Huge lithium degradation

[91cavgt]

Was this the resting voltage?
What was the charge cutoff voltage?

Yes, that was the voltage after sitting all night. Charge cutoff voltage?

The charge controller charges the batteries to 14.6 volts, then it drops down to 14.1 volt float charge. Every morning, with nothing running, the batteries drop down overnight to around 13.4 to 13.3 volts. I’m not sure of this charge cutoff voltage that you speak of.

 

[91cavgt]

Results from the second 100 amp/hr battery are now in.

13.32 volts at the start.

Same 250 watt heater as the load on the same 2,000 watt pure sine wave inverter.


2 hours 23 minutes and 28 seconds total run time.

The BMS shut the output off at 11.4 volts, so 0.1 volt lower than the first battery tested. Of course, the time it took this second battery to go from 11.5 volts to 11.4 volts was quit literally only about 20 seconds so this seems to be an inconsequential difference.


I’ve got a 2 hour Zoom call coming up at 1 pm so I will start the test on the third battery after that.

 

[Vigo]

At least we know real data now. They have definitely suffered capacity loss but they're far from useless and it may still be an issue of imbalance shutting you down early, so its possibly recoverable. Fingers crossed for the third one.

 

[42OhmsPA]

Yes, that was the voltage after sitting all night. Charge cutoff voltage?

The charge controller charges the batteries to 14.6 volts, then it drops down to 14.1 volt float charge. Every morning, with nothing running, the batteries drop down overnight to around 13.4 to 13.3 volts. I’m not sure of this charge cutoff voltage that you speak of.

The cutoff voltage would be the 14.6V.
It would be really nice if you could see voltage off each cell bank. With the voltages you mentioned they should be balancing and charging to "full".

The BMS shut the output off at 11.4 volts, so 0.1 volt lower than the first battery tested. Of course, the time it took this second battery to go from 11.5 volts to 11.4 volts was quit literally only about 20 seconds so this seems to be an inconsequential difference.

It would be nice to know the BMS cutoff setpoints for individual cells, this would establish if a low cell bank is causing the disconnect or they are cutting off above 2.5V.

 

[91cavgt]

Results are now complete for this round of testing. I’ll put all of the results here to make it easier to see all of the data in one place:

Battery #1

13.29 starting voltage
11.5 volts BMS shutdown
2:48:50 total run time

Battery #2

13.32 starting voltage
11.40 volts BMS shutdown
2:23:28 total run time

Battery #3

13.28 starting voltage
11.38 volts BMS shutdown
2:15:10 total run time


All tests were performed with using a 250 watt 120 volt personal electric heater being powered by a 2,000 watt pure sine wave power inverter. Total run time is hours:minutes:seconds.


So the #3 battery lasted 33 minutes less than the #1 battery. That is pretty significant in my opinion. Now we have to find out the why. The next step is to disassemble the box that these 3 batteries are mounted inside of and check individual cell voltages of all 12 cells.

 

[740GLE]

When charging, are you able to verify if the BMS went into high cell voltage protection? 14.6v is hard to hit unless each cell bank is nice and balanced

Knowing each cell bank voltage is paramount to determining battery health’s and trying to pull each Whr from it.

 

[Zwy]

I would continue charging/discharging. Most likely cells are out of balance.

A pack of cells is only as strong as the weakest cell which can be undercharged or just doesn't have capacity. Absent individual cell testing, the pack has to be sufficiently charged to enable balance across all cells to occur. Starting at a lower charge voltage but at the low end of enabling balancing allows time for cells to balance is the first step. After that, voltage needs to be increased in small steps at constant voltage. When current tails off, then increase charging voltage another step. Repeat until the pack reaches full charge and the cells should be balanced. Without BMS communication, this is the only way to attempt cell balance for testing capacity of a pack.

 

[91cavgt]

When charging, are you able to verify if the BMS went into high cell voltage protection? 14.6v is hard to hit unless each cell bank is nice and balanced

Knowing each cell bank voltage is paramount to determining battery health’s and trying to pull each Whr from it.

I can reprogram the EPEver charge controller pretty easily with the MT50 controller. Would raising the charge voltage up to 14.8 volts be enough or should I set it higher?

 

[91cavgt]

I would continue charging/discharging. Most likely cells are out of balance.

A pack of cells is only as strong as the weakest cell which can be undercharged or just doesn't have capacity. Absent individual cell testing, the pack has to be sufficiently charged to enable balance across all cells to occur. Starting at a lower charge voltage but at the low end of enabling balancing allows time for cells to balance is the first step. After that, voltage needs to be increased in small steps at constant voltage. When current tails off, then increase charging voltage another step. Repeat until the pack reaches full charge and the cells should be balanced. Without BMS communication, this is the only way to attempt cell balance for testing capacity of a pack.

Charging voltage has always been set to 14.6 volts so they have been charged fully. Continuing what I’ve been doing is not going to help them. At this point in time, it looks like my next step is going to have to be disassembly so I can test each of the 12 cells with my Fluke.

 

[Zwy]

Charging voltage has always been set to 14.6 volts so they have been charged fully. Continuing what I’ve been doing is not going to help them. At this point in time, it looks like my next step is going to have to be disassembly so I can test each of the 12 cells with my Fluke.

Charging to 14.6V does not mean all the cells are fully charged. This is why you have this problem, you are not the first one to post this occurring.

Absent individual cell voltage data from the BMS, you are guessing and the only way to correct cell imbalance is to start at a lower charging voltage, usually with a power supply limiting both voltage and current to a level where cell balancing can occur and a passive balancer can reduce runner cell voltages.

You can open it up and check/charge individual cells in some cases. If you want warranty, you might void it by doing so. Using a power supply with limiting voltage plus current to a level where the passive balancer can do it's work will correct the problem if it is cell imbalance.

 

[740GLE]

I can reprogram the EPEver charge controller pretty easily with the MT50 controller. Would raising the charge voltage up to 14.8 volts be enough or should I set it higher?

Nope

I’m saying at 14.6 we need to know if the BMS is stopping charting due to cell imbalances.

A charger can sit at 14.6v all day long, but if the BMS stops current going into the battery at say 14.2v due to imbalance, the charger will say “full” but you could be missing 20ah.

 

[91cavgt]

Charging to 14.6V does not mean all the cells are fully charged. This is why you have this problem, you are not the first one to post this occurring.

Absent individual cell voltage data from the BMS, you are guessing and the only way to correct cell imbalance is to start at a lower charging voltage, usually with a power supply limiting both voltage and current to a level where cell balancing can occur and a passive balancer can reduce runner cell voltages.

You can open it up and check/charge individual cells in some cases. If you want warranty, you might void it by doing so. Using a power supply with limiting voltage plus current to a level where the passive balancer can do it's work will correct the problem if it is cell imbalance.

This is not a factory built battery, as I’ve already covered. Since the 3 BMS units that are being used do not have any way for me to see the individual cell voltages, changing charging parameters is just shooting in the dark. It may help, or may not. Finding out where each cell is at in comparison to each other is going to be the most time expedient way to move forward. If the cells are in balance and the cells have just degraded then changing charging parameters won’t do any good. If the cells are in fact out of balance, then the next step is to find out why. Is it due to a charging setting or is it due to faulty BMS units?

This is the direction that most members here seem to be guiding me to, and it makes sense to me. Or are we wrong on this?

 

[91cavgt]

Nope

I’m saying at 14.6 we need to know if the BMS is stopping charting due to cell imbalances.

A charger can sit at 14.6v all day long, but if the BMS stops current going into the battery at say 14.2v due to imbalance, the charger will say “full” but you could be missing 20ah.

Gotcha! Makes sense.



Everything seems to be pointing to needing to know where the individual cells are at, so the next step is to disassemble the housing to check each of the 12 cells.

 

[91cavgt]

Alright, y’all were 100% right! I’ve got an imbalance of cells with all 3 batteries.


Battery #1

Cell 1 - 3.55 volts
Cell 2 - 3.67 volts
Cell 3 - 3.30 volts
Cell 4 - 3.34 volts

13.86 total voltage


Battery #2

Cell 1 - 3.50 volts
Cell 2 - 3.60 volts
Cell 3 - 3.29 volts
Cell 4 - 3.35 volts

13.74 total voltage


Battery #3

Cell 1 - 3.47 volts
Cell 2 - 3.59 volts
Cell 3 - 3.30 volts
Cell 4 - 3.40 volts

13.76 total voltage



While it was apart, I was also able to read the info on the BMS modules. They are xiaoxiang model# JBD-TP04S006 and are rated at 100 amp each.

Each battery is 4s1p, each battery has its own 100 amp fuse, and all batteries are then wired in parallel outside of the enclosure.



Now that I see we have a big imbalance issue, I’m thinking about changing things up. It appears as though these cells may in fact have a pretty good capacity still. So, I’m thinking about rebuilding this battery into a 4s3p pack and buying a 300 amp BMS that has Bluetooth so I can remotely monitor cell voltages. I’m also thinking about buying a couple of 4’ long sticks of 5/16” all thread, connecting all of the cells in parallel with each other, and putting a small 1 amp charge(that’s all that I can do with what I’ve got) on them to help balance the cells for a couple of days prior to building the new battery. I’ve learned a hard lesson of the importance of being able to monitor individual cell voltages!!!



What do y’all think on this?

 

[Dagoth Ur Does Solar]

I've never built my own pack, but if I was going to build one from a bunch of cells I am not sure on, I think I'd want to pull them all out and figure out a way to individually capacity test each cell first just to be sure they're all in similar condition.

Experienced people here could correct me if that's not a worthwhile thing to do.

 

[42OhmsPA]

Alright, y’all were 100% right! I’ve got an imbalance of cells with all 3 batteries.


Battery #1

Cell 1 - 3.55 volts
Cell 2 - 3.67 volts
Cell 3 - 3.30 volts
Cell 4 - 3.34 volts

13.86 total voltage


Battery #2

Cell 1 - 3.50 volts
Cell 2 - 3.60 volts
Cell 3 - 3.29 volts
Cell 4 - 3.35 volts

13.74 total voltage


Battery #3

Cell 1 - 3.47 volts
Cell 2 - 3.59 volts
Cell 3 - 3.30 volts
Cell 4 - 3.40 volts

13.76 total voltage



While it was apart, I was also able to read the info on the BMS modules. They are xiaoxiang model# JBD-TP04S006 and are rated at 100 amp each.

Each battery is 4s1p, each battery has its own 100 amp fuse, and all batteries are then wired in parallel outside of the enclosure.



Now that I see we have a big imbalance issue, I’m thinking about changing things up. It appears as though these cells may in fact have a pretty good capacity still. So, I’m thinking about rebuilding this battery into a 4s3p pack and buying a 300 amp BMS that has Bluetooth so I can remotely monitor cell voltages. I’m also thinking about buying a couple of 4’ long sticks of 5/16” all thread, connecting all of the cells in parallel with each other, and putting a small 1 amp charge(that’s all that I can do with what I’ve got) on them to help balance the cells for a couple of days prior to building the new battery. I’ve learned a hard lesson of the importance of being able to monitor individual cell voltages!!!



What do y’all think on this?

I'd add an active balancer to each pack and call it a day.
I think you'll have more issues with balance if you reconfigure them into one big battery.

I don't remember how they were built... Each 3.12V cell group is spot welded together and a cable lug is on the end? If so your top balance plan sounds good. A 1A charger is going to take a long while.

 

[Zwy]

I'd swap in a BMS with bluetooth, possibly even active balance, charge slow and let the BMS balance the cells. Power supply would be preferred but 1A will work, it just takes a long time.

 

[Hedges]

Alright, y’all were 100% right! I’ve got an imbalance of cells with all 3 batteries.

While it was apart, I was also able to read the info on the BMS modules. They are xiaoxiang model# JBD-TP04S006 and are rated at 100 amp each.

Does that BMS have a balancing function?
Maybe not. Maybe so but can't keep up anymore.

Each battery is 4s1p, each battery has its own 100 amp fuse, and all batteries are then wired in parallel outside of the enclosure.

So, I’m thinking about rebuilding this battery into a 4s3p pack and buying a 300 amp BMS that has Bluetooth so I can remotely monitor cell voltages.

That would be 3p4s.

I have a JK BMS 4 ... 8s, 300A, with active balancer. Haven't put it through its paces yet. It could work for you. Optional display too.

Your battery has worked until now. Maybe you just need to charge all cells individually up to 3.65V, waiting for low tail current, then go back to using it as before.

 

[91cavgt]

Thank y’all for the input so far.

The only charger that I’ve got can only do 1 amp, and I’ve got no way to do a capacity test on each individual cell.

Each of these 100 amp/hr cells is 25 cylindrical cells, all spot welded in parallel, then there are 4, probably 12 gauge wires, all crimped on to a single ring terminal. The entire 25 cell pack is then shrink wrapped. I’ve got 12 of these 100 amp/hr cells.


I went online a couple of days ago and looked for active cell balancers, but the only ones I found that have cell voltage readouts are in the $50+ range. Ones I see people recommending the most don’t have individual cell readings. Now that I know where I made my mistake, I don’t want to repeat that mistake again. I’m half tempted to go with 3 of the cheaper active cell balancers added to what I currently have, but run a positive and negative for each of the 12 cells to the outside of this box to 12 separate spring loaded speaker terminals. That way I can easily check each cell for voltage, and if needed I could charge one cell up at a time without having to tear the entire assembly apart again.

I don’t know if that would be good or not though. I don’t want to dump a lot of money into this pack at this point. Primarily because lithium has come down in price so much over the last 3 years.

 

[Vigo]

If the cells are already spot welded in parallel, i would simply find a safe max cell voltage and use an adjustable power supply to bring all the 25p packs up to that voltage. May take days depending on amps. As long as you don't have any suspiciously warm cells during this process, you will probably be pretty close to back where you started and can reassemble the batteries exactly as they were and expect a few more years of service before you have to intervene again.

It's also possible that your charging regime simply wasnt creating the conditions under which your BMS's will actually do balancing (or not often/long enough), so i'd research that.

 

[Drunkin Solar]

I would just buy three of these and call it a day. With monitoring, charge adjustments, you can bring the packs back into balance on the cheap at $35.60 apiece for 100a model.

4S 14.6V lifepo4 Smart Bluetooth BMS or 3s 12.6v li ion BMS for jump start Battery pack with 80A to 150A constant current - LLT POWER ELECTRONIC

Application for 3s or 4 series cell protection for Lithium Battery pack (3*4.2=12.6V 4*3.65=14.6V Lifepo4 battery pack example ),client can select suitable type based on your battery characteristics Dimension (L*W*T)=160*105*10mm Various protection functions for charging and discharging (...

www.lithiumbatterypcb.com

 

[Hedges]

Alternate way to balance is to bleed off high cells.

If you charge (unregulated) a cell, when it gets full the voltage will run away high and cause damage. Like filling a bottle getting faster as it goes up the neck.

If you start with a charged back and bleed off the high cells with a resistor, they will come down fast at first then voltage drop slows, so not likely to run away from you. SoC, however, will continue to decline.

Just clip a suitable resistor onto the highest cell in the pack. An 1800W space heater, 15A at 120V, would only draw about 0.4A so slow but could work eventually. 10 Ah/day. Would rather have something 1/10th the resistance. You could buy ~ 1 ohm 25W or larger. Get three and do all three batteries at once. Set a timer to remind you, based on your estimate of how many Ah to bleed off.

3x 1.0 ohm 25W, $17.25 delivered https://www.ebay.com/itm/122050679898

When highest cell gets down near voltage of lowest, charge the battery up again and repeat.

 

[91cavgt]

If the cells are already spot welded in parallel, i would simply find a safe max cell voltage and use an adjustable power supply to bring all the 25p packs up to that voltage. May take days depending on amps. As long as you don't have any suspiciously warm cells during this process, you will probably be pretty close to back where you started and can reassemble the batteries exactly as they were and expect a few more years of service before you have to intervene again.

It's also possible that your charging regime simply wasnt creating the conditions under which your BMS's will actually do balancing (or not often/long enough), so i'd research that.

I don’t have an adjustable power supply. All I’ve got is a 1 amp charger for 18650 cells, a 12 volt trickle charger, an old Montgomery Wards 10 amp 12 volt battery charger, and my EPEver MPPT charge controller that I can change charge parameters on. That’s it.

Now it is possible that my charging parameters were not set right on the solar charge controller. The BMSs that I bought did not come with ANY documentation though so I have no idea as to what settings are best for them.

 

[Hedges]

If your 1A charger is for a single cell of desired voltage, then just charge each cell individually with that.

(First fully charge 12V battery with a reliable voltage-regulated charger. The Epever, I think. Not the Monkey Ward, which might drive excessive voltage perhaps that is OK one time relying on BMS to disconnect, but best to use BMS as last line of defense in case regulator fails or a cell goes high.)

The target is probably 3.65V/cell to top balance, 14.6V for the pack or until BMS disconnects for high cell voltage then charge individual cells to 3.65V

 

[robbob2112]

@Will Prowse - I am curious what your current stance is on active balancers inside batteries?

As part of the BMS verse add-on. Used to be you were firmly against it in your videos and I am wondering if that has changed as battery cell capacity has grown and more BMS have started incorporating them.