Zeke's last paragraph says it all. The AIOs from Sol-Max and EG4 are pretty good at reporting an accurate SOC, regardless of charge and discharge rate, as long as the manufacturer-matched batteries reach 100% reasonably often. Even when the batteries sit in the mid-range for several days, the SOC seems to remain pretty accurate. When they hover near the low end for extended periods is where they will drop to zero without warning.
I find it hard to trust external shunts for Lead Acid over time for the reasons Zeke mentioned, and for the reasons I mentioned previously regarding aH capacity depending on rate of charge/discharge. And for LFP, it generally makes more sense to rely on the battery BMS stated SOC, even though that can't be completely accurate, either. It's pretty wild to watch the discrepancy between a large EG4 rack-mounted bank versus what a Sol-Ark 15K reports in open loop. I have watched this pretty closely at a few sites, and the Sol-Ark seems to be all over the map. I have pretty limited experience with the Victron shunts, so I can't speak to those. I am getting ready to do a small Multiplus installation where SOC is going to be important due to tight margins between solar production and load, so I will be interested to see how well it does. Jason On Wed, Aug 26, 2026 at 4:09 PM Zeke Yewdall <[email protected]> wrote: > I did not expect this to be such a controversial topic... but since it > is... here goes :P > > I consider having an AH counting SOC meter absolutely vital on most LFP > projects -- at least on any older equipment. Voltage is just way too flat > on LFP to make any reasonable decisions based on it, except below 25% SOC > maybe, and above 90% SOC. But that entire range in between there is very > difficult to tell where you are based on voltage. And most of the older > LFP had no display for SOC (or at best a couple segment bar graph), and no > way to do comms to them for many of them either. So to allow the customer > to actually know what their batteries are doing, and make decisions on > operation, they need to know SOC -- especially at middle levels where > decisions can still be made (other than it's full, or it's empty, start the > generator asap). > > Even on lead acid batteries, I would almost always do an AH counting SOC > meter rather than relying on voltage, as it was just too complicated for > about 75% of customers to grasp voltage -- that's it's non-linear, 0 volts > is no 0% charged, and that it changes at the same SOC depending on whether > you are charging or discharging. > > I've used the FNDC, Victron Smartshunts, and Trimetrics on LFP batteries. > And even a Schneider conext battery monitor thing once or twice. For the > most part, I've been quite happy with how they've functioned. > > Thats not to say that there aren't a number of issues with AH counting SOC > meters, whether on lead acid or lithium batteries, and there are some cases > where they just don't work. > > With lead acid, the biggest issue was usually declining battery capacity > over time lead to the SOC being unreasonably high compared to reality. So > it worked great to begin with, but 4 or 5 years later, it would start > kicking the generator on (or shutting down) at much higher indicated SOC > than it used to. > > For lithium, SOC drift can definitely be a problem. I do tell people they > need to get the SOC to 100% once every two weeks minimum, in the winter, to > prevent this -- 1% accuracy error per day doesn't sound bad... till you go > a full month without recalibrating.... > > Many BMS's built into the batteries experience the same issues with SOC > drift too. Most internal BMS's are blind to currents less than 1 amp or so, > so people who are very good at conserving electricity, reducing their loads > to less than an amp per battery for large portions of the day or at night, > seem to have bigger issues with SOC drift than heavy users. > > The new heated batteries are a big problem with AH counting, and can cause > the SOC meter to be completely wrong and useless at times. I have not > figured out a way around that yet (other than not using heated batteries in > cold climates... they have many other drawbacks IMO, compared to a separate > heating system that's not internal). > > SOC meters that reset to 100% if they lose power, rather than retaining > SOC, can also be a problem if you ever trigger the emergency protection on > the lithium battery. Battery shuts off from low state of charge then SOC > meter resets to 100%, then customer is super confused. The Victron shunts > are the only ones I've found that can be programmed to keep the same SOC > over a power outage. Batteries that disconnect just the charging FET due > to cold temperatures, can also cause a unintended reset, by making it > appear the battery rises to absorb voltage with very low amperage while > charging, when really it's just the charge FET is off. > > For any new systems, I am coming around to using closed loop batteries of > the same manufacturer as the AIO inverter, and everything is just built > in... the BMS reports to the inverter, everything is based on the BMS > control, and the same monitor shows SOC along with everything else. My > first attemps at closed loop control were disasters, and I stayed away from > that for several years, but recently I've been much happier with it and the > level of control it gives you. So that's a different case, and eventually > that'll be all the systems out there. But for now... there's a lot of > legacy systems still out there, and first gen lithium batteries on legacy > equipment -- no SOC display on batteries and no comms. So, that's where I > still believe that AH counting SOC meters are critical for system > operation, as long as the customers know the potential trip-ups and don't > view the SOC meter as gospel. > > Zeke > >
_______________________________________________ List sponsored by Redwood Alliance Pay optional member dues here: http://re-wrenches.org List Address: [email protected] Change listserver email address & settings: http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org There are two list archives for searching. When one doesn't work, try the other: https://www.mail-archive.com/[email protected]/ http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org List rules & etiquette: http://www.re-wrenches.org/etiquette.htm Check out or update participant bios: http://www.members.re-wrenches.org
