MosFet leakage from Battery Protect activates Orion-Tr Isolated DC-DC converter remote switch

Bit of a niche one this, but I've run into something similar on my narrowboat setup and I know a few van builders here have too.

Running a Victron Battery Protect as a virtual load output from a SmartSolar MPPT — great in theory for unattended systems. The MPPT controls the BP via VE.Direct, BP sits between the battery and the load bus, job done. Except...

The MOSFET leakage from the BP when it's supposedly "off" is enough to ghost-trigger the remote L pin on an Orion-Tr Isolated DC-DC converter. We're talking maybe 1-2V of leakage, which shouldn't be enough, but apparently is on some units. My Orion kept waking up intermittently overnight when the BP was commanded off. Took me ages to figure out what was happening.

My workaround was adding a small bleed resistor to pull the remote line properly to ground when the BP cuts out. Crude but it works. Seen others suggest an intermediate relay instead, which is cleaner but adds complexity.

Few questions for the collective:

• Anyone else seen this with newer BP firmware? Wondering if Victron have tightened it up
• Does the non-isolated Orion behave differently here? I'd expect the remote input threshold to be similar but not sure
• For those running unattended installs (holiday let, remote cabin, that sort of thing) — do you just accept the relay approach as best practice?

Would be especially curious to hear from anyone running Fogstar lithium banks with this config — wondering if the BMS cutoff interacts with this leakage issue at all.

Classic Victron "it's not a bug, it's a feature" moment — my Orion-Tr did exactly this until I wired a proper physical disconnect rather than relying on the Battery Protect's leaky MOSFET to behave itself. 🙄

The fix that sorted mine: add a small bleed resistor (10kΩ works) across the Orion remote input to drain that parasitic voltage to ground — keeps the threshold below the ~2.9V activation point.

Fogstar cells staying healthy is the whole point of the Battery Protect, so having the Orion sneak around it because of MOSFET leakage rather defeats the purpose, doesn't it. 😤

@PanelGraham narrowboat installs are notorious for this — damp environment probably not helping your leakage figures either, pal.

@RetiredNurse49 has already covered the workaround but the root cause is worth nailing down properly.

The Battery Protect's load-side output sits at roughly 1-3V even when "off" due to MOSFET body diode leakage. The Orion-Tr Isolated's remote H pin only needs ~3.5V to trigger — so you're right on the threshold depending on temperature and unit variation.

The clean fix is a 10kΩ pull-down resistor between the remote H pin and ground. Costs pennies, takes 10 minutes. Bleeds off the leakage current without affecting normal operation when the Battery Protect actually closes properly.

Worth noting this catches people out specifically with isolated DC-DC converters because the floating secondary means there's no incidental path to ground that would otherwise swamp the leakage. Had this exact scenario on my shepherd's hut EV charging setup using a BatteryProtect 65 feeding an Orion-Tr 24/12.

The MOSFET leakage figure Victron quote in their BP documentation is up to 1mA at the output when nominally "off" — enough to pull the Orion-Tr's remote L-wire high if it's floating or weakly pulled down internally.

I've seen this on my narrowboat build where a BP-65 was sitting upstream of an Orion-Tr Smart Isolated 12|12-30. The fix that actually worked long-term was adding a 10kΩ pull-down resistor between the remote H/L wire and ground, rather than relying on the Orion's internal impedance alone.

Worth checking your specific Orion-Tr variant too — the isolated models behave differently to the non-isolated regarding remote input sensitivity. @SolarNotSure is right that understanding the root cause matters, because @RetiredNurse49's physical disconnect approach, whilst valid, doesn't scale well once you start adding automation logic downstream.

Interesting one this — I had something similar crop up on my static caravan setup last summer but never traced it back to this specifically. So just to confirm my understanding: even with the Battery Protect showing as "off" on the VictronConnect app, that leakage is still present at the output terminals?

Is this a consistent behaviour across all BP models, or more prevalent on certain ones? I've got the BP-65 feeding an Orion-Tr Smart 12/12-18 and I'm now wondering if what I thought was a firmware glitch was actually this all along.

Would a simple pull-down resistor on the remote pin be the cleanest fix, or is @RetiredNurse49's physical disconnect approach the more robust solution long-term?