Lithium battery charge profiles

[Westbranch]

Hi North of 56, I guess I am back to ''why can't you just set the CC to your safe values'' and let 'er rip with a WhizBang to control the cutoff with float at 3.32V?

As in Karraks example numbers, Charge at 3.45V (or less) and go to Float (3.32V) when Amps into the battery drop below C/20 (5%)

[North of 56]

Hi North of 56, I guess I am back to ''why can't you just set the CC to your safe values'' and let 'er rip with a WhizBang to control the cutoff with float at 3.32V?

As in Karraks example numbers, Charge at 3.45V (or less) and go to Float (3.32V) when Amps into the battery drop below C/20 (5%)

I guess one could do that, but the BMS is there for a reason, to protect your valuable investment. And the LiFePo4 cells can go out of balance, as you are dealing with gradual cell degradation over time.

The way I see it is, you want to charge the cells to about 90% state of charge on a regular basis, to maximize usage of the storage capacity of the bank. At that charge state, you are right in the neighborhood of cell degradation. If you rely on just pack voltage an even slight cell imbalance could put one or more cells into that 90 to 100% range, even though pack voltage may indicate otherwise. That's where the BMS comes into play. As soon as one cell goes above the 90% level, the charging is triggered to go to a lower float voltage.

The imbalance will grow worse over time if you just base the charging on the pack voltage, and a lower capacity cell, is continually reaching a higher state of charge. THe degradation will be most pronounced with that cell, and it's capacity reduced more than the higher capacity cells over time. That's also why I think it's a good reason to top balance (as opposed to bottom balancing). All the cells will reach the same state of charge and cell degradation will be more or less equal over time. Cell degradation is more prevalent at a higher state of charge, than at a lower one. Of course, you don't want to go too low either. Does that make sense?

[Westbranch]

Yes and to make sure you don't go over the top so to speak, I agree that a BMS is essential otherwise you end up like WRT with a .... dead , verrrry dead LiFePo4...

battery.http://midniteforum.com/index.php?topic=2270.0

so you use the CC to keep your Volts low and the BMS is the safety valve just in case ...

[North of 56]

Yes and to make sure you don't go over the top so to speak, I agree that a BMS is essential otherwise you end up like WRT with a .... dead , verrrry dead LiFePo4...

battery.http://midniteforum.com/index.php?topic=2270.0

so you use the CC to keep your Volts low and the BMS is the safety valve just in case ...

Thanks for the link Westbranch. That was an interesting read and unfortunate for the OP to lose his battery pack. Looks like having an extra back up just in case, would not be a bad idea!

About keeping the volts low with CC, I still think that you want to maximize storage, and get up to about a 90% state of charge. The issue I see is that 90 to 100 % state is an area you want to avoid, and reaching that with one or more cells may be a more frequent event than one would hope for. The other question I have is how well can you judge the charge state of each cell, with varying charge currents? I'm thinking that a higher charge current will skew the voltage higher, whereas a very low charge current will allow the charge state to climb higher, perhaps than interpreted?

[Westbranch]

I just picked up on a missed fact.  see the graph on discharge mentioned above.  Note the NOMINAL VOLTAGE!   = 3.25V
Now think about a FLA battery, its Nominal V is 12 yet we know it is able to take 13.2V to get to 100% charge..  This makes me think there is a lot of head room for topping off a Life battery.  Note the

3.5 Vdc Bulk Charge | 3.6 Vdc Absorption Charge | 3.3125 Vdc Float Charge
28.0 Vdc Bulk Charge | 28.8 Vdc Absorption Charge | 26.5 Vdc Float Charge

very much like my AGMs...

[North of 56]

So a question then. Do you think it's possible to fairly accurately predict the charge state of an LiFePo4 cell (or pack) by the voltage, when your charge source is not constant?

I know that with FLA batteries, you may have an idea of the charge state when charging, but you would also look at the amount of current going in at the time. Even then you wouldn't be certain? The voltage can be skewed up or down considerably, depending on the solar production. I know that LiFepo4 voltage outputs fluctuate to a lesser degree than FLA, but there is still quite a range from resting (ie 0 amps in) to bulk, the posted charge rate.

I'm thinking that getting a bank to a 90% charge state may not be so cut and dried, as you're dealing with varying charge currents with solar power? The voltages you referenced are based on a constant charge rate.

[Cniemand]

I haven't had any issues with my LiFePo4 (Calb CA) battery bank without BMS management. It has functioned the same since day one of having its bottom balance amongst the 16 cells I have in series. (48v setup)

Frankly. I do not see the need for them. The cells all stay balanced within 15mv of each other. This is using 2 years of Off-Grid, full time household usage of them.

// MODIFIED to reflect 15mv //

[inMichigan]

Being new to solar and battery systems, I sure appreciate this forum as a source of information and debate as we designed our grid-tie hybrid system.
inMichigan

[Cniemand]

Hello everyone!

Due to LIFEPO4 chemistry having two different stark contrasting camps regarding as best setups I wanted to add my two cents. It has been slightly over two years now since I installed a battery bank consisting of 16 - 100aH CALB CA series cells. A theoretical bank of just over 5 KWH. The paperwork has it listed that if you only routinely use 80%DOD you are supposed to get a 2000+ cycle-life. 

In my setup, I purchased a couple acres in Colorado in the beautiful Rocky Mountains and have been powering a Travel Trailer/Tiny House. I have 2000 watts worth of Solar coming in and a 200v Midnite Solar C.C to regulate it all. I do not have the Whizbang JR. However, I plan on getting it.

When I first got the cells I opted to bottom balance them. The idea was that even though you purchase a set of 100aH cells from the same batch, it does not mean they all are exactly 100aH. They very well could be broken down as such: 100, 102, 105, 97 etc. I have heard it described as having a bunch of sticks to varying lengths. (Placing one end of them against a surface so you can see the "jagged" variance at the top)

We all know for longevity with any battery chemistry that you get the most cycles from not fully utilizing the battery capacity. IMHO I went with the idea that through bottom balancing I would be able to charge all the cells up to a point where they are all at, say, 90aH with no variance. Removing the runaway at the top. I felt that in a real world situation it is more likely that a bank will be accidentally over-discharged rather than over-charged. Through cells that match you would have very little chance the stronger cells would kill the weaker cells in that condition.

Two years ago I pulled all the cells down to 2.700v and managed to get all the cells balanced within 5mv at that level. I have the CC set to charge them back up to 3.5vpc.

NO Equalize. Absorb = 3hr BULK=3.5vpc FLOAT= 3.34vpc REBULK=2.8vpc ENDAMPS=8

Using a multimeter on each cell when in Absorb mode (And nearing the END AMPS) after two years of daily cycle-real world usage I see them still within 15mv of each other at their upper charge states (where they would start to deviate from each other).

I think so far that is fantastic to see!

As far as ideas concerning whether you should use FLOAT for LIFE cells, I have not had any problems with it. For myself I chose a voltage that closely matches their resting voltage. Something that over the last two years has shown to not slowly increase their voltages when I am not around to monitor. The cells would always have a small load on them to take off a bit of charge. The CC and Inverter.

What about ENDAMPS? According to spec sheets 0.05C is supposed to be where you consider the cell full. In my case of 100aH cells that would be 5 amps. However, because I do not have a Whizbang Jr, I find that if I use 5amps the voltages on the batteries shoot up further than I like. This is because there is always a couple amps worth of background load in the morning before it is fully charged. As the CC cannot tell that by itself it ends up putting in more aH to the bank than I want. To fix that I just up the ENDAMPS and the voltages stay where I want them.

- Cloud

[CDN-VT]

Cloud , How is it you DON't Have a WB-jr , but all that $ in storage . You can have one , but leave it unhooked & / or set it off .
You do need the shunt also ..
Im not getting this thinking . I think there 50pluks and really programmable , but them new style potatoes with nails are WAY more Pluks !?
VT

[Cniemand]

Haha @ CDN-VT :

I can easily see your point! I have $2500 USD in 5 kwh worth of storage, but no $80 USD WB-JR? The WB came out after I got everything setup. Everything has been running smoothly that I guess I sort of figured out about where I should take the pack down to voltage wise. I typically only use 2 KWH a day unless I make a point to use a space heater. 

I WILL get one shortly though as the idea of knowing a much more precise number of watts going in and out makes for being able to fully utilize what I purchased. The other main idea about having one is to be able to set the ENDAMPs to 5 rather than 8.

I'll get right on that! 

[ClassicCrazy]

I see the new Morningstar controller is advertising it charges lithium batteries but I could not find one mention of anything lithium in the owners manual. It does no more than the Midnite controllers do and seems to be just their version of the Kid .
http://www.morningstarcorp.com/products/prostar-mppt/

Larry

[Westbranch]

From reading the email I got it looks like it is a KID minus ~5 amps, ....didn't dig too far...

[ClassicCrazy]

What features do you think would have to be added to the Classic to make it the perfect lithium battery charger ? 
I know for my LiFePo4 ebike battery the critical features are that it go to top voltage but at that point it needs to be able to turn on and off in order for the BMS to do it's thing for balancing. So I guess that means it goes from constant current to constant voltage at certain point.

Larry

[Westbranch]

What features do you think would have to be added to the Classic to make it the perfect lithium battery charger ? 
So I guess that means it goes from constant current to constant voltage at certain point.

That sounds like the transition point from BULK to ABSORB to me........??