Disparity between amount consumed and amount being used

Hi,
If I look at the graph for the amount of electricity drawn from the grid and the amount the house is using, the two figures are not the same. This is at night when there's no solar power and my battery has ran out of juice so they are not a factor.

There is a difference of 150W which is a bit high to be a margin of error or Inverter idle usage (which I read is less than 15W - I'm not sure if that somehow bypasses the house consumption figure). I've got a Gen 3 8kW model just fitted a few days ago.

Is this normal?

Thanks

petersmithson you are looking at a slightly inaccurate reading and a calculated (and thus definitely inaccurate) reading and trying to compare them.

Electricity meters are not 100% accurate. There is a small margin of error (if you are comparing against your actual supply meter)
The inverter grid import/export meter is driven by a CT clamp on the grid supply. These are less accurate than the grid supply meter.
House demand is not measured by your inverter, it is calculated based on import-export+solar gen-battery charge+battery discharge.
Your inverter and battery both do use some power. I think it’s more than 15W though.
The inverter data is sent to the portal is reported in kWh not watts, so there’s a risk of rounding errors

I’ve learnt to not bother too much about the graphs and losses that occur in the process. Having solar and batteries I import less than I would have done without it, thinking about the losses are just a bit depressing

geoffreycoan hi. Thanks for the reply.

I'm not comparing the meter reading with the clamp thingy. I'm looking purely at the graph on the GivEnergy app which does not have access to the meter provided by the electricity board - as far as I know.

The calculations you mention about the battery etc shouldn't apply at night with a battery that is not charging or discharging.

The inverter will certainly use more when it's doing it's job - around 200W. But when idle, it uses less than 15W.

So as I'm looking at a graph that only has one source - that clamp thing, theres not much scope for error. The point about rounding errors is really clutching at straws - even a 4 byte floating point number (let's face it - most would be 8 bytes now) would not loose 150. It would be tiny fractions. Belive me, I'm a computer scientist! 😁

Cheers
Peter

petersmithson I agree, but you have ignored the most important thing, its a GivEnergy graph!

They’re not great. Have a look at other forum posts and you’ll find lots of comments about issues with the graphs. They make a number of assumptions and often have data formatting issues. I can’t speak specifically to your issue but it doesn’t surprise me that it doesn’t look correct.

Sorry

Can you measure how much power the setup is consuming using your smart meter? Either by turning everything else off, or by turning off the AC isolator and seeing how much the power drops?

For reference, a gen1 inverter consumes about 15W, the Gen 3 may be different.

Measuring the batteries is more difficult as they draw power from the batteries themselves. My estimate is the old 48V batteries consume about 15W each when idle. How many batteries do you have?

Rubikcube I've got 3 batteries and the manual for the inverter says it draws up to 15W when idle. I think the graph drawing is just off. There's only one place it measures the current (someone else made me realise this) so it's just drawing the graphs in an odd way. Perhaps it thinks my baseline is 150W so there's no point in showing that all the time and it treats 150W as being zero when drawing the load - that's the only thing that makes sense! :-)

geoffreycoan I think you're right - it's just drawing the graph in an odd way.