Yesterday I experienced once more a "weird behavior" of my to-be-experimented-upon-Quattro (2658, firmware 408): after modifying the battery voltage threshold for inverter shutdown from ridiculously low 37-something to 41-something and the subsequent reset of the device, the Quattro stubbornly refused to even look at (the present) mains.
I mean like unexplained behavior - to me.
As over the years I developed a somewhat close relationship to this specific Quattro, I might even say to have some kind of intuition about its feelings and desires. So I intuitively switched it to "charger only" mode" (1st blackout). Et voila! It recognized main and started charging the battery.
I then switched it back to on (2nd blackout) - Et voila again! Now it saw mains and started to behave as expected. Maybe I will start calling her(?) "my little diva".
Now: I know Victron is a little reluctant with people updating their firmwares, although it seems not as rigorous anymore as it was in the past:
which I find a good development. It makes the user feel being taken more seriously. While I agree with the general sentiment that firmware upgrades shouldn't be done lightly or as some trial&error a.k.a. "cargo culting", I think that giving users even more information at hand so they could make an informed decision whether to upgrade or not would be a good idea.
There is a quite nice "changelog firmware.docx" file under https://professional.victronenergy.com/downloads/firmware/ and I am not aware of it being available publicly. If it isn't - here's the thing: It may seem relevant only to those who actually do have access to the .vff files, but in fact this changelog file should be available to everyone.
If read carefully, one has a so-called "diff" of the features of a current firmware and the present firmware in one place that should ease the decision "Yes I do/do not need a new firmware." It also - if you are able to read "between the lines" gives proof that it is NOT desireable (as Victron points out) to really keep your firmware updated to the newest version.
I read the changelog today, to see if something in there could be applied to my "ignoring mains" situation. And there it is:
At least I believe that is relevant to my situation. But wait, it doesn't end there. While reading through the changelog (since 408) thoroughly, I encountered
xxxx428: Bugfix "parallel and 3-phase not working because of false E11" (This error was introduced in version xx25425)
And this is the very reason one shouldn't simply update to latest. because if you were operating a parallel or 3-p system and updated to 425 "just because", you might have experienced a bad time until 428.
So with all that said, it would seem I need 456. Available versions are 433, 459 and 465. 433 too low obviously, 459 would be ok, changelog between 459 and 465 shows a nice feature "display the remaining waiting time before connecting to grid and also to display the cause of a grid rejection" on the other hand it also shows "A lot of under the hood changes." while I'm sure the latter is good for future enhancements, it also tends to introduce new errors. And indeed, 461/462/463 seem a little bit on the stabilization side. As there also are no critical bugfixes between 459 and 465, the conservative choice is on.
So 459 it is! For me.