As you can see: the dashboard shows a lot of consumption: it rises when the solar production rises.
The MP1 however shows almost no use at all.
The story is this: I was researching the capabilities of victron in preparation of a future force off-grid situation.
I did a lot of tests with ess, as I did not have any normal solar setup.
Eventually I let professionals install the victron equipment into a container so I can move to the new situation where I am forced to be off-grid for the next 4 years (after a promise of maximum 6 months).
Ok, too much info. But: I turned off ess, and thought, that’s it.
But the dashboard still hangs on to ESS use. My battery charging is shown as consumption. While I actually want to see my 3 phase MPII out as real consumption.
I mean, in the end the numbers still add up. But I am really almost using no energy during the day, and usually also not during the night. However, the cold seasons set in and now I have energy use during the night. and I need to get a good grip on these graphs to get some more insight.
The lingering ESS is visible on the dashboard on the To AC Input and From AC Input.
What can I do to remove the lingering ESS profile from my dashboard in order to make it more useful to me without purging all the data.
I mean, the data stored seems to be correct, the advanced overview is just correct information.
I got to e-mails from VRM that triggered me wanting to fixed this:
“DESS error code: ESS functionality not available”
I used the dbus command to really turn ess off yesterday. dbus -y com.victronenergy.settings /Settings/DynamicEss/Mode SetValue 0
as the value was still 1. (Which was suggested by the ESS FAQ).
Setup:
3 phase MPII 5000, smart solar mppt 250/60, cerbo gx, some victron lynx stuff, 3 seplos 10E BMS’s as battery, DG75E diesel generator for emergency.
As a sight for the future: I will probably add a lynx and a smart solar before the winter starts and 12…16 solar panels to minimize the freezing moments I will have before my new house is usable.
From then on, I project to go to about 30…40 panels of around 500Wp to get around the winter with minimal emergency feeding with a generator in my new house. It has no use to wait 4 years for an unreliable grid connection. But this is what I didn’t know a few months back.
To be clear: forecasted total use is 11kWh, forecasted total solar is 11kWh.
My battery has been charged by 16% (1 percent is roughly 450Wh) so actually my battery has been charged by 7.2kWh and I will have used 5kWh of power, which is something I in no way can find back in the dashboard (and not in the advanced).
The reason I want to know my real use, is that if it’s so cold that I had to use a heater, I need to know how much it used, so I can predict my generator use and the length of my battery use.
If you are getting messages from DESS that is enabled on the VRM so probably should disable it there.
Using the dBus does not remove ess assistant from the inverter itself. So re program that.
Besides your consumption the system itself has self consumption.
You would need to add energy meters for specific loads if you really want to know what they use.
With ESS assistant the battery is raised to absorption voltage everyday then it drops to float. In a keep batteries charged system you still see power going in and out of the battery that way even without discharging to a certain SOC.
The inverters do not contain any assistants for a long time. ESS was nowhere to be found in the VRM page.
This is I think pure a problem with the online VRM dashboard.
Self consumption can not be that big.
I have 3 big energy meters (MPII) that can not give a total of used power.
According to the dashboard, the battery is hardly used or charged.
The advanced page, or actually just the graphs of all the data, that one does make sense, but it’s not really handy.
The kwh csv download is broken as data is put in the wrong columns, but the all data csv download is correct. The kwh csv actually shows that PV to battery looks more like my real use and the PV to USE is what is going to the battery. My cerbo console agrees with me.
It’s like the ESS mode I had enabled as a test, has created some aggregation flows that really makes no sense in off-grid, or I did configure something wrong.
So for now I have to download the all data csv, and try to make do with that.
The worst about it, is that I have no idea what the loss is of the system.
Is there a way to figure out how consumption is calculated? Because the POV of consumption currently has no use in my case.
Or is there a way to change the behaviour of the dashboard, so I can change the formulas used there?
Ess is programmed on the inverters. Not sure what you have changed there so can’t help there with that.
Its a difficult one as the power consumption of the system changes as the load goes up and down.
Most calculations use heat losses among others.
Your set can also be efficient/inefficient depending on physical layout etc as well as inverter efficiency at certain loads since inverters are more efficient in certain ranges.
Roughly a 3 unit three phase system uses in self consumption and looses 1kwh (or more) of power per hour from what i have seen with the transformer type.
It is also load based since obviously the larger the load to inverter ratio the more heat is produced etc. And of course how many other devices are in the system (mppts gx etc
There is no way to change the VRM dashboard it is really a one size fits all solution and obviously that includes on an offgrid systems, so the boxes are there for all.
To avoid calculated figures, metering is the best way.
Metered out the battery, and metered to loads. Then the rest are losses due to heat inefficiency etc.
It is possible one of your meters is sending incorrect data to the system and it is being included in the wrong data set.
Otherwise self consumption as i mentioned before increases with load as in erter efficiency changes and heat is made
It is possible one of your meters is sending incorrect data to the system and it is being included in the wrong data set.
None of the meters is sending incorrect data. However, it is being included in the wrong data set.
This is obviously a bug somewhere in the VRM dashboard that might have been caused by having used ESS.
The total consumption this morning does not even add up to 1kWh, while my battery SOC clearly show a decline of 3.5kWh. So yes, as I already said data/formulas are not correct. And an insight in the formula would be handy.
None of the meters are incorrect, it’s the formula that’s wrong and I am trying to determine why, and for that I need to know, what it actually tries to calculate.
As far as I can see, it sees the battery as a consumer, and it sees the MPII as battery when there is solar power. The kWh csv download from advanced suggests this, as it shows very high values as consumption which according to the cerbo and all metering went into the battery, and it shows very low numbers as battery charge, which actually is what I am really using.
So: PV to battery in the kWh csv should be PV to consumer, and PV to consumer should be PV to battery. This can only be caused by mixed up properties, not with wrong metering.
I just need to know how I can fix those properties.
Or a link to the right part of the documentation of the vrm.
So to conclude:
advanced->download->kWh => wrong aggregation,
advanced->download->raw => correct metering.
I have the same or similar problem. My situation:
3 phase MP-II 5000, smart solar 250/70, cerbo GX, AC-PV systems, 3 Seplos BMS 10E (3x16kWh) connected via CAN, 3phase grid meter, Wallbox EV-charger on AC-in.
Everything is looking good, except the values for total consumption. The value is far off especially when the sun is shining. This was also the case before I installed the smart solar. A possible but not so probable cause could be an apparant bug in Seplos BMS: it reports only charged / discharged of the master BMS, not of the three slaves. Other values are correct from all three BMS’s. This is not a likely cause, because the high consumed values occur also when there is no charging / discharging.
What I’ve noticed is that it only updates charged or discharged, if the counter actually changes.
If I graph the MPII kwh inverter to AC out it shows the correct MPII data.
It also has some point where the Seplos has reported results.
But it is an interesting suggestion: The seplos does have 2 counters: charge counter and discharge counter. In my case they are pretty close to eachother.
So I’ve put them all in a graph, and that makes more sense than the dashboard ever did.
This image is a bit confusing, but that’s because the mppt has a scale that’s 4x as compressed as the others.
The MP2 kWh counter is reset on every tiny change.
So these are for me the important numbers. I now realize the seplos does give the wrong charge/discharge info, although the current and voltage are correct.
Stalking on your other thread, I think that you might be correct with has DC-system that should be turned off in our case.
So the consumption in my case at night is almost 0.10kWh per hour, while it should be 0.3…0.4kWh, especially due to the use of heating.
The hard find is that the battery reports discharging correctly, just not the discharge counter, which isn’t shown.
I was already looking at the rs485 connection, as the seplos does the same querying as that pc software would do. And then implement the “master” as a third party device and let the seplos just do what it should do: protect the battery and balance it a bit.
To conclude: @lxonline was right about the measurements. However these measurements are not shown in the advanced dashboard. You have to add a widget with the kWh counters to be able to compare these numbers. @robmeerwijk was spot on on where it went wrong and had a solution: turn off “has DC system”. A for me long forgotten option that I (maybe together with my installer) long ago enabled because I was certain I would be having loads of 50V to 12 or 5V DC converters.
When has DC system is turned on, it will add all kWh counters together and the remainder is booked on DC system. For me this block was new and I never questioned it, thinking it was the step from ESS to HUB1 mode.
So if you have a multi Seplos 10E, do not use has a DC-system, or do not use the Seplos 10E can measurements.
I wonder if we can just do multi-can bms, because that makes it good. The battery boxes are independent anyway, and I rather want to be able to disconnect one box for maintenance while the other 3 are still working.
The relevant thread with the discussions about what is here:
