14.2, 14.4, 14.6 - seems like there is no agreement on charging voltage

[Browneye]

The old A123 ANR26650 M1 specification stated it but I'm having trouble finding a copy of the old datasheet. HKJ copied the info and put it on his website
This site also talks about it

Technical Information LiFePO4 Care Guide: Looking after your lithium batteries

LiFePO4 Care Guide: Looking after your lithium batteries - Technical Information LiFePO4 Care Guide: Looking after your lithium batteries

www.evworks.com.au

Then there is also a couple on YouTube that put lithium in their RV a decade ago and the distributor was telling them (per manufacturer) to charge up to 3.8 or 3.9V
Yes, this is "outdated" information as we now know that there is no benefit above 3.65. That might even change to 3.60 one day
Just because it's old information does not mean it is incorrect. It is the max safe charging voltage. You have nothing to gain and you have no body parts to lose but cycle life is something you will very likely lose.
I had thought this was well known information on this forum.

Cuz this...and then they fail.

 

[archjeb]

Thanks everyone for the help and pointers. I think I'm ready to move these from the bench and install now.

 

[John Frum]

Thanks everyone for the help and pointers. I think I'm ready to move these from the bench and install now.

Before you do I suggest a couple of tests.
1. find out where the weak cell nose dives. that will tell you your cycle bandwidth.
2. re-charge and see if the cells re-converge sufficiently. that will give you an idea if your BMS can maintain the pack balance.
3. make sure you can administratively disable your charge and discharge current paths through the BMS.

 

[archjeb]

Before you do I suggest a couple of tests.
1. find out where the weak cell nose dives. that will tell you your cycle bandwidth.
2. re-charge and see if the cells re-converge sufficiently. that will give you an idea if your BMS can maintain the pack balance.
3. make sure you can administratively disable your charge and discharge current paths through the BMS.

I did a full discharge and recharge, this time with my Minn Kota charger that is 15Amps @ 14.2volts. This time it charged without BMS cutoff and it finished the charge without flashing red at me. So that is a good thing.

As far as re-converging sufficiently; that cell 3 is still hitting peak before all the other cells and cell 4 is always the lowest. When I did the initial manual balance they were more equal; but after a full discharge/charge cycle, its showing up as the weaker cell and #3 is still the strongest. At least I'm closer now and the Minn Kota charger is not complaining...so that is an improvement.


What do you mean by #3? Is there a menu option in the settings on this SOK BMS to do that?

 

[John Frum]

I did a full discharge and recharge, this time with my Minn Kota charger that is 15Amps @ 14.2volts. This time it charged without BMS cutoff and it finished the charge without flashing red at me. So that is a good thing.

As far as re-converging sufficiently; that cell 3 is still hitting peak before all the other cells and cell 4 is always the lowest. When I did the initial manual balance they were more equal; but after a full discharge/charge cycle, its showing up as the weaker cell and #3 is still the strongest. At least I'm closer now and the Minn Kota charger is not complaining...so that is an improvement.

1. Did you determine the pack voltage where the weak cells drops away from the rest?
2. What was the cell delta at charge termination when you recharged?

What do you mean by #3? Is there a menu option in the settings on this SOK BMS to do that?

I don't have a sok bms user interface to look at but most of these smart BMSs have an on/off widget that allows you to disable/enable the charge through the bms and another to disable/enable the discharge path.
If one or both of these don't work then your BMS will not be able to protect your battery.

 

[KITROBASKIN]

Image is from basic settings (passcode 200010)

 

[archjeb]

Image is from basic settings (passcode 200010)

Are there any docs on the these settings?

 

[KITROBASKIN]

That would be nice. Turns out I had accidentally messed up idle calibration on one battery, explaining the error in a previous screenshot of the Bluetooth display. So a guide would be helpful. SOK has a blog, and there may be something here on diysolarforum?

Even with the very helpful information in this thread, I am still wondering about the way forward for our backup system. Lowering the absorption voltage to 14.4V and float to 13.7V at this time but would not mind having a long term goal of getting that one cell in balance to the other three. Thinking of separating the good battery and letting it sit less that <80% and trying strategies with that ornery 12V 100Ah SOK, hopefully without opening it up and placing a load on the runaway cell.

This is the first winter without any generator backup at all for our off grid home, and the backup system has been great, along with shutting down the home Outback inverter during sleep time when cloudy days happen about 20% of the time these days. As we get more sun, it is hoped that the rascal battery will spend more time at the 13.7V point and the BMS will slowly improve the cell balance? Can I trust the Bluetooth voltage numbers of individual cells?


Thanks archjeb for all your inquiries.

 

[archjeb]

Celebrated a little too soon. It is better...but not good enough to call it good.
After some discharge/charge cycles, its not as good as that first manual balance. Good thing is, the charger (Minn Kota charger) is not erroring out when doing a full charge. The BMS is not cutting out with the 14.2v charger, like it was before -- so this is an improvement.
The 14.6v Noco charger is pushing it over the edge and the BMS will do a CMOS cutoff. But with the more conservative charger (Minn Kota at 14.2v), the charger is actually finishing its charge cycle...so this is an improvement. But after some discharge/charge cycles, I'm still around 200mV of imbalance. Does SOK test/check this stuff before shipping? Seems weird that I"m having to go through all this work for a so called, 'drop in replacement'...

I am getting over 100Ah capacity, so that is good; so either I can continue to worry about this...or just go with what I have and pray that the BMS will eventually balance this thing out and go fishing. I'd like to do the latter...assuming the weather will cooperate

Its only the 1 battery that is creating this headache, the other 2 are just fine...

 

[sunshine_eggo]

The 14.6v Noco charger is pushing it over the edge and the BMS will do a CMOS cutoff. But with the more conservative charger (Minn Kota at 14.2v), the charger is actually finishing its charge cycle...so this is an improvement. But after some discharge/charge cycles, I'm still around 200mV of imbalance. Does SOK test/check this stuff before shipping?

Not based on the multiple accounts of folks having imbalanced batteries out of the box.

Seems weird that I"m having to go through all this work for a so called, 'drop in replacement'...

"drop in replacement"... for lead acid is what they meant.

I am getting over 100Ah capacity, so that is good;

Yep.

so either I can continue to worry about this...or just go with what I have and pray that the BMS will eventually balance this thing out and go fishing. I'd like to do the latter...assuming the weather will cooperate

It will provided you regularly charge it to 14.XV.

 

[Checkthisout]

Jesus. I thought lipo4 was supposed to be less hassle than la?

I'm seeing little lights for draining cells, arguments over voltages, tripping bms.....ugh

 

[sunshine_eggo]

Jesus. I thought lipo4 was supposed to be less hassle than la?

I'm seeing little lights for draining cells, arguments over voltages, tripping bms.....ugh

We're only seeing the problems - not the 90%+ that doesn't post about a problem or sing of their success on every corner.

There can be some initial discomfort with a new LFP battery, but if you follow a few simple rules about charging, they are generally maintenance free after that.

 

[Browneye]

We can't tell from the second picture if your pack is balanced as all the cells are in the flat part of the curve.

Finally got around to plugging the RV in and turn on the new Meanwell charger - it's set to 'Gel' which is 14.0V and terminate at 5% of charge capacity.

Here's a series of bms screen-captures over about a 2 hour charge session. It's a 25A charger. Notice as the battery voltage came up the bms decided it was full. The smart shunt meter was already showing it at 98%. So once it gets to near full charge they then start to agree on the SOC. Over the past few months the SCC has been set on my 'storage' profile which floats at 13.25V, the battery tends to sit idle at about 75%. For active usage I have a higher-volt program I load into it.

The new charger is pretty slick - it replaced the old Powermax brick.

Since charging cuts off before any unbalanced cells can shoot up, it's still hard to tell if they're balanced at high knee. They tend to stay this way all the way up to about 3.55Vpc, but I don't bother with trying to charge any higher. I didn't catch the cell delta at charge termination, but the log does show it reached 13.95volts - that would be 3.49Vpc. I'm satisfied with the cell balance.

 

[hwse]

Finally got around to plugging the RV in and turn on the new Meanwell charger - it's set to 'Gel' which is 14.0V and terminate at 5% of charge capacity.

Here's a series of bms screen-captures over about a 2 hour charge session. It's a 25A charger. Notice as the battery voltage came up the bms decided it was full. The smart shunt meter was already showing it at 98%. So once it gets to near full charge they then start to agree on the SOC. Over the past few months the SCC has been set on my 'storage' profile which floats at 13.25V, the battery tends to sit idle at about 75%. For active usage I have a higher-volt program I load into it.

The new charger is pretty slick - it replaced the old Powermax brick.

Since charging cuts off before any unbalanced cells can shoot up, it's still hard to tell if they're balanced at high knee. They tend to stay this way all the way up to about 3.55Vpc, but I don't bother with trying to charge any higher. I didn't catch the cell delta at charge termination, but the log does show it reached 13.95volts - that would be 3.49Vpc. I'm satisfied with the cell balance.

Thanks for the update.
The shot that shows 13.85v has a delta of 24mv. That is about the point that my packs start to get a big divergence. If I when to 14.1v I had a OVP cutoff with a couple of hundred mv delta. It would be interesting to see what delta you have at 14.0v.

 

[Browneye]

MeanWell chargers have a dip-switch defined charging profile in either 2-state or 3-stage. I use the 2-stage - charging terminates when battery reaches the boost voltage level. 3-stage would then reduce charging to a float level to maintain, which I don't need because the solar charger does that. They also offer a BT interface device that lets the user set infinite charging parameters - it's about $70.

So far I really like the charger. Most will want the 750W one - it's rated for about 45amps. All I need is a quick boost after a bunch of cloudy days and plug in or run the generator for 120VAC. My only caveat is that it does not re-boost. Once charging is terminated the unit has to be power-cycled. Which is fine for an RV, might be cumbersome for other applications.

Presets for charge voltage, current tapers as battery reaches the setting:
Default - 14.4
Gel - 14.0
Flooded - 14.2
AGM, Lifepo4 - 14.6

I could switch it to the flooded 14.2 setting and see if charging terminates without the bms shutting down for cell over-voltage.
The higher the charge voltage the more I'll get a cell that will reach full and run up, so bms terminates charging when one of them reaches 3.65V.

There's not much point in charging past 3.5Vpc, or 14.0V for a 12V battery - that is arguably full state of charge. My pack would never reach 14.6 volts - the bms would terminate charging before it ever got there.

 

[hwse]

MeanWell chargers have a dip-switch defined charging profile in either 2-state or 3-stage. I use the 2-stage - charging terminates when battery reaches the boost voltage level. 3-stage would then reduce charging to a float level to maintain, which I don't need because the solar charger does that. They also offer a BT interface device that lets the user set infinite charging parameters - it's about $70.

So far I really like the charger. Most will want the 750W one - it's rated for about 45amps. All I need is a quick boost after a bunch of cloudy days and plug in or run the generator for 120VAC. My only caveat is that it does not re-boost. Once charging is terminated the unit has to be power-cycled. Which is fine for an RV, might be cumbersome for other applications.

Presets for charge voltage, current tapers as battery reaches the setting:
Default - 14.4
Gel - 14.0
Flooded - 14.2
AGM, Lifepo4 - 14.6

I could switch it to the flooded 14.2 setting and see if charging terminates without the bms shutting down for cell over-voltage.
The higher the charge voltage the more I'll get a cell that will reach full and run up, so bms terminates charging when one of them reaches 3.65V.

There's not much point in charging past 3.5Vpc, or 14.0V for a 12V battery - that is arguably full state of charge. My pack would never reach 14.6 volts - the bms would terminate charging before it ever got there.

Thanks for the info on the MeanWell chargers. I have not heard of them before. I am looking for a supplemental high-amp charger to add to the 100A output of my Magnum for when I am dry camping. I recharge primarily with my 10kw genset so 100A is a lot of runtime. My BMS's will each handle 200A charging so pushing it to 200A is still fairly conservative.
Do you have charge balance turned on or off? I have found that off does a better job of balancing.

 

[Browneye]

JBD/Overkill default balancing is over 15mV delta at 3.4Vpc or higher.
As mentioned, I don't charge past 3.5Vpc anyway.

We use about a hundred amp hours of power over a day/night. All the major appliances are still LPG - 15yo winnebago. If it's hot and the AC needs to run the generator runs, or we leave.

 

[hwse]

My coach is all electric, so it is a total hog. I use 300AH to 400Ah in a normal day when you include all of the computers and monitors I have running. When I got my two batteries, they tested out with 925Ah total which is 5Ah more that rated capacity. The problem was moderately balance battery had 475Ah and the badly balanced battery only had 440Ah and could not be discharged below 12.0v because that was where the low cell cut off. After working at getting them balanced I can now charge both batteries to 14.3v and the capacity of both increased. I have not had the opportunity to retest for capacity, but I am sure that the the weak battery is at least as high as the stronger battery, and it should have a marginal gain.

For me, I have probably seen a +40Ah gain (5%) to my capacity so it was not trivial.