Probleme bei Cerbo GX Auswertung mit kombinierter AGM + LiFePO4 Installation

by VDH_Boats · 2 weeks ago 24 views 4 replies
VDH_Boats
VDH_Boats
Active Member
13 posts
thumb_up 3 likes
Joined Nov 2024
2 weeks ago
#19013

Mixing battery chemistries on a Cerbo GX is something I stumbled into almost by accident on the boat last year, and honestly it's been a rabbit hole ever since.

I was running a pair of Victron AGM Super Cycles as my "buffer" bank alongside a LiFePO4 bank — different situation to what's being described here, but the monitoring headache is familiar. The Cerbo simply doesn't know how to present a sensible state of charge when it's seeing two fundamentally different charge curves. It'll latch onto whichever battery monitor input you've configured and just... ignore the nuance.

The core problem as I understand it:

  • AGM and LiFePO4 have completely different voltage-to-SoC relationships
  • A single SmartShunt can only be set to one battery type profile
  • The Cerbo's dashboard will show you a number that feels authoritative but is essentially a best guess

What I ended up doing was running separate SmartShunts on each bank and treating them as distinct systems within VRM. Messy on the dashboard, but at least honest.

Has anyone found a cleaner way to handle this? I've seen some folk use a DC-DC charger between the banks to essentially isolate them electrically whilst still letting the LiFePO4 top up the AGMs — curious whether that actually helps the monitoring side or just shifts the problem.

Would love to know how others are configuring their Cerbo settings for mixed installations. Feels like Victron's documentation assumes you've committed to one chemistry and never looked back.

Camper Clive
Camper Clive
Active Member
23 posts
thumb_up 11 likes
Joined Oct 2023
2 weeks ago
#19020

@VDH_Boats interesting one — I've been tempted to do something similar on my boat where I've got legacy AGMs I don't want to bin yet but want to add LiFePO4 capacity alongside them.

Quick question though — how are you handling the Cerbo's SOC calculation when it's seeing two completely different charge curves? I'd imagine the voltage-based estimation gets completely confused, particularly at the top and bottom of the AGM range where the curves diverge most dramatically.

Are you running separate battery monitors (SmartShunts on each bank) and feeding both into the Cerbo, or trying to let it manage everything from a single shunt?

I've read you can use the VE.Bus and VE.Direct inputs to distinguish banks but never actually tried it myself. Would love to know what your wiring topology looks like before I attempt anything similar!

ShortCircuit
ShortCircuit
Member
8 posts
Joined Mar 2025
2 weeks ago
#19042

@VDH_Boats the Cerbo really struggles here because it wants a single battery type for its SOC calculations. Mixing chemistries means the voltage-based SOC readings become meaningless — AGMs and LiFePO4 have totally different discharge curves.

What I've seen work is running the LiFePO4 as your primary bank with a proper shunt (BMV-712 or SmartShunt), then treating the AGMs as a separate "load" essentially. The Cerbo can then at least accurately track the lithium side.

@CamperClive the legacy AGM problem is real — I had similar with some old Trojans before I ripped them out. Sometimes the cleanest fix is just... not mixing at all 😅

Main issue with Cerbo displaying combined SOC on mixed banks is it'll just lie to you half the time. Two separate inputs if your install allows it.

Wez
Wez
Active Member
42 posts
thumb_up 30 likes
Joined Aug 2023
2 weeks ago
#19060

@ShortCircuit is right about the SOC issue but there's a practical workaround — use a SmartShunt on each bank separately, then feed only the LiFePO4 shunt to the Cerbo as the "main" battery monitor. Let the AGMs sit on their own circuit with a separate VSR or manual isolation.

I did something similar in my garden office setup, not boats, but same principle. The Cerbo then reports LiFePO4 SOC accurately and you just keep an eye on AGM voltage manually or via a second shunt in the VRM dashboard as an "extra sensor."

Not perfect but it works. The key thing is making sure your charge sources aren't fighting over voltage setpoints — AGMs want higher absorption voltages than LiFePO4 tolerates long-term. That's where most mixed installs quietly damage the lithiums over time.

Paul Jackson
Paul Jackson
Member
7 posts
Joined Apr 2025
2 weeks ago
#19077

Great thread, this. @Wez1961's SmartShunt approach is the right direction — I'd add that you can feed both shunts into the Cerbo via VE.Direct and then use the 'has DC system' option in the settings to at least make sense of what's happening across the whole installation. The Cerbo will still only use one bank as the primary SOC source, so you'll want your LiFePO4 designated as the main battery instance given it'll have the more reliable SOC curve. Your AGM bank then shows up as a DC system load/source which isn't perfect but is workable.

Also worth checking your MultiPlus charge profile — you really don't want a lithium algorithm hammering the AGMs or vice versa. Separate charge sources for each bank is the cleaner long-term solution if you can manage it.

Log in to join the discussion.

Log In to Reply