[RE-wrenches] Lithium Iron Phosphate

Tue Aug 18 07:45:48 PDT 2015

Hello Daniel,

With such an expensive battery, I would not recommend trying to use the controllers you mentioned. The voltages are wrong and, more importantly, temperature compensation and automatic EQ will damage your battery. To date, there are no PV solar charge controllers on the market that provide the ideal profile for LFP battery charging. But, you can get close with some. We use Blue Sky Energy and Midnite Classics for all our Lithium-ion systems. No need for you to invent something new. 

I might be offering TMI here but, you asked. When you reduce the charge voltage to less than the potential voltage of the bank, you will begin discharging. As soon as there is an equilibrium of voltage, your controller will maintain a CV float charge on the battery as well as provide power to loads. Of course, this is how all battery charging works but something different happens with LFP batteries.

With this float voltage, you will see an oscillation of charging and discharging because LFP batteries have such a narrow voltage range compared to a charge controllers regulation. Just 0.05 volts will make the difference here. This can begin to unbalance the bank because efficiency of the cells is not exact; some absorbing, some not based on voltage. I have studied this but it's not a problem since all the all the systems we have designed use PV solar power. During the next charge cycle, the battery voltage will rise above the balancer voltage (our systems use 3.55V/cell) and bring up the lower cells.  For our systems, I program a CV (3.6V/cell as I use LFMP batteries) for about 20 to 30 minutes. This is normally long enough to allow any higher capacity cells to catch up to the lower ones but due to the low voltage (LFMP can be charged up to 3.8V/cell) no cells are 100% charged. In fact, the charge parameters are specifically set to prevent 100% charge to provide optimum life.   

You said when you hit your target voltage (3.65/cell) you are stopping the charge. At that point the LFP battery is not fully charged and that is good for the life of the cells. However, unless your balancers are starting at a much lower voltage, they won’t have any time to do their job. This is because charging LFP batteries you are always in the bulk mode at a voltage much lower than your cut-off setpoint. During the last 5% or so, the voltage will rise quickly, thus no time to balance.

The way to “blow up” a battery is apply too high a voltage and/or go beyond 100% absorption thus creating excessive internal heat (again, like all batteries). Keep voltages low while you experiment and never leave a CV bulk or absorb voltage on the battery.

Let me know if you have more questions about LFP batteries. I’m looking forward to the coming price reductions where this technology can become mainstream as the benefits are far superior to lead acid technology. Primarily 40% faster charging, always using the full current available and never a need to fully charge. Hopefully some RE equipment manufacturers will see this coming potential and design suitable chargers.

Larry Crutcher
Starlight Solar Power Systems

On Aug 17, 2015, at 3:29 PM, Daniel Young <dyoung at dovetailsolar.com> wrote:

Hello Wrenches,

I’ve been trying to think through how I can mate some lower cost charge controllers (prostar 15A 24v PWM units, or the new 30A Midnite solar Brat) to some 8s LiFePo battery banks.

As I understand it, lithium batteries meant to be charged to around 3.65V per cell (29.2V for the 8 cell bank) and then the charge is terminated, not put into a float mode.

Assuming I use a PCM to protect the batteries, so that each cell has a balancing function and also over/under voltage protection. Would the battery bank accept a simple charge controller that in essence, charged to 3.65V/cell, then backed down to a float voltage that was lower than it’s resting voltage, say 3.3V/cell. In theory at that voltage, there would be little/no charge current going to the batteries. Or would the cells still take in a small amount of power at these voltages and start to overcharge/degrade?

Again, I’d plan to have a full PCM circuit protecting cells and balancing them, but I’d like to have a good way to charge the batteries with solar (right now I’m shooting for using a 72cell module and 24V nominal LiFePo battery stack, so maybe there is another way that I can use, but I’ve not decided to buy any batteries I am ok with accidently blowing up if I’m wrongJ).

With Regards,

Daniel Young,
NABCEP Certified PV Installation ProfessionalTM: Cert #031508-90

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20150818/e43feb56/attachment-0002.html>