I have connected Multiplus-II 3000 to an old LiFePO4 battery pack and is being used only for trading. after the recent update the dashboard shows the money in net. I have observed that there has been days that it makes money for me, but there are also days I loose money.
I fail to understand why is the system trading if it cant make money? Is my configuration broken or the DESS?
Regards.
I just dont think DESS is intelligent enough to do what we expect, or think we expect, I just made a post where DESS is taking some electricity from the grid, rather than taking it all from the battery and solar, so costing me rather than using the free energy I have already generated,.
I switch DESS off and all electricity comes from battery and solar, and zero from grid
Thanks for replying. I think the battery efficiency is also to blame here. It would be nice if we could set a minimum difference between buy to selling rates ratio which could ensure profit. Is there any such option to configure?
I’m not sure it would affect VRMs internal profit calculation but you seem to have the same price calculation for both buying and selling. At least I don’t get to add VAT, grid fees, electricity taxes to my sold electricity so that might artificially inflate the sell price vs reality.
This is my formula: (i get only spot price minus the operators margin (0.002€/kwh) for my sold electricity)
Also you might want to restrict the maximum invert and charge power of your inverter to make sure you maximize the round trip efficiency.
Can’t tell from your screenshots what your inverter power is during selling but the efficiency seems to dip below 90% little over 2kW output power.
I think efficency is the issue here.
your example shows 8 kWh pulled from grid, but 5.8 kWh only feedin. That is 72% efficency, if we expect nothing went from bat2consumption.
To avoid this, you should increase the battery costs, which would make the window between “buy/sell” for trading larger.
I.e. considering your first screenshot, where you “lost” 41 cents on a day with 5.8 kWh Feedin, I would probably increase battery-costs by 7 cents, making dess aware that having a price difference bellow ~7 cents does indicate “no trading opportunity”.
At that day it seems, that DESS bought for:
and sold for
which WOULD be a well profit of 3 cents with the 1 cent battery costs you’ve set up. But since efficeny doesnt allow a 1:1 relation on buy/sell volume, the overall result was negative.
This would align with the thinking that “efficency losses” of 30% are costs caused by the battery-roundtrip: 32% “waste” on a 27cent purchase is 8.64 cent of additional “battery costs” per kWh.
So, i’d play around with costs of 6-9 cents there, and see what is most satisfying.
Instead of playing around with the battery costs, I would prefer to set the system efficiency correctly. It is set to 90% by default, which is to high in many cases.
That’s also my opinion! Because that works regardless of whether prices rise or fall more sharply later on.
Hey fellows,
thank you all for your input.
I haven’t delved into the formula as such, it was the default formula that comes up when I select the Provider and at a glance I think it is right.
Thanks for bringing up the efficiency chart. and as for the efficiency of the Multiplus is concerned I think i am in the clear since its capped at 7A. so somewhere around 1.6kW. But I know that the battery efficiency is bad.
I too would like a multiplication factor in the calculation rather than just the battery cost. I did some reading from the above given links, got me thinking if the systemefficiency setting is round trip for the battery or just one way into the battery. I saw a topic on the same discussion but didnt see a concrete answer. my setting in it was 90% I changed it to 72% (via ssh and dbus-spy), its just been a few days but I did see a day with loss, but i think its too early to report.
what are your recommendation? is 72% a good number? there is no battery to load. since is no load connected.
Thanks to one and all for the response.
I think you losing money with pv is the target of our governments 
just download your Victron production report over some month and calculate your efficiency. Here for example:
1206,6 / 1644,3 = 74 %
You can set the efficiency also via Node-RED instead of ssh and dbus-spy if you want.
What also bothers me regarding this topic is that these charge / discharge counters can be very much inaccurate. Reported kWh from my Pytes Ebox 48100R via BMS-CAN are proven to be based on “Coulomb” counters with fixed voltages.
That means it will count only As (Ampere seconds) and multiply that with a fixed voltage to obtain kWh.
But in reality the voltage is not constant. Depending on the currents it will bump while charging and drop while discharging. In general, charge voltages are higher than discharge voltages. This is also where the charge / discharge losses are coming from (related to internal resistances).
IMHO, counting with a fixed voltage means underestimating these losses and thus overestimating efficiency.
Imagine 105Ah are being counted for a full charge and 95Ah are being counted for a full discharge.
With fixed 53V this means 5565Wh charge and 5035Wh discharge, hence 90%.
With 54V charge voltage and 52V discharge this means 5670Wh charge and 4940Wh discharge, hence 87%.
A smart shunt could help here.
But still we have the charge / discharge losses from the inverter which will not be taken in to account by the shunt.
So in the end it’s truely reasonable to be pessimistic about round trip efficiency and I also started to limit charge power to halve of inverter power.
PS: Pytes is stating 95% round-trip, but counters say it’s 82%. I guess they measured 95% with 1A charge/discharge currents 
Yeah, it’s the same as a car manufacturer says “You’ll can drive this with 4.5l / 100 km”… yeah… you “can”.
At the end, a “real” efficiency is “In vs Out” under your very custom load-profile. Tracking my system since a few months yields about 80% Round-Trip efficiency which is quite good and something i’m totally fine with.
Apparently, lowering the CVL resulted in less burned energy for balancing (apparently, not sure if that is the real cause)
Interesting! I did’t feel balancing is so much of a power hog but real numbers are much more trusty. Depends on the battery probably.
But I’ve limited charge current to 60% of max. My PV power won’t exceed this anyways but DESS will not charge with max power anymore where the inverter itself has only 87% efficiency.
