Lithium battery charge profiles

[lestef]

I'm from Europe, and here by far the best price and availability have this guys: http://www.ev-power.eu . The problem is that they have both manufacturers at about the same price for the cells and is difficult to choose... I think i will go with Winstons, for start: 16 x 100 Ah battery bank...
Now the BMS, any recomandations ?
I found this: http://www.ev-power.eu/BMS123-System/BMS123-BMS-Main-Controller.html ,the complete system is about 1000 $
Is enough to program the charger and the inverter's limits or I need to search for a BMS that have the output relays to connect or disconnect them, like this model: http://www.ev-power.eu/BMS123-System/BMS123-OFFGRID-BMS-Main-Controller.html

[vtmaps]

I'm from Europe, and here by far the best price and availability have this guys: http://www.ev-power.eu . The problem is that they have both manufacturers at about the same price for the cells and is difficult to choose... I think i will go with Winstons, for start: 16 x 100 Ah battery bank...
Now the BMS, any recomandations ?
I found this: http://www.ev-power.eu/BMS123-System/BMS123-BMS-Main-Controller.html ,the complete system is about 1000 $
Is enough to program the charger and the inverter's limits or I need to search for a BMS that have the output relays to connect or disconnect them, like this model: http://www.ev-power.eu/BMS123-System/BMS123-OFFGRID-BMS-Main-Controller.html

I know very little knowledge of lithium batteries, but I would suggest you head over to the NAWS forum and search for member "pnjunction'.  He has written a lot about LiFePO4 batteries in RE systems.  One of the things I learn from his posts is that a RE system is very different from an EV system in terms of the need for a BMS.  In a RE system you are not handling the extreme high currents that occur in a EV battery.  Also, if you get balanced, large prismatic cells, and stay within the "knees" of the charge curve, you do not need a BMS.

--vtMaps

[mahendra]

think he is right, you would probably only need a cell protection circuit . just in case .

[lasolargroup]

Given lithium battery's are now a viable cost effective option. Also now a practical option due to large cells format with equivalent capacity to the led acid cells on offer.

Are there any future plans for the classic range to offer a charge profile for lithium battery's?

I have had a very possessive experience for many years using lithium battery's to power my electric bikes, camper van and many other devices that were previously led acid. When my led acid battery's that power my off grid home are dead there is no question about them being replaced with lithium.

As it stands now before using my classic to charge by house battery's. I  experimented with it for a few days charging a small 40ah 12v lithium pack with a small 120w PV and its possible to tweak the led acid profiles to work with the lithium though a dedicated lithium profile would be much better.

Kurt.

Hi,
Before reading this post i was really not aware to all this. you really give us a brilliant information... great post

Link removed as it resembled advertising

[Westbranch]

Here is some not so good news about a 'memory effect' in LiFePo4 batteries, especially in partial state of charge situations.

from 2013...

http://phys.org/news/2013-04-memory-effect-lithium-ion-batteries.html

Add: Bold text

[North of 56]

The Classic does not need a separate LiFePo charging profile.   Batteries with the BMS (which is necessary anyway) mean that it already works for these.  You will have to change the Absorb time or maybe use the Aux 2 input force to stop charging function from certain BMS's.

boB

Is it possible to use the HVD trigger from a BMS to force the Midnite Classic into Float (ie at a lower voltage) rather than stop charging? IMO that would be a better option.

[Westbranch]

North of 56, what is it you want to do/achieve from Float charging?

It is important to achieve the  float stage but not so much the charge that happens after getting to float, over a long time period, and not much is gained from what I have read in an RE setup... very similar to FLA Float stage ie 95 to 100%

[North of 56]

North of 56, what is it you want to do/achieve from Float charging?

It is important to achieve the  float stage but not so much the charge that happens after getting to float, over a long time period, and not much is gained from what I have read in an RE setup... very similar to FLA Float stage ie 95 to 100%

I did mention to drop down to a float charge, and one that is similar to the resting voltage of the pack. That way, if your LFP batteries are charged up to your desired level, say 85 or 90% state of charge, you won't discharge the battery pack with any further loads (they will instead come from the excess power from your solar, etc..)

The current option is to have the BMS signal (using the HVD if triggered) will shut down the charging function. This will likely have the charging function alternating between off and on as the HVD gets triggered every time the charging function brings up the voltage of the cell. This would have it hovering around the HVD trigger point. I have read that once an LiFePo4 cell is charged, it's best to drop down the voltage to a "float voltage", which will help prevent cell degradation over time.

Another option using the HVD signal and AUX2 line, would be to latch it off, until it's reset again, but that could mean that your battery pack may be discharged significantly, it there isn't someone around at the time.

I realize that an HVD trigger event should be rare if you design your system with safeguards in place, but it could still happen. Particularly if cells are or become out of balance, in combination with various charging currents as well.

[Westbranch]

I think I follow your idea,

you won't discharge the battery pack with any further loads (they will instead come from the excess power from your solar, etc..)

That will work for light loads ie radio, internet etc... I call it 'The Dance of Opportunity Power', just enough , not too little, not too much...

But what happens when your PV can NOT support the load, eg a fridge startup?  the battery has to/will supply the demand not met by the PV, especially the amps needed, putting your Classic back into Absorb or maybe even Bulk.. No?
As I see it, Setting the Float V (too) Low will cause you to draw more power out of the battery and thus cause unintended results...

[North of 56]

I think I follow your idea,

you won't discharge the battery pack with any further loads (they will instead come from the excess power from your solar, etc..)

That will work for light loads ie radio, internet etc... I call it 'The Dance of Opportunity Power', just enough , not too little, not too much...

But what happens when your PV can NOT support the load, eg a fridge startup?  the battery has to/will supply the demand not met by the PV, especially the amps needed, putting your Classic back into Absorb or maybe even Bulk.. No?
As I see it, Setting the Float V (too) Low will cause you to draw more power out of the battery and thus cause unintended results...

With LiFeP04 batteries, the voltage after charging will drop down to a resting voltage, and that is what I was referring to. You won't see much voltage sag with LFP, plus you can set the rebulk voltage  point with the Classic.

Actually, what I was getting at was a proper way to handle a HVD from an LiFePo4 battery bms. Turning off the charging function of the Classic, IMO, is not the best way to handle this. Would it not work better if one could force the Classic into float? Is that possible?

[Westbranch]

I see what you are want to do but I don't think I would call it, ''forcing'' , it only does what you tell it to at any given time or parameter met... eg a min Absorb of 5 minutes and a max of 1 hour or an End Amps setting. It's sort of like,  "Float" is a state of mind...

Would it not work better if one could force the Classic into float? Is that possible?

So, if you have a WBjr and set the end amps appropriately,  not sure of what number you would need/use for LiFe, wouldn't that achieve what you want? 
Reason I ask  this is that the discharge curve here looks quite flat. See page 5, http://www.balqon.com/wp-content/uploads/2013/07/35_35balqon_battery_2013.pdf and the table above it indicates the max charging rate , like <3C for cells up to 1000Ah so do you say 2C to stop??

[North of 56]

I see what you are want to do but I don't think I would call it, ''forcing'' , it only does what you tell it to at any given time or parameter met... eg a min Absorb of 5 minutes and a max of 1 hour or an End Amps setting. It's sort of like,  "Float" is a state of mind...

Would it not work better if one could force the Classic into float? Is that possible?

So, if you have a WBjr and set the end amps appropriately,  not sure of what number you would need/use for LiFe, wouldn't that achieve what you want? 
Reason I ask  this is that the discharge curve here looks quite flat. See page 5, http://www.balqon.com/wp-content/uploads/2013/07/35_35balqon_battery_2013.pdf and the table above it indicates the max charging rate , like <3C for cells up to 1000Ah so do you say 2C to stop??

This has more to do with protecting from a cell imbalance where one (or more) cells are about to be overcharged. The BMS is there to signal an alarm of a HVD in order to shut down the charging source, when that cell(s) voltage climbs above a preset level.

If you use the AUX2 input to shut off the charging source, that would work temporarily, but as soon as the cell's voltage drops down again, the HVD signal will disappear (unless latched on) and the charging will resume, bringing the cell back to HVD again. This could continue, keeping the cell at the threshold of HVD for quite some time?

Alternatively, one could latch on the HVD signal, but then you would lose the charging from solar, and thus the battery pack will begin to be depleted, until the HVD is manually reset. You could have it automatically reset at a preset lower voltage. The best solution though, would be to "force", ie signal, the Classic into float mode. In float mode, the charge controller will keep the battery pack at it's lower resting voltage, and the charger will still work to supply solar power to any loads that come on. Is it clearer now?

[Westbranch]

Not at all.... I must be heavy headed today... here are some specs from a Balqon 18KwH battery
http://www.balqon.com/online-store/#!/Lithium-Battery-Storage-18-kwhr-1-week-Lead-Time/p/34642671/category=2860254

Charge Voltage: 28.0 Vdc Bulk Charge | 28.8 Vdc Absorption Charge | 26.5 Vdc Float Charge
Discharge Voltage: 24 Vdc Max

which uses 8 cells @  700Ah cell  @ 3.25V/cell
= / cell

Charge Voltage: 3.5 Vdc Bulk Charge | 3.6 Vdc Absorption Charge | 3.3125 Vdc Float Charge. Discharge Voltage: 3 Vdc Max

So using these real life numbers, what values would you expect to use in a charge regime?

[North of 56]

Not at all.... I must be heavy headed today... here are some specs from a Balqon 18KwH battery
http://www.balqon.com/online-store/#!/Lithium-Battery-Storage-18-kwhr-1-week-Lead-Time/p/34642671/category=2860254

Charge Voltage: 28.0 Vdc Bulk Charge | 28.8 Vdc Absorption Charge | 26.5 Vdc Float Charge
Discharge Voltage: 24 Vdc Max

which uses 8 cells @  700Ah cell  @ 3.25V/cell
= / cell

Charge Voltage: 3.5 Vdc Bulk Charge | 3.6 Vdc Absorption Charge | 3.3125 Vdc Float Charge. Discharge Voltage: 3 Vdc Max

So using these real life numbers, what values would you expect to use in a charge regime?

A voltage that corresponds roughly to a 90% or slightly more state of charge is one I would use to trigger a HVD, to prevent overcharging a cell. I would not use the BMS that you referenced, as they  would likely bring the state of charge up closer to 100%, which will impact the lifespan of the battery pack. I think that the trigger voltage for HVD will vary somewhat, depending on the rate of charge at the time. Ideally, you would want the pack go to float before an HVD is triggered. Again that voltage may be dependent on the rate of charge, and a charging regime specific to the type of LiFePo4 cell would be ideal.

[North of 56]

So using these real life numbers, what values would you expect to use in a charge regime?

Westbranch, you may want to check out this link if interested in charging voltages. A member there (karrak in post #9) has posted voltages he uses to manage his Winston cells. Looks like he occasionally goes to 3.45 volts per cell, which he figures corresponds to about a 95% state of charge, and then drops down to float at about 3.32 volts per cell. He rebulks at about 3.27 volts.

http://forum.solar-electric.com/forum/solar-electric-power-wind-power-balance-of-system/off-grid-solar-battery-systems/20447-charge-float-and-cutoff-voltages-for-large-lifepo4-cells?21701-Charge-Float-and-Cutoff-Voltages-for-Large-LiFePO4-Cells=