Mismatching load data

I've more recently noticed that the load figures no longer add up in both cloud and local on the portal and API. Is this a temporary bug?

Note the app reports correctly

i.e. in API/portal:
537W + 48W != 612W - incorrect data
while in the app:
692W + 21W = 713W - correct data

Nope, this is DC/AC conversion losses. There will also be losses going the other way. Important to factor this in when you're working out the true cost of shifting energy around your home and charging/discharging your battery.

I think the app is sanitising the data to make it look like there aren't losses - that just an opinion btw, it would need to be verified by GE.

browellm interesting if that's true, maybe GE could verify? I'm keen to see if this is a linear percentage with usage.

jd notwithstanding inverter losses I think that the Givenergy plant knows PV output, battery input/output and grid import/export (from CT and Modbus Meter) and the figure in the middle (house load) is simply their mathematical balance of those flows for a point in time. I find my myEnergi app represents flows better simply because of the update frequency. Local inverter access will improve the picture I’m sure.

I'm also monitoring via givtcp locally which matches the portal information. Would be nice to get peace of mind on its accuracy i.e. where is the actual loss. I can imagine customers asking if they were to do the math!

jd Further to this, I've been looking at the values in the app and MQTT explorer too.

During these sunny days with the battery SoC at 100% and a solar surplus,

In app house demand = Solar_to_House + Solar_to_Battery.

Solar_to_Battery is always consuming around 70-180W. So I don't know if this is just the "price of doing business" through the inverter and covers the DC/AC conversion losses, or whether there's scope for optimising how the modes work.

My conclusion is that the timings of the data points are out so they don't match up in real-time so I've resided with that fact. The totals seem to match up after a day showing only a small % loss by the inverter efficiency.