Larger ESS system behaviours

DIY
1

I fully understand that maintaining a perfect physical zero on a real 3-phase grid is impossible.

But my issue is not +/- 50W around zero.

The real issue is:

  • MPPT power occasionally collapsing to 0W at full SOC,
  • while the Fronius inverter continues producing,
  • ESS behavior becoming unstable,
  • and then everything immediately returning to normal after a Cerbo/VE.Bus restart.

That behavior cannot be explained only by:

  • EM540 measurement tolerances,
  • grid asymmetry,
  • or dashboard averaging.

If it were purely a meter physics issue, a software/service restart would not instantly restore correct ESS behavior.

That is why many users suspect there are still unresolved software or ESS state-management issues involved, especially in larger AC-coupled systems.

2

I think you should probably create some threads for dedicated issues, provide some data and relevant settings and a lot of the enthusiasts around here will have prompt solutions or improvement suggestions.

Just a general “it’s not working as i expect” is neither phrasing a concrete problem to look at, nor allows to suggest where to start…

This for instance alone can have many many many reasons:

  • Solar string voltge to close to battery voltage
  • Peak Shaving settings requiring curtailing of production to obey limits
  • Simple MPPT Tracking doing a fullscan. (All inverters do that, they usually just don’t reflect that in their reported power output)
  • String voltage raising to high, making the MPPT going into overvoltage protection.
  • String current raising to high, making the MPPT going into overcurrent protection.
  • Thermic throttling on the MPPT
  • Thermic throttling on the inverters requiring curtailing on mppts to avoid overvoltage of the battery
  • … and many more
3

I understand all of those points — and yes, any MPPT can temporarily reduce power for many legitimate reasons.

But in my case:

  • there is no thermal throttling,
  • no overcurrent,
  • no overvoltage,
  • no peak shaving limitation,
  • battery voltage is within normal range,
  • temperatures are normal,
  • and the system can immediately return to correct behavior after a Cerbo/VE.Bus restart.

That is exactly why many users suspect this is not simply “normal MPPT behavior”.

Also, this is not an isolated case anymore.

There are already many discussions from users running:

  • large AC-coupled ESS systems,
  • Fronius integration,
  • fast energy meters,
  • modern batteries,
  • and seeing similar full-SOC ESS behavior.

So while some cases are certainly configuration related, I do not think every report can simply be explained by:

  • thermal throttling,
  • MPPT scans,
  • or measurement tolerances.

Especially when:

  • the issue appears mainly near full SOC,
  • affects ESS coordination behavior,
  • and is temporarily fixed by restarting the ESS/GX side of the system.
4

I absolutely agree - but the case that there are systems NOT showing this issues also disqualifies the thinking that “it’s a general control / architecture problem”.

Yes, there have been bugs in firmwares and control-logic. There are some presently. And there will be new in the future as well.

Overvoltage and overcurrent are btw damn hard to identify - because the MPPTs are exactly avoiding that to happen at all and stay within specification. (Which just means, measurement data won’t show this, external measurements won’t show it as it is literally prevented to happen to avoid damage)

5

I understand your point, and I agree that not every issue automatically means there is a fundamental architecture problem.

But at the same time, the fact that some systems work perfectly does not automatically invalidate the experiences of many others either.

In my case — and in several very similar installations I personally know:

  • the systems operate within specification,
  • temperatures are normal,
  • voltages are normal,
  • currents are within limits,
  • and yet the behavior still appears mainly near full SOC and ESS coordination scenarios.

That is why many users suspect there may be a combination of:

  • firmware edge cases,
  • ESS state-management behavior,
  • AC-coupling interaction,
  • and specific real-world operating conditions.

I fully accept that some problems are configuration related.
But when:

  • multiple experienced installers,
  • similar hardware combinations,
  • similar ESS topologies,
  • and similar symptoms
    appear repeatedly across discussions,
    then I think it is fair for users to question whether there may still be unresolved ESS/software behavior involved.

And honestly, I appreciate this technical discussion much more than simply pretending the reports do not exist.

6

Just to clarify — I am not coming at this as a beginner user.

I have been operating PV systems since 2011 and hold an official license for electricity production from PV.

So my comments are not based on “one bad setup” or a basic configuration misunderstanding.

I fully understand:

  • AC coupling,
  • ESS behavior,
  • export limiting,
  • battery charge control,
  • MPPT operation,
  • grid interaction,
  • and real-world residential PV operation over many years.

That is exactly why I am trying to describe the difference between:

  • normal ESS physics and expected regulation behavior,
    versus
  • situations where the system behavior appears inconsistent and temporarily resolves after GX/ESS restarts.

I am not trying to attack Victron.
I am simply describing what many experienced users and installers are observing in larger modern ESS installations.

7

The community is not a real barometer, it has skewed representation.
There are literally millions of systems, here it is mostly biased to a small diy and self-install sub-section.
The vast majority of issues are self-inflicted.
Sure we solve real issues, but those are quite apparent and get addressed and are a minority, beyond betas.

You may also notice the community trend of members tagging themselves as expert and professional, when the topics posted clearly show the opposite.

Large scale flaws get addressed, there simply is way more work happening behind the scenes beyond this community.

8

I understand your point regarding community bias, and I agree that forums naturally attract users experiencing issues.

And yes — many DIY/self-installed systems certainly have configuration mistakes.

But at the same time, I do not think it is fair to assume that most reports automatically come from inexperienced users or fundamentally broken systems.

There are also:

  • experienced installers,
  • licensed PV operators,
  • long-term ESS users,
  • and professionally maintained systems
    reporting similar behaviors in certain ESS scenarios.

In my own case:

  • the system is professionally configured,
  • monitored long-term,
  • and even accessible directly by Victron staff through VRM administrator access.

So I fully agree that forums are not perfect statistics — but I also think repeated reports from experienced users should not simply be dismissed as “community noise”.

Especially when many of the discussions revolve around very similar:

  • full-SOC behavior,
  • AC-coupled ESS interaction,
  • and GX/ESS coordination scenarios.