I don't know how to get individual cell voltages, especially when it is set to lead-acid mode... can you help?
EG4 wall mount batts (4) go from 54% SOC to 0% at night when set to go down to 30%
- Thread starter mjcother
- Start date
You can only get it if running under SOC. On 7-7, were you running under soc or lead acid?
If soc, you can export the data history, rename the file by adding .txt to it and you can post it here or dm me if you want a bit of privacy. I've helped out a lot of people via messages here .
If you don't want to do that, or if running under lead acid ( it formats the spreadsheet to eliminated cell V info), then you have to check for cell V's manually like you posted a pic of the battery display.
Based on the pics you posted, set your absorb, float, system V all to 56.4 for now. That should get a few cells close to 3.55V,
Since you have grid, I would do ac charge tonight. You will need to get cell V of all 4 packs.
Once charge hits the first time @ 56.4V, it'll quickly stop and then will discharge for like 10 minutes and float will kick in. The V will slowly come back up to 56.4. This is the time to take pics of the packs. If you want to slow down the charge, you can set the charge current under Charge Setting to like 20A and each pack will draw close to 5A.
Do this and post back pic of cell V.
Last edited:
TM48
Solar Wizard
EG4's documentation says to use P01 EG4/LUX. The three I have use this and have run without problems for a year now.
TM48
Solar Wizard
delete
I am running under lead acid, as directed by support to fix the SOC calibration. This is noted in post #11.
TM48
Solar Wizard
You started in SOC mode and had problems running the wrong protocol. No?
wpns
Solar Joules are catch and release
That's what I'm wondering, my 18Kpv/PowerProOutDoorWallMount ESS works fine in closed loop with the (non-obvious) Lithium Brand of 1
I get to 100% almost every day, I've been down to 5% without any loss of function, and the REPORTED SOC delta between my six batteries is usually around 5% until they reset at 100%.
TM48
Solar Wizard
Yeah I get to 100% and stay in balance mode for generally a good while almost every day so apparently my usage patterns fit the OEM's methodologies. The three times it shifted to grid the batts were at 19 and 20% so within 1 % of each other. It would be really useful to know the BMS brand and model. I believe the battery is manufactured by Hinaess but there is no info regarding this on their website.
Ok, So you have to get cell V's manually.
@EG4TechSolutionsTeam should have had you downloaded the data history when you were in SOC mode to review the cell V's.
Since you have switched to LA mode, the data doesn't show. I think its still there if you switch back to comms/SOC, but don't recommend that now since you want to fix the imbalance 1st.
So again, set all charging Volts to 56.4V ttl, and when the packs get there, take a pic of all the cells Vs. I think it's doing the 100% reset at 3.550V, which would make the most sense to be same as other EG4 packs but it would also depend on firmware.
Need to find out how cells are. Might have to run in lead acid mode for a couple weeks if you have sufficient time for balance to work. Just need to get it close enough so u can go back to soc mode if you want and let bms handle rest. Default bms balance parameters 3.45V, 40mV is not good.
You'd have to take cells close to 3.65V for decent balance. If you can change parameters to 3.41V, 15mV, it would be way better.
@mjcother
Here's a visualization of what I consider proper balancing through correct bms parameters and regular use. No stress on the cells and eventually will get balanced. This was a brand new pack, built intentionally with a slight amount of imbalance to verify BMS does its job.
Parameters: BMS balance 3.41V balance, diff = 10mV.
Charger set to 3.525V/cell (56.4V)
You can see the imbalance starts out hi (>75mV) and drops about 12mV per day as it hits "full charge" @ 56.4V. You can see each peak is a bit lower each day as cells get closer in V. At the end of 7 dyas, the peak is at a lower cell V, so at this time, I can adjust peak ttl V higher if I want to balance at 10mV at higher ttl Vs.
Here's a visualization of what I consider proper balancing through correct bms parameters and regular use. No stress on the cells and eventually will get balanced. This was a brand new pack, built intentionally with a slight amount of imbalance to verify BMS does its job.
Parameters: BMS balance 3.41V balance, diff = 10mV.
Charger set to 3.525V/cell (56.4V)
You can see the imbalance starts out hi (>75mV) and drops about 12mV per day as it hits "full charge" @ 56.4V. You can see each peak is a bit lower each day as cells get closer in V. At the end of 7 dyas, the peak is at a lower cell V, so at this time, I can adjust peak ttl V higher if I want to balance at 10mV at higher ttl Vs.
Attachments
Last edited:
no, not at all... the original topic was that the batteries shut off when the 18kpv was showing they hand 54%
I have shown pictures of the cell voltages directly from the batteries, they are not that far off, so balancing is really not the issue, drift in the SOC calibration is the issue... somewhere the inverter was thinking that the batteries had 54% left, but in actuality, the voltage was so low the BMS int he battery disconnected it from the system... and all 4 batteries did it near the same time....
I did this, it is shown in post #20
I am running in lead acid mode for a week, per the support I got when I called SS.
The issue is that "put it in lead acid mode' is a grossly inadequate statement, which I learned the hard way. You must also manually change the absorb and float voltages or the batteries will never get above 54v, or 3.375v per cell. To make it worse, these settings are not in the regular 18kPV maintenance menu, they are only in the legacy "Old setting page", they are not in the "working mode" menus, this is also covered in post #20.
Zapper77
Renaissance Man (EG4 as of Jan 2025)
- Joined
- Jul 23, 2022
- Messages
- 2,739
Redundant post
Zapper77
Renaissance Man (EG4 as of Jan 2025)
- Joined
- Jul 23, 2022
- Messages
- 2,739
54v resting is a good 100%.
DougL
Solar Enthusiast
If it hasn't been said yet, you have to get your batteries balanced because if you don't, one or two cells will be at a higher charge and they will hit the upper Protect or Warn threshold and trigger the battery BMS to stop taking charge. That will leave all other cells at a mix of state of charge and your total SOC will be inaccurate. If the battery cells are balanced then it's still important to hit 100% periodically( every 4 weeks or so ).
Others have relied on a single "smart" shunt to measure the SOC since it would be inline and measuring total current which is higher and more accurate rather than the lower currents a bunch of batteries themselves see.
Others have relied on a single "smart" shunt to measure the SOC since it would be inline and measuring total current which is higher and more accurate rather than the lower currents a bunch of batteries themselves see.
The "not that far off" is with pics with ttl volts at 50ish. However anything between 3.1- 3.33V per cell is usally within 5mV even with >10% imbalance = doesn't give good data. The only time the inverter acts on SOC is in coms mode which is not lead acid. As of July 8th you were still in coms mode per your pics.
Your V did not correspond to what bms soc reported (higher than reality) because of 2 things.
1. Low current draw that bms can not account for, and
2. reporting 100% soc when a cell reach 3.55V whle other cells are lower.
If all cells hit around 3.55V at the same time, then there's no prob. If not, need to find out why.
No, you were at 54.7V, though you also wrote "peaked" at 55V. I am asking 56.4V. There is a difference at 3.42ish compared to 3.55ish dependent on current. But you only showed just 1 pack, what are all the others? There could be the cascading effect as packs shutdown.
Also, at the 54.7V, I don't know how long it was siitng there. It takes about 4-5 mins for the V to settle on LFP from the Overpotential bump/slump. So if yu still want help, get any of the 64 cells to 3.55 and see what rest looks like.
455dan
Solar Wizard
My Power Pro Wall mounts started doing similar to OP's when left on the as delivered settings. It started happening in the fourth month.
This was with the 59.5 V?? factory delivered max system voltage.
The batteries/ 18K PV started to report "W 026" low Voltage and disconnect at a higher and higher SOC , finally when it set the notice at 21 % SOC going through the night and BMS disconnected the lowest cell was @2.695V. Yikes
All I did was lower the max system Voltage to 58.5V ( should have done that day one) and increased the SOC limit from 100% to 108%, after a few days things improved drastically the reported battery voltage at 100% charge originally was stopping in the 54.3-54.4 Volt range.... cells never made it to above 3.4V so the batteries were never reaching the upper knee voltage.
Took two weeks until the daily solar charging caught the batteries up to 56.4V @100%, then stopped the charge and back up to where the system could be run down to 0 % if wanted and not BMS disconnect and stopped setting the low voltage W26.
Never did run in lead acid mode and still set to lithium and Brand 1.
I lowered the SOC limit to 103% and just track it and watch the graphs for the highest Voltage reached which still has never hit 56.8V and max cell V of 3.55 within 20 minutes of charge stop. It seems to work for my 18K PV and Power Pros
Last edited:
wpns
Solar Joules are catch and release
I've never been out of closed-loop:
Code:
select max(EG4$Alpha$Vbat),max(EG4$Alpha$BatVoltSample_INV), max(EG4$Alpha$MaxCellVolt_BMS) from EG4;
+---------------------+----------------------------------+--------------------------------+
| max(EG4$Alpha$Vbat) | max(EG4$Alpha$BatVoltSample_INV) | max(EG4$Alpha$MaxCellVolt_BMS) |
+---------------------+----------------------------------+--------------------------------+
| 57.00000 | 57.30000 | 3643.00000 |
+---------------------+----------------------------------+--------------------------------+System Charge Volt Limit is 59.5, but (hopefully?) not used, as I'm in SOC mode.
I wish I could get to 57.... I added 18 more 480w panels today, so I get to a reported 100% quicker and staying longer there on my 4 batts, I am trying the 108% "System Charge SOC Limit(%)" value and see if it helps.I've never been out of closed-loop:
This is in SOC mode with limit of 101%, which was the recommendation a year ago when I installed it, IIRC.Code:select max(EG4$Alpha$Vbat),max(EG4$Alpha$BatVoltSample_INV), max(EG4$Alpha$MaxCellVolt_BMS) from EG4; +---------------------+----------------------------------+--------------------------------+ | max(EG4$Alpha$Vbat) | max(EG4$Alpha$BatVoltSample_INV) | max(EG4$Alpha$MaxCellVolt_BMS) | +---------------------+----------------------------------+--------------------------------+ | 57.00000 | 57.30000 | 3643.00000 | +---------------------+----------------------------------+--------------------------------+
System Charge Volt Limit is 59.5, but (hopefully?) not used, as I'm in SOC mode.
This is when it has been at 100% for an hour..
Batt1 - 55.92 (min3.456 max3.506)
Batt2 - 55.42 (min3.437 max3.469)
Batt3 - 55.29 (min3.431 max3.459)
Batt4 - 55.98 (min3.464 max3.522)
rogerheflin
Solar Addict
There is very little power difference between 3.45 and 3.65v and the BMS settings will control what voltages the inverter provides to charge the battery (unless you are in lead-acid mode).
455dan
Solar Wizard
Interesting , I may move the Volt limit higher again, to see if the 16s cells will get just a bit higher but under 57V. I upped it to 59V now.I've never been out of closed-loop:
This is in SOC mode with limit of 101%, which was the recommendation a year ago when I installed it, IIRC.Code:select max(EG4$Alpha$Vbat),max(EG4$Alpha$BatVoltSample_INV), max(EG4$Alpha$MaxCellVolt_BMS) from EG4; +---------------------+----------------------------------+--------------------------------+ | max(EG4$Alpha$Vbat) | max(EG4$Alpha$BatVoltSample_INV) | max(EG4$Alpha$MaxCellVolt_BMS) | +---------------------+----------------------------------+--------------------------------+ | 57.00000 | 57.30000 | 3643.00000 | +---------------------+----------------------------------+--------------------------------+
System Charge Volt Limit is 59.5, but (hopefully?) not used, as I'm in SOC mode.
Something off with how the reporting is a bit mismatched with the 18K PV , between the cell voltages displayed at the batteries themselves versus what the 18K charges to. I have probably thrown things out of whack by somewhat recently adding a 314AH Ruixu and a Eco-Worthy 280 AH to the mix with a set of buss's but now only have equal length cables on the EG 4 batteries. Still running in closed loop and do have some variance in the Eco - Worthy as far as matching the SOC of the EG 4's. The Ruixu usually displays the same SOC as the EG 4's even though it is a higher AH battery. I need to upgrade the supplied Eco Worthy cables and get the Solar assistant hooked up still.
The above is when it is still charging correct?
Let us know how it goes, and I would keep checking the voltages - but hopefully all of the cells will start creeping up into the 3.6 V range.
I also played with lowering the charge current once the batteries were @ 100% SOC to see if it would help to not trigger a charge shut off .
IIRC originally the 100% charge was being reached before ever getting to 54.8 Volts as displayed by the 18K PV, this was before bumping up the SOC limit to above 100%. The batteries were disconnecting at higher and higher displayed SOC.
Below some of the readings before moving the SOC limit up during the 4th month of operation. SOC showing ~18% and then battery disconnect cells were around 3 Volts. The 18K has never been run in lead acid mode so for whatever reason the Power Pros were brought back into reality by moving the SOC limit, imo.
Other members know a lot more about LiFePo4 than me, just seems to have worked for me with no updates on either the 18K PV or batteries, still running on what firmware shipped in August of 2024- the as delivered firmware, and in no hurry to change it. No new events since 5/24 and can discharge to 0 SOC and get full KWH use.
Attachments
Last edited:
not really, it is bouncing on and off very small charge amounts, but 4-5 hours at 100% the next day, and 55.5 is as high as it got..
My min and max cell voltages are now within 0.010 v, so I think the cells are balanced pretty well.
455dan
Solar Wizard
That sounds a lot better than earlier in the thread.
Maybe time to try a discharge cycle, and see if the SOC will now allow full discharge without showing an SOC above 0 and stopping the discharge . ?
55.5v seems a long way from 57v. I take it to 25% nightly and that is reading 51.9V (3.249min-3.251max), what definition of 'full discharge' voltage are you saying to do? EG4 18kPV seems to have a limit of 20% minimum SOC.
I am assuming the BMS protection will protect the battery if any cell gets to 2.7v.
SOC has a limit of 20%, below that, it gives an error. I and hesitant to change it back and forth between Li and lead-acid mode all the time as the end user on a production system.
I am assuming the BMS protection will protect the battery if any cell gets to 2.7v.
SOC has a limit of 20%, below that, it gives an error. I and hesitant to change it back and forth between Li and lead-acid mode all the time as the end user on a production system.
wpns
Solar Joules are catch and release
That’s the default at which the inverter switches to the grid (or shuts down in off-grid mode, sigh). You can change it to whatever you want, but be wareful of firmware updates setting it back to the default.