question

What does "Lifepo4 lithium battery internal bms- no communication" mean?

Have you not connected the battery's management to the GX?

If not, that would really be a good place to start.

Update 2: I reverted back to older firmware and assistants and the same behavior was observed. I am back on the latest and greatest. I lowered the absorption and float voltage to what the battery selection defaults to. 56.8 and 54 volts respectively. (Which is a little weird, when you go to change those values, the dialog actually says the default is 57.6 and 56.4, and also what my battery manual recommends.) Today was mostly sunny, few if any clouds. I didn't see the solar charger drop to 100 so that's good. The behavior of charging and discharging did happen. I think it was only during the absorption phase. When the voltages gets to the absorption set point, it switches to grid feed in, at which it uses all the power from the MPPT, therefore starts discharging the battery, to cover all the AC loads. The battery SOC drops, and then drops grid feed in, and starts charging battery again. Over and Over. Does the algorithms need tuning? Anybody go an idea?

For anyone else that might run into this problem:

I was having a problem with High DC Ripple on my system as well. I have two 24 volt batteries in series, and would start getting errors after bulk charge finished and absorption started. It was effecting my MPII and MPPT behavior. It ended up being one of my batteries getting full before the other and the voltage was spiking, going over 30 volts on the one battery. The MPII would stop the charge, volts would lower 3 or more volts, and the system would try to charge again. With this happening over and over quickly, it caused the behavior described above, and the ripple errors. I found the culprit by setting up a midpoint voltage monitor with my BMV-712 and started logging on the VRM portal. I had checked the two batteries before with a volt meter, but not right when they were close to fully charged during the absorption phase, and didn't catch it for awhile. I ended up lowering my absorption voltage lower (56.2) and it solved the problem. I picked this voltage by looking at the graphs and seeing when the midpoint deviation started to get above 2 percent between the two, and setting the limit going right before that point.

If you see this on your system, it may be that you have one battery in your set getting fully charged slightly faster than the rest and spiking the voltage.

it seems i have a similar problem.

3h metrics

https://i.imgur.com/UONKXO6.png

https://i.imgur.com/tfp6NsE.png

after arriving to 100% the system starts discharging right away to 97-98% and after they start charging again.

https://i.imgur.com/FeBAovx.png

How can i calibrate that midpoint you mencione?

for me it also look like the same problem, when i set ess to battery life mode Battery is 100% it jumps -5000 to grid than in load battery with +5000 w and it beginn all 2-3 sec it change, sorry but that is realy disapointed from all, when i kill all 3 phases it works but this ess looks crap for me sorry

@Matthias Lange - DE

ich finde das gehört geklärt, echt jetzt ich kauf mir doch ned so ein system und nun wo ich alles habe 96kwh batterie 3 multiplus 5000 1 fronius 5kw on ac-out , 2 mppt.

Dann hüpft der bezug und einspeisung total sinnlos herum!!!!

ich werde auch einen eigenen thread aufmachen mit Video das sehst das, dies sinnlos ist ständig rein raus, und das bei KONSTANTEN Verbrauch ich sag ja nix wenn ständig irgendwas ein aus geht aber, es ist eben bei Ruhigen verbrauch von 980w - 1070 w das selbe spiel.