question

I would conservatively assume that with caps at zero voltage that they're literally dead shorts (likely to be a bank of low ESR electrolytics). The inrush current is then limited by the battery internal resistance and the cable loop resistance. Should be able to get a ball park estimate from your system design.

Look into sequencing a pre-charge if the above numbers are not to your liking.

I ran into the same issue with the quiescent draw in my design. I now use the highly technical mitigation of leaving the switch in the off position when not in use.

Multiplus 2 has lower quiescent but its only 48v at present.

Any contactor will die if you don't use a pre-charge device.

I'm using a manual switch in my test-setup with a 25ohm resistor to charge the inverters before turning on the breaker, another trick is to supply the inverter with AC so it starts as a power supply and charges the capacitors for you.

I don't know the current draw during different 'off' states.

This is a question, not an answer. If the Multiplus turned off, either by the rocker switch on the Multiplus, a Venus device, or a wired remote On-Off-Charger Only switch. And then you turn off the power, through a contactor, to the Multiplus.

When you then, through the contactor, turn on the power to the Multiplus, do the capacitors charge, or do they wait until the Multiplus is turned on?

I don't know, but I would like to.

Pat

Of all the Multi's and Quattro's I have connected to a DC Supply, the initial DC inrush (Big Spark and Crack) happens as soon as the unit is connected to a DC supply, i.e. inserting the fuse.

The units have always been switched off on the front panel power switch before connecting to DC.

I assume the capacitors are being charged before the unit is turned on. (I Agree with Boekel on the pre-charge device)

We use a 5watt resistor to precharge the variable speed low voltage DC winch controllers, this is permanently connected, (don't permanently connect a precharge resistor to an inverter).

Hope this is of some help.

Darren

Maybe a battery cable loop through a ferrite ring might reduce the inrush current?

We did that on a much smaller scale. We had a 24/12V 5A Victron Orion DC/DC converter connected to the output of a 10A solar charge regulator, but it would not start because the inrush current of the Orion caused the charge regulator output to turn off for overcurrent. Then we added a loop through a ferrite ring in the cable between the charge controller output and the Orion, and it worked fine.

As the working DC current from the battery to the inverter is not continous/fix, but rather oscillating with 120Hz (100Hz in 50Hz output inverters), I don´t know if that might cause additional loss or heating up of the ferrite. Would have to be thoroughly tested under load.

I have performed a very quick and dirty experiment to measure the instantaneous inrush current. I don't have much time to do a proper one. I am not expecting a very high current here, but it will give an idea at least.
1x VE.9bit Phoenix MultiPlus 12/2000/80
1x 12 12Ah SLA battery (a tiny battery).
1x 500A / 50mV shunt (the one that you get with a BMV).
Spindly thin wires (6mm2 aka 10AWG AFAIK).

The MultiPlus switch os in the OFF position, I am simply slamming the connector onto the battery terminal with the oscilloscope connected across the shunt.

So with 200mV per division there is 500mV zero to peak. With 50mV / 500A there is 5000A of inrush current being measured which goes on for roughly 50uS before the oscillations stop.

5000A is quite reasonable I suppose for a little setup, but with a proper setup, a good size battery bank, proper cables and a faster switching time I'm sure much higher inrush currents can be achieved.

Thanks everyone for the input, greatly appreciated and very helpful. I have some good input now to help me investigate my setup further and develop a workable solution.

Thanks again.