What would happen to the ESS if one uses the new EXTERNAL transfer switch ability as in the image attached? Would the setup work?
SETUP NOTE: Regular “AC Loads” setup and measurement with a carlo-gavazzi meter
Load Split “Essential Loads A” and “Essential Loads B”
Am I missing something?
Do people realize the implications of this new functionality? Absolutely awesome (if it can work this way…)
I am planning to make a setup with an external transfer switch.
You only connect your loads and grid to AC Out on the Victrons.
AC In is just for synchronization when the Grid power restores.
And I was wondering too if ESS works in such setup.
Does someone know this or even has build this setup?
Take a look at this news at Victron Professional (free account needed):
It’s certified for grid use (ESS) with an external NS protection (Ziehl UFR1001E)
This is awesome honestly. Opens up a lot more flexibility in terms of installation possibilities.
Looking forward to implement this setup where I get the chance.
The CTs are obligatory in the external transfer mode with firmware s99?
There is a grid meter that gives the data to the mp2’s, would this meter replace the CT function?
Anyone?
Before I will order the Ziehl, I had recently set up a test system with the S99 firmware in a 3-phase setup.
The master unit uses the Aux Relay to switch the External Relay.
I encountered a phase rotation issue between the phases, which initially prevented two of them from synchronizing.
I adjusted the phases to a clockwise phase rotation, after which the phase synchronization was successful.
You can monitor this process via VE.Configure for the VE.Bus system.
The MultiPlus system syncs the phases, and internally (on the master unit), you hear the Aux Relay trigger for a very brief moment.
This process repeats itself. I could hear a ‘changing, faint buzzing’ sound indicating that the phase syncing checks are being performed, and after a short delay (or “after a few cycles”), it triggers the Aux Relay again.
You can see a small time bar on the VE configure monitoring screen during this.
This process repeats as long as the External Relay fails to connect the Grid to the Victron System.
Is this mechanism designed to ensure that a very quick-responding Relay is used?
How much time does this external Relay have for connecting the Grid, so the Aux Relay keeps the signal on?
The CT measurement is also used in the algorithm to determine if the External Relay should be open.
If current is still flowing (when it shouldn’t be), it results in Error 24, and the system enters a safe mode.
I will set up the test system again with a separate relay to gain further insight.
20250902
Remote Transfer Switch application works as intended, at 26.5m distance from the fuse box.
Not even a blink is visible, you wouldn’t notice if the grid disconnects also the SMA Tripower x15 keeps running without interruption and power decrease. After switching the Grid back on the voltages are different and therefor the PV overproduction flows into the Grid the moment the contactor closes.
Also noticing that PV power regulation gets very slow adjusted to load the batteries.
Don’t know if this is a bug or something from the SMA / Multiplus combination.
Need to do some more testing with grid interruptions.
Components
The rest of the electrical installation is as before secured with RCD’s and some parts with RCBO’s.
After receiving the Ziehl i will install it as required in NL and put the ground relay in place.
Firmware updates do interrupt the complete installation as the contactor will be off.
You can build a bypass with a second relais with auxiliary contacts to prevent bad things.
Thanks for the testing. I am very keen on building my system with a single 8kVA this way. I was having the same questions you asked on June 24th, And I still have…
After reading the certificates again, is it this what it boils down to? MPII 8k, 10k and 15k single phase system: you do not need a Ziehl. When three phase: Ziehl required. If so, can the manual be updated so that is written more clearly? And while updating those manuals, please consider to add a hearty advise for double separation in grid supplied systems.
You need the CTs connected as they cannot be replaced with a external Meter! The CT’s are for the Multiplus to natively check for any current flow before the transfer relay is triggered, because this will prevent it from energizing the AC OUT as a precautionary measure.
In the Cerbo menu you can choose internal or external for the measurements displayed.
With a single 8kVA multiplus i think you can implementing the S99 fw, give it a try without the Ziehl at first, follow the manual without the Ziehl.
SMA Tripower x 15 does communicatie over modbus normally and in my case the battery capacity is 90kWh and from next week on 150kWh, at the moment it does not need to be ramped down now the summer is ending and i cycle them every day.
Thanks a bunch!
Little by little everything gets a bit more clear. From the specs i read about your setup, I think a “show your system” topic is in order….
I hope one of the Victron Experts (@nickdb for instance) is able to pass my remarks from my previous post in this topic to someone who will be able to review them.
Well, my system is temporarily built in a position where it won’t stay. And since it’s temporary, I didn’t put much effort into wiring it up nicely. It’s a quick and dirty solution to gain experience and do it right at the right spot later on.
I have tested some random disconnects, and the system works flawlessly. When the system is injecting power into a phase and that specific phase is disconnected from the grid by a 25A fuse, it takes a second or two before the contactor disengages. However, if a phase is switched off from which the system is drawing power, it triggers the disconnect almost immediately.
Today I noticed that the MPPT did not respect the 0W grid feed-in. The SMA did, but somehow the MPPT kept delivering between 300 and 400 Watts of power. The 0W grid feed-in was working fine before I changed to the S99 firmware. This is something to investigate in the settings or I’ll have to go back to 3.65 from 3.70~28.
I’m considering expanding the system to 6x 8kVA, but I first want to understand the efficiency of these units. I need more charging power for the additional batteries, and in a few years, I’ll require the extra output capacity to power all the households in the building. I’ve considered selling the three 8kVA units and replacing them with three 15kVA units. However, if there’s no market for ‘used 8kVA’ units, expanding with three more of the same units would likely be a more practical option.
One more series of questiond: have you implemented a bypass for when the Victron installation needs maintenance? Can you share a schematic?
Like I said, I am very interested in this way of building the ESS. I figured it would make it more easy to make a bypass but after some thinking that appears not to be the case. When the Victron is running there’s no problem. But when coming back from grid only, when will the MP-II’s synchronize?
Have you tried to switch the MP-II’s off with the mode switch? Do the grid relays stay closed? What if you disconnect the AC-out from the loads and grid relay?
I tried to explain it as clear as possible, i hope it is a bit understandable.
I haven’t implemented a permanent bypass yet. For now, I have a manual bypass using local wiring that I connect to feed the grid relay when needed. However, after servicing the Multis, I have to power off the entire house to let them sync and switch the grid relay. I haven’t had the time to create a proper bypass schematic yet but have some thoughts about how to realize this.
For Syncing there are 4 wires 3p1N from the grid to AC-IN to sense AC and start syncing, as soon these wires are powered up the Victrons start syncing and switch AUX. So if you switch the 6A MCB off the multi’s do not sense any AC that triggers syncing.
If you switch the mode to “Off,” the grid relay will release. If you disconnect the house from the AC-OUT while the Victrons are active, you’ll get an overvoltage error and will need a VE.Bus restart to get everything back online. The reason for this is that the grid meter continues to measure the loads, and the Victrons try to handle that load. If the AC-OUT is disconnected, the voltage rises too quickly, resulting in an overvoltage error.
My thoughts about a bypass circuit:
It could be implemented with a second relay in place, a grid relay (call it A) to isolate grid from the house and one relay (call it B) just after the multi’s to isolate AC-OUT from the house.
You need a switching circuit to manage the AUX signal changeover from relay A to relay B.
In normal operation the relays are used as following:
Let’s say the Multi’s are in normal operation and the Service Bypass is switched on, now the AUX changeover from relay A to Relay B should be uninterrupted:
And of course with bypass switched to off:
I’vI recently installed another 3x 8kVA units in parallel, and I’m very happy with the power. I finally have enough charge current to handle both grid and PV loads at the same time. The parallel setup was also why I was so keen on the RTS implementation. I really should have just bought the 15kVA units from the start. For now, the installation is a bit quieter than before.
Thanks. I’ll go simmer on that for a while. If I get the golden idea, I’ll come back here. I also noticed your nick on Tweakers. I will share it there too. It might help someone.
Indeed used the same name.
