Victron ESS: Legacy Compatibility Is Now Holding Back Innovation

Victron ESS: Legacy Compatibility Is Starting To Hold Back Innovation

Please take this as constructive feedback from a long-time Victron user running a large AC-coupled ESS system.

I fully understand why Victron maintains compatibility with 10–15 year old systems, VE.Bus architecture, older MultiPlus units and thousands of existing installations worldwide.

But modern residential ESS requirements are now completely different.

Today customers expect:

  • ultra-fast ESS response,
  • stable zero-export regulation,
  • seamless wallbox integration,
  • V2H/V2G readiness,
  • advanced energy management,
  • smooth full-SOC behavior,
  • and modern UX/app experience.

At the moment it feels like the current architecture is becoming limited by legacy compatibility requirements.

The issue is not Victron hardware.
Victron hardware is still excellent.

The problem is that modern ESS logic is now running on top of:

  • old VE.Bus architecture,
  • historical compatibility layers,
  • and software decisions designed many years ago.

It reminds me of the transition from the MS-DOS / Windows 3.1 era running on 8088 processors to modern operating systems.
At some point software architecture must evolve together with modern hardware capabilities.

Today:

  • GX devices are much more powerful,
  • EM540 meters react extremely fast,
  • batteries respond instantly,
  • Fronius/SMA AC coupling is common,
  • and customers expect real-time behavior without hunting, unnecessary grid imports or MPPT instability.

Meanwhile competitors such as:

  • Sigenergy,
  • Huawei,
  • Sungrow,

already offer:

  • significantly faster ESS response,
  • smoother full-SOC behavior,
  • modern EV/wallbox integration,
  • no unnecessary grid imports,
  • stable power limiting,
  • and highly integrated modern platforms.

Many installers are already starting to move away from Victron for residential ESS projects because of:

  • ESS software limitations,
  • delayed wallbox development,
  • slow software progress,
  • and issues that continue for years.

And honestly — many users and installers are starting to become tired.
Not because of Victron hardware, but because the ESS/software development no longer seems to keep pace with today’s market.

My suggestion:

  1. Keep the current platform as a “Legacy ESS Platform”
  • full backward compatibility,
  • support for older VE.Bus systems,
  • marine/off-grid stability,
  • long-term maintenance.
  1. Create a separate “Modern ESS Platform”
    for newer hardware only:
  • new ESS engine,
  • redesigned control architecture,
  • faster real-time response,
  • optimized AC coupling,
  • improved full-SOC behavior,
  • modern EV/wallbox integration,
  • future-ready platform design.

The current approach increasingly feels like:

  • legacy compatibility continuously slows innovation,
  • new ESS changes fix one issue while creating another,
  • and competitors continue moving ahead rapidly.

Victron still has:

  • excellent hardware,
  • VRM,
  • modularity,
  • openness,
  • and one of the best technical communities in the industry.

And that is exactly why I still believe Victron can approach this intelligently, acknowledge the limits of the current architecture, and build a truly modern ESS platform from the ground up — because the potential is absolutely still there.

The RS is already a new implementation. No assistants like that which powers the vebus systems.
You seem to be conflating different concepts.
ESS will always be hamstrung by compliance and certification requirements. Has little to do with legacy support; which is a strong point for many; many users out there.

The slow ESS grid set point keeping has been solved with the faster VM meter, the issues holding back response times was the meters were slow.

So theae are technically solved

And i found this varies with battery manufacturers. But usually a few set up tweaks and we are away again.

I have a bunch not on the list integrated and have no issues there. Not such a fan of the fronius. Have done maybe 1. Have better response from other PV inerters so maybe that is a consideration..

The brand agnosic way (gx software) vrm is set up makes it so versatile compared to others.
But then i guess for some the no think locked in way is what is preferred.

I know several of these users who are using the vrm for their stuff… Just saying.

I for one do not like proprietary software solutions that basically block me from doing what I want.

The Victron energy meter is blistering fast. Note that I’m not using Ethernet for this. And there seems be a new energy meter on the horizon (V3P5A) as well. Possibly one without the need for CT’s.

Also have different brands of AC coupled inverters running flawlessly. Even the Enphase Envoy was running great, until I swapped it out for another blue box.

Thank you for the replies — and I do understand the certification and compliance aspect.
I also understand that RS is already a newer implementation and not simply “old VE.Bus assistants”.

But from a real-world installer and user perspective, the problem still exists.

I already use:

  • fast EM540 metering,
  • Fronius AC coupling,
  • modern GX hardware,
  • and still see:
    • unnecessary grid imports at full SOC,
    • ESS hunting behavior,
    • occasional MPPT instability,
    • and slower ESS response compared to newer integrated platforms.

That is why many users feel the issue goes beyond only meter speed or certification constraints.

The important point is this:
Competitors such as Sigenergy, Huawei and Sungrow also operate under the same grid-code and certification requirements — yet many installers and users report noticeably smoother real-time ESS behavior.

So from the outside, it increasingly feels like:

  • certification/compliance is only one part of the story,
  • while the overall ESS control architecture and platform integration may also play a significant role.

I am not attacking Victron — quite the opposite.
I still believe Victron has:

  • some of the best hardware,
  • VRM,
  • modularity,
  • openness,
  • and one of the strongest technical communities available.

That is exactly why many of us care so much about the future direction of ESS development.

The concern is simply that:

  • installers are starting to move toward competitors for residential ESS projects,
  • mainly because customers today prioritize:
    • seamless ESS behavior,
    • modern wallbox integration,
    • smooth full-SOC operation,
    • fast response,
    • and “it just works” user experience.

I think many users would simply like to see Victron make a bigger architectural step forward in modern residential ESS — because the hardware platform itself absolutely has the potential.

I completely agree that Victron’s openness, modularity and VRM ecosystem are major strengths.
That is actually one of the reasons many of us originally chose Victron.

And I also understand that not every installation experiences the same ESS behavior.

My point is simply that:
these “edge cases” are becoming less and less rare in modern residential ESS environments.

Today more systems are running:

  • large battery banks,
  • fast smart meters,
  • AC coupling,
  • dynamic tariffs,
  • export limiting,
  • EV charging,
  • and high-speed load changes.

In these scenarios, users start to notice:

  • unnecessary grid imports,
  • slower ESS stabilization,
  • hunting near full SOC,
  • or differences in realtime behavior compared to newer integrated platforms.

I fully respect that some installations run perfectly.
But the growing number of ESS discussions, firmware rollbacks and installer feedback suggests there are still areas where the platform can evolve further.

And honestly, this discussion only exists because many of us still care about Victron and want to see it remain a leader in modern residential ESS.

My 2 cents…
All of you have valid points.

The ones that live mostly from Victron products and services, take company side.
The users want more from the company.
Two views that are not excluding each-other.

The (other) companies mentioned here make HF inverters. These are more efficient and easy to control from many points of view.
Victron lovers say big LF tranformers are better than HF transformers, because can handle more overload.
This is a nowadays a myth. Think of electric cars. All are HF based and can handle tremendous overloads when full-throttle.

Now back on Victron… and last two years.
What is new on the inverter universe?
Just two new inverters that swapped copper with aluminum on the same LF technology…
Anything on HF?.. Quite nothing, just promises…
And from my point of view, HF is the future, but in that sector Victron has a lot to recover… and competitors are not waiting.

Like the OP, just a (little) disappointed user that is thinking of buying a third Victron inverter, for sure a HF one.
But still waiting it to be available (the RS19), together with the wow factor.

PS:
If a Multi RS GX (!!) rack mounted will appear, it will be awesome…
I know, ESS inside Multi RS now, but, no, a limited ESS, not the versatility offered by a linux system like GX card.

PS seconded.

I think a big part that is being missed is its a much smaller company than the others mentioned, with a (significantly) smaller core of people working there. So expectations of keeping up does need to be tempered by that thought.

To be honest not that ‘far behind’ if you look at the billions vs the millions…

If i recall some stats about Huawei their r&d is bigger than the whole of victron by at least 10x

You should probably talk with your installer then. Some missconfigurations could lead to this as well.

I’m running similiar hardware and my worstcase gridpull on a all sunny day is 100Wh - the smallest delta the gx is reporting at all.

Ofc there is also heavy impact of your average grid and consumer stability. Fluctuating loads and voltage will impact setpoint precision. The common thing for any ESS is, that they can only start adjusting their inverterpower if something HAS affected the grid values already. So it will never be a stable 0.

And be aware, victron just shows the truth. Some of the “apparently more stable competition” just display a 0, if it’s +/- 50W, average dashboard values per minute or anything that is “good enough for a end user watching his system” :wink:

And just to add: Gridmeters themself “lie” around 0. Check the docs, and you will see, that the EM540 for example denotes a minimum measurement current of 0.25A, which actually means: between -150 and 150W on a 3 phase grid it may be anything but accurate.

On a saldating grid, where you run with asymmetric phase currents, that’s less of an issue. But for phaseindividual or accidently high symmetric loads, keeping every phase at a physical 0 is impossible by just not beeing able to measure it at all.

Its funny you say that. I added a fluke power meter to a site and also had some interesting findings.
Things were not as they seemed.

Just my 2 Pence a lot of those manufacturers do not care about backwards compatibility which was one of the things I like about Victron. I know Sigenergy has said they would make batteries backward compatible.

I for one as a user/consumer am not interested in the constant change of products, I just want something that works and reliably so. I like the modular build as it means I can add what I want to my system.

I have 2 AC coupled Solis inverters in my build that do frequency shift so can tie in and work when the grid is down. I did find it not to be as responsive when there was a drop in solar production it would pull from the grid a lot the only way I could get it to work better was to take out the grid meter shift everything to critical load and I did use the CT for the inverter and also the VM3P75CT I’m still trying to get someone from Victron to actually enable the use of the other CT’s for there Energy meter so you can monitor three devices on the single energy meter for single phase so we are not just looking at two CT’s that we can’t use.

The single stack approach looks good but losses all the flexibility, one now I’m subscribed to your battery pack and design, the choice of me choosing another battery has gone. I run away from integrated vehicle chargers as a lot of them are so heavily integrated into an eco that they do not function on their own and a lot are cloud based.

All of this is a single point of failure but in my approach many would not take ie AC solar separate EV chargers and the Victron Whole house. Knowing that if the Victron fails I can throw a switch to the grid still save a little on energy as the solar will just continue to work and probably the more important thing is knowing the car can still be charged inorder that I can fund all of this stuff.

Here in the UK you might know of a company that was churning out new models/versions of systems they where almost constantly pumping out stuff started Low voltage, High Voltage single phase 3 phase and no backwards compatibility so systems with design life expectancy that was changeable how is it changeable you as? Well the manufacturer could change to a new version of battery or inverter making it hard to source an older version leading you to gut the system and start again that is a lot of :money_bag: .

GivEnergy was one such company that had a lot of versions they are currently in administration as I think a lot of others may go this route too, saying that at least you can pay for there cloud use when your not at home, but they left a lot of people just looking at that long warranty period well that’s what it says on paper but kind of irrelevant if the company no longer trades

All RS models are HF and there is a 19" version coming.

The HS19 has been announced waaay too early but as far as I know it’s now on the finish line for real.
Most R&D has been focused on the HS19 over the last months to finally get it to release.

After that R&D resources are free for other projects like the RS19, the Multi RS without MPPT and better ESS with the RS models.

I’m sure we will get some more information at the Intersolar in a few weeks.

Exactly this I have been saying since two years now… Each time InterSolar shown almost the same…
First year RS19, second year RS19 and HS19, and so on…
Distributors are shrugging all the way, all the time when I am standing in front of them wanting to buy the RS19…

As per topology, you can say that my inverter lost warranty in the moment I’ve transferred the money… :smile:
Among the first things I’ve done was to open it and reverse engineer the main schematic for the bidirectional LVDC/HVDC converter, bidirectional HVDC/AC converter and MPPTs.
Here I was surprised by the interesting choice of Victron hardware engineers concerning the HV part of the bidirectional DC/DC converter.
They used a half bridge push-pull configuration instead of a full H-bridge configuration like almost all others manufacturers are using…
I will welcome any comment and brainstorming from you guys here on this choice, because I doubt any HW engineer from Victron will see this message and care to comment.

If I remember correctly the the HS19 has been shown (not for public) long before the first RS models (was it before Corona? 2019 2018?).

The RS19 was shown 1 or 2 years ago for the first time.

If you are working with Victron for a few years you should know that you don’t should rely on announced release dates until you actually can order it.


I’m not a HW engineer so I can’t comment on the HW design of the RS products (and I don’t care, as long as it’s working and reliable).

That pretty much says it all for some…

Very interesting conversation - as a software engineer I just love the open concept and node-red integration. There are tons of regulations over here in GER that probably forbid super fast responses. Anyway, I am happy with my system and its countless integrations and data access to/from influxDB, … Even data scraping from mobile-alerts sensors is working. The only thing to improve the WAF - would be a quiter setup. But for me the 3-phase setup is working like a charme in the basement w/ very limited energy import and very flexible energy distribution as I want/need/program it.

Totally agree.

And yet here in North America we’d still like to have a Quattro-II or ESS. Or a split-phase inverter. I would love to have the HS19 for commercial projects and the RS19 for mobile or smaller scale off-grid in this decade. But I won’t hold my breath. Unfortunately, here in the USA, Victron is a poor choice for grid connected residential except maybe to backup 120V loads. And even then, the grid connection is only available for charging batteries or AC pass-thru to loads.

Sigenergy is a great solution for non-technical customers here. And I like the MID architecture. Their software is pretty good, but you need Home Assistant to replicate the programmability and automation capabilities of VenusOS. It’s an excellent solution for customers who just want the lights to come on when the flip the switch, especially if they have one or more electric vehicles. The V2X module is the secret sauce.

I think that some of the issues are not ‘victron’ only. My personal experience where I had existing Enphase and added VE later is similar. But we have to take into account that with two different devices and different manufacturers the systems are also reacting to each other, with the measurement & controls delays from each adding up.

enphase ‘pulses’ power in low light conditions, which makes the MPii react via the VEmeters: always a bit after the fact, because it can’t know before the peak is there. And as others have mentioned the construction of my MPii doesn’t change neither it’s strength and weaknesses.

as for the EVCS, I did see some beta-firm/software dialogues here in the forum. Didn’t go there myself since it is the one I depend on aka ‘running on stable’. The forecast has improved in a new direction, car/guest appearance @Oldrich77 what further improvements would you like there? And how would it be done if you had your say.