Earthing issue or faulty system? https://vrm.victronenergy.com/installation/511202/share/0519c465

I have an extremely strange issue and have worked out part of it but is there something unique in my build which is causing it? Background Im not an auto electrician or sparky but I can diagnose and fix anyones off grid system (cheaper Hybrids or cheaaper systems in general)and even improve on professional installs and tweak settings that i have come across until it has come to my own system. I had very little experience with Victron before my purchase and relied on a reputable seller and installer to supply me with my setup apart from 2 x Ve.Can 250/100 MPPTs I just purchased from another very reputable seller and installer no questions asked and till today theres nothing I would of changed or till today can see if anything I have done to cause my issues which are now ongoing for almost 2 years or have I? Victron was very overwhelming for me personally when it came to understanding Ve.Can Ve.Bus Ve.Direct does the shunt control charge what is the start up sequence ect ect so spent a month studying it and purchasing Rj45s cables lugs ect and installed it. Lynx 1000 Shunt is connected directly to batteries which I already had OPZv 1000Ah all with manufactured cables and buzz bars all extremely short torqued over killed cold hydrualic pressed lugs ect battery disconnect Victron brand (now its a more heavy duty less resistant switch making the little Victron look and perform like something off Wish) on positive side bolted direct to Shunt batteries have individual BMSs with no bluetooth or communications ect and isn’t required. Do not question the battery condition or the cabling (explaned later) batteries are placed in steel racking specifically made for 2V tubular gels and batteries are not earthed to this steel rack or anywhere else. Chargers are bolted to this steel rack and also the Quattro 15000 is and just for overkill the quattro and chargers have an extra earth from body to this steel rack which is isolated sitting on carpet. House is full steel/iron solar panels are all mounted to steel roof and all solar arrays 6 of 6 series are wired individually to purchased combiner boxes (2 into 1) with individual ceramic fuses on both negative and positive sides and combine into a lightning arrester which has an earth output also connected to steel rack and this rack was earthed to the house. Generator has RCD on it and is connected to Quattro via 4mm 2c plus earth. House wiring although is an off grid dodgy second hand cut in half machinery shed the wiring is professionally done and taged with 30ma ripple detectors ect and earthed with a 2 meter stake ect and I cant fault it if anything its way over kill with individual circuit breakers per light switch and power points any questions Ill answer best I can and please take note of earthing system.

Problems from day 1 and keep in mind I studied the hardware connections not the software apart from charging algorithms at this stage and anyone honest will tell you its a complex system with endless features and compatibility and yet to speak to an expert who knows it all but my setup is extremely simple and intermittent issues was causing me grief.

Cerbo wouldnt connect to Wifi for 4 hours I walked away came back and it connected? Intermittent shunt and MPPTs connected Via Ve.Can appearing on GX touch I thought it was my lack of navigating or a setting maybe but system powered fine.

few months later Update was required for Quattro which i did and wouldnt boot back up once installed but disconnecting Ve.Bus RJ45 from Cerbo and that did it. Heres where the fun began

Generator would trip intermittently as drawing to much and I couldnt for the life of me find the settings via GX and it wouldnt charge in bulk just cruise in absorption not to concerned as i didnt really require it i now know the update locked out the feature and was disabled in Ve.Configure

8 cloudy and consecutive days and visitors over and me being over the system not monitoring it house shut down due to batteries dropping to 36v. No alrams no emails or any warnings start up generator for 12hrs and check settings which are all correct low/high soc voltage ect ect angry is not the word to describe and I start again studying the software and long story short finally worked out fualty lynx shunt and a fualty cerbo. A week later cerbo alarm going off all chargers are off and i could of sworn for the life of me there was a short coming out of one of the mppts Can connections? Fuses for pv input blown and PV over volt as an error. This is impossible to happen on all 3 arrays as 1 they are well under specs via volt meter regardless of weather I had this array running on several other systems and the array that over volted is 50VOC 4 series parralled and also no where near 250V via volt meter or software/hardware the fuses that blew were to small at 15A and there was days of 17A but doesnt explain the over volt error. I did find that a firmware update ran for a short time could throw this error maybe 3.13 or there abouts? regardless till today array is the same and monitored and hardly likely. That charger wasnt charging even though appearing on cerbo that charger also had an intermittent voltage reading and swapped to others it was the charger not the wiring and charger replaced under warranty. Cerbo Ve.can coms port also non existent so I wasnt going mad just intermittent. Cerbo replaced and new Cerbo not laggy intermittently like the old one which I thought was normal all RJ45 cables tested and all fine and still have them and had them tested professionally where I learned purchase quality cat8 and a link given to me ect explaining differences.

New cerbo new shunt new MPPT….. Gets more fun now

Voltage irregularities only minor but intermittent but now non existent

Shunt reads fine but I dont ever recall seeing average Ah and its not there and asking questions no1 has replied (via sm)

Temperature readings between Quattro and Shunt but Shunt usually displaying 3 degrees more but also intermittent but now stable and has been for a few week?.

Charging algorithims not responding intermittently

MPPTs via local showing networking intermittently (via iphone)

MPPTs not showing on GX intermittently

Over voltage of batteries intermittent

Batteries not charging correctly stating in bulk stage but at float voltages

I get errors of VeBus warnings overload but nothing Im doing can cause it and looking at VRM history of time and date the information is not avaiable?

I can go on for months about this system and its glitches but my concern was charging. 1 charger on its own runs as it should plug in ve.can and voltage races above set limit and goes into float BUT intermittently. Whats intermittent? I worked out if I do a PV and battery disconnect from chargers they will work as should for a day and the next day or 2 go back to over volting and skipping absorbtion. Months and months of research and trial and error of combinations and that intermittence was consistent. Which charger is doing it it ends up being they all do it even the new one? They even do it without daisy chaining also?? 1 charger I found is the culprit of the skipping absorption and Iv now removed it sending it back for warranty. It gets better 1 of my chargers has had the total kw erased and another or maybe the same shows 12v at 100apms and fairly stable also with a 5000w array and 54ish V max and minium but only via local trends. When daisy chained they all network and show local until they start charging than the total W starts to drop off as PV increases disable charger and its networked again. All 3 chargers do this consistently. I have a 35/150 pluged in cerbo via Ve.direct as its only coms port and it also will network but cant see others network once they get some input. If all 3 chargers daisy chained into cerbo they all work as they should but cant be seen as networked locally and the 35/150 wont follow algorithim staying in bulk regardless of tail settings. At the moment 35 is connected via cerbo the 3 250/100 are all networked locally and all are working as they should but they cant be daisy chained or seen via VRM. Youll notice a lot of comms errors in logs thats due to me just plugging and unplugging cables while powered up Iv never done this and was always cautious but Im so over the system its done my head in. Im off grid Im remote and rely on this system heavily. Theres gaps missing on VRM history i have rebooted 1000 times I have defaulted back to original I have reistalled firmware over and over it has taken me 2yrs thousands of hours many sleepless nights to diagnose what is going on. There are some things Im unsure of as to whats caused this or how this can happen was it a faulty MPPT/Shunt/Cerbo that ran riot via Ve.Can or is it the Quattros Ve.Bus or is it simply caused by the earthing which in this unique combination of house material builds ect caused this? If you dont know 110% please do not tell me as I have ripped out all my PV array wiring out of its conduit climbing in and out of a 400mm roof space 100 times during peak of summer with an iron roof and laying on rock wool fat bat insulation i have replaced all of my cabling, connections, fuses, coms cables bms moduls pv combinations ripple detectors DC disconnect switches ect and its not from my end. I even the other day connected DC via negative first seeveral times to debunk the thought of thats what could cause this or will damadge the shunt or comms ve Ve.Can and I assure you from someone with a vast welding background 5 chargers and a quattro 15000 when fully depleted than connected via 1400kg of old school batteries with 53v and over 1000Ah that 90mm lug hitting an M10 bolt welded itseelf to it and no shunts still the same and working apart from the average consumed which still isnt there. Im not exactly sure how or if the link is correct or shared but hopefully its done correctly.

Update. Chargers wouldnt go into float but did once daisy chained and all plugged in including the little 35/150 connected via Ve.direct. Itll go out of syn but worked for now.

Any genuine help or insight would be appreciated.

Regards Greg

Quite a book you wrote here. But I think my eye fell on this: did you say you had the Shunt in the positive battery leg? That is not correct. It should be in the negative as last device before the battery.

In the positive leg the Shunt will not measure the battery voltage.

Theres more to it and the exact sequence I had issues with may be a little jumbled but youll get the nightmare Iv been going through. No nothing like that. The battery disconnect is on the positive leg between the shunt and the battery bank. Shunt reads correctly.
looking at it

Battery to shunt to distributor. Distributors have inverter chargers and cerbo power connected. I dont have a lynx power in because its just one bank of batteries.

Make sure the cerbo is connected to a local wifi network and you’re not connecting over bluetooth.

After a firmware update inverters are reset to factory settings. Also don’t update the firmware on inverters unless there is a good reason to. (in your case troubleshooting is a good reason).

Maybe obvious but have you installed the VE can terminator plugs?

Victron also has a book on wiring systems with a section on grounding, maybe it can be of help.

Are you using DVCC or another method to control your chargers?

Have you checked the MPPT calculator to see if the solar arrays are within spec?

I tried the VRM link but it is password protected.

Hi thanks for your reply and yes van connectors are in place as well as I have tested the resistance of all these end termination plugs I have about 8 spares and 2 of them have a white coloured wire visibe compared to the others which are green their resistance was slightly higher than the others. Also tested original RJ45 and new cables and genuine Victron cables and the Cat8 cables im running now which is just in the last month are no different in the sense of broken wiring shorting or insuficient insulation which unlikely in this system may transmit erratically.

Solar arrays I know more about than most installers and i know although rare that a panel can short in rare occasions and increase voltage. Extremely rare and even more rare if it can than carry enough amperage to do any realy damage and youll be more lucky to win the lottery 5 times in a row for it to become intermittent. The only other thing it may have been was mice chewed wiring or I nicked wiring while installing and I went through it with a fine tooth comb and nothing. Either that charger or this one Im about to post back for warranty caused some issues. . It aint my array it aint my wiring it never has been thats elimanted burried dug up from the dead revied killed again several times. Not my wiring not my settings never has been. What can cause a charger to locally not hard wired tell other chargers what to do? How can a charger almost elimante total hours of another charger? The things possesed! I have taken screen shots and left all its memory and told the supplier the specifics to get it to….. I cant explaine it without sounding like a looney but i found a way to get it to show what its doing consistently. Its not DVCC that seems to be working but SCS dissaperers if Ve.can chargers are plugged in but comes back up as i believe it should when the ve.direct charger is installed. Is that normal? Makes sense the charged algorithim is in the chargers that are daisy chained and communicating with the shunt so that tail current can be set and work… Which it does now because the gremlin charger is eliminated. But intermittent shared current isnt working with the odd charger and it should and it does but intermittently. Are all 4 chargers faulty? Is everything fauly and replacing 1 item is going to be damadged again? The communications side I dont understand and how do i test units so this doesnt happen again?

Regarding overvoltage on the PV input of a MPPT, there is no margin for error. VOC should be below the max even in low temperature. The MPPT calculator takes this into account.

You mention local networking, are you using a VE smart bluetooth network? If so this should be deleted when working with a cerbo.

So if i understand correctly this is your setup;

Lynx Shunt VE.Can and 3x MPPT 250/100 VE.Can connected to cerbo over VE.Can with terminators on both ends of the chain.

SmartSolar MPPT 150/35 connected to cerbo with VE direct cable.

Quattro 15000 connected to cerbo over VE bus.

Are all devices appearing in the cerbo under devices?