lithium batteries, bulk float absorb time? end or return amps?

[Living Large]

So what are the differences between say solar and EV application?

{snip}Intercell connection resistances have absolutely no effect on capacity simple because the cells are wired in series. In a series, current in all cells is equal. It is impossible for the current to be different.

If I have a 8S2P configuration, does that change the argument? On a related topic, will each of my 8 pairs necessarily stay in balance? Seems like if one battery strays compared with the one it is paralleled with, I have a parasitic loss of sorts. Hmm, if pairs of 300 Ah used to make 600 Ah, and the two batteries in a pair differ by 0.05V, that would be about 0.05/.006 or 8+ A.

A 100 AH LFP has an Ri of less than 1 mill-ohm and ranges from .9 to .6 mill-ohms. The Ri scales up or down with capacity proportionally. The two most important characteristics of Ri are Voltage Sag/Rise under high Load/Discharge currents. The other is heating and power losses of heavy load or charge currents.

At 1C rates on any size LFP battery ranges in a Sag/Rise voltage up to .1 volts. That makes them extremely effective at very high C Rate applications. Not so important for solar, but extremely important for EV's when accelerating and regen braking where C Rates exceed 1C. The difference in how that is handled is in solar applications is a non issue on the charge or discharge side. In solar you only need simple trigrer tates for HVC and LVD. In a high current application you use the same voltage trigger set points, but add an element of time delay with special attention on the discharge side. Regen braking and charging on the EV side is a a non issue because charging is slow, and after you go a very short distance in an EV there is plenty of room at the top for the regen current to be absorbed.

Is that supposed to say "trigger points"? I believe the point here was for solar LFP apps with lower C rates, we don't need a time delay to allow for voltage sag. Just want to make sure I am on the right page.

There was a lot in this post to add to the library - thanks!

[Sunking]

If I have a 8S2P configuration, does that change the argument? On a related topic, will each of my 8 pairs necessarily stay in balance?

Doesn't change a thing because you do not wire parallel LFP cell like FLA. On LFP you parallel each cell in a Ladder configuration so it becomes one cell. It would be impossible for a cell to drift.

Is that supposed to say "trigger points"? I believe the point here was for solar LFP apps with lower C rates, we don't need a time delay to allow for voltage sag.

You got it right.

[Willy T]

3. In a Bottom Balanced we reference at 0% SOC of 2.5 vpc. On a 12 volt system is 10 volts. Not only do we know the SOC voltage, but we ALSO KNOW THE CAPACITY OF EACH CELL = 0 AH. That is completely different than Top Balance. We now now capacity and SOC are equal at the BOTTOM.

How often does a Prismatic Battery / Cell change in capacity because of cell fade or the failure of one or more of the many elements that make up the cell / battery. What does this change do to your bottom balance ? Can this cause or create a cascade effect ?

[Sunking]

How often does a Prismatic Battery / Cell change in capacity because of cell fade or the failure of one or more of the many elements that make up the cell / battery. What does this change do to your bottom balance ? Can this cause or create a cascade effect ?

As cell age, they loose capacity, but SOC voltages do not change.

Example let's say we have two 100 AH cells, and both brand new out of the box have 110 AH capacity. 5 years later they are down to say 90 AH capacity. At the bottom the voltage and capacity have not changed as voltage still = 2.5 amps and AH = 0. At the top the voltage is still the same but the capacity is now 90 AH. Bottom never changes, the top is a moving target in terms of capacity.

In operation you might have a cell that ages faster than the others. Thing to remember the pack capacity is determined by the weakest cell, and at the top the weakest cell will always be the highest voltage in a Bottom Balanced system. In a Bottom Balanced system all cells have the exact same amount of capacity which is determined by the weakest cell. At the top the voltages will not be equal, but are equal at the bottom.

What give batteries users fit is trying to apply Lead Acid techniques to lithium batteries. With lead acid you TOP BALANCE because anything less than 100% your batteries degrade. If you were to Bottom Balance a Pb cell, you would destry every cell except the weakest cell in th epack in a short period of time because every cell except the weakest cell would never see 100% SOC. They would be in a constant state of Partial State of Charge aka PSOC. That is the downfall of all lead acid batteries as they do not like being in PSOC.

Here is the irony of lead acid batteries. To maximize battery life, lead acid batteries must be kept at 100% SOC and never used. If used, must be a shallow discharge and recharged immediately. How does that work? LFP on the other hand works best in a PSOC environment. You apply a great amount of stress to LFP batteries taking them to 100% SOC. If stored, you only store them at 50 to 60% SOC.

Seeing the picture here? TOP BALANCED comes from Lead Acid battery mentality. Really it comes from all battery chemistry because all other batteries are chemical reactions. Lithium is not a chemical reaction, it is a electrical exchange of Ions. You don't treat Lithium like any other battery.

[Willy T]

SunKing, have you opened a prismatic cell / battery to see what they look like inside ? Seems like to me with your bottom balance you keep getting a diminished capacity as you have to lower the charged voltage to compensate for the lower capacity cells. like was said, it's everyones choice as to what they use. Me I like to watch the led's on the balance boards and they have worked ok without a failure.

[Sunking]

SunKing, have you opened a prismatic cell / battery to see what they look like inside ?

Yes. Shot one with a .22 and soaked it in salt water for a couple of days.

Seems like to me with your bottom balance you keep getting a diminished capacity as you have to lower the charged voltage to compensate for the lower capacity cells.

Who told you that you get diminished capacity from lowering cell voltages? That is only true for lead acid and all other battery chemistries, not lithium. Who ever told you that is stuck in a Pb box.

You get the same usable capacity from either method. Go top balanced and you get shorter battery cycle life.

[Living Large]

Doesn't change a thing because you do not wire parallel LFP cell like FLA. On LFP you parallel each cell in a Ladder configuration so it becomes one cell. It would be impossible for a cell to drift.
.

Wait a second. Maybe I am using the wrong terminology, but I did describe in words what I meant. 8 pairs of 6V 300 Ah cells. So each pair of cells is paralleled, then the 8 pairs are in series to make 48V.

My question has to do with any one of the pairs and the fact there is a non-zero internal resistance in each cell. Can an imbalance occur between the two paralleled cells, such that a current will flow from one into the other? Or will this effect naturally "self-balance" the two paralleled cells? I've never thought anything about internal battery resistance, obviously. My assumption may be way off - that you could even get to a point where they are imbalanced, because they are always tied together.

[Willy T]

Yes. Shot one with a .22 and soaked it in salt water for a couple of days.

Who told you that you get diminished capacity from lowering cell voltages? That is only true for lead acid and all other battery chemistries, not lithium. Who ever told you that is stuck in a Pb box.

You get the same usable capacity from either method. Go top balanced and you get shorter battery cycle life.

What was in it ? How many cell elements were in it ? Was it spiral wound or flat ?


So if you lower the charged voltage to compensate for the lower capacity cells that have a higher voltage, the pack as a whole doesn't have a lower overall capacity ?

[Sunking]

What was in it ? How many cell elements were in it ? Was it spiral wound or flat ?

Well Prismatic cells are not rolled like a jelly roll. The electrodes, insulators and liquid like electrolyte are stacked like sheets of paper in a sandwich like fashion.

So if you lower the charged voltage to compensate for the lower capacity cells that have a higher voltage, the pack as a whole doesn't have a lower overall capacity ?

Well stop and think about it for a minute. Let's use 100 AH cells as an example. You go buy say 100 cells, and measure the capacity of each. What you will find is what the manufactures claim of tolerance is correct of -0 +15% tolerance. So you measure and find 1 cell at 100 AH. All the rest are from 101 up to 115 AH. With me so far?

The design goal for LFP is 80% usable capacity, we bought 100 AH cells and expect to get 80 AH usable from the pack. Makes no difference if we Balance from the Top or Bottom we will get 80 AH. Still with me?

So now let's put the plan into action and all we have to talk about is two cells the lowest cell of 100 AH and highest of 115 AH. If we Top Balance the weakest cell has 100 AH, and the strong cells are 115 AH. The voltage of all the cells are equal but not the capacity right? Now we discharge and the voltage of the weakest cell is lower than all others. If we are not monitoring all cells and only use the pack voltage like all systems out there use we can completely drain the weakest cell to 0% and never know it because the other higher capacity cells still have 15% capacity left and there voltages are say 3 volts. So if that is a 16 cell 48 volt pack and have LVD set for 40 volts (16 cells x 2.5 volts). So now we have 1 cells at 0 volts, and 15 cells at 3 volts for a pack voltage of 15 x 3 = 45 volts we think everything is OK from an equipment POV. But in reality we just destroyed a cell. Why because we used the lead acid mentality.

So what if we change the reference to the Bottom of 2.5 volts per cell or 40 volts on a 16 cell 48 volt battery. At 2.5 volts and 40 volts = 0 AH capacity. We now charge the pack at a constant current and monitor all cell voltages. We terminate the charge when any cell reaches 3.5 volts or 90% SOC. The first cell to reach that voltage is going to be the weakest cell which is the 100 AH cell. All other cells will be a a slightly lower SOC voltage less than 3.5 volts. But guess what? Every cell still has an equal amount of capacity referenced to the weakest cell which is at 90 AH. Voltages are not equal but capacity is. So now we discharge and if we screw up all cells arrive at 2.5 volts at the same time and the lights go out, and all our batteries are safe and happy.

So what have we lost?

• Nothing in terms of capacity because the capacity is determined by the weakest link in the chain of series cells.

What have we gained?

• Well money to start with because we did not have to spend the big bucks on a factory BMS and all those Vampire Boards that over charge every cell every time you charge. All we need is a simple cell monitor to turn the charger off.
• We also gained significant battery cycle life (roughly 50%) because none of our cells never see 100% SOC.
• We eliminated the risk of over charging and over discharging.

[Willy T]

Sunking, I guess this is where we hit the fork in the road, good luck on the path you'v chosen. Time will tell the one that was best. I only have about 6 months on the top balance I chose and I'v never seen it have a need to balance yet as the BMS only goes to 3.45 V daily. The cells have stayed at .01 - .02 v difference, right where they started on my cheesy meter.

[PNjunction]

In a Bottom Balanced system all cells have the exact same amount of capacity which is determined by the weakest cell. At the top the voltages will not be equal, but are equal at the bottom.

Unfortunately, lifepo4 top/bottom balance threads usually get exciting no matter where you go.

Yes, bottom-balancing is technically the best, but not the only way to do things. It all depends on your lifestyle, risk, and willingness to do homework first. I think we can all stay friends no matter which way we choose to operate.

Bottom-balanced - the obvious choice for making sure all the cells reach the bottom at the same time. The other advantage not really mentioned is the fact that you get the charge over and done with relatively quickly! TIME spent "top-balancing" cells near 95-100% SOC adds up with degradation and eventual loss of capacity. This gentleman found out the hard way, especially with a very high 3.7v as his top-balance was set to!

http://www.technomadia.com/2015/02/l...attery-update/

At rates we commonly use for charging, say 0.25C or less, the time spent trying to reach 3.7v meant that if he was watching a current meter, the cells were trying to absorb down to zero amps, which is bad. The lithiating process is essentially over with by the time you reach .05C, and anything below that is so inefficient that all you are doing is heating the electroloyte, and increasing the size of the passivating layer. As this layer grows, it becomes harder to charge / discharge - and in some cases "sudden death" syndrome occurs when the last little open pores among the rest that are totally clogged, can no longer pass all the ions. (the best I can put it from a layman's standpoint ....) I'd love to be a student of Prof Jeffrey Dahn at Dalhousie, but I digress...

When the electrolyte heats after charging has finished, the voltage shoots up even further! I have witnessed this with my own experiments on my cells. Boring to watch, but exciting when that Fluke suddenly took off like a rocket AFTER the cell was already fully charged and the heating process really began about a half-hour later! But I am sooooo glad to have sat on the floor like a mindless robot next to the cell and witnessed it before I whipped the charger off. Basically, this is where all the secondary reactions which increase the passivating layer takes place.

Secondary to that, is that at 3.7v setpoint, the so-called "balancing", was only stopping the heating and not the charge / balance process! In other words, no balancing was taking place that high up when electrolyte heating was the determining factor for voltage, NOT the SOC, which was already 100%!

I guess what I'm trying to say is that IF you are going to do a so-called "top balance" - which I do on a normal basis for convenience and get away with it, keep an eye on your current. If your vampire bleed off boards are still blinking when the overall current is .05C, your balancing may be for naught as you are now overcharging and heating electrolyte. That's one reason I don't go up to 3.6v - mostly 3.5v max at my low charge rates. I don't want to spend a lot of time at the top end of charging, which when accumulated over time, can lead to loss of capacity. That, and of course along with heat, but my bet is that the guy above just plated his cells prematurely.

Remember though - I'm the 12v guy that uses NO vampire boards - just an initial sanity balance, a charger/controller set to 3.5v/cell max, and an LVD. Rare.

Either way works - bottom or top balance - just be conservative with top balance!

[Willy T]

In other words – our batteries seem to be aging way faster than they should – now giving us just 75% of their original rated capacity.

This is where it begins to get interesting. At 75% you begin to move away from your design capacity. The question would be, can you augment the capacity by adding new cells to get it back or would the charging of the older cells prohibit this mixing ??

Generally with FLA you can build in some extra capacity, with LiFePo4 it's high dollar to do.

We’ve spent the past year using 14.2V bulk charging, and a 13.55V float. I’ll probably soon change to even lower – 14.0V bulk and 13.4V or even 13.2V float.

This is just using off the shelf settings for flooded batteries, probably not smart, especially in a RV where the use is intermittent.

[Sunking]

Another EV Lithium Battery Bites the Dust

Well today another EV lithium battery bit the dust. Friend of mine top balances and over discharged a cell destroying it. Read all about here. He is now converted to Bottom Balance and understands the risk of Top Balance.

[Living Large]

Well today another EV lithium battery bit the dust. Friend of mine top balances and over discharged a cell destroying it. Read all about here. He is now converted to Bottom Balance and understands the risk of Top Balance.

I read the thread, and while your friend now seems to understand what happened in this case, I am wondering - is he really converted?

[PNjunction]

This is where it begins to get interesting. At 75% you begin to move away from your design capacity. The question would be, can you augment the capacity by adding new cells to get it back or would the charging of the older cells prohibit this mixing ??

Well, he was overcharging by having an early balancing system set too high at 3.7v. Conservatism pays off - in fact when I do my power budget, I de-rate the cells to only 80% of the stated capacity to begin with when figuring that out. Why? Most of the stated capacities are at 100% DOD, something one should never plan to go to in the first place. Derating to only 80% of stated fits like a glove with the commonly accepted 80% DOD threshold.

Yes, you can "marry" in a good cell for a bad one, but just like doing so with lead, you now have to be super-vigilant and keep track of it, and what that may do to overall performance as the pack itself ages having new and old installed.

This is just using off the shelf settings for flooded batteries, probably not smart, especially in a RV where the use is intermittent.

A big mistake with lifepo4 is thinking along the Pb lines, or the "drop in" mentality that salesman like to pitch.

If we look at it from the typical 12v standpoint, the fellow in the article was doing the near equivalent of charging a 12v lifepo4 battery to 14.8v (3.7v per cell for a 4-cell lifepo4). BAD! That is near equivalent to using an AGM setting to charge your lifepo4 with. Not good. If one is going to wing-it with lead-based chargers, then by all means try to use the GEL settings, which are near 14.1v. Keep wallet open.

Anyway, hats off to that guy for being a pioneer, sharing his experiences, and learning from them.