rrroonn avatar image

High grid voltage disconnect with Multiplus II in AU, why?

Lets be clear, I understand that power generators should not feed in energy if the grid voltage is already high, hence the regulations around handling high voltages. Hence the locked-in grid parameters.

I have recently has "grid lost" for short periods in the middle of the day when high voltages were sustained. Circuits upstream of the inverter were not affected by a power loss - that made it was clear that the MP2 disconnected itself from the grid. The MP2 went into "Inverting" mode.

The question is "why disconnect from the grid?". Why does the MP2 not simply go into "pass-thru" mode or a "zero-export" mode? To me this seems much more sensible. It achieves the desired outcome of removing the generator from the grid while not drawing down on batteries needlessly.

Is there any way to change this behavior in response to high (10 min moving average) voltages? (yes, ideally, fix the voltage issues but ...)

thank you,


2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
Guy Stewart (Victron Community Manager) avatar image
Guy Stewart (Victron Community Manager) answered ·

Hi @rrroonn,

This behaviour is a requirement of the Australian Standard AS4777.2:2015 - Grid connection of energy systems via inverters" - Section 7.1 (b) -

Automatic disconnection device (ADS) shall operate when grid goes outside preset parameters (e.g. overvoltage).

Section 7.2 goes on to specify the ADS will isolate all live conductors. And that this must be a physical separation, with even a semicondutive isolation being insufficient.

Those are the rules. There is no way to change this behaviour and still have the unit compliant with the Australian grid codes.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @Guy Stewart (Victron Community Manager)

I believe you confirmed the MultiPlus behaviour to disconnect from the grid in times of high voltage but you have not addressed the issues that are the reason this occurred.

I too have this issue of high grid voltages during peak daylight hours where it is not possible to export 5kW to the grid without exceeding high voltage limits.
To me, the problem lies with throttling back PV sources.....and I question whether it is the Grid code, V1-V4 values for Volt-Watt response curve that trigger the throttling?


My settings (pictured) do not match those that my distributor (Powercor) recommend, where they say V3=253 and V4=259.

In my case, I have a DC-coupled MPPT charger which does throttle back, albeit a little late....

Also I have an AC-coupled Fronius inverter which I was told before commissioning that it would be controlled via frequency shifting. This I believe cannot happen whilst connected to the grid and instead can only happen through a Modbus TCP connection. Throttling will work using Modbus, but again too late!

--> What settings will allow throttling to occur sooner?

Furthermore, when using Modbus the GX device shows the Fronius connection as (sunspec) and the Fronius status as "Running(Throttled),
versus not using Modbus it shows the Fronius connection as (solarapi) and the status as "Running".

When using Node-RED to retrieve data, sunspec & solarapi behave differently - with solarapi giving better information for Status and Error codes.

--> Can we use Modbus for throttling, and also use the solarapi connection?

1635941129786.png (69.4 KiB)
rrroonn avatar image
rrroonn answered ·

Hi @Guy Stewart (Victron Community Manager),

Thank you for your thorough answer and I do understand the reasoning behind the current behavior. However, do you think there is scope in the interpretation of the rules to improve the customer experience?

The ADS (or ATS) can disconnect the 'device' (where 'device' is the inverter) by bypassing it through the action of the ADS? IE the ADS is not the 'device' or part of the 'device'. Indeed, the wording assumes that the ADS is separate from the 'device'. I guess it would come down to the testing agency?

Maybe something to consider for the next iteration of approvals.

I know that lots of country areas in AU are now suffering from high peak-sun line voltages and the current behavior is certainly not optimal. In fact, recent news in AU is that grid operators are planning to deliberately push up line voltages to switch off inverters when grid nett energy gets too low - IE supply outstripping demand. This is due to the massive uptake of solar in AU and the ready supply of clear, sunny days.

Most installations are AC-coupled solar and switching the inverter off just means that power from the sun is not available and that grid power will then supply 100% of household consumption. Users of your (Victron) grid-connected inverters are in a different position as consumption occurs downstream of the inverters (rather than in parallel for an AC-coupled solar inverter). Grid operators pushing up the voltages, or just routine high grid voltages, will disconnect these customers from the grid. Customers have a right to be unhappy about this:

  1. They may not have very big batteries and risk their lights going out.
  2. Their inverter capacity may not support starting big loads that they rely on - such as water pumps etc.

From my own personal experience, the resellers do not identify these risks to customers. Customers buy a system that they think is reliable but it turns out to be the opposite - not because the Victron inverter itself is unreliable, but because of the architecture involved and the implementation of the ATS function.

As a minimum, I suggest that Victron recommend to all AU distributors and installers that a manual transfer switch be mandated as part of the installation. My installation did not have this and I suffered a complete outage when sparks emerged from the inverter. Neither the supplier/designer nor the installer even recommended or offered this. Having a manual transfer switch would also help customers where elevated grid voltages disconnects them from the grid.

Slightly 'off-topic': I also found out the hard way, that losing the grid (due to high voltage) with a full battery and max ac-coupled solar output is a bad thing. Again, something that the supplier (designer) and installer did not even consider. I have implemented a work-around using my home automation system, but this is also something that system designers and installers need to be educated on. And, better handling for common AC-coupled solar inverters is so easily added to your GX range. In the case of my SMA invertor, all it takes is a few lines of code and connection to the SMA via its modbus interface. It makes the difference between lights staying on - or going off until the BMS is reset.

My thoughts of the issue.

thanks again,


2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.