Battery power in vs out...

So I have had the SOC bug in the past and my system does appear to be fixed in terms of sudden drops. However the graphs on the dashboard are not tallying with what I would expect. Today I was looking at the data as I am considering getting a second battery, however I noticed something. I charge on a cheap rate nightly and here is that charge in the energy graph (battery in):

Here is the power graph for today also.

The battery can start discharging from 5am and tou see that usage over the morning , it even got minimal solar this morning at 9am ish... recharging slightly. Then I put on the dryer and the battery discharged down to 4% .

Here is the energy (battery out):

There are often missing bars which I can understand, the ones between 8-9 would be similar trend to the surrounding data so would not significantly increase the output, the totals are not exactly 2.4KwH which I also can understand. However they are significantly off here (.65KwH)

So The question, is this common with people? I suspect it's software rather than hardware but is there a fix coming? As a software developer I find these things infuriating because it causes me to doubt the hardware .

Thanks
--John

Sure is some serious issues with Energy Graph data atm for AC coupled systems.
And the Power Graph tab remains corrupted on my Dashboard, in spite of clearing caches etc etc.

CaribouJohn so with energy in and out of a battery, they won't be equal due to the conversion losses, however yours are on the high side (1.66/2.31 = 71%) so a small bar missing from the graph wouldn't explain that. Did a calibration discharge/charge get done when you updated the firmware?

Cdent Yeah it did happen and I could see where the upper and lower limits were "reset" - I was getting a much better output. What are the ballpark expected conversion losses, I do understand that the DC->AC conversion will drop the actual output but I'm not sure what that would be.

Depends really. If you charge the battery from the grid then you're probably looking at 16-17% losses for the charge-discharge cycle. If its from PV then it'll be a bit less as its DC-DC charging. Its going to depend on a fair few things really, including the rate of charge/discharge amongst other things like ambient temperature.

I just think of it as a 20% overhead in terms of choosing when to charge from grid on Octopus Agile, easy to work out in my head that way.

CaribouJohn as Vestas mentions, battery to home losses in the 15% area are the norm. The inverter will be powered by the battery as well which would account for some small losses too

Cdent The losses may not be so small if you're charging battery from grid/PV while pulling a couple of kW to the home as well.

Inverter is in a nice 20C environment (insulated integral garage) and the last "price plunge" on Agile (been a couple recently) saw me pretty much max out this G1 inverter in terms of battery charge/pv/grid to home supply. Heatsink at the back got pretty toasty but when its negative or zero import costs then, well it'd be rude not to "help out" 🙂

I have to say its an unexpected side-effect of getting a PV system/battery - wasn't really expecting to want bad weather (wind etc) but overnight wind results in some very low unit costs on Agile....

Vestas Cdent thanks guys its helpful to understand this 🙂

Actually if GivEnergy are watching this thread an entry to show "losses" would be useful , it could be married up to the Battery In total so that we can see at a glance whether there is (a) a problem, and (b) what the losses are in reality - presuably including the power requirements of the unit itsself and the conversion from DC to AC (AC-DC if charging) .

The app has a setting to show calculated load, portal shows the same on the graphs.

Losses are mainly in the inverter rather than the battery chemistry under normal usage. eg - you'll have about 100W being consumed by the inverter at very low home load levels when taking power from PV/battery.