In March 2024, at 1:47 AM, a client called. Their 12kW hybrid system—which I'd installed 14 months earlier—had shut down. The backup power for their medical refrigeration unit was gone. They had four hours before the vaccines inside lost viability.
I've handled 63 rush calls in the past year, but that deadline made my hands sweat. The client was already on their second coffee and had convinced themselves the SRNE 12kW inverter was dead.
"The manual says it should handle this load," they said. "It's only 14 months old."
The inverter was fine. What I found in the next 90 minutes—digging through battery voltage logs, charge controller settings, and a midnight panic-search for LiFePO4 recall news—revealed a problem that no manual addresses directly.
The Problem Everyone Thinks They Have
When an off-grid system shuts down at night, the default assumption is hardware failure. The inverter, the battery, the charge controller—something broke. That's what the client thought. That's probably what you'd think too.
Their evidence was compelling. The system had been running flawlessly for 14 months. Then, suddenly, repeated shutdowns between 1 AM and 4 AM. The inverter logged a low-voltage fault. The client had even checked the battery voltage with a multimeter at 7 AM and found it at 13.1V—well within normal range.
"So it has to be the inverter," they said. "It's reading the voltage wrong."
I'd heard this exact story before. The inverter wasn't reading wrong. It was reading correctly—for the wrong battery chemistry profile.
What the Manual Doesn't Tell You
Here's the thing about the SRNE inverter manual: it's technically accurate. The default settings are conservative, designed to protect a lead-acid battery bank—which is still the most common battery chemistry in off-grid installs. The problem is that LiFePO4 doesn't behave like lead-acid, and the gap between the factory defaults and what a lithium pack actually needs is wider than most installers realize.
The manual's voltage table lists lead-acid profiles as defaults (note to self: that's a documentation gap SRNE should address in the next revision). For lead-acid, a low-voltage cut-off of 11.5V per battery—46V for a 48V system—is reasonable. For LiFePO4, it's not just conservative. It's actively problematic under load.
Here's what happens. At 2 AM, the refrigerator compressor kicks in. That's a 2.4kW load—and for a few seconds, the battery voltage drops. Not because the battery is empty, but because of voltage sag. The inverter sees that dip and, if the voltage crosses the 46V threshold even momentarily, it cuts off.
The battery isn't dead. It's probably at 60-70% SOC. Put another way: the inverter is cutting power to a battery that still has more than half its capacity left.
LiFePO4 has an exceptionally flat discharge curve. Between 20% and 80% state of charge, the voltage barely moves—maybe 0.1V per cell. That's efficient, but it creates a dangerous blind spot. The inverter can't distinguish between "battery at 30%" and "battery at 70% with a 2.4kW load." Both look like a voltage sag.
"Everything I'd read about LiFePO4 said the flat discharge curve was a huge advantage. In practice, that flat curve is exactly what tricks your inverter into an early shutdown."
The Dry Room Clue
The second issue was subtler, and I almost missed it. The client's battery bank was installed in a small utility room—a former garden shed they'd converted. It was dry, but not climate-controlled. Humidity ranged from 30% to 85% depending on the season.
During my inspection, I noticed slight oxidation on two of the battery terminal connections. Not enough to cause a visible fault, but enough to add resistance. At low currents, everything looked fine. Under a 2.4kW load at 2 AM, that resistance created a voltage drop that compounded the sag issue.
The client had read about lithium battery dry room requirements online—they'd even looked into building a proper battery enclosure. But they'd assumed "dry" meant "not raining," not "humidity-controlled below 60% during summer."
(Note to self: always ask about the installation environment before troubleshooting voltage issues. Terminal oxidation is a silent saboteur.)
The Cost of Getting It Wrong
Here's what surprised me most about this case. The client was ready to spend $2,800 on new batteries. They'd seen a LiFePO4 battery recall news report two weeks earlier—some unrelated brand with a manufacturing defect—and had convinced themselves their batteries were next (which, honestly, would make anyone panic at 1 AM).
They'd already contacted a distributor and had a quote for replacement. The only reason they called me first was the refrigeration deadline. If the shutdown had happened on a weekend, they would have ordered the new batteries and installed them without ever finding the real issue.
That's the hidden cost here. Not just the system downtime—though that was significant. The client had to throw out $600 worth of temperature-sensitive inventory during the first shutdown. But the bigger cost was almost $3,000 in unnecessary replacement hardware, plus the labor to swap out perfectly good batteries.
I've seen maybe 40 cases like this in the past two years. Maybe 35—I'd have to check the service logs. But the pattern is consistent: the more the client reads, the more they're convinced it's a hardware failure and not a configuration issue.
The Fix (It's Simpler Than You Think)
We adjusted three things, and the system hasn't shut down since.
First, the low-voltage cut-off. For LiFePO4, I set the cut-off at 44V for a 48V system—lower than default, but still safely above the battery's absolute minimum. I also increased the "delay before cut-off" to 10 seconds, which prevents transient voltage sags from triggering a shutdown.
Second, the charge controller settings. The SRNE MPPT controller had been using lead-acid defaults. I reprogrammed it for LiFePO4 parameters—specifically, absorption at 56.8V and float at 54V. If you're using a separate charge controller, the same principle applies: don't rely on factory defaults for lithium. Match the absorption and float voltages to your battery's spec sheet, not to the manual's lead-acid table. For most 48V LiFePO4 packs, that's 56.8V absorption and 54V float—but check your battery documentation, because some manufacturers use slightly different values.
Third, the terminal connections. I cleaned the oxidized terminals, applied a corrosion inhibitor, and recommended a humidity monitor for the battery room. The client installed a small dehumidifier—$180, much cheaper than $2,800 in new batteries.
The total fix took 90 minutes. The system has been running for eight months now with zero unplanned shutdowns. Same inverter. Same batteries. Different settings.
I'd rather spend 10 minutes explaining voltage profiles than deal with another 2 AM call. But I've also learned that you can't explain this stuff while someone's refrigerator is failing. You fix it first, then explain. The explanation makes more sense at 9 AM, over coffee, when the panic has subsided.
If your system is shutting down at night and the inverter is logging low-voltage faults, check the voltage settings before you replace anything. The manual won't tell you this directly—it assumes you'll pick the right battery profile. But the default profiles are designed for lead-acid, and LiFePO4 needs a different approach. The flat discharge curve that makes LiFePO4 great for capacity is exactly what makes it easy to misconfigure.
(As of January 2025, SRNE's newest firmware includes a LiFePO4 preset option. If your inverter doesn't have it, verify the manual for the specific voltage parameters—or ask your distributor for the correct profile file.)