Mike Casey avatar image

VRM Numbers not adding up over 24 hours

8175a3d4-065d-4b5a-bbac-1b65cefb4a49.pngI’ve written my own custom controller to set the grid set point based on spot price here in New Zealand which has been through the roof recently because of a gas shortage.

ive been reviewing the Victron aggregated logs and have always assumed that grid input + production ~= grid export plus consumption although this isn’t the case. I was wondering if someone could explain where the remaining kWh go? I know inverters are 95% efficient but that doesn’t account from missing kWh in my mind anyway.

In the first photo you can see around 4-5kw is missing from live view based on discharge amount of battery and pv production

In the second photo you can see that this seems to also be the case in data aggregated over the last 24 hours. I consumed or exported 193kwh but produced or imported 246kwh.

Thanks for your help!


victron community
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
Guy Stewart (Victron Community Manager) avatar image
Guy Stewart (Victron Community Manager) answered ·

The missing quantity goes into efficiency and transformation losses.

Sorry to say that a round trip from an AC PV inverter, through a Multi, into charging a battery, then back out discharging a battery and then inverted back to export has an efficiency of MUCH lower than your expected 95%.

In your real time snapshot example - you're supplying 28kW from the battery, and 10.1 kW from the AC PV array (strange that nothing is coming from the DC PV?). That production is supplying 29.9 kW to the grid as export and 5.5 kW to loads.

Total production = 38.1 kW

Total consumption = 35.4 kW

35.4/38.1 = 0.93

Giving you a total real time efficiency of 93% - not too bad.

Now that is only going to get worse over time as instead of just counting that production as discharge from a battery, some of it goes into charging a battery (and then discharging it again - experiencing losses on both sides).

In particular the step from charging a DC battery from the AC supply is much less efficient that charging a DC battery with a DC supply, or using the AC supply direct to load or export.

That round trip all told can be 20% loss, depending on temperature, voltage, age of cells, measurement accuracy, wiring run length, copper size, and connection integrity.

1 comment
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

My follow up question ended up not being posted inline to yours so just messaging to make sure you get notified of it:)

Mike Casey avatar image
Mike Casey answered ·

Aha! This may be the case where a computer programmer with know electrical engineering knowledge thinks he is clever than he really is, but if you have a moment I’d live you to sanity check what I am doing.

I have a contract to buy power at 12-19c nzd during the day from my energy retailer. I then have a seperate contract to export power at 10% under wholesale. The spot price has consistently been at 0.45c due to a natural gas shortage.

So I’ve realised (by mistake - my program ran for days doing this after months or running smoothly and discharging only occasionally when market conditions were right) there is a way to potentially make some pocket money / pay for my batteries if I get $.405 on everything I export, why not just import power at .19 and export it again immediately? What I was not aware of though was the AC to DC penalty but I think my figures still check out?

Since 20% gets wasted I’ll add that to the cost of the power I import, bringing that up to circa $.25 so there is still a $.15 margin per kWh.

now I’ve said the cost of wear and tear on my batteries at $2 a cycle (byd batteries guaranteed 6000 cycles to 80% DoD and cost 10500 nzd each) - not sure if that’s adequate?

$2 over 15kwh is 0.13c a kWh which now only means a 2 cent profit per kWh, however of I move to only charging between midnight and 4am I can get that power at 12c and increase profit to 7c per kWh.

Still not as good as I originally thought!

2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Please take this only as a curious intellectual exercise, and not manufacturers advice :)

It's not as bad as you might think, the economics of lithium batteries are such that they are going to be degrading anyway over time, regardless of cycles - so cycling them as much as you can, (as long as temperature and charge rates are controlled) will mean you can extract as much value from them before they inevitably fail anyway. Read your battery warranty conditions carefully though, as there might be some parameters that you want to stay within there.

From a design perspective, get as much DC charging going as possible, so your AC PV can efficiently export its entire capacity as AC out to the grid (without the transforming), and have the DC MPPT chargers to charge the DC battery efficiently.

You definitely want to design around a charge doing that off peak low cost midnight to 4am, and then be pushing that back out again to allow a charge from the PV when it comes up (leaving yourself a little buffer for blackout protection).

Mike Casey avatar image Mike Casey Guy Stewart (Victron Community Manager) ♦♦ ·

Very interesting indeed. I acknowledge any all communications in this thread are strictly of a theoretical nature and indemnify you and Victron from any liability!

Three questions if you don't mind

1) When you say charge rates, are you referring to controlling how fast the batteries are charged and discharges or something else?
2) What setting is it to instruct MPPT to always export (or consume on AC loads) power coming from AC PV?
3) Do you think a 50/50 AC do DC PV split is the right way to go here?