First of all my apologies for yet-another-topic about the AC200Max and its (in)accuracy battery level indicator. But I monitored and performed research over 1 year period of time and came to the following in-depth technical observations and conclusions that may be of interest for you all.
So for 1 year I own a AC200Max main unit with the maximum number (2) additional B230 batteries. Making a total of about 6 KwH of capacity. Regarding the acclaimed capacity: it is fully correct → in reality I even measured 5% more capacity than Bluetti advertises. So that’s really cool. During my tests and recalibration attempts I applied stable charge (400W) and discharge power levels (366W).
Anyhow, I also ‘suffered’ from wrong SOC indicators of each single battery pack (main and 2x B230) and the general one shown on the display. I applied the recalibration procedure a couple of times during the year but without definitive success. I asked for a firmware upgrade and Bluetti support tried their best but in the end they had to admit with “your BMS version starting with 1010 cannot be upgraded”.
So what’s the deal in my case: ideally with an accurate SOC level during the day and reaching 100% I can decide to reroute the Solar input (or drain the AC200Max with a decent AC load) to some other devices. Just not to waste any Solar input surplus. But the glitchy SOC levels would jump from about 75-80% to 100% in one go and the PV input was being throttled and cut off. These cases were noticed by me ‘too late’ and could not anticipate. I have connected two Hyundai 490W panels in series (85V OCV) that often deliver the max PV Power input of 922 Watts. So I really want the utilise all input.
With great MQTT/Bluetooth/RPi support (tip: seek further on this forum) I can monitor anything in great details. For instance the SOC and Voltage of each battery pack. I noticed something interesting that is equivalent to my recent purchase in which I extended my total battery capacity to 11 KwH (a PowerQueen 51.2 V 90Ah rackmount unit + a Victron SmartSolar 100/20).
It’s about the charging curve of the LiFePO4 ‘48Volt’ battery. It happens to be suspiciously equivalent to the Bluetti internals. So during bulk charging the battery voltage will ‘slowly’ increase to say 54.0 Volt but as of then an ‘exponential’ charge curve is being observed when you reach 100% SOC. This period takes only about 5 to 10 minutes. It’s about squeezing the last 1% into the battery. The SmartSolar battery charger will stop the bulk charge as of 56.40 Volt and goes to Absorption mode and finally to Floating (in case of LiFePO4 it means it will re-ignite Bulk charging when the voltage drops below, say, 52.25 Volt).
The Bluetti B230 battery has the same charge curve. It exposes the same ‘slowly increasing’ charge curve and accelerate as of 54.0 Volt and ends around 56.5 Volt to report a 100% SOC as of then.
The Bluetti AC200Max battery however considers the battery full at around 54.50 Volt.
When the AC200Max orchestrate the charging of its own battery plus the 2 B230 batteries it perfectly keeps all charge voltages at the same level. They all go together synchronized. Honestly that is beautifully to see! As you can imagine the AC200Max battery pack reports 100% SOC first and 5 mins after the B230 units report 100% SOC.
Nowadays I only look at the battery pack (charge) voltage to see the progress and in between 54.0 and 54.5 Volt you can consider all battery packs 99% full.
In more detail (15 seconds period scale):
My conclusion: I have ‘to live and cope’ with the glitchy SOC indicators. But luckily all battery pack voltages are kept in line by the AC200Max unit. I use MQTT Explorer with some handy graphs to keep an eye on the current status. For instance:
Lastly (1), you may wonder how I utilise the extra 4.5 KwH from the PowerQueen 48/51.2V battery. It perfectly can be connected via the Bluetti D050S device. Nearly no loss (about 5 watts) and the D050S stays very cool! (because the PowerQueen voltage is quite close to the D050S output).
So when the AC200Max ‘surprisingly’ decides its totally SOC is 100%, it just cuts off the ‘ac input / D050S’ but my 2x Hyundai panels keep on charging the PowerQueen battery.
Lastly (2), when both the battery solutions are almost full I start ‘dumping’ to an Enphase IQ7+ or IQ7a microinverter (back to the grid) which can also be directly connected to the PowerQueen.
Don’t hesitate to contact me for more technical details!
Greetings from the Netherlands