MultiPlus-II 4k5 + one Dyness DL5.0C Pro shuts down on grid loss with Low Battery and High DC Ripple

DIY
, ,
1

Hello everyone,

I need help diagnosing a grid-loss issue with the following system:

  • MultiPlus-II 4k5
  • 1 × Dyness DL5.0C Pro, 51.2 V / 100 Ah
  • SmartSolar MPPT 150/45
  • ESS installation

During some scheduled grid outages in Ukraine, the MultiPlus-II goes into Off for around 10–30 seconds, causing the lights and other loads to lose power briefly. The unit then restarts and enters Inverting mode.

image
image927×405 25.8 KB

VRM reports Low Battery and High DC Ripple, although the battery voltage remains around 52.7–53.3 V, the state of charge is normal, and the load is only a few hundred watts.

image
image1387×408 37 KB

The problem appears only during some scheduled outages and has not been reproduced by manually opening the AC input breaker. Sometimes the voltage relay ZUBR turns the grid off due to high input voltage (>260) without issues for MP2, which instantly goes to inverting mode.

My settings are: grid code other, ups function on, dvcc activated, LOM B.

The difference between MP2 battery measurement and BMS is ~0.10 - 0.15v.

DC Ripple in reality is from 0.00 to 0.2v, usually ~0.1v.

image
image1392×405 16.7 KB

I initially installed the system in March with a MultiPlus-II 3000 VA GX, and I did not experience any problems with it. However, there were no scheduled power cuts during that period, so this comparison may not be conclusive.

The battery was connected using the cables supplied with the Dyness battery. Around September, I replaced the 3000 VA inverter with the current MultiPlus-II 4.5 kVA.

Could anyone help diagnose what may be causing the MultiPlus-II to enter Off during grid loss and report Low Battery and High DC Ripple, despite normal battery voltage and a relatively low load?

I would appreciate any suggestions regarding further diagnostic steps or possible solutions. I can provide detailed VRM data, VE.Bus logs, screenshots, the VEConfigure file, and Dyness BMS parameters if required.

2

This used to be a common report.
The simple answer is that a grid brown out, which is what you are experiencing as opposed to a blackout, like when you drop the breaker, causes a large transient demand on the battery which triggers BMS protection and the system restarting.

The reality is that battery has a recommended discharge of 50A (~2500W before losses, check the DCL it is reporting) and is too small for your system and coping with a brown out.
You also upgraded from a 2400W inverter to 4000W one (the new unit ratings are Watts not VA) without adding more battery.
The solution has always been to add more battery.

This is a quirk of how low-frequency inverters react to sudden changes in demand.

3

Hi and thank you for this answer.

Currently, DCL is 100A, with such a small AC load of ~200-300w… Actually, there is no change in demand, because setpoint is 15w, so all the AC load is going from the battery at the moment of grid loss…

image
image1372×393 35.9 KB

4

The load isn’t a factor in these scenarios.
You also cannot rely on vrm stats, the resolution is too low.
Battery specs state a max of 100A, sounds like a textbook battery issue.
No shortage of topics about this.
Low battery and ripple are the perfect markers for this.

Get dyness to check the battery logs.

5

OK, thanks for this explanation!

I will also try to limit the max Inverter power to 2500W and see what happens.

For the time being, I would like to keep the Dyness 5.0c pro, but will consider changing it to 2x Felicity FLA48100UG1.

Another solution could be Node-red - to switch between ‘Inverter only’ and ‘On’ mode before and after the scheduled grid loss.

6

Hi @Oleksiy ,

Take a look at this video: https://youtu.be/8NFxRJc7f0Q?si=JVUIBz_gnmtiZZkA

Good explanation of your issue.

When the grid is down, MPII 4k5 tries to get max energy from your battery for maintenance 230V on AC Out1, but the BMS can not produce it.

MPII 4k5 peak power 7000W => 7000W / 51.2V = 136.7A

image
image1443×665 55.3 KB

7

Yes, makes sense.

At the same time, my friend is using the same 4k5 model and 1x FLA48100UG1, he has not experienced any shutdowns during scheduled grid outages..

8

Brown-outs can vary by area, depends on the area, density, grid quality etc. We had many here, but across the road on a different transformer, they did not.
Different batteries have different BMS’s, they handle the same conditions differently, some use 15 cells others 16 and so forth.
There can also be subtle differences in the installation as well that can make a difference.

9

I had a similar problem:

  • My system:
    • Victron MultiPlus-II 48/6k5/100-50, VE.Bus firmware v560
    • Cerbo GX MK2, Venus OS v3.81 (during the latest incident)
    • 2 × Dyness PowerBox G2, 10.24 kWh each (20.48 kWh total), CAN-BMS communication
    • SmartSolar MPPT 250/100 VE.Can
    • ESS configuration

Here are two similar incidents:

  • October 1, approximately 05:01:
    • Low battery: Alarm (05:01:02) – 48s
    • High DC Ripple: Warning (05:01:02) – 48s
    • Grid alarm: Grid lost (05:01:31) – 1m25s
  • October 11, 22:39:09:
    • Low battery: Alarm (22:39:09) – 51s
    • High DC Ripple: Warning (22:39:09) – 51s
    • Grid alarm: Grid lost (22:39:42) – 1m6s

During both incidents, the AC output stopped and my house lost power for less 1 minute ((

10

Hello @VVKo ,

Do you use a DC bus bar (Lynx Power In) to connect batteries in parallel as recommended in the Dyness user manual?

image
image931×645 86.4 KB

Did you set the correct settings in VE.Config?

image
image492×415 41.6 KB

Dyness documentation - https://www.dyness.com/Public/Uploads/uploadfile/files/20241023/PowerBoxG2UserManualENV1.pdf

11

Hi Vladyslav

No, I don’t use a Lynx or DC bus bar. My two PowerBox G2 are connected directly toeach other. Both DC cables going to the Multiplus are connected to the same batery.

I haven’t checked those exact settings in VE.Configure yet. My system uses ESS and CAN-BMS communication.

The shutdowns are not a regular problem. For now everything working normally

Thanks for your advice. I will check the DC connections and settings.

12

The standard Dyness DC cable is 35 mm², which is undersized and causes excessive voltage drop. For the MPII 48/6k5, 70–95 mm² cable is required.

image
image1405×413 47.6 KB

You can use the 35 mm² Dyness DC cables to connect each battery to the Lynx Power In, and use 70–95 mm² cables (depending on the distance) to connect the MPII 48/6k5 to the Lynx Power In.