Nexuss

I recently saw a video going though a study they did on LFP cells and it was quite informative , among one of the important points was that cycling from 75-100% caused the most cell degradation. Cycling at a lower SOC range, like 0-60%, resulted in less degradation. Cycling from 0-100% is better for the battery than 75-100% which most people probably wouldn't expect. Here is the study and video : https://iopscience.iop.org/article/10.1149/1945-7111/ad6cbd/pdf https://youtu.be/w1zKfIQUQ-s

The "power" column in the system work mode 1. This will set charging and discharging watts. i set mine to 1000w through the night to prevent draining the battery quickly by larger loads.

4 hours ! that's very long , what is your charging current like? if there is plenty of solar available that is not normal. Paste the history log here so we can check it ,erase serial number for internet weirdos . Also check the firmware version on each battery they should all be the same version ideally. Might be worth upgrading them all .

Seems like that was already confirmed.

If this geyser still uses grid /like in winter it would be useful to untick "limit to load" to be able to supply it with solar power as well. Also just so you know the "grid signal" and "grid charge" functions on the battery setup page needs to be ticked for grid charging to work if you ever want to do that(in work mode you set the time slots which you want grid charging for). I would recommend ticking both so the battery has access to grid charging very low SOC like below 5% the BMS can then request a force charge automatically if that ever happens.

You wont see the CT coil in the flow diagram when "limit to load" is ticked, Since you have everything on the load side of the inverter you can leave it as is. Only thing i would change is the grid tricle feed to 0 watts and on the work mode screen set the SOC much lower . As it is now you will only use 20% of the battery, unless there is a grid failure/loadshedding. I would set them all at 40% as previously mentioned. I have mine set to gradually go down to 10% to get maximum battery utilization. Also tick "Priority load" so the load is priority above battery (battery wil charge with the solar thats left after servicing the load) . Currently it will charge battery first meaning you will use grid till you reach the set SOC in work mode timer.

BMS comms looks good and is working fine , all 4 batteries are there if you look at the discharge current limit ..Your charge voltage is high but i dont think you can do anything about that , i am unfamiliar with your batteries/firmware, might look into using voltage mode instead of BMS comms. Regarding the dip switch settings ,refer to the battery manual. The next screen shows that you have no non essential ? do you have your entire house connected on the load/ups side of the inverter ?

That is not what the "activate" function does... "The "Activate" function (often a checkbox) in the battery setup page of a Sunsynk (also stylized as SunSynk) hybrid inverter is primarily used with lithium battery configurations . It helps recover or "wake up" a lithium battery that has become over-discharged. In such cases, the battery's built-in Battery Management System (BMS) shuts down the battery for protection, preventing it from charging or discharging normally .Enabling "Activate" forces the inverter to provide a low-current charge (from solar panels or the grid) to the battery, even when the BMS is in a protected/shutdown state. This slowly reactivates the battery and allows normal operation to resume." From Sunsynk website.

You can get a new US3000C for R 13,282.50 from powerforum store , you wil have to come down lots on that price for the 5 year old batteries IMO.

On the battery setup , the charging amps is set way too low for your massive battery bank, this is what is causing you to lose out on power , i would set the charge current at 100-120A . Next on the system mode screen tick use timer and set the soc on 40% for all lines . This will make sure you actually use your batteries. Also tick "signal island mode" to make sure it does the N-E bonding if there is no hard bond . It does not seem like he installer was familiar with this inverter so i would make sure the CT coil placement is correct and also the settings on the second work mode screen so you are feeding back to non essentials if there are any . Give some pictures from the other work mode screen and the Li BMS screen please. Also post a graph of both solar stings voltage through a typical day , 4 panels might be too little voltage to keep the MPPT running properly so that could also make you lose power.

Yes your LiBMS screen indicates your comms is working so the float voltage and other voltages on that settings page does nothing(only the charge and discharge amps is what u can/should set). You as the end user can not change the charge voltage unless you disable comms and switch to voltage mode , or install the newest firmware which limits charge voltage (this should not be needed usually). You as the end user did nothing wrong is what i am saying . Somewhere between the inverter and batteries there is a fault causing the charge voltage to go too high. This seems like it could have something to do with the problem , i would go back to Segen and provide logs to them , battery 2 needs to be swapped or have that cell 13 replaced and they all need a firmware update for lower charge voltage. Initial looks at the other packs history does not indicate any other weak cells but i am still going though them all . Can you post some historical graphs from solar assistant for battery voltage /BMS voltage? would like to see what the inverter is recording voltage wise daily. I also notice the logs only provide data for this year ? Seems like Segen deleted the old data which is not ideal. edit * just remembered the history is cleared with firmware updates .

haha ok i admit i did not look at the other batteries and add up the voltages i apologize for missing that ! I only looked at the history data of pack 2, from the picture he posted in post nr one the voltages of the packs are as follows: Battery 1: 54.196 V Battery 2: 52.87 V Battery 3: 53.972 V Battery 4: 54.118 V Very high indeed .... lets see what the other packs history shows if their cells are more balanced . Pack 2 being lower voltage and the rest that high is certainly interesting. i would keep a close eye on that to make sure they dont go that high again , maybe even disable comms and use voltages if necessary. flashing the new firmware for lower charging voltage could be helpful i agree ,just doesnt make sense why it would go this high in the first place .

The highest pack voltage recorded across all provided logs is 53631 mV (53.631 V) Well below 54V max and the exact same value mine goes to and many thousands of other people using Sunsynks/pylontech. Here is a more concise version as it seems you are having trouble understanding what the issue is. Pylontech US3000C Diagnosis Issue: Severe cell imbalance — cell 13 has significantly lower effective capacity (~30–50% less than others). On charge: Cell 13 rises fastest → triggers OV protection early (pulsing/throttling, high OV counters). On discharge: Cell 13 drops fastest → limits usable capacity (pack hits ~6% while others ~78–91%). At sustained 100%: Passive balancing partially masks it (smaller delta, other cells appear high). Cause: Common degradation in these packs after cycles/partial charging; weak passive balancer can't fully correct large mismatch. Current Impact: Reduced real-world Ah, frequent protections, but pack still functional/safe (SOH 93%, no faults). Recommendations Immediate: Avoid discharge below 20–30% SOC. Slow charge to 100% and hold (float ~53.4–53.6V) for days/weeks. Best fix: Install a dedicated active balancer (strongest community solution): Servtec 1A (plug-and-play retrofit for US3000C) — highly recommended. Alternatives: Heltec/JK 5–8A capacitive (wired to cell taps). Long-term: Contact Pylontech for warranty/service if possible. Regular full charges for prevention. This resolves/manages the issue for most users without replacement.

The whole reason it is showing all those HV events is because of the weak cell 13 . He as a end user followed normal use procedure with BMS comms working and amps even limited to 90A so there is absolutely no reason they should not honor the warranty . The logs prove it , dropping from 90% to 6% indicated a clear problem and upon investigation the problem was found , it has nothing to do with having the latest firmware or incorrect charge voltage . Feed those logs into AI for yourself and verify if you want . @JuliusK paste the logs for the other 3 batteries as well .

No problem glad to help! it seems like that cell 13 was weak from the start with that amount of HV errors ... hopefully they honor the warranty . Its also worthwhile to download the logs from the other packs to make sure they are fine . If they wont honor the warranty installing an active balancer could maybe solve the issue, depending on how weak that cell is .

After consulting AI here is what its come up with : Historical counters reveal a chronic issue: Bat OV Times: 11,862 Bat HV Times: 12,932 Pwr OV/HV Times: high (6,923 / 15,041) These indicate thousands of overvoltage events over the battery's life (739 cycles, SOH 93% — still decent capacity). Key Patterns Across All Logs At high SOC (99–100%, Feb 5–6): Cell voltages group reasonably well (ΔV ~0.02–0.15 V during low/no current). Various cells hit High/OV during charge (often cell 11 or 12, e.g., 3.613–3.647 V). Cell 13 frequently appears as the lowest (~3.392–3.465 V in some entries), with individual SOC sometimes lagging at 99% while others are 100%. Frequent OV protections and throttling indicate ongoing passive balancing attempts. Transition to deep discharge (Feb 7): As SOC drops below ~90%, cell 13 voltage diverges downward rapidly. By 56% pack SOC: cell 13 ~3.179 V (individual SOC 56%), others ~3.318 V (91%). At 6% pack SOC: cell 13 3.143–3.251 V (SOC 6%), others ~3.287–3.314 V (78–87%). This matches the earlier low-SOC logs (~19–23%) where cell 13 was the clear runner (OV ~3.666–3.692 V during charge pulses). This full-cycle view confirms cell 13 has significantly lower effective capacity (likely 30–50% less than the average, based on SOC divergence). It behaves as a classic "weak/runner" cell: Rises fastest on charge at low/mid SOC → triggers OV protection early. Drops fastest on discharge → limits usable capacity. At sustained high SOC, passive balancing (bleeding high cells) partially aligns voltages, making the delta smaller and shifting apparent "high" status to other cells. The pack's high historical OV/HV counters (11k+) stem from repeated top-end protections over 739 cycles, exacerbated by this mismatch. SOH 93% indicates overall decent health, but usable Ah is now bottlenecked by cell 13. I would contact Pylontech for warranty/service (logs show chronic protections).

The capacity is still wrong at 290Ah , it should be 296 Ah . The charge and discharge amps on the batt type screen is what the inverter will limit the battery to . The charging amps on the Batt charge screen is the amps it will use for grid charging only . There is a option to download "events" under the history data window in battery view. Edit* you are welcome to post the events and history here so we can have a look..

How exactly did you do this ? Disable the work mode timer ? the cells being at 3,6V each is unusual as @Tinbum is saying . Not sure but maybe it has something to do with your capacity being set incorrectly . So i would change that and do a load test , also paste a picture of your LiBMS screen to confirm that comms is working .

It doesn't have to be super accurate run a load maybe +- 2-3kw and watch for errors on battery view(on the battery that was showing 7%) , the capacity is not that important just check that the packs stay close to each other in terms of SOC all the way down the SOC curve from 100% to however low you want to go. They should stay close to each other(within 5%) all the way down to 5-10% SOC.

Changing the float voltage wont do anything as you are using BMS comms... Your capacity is not set correctly , 74x4=296Ah I would also set the charge at least 100A and discharge at 148A.

I would do a discharge/capacity test while watching what they do in battery view.

Those panels are cooked , why would you rub them with vinegar haha ? water and dirt is under the glass layer . Solar panels have a anti reflective layer that is not acid /detergent friendly btw . Those panels need to be replaced and the new ones need middle support IMO.

The top and bottom battery in a stack will have the least resistance(because of the daisy chaining) and thus provide a little more current than the middle batteries. that along with different age /cycles on batteries will either cause the pause at 89% for balancing or not , i hardly every see the 89% pause anymore as i cycle past 2 balancing points every day . Atleast thats my theory haha!

Other way around , Nr 1and 3 disconnects because they are full first there is no overvoltage, then Nr 2 catches up,the 89-90% pause to balance packs is a pylontech thing , i see the same behavior with the middle battery in my stack as the top and bottom battery receive slightly more current. A graph showing the current flow into each battery would confirm this.

There is nothing wrong with using BMS comms and going to 53,5v . Been doing it for 5 years , over 10MW discharged at 93% SOH currently . What was the highest pack voltage that you saw ? seeing one or two cells at 3.9V would be possible under specific conditions but for the entire pack to be at 58,5 would be very unlikely as the BMS would have to request that charge voltage which it would never do with comms . What i suspect happened is that when you were using the lower charge voltage on the Axpert for those 2 years the individual cells were not hitting the top end of the charge curve thus they were not balancing properly , over time this caused great imbalance. Then when you connected the comms and the higher charge voltage was requested some individual cells hit 3,9 as the pylons really suck at balancing the cells if they are severely imbalanced. Post the batteries cycle history/log files here if you can.