loxley avatar image
loxley asked

Strange oscillation from 2 parallel Multiplus-II 48/5000 when negative setpoint in ESS mode

Hi all,

I've recently added a second Multiplus-II 48/5000 unit in parallel to one that's successfully been running for nearly 2 years, in ESS mode.

I've read through the guide (, ensuring the wiring for each unit was the same length. They both have the same firmware (updated to latest version, 494), and have successfully been configured in parallel mode by VE Quick Configure. They also both have the same settings (through VEConfig), including the same ESS assistant.

ESS mode is set to "external control", like it was for the 2 years running on the single unit.

The system works well to charge (can get the full charge current), as well as power loads. The slave unit is showing as slave (correct LED codes).

However, the problem is when I try and set the grid setpoint to negative: after a while (timing of how long it takes is random, but a few minutes is the best I've had) I get a strange oscillating behaviour. Within about 0.8 seconds, the power draw from the battery can spike to a few kW, both positive and negative (and lights flicker accordingly on AC out). The transformers within both Multiplus units also emit a very noticeable crescendoing buzz. This keeps cycling, until the minute timeout of sending a grid setpoint.

Even applying a grid set point set to a value within what the single Multiplus unit was able to achieve before exhibits the behaviour.

Any idea what could be causing this issue?

Multiplus-IIbattery chargingESSmultiplus in parallel
2 |3000

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

3 Answers
Alexandra avatar image
Alexandra answered ·


What are you trying to achieve with a negative grid set point? For self consumption it should be set to 0W

It would be better to set the feedback from ac or dc to a limited amount. If that is what you are wanting to do.

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.

loxley avatar image loxley commented ·

I'm deliberately wanting to feed back into the grid (hence the negative setpoint, which is dynamic). I've had this working fine on a single Multiplus for about 2 years.

The issue I'm having is the erratic/unstable behaviour of the (newly parallel) system, when a dynamic negative grid setpoint is set.

0 Likes 0 ·
loxley avatar image
loxley answered ·

I just did a check on hardware revision numbers.

On the old Multiplus, it's rev 05.

On the new Multiplus, it's rev 07.

Is this likely to cause an issue?

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.

nickdb avatar image nickdb ♦♦ commented ·
Units should be identical.
0 Likes 0 ·
tnt369 avatar image
tnt369 answered ·

What grid code do you use? Maybe try to deaktivate the LOM function temporary for testing (grid password is necessary for that).

The LOM has caused a similar problem at my system (with one quattro fine, after adding a second quattro the system sporadically oscillated).

2 |3000

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

loxley avatar image loxley commented ·
Grid code is set to " New Zealand: AS/NZS 4777.2.2015 (AC Neutral Path externally joined)"

What were the symptoms you observed when you added your second Quattro in parallel? Did you also have ESS set to external control (mode 3), with a negative grid setpoint (to export) ?

0 Likes 0 ·
tnt369 avatar image tnt369 loxley commented ·

The LOM function of my two Quattros apparently interfered with each other. Perhaps my grid is also too weak. As soon as there was a somewhat higher load on the grid, the Quattros became sporadically and temporarily very loud. A lot of power was drawn from the mains or fed into the mains.

I have ESS activated (no external control) with a positive value of approx. 30-100W. I do not want to feed in power.

After I deactivated the LOM function, the system worked without any problems.

What I also do is read the SoC and write the grid setpoint depending on it (with a control via Modbus).The range for the grid setpoint is +30...+3500W.

1 Like 1 ·
loxley avatar image loxley tnt369 commented ·
Interesting. I don't seem to have a problem, until I set the grid setpoint to a negative value. A full charge (max current per multiplus) doesn't exhibit the issue. I'm not sure if removing LOM will make a difference, but if it does, it'll only be a temporary thing (LOM is a requirement here to be grid compliant)
0 Likes 0 ·
jtynan avatar image jtynan loxley commented ·
Hi, I am seeing the same behavior. Did you manage to get a fix for this? I spent way too much time trying to figure out what causes the issue. I have noticed if you set a lower negtive 700 ish watts, I think the two parallel devices act as against each other. it then trips and pulls about 2k from the grid and restarts the inverters.

Do you have a GX deployed also and does it also reboot?

Any feedback would be great.....



0 Likes 0 ·