I could see a momentary spike if one of the batteries hits High Voltage Disconnect. But if it stays at that voltage for longer than that spike then I suspect a hardware or firmware issue.
I would pursue a warranty return with the dealer you bought the solar charge controller from.
Thanks Jim, yes it will stay at 26.5ish until a load kicks in. I haven’t ever even seen a spike of more than about 0.5V before. Will pursue warranty claim as I can’t see any way it can be a battery issue. I’ve got an easysolar unit coming next week anyway so I guess the built-in 100/50 will soon tell me if it’s my Smartsolar that’s letting the team down.
Left a couple of fans on and all the lights today, no fault…
I did notice that the voltage seems to quickly fluctuate more than usual by as much as 0.5V when on float. Usually it is fairly steady or fluctuates by 0.1 or 0.2 on an upward or downward trend, just seems a bit erratic now. Been using these for 10 years, literally the first issue I’ve ever had. You can imagine the panic it induces to see your 12V bank at 26V!
I’ve had a similar issue with a lithium set.
I use a Easysolar and a Daly balancer.
I was having issues with 1 cell running away, then going into overvolt alarm. Thinking it was the balancer the only fix was to turn off the solar array.
I altered my float down 27.1v from 27.2v
Also my absorption is now 27.7 from 28.4v.
Not ideal but we don’t use that much power. I still keep voltages well within limits.
Personally I wouldn’t go much higher than 3.45V per cell for max charge. That final .2V is maybe 2-3% of the total capacity and the cells will still balance @ 3.45V (so 13.8V in total). Especially as you aren’t using the full capacity. So absorption set to 13.8V and let it float at about 13.6V, even lower if you like - most of the discharge curve sits between 3.4V and 3.2V (very flat).
Still it doesn’t explain your issue unless you have a faulty battery that is actually spiking up at some point. That seems unlikely with 4 in parallel as even if you do have a faulty battery the other 3 tend to act as buffers. If the voltage from your charge controller truly is spiking up to 26V you would expect the BMS on the batteries to shut them down (or to blow up, whichever comes first 
)
Ken; when your LiFePO4 batteries get close to full, the SCC should change mode to float. Once the batteries are “fully full” the drop in current should let the SCC know to stop, but there are a few cases where the SCC continues to charge, and the BMS doesn’t want that;
If one of these happens, your battery doesn’t want any more charge, and the “last resort” is to throw back at the scc a very high voltage. On Growat’s on a 48v LiFePO4 i’ve seen over 65v, but so far i’ve never seen 26v on a 12v system.
I would be following the advice to remove all batteries, and run one at a time, so maybe A for a couple of days, then B, etc. Mark the ones that are in the system when a fault occurs.
Or … do something like put batteries A and B in the system, run for a day, then put B and C in, then C and D, then D and A, noting which combos fault and which don’t. This way you still have at least 200Ah capacity.
The other theory i would be chasing is somehow 24v from your other sources is getting into the 12v system. I can’t think of a way except for the orion faulting.
The Orion is currently isolated on the 24V side. I’ve swapped out the batteries for 2 new 100Ah lifepo4 batteries that run my freezer in our trailer (separate system, 100/30, 425W panel). Tested the 4 batteries that were removed and all were still within 0.3V of each other. New batteries went fine for 4 days, then faulted again with same results. Retailer is accepting a warranty claim as soon as my Easysolar unit arrives.
Just checking we are on the same page; I wasn’t very clear in my post, but i’m saying that it can also be that the settings on the SCC or MP might not be perfect for the batteries - i’ve had SCC set up for the documented settings for a LiFePO4, but the battery disconnected repeatedly (or in another case, threw a high voltage back to the scc), and the solution was to lower the float voltage by 0.1v at a time, until they ran without faulting.
Yes, I tried lowering absorption and float voltages to 14.2 and 13.8 respectively but it is still happening. Keep in mind that this worked fine for almost 2 years without issue. The only thing I’ve found that stops it happening is by having a constant load on the batteries.
That tells me its very much a BMS complaining issue, or there is a high voltage source (like a faulty orion) that has a high (~26v) voltage but so little current capability that a load will easily squash it. Good examples are HV leakage from adjacent AC cables that can make a warm compact fluorescent light pulse every 30 seconds or so all night - granted that is an AC example, but its still a high voltage on a cable that is theoretically switched off by a mechanical switch, so should be exactly zero volts, and yet has enough power to excite the bulb repeatedly, and can be detected on just about any multimeter.
One test you might do is to remove the batteries from the system - i.e. turn off the battery breaker. The SCC should immediately go to float (provided there isn’t much load), and the voltage should sit at the float voltage point. You might have to remove the load on the inverter by turning off the AC output breakers that feed your house/vehicle/whatever.
If there is a high voltage source, this will probably then be apparent, because the voltage will be well above the set float voltage.
If this test still results in a 26v reading, then you have proved that its not the battery’s BMS throwing that voltage - its coming from somewhere else, and because you can reproduce the situation, it should be easy to find.
If there is not high voltage source, you haven’t proved its on the battery side - you might just not be recreating the right conditions, but it certainly nods and winks towards the battery. At that point i’d be looking for the software to connect to the batteries and see what logs you can pull.
I’m always weary of a situation where a battery retailer takes back the battery, because it leads you to immediately conclude that the battery is at fault, and stop investigating. Many times it has resulted in costing 2 x the shipping costs, but not fixing the issue, because the source of the issue was not the battery.