question

Do you mind me asking what type of batteries you have?

Similarly what settings have you got within your smart shunt?

You're right it does sound odd. May it be that the smart shunt settings aren't calibrated correctly for your battery type, could you share those settings? (I originally had done something similar with my LiFePO4 batteries which caused odd VRM behaviour)?

I have victron gel total 660 Ah at 48V. As far as I know settings in shunt affect only estimated time to go and help with SoC being more accurate during discharge (peukert and efficiency). But never anywhere do these interfere with the current passing through.

I was thinking more that the efficiency/peukert settings may be incorrect so SoC of battery wasn't accurate. That way victron VRM could think your battery is say 0% and stop discharging but in reality, it could be 20%. Which might explain why usage from battery is lower because cerbo/Venus os is stopping discharing based on SoC which is incorrect.

If you're confident in your efficiency/peukert settings then I'm out of ideas sorry. They sound awfully big losses for anything like self consumption etc.

When your batteries are at 0% SoC as per VRM/smart shunt is the voltage what you'd expect to see at 0%. If not then maybe it is those efficiency settings.

This past 6 months reporting got me wondering so I bought the shunt specifically to decide if batteries are crap or not. And they are OK, VRM is wrong. I also checked the shunt and summed up everything leading to the quattros with a clamp ampermeter and the shunt is spot on, very accurate.

It acts as if VRM only gets data reported by quattros which is ~300W less than what actually passes through the shunt.

Maybe I should enable in cerbogx "has DC system"? Maybe this way VRM will be forced to get kwh data from the shunt instead of quattro estimate?

What do you see in Victron connect app for usage in Smart shunt? If that differs from VRM then i think you've worked it out that the data isnt coming from Shunt.

If that's the case i think there's a setting in cerbo (settings->system setup->battery monitor). See if that's not on Smartshunt, might be the solution

I think the "has DC system" is more for devices that have another shunt like a 12v system/load.

In cerbo battery monitor it is set on shunt.. I set it like this when I put the shunt in, hoping over night vrm will show same as shunt, but no... (it was quattro before)

Hi @ikeakayke

What VRM is missing in these screens is the amount Grid has fed to the batteries. Only Solar gets shown as a battery input. This is unfortunate, but you should be able to calculate it out. When doing so, consider that with your ESS there's a possibility that some of your export may actually have come from the batts.

The historical shunt figures for Charged/Discharged Energy can be seen in Victron Connect. They're cumulative from the last history reset, and you can't pick a time period. This is gospel though for the batteries.. I find the VRM-derived calculated figures show about 1.5% lower in efficiency return from the batteries, but that's got the losses too from the whole system.

How far you draw your batts down each day will alter the return you get, but don't be surprised if you see down to like 80%. Such are Pb's.

@JohnC During night there is no export. First night with the new shunt: vrm 22kwh, shunt 28kwh. This no longer involves solar, grid, nothing, just pure current drawn from the batteries for the sole purpose of inverting. This is roughly in-line with what VRM collected all these months. VRM calculated efficiency 57% for brand new victron gels, shunt calculated efficiency (over the last few days since I got it) 94.35% so as they should.

I marked as 'visible' in vrm dashboard quattro monitoring as well and during nights quattro reports less realtime kw than the shunt by about 300-400w.

I guess at this point is pretty obvious there are 2 problems:

1- quattros do not measure input current on the DC side when they invert/idle but probably whatever goes out after the inversion takes place and somehow it estimates battery usage

2- Although primary battery measurement is the new shunt in cerbogx and the one displayed in main VRM dashboard, VRM data collector does not appear to use it but still favors quattros. Will see if this changes by enabling 'has DC system' as I don't have any other ideas.

Everything is latest firmware except quattros which are 1 or 2 versions behind but reading release info nothing that I would need/impact/solve/improve to make want to reboot them.

From the manual of ESS:

 4.3.10. Limit inverter power

Limit the power drawn by the Multi: ie. limit the power being inverted from DC to AC.

Notes:

    The losses in the inverter/charger are not taken into account. If you want to limit the amount of power being drawn from the battery, you will have to set this limit slightly lower to compensate for those losses.

So victron is aware these 'losses' are drawn from the battery but is unfair to not be accounted for in VRM data collector. This can even have marketing implications since people like me buy batteries then in a few months they're mislead into thinking they're not worth it when in fact they saved energy from the grid when they kept quattros internals running at night...

I just checked a public VRM from a company in my country that specializes in PV installations.

They have tesla 40kw battery:

https://vrm.victronenergy.com/installation/89417/dashboard

Checking this year solar to battery 4305 kwh, this year consumption from battery 3128 kwh which gives an efficiency of 73% :))

This is so wrong... am I missing something? They also have multis 5kva which have less 'losses' compared to my quattros 10kva.

I've not seen losses comparable to yours on my system but I'm out of ideas as to why that might be sorry. Mine has 90% using 1 year data (all I've got).

Do you have a current sensor? Maybe a good test would be turn on a load and measure what's coming from battery vs what's coming on ac side. That would confirm the efficiency of the inverter. And confirm if the issue is in fact inverter inefficiency or a vrm data issue.

@matt1309 You probably have only one inverter with one phase and not a quattro, something smaller like a multiplus 3-5kva? If you add another 2 of them you'll reach 70% as well then. 90% for whatever battery technology you have in their first year is pretty low.

You're right multiplus ii 48v 5kva however I thought the maximum inverter efficiency of that inverter was 93% so I was quite pleased with 90%.

My only suggestion would be to do the test with a current sensor on dc side into inverter and then ac out. At least then you'll have 100% confirmed the issue to be the inverter efficiency and not some vrm/data issue.

You maybe then also be able to discuss with whoever sold you the inverter if it's acting much below spec.

Thanks for explaining @ikeakayke

That is odd, I'd never considered that before.

Just to confirm I'm following, so without that "has dc system" setting enabled the "out of battery" data in VRM is just AC Load data (assuming no solar incoming like you said post sunset)?

So now that you have enabled that setting your "out of battery" kwh in consumption data on vrm will match the DC kwh passed through the shunt, is that correct?

Does that not mean that your VRM consumption data is now inflated compared to what AC kwh was actually output by the inverter, because it's kind of "excluding losses".

ie if you put an energy meter on inverter output it would now not match what vrm is saying? Have i understood correctly here? (Not saying that's a bad thing as I can see why you want the data in the form you do, just wanted to make sure I'm understanding the limitations of each form of data ie with or without "has DC system" setting enabled)

Without 'has dc system' VRM doesn't care what comes out of your battery, just what the inverter/cerbo estimates used from battery to power AC loads on output (grid feed included if any).

With the 'has dc system' VRM consumption from battery will match exactly the primary shunt DC kwh out (for the shunt that is set as battery monitor/measurements).

So instead of estimated battery consumption used exclusively to power AC output loads and not the inverter themselves, VRM will from now on show exactly the kwh from DC battery to power anything: inverters, AC loads on output etc and no estimates needed, just actual DC kwh as reported by shunt.

After all why record DC kwh INTO the battery but lie about what actually comes out...

As to your last question, energy meter on inverter output will report the same thing inverters report, so will NOT match DC kwh.

But in my opinion since inverter electronics themselves are AC loads just like anything else on AC, the new 'has dc system' measurement makes total sense to me. If I read correctly inverters can start with 0/dead battery if connected to grid, but can't start with dead battery if not connected to grid. So I suspect the inside electronics are powered by AC - whether from grid or self fabricated by inverting - therefore 'AC loads'.

Ahh I see your point. I can see why they do it though. Imagine a consumer seeing the data in VRM with "has dc system" setting "on".

As a consumer if VRM was saying 10kwh produced from battery as a consumer i would expect 10kwh less of an electricity bill however with "has dc system" that would not be the case.

If "has dc system" was on VRM would now say circ 12kwh produced because it's using DC Kwh, as a consumer I would be more confused/outraged that my solar system is then seen to be "overestimating" production as it's not discounting it's own loss/usage/inefficiencies first.

Suppose it depends on what your focus is on AC kwh or DC kwh. I see your point that DC makes more sense when you think about what's actually happening but if i was selling to consumers i can see why they've gone the opposite way.

I wonder what data is used if the inverter/charger puts DC into the battery rather than a Mppt.

A quick solution to make clearer would be to add a "AC" or "DC" indicator on VRM page.

With the consumption page showing all information in AC when "has DC system" set to off.

I think you complicate things more that it needs... say 10 kwh/night 'losses'. If it weren't for those unaccounted DC kwh then grid meter whould show 10kwh more (inverters need to stay alive with energy from somewhere right?). So all it does is lowers grid meter usage due to kwh drawn from DC while meanwhile unfavoring battery making it look totally un-efficient. Just because it uses DC kwh for itself doesn't mean it's not used. Simple as that.

If anything a normal customer's reaction is mine looking at VRMs of people: why do you even bother with cycling batteries each night when you get just 57-70% return in useful kwh. By these measurement all should just get Fronius with no batteries. One can barely justify batteries for ESS at manufacturer efficiency as it is, imagine at 57 or 70% noone would ever use batteries with ESS (unless really needed by outages).

It depends on what you're concerned with I suppose.

If for example you wanted to calculate monetary savings using vrm data you would now get an incorrect number because the "from battery" data is as per smart shunt on DC side. ie before its been inverted to AC.

Unless with the setting enable the inverter powers itself via the grid rather than battery?

"has dc system" enabled = to see actual dc output from Battery which makes ac information not accurate.

"has dc system" disabled = to see actual ac output from inverter which makes dc information not accurate.

"If for example you wanted to calculate monetary savings using vrm data you would now get an incorrect number because the "from battery" data is as per smart shunt on DC side"

Incorrect because now I truly know how much money batteries saved me. Realize VRM didn't report some ~2000 kwh from solar PV stored in battery and then used to power the inverters when no solar available.

"has dc system" enabled = to see actual dc output from Battery which makes ac information not accurate.

False. It never makes AC info inaccurate. Grid meter stays the same. 'From Battery' consumption will now be real and Total Consumption will also be increased adding the 'losses' as it should.

"has dc system" disabled = to see actual ac output from inverter which makes dc information not accurate.

Correct, up to a point. AC output is correct, but then you could argue it is labeled incorrect in VRM as 'Consumption'. When in fact Consumption should = AC output + self consumption by the inverters since they are devices plugged into my AC lines.

Incorrect because now I truly know how much money batteries saved me. Realize VRM didn't report some ~2000 kwh from solar PV stored in battery and then used to power the inverters when no solar available.

By savings I mean comparing having a Solar system vs Not having a Solar system.

So the batteries wont have saved you an extra 2000kwh because you wouldn't have used that 2000kwh if you didn't have a solar system.

That 2000kwh is running cost of the system so to get true savings, the 2000kwh running costs also need to be deducted from your gains from solar (ie "has dc system" disabled)

Normal savings calc is:

(VRM consumption data from battery + from solar) * $/kwh

However with "has dc system on" you now need to deduct inverter running costs to get true $ savings.

So to get the actual monetary savings I'd then have to deducted the 2000kwh * $/kwh.

True Savings = Gains from having solar less running costs of solar. (less initial cost of system ofc)

False. It never makes AC info inaccurate. Grid meter stays the same. 'From Battery' consumption will now be real and Total Consumption will also be increased adding the 'losses' as it should

Sorry by inaccurate i mean VRM consumption from battery + from solar data would not reconcile to an external energy meter on inverter output.

It really depends on what you care about i suppose. I use VRM consumption data to calculate my savings rather than check my battery efficiency (i use smart shunt data for that).

My overall logic is the "useful power" in these systems is AC power, that's the output of this whole system. So in my opinion the default view in VRM should show that power as AC.

AC power in my head is the output of the whole process.

For example if i as a consumer buy a carton of apple juice the packaging shows me 900ml of liquid (the output of the apple juicing process), and not the input of the process ie "9000 apples less losses".

Correct, up to a point. AC output is correct, but then you could argue it is labeled incorrect in VRM as 'Consumption'. When in fact Consumption should = AC output + self consumption by the inverters since they are devices plugged into my AC lines.

I agree with you on this one. An additional "self consumption" heading would solve the problem but i see why they've just put it into battery.

A solar system is an investment/ improvement for the home allaround. If you only care about in how much time you can see if you recovered the initial cost, then you might be correct. But then what do you do afterwards?

Let me put it another way: would you prefer your car computer to record worse gas mileage because it knows you have hidden losses like insurance, maintenance and add them to gas? Because this is what it boils to.

Whenever you install any electrical appliance that adds value, like a tv, do you measure only watched hours diregarding standby use?

Not to mention that losses that are heat, actually warm up my house, so they even save on the heat pump by that ammount.

Your car literally does that though.

The fuel is the perfect example. The car engine isn't 100% efficient at converting the potential energy store in the fuel to kinetic energy/moving the car. The mileage calculation factors that in. The input in the car example is fuel measure in Gallons.... the input in an inverter is DC in kw....

Only difference in our example is both sides of the system are measured in kw except one is DC the other AC.

"has dc system on" is like the car mileage not factoring in it's own weight into the mileage calculation. If your car didn't factor in it's own losses/usage, the milage would be way further. What's more useful knowing real world mileage (output of the car) or a theoretical mileage excluding car usage/losses.

It's like asking someone how far your car go and them saying "12 gallons" which is the input into the system not the output which is mileage. If engines were 100% efficient we could get away with saying things like that as it would mean something to us.

I'd say Insurance isn't a direct running cost of the system. It's not charged per meter moved by the car. It's charged yearly. If i wanted to do a pure savings calc of car vs no car I'd agree should be included.

VRM shows power output including direct running costs/losses of the direct system. Same as a car shows mileage and not gallons as it's measure of output.

Your TV example you're confusing costs and "savings". The benefit (aka saving) of a TV is only when you're watching it. Otherwise, an off TV is literally an ugly picture....

Costs you're right should be factored in all time regardless of on or off. But you're not doing that with a "has dc system on" You're removing the "cost" (inverter usgae/losses) from the equation.

In one example you say solar vs no solar, in the other you disregard total cost of ownership for the car which is basically car vs no car.

If car would do what VRM does by default, it would report 5 gallons used for trip instead of 25 literally missing from your gas tank (typical engine is 20% efficient) because 'losses' :)) Now you understand?

This is perfect analogy to my findings.

Again it depends on what you care about:

• Mileage (output of the system) includes losses. What VRM shows by default. This is more useful if you want to know how far you're travelling ie the aim of a car.

• Fuel level (input to the system) excludes losses. More useful if you're a mechanic/buyer and want to know the fuel efficiency or you know the exact losses so you can work out mileage data.

You're saying VRM should show gallons, I'm saying it should show millage. The confusion is coming from the fact that it's kw on both sides except one is DC the other AC.

I'd rather know the output of my system at a glance. You'd rather know the input data.

I imagine Victron chose output data of the system to be displayed on VRM, as "from grid" is already in AC. and if "has dc system" is on the "from battery" data is essentially DC. So "total consumption" is summing ac data and dc data.