How much Cell Voltage drift is acceptable for LiFePO4?

[Jman]

I have an 125AH dropin battery I use for camping and charge it at just 14V, 0.3C charge rate. I have done 4 cycles to around 5-10% soc & back to full, after letting sit for 12hours at 60%Soc,
the bluetooth app shows the cell voltages are 3307mV, 3311mV, 3312mV, 3321mV, another few seconds later they are: 3309mV, 3314mV, 3314mV, 3322mV,

Is this fine or too much drift between them? cheers

[Ampster]

What I look for in my Nissan pack is the cell voltage deltas at various SOCs of the pack. In my case I have 6 cells on parallel and if one group has a high delta above the average at close to 100% SOC and then also has a high delta but at a level below the average when the pack is at 50% then I know one or two cells in that group have less capacity. Some people would not call that drift but rather a weaker group of cells exhibiting normal behavior that is symptomatic of less capacity. You could make these observations under load, after charging or at rest.

EDIT
To answer your original question about how much drift is acceptable the answer depends on your application, how good your Battery Management System is and what safety controls you have installed. You and a multimeter can be a very inexpensive BMS. Good BMSs can run as much as $1,000 or more depending on cell count and complexity..I don't worry about a 1/10 volt (0.1 volt) over a SOC range of 30% to 90% because my pack never gets that low or or high and if it does the BMS shuts it down and sets off an alarm. I also have low and high voltage cell cutoffs.

[NochiLife]

I have an 125AH dropin battery I use for camping and charge it at just 14V, 0.3C charge rate. I have done 4 cycles to around 5-10% soc & back to full, after letting sit for 12hours at 60%Soc,
the bluetooth app shows the cell voltages are 3307mV, 3311mV, 3312mV, 3321mV, another few seconds later they are: 3309mV, 3314mV, 3314mV, 3322mV,

Is this fine or too much drift between them? cheers

I wouldn't worry too much about drift less than .1v. Those are actually really close.

The core issue is the voltage range of a LiFePo4 cell (generally given as 2.5 to 3.65v), and that unlike lead-acid batteries, LiFePo4s don't self-balance. When/if the cells get imbalanced (particularly with deep cycling), it is very easy for one cell's voltage to go outside of the safe specification (i.e. 3605, 3599, 3598, 3942). When that happens, it doesn't take too long to irreversibly damage the cell--and then the whole pack becomes rather useless. (Heard of a Prius battery with one bad cell: the car was all but undrivable because the one cell was going negative under acceleration, and way overvoltage when charging. When the one cell was replaced, a perfectly working Prius resulted.)

[Ampster]

Yes the stated voltage range of LifePO4 cells is 2.2 to 3.65v but I never exercised mine that much. In fact the OP said he only takes his to 3.5 per cell (pack =14v).

[NochiLife]

Ditto with my LFP bank; I run it at 3.5vpc tops.

[PNjunction]

Just be careful since like many, one can get obsessed with voltage alone and forget about *time*.

In other words, you can charge a cell to a conservative 3.45v, but if left on charge long enough - even with the voltage limited to 3.45v, you will end up over-charging and damaging the battery. The only reason people don't notice at first is because it takes much longer to achieve full charge at 3.45v vs a higher 3.6v.

From a cell standpoint, *current* decides when you are done, (as long as you use the minimum CV setting of 3.45v) not battery voltage. Typically .005C current. Anything more, and you are oxidizing it.

So - can you overcharge a 12v drop-in if your CV is set to a conservative 14.0 volts? YES, given enough time. This voltage is low enough to allow some leeway if one is cycling it daily so you'll never reach full charge anyway. Leave it on charge at 14v for a week? You'll reach a full charge condition eventually - even though never rising above 14v - and damage the cells.

[NochiLife]

Just be careful since like many, one can get obsessed with voltage alone and forget about *time*.

In other words, you can charge a cell to a conservative 3.45v, but if left on charge long enough - even with the voltage limited to 3.45v, you will end up over-charging and damaging the battery. The only reason people don't notice at first is because it takes much longer to achieve full charge at 3.45v vs a higher 3.6v.

From a cell standpoint, *current* decides when you are done, (as long as you use the minimum CV setting of 3.45v) not battery voltage. Typically .005C current. Anything more, and you are oxidizing it.

So - can you overcharge a 12v drop-in if your CV is set to a conservative 14.0 volts? YES, given enough time. This voltage is low enough to allow some leeway if one is cycling it daily so you'll never reach full charge anyway. Leave it on charge at 14v for a week? You'll reach a full charge condition eventually - even though never rising above 14v - and damage the cells.

OK, but then I have an issue. Initially when setting up the charging settings for my LiFePo4 bank, I had the MPPT drop a volt from "Absorption" to "Float." Problem was, when that happened, the MPPT would go straight to zero current output, and all the loads would run off the battery until its voltage fell down to the new low (5 minutes later). If the voltage was "too high", I would expect it to fall if removed...?

[Ampster]

OK, but then I have an issue. Initially when setting up the charging settings for my LiFePo4 bank, I had the MPPT drop a volt from "Absorption" to "Float." Problem was, when that happened, the MPPT would go straight to zero current output, and all the loads would run off the battery until its voltage fell down to the new low (5 minutes later). If the voltage was "too high", I would expect it to fall if removed...?

If I understand the earlier post by PNjunction you don't want to be floating Lithium cells for long term any way. It sounds like your settings are cutting off current at a set voltage. If that is what you want then leave it alone. I am not as familiar with the Lead Acid charging terminology and I am not sure you even need absorption with Lithium. It is often a timed cycle and maybe that is what you are observing. To make it simple I have set my integrated charge controller to bulk and a refloat voltage slightly below the bulk max voltage. That way I get the benefit of some charging from the sun until the end of the day. I have loads on the system and when thu sun goes down my batteries are topped off. There is not much risk of extended float charging in that scenerio.

[NochiLife]

Yes, I can understand not wanting to "float" Li-cells for a long time. My MPPT configuration is currently set to 3.53vpc (56.5v) "absorption" (which is a timed cycle), then 3.50vpc (56.0v) "float." ("Equalize" disabled.) Overnight, the bank falls to around 3.31vpc (53v). I previously had the "float" voltage lower (55.5v), and observed zero MPPT current when it switched to "float" mode for quite a few minutes. Is this a good (or bad) configuration? Suggestions?

[Ampster]

3.32 is a normal resting voltage of a fully charged LFP cell. I dont see an issue there. Why are you using absorption instead of bulk?

[PNjunction]

Float has nothing to do with it. And this isn't lead-acid specific terminology.

Many of the problems occur because end-users aren't even familiar with the standard charging technology of how CC / CV works. It's pure physics regardless of chemistry.

What complicates matters is that instead of using CC / CV, manufacturers and battery suppliers use some looser terms, like "bulk" and "absorption".

Here's the quick rundown:

CC mode - Your charger/array dumps as much current into the battery as it can, without regard to overall terminal voltage.

CV mode - As you charge in cc mode, the cell terminal voltage rises. But we don't want it to rise too far. So a cap, or limit is set in the charger itself.

CV mode continues - aka "absorption". Not float, but absorption. When the cells are held to an upper-limit cap, the cells are still accepting current. BUT, the nearer in voltage between what the cell voltage actually is at that very moment, and what the CV cap is set by your controller, the current gets smaller and smaller over time.

Why? Physics - to ALL battery chemistries. It takes voltage to drive current, and voltage is represented as the *difference* between two sources. So, as the cell is being held to an upper limit by the charger, and the cell voltage is rising due to being fed current, the *difference* between your CV limit and the cell voltage gets less and less. Less voltage difference, less charge current.

In this way, we see that the charger itself is *only* limiting the upper voltage cap, but it is the cell itself which is the cause for the absorption to become less and less the longer we keep it on charge. At some point, typically 0.05C current for lead-acid, or 0.005C for li-ion, we can consider the battery to be "full".

What is "full"? It is when the battery has enough amp-hour capacity to satisfy the ratings you bought it for.

What happens when we just "pull the plug" at a certain voltage, and don't let it finish absorb? Remember the real battery nerd way of talking about this is the CC (constant current), or CV (constant-voltage) stage.

It merely means that if you pull the plug without finishing absorb, if you do an amp-hour capacity test on it, it will be smaller than if you had let it finish.

LITHIUM: Unlike lead-acid, we don't *have to* fully recharge a lithium battery to full amphour capacity - and for longest life it is best not to - unlike lead.

So.... A quickie way to accomplish this is to just pull the plug when you reach a certain voltage, but the warning is that *eventually* you will reach the end of absorb at some point and do harm going further if you never pull the plug, or the charger has no absorb current trigger - like detecting 0.005C current.

The physics of CC / CV can be seen that if you set your CV to a low voltage point - it will take much longer to finish the absorb current. Use a higher voltage for CV, then absorb times will be lessened.

Cycling a LiFeP04 for say a quick turn-around in a utility vehicle? You'd set your CV to 14.6v (3.65v per cell) to get it charged up quick so the other forklift operator can take over. Goofing around at home where you have time in reserve? Charge to 14.0v (3.5v per cell) where just that .15v lower difference means a LONG time to actually finish. You may have enough time overnight to just pull the plug - and be satisfied that you never truly finished the charge.

Another common example would be LiFeP04 motorcycle starting batts - charged at 14.6v because you don't want to hang around the city for another day when your batt went low. Or see the sights for another day while your bike is in the shop charging at 14.0v Heh, not realistic since you don't *need* to fully charge to start, but this is the example.

THAT is why the quick way to ensure you don't actually overcharge is to set a low voltage - it gives you time to never truly complete the charge - again as measured by amp-hour capacity.

But don't take my word for it. You can prove it to yourself with any hobbies li-ion charger that has both voltage and current displays.

Watch in amazement as it the cell voltage eventually rises to 4.1v or so (we're talking NON lifepo4 - use 3.6 for that), and sit there while the current slowly decreases and at some lower current point, the charger shuts off or indicates full with an led or whatever.

It isn't the charger decreasing the current, but the physics of CC/CV which applies to all chemistries.

So yes, pulling the battery when it reaches the CV limit is one way of defeating absorb. But if you look at any major charger - or perhaps more conveniently a hobbiest lion charger with both voltage and current displays, the quality ones don't just stop when they reach the CV voltage point, but look at current as the trigger.

[PNjunction]

KNOW YOUR PHYSICS or get robbed.

It's been this way since batteries became popular.

Knowing that consumers don't know their physics about batteries, the consumer has been hoodwinked for centuries now.

The latest twist, is that in order to make a buck, unsafe online project scam videos abound. Ah, this is where you make your bucks by "liking" me, "subscribing to my channel", all the while pushing click-bait, unsafe used trash, and providing data to analytical companies.

NO, don't learn "E over I times R" (E/I*R), nor do you want the consumer knowing about the "P over I times E", or (P/I*E).

Keep them in the dark or sales will plummit!

[NochiLife]

3.32 is a normal resting voltage of a fully charged LFP cell. I dont see an issue there. Why are you using absorption instead of bulk?

Because the MPPT runs "bulk" (constant current) until it reaches the "absorption" voltage, at which point it switches to constant voltage mode. After a configurable period of time in "absorption", it switches to "float." About all I can control is the charge current in "bulk", and charge voltage in "absorption", "float", and "equalize." (I have disabled "equalize", and also set the voltage temperature coefficient to zero.)

PNjunction ...I don't mean to rub you the wrong way, but I don't get the connection between "batteries being charged and then disconnected" (i.e. hobby charger), versus off-grid solar batteries that physically can't be disconnected from the charger, a.k.a. MPPT. My use of the "float" and "absorption" terms were by no means intended as battery terminology, but instead were references to the settings available to me in the TS-MPPT-60's configuration settings. Basically, in an off-grid system, there is no way to disconnect the battery from the charger once full, nor would it be practical even if you could (think a sunny day with a bunch of puffy white clouds). The the enigma I'm trying to figure out is how to properly configure the MPPT for longest battery life.

[Raul]

You do not disconnect the mppt charger from the battery. Use instead a SSR big enough to handle the PANELS that feed the charger.
Most of intelligent chargers can be programmed to send a signal for HVD. The trick is to get it right the HV reconnect for loads with minimal cycling.

[NochiLife]

You do not disconnect the mppt charger from the battery. Use instead a SSR big enough to handle the PANELS that feed the charger.
Most of intelligent chargers can be programmed to send a signal for HVD. The trick is to get it right the HV reconnect for loads with minimal cycling.

?!?!? . If HVD (high voltage disconnect) is encountered, there's something wrong with the system setup. I agree that the MPPT never needs disconnected from the battery; I can very easily shut my MPPT off programmatically at any time or for no reason at all via MODBUS--the problem is that that is not practical. If I turn on the microwave, I need about 2,000W of solar power to balance that out--or drain the batteries. Actually, there are small loads 24/7--if the solar is disconnected from the MPPT, the battery will start to drain the instant the solar is disconnected.
"Floating" the batteries in my context would have to do with maintaining a certain voltage on them when the sun is shining, so that loads don't pull from the battery unless they exceed the available solar power.