Accelerated degradation of LiFePO4 batteries when stored fully charged

[Cpt Pat]

There's a common belief that storing LiFePO4 (LFP) batteries in a fully charged state is a poor practice. Linked below is a good thesis on factors that degrade the life of LiFePO4 batteries, with a detailed explanation of the chemistry involved.

Here's an excerpt from the abstract (emphasis is mine): Capacity loss in cells stored at high temperature and fully charged conditions resulted in faster degradation rates. Temperature had the most significant role in the degradation of the cell and then the cell’s SOC. Comparing capacity losses between cells stored at the same temperature, but with different SOCs, found that the cells with higher SOC experienced increased rates of degradation in comparison to their fully discharged counterparts. In addition, storage at high SOC and high temperatures promoted such severe losses that the cells in question were unable to recapture capacity that they had lost reversibly.

This calls for a storage regime that is the opposite of lead/acid batteries, which should be stored fully charged to prevent sulfation. It also indicates that simply "dropping in" an LFP to replace a lead/acid battery - without changing the storage regime - will result in degraded life.

I personally never charge my LFP battery above 80% SOC or discharge below 20% SOC, and I store the battery at 20% SOC; based on a combination of terminal voltage measurements and Coulomb counting (counting amp/hours in and out). I charge my battery on shore power just before going out to avoid storing it in a charged state between sails. Coulomb counting is especially important to prevent over charging: at very low charge currents, you can overcharge an LFP battery with a terminal voltage as low as 13.65 volts.

https://uwspace.uwaterloo.ca/bitstre...pdf?sequence=3

[john61ct]

I've never heard anyone claim it wasn't true.

That 80/20 idea is spurious IMO.

My zero SoC is 2.99Vpc, my 100% Full is 3.45Vpc, I do not even hold Absorb, no Float, just "charge to then Stop".

That gives me **over** the rated AH in most cases, never more than a few percent under with a new bank from a quality vendor,

like ​Winston/Thundersky/Voltronix, CALB, GBS, A123 & Sinopoly.

So, if you're getting the promised AH capacity while being so safe & gentle with your bank, why state you're "sacrificing" 40%?


Compared to a Full definition at 3.65Vpc, maybe lose 3-4%

Compared to defining 2.5V as zero, max "loss" is 10%, but you are risking major damage, certainly losing lots of lifetime cycles, so no one should really consider going that low, even for testing purposes.

[Cpt Pat]

Originally Posted by john61ct

.

So, if you're getting the promised AH capacity while being so safe & gentle with your bank, why state you're "sacrificing" 40%?

I don't recall ever saying that I'm "sacrificing" anything. I'm getting what I expect from my LFP, while keeping it "off its knees." I have no argument with the rest of your post. Yours sound like reasonably good practices to me. The rest is just semantics.


This may not apply to you, but watch out for charging over 13.65 volts (3.4125 volts per cell) at very low currents, such as you might get from a PV array at day's end. You could risk going over 100% SOC without Coulomb counting (counting amp/hours in and out).


The whole point of my post was to present a good study supporting the belief that storing LFPs at a high SOC can accelerate aging. I believe the thesis is well researched and may be of benefit to some readers. If you read an argumentative tone in my post, it wasn't intended by me.

[coopec43]

Maybe this is relevant

https://www.cycle9.com/wp-content/up...ery-manual.pdf

If you need to store the battery for long periods, start by fully charging it up, and then
make sure to charge it at least once a month to keep it topped up. These batteries will
discharge themselves a bit during storage, since the BMS draws a small amount of
power to maintain cell balance. If you don't occasionally charge them, one or more cells
may fall below the 2.0V per cell critical threshold, permanently damaging the battery
and voiding the warranty.

[john61ct]

Originally Posted by Cpt Pat

I don't recall ever saying that I'm "sacrificing" anything.

Someone tossed out the way they ensure longevity is to only cycle between 20% SoC and 80%. It might be "just semantics" but I hate for newbies to falsely interpret both the top and bottom 20% are being sacrificed from rated AH capacity.

> I have no argument with the rest of your post. Yours sound like reasonably good practices to me.

That's how I know you're a clever fellow 8-)

> This may not apply to you, but watch out for charging over 13.65 volts (3.4125 volts per cell) at very low currents, such as you might get from a PV array at day's end. You could risk going over 100% SOC without Coulomb counting (counting amp/hours in and out).

I can't see that being a problem charging **to** 3.45Vpc and then stopping without any Absorb.

But for me personally one of LFP's biggest advantages is eliminating any dependence on solar.

> If you read an argumentative tone in my post, it wasn't intended by me.

Same here. Getting hard evidence is A Good Thing, even of facts already universally accepted as beyond dispute.

[john61ct]

Originally Posted by coopec43

Maybe this is relevant

Not IRL for LFP in general, and in fact harmful advice 99.99% of the time.

As long as the cells are completely isolated, **including from any BMS circuitry**, and temps are cool, self discharge rates are very low.

If someone were to ask "how should I store my LFP bank disconnected for years without anyone looking at them",

IMO the only safe answer would be "don't do that".

[Waveguide]

For long term storage, I've connected a 12 volt "wall wart" power supply to the pack through a steering diode. If the pack voltage drops below 12.0 volts (3.0 Vpc), the supply automatically takes over to supply a small amount of current to compensate for the very low self-discharge rate. Except in such an extreme, the diode is always reverse-biased and passes no current.

It's important that the power supply be regulated and free of ripple. I use this supply: https://www.alliedelec.com/product/s...812a/70213351/

If you have "balancing boards" connected to each cell, remove them before storage. They (and the rest of your BMS) are a source of quiescent current drain. If any component of your BMS fails it can discharge your pack to destruction.

For short term (weeks) storage, I have a circuit breaker that I open that isolates the pack from the boat. I have lead/acid batteries on a shore power charger that carry the loads during storage. In these short term storage periods, it's most convenient for me to just switch the breaker open at the end of the sail before I connect the shore power. I leave the LFP batteries in whatever state of partial discharge they happen to be at, and recharge before my next sail, making sure to fully recharge the batteries before each sail to prevent accumulation of memory effects.

This is a practical approach to preserving my LFPs. I don't treat them with the delicacy of museum pieces, but they still test good after 5 years and more than a thousand cycles.