PV Assistant remains too long in higher frequency

[PetaJoule]

AC coupled Fronius Symo Advanced 15 on 3 Multi-II 10kVA with PV assistant.
Works well so far, except in the evening the Multi-II do not lower the frequency when they already could/should IMHO.

Currently the PV capacity connected to DC MPPTs (MPPT RS, Multi RS) and AC (Fronius Symo) is about the same and they also deliver about the same yield.

After a short burp in the morning, when this insanely overdimensioned installation has filled the batteries, the system keeps battery voltage constantly at 54V - float. Most of the time, the 15kW Symo is kept on a leash so it delivers around 75W (there might be optimization potential, but that’s for another discussion).

Now in the evening when the Victron MPPTs can’t deliver enough power to keep the battery at 54V - see picture - the PV assistant still keeps the frequency at 52.6 to 52.8 Hz despite V_bat falling.

Now it DOES start to lower the frequency, somewhere in the range of 53.85V, but that’s IMO a tad too late. Why the hesitation? I wonder if the system has the information whether the Victron MPPTs are throttled or not (anymore). If - as is in evening and low light conditions the case - they are not, the system should have a pretty solid hint not to throttle as much and/or as long.

[lawrence (Lawrence Wallis)]

Did you get this problem sorted? In the very early days of the BYD LVL 15.4 i had the same problem that was due to the setting in the battery charge settings in VE config. the instruction from Victron was slightly wrong, since corrected.

When configured correctly, as the battery approaches 100% the MPPT will shut down and the frequency will rise, until the AC coupled power matches the load. More efficient to supply AC loads with AC solar being the theory.

[PetaJoule]

Unfortunately, the problem persists. And as we get winter here on the northern hemisphere, this “leaving energy unused” makes me more and more uncomfortable.

The PV generator capacity is there, to fill the battery - no problem, but it seems once the system went near the 55,2V threhold
(what is considered Absorption voltage limit), it seems to throttle everything and fall back to “holding” 54V.

Even though the system is aware of being only at 95,2% SoC

And even though the “near 55,2V” was recorded only for the voltage @ Inverters, not the SmartShunt

IMHO that’s a bug.