical domestic loads. If the fittings still have their PFC capacitors, which no longer have the lagging power factor of a fluorescent ballast to correct, they pass a wattless current at near zero power factor that doesn't register on the 'proper' electricity meter. But the Owl does not know this because it ignores the actual power factor and uses its average figure, so it thinks the capacitors are consuming real power. Obviously the capacitors are in circuit even when the LED tubes are removed.

Whether this is the explanation for this anomaly or not, it's something to watch out for with any energy monitor that does not measure power factor. People report odd findings where appliances seem to have outlandish standby power usage, which turns out to be incorrect because the power factor has not been taken into account by the monitor.

 

[charliehorse]

Thanks James and Lucien for your replies.
I could not find any other wiring.
I disconnected the light fittings entirely from the circuit by disconnection all the live feeds. The monitor still shows an increase if I turn on the switch.
Surely this should isolate the lights. I will disconnect the neutral also and see if that cures the problem (if it in fact is a problem). From what Lucien is saying it could be the Owl monitor throwing a false.
I got the Led tubes with a "starter" to fit instead of starter that CFL tubes have.
They seem to work ok with those. Should I disconnect the PFC capacitors?
Strange how the monitor shows an increase in watts with no lights connected. Maybe a better thing to do would be to monitor the meter. The reason I investigated usage in the first place was an large power bill I got.