question

kjell avatar image
kjell asked

3 ph Multiplus-ii totally dead after over voltage

So....

Today my 2-signal bms threw the contactor while there was high pv-production(~8kW) and moderate windproduction(~3kW). There was probarbly an overvoltage beacause now all three multipluses are dead and so is the wifi-AP connected to 48V.

Is there any decent protection against these situations?

So far when the battery has been cut I have had 1 cerbo, two network switches, one AP and three multipluses have failed...

The wind turbine is a Bornay 5kW with an MPPT

5x Smart solar 150/35

3x Multiplus-II 48/35/3000

1120Ah LiFePO4 48V

Cerbo/RaspberryPi

Batrium BMS

I have moved the GX, BMS and all the extras to an isolated 24V system that isnt connected to the Lithium-bank so that nothing more will break, but any ideas would be helpful.

Can a multiplus be repaired after over voltage?

Is there an obvious way to protect the setup?


Thankful for all help!

Multiplus-IIovervoltage
2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

Alexandra avatar image Alexandra ♦ commented ·

On an overvoltage from where? AC or DC?

If it is AC then SPDs are useful.

Have a feeling something else is not quite right though, have seen quite a few systems loose battery reference and been ok in those circumstances.

If all components are programmed then usually the system shuts itself down and throws out errors. Cerbo being dead as well sounds like a neutral issue though? Shot in the dark there. Working out what kills AC and DC components is usually neutral related.

1 Like 1 ·
Al avatar image Al commented ·

That sounds bad! Ouch.

The most likely issue is your wind turbine, or how it is/isn't regulated well with its dump load, everything else can be controlled nicely with DVCC.

1 Like 1 ·
2 Answers
kjell avatar image
kjell answered ·

Its on DC so no neutral issue here, everything that has died has been on the 48VDC

11 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

phoenix avatar image phoenix commented ·
Any indication from the BMS or the MP what voltage it was, although TBH I'm curious. Inside my 24V 3000 there is a tiny fuse, orange/brown about 9mm diameter and 9mm high on the main PCB at the bottom near the heatsink. It cant be tested without removing the PCB. It's a long shot as it would only blow if Victron put crowbar protection in, which you rarely see.

How are all the mppt devices connected to the battery - busbar, for example? I've certainly heard of wind turbines going over voltage when SOC is high, and you get a gust on a diversion controller. Check the dump resistor hasn't got loose connections or dodgy crimps.

Whilst you could put in an independent contactor and voltage trip, contactors are slow in the scheme of things. Some are much faster than others.

1 Like 1 ·
kjell avatar image kjell phoenix commented ·
I found the F4 fuse that you are referring to and it measures good continuity.

The Pv-MPPT:s, Wind MPPT, multipluses and a small fusebox for Cerbo, networkswitch, AP are all on a busbar, then > shunt > contactor > LiFePO4-bank

The wind MPPT is behaving just as it should as long as the batteries are there, I had it configured wrong for a little while(24V) and it used up all energy on the dumploads so it would have shown then.

I am thinking maybe a node-red implementation where the wind is diverted into the dumploads after about 90% SoC, then everything should be protected from OV as long as the contactor doesnt release in the 20-90% interval.


0 Likes 0 ·
phoenix avatar image phoenix kjell commented ·

If the PCB is out the only other thing to do is trace the track from the fuse and that will most likely go to a buck converter. They tend to crack or blow a piece off when over-stressed. Change that and the associated capacitors. If it has an external mosfet, change that too. It'd bet it will power up ok once repaired.

I have a LiFePO4 battery and a home-made BMS. I put all my local loads eg GX direct on the battery, so the main BMS contactor only disconnects the MP. This keeps services available so I can fault find remotely if need be. The only risk is that I could over-discharge the battery with these loads.

How about using node red to pre-detect over-voltage on the battery and disconnect the turbine (or all generation) with another contactor. Might be wide to use a separate contactor for the turbine so you don't stress the solar.

1 Like 1 ·
Alexandra avatar image Alexandra ♦ phoenix commented ·
Small fyi there is also DC spd.

But you would definitely need to add some production shutdowns for safety on the wind turbine.

0 Likes 0 ·
Show more comments
kjell avatar image kjell phoenix commented ·
I traced it and it goes onto the 52-pin soldered controllboard. If I could buy a fresh one I would absolutely change it in full, but desoldering it to try to fix feels dodgy at best.
0 Likes 0 ·
Show more comments
pwfarnell avatar image
pwfarnell answered ·

Something you could look into are the alternator protection modules offered by the likes of Balmar which are designed to limit over voltage spikes from an alternator when a BMS disconnects. They claim up to 200A and have a 48V unit.

https://balmar.net/alternator-protection-modules/

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

Related Resources

MultiPlus-II Product page 

MultiPlus-II Manual

Multiplus-II GX Online Manual

MultiPlus-II 230V Datasheet 

VE.Bus Error codes

Additional resources still need to be added for this topic