3 phase supply, single phase monitoring

Hi All,

have just installed an emontx and emonpi with 4 CT’s. I originally planned to monitor grid power, solar, hydronic heat pump and HWS heat pump. I was then going to be able to calculate total house consumption as grid-solar, as well as monitor the consumption of the heat pumps directly.

I had assumed single phase power, and was rather embarrassed to find I actually had 3 phase power when the electrician arrived to install the CT’s behind the switch board.

I decided to skip monitoring the grid connection, and instead monitor the induction stove and oven consumption, as they were fed off one phase, and are the other major source of energy usage. The heat pumps were on another phase and the solar on yet another. The 9v supply was on the same phase as the solar.

I then realised that due to the phase shift between the 9V supply phase and the others, the power was out by a factor of 2 for the heat pumps and stove/oven. I fixed this in the feed with a x2 multiplier.

Thought this may be of interest to others with a similar set up.

[Edit: The “solution” in the final sentences above is likely to give false readings, and is not a recommended solution. - Moderator (RW) ]

That’s NOT a solution. If the power factor of your load changes - and let’s face facts, it won’t be a nice value of unity for the induction hob even if it is for the oven - then what you read will be in error, and the error could go either way depending on whether the load has a leading or lagging power factor.

You only have to think of the phasor diagram for your situation to realise what’s going on.

Thanks for the feedback. I see your point that this really only works for resistive loads. Interestingly I can measure the HWS heat pump consumption directly, and it is very close to the value that the emontx measures, which is surprising given I would expect most of the load to be the compressor motor. I can’t check the induction cooktop directly as it is hard wired in, but I will check the specs and see how they compare.

Can you suggest an alternative approach? It seems to me that with a 3 phase supply but the loads split over different phases, this will be an issue given the 9V supply will only be synced to one of the phases.

I think you’ve missed the point regarding the 3-phase sketch. It doesn’t know the amplitude of the other two phases of course, because there’s only one voltage input available, but it phase locks to that and shifts it by the nominal 120° either way to use as the reference for the real power calculation on the other two phases. So when we describe it as “approximate”, the approximation is only in the voltage balance across the phases.

You could buy Robin’s 3-phase energy diverter and customise his software to produce the numbers you require. That’s genuine 3-phase with 3 voltage and 3 current inputs. He might even put together a “monitor-only” kit for you (that’s up to you to negotiate).