MultiPlus II 3-phase system struggling to connect to Perkins genset above 15kVA load

Hi Victron community, I need some assistance troubleshooting a generator integration issue.

System setup:

  • 6 × Victron MultiPlus II 10kVA configured as a 3-phase system
  • Fronius 27kVA PV inverter connected on the output side
  • Cerbo GX (latest firmware)
  • All MultiPlus units running the latest firmware
  • Freedom Won 60/48 battery system
  • Perkins 80kVA genset recently integrated using Generator Assistant

Genset settings:

  • PV Support enabled
  • Genset current limit: 80A
  • Weak AC enabled
  • Dynamic current limiter enabled
  • Battery charge current limit: 65A per inverter
  • MultiPlus units synchronize with the genset correctly
  • Phase rotation has been checked and is correct

The system works normally with solar and battery. The problem starts when the genset is connected and the load increases above around 15kVA.

The MultiPlus units try to connect to the genset, synchronize, then disconnect again. This cycle repeats continuously. However, once the connection is finally established, the system can handle the full load and run up to around 48kVA without any issue.

It seems like the initial AC input acceptance/synchronization is the problem, not the ability to carry the load.

What settings should I check?

  • AC input current limit?
  • Weak AC settings?
  • Dynamic current limiter behavior?
  • Generator Assistant parameters?
  • Frequency/voltage tolerance?
  • Any recommendations for large 3-phase MultiPlus systems with an 80kVA generator?

Any advice or similar experience would be appreciated.

Thank you.

FLYING_ODYSSSEY_BASE_CAMP_c0619ab784e6_inst_ttyS4_Interfaces_Mk2_Tunnel (6).rvms (8.3 KB)

Gensets are always a trial and error one.
Had a wierd one a few years ago like this.

Can i assume then it is not ess but offgrid? So no grid code?

We took off dynamic power limit and weak ac.
Wide frequency on, no ups.

Recently with the generator warm up in the GX things got a bit better but the genset has aged as well.

We have added a node red flow that connects it at a lower input current and slowly ramps it up to what ever max it ‘feels like’ on the day. (Based on how the frequency and voltage behave as we load it more)

Its used so occasionally so the owners are not keen on buying a replacement of course. But it has stretched the creativity a bit.

The geneset controller may need to be adjusted, had many a unit that has needed a tune up on the genset side. A basic oscilloscope will show what its doing under load.
One of them only had an issue when the battery charging kicked is. Running loads was fine through the inverter. Go figure.

what solution did it work cause if i turn off dynamic current limiter makes things worse

Does the genset have an active frequency/speed regulation? Does it hold 50 (or 60) Hz actively while the load changes?

Since you also have an AC coupled PV, you might want to try to run the genset at a bit over 50Hz. Synchronizing to the genset requires the MP to lower its frequency down to what the genset produces. That could also mean that at the low frequency the AC coupled PV is increasing its output (due to the frequency-power-curve).

Try to see how the system as a whole behaves in this situation, how the power flows. Also worth a try with the AC PV disabled/switched off, just to see how it reacts

I also just now looked at the file. Im having trouble opening it, the assistant reconstruction fails. But theres is gridcode “Other” set (i hope the file is correct at least in that regard), which would imply that you are using ESS? You cant use ESS when offgrid. You cant feed energy back to the generator, so using ESS in this case is hindering the function, not helping.

Also, even gridcode “Other” sets narrower voltage and frequency windows, which again hinders the use of a generator.

Trial and error.
There is no formula only a combination that works with what you have.

If dynamic limiter had a poor effect leave it on.

One thing i do see since i have now had a chance to sit in the office with a laptop is you do need to take off LOM detection in one of your tests.

Sometimes weak ac messes with it as well. Especially when there is battery charging. What firmware are you using?

Change one thing at a time and see the effect.

am using latest firmware both on gx and mp’s , there is no any change in frequency or voltage change i can probe with and oscilloscop but i dont have it on site , grid code"not compliant to any grid standards , no ess am using vp support as assistance . genset avr has frequency and voltage adjust RPM set to 1580 frequency 51.2HZ .. bellow are attached images of genset with watchman (Hyena),AVR board,thermal regulator for heating lpg tanks (hot air ballons ). genset looks well build and stable i think am doing somthing wrong on the setup

If the cable is long to the set up other things come into play over the cable length.

LOM detection off may help.

Because you have ac coupled PV you need to leave these at their max to take advantage of charging with ACPV. So you are limiting them unnecessarily.

With the the generator on the input this setting doesn’t help as the system automatically does power control (runs load first then uses left over power to charge)