LiFePO4 - The future for off-grid battery banks?

[steveg]

Sorry Steve, you can't read accurate cell balance with the cells under 3.45v and the charge rate less than 10 amps, anything below full voltage except for fully discharged voltage is relatively meaningless accept to show a failing cell because the difference will be big. Fully discharged (2.8v or under) is used by a section of the EV people because it gives a more accurate out of capacity reading, saves a cell being driven into reverse charge and killing it.
As far as needing a full time active BMS on an off grid system, why? Monitor the cell voltages with a Jusi Cell Logger 8 does 8 cells and cost about $28, it can record data files you later view on your computer so you can see just how the batteries going and make minor balance adjustments if needed. I have set mine to 30 sec readings and I can record a weeks battery activity at a time. It has an alarm port that can be configured for dump loads to bring high cells down or turn off the charge till the high cell drops back. The dump load system seems to work better because it self balances the cells very quickly and they don't seem to wander much after that.
Here is a sample graph showing 4 cells of a 12v nom. 720Ah battery bank being recharged daily by a 2 kw solar bank.[ATTACH=CONFIG]1613[/ATTACH]

T1 Terry

Huh?

Voltage is not meaningless as this is the real voltage of the cell (at the cell terminals with my very expensive DVM) at that charge state - I am an electronics engineer by trade and have more than a laymans idea of how to read a battery. You are referring to internal impedance under charging/load I am guessing - if my full charge rate is 60A (and no higher in any condition) then this is the true variation I will get as of now 70mV maybe later as it ages I will get a larger result between cells - with a large load I get similar results from what I have seen so there is no 'hidden' condition with my cells that I am aware of.

Why you so against me having a BMS? (that came with my cells in any case) - why would I now go off and purchase a logger etc to replace something that is working now (yes I have tested it in overcharge and discharge modes and cell balancing mode)

Dump loads from battery?? You either have a shunt across the cells that are higher (usually under micro control) or you only push charge to the ones that are lower (very expensive flyback converters for each cell) ... a shunt across *all* cells pulls current from the entire stack, and ages the battery un-necessarily I would think? and if it is up to charge then no current flows into the cell because it is the same potential as the regulator output, in my case 56.4V

Why you so against me having an BMS? It *does work* you know - and you not having one works well for you - great news, but I dont want to obsess about my batteries I want to forget about them for months at a time.

A BMS monitors every individual cell on charge and discharge, cutting the load/charge if any cells goes beyond 2.7/3.65 - I dont see a down side - I am only drawing a MAX of 150A that's 0.5C per cell and charge is 60A that's 0.2C per cell.

You cant compare home use and EV - EV draws much more current, and hence exaggerates any cell to cell variation - 20kW loads for periods of time and is very hard on cells, and re-gen braking moreso, so the brand (cell construction consistency) and type of cell and BMS is more important in this use case.

Regards,

Steve.

[steveg]

How does a dump load take excess amps out of only a few cells ??

It doesnt, it takes current out of all cells including the ones that are too high/low ... reducing the head/foot room until balanced.

[T1 Terry]

Lithium iron cells have a 100% full point, 3.45v, any voltage higher than that is 100% full, just wasted energy that eventually heats the electrolyte and damages the cell. By balancing the load to being slightly higher than the charge rate the excess voltage is burnt off the high cell using up the wasted energy used to get the voltage up there but none of the other cells have any issues with wasted energy, it is all used for charging. Just a hypothetical example, if 0.5Ah is needed to get a cell from 3.45v to 3.75v as soon as the load is applied that 0.5Ah would appear as though it was never there, the voltage on the high cell will immediately drop to below 3.45v. Each time this cycle happens the lower cells get 0.5Ah closer to fully charged.
If the boost voltage was set high enough eventually all the cells would reach above 3.45v together but it's not really necessary to pull all the cells up that high every recharge, they are not damaged at all if they never reach fully charged. Maintained high cell voltage (above 3.5v) and below 2.0v are the thresholds that damage lithium iron cells, they stay healthy between these two limits.
The only reason to bring them all as close to 3.45v as possible at the same time is to know where the 100% charged point is, that

[russ]

I dont want to obsess about my batteries I want to forget about them for months at a time.

Regards,

Steve.

That is the case with maybe 95% of users - they want to use and not have to tinker.

Russ

[steveg]

[QUOTE=T1 Terry;38292]Lithium iron cells have a 100% full point, 3.45v, any voltage higher than that is 100% full, just wasted energy that eventually heats the electrolyte and damages the cell. By balancing the load to being slightly higher than the charge rate the excess voltage is burnt off the high cell using up the wasted energy used to get the voltage up there but none of the other cells have any issues with wasted energy, it is all used for charging. Just a hypothetical example, if 0.5Ah is needed to get a cell from 3.45v to 3.75v as soon as the load is applied that 0.5Ah would appear as though it was never there, the voltage on the high cell will immediately drop to below 3.45v. Each time this cycle happens the lower cells get 0.5Ah closer to fully charged.
If the boost voltage was set high enough eventually all the cells would reach above 3.45v together but it's not really necessary to pull all the cells up that high every recharge, they are not damaged at all if they never reach fully charged. Maintained high cell voltage (above 3.5v) and below 2.0v are the thresholds that damage lithium iron cells, they stay healthy between these two limits.
The only reason to bring them all as close to 3.45v as possible at the same time is to know where the 100% charged point is, that

[T1 Terry]

Hi Steve,
The cells really are full at 4.45v, well, 3.40v actually but the 0.05v over voltage is required to charge and 0.05v under terminal voltage is required to discharge.
Here is a charging graph that shows just how fast the voltage climb is once they are full at 3.45v and how little energy is there at any higher voltage. If you are reaching 3.5v on every cell at trickle charging rates the battery bank is 100% full.
The boost charge voltage was raised to 15v for a nom 12v, sorry but I don't have the amps graph to go with it, it's off powering the ether or something because buggered if I can find it.

T1 Terry

[T1 Terry]

Hi Steve,
The cells really are full at 4.45v, well, 3.40v actually but the 0.05v over voltage is required to charge and 0.05v under terminal voltage is required to discharge.
Here is a charging graph that shows just how fast the voltage climb is once they are full at 3.45v and how little energy is there at any higher voltage. If you are reaching 3.5v on every cell at trickle charging rates the battery bank is 100% full.
The boost charge voltage was raised to 15v for a nom 12v, sorry but I don't have the amps graph to go with it, it's off powering the ether or something because buggered if I can find it.

T1 Terry
EDT: Here is another graph showing the charging curve.

[Northernflyer]

Very interested in lithium battery setup for a new 6.5kw system

I am working on a 6. 5 kw system and are considering lithium nmc or iron phosphate batteries. I am currently doing research and have found Corvus energy that offers these for actual solar systems. I am waiting for more info and pricing from the manufacturer. Will get back when I hear more.

[T1 Terry]

I am working on a 6. 5 kw system and are considering lithium nmc or iron phosphate batteries. I am currently doing research and have found Corvus energy that offers these for actual solar systems. I am waiting for more info and pricing from the manufacturer. Will get back when I hear more.

Do you have any info/specs on the lithium NMC cells? What is their stability like? The LiFeP04 chemistry is quite stable, the newer lithium sulphur batteries are showing good stability but life cycles are a bit of an issue. The stability of the other chemistries at each end of the charge cycle isn't real good so it will be interesting to see reports about the lithium NMC construction.

T1 Terry

[Northernflyer]

Lithium nmc

If you go to www.corvus-energy.com they have a good amount of info on these batteries. They are in the same neighborhood for safety as the lithium iron phosphate and seem to be better in all regards such as having a much higher cycle life and a 99% efficiency. I would imagine they are quite expensive.

Steve

[T1 Terry]

Thanks for the link Steve. after reading through heaps of spin I found this bit

The Technology

Corvus utilizes advanced lithium ion polymer cells in multi-cell module configurations. These configurations are monitored by a proprietary battery management system to control charge, discharge and monitor all aspects of the module, pack and array.

Lithium Polymer is what the model aircraft batteries are made of, they have special bags for them when recharging to contain the fire if the explode into flames. this is an extremely dangerous form of lithium ion battery and these units must rely on the battery management systems never stuffing up. Anyone who has used a computer knows that microprocessor control can hang or shut down for no real reason, just a glitch or spike or piece of bad information in a packet can cause it.
I sure wouldn't be risking them, the cell type certainly are not in the nieghbourhood of safety compared to lithium iron phosphate, with out an active and fully functioning BMS they are time bombs.

T1 Terry

[Northernflyer]

NMC Batteries

I have another link from battery university on the different lithium batteries and the nmc ones are grouped with the iron phosphate ones for safety. I am still researching but they appear to be one of the better types to use. http://batteryuniversity.com/learn/a...er_lithium_ion

Steve

[Sunking]

Lithium Polymer is what the model aircraft batteries are made of,

For model aircraft energy density is the most important factor in such devices because of the weight issue.

For solar, energy density is not important, price per Watt Hour and safety are the most important factors. For that reason LFP is the best choice as it is the safest of the lithium chemistry and the least expensive as prices have dropped to around $0.50/wh for chi-com units.

[markfoy]

Hi all,
Just to appreciate steve and others. I am learning a lot on this thread

[buchhla]

This is all great info. I am thinking of doing a very similar install to Steve on my boat which is currently in Auckland, though I will only have about half the solar capacity...