Best LifePo4 charge controller settings known to man for Maximum Service life and Minimum battery stress!!! 5,000-10,000+ cycles?

[Substrate]

Honorable mention - when running conservatively with a CV of 3.375v, when doing the initial "sanity-check" for balance at higher voltages, like 3.5v or higher, if you are within 100mv or less of delta, you are done.

Now you don't have to obsess over micro-sized balance deltas either. Still wise to check for sanity once a year or so. 100mv or less delta? Fine, drop back down to 3.375v cv for cycling.

 

[Gazoo]

Please help me understand...
When my Li battery is in float, the charge current is 0. How will this damage the cells?
I set my float below my charge voltage to allow for the "rest period"

If the voltage doesn't drop below your float voltage then there is a small current going to your cells. Whatever you are using to measure the current might not be able to measure that small a current.

 

[Cheap 4-life]

If the voltage doesn't drop below your float voltage then there is a small current going to your cells. Whatever you are using to measure the current might not be able to measure that small a current.

The voltage will not drop below float voltage if the batteries are being floated like they are everyday for solar storage so the loads can use the pv power first..

 

[justgary]

The voltage will not drop below float voltage if the batteries are being floated like they are everyday for solar storage so the loads can use the pv power first..

What float voltage do you prefer? Float above 3.370 or so = potential big trouble with longevity.

Assuming you got to float mode from a charge cycle that stopped at a recommended volt/amp profile, you started float with a full jug. At that float voltage, your jug is still full. Attempting to add more to a full jug will damage the jug.

 

[Substrate]

What float voltage do you prefer? Float above 3.370 or so = potential big trouble with longevity.

Emphasis on that. Floating at 3.4v / cell (13.6v nominal 12v batt), if given enough time like days, will eventually reach full capacity. That's why I disagree with any recommended float-voltage of 13.6. Only 100mv drop to 13.5 (3.375v) makes a big difference, because chemically, the cell simply can't achieve it.

Worse yet, is that at 3.4v/ cell, some cells react differently. Some will be willing to be recharged to full in days at this voltage, and some won't. A guaranteed way to imbalance the cells. Some get lucky and they all react the same. Some don't at this knife-edge. Don't take the chance at 3.4v for float OR balance triggers.

 

[Cheap 4-life]

What float voltage do you prefer? Float above 3.370 or so = potential big trouble with longevity.

Assuming you got to float mode from a charge cycle that stopped at a recommended volt/amp profile, you started float with a full jug. At that float voltage, your jug is still full. Attempting to add more to a full jug will damage the jug.

3.368v (64v for 19s) is what I’m floating at. That seemed to be where my cells would float/rest and not need current to stay at that voltage.. I assume that means the cells are basically full, full enough anyways.. Before my cells go into float I absorb to 3.421 (65v) for 30 minutes with end amps set at 7amps. I don’t know exactly how full that makes the “jug” but I am floating at a correct voltage from what I have read and from what the charge controller is trying to tell me..
people saying that these cells shouldn’t be floated doesn’t make sense to me. That means that everyone using them for solar storage is using them wrong. Yeah maybe there’s some very very minimal (shouldn’t even be mentioned it’s that minimal) degradation over time from the very small power impressed on cells when they are being floated at the correct voltage.

 

[justgary]

I think you may be correct if your use is more of a daily load since the cells get a chance to discharge overnight. The real problem comes from keeping them near 100% SOC for a long time without ever letting them discharge through a load.

This could easily happen at my small cabin since I am not there on a regular basis to turn on big loads. Having the SCC wake up and stuff the jugs full, then float near the top to keep them full, then go to sleep with the jugs full is not desirable if nothing drains the jugs overnight.

I'll keep small loads on and set float to just above rebulk so they can discharge a bit before the cycle begins again the next morning. Absorb will also be set below 3.37v/c so I'm not trying to fill the jugs all the way.

 

[Cheap 4-life]

Yeah that makes sense that floating the cells when they are basically full for long periods (not cycled every day) could cause problems long term as you explained

 

[Substrate]

Even Andy at his offgrid garage when he was playing around with 3.4v / 3.5v with and without absorption noticed this. He did a video about it. But he's not aware of the knife-edge.

At 3.4v CV, (13.6v nominal for 12v batt) and allowed to absorb for days, he achieved a full SOC.

So the worry here is that anyone who thinks that 13.6v is ok to float at, might find themselves at a full SOC (full capacity) and sitting there basically at full charge without knowing it!

But Andy only tested one cell. Had he taken the 4 cells from a 4S bank, THREE of them might do the same - take days to reach full capacity. But one of them *might* not, and just hold back as if it was a bit lower. Depends on the cell. It's the knife edge.

Lower than 3.4v (like 3.375), the cell just cant achieve that full SOC chemically no matter how long you float (or do a conservative CV there.)

Put it this way: 3.4v used for anything is pure evil to LFP. Either go above or below depending on your needs.

 

[RandyP]

Even Andy at his offgrid garage when he was playing around with 3.4v / 3.5v with and without absorption noticed this. He did a video about it. But he's not aware of the knife-edge.

At 3.4v CV, (13.6v nominal for 12v batt) and allowed to absorb for days, he achieved a full SOC.

So the worry here is that anyone who thinks that 13.6v is ok to float at, might find themselves at a full SOC (full capacity) and sitting there basically at full charge without knowing it!

But Andy only tested one cell. Had he taken the 4 cells from a 4S bank, THREE of them might do the same - take days to reach full capacity. But one of them *might* not, and just hold back as if it was a bit lower. Depends on the cell. It's the knife edge.

Lower than 3.4v (like 3.375), the cell just cant achieve that full SOC chemically no matter how long you float (or do a conservative CV there.)

Put it this way: 3.4v used for anything is pure evil to LFP. Either go above or below depending on your needs.

Amazing. So scientific. Have you any test data to backup your words, or do you just have deep beliefe in your convictions ?

 

[Substrate]

Not belief my friend, but elemental battery basics for any chemistry. Get a pencil, and next time don't be late for class.

To FULLY charge a battery, there is a minimum voltage potential necessary to achieve it. In the case of a single LFP cell, that minimum is 3.4v. You can charge anywhere from 3.4 to 3.65 to achieve this, the only difference is the matter of time it takes. But quite often you don't see this printed on the case - only the maximum voltage.

But nothing in the world is perfect. If you skirt this minimum edge at 3.4v, some cells will charge to full if given enough time, and some may not. This goes the same for trying to balance here at a trigger level - same problem, because it is skirting the edge of the upper charge knee, where results can be variable.

Trust me, I'm not a measure-bator. This is real-world experience. The best part is you can do this kind of test yourself.

If you want to get insulting, well you know where I'm going to send you. Straight to the principal's office. Grow up. Or will you embed a doo-doo emoticon in your thread messages and epose your age?

 

[Goboatingnow]

Honorable mention - when running conservatively with a CV of 3.375v, when doing the initial "sanity-check" for balance at higher voltages, like 3.5v or higher, if you are within 100mv or less of delta, you are done.

Now you don't have to obsess over micro-sized balance deltas either.

Couldn’t agree more

Still wise to check for sanity once a year or so. 100mv or less delta? Fine, drop back down to 3.375v cv for cycling.

 

[justgary]

To FULLY charge a battery, there is a minimum voltage potential necessary to achieve it. In the case of a single LFP cell, that minimum is 3.4v. You can charge anywhere from 3.4 to 3.65 to achieve this, the only difference is the matter of time it takes. But quite often you don't see this printed on the case - only the maximum voltage.

I think that 3.4v still has a required end current to it. In other words, you can't charge at 3.4v indefinitely without causing damage. The end current is very low at 3.4v (e.g. ~0.005C), but still non-zero. The reason is that 3.375v is basically the full charge voltage for LFP chemistry. When charging above that voltage, you are utilizing the fact that the cells have a non-zero internal resistance. The resistance is the source of the "knee" seen when charging.

We really have no reason to attempt to stuff our cells full on a regular basis. Even floating at 3.375v is probably still too high depending on the actual resistance in your particular cells.

 

[justgary]

Let's state that again using math. Charging has to stop when the charge current (Ic) is less than Ic = (Vc-3.370)/Rb, where

Ic = Charge Current Limit​

Vc = Charge voltage​

Rb = Battery internal resistance​

Stated as a proportion of the cell's capacity, Ip = Ic/C, where

Ip = Charge Current per Capacity​

Ic = Charge Current Limit​

C = Rated Cell Capacity​

The reason charging has to stop is that the cells are full and can no longer move lithium. For my 230 Ah cells with an internal resistance of ~0.024 Ohms (24 milliOhms), the charge current limit looks like this at a few different voltages:

Vc​	Delta​	Ic​	Ip​
3.650v​	3.650-3.370 = 0.280v​	0.280/0.024 = 11.46A​	11.46/230 = 0.050C​
3.500v​	3.500-3.370 = 0.130v​	0.130/0.024 = 5.42A​	5.42/230 = 0.023C​
3.400v​	3.400-3.370 = 0.030v​	0.030/0.024 = 1.25A​	1.25/230 = 0.005C​
3.370v​	3.370-3.370 = 0.000v​	0.000/0.024 = 0.00A​	0.00/230 = 0.000C​

So, let's all say it once again: Once the charge current is below the calculated limit at your charge voltage, charging has to stop because the cells are full. Any impressed voltage after that *must* be at or below about 3.370v or damage will occur.

By the way, the first line above shows the cutoff at the factory specification, which is 0.050C at 3.65v. Since I have two batteries in parallel, I have to stop charging at twice the current shown above because each battery is taking half of the charge current.

 

[John Frum]

For my 230 Ah cells with an internal resistance of ~0.024 Ohms (24 milliOhms), the charge current limit looks like this at a few different voltages:

This spec sheet

EVE LF230 Product Specification

EVE LF230 Product Specification

diysolarforum.com

shows...
"ACImpedanceResistance ≤ 0.30mΩ"

I make that to be .0003 ohms, why is yours two orders of magnitude higher?

 

[Horsefly]

Not belief my friend, but elemental battery basics for any chemistry.

Just catching up on this. @Substrate, you've given us quite a bit of new info here. Thanks for that. Do you have some source material you can point to about this? Like maybe some academic material or write-ups from the manufacturers? I guess I'm looking for whatever you learned from, since this is a bit of a different take than what everyone here has been assuming to this point.

 

[justgary]

This spec sheet

EVE LF230 Product Specification

EVE LF230 Product Specification

diysolarforum.com

shows...
"ACImpedanceResistance ≤ 0.30mΩ"

I make that to be .0003 ohms, why is yours two orders of magnitude higher?

Never do math in public. Let me stew on this for a while. At least the numbers work out....

 

[justgary]

This spec sheet

EVE LF230 Product Specification

EVE LF230 Product Specification

diysolarforum.com

shows...
"ACImpedanceResistance ≤ 0.30mΩ"

I make that to be .0003 ohms, why is yours two orders of magnitude higher?

Because AC resistance and DC charging resistance are very different. I didn't actually measure the DC resistance of my cells (yet), but this paper suggests that one should expect DC resistance up to the tens of milliOhms in LFP cells.

 

[Goboatingnow]

Because AC resistance and DC charging resistance are very different. I didn't actually measure the DC resistance of my cells (yet), but this paper suggests that one should expect DC resistance up to the tens of milliOhms in LFP cells.

The internal resistance model for LiFePO4 is very complex and the models are not good either.

Resistance changes by charge rate , temperstute SOC , SOH. Hence picking figures “ out of the air “ for particular situations is rather meaningless
On fact On high fractional C to 2V absorption phase od very short and can be ignored. Simple voltage charge termination cutoff is a good a metric.

 

[justgary]

Keep in mind also that DC resistance is rather dynamic depending on SOC and charge/discharge current. What we are discussing here is the resistance that the cells exhibit when they are full and you attempt to keep charging them.

I can't directly support my use of 24 mΩ except that it does correlate well with Eve's recommended charge cutoff at 3.65v (0.05C).