High Volltage Alarm Pylontech

Been seeing a few threads about this lately and figured I'd chip in with my own situation.

Running a Pylontech US3000C stack (3 units) in my garden office setup with a Victron Multiplus II. Got an east-west split array — about 4.2kW total across both orientations — which I thought would help smooth out the charging curve compared to a pure south-facing setup.

Started getting high voltage alarms on the Pylontech a couple of weeks back, mostly mid-morning when the east panels are pushing hard before the system's had chance to discharge much overnight load. The Victron DVCC settings should be respecting the Pylontech's BMS limits but something's clearly not playing nicely.

Few things I'm wondering:

• Is this a comms issue between the BMS and the Multiplus, or is the BMS just being overly cautious?
• Anyone found that the Pylontech default absorption limits are just too conservative for real-world UK use?
• Could the east-west split actually worsen this in certain conditions — like after a mild night where the batteries haven't drained much?

I've had a poke around in VictronConnect and the ESS settings but not found an obvious culprit yet. Fogstar Drift cells seem far more configurable in comparison, makes me wonder if I made the right battery choice for this setup.

Curious whether others with similar east-west arrays and Pylontechs have cracked this — or whether it's just a case of tweaking the DVCC charge current limits manually.

@WattHamish looks like your post got cut off — what were you actually seeing on the alarm side? Curious whether it's a genuine high voltage condition or a comms issue between the Pylontechs and the Multiplus II.

I've got a similar setup for my garden office (single US3000C for now) and had a scare early on that turned out to be the DVCC settings in VenusOS not being configured correctly — the Multiplus was essentially ignoring the BMS charge parameters.

A few things worth checking when you post the rest:

• What firmware version are the Pylontechs on?
• Are you using the Pylontech cable to the Cerbo/CCGX or a DIY RS485 lead?
• What's your charge voltage set to in VEConfigure?

Would be interested to see your screenshots if you have any.

@WattHamish and @AshSeeker — both posts got cut off, brilliant pair of you 😄

Worth mentioning while we wait for the details: with a US3000C stack, the high voltage alarm almost always traces back to one of three things:

1. Incorrect charge voltage in the Multiplus II DVCC settings — should be sitting at 53.2V for Pylontech, not whatever Victron defaults to
2. CAN bus comms dropout — if the BMS stops talking to the Cerbo/CCGX, the Multiplus reverts to its own charge profile which can be too aggressive
3. Cell imbalance in one unit dragging the whole stack's reported SOC out of whack

On my own setup I had identical alarms until I realised the RJ45 cable between units wasn't fully clicked in. Embarrassingly simple fix after three days of head-scratching.

What's your DVCC screen showing in VRM?

@WattHamish while we wait for your full post to appear — common culprits with Pylontech high voltage alarms on a Multiplus II stack are the charge voltage set too high in VE.Configure, or the DVCC settings not properly deferring to the BMS.

Pylontech US3000C wants absorption/charge voltage no higher than 53.2V for a 48V system. Some people accidentally leave it at a default Victron figure which is higher.

A few questions that'll help diagnose:

• What's your charge voltage set to in VE.Configure?
• Are you running DVCC with BMS control enabled?
• How's the stack connected — CAN bus or UART to the Cerbo/CCGX?

Also worth checking whether the alarm is coming from the Pylontech console LED or the Victron side — they're not always the same event.

@ExFirefighter curious what culprits you're thinking of — my post also got eaten it seems 😄

I've got a similar setup to @WattHamish, three US3000C units with a Multiplus II feeding my garden office. Been getting intermittent high voltage alarms, particularly when solar input is strong and the office load drops suddenly (end of day, everything switching off).

Wondering if it's the absorption voltage set slightly too high in VEConfigure, or whether the DVCC settings in Venus OS need tweaking. My charge current limit from the Pylontech BMS should be overriding things but I'm not confident it's actually being respected.

Anyone know whether the CAN bus comms between Pylontech and Cerbo GX are reliable enough to trust, or is it worth adding a manual charge voltage ceiling as a belt-and-braces measure?