mafiose13

Perfect, from your answer i guess I won't take any chances yet. 
How can I incentivize you to start with mine 😁😁😁? Ill be your tour guide, hiking, climbing if you're ever in Lebanon

No wonder why I can't get any internet connection from connecting to the inverter "LAN" 😆😆😆

On this particular installation, the LAN goes to a dedicated pc with Watchtower software always on for continuous monitoring from a nearby shed to inside the house about 20m length cable. 
The wifi watchpower or other wifi apps are not live data ....they typically update every few minutes, as opposed to pc version. 

So guys I want to report back something positive. After struggling for quite sometime on this particular installation, I decided to remove those panels (6x550w jinkos), move them to the building's rooftop, make a metal chassis with "perfect" orientation (Dead south 180) n Azimuth (35) for the location. Previously west facing @ 15 degree tilt. 
Max PV output at peak previously was 1580-1600 watts!
Max PV output 30min pre-peak today post move was 3017 watts!
We were averaging mediocre 2.5kw/day generation....we averaged a whopping 12.5kw today...5 Fold  increase.....WOW! 
It's not all & always firmware issues...
 
 
 

I should be able to. I've just finished patching an Axpert Max 8kW '28066 for Georg's shade management patches. My head is still spinning after that one, phew.
Perhaps next. These seem like popular units.

Duh, of course. I'm spoiled; grid charging is a rare event for me, so I didn't even think of that.
Ah, you're confusing bulk and absorb stages:

The bulk stage is where the bulk of the charging takes place. How long the battery stays in bulk stage depends on how discharged it was, and the net charge current (basically solar power less load power). I just assumed that the battery was more discharged in the second graph than the first. I don't understand why you seem to be getting consistently longer bulk stages in FLD versus USE battery types.
Can you check the battery charge current for each? I can't see why they would be different.
Maybe in USE mode, you are somehow getting a much larger charge current, and somehow that's causing the BMS to do something strange.
Unfortunately, one of my inverters has a dead power supply, so that has to take priority.

@Coulomb just an fyi (since you like keeping track of things). I just finished updating the U2 to 19.16 (from 19.12) on all 6 units (3 for me + 3 for friends who have the same unit). I can confirm that this update fixed: 
- Odd menu setting 38 with 00/00 is no longer there.  
- No longer prematurely terminating equalization charging at 30-40min randomly, regardless of time originally set. 
Obviously, no effect on premature float bug or stuck at 120v symptom. I am still counting on your help or @Georg594 on that one, when either of your time permits to patch 55.06. 

excellent result, I was going to use it to charge the ev, I think it is resistive load

I did a load test today on one of the units, i went up to 4.1kw all resistive load with no issues for 15min.
Running a capacitative load at 3.7kw tripped a 58 code within 10 seconds. 
So to answer your question about max load, i guess it depends on your load. 
Maybe someone else more knowledgeable can explain why

Yes.
Hard to say. At a wild guess, 70% chance of fixing it.

Duh! That was really late at night, and I forgot to open up the display update files the next day. I'll have a look now.
Briefly. I'll have another look now too.

Couple of more bugs to report on this firmware:
- Battery equalization is randomly dropped or prematurely terminated. Sometimes it works just fine, most of the time I find it ends at in 20-40min range regardless of time set to equalize. 
- When batteries reach 100% in floating with 0A flowing to them, something strange starts to happen. 
10 seconds of 0A and floating 27.2v, all of a sudden i hear a click noise/switch, fans kick up, charging amps kick up to 12-15A for 1 sec (kicking battery volt to near bulk), then taper down 8A 4A back to 0A within 3 seconds....fans taper down. 10-12 secs later same process. So instead to maintaining a super low current to maintain float, it seems when current needed to maintain float drops to 0A, the charger enters into a bulk type mode and realises battery are full and tapers down.
Those 2 issues are added to the current bugs on this firmware:
- Speedy absorption/bulk, Premature float while solar charging.
- Super crapy mppt performance but especially in cloudy weather/morning/afternoon, the "stuck bug"
 
WI-55.06.7z

Just went through this amazing thread post by post. I just have to say that I am amazed at the wealth of information and learning curve on my end. Thank you for all the contributions. 
@Coulomb can you point us to that "thread" with a link, i was not able to find it.