Is floating LFP batteries bad for them?

[simat]

If you charge an LFP battery to nearly 100% full using solar power with a charge voltage of 3.45 V/cell at an end current of less than C/50 and then float the battery at 3.34 V/cell are you substantially decreasing the lifespan of the battery rather than just terminating the charge?

I have seen quite a few recommendations not to float LFP batteries but haven't found any information as to why floating an LFP battery at such a low voltage would decrease the lifespan of the battery.

Simon

 

[Hillhater]

I suspect they will sink ! :wink:

 

[dogman dan]

I think mostly, there is simply no need to float the battery. Once it's charged, it will hold what it can for a long time. But I can't see myself, in my ignorant common sense way, why floating them at that low a voltage would actually harm them any more than storing at the same voltage. As long as the voltage does not climb past 3.65v, I don't see the harm.

I'd monitor closely at first, make sure no individual cell charges too high. This is what I did, when I ran a lifepo4 bike battery on lead chargers for almost two years. They undercharged the battery, so having them float overnight never did anything the bms could not keep up with. Mostly, the bms did nothing, since it was very rare for any cell to get charged high enough to kick in the bms discharging.

Bear in mind, I'm an Ignoramus, with an agriculture degree. But it worked fine for me.

 

[simat]

I suspect they will sink ! :wink:

How full do they have to be before they sink?

 

[simat]

I think mostly, there is simply no need to float the battery.

In my application in an off-grid power system charged by solar the battery can be fully charged before midday. Keeping it on float and fully charge for the rest of the day with intermittent loads is useful as I don't know how cloudy it is going to be over the next few days.

But I can't see myself, in my ignorant common sense way, why floating them at that low a voltage would actually harm them any more than storing at the same voltage. As long as the voltage does not climb past 3.65v, I don't see the harm.

I tend to agree with you, but these are electrochemical devices and anything is possible. I am not sure if keeping the battery at this high energy state can lead to some lithium plating or other unwanted side reactions. I would also be interested in more details about the lithium plating that I understand happens at the anode above 3.65 V/cell from some of the very knowledgeable folk on this forum.

Bear in mind, I'm an Ignoramus, with an agriculture degree. But it worked fine for me.

I like your disclaimer, I have been called an idiot by someone who occasionally lurks on this forum so you are in good company.

Simon

 

[dmwahl]

Every academic or industry study I've ready on aging suggests that storing lithium ion batteries (LFP included) above about 40C and 100% SOC reduces calendar life. Below 40C the reduction in capacity over time becomes much less to the point that it's probably not a big factor. Even at 40C and above, the capacity reduction over time is mostly a factor of temperature, storing at lower SOC has some effect on the capacity reduction, but the primary variable is temperature. I have a LFP pack in my lawn mower, which I plan on charging via solar to ~3.45V per cell. I've sized the panel such that the charge will take 3-4 days, but even so it will sit and "float" for a few days.

There's no real reason to float them other than it simplifies the charging system. If you are going to float them though, I'd suggest doing it at a voltage just barely above the resting voltage rather than the standard charge voltage of 3.65V/cell. There's very little capacity gain above that point anyway, so you're not really losing anything.

https://hal.archives-ouvertes.fr/hal-00876555/document

[EDIT] I did some more searching, apparently the A123 26650 cells at one point specified 3.45V for indefinite time floating and CALB spec'd 3.40V. I haven't been able to find the actual documents with those numbers, but a fair number of forums have claimed them so probably reasonable. I also have a commercial battery backup pack from Lifebatt that uses a bunch of 8Ah cylindrical cells, the onboard charger is set to 3.45V/cell. It was designed as a starter battery for a multi-megawatt diesel generator and stayed on float for it's entire life outside of power outages.

 

[dmwahl]

Here's an A123 document stating the 3.45V float voltage.
http://liionbms.com/pdf/a123/charging.pdf

 

[dogman dan]

The main reason I think it will be ok, is the top of charge is only going to be 3.34v. As for storage degradation, well, you kind of want it full all the time in this application, so it is what it is. Just don't float them higher than 3.65v for sure.

Definitely, if possible keep them cool. Don't have the cells in some tin box that gets hotter than ambient when the sun shines on it.

 

[simat]

There's no real reason to float them other than it simplifies the charging system. If you are going to float them though, I'd suggest doing it at a voltage just barely above the resting voltage rather than the standard charge voltage of 3.65V/cell. There's very little capacity gain above that point anyway, so you're not really losing anything.

https://hal.archives-ouvertes.fr/hal-00876555/document

Thanks David for all the information.

The 3.34 volts float voltage I mentioned in my first post was arrived at experimentally by Gorden who posts on the Energy Matters forum as being the voltage at which to float an LFP battery after it has been charged to 3.45 volts with I think an absorb time of half an hour to maintain it at close to the same charge level throughout daylight hours with intermittent small discharge cycles.

I also have a commercial battery backup pack from Lifebatt that uses a bunch of 8Ah cylindrical cells, the onboard charger is set to 3.45V/cell. It was designed as a starter battery for a multi-megawatt diesel generator and stayed on float for it's entire life outside of power outages.

I would be interested how long the battery was in service and if there were any tests done on it at the end of its life to ascertain its end of life capacity and internal impedance.

Thanks,
SImon

 

[dmwahl]

I would be interested how long the battery was in service and if there were any tests done on it at the end of its life to ascertain its end of life capacity and internal impedance.

Thanks,
SImon

The one I have was never placed in service, it was a sales demo. Has maybe a couple shallow cycles on it, otherwise brand new except for being a few years old. I haven't done extensive tests on it yet, but plan to some day. Point was that they decided on the 3.45V float voltage and presumably (hopefully) had a good reason for doing so. The cells are all from some german company that I forgot the name of, and are very high quality. I know a guy who has had them in his mower and ebike for over 10 years and says they still perform like new. The A123 doc (attached) I mentioned was the other reference for the 3.45V/cell float.

 

[dmwahl]

Some more data which may be helpful...

I had been charging my mower pack using the B&D charger for some time, and a couple months ago mowed the lawn to the point that I actually depleted the cells (7 out of 8 down to ~2.8-2.9V resting, the last around 2.0V). I hooked up my watts-up to measure current to recharge, and using only the B&D charger, put about 18Ah into the 20Ah pack, at which point the cells were all around 3.42V (which is the float voltage of the B&D charger. I then hooked up my lab supply set to 29.2V (3.65V/cell) and set current limit to 1A to top off the pack. After that it took an additional 2.5Ah before 7 out of 8 cells pretty much simultaneously reached 3.65V. Another 0.5Ah was needed to get the last cell up to 3.65V.

The point being, at a float voltage of 3.42V/cell, the cells charged up to roughly 90% and then stayed there. I measured no current flowing at that point. Unlike lead-acid cells which at float will continually accept a small current, the LFP cells don't require a measurable current to remain there.

 

[jonescg]

Hi Simon,

I know Kurt (OffGridQld on the AEVA forum) has some LiFePO4 cells in a PV-battery system and he's floated at no more than 3.5 V per cell.

Provided your charge regulator delivers this exact voltage it should be safe. As the others have said, keeping it cool will be critical.

There are a couple of good threads on the AEVA forum about this stuff, so check them out.

Harvey gets hot in summer, so an insulated/ventilated spot would be ideal.

Chris