Hello colleagues. Half a year ago I ran into a problem that I have now managed to solve. I want to leave this here, maybe it will be useful to someone. It would also be great to get official comments from Victron in order to better understand the reasons for what happened.
So, this is a house in the forest, far from civilization. Autonomous system with EasySolar-II GX 3000 with built-in MPPT 250/70 + four AGM batteries of 130 Ah + battery monitor BMV-700 + PV modules 1980 Wp. The system was put into operation at the beginning of 2021.
The system worked great, but two years later a client complained to me that after sunset the system turned off too quickly due to low battery voltage. At first I thought the batteries were dying. But it turned out that even on a sunny day and with minimal consumption, SOC rarely reached 100%, since something was limiting the charge much earlier. As soon as the water heater was turned on, the PV power immediately increased to normal values.
Having studied the charts in detail in the Advanced section, I found that the built-in MPPT incorrectly reads the battery voltage, overestimating it.
Fortunately, there are three devices in the system that measure battery voltage: MPPT, VE.Bus and BMV. The VE.Bus and BMV readings are completely the same, and the MPPT readings are 2.5-3 volts higher. If this only happened during the day, when the MPPT is a generator, one would think that there is a problem in the wiring that is causing the voltage drop. But the MPPT shows a higher battery voltage even at night, which means it's not the wiring.
As a result, the MPPT "thought" it was supplying the correct voltage to the batteries, when in fact it was almost 3 volts lower. From this, the batteries were chronically undercharged. I checked all the settings, everything was correct there. Then I started looking on the charts for the moment when this problem first appeared. It turned out that the minimum difference in readings began about 6 months before the client complained to me about the strange operation of the system. Since then, this difference in the measured voltages has gradually increased, reaching 2.5-3 volts.
The solar controller built into the EasySolar-II is connected to the batteries not directly, but via a board with electronic components. Something must have happened there. I updated the firmware of all system components to the latest versions, but nothing has changed.
Then I created the E-RMA. Victron told me to remotely disconnect the voltage sensor in the inverter as a test, but that didn't help. The case was declared under warranty and I could send the EasySolar to Victron for repair.
The object is far from us, and all transport costs with dismantling and installation would cost our client at least 300 euros, which did not inspire him at all. In addition, we could not leave the client without electricity, because the system is autonomous. So it was necessary to provide a replacement EasySolar, which at that time we did not have.
And now we have a replacement EasySolar, and before organizing the repair, I decided to try to solve the problem remotely again. By this time, the new version of Venus OS 3.00 had just arrived, but the update did not change the situation. Then I reset MPPT to factory settings and set it up again, but that didn't help either, as did rebooting the entire system multiple times. I did many other strange things, even activated the equalization in the MPPT settings for a few seconds, hoping to somehow stir up the MPPT and reset its error: the battery voltage jumped, but the difference was again almost 3 volts.
Then I had the idea to just change the charging voltage settings in the MPPT configuration so that the actual values on the battery would be correct, but this was not a desirable option, because over time the difference can become even greater. And I don't want fake data from MPPT on the charts.
The last thing I decided to try was to take away the MPPT's ability to make decisions on its own by switching it to "External control" mode, and for this you need to install an ESS assistant, which, in general, is not intended for off-grid systems. As soon as I installed the ESS assistant, the problem was immediately solved.
It's hard for me to say why this happened. I can only assume that some components on the board through which the built-in MPPT is connected to the battery did not work properly, but these components do not affect the operation of the MPPT in "External control" mode. Maybe, of course, the problem is in the MPPT itself, but I have never encountered something like this when I installed MPPT separately. It would be interesting to know the views of fellow installers and Victron employees on this matter.
