acdoctor
Solar Enthusiast
This is correct.
Sudden battery voltage drop, only when epever connected
- Thread starter DaniSe
- Start date
acdoctor
Solar Enthusiast
For today look at the mt- 50 not sure if it will show back current though. Also with inverter disconnected shunt should show nothing in or out.
The 1 cell being 0.02 volt off is not a problem at all but it could show a lot worse in either of the SOC knees. I would want to see the all cell voltages when battery is at 12 volts and 14.2 and still charging.
To be sure I am understanding: with the multimeter, check the cells while in the upper knees (top 10% SOC) as it is charging. So physically (carefully!) tip the lid up when it reaches that voltage and check the cells. I do have an extra set of hands to call on for help with this.
And to test at 12V, run load until depleted, then check the cells?
acdoctor
Solar Enthusiast
Correct
If we can see what the cells are doing at the ends we can either determine the problem or rule them out.
If we can see what the cells are doing at the ends we can either determine the problem or rule them out.
I had been using 14.4
acdoctor
Solar Enthusiast
14.4 volts/ 4 cells = 3.6 volts per cell. The BMS should disconnect if any 1 cell exceeds 3.65 volts per cell. That only leaves you 0.05 VPC my 280AH cells get out that much at 3.6 I can see all of them in the blue tooth app. I stop at 3.55 on a daily basis.
acdoctor
Solar Enthusiast
Epever cutting out on equalization/ high volt
I have an AN50 that’s been working great. No complaints other than a broken terminal screw out of the box. EDIT: this thread has developed into solving problems with equalize mode which is apparently the issue. Latest posts (Feb2022) discuss my desire to get a fix from Epever I have a question...
diysolarforum.com
Hmm, this does sound a lot like my issue. Could having this ongoing problem have thrown the batteries out of sync? This morning (after the batteries were charged together in parallel then put back in series after sundown) one was 13.2 and the other 13.4 with the combined voltage flicking between 26.6 and 26.5. We used it before bed to charge devices then removed loads and left inverter on overnight. It was 26.8 on going to bed with batteries appearing equal at 13.4, and just under the inverter dropped to the above voltages by morning. Running more loads this morning in an attempt to trigger the LVD the gap has widened and we are at 13.0 and 12.6.Please read through this thread and see if it sounds anything like what you are seeingEpever cutting out on equalization/ high volt
I have an AN50 that’s been working great. No complaints other than a broken terminal screw out of the box. EDIT: this thread has developed into solving problems with equalize mode which is apparently the issue. Latest posts (Feb2022) discuss my desire to get a fix from Epever I have a question...
My equalization is set to 0min, as per LiFePo4 recs. I saw the OP in this thread has it at the auto 120m, for sealed batteries, did I understand that correctly?
Doing more reading on using the EPEVER I came across this thread, which also cleared up some terminology for me - I always see "absorption" mentioned but it seems on the EPEVER that is called "boost" so my settings may have been incorrect due to faulty understanding. My values were as follows:
Batt. Type User
Batt. AH 100
Temp. Comp. Coeff 0 mv / °C / 2V
Voltage auto
Over disc. 29.2
Limit 28.8
Over volt rec 28.8
Equal Charge 28.6
Boost 28.8
Float 27.2
Boost rec. 26.4
Low rec. 25.2
Under rec. 24.4
Under warning 24.0
Low disc. 23.6
Limit 22.4
Equalize 0
Boost 120
Struggling with basic LiFePO4 settings in Epever Tracer
Hello all. I am struggling to justify some basic LiFePO4 charging parameters, and relate them to how to set up my Epever Tracer 2215BN. The Epever has many charge parameters geared toward charging older chemistries. This is very confusing to apply to LiFePO4, and the fact that terminologies and...
diysolarforum.com
I did find the proper cable to connect my epever to my laptop while rifling through my box-o-equipment (some was given to us so I didn't realize I had it) so I will be monitoring the charging today with more access to real data. (Once I measure the cells after LVD) On that note, with the discharge sending them so out of sync, will I have to charge them both individually to their upper knees?
Okay, I think I have a definitive answer. I got the LVD to open, and the cells in suspect battery were + 3.2 / 3.2 / 2.5 / 3.2 - and on the good battery + 3.3 / 3.2-3 / 3.3 / 3.3 - reading 13.1 total.
Should I even bother checking the upper knees?
The cells with higher charge being fully charged would trigger OVD and cause my SCC to cut input, is this understanding correct? We have had the cutting out issue since day one, which leads me to believe the battery was out of balance this whole time and just getting worse until LVD ended up triggering
Time to call in on that warranty... any tips on shipping LiFePO4 batteries?
Should I even bother checking the upper knees?
The cells with higher charge being fully charged would trigger OVD and cause my SCC to cut input, is this understanding correct? We have had the cutting out issue since day one, which leads me to believe the battery was out of balance this whole time and just getting worse until LVD ended up triggering
Time to call in on that warranty... any tips on shipping LiFePO4 batteries?
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,572
You are correct.
acdoctor
Solar Enthusiast
Sorry I had to replace a HVAC system today. The second set of numbers look marginal. That cell #3 in first pack must be trashed I would contact SOK if they won’t make good on it. It looks like the individual cell could be replaced. Someone on here may have a spare or SOK may just send you one.
You can check upper charge but it will either run lower than the rest or go high first. First is my bet.
You can check upper charge but it will either run lower than the rest or go high first. First is my bet.
No worries, you have been so incredibly helpful
They advertise on the replacability of the cells, but of course it voids warranty - though with the cost of shipping a lithium battery it may be best to just go that route.
Talking with them now, they want me to try using a 14.6v LiFePO4 charger on the individual batteries to allow BMS to balance the cells
I don't have access to grid power (seriously considering a backup generator now) so have asked them if using my smallest panel, a 100W renogy suitcase, will simulate that slow charge well enough to satisfy them.
My guess is they will have me try everything to prove it isn't the battery before allowing me to collect on the warranty, but we will have to see.
acdoctor
Solar Enthusiast
Not likely that it is just out of balance but it could be. Most BMS bleed off highest cells during upper knee of charge allowing one’s that are behind to catch up. The bleed wattage is very small. The heat is dissipated in the BMS. The slower you can charge during that time the greater chance of it catching up. They may be able to tell you at what voltage balance starts. I have mine start at 3.45 volts. That is when you would want to slow down.
You can break down and set up like you did before to 12 volt parallel and charge from solar. When you get to the upper knee set SCC to low amps and high voltage and charge till bms shuts both off.
Then I would want to know what the suspect cell voltage was just before the highest cell tripped bms. Should trip at 3.66 volts.
You can break down and set up like you did before to 12 volt parallel and charge from solar. When you get to the upper knee set SCC to low amps and high voltage and charge till bms shuts both off.
Then I would want to know what the suspect cell voltage was just before the highest cell tripped bms. Should trip at 3.66 volts.
I found this thread on the SOK specs that talks about balancing <14.4
SOK LiFePO4 specs: 100Ah vs 206Ah
The SOK website lists basic specs for each of their batteries. But, apparently the only place you can currently download detailed spec sheets is the SOK Facebook Group. I've attached the spec sheets for their two original 12v batteries (100Ah and 206Ah). Specs sheets for their two newer...
diysolarforum.com
Relevant passage: "Under Balancing Voltage it says <14.4V. I found that odd given many BMS balancing systems only begin to balance at about that voltage. I asked the question on the SOK FB group and got an answer from the SOK rep. They said the SOK BMS does continuous balancing up to 14.4V. Apparently it's a real thing, but I don't know how it compares qualitatively to other balancing systems."
So as long as they are charged in parallel, they can be balances together and if it is just a balance issue then the low cell would theoretically catch up? I don't have to do them individually?
EDIT: I inspected the cables and the PC cable seems damaged? I did not specifically inspect it before inserting so don't know if was already thay way. Images attached.
- - -
EDIT 2:
Read up on RS485 ports here https://diysolarforum.com/threads/triron-4210n-and-multiple-rs485-devices-noob-sorry.23899/
Could be corrupted data readout on MT50? Also it seems the cable heads just look like that from the start for RS485 to USB?
- - -
Something exceptionally alarming happened today while doing a charge in parallel.
I have both MT50 and computer monitoring the system, the computer had been continuously running the last 60min, taking data every 10s. I walked over and saw the overheat alarm flash and then clear. So I stuck around to see it happen again and saw on the MT50 the PV volts shoot up to 650+ (I have two panels on, at 80voc , 10isc), before giving the overheat warning again. I of course immediately disconnected my array.
The SCC internal temp was reading 38.7°C and the manual states that the controller will reduce charging current at 81°C and shut off at 85°C, to restart when below 75°C
I caught the beginning of the spike in the program, but with the 10s interval it did not capture the full progression. The PV and charging cut to 0 at the spike, as did panel wattage. I have attached images from the readout.
Is it physically possible for that voltage reading to have been accurate or is this more likely a faulty readout? Everything was normal in the readouts up to that moment . The first voltage dip was a cloud, been stable lighting since then.
- - -
EDIT 2:
Read up on RS485 ports here https://diysolarforum.com/threads/triron-4210n-and-multiple-rs485-devices-noob-sorry.23899/
Could be corrupted data readout on MT50? Also it seems the cable heads just look like that from the start for RS485 to USB?
- - -
Something exceptionally alarming happened today while doing a charge in parallel.
I have both MT50 and computer monitoring the system, the computer had been continuously running the last 60min, taking data every 10s. I walked over and saw the overheat alarm flash and then clear. So I stuck around to see it happen again and saw on the MT50 the PV volts shoot up to 650+ (I have two panels on, at 80voc , 10isc), before giving the overheat warning again. I of course immediately disconnected my array.
The SCC internal temp was reading 38.7°C and the manual states that the controller will reduce charging current at 81°C and shut off at 85°C, to restart when below 75°C
I caught the beginning of the spike in the program, but with the 10s interval it did not capture the full progression. The PV and charging cut to 0 at the spike, as did panel wattage. I have attached images from the readout.
Is it physically possible for that voltage reading to have been accurate or is this more likely a faulty readout? Everything was normal in the readouts up to that moment . The first voltage dip was a cloud, been stable lighting since then.
Attachments
Last edited:
I've never seen an Ethernet cable look like that.
I was looking up listings for getting a new one and they all the USB ones seem to have that dark substance under the clear plastic in the image, so I assumed it may just be a manufacturing thing
Latest info gathering from the software shows the following tracking at the end of the day. It only holds around 30min of data at 1s intervals so I unfortunately don't have the moment captured when this behavior started. Having never looked at this data before, I am unsure quite how to interpret these rapid swings. Could this be the OVD engaging in the upper knees? Sunset was around 5:25pm
Voltage was right around 14.0, appearing to charge normally with constant PV voltage and low incoming amps as the sun was low and I went to start dinner. Checking now is 13.33, not connected to any loads just the controller. Is that just the battery settling? Holding steady around 13.30-13.33 with controller and panels connected now that it is dark.
Voltage was right around 14.0, appearing to charge normally with constant PV voltage and low incoming amps as the sun was low and I went to start dinner. Checking now is 13.33, not connected to any loads just the controller. Is that just the battery settling? Holding steady around 13.30-13.33 with controller and panels connected now that it is dark.
acdoctor
Solar Enthusiast
I am not seeing what you can in post # 55. Post #58 I think that is just mppt searching with little watts to work with. If you have graph of amps at the same time I think you would see it would be very small to none in a reverse manner with tiny bumps.
The 13.33 is normal settling.
The only way to explain the high PV voltage would be possible loss of battery voltage to SCC. The graphs show constant voltage, solar flair, emp?
The 13.33 is normal settling.
The only way to explain the high PV voltage would be possible loss of battery voltage to SCC. The graphs show constant voltage, solar flair, emp?
The high readout I only saw on the MT50, the interval for data collection didn't time to that reading. However, looking back through the manual it says not to use both RS485 ports. It has not happened again since using just the software.
After a couple days charging in parallel, we are back on 24v to see how they behave together. The low battery is lower on the low end and and higher on the high end, but in neither case is the overall voltage at the specified cutoffs. Today shortly after the ovd, the good battery read 13.6 and the suspect battery read 13.8. In float at 26.8 the two batteries are reading 13.35 and 13.34.
To use the system and keep our essentials powered while I work with SOK, I have adjusted the max charge to be below where the ovd occurs and we are not using any extra power at night. This has stopped the disconnects on either end, of course at the cost of capacity.
SOK suggest we get the Victron balancer for our battery bank (probably a good idea in any case), but are first having me call tech support before proceeding. If I can convince them it is the battery, we have to pay shipping round trip so I am leaning towards picking up replacement cells. These SOK cells I believe are 100ah 3.2v grade A?
Last edited: