What is the optimal SOC to recharge LFP batteries?

[Ka_sea_ta]

I have a LFP bank that is only charged by an AC powered charger. The BMS outputs the batteries' SOC so I can turn the charger whenever needed. What I can't determine is what is the optimal SOC to start the charge cycle to prolong the life of the battery. The manufacturer just says to expect 3500 to 5000 cycles, the BMS does count charge cycles.

 

[Comodave]

Marinehowto.com has a very good article that might help you.

 

[twistedtree]

I have a LFP bank that is only charged by an AC powered charger. The BMS outputs the batteries' SOC so I can turn the charger whenever needed. What I can't determine is what is the optimal SOC to start the charge cycle to prolong the life of the battery. The manufacturer just says to expect 3500 to 5000 cycles, the BMS does count charge cycles.

One of the nice things about LFP is that it doesn't really matter.

 

[Delfin]

I have a LFP bank that is only charged by an AC powered charger. The BMS outputs the batteries' SOC so I can turn the charger whenever needed. What I can't determine is what is the optimal SOC to start the charge cycle to prolong the life of the battery. The manufacturer just says to expect 3500 to 5000 cycles, the BMS does count charge cycles.

The data suggests that the deeper the discharge, the fewer the number of cycles. I think the best way to look at it is not to think about cycles, but total Amp hours delivered. For example, assuming a 1C discharge rate, which never happens, but is the metric used to calculate cycle life, if you have a 200 Ah LFP battery, it will deliver 780,000 Ah if recharged at 70% SOC, 580,000 at 50% SOC and 432,000 Ah at 20% SOC. No clue on the effect of a more realistic discharge rate, but I suspect it changes the numbers, so it's your choice to define optimal SOC.....


Look under: NeverDie BMS Standard Series User Guide https://lithionicsbattery.com/support/


And: https://oceanplanetenergy.com/wp-content/uploads/sites/86/2015/10/WhatIsLifeExpectancy.jpg

 

[Ka_sea_ta]

The data suggests that the deeper the discharge, the fewer the number of cycles. I think the best way to look at it is not to think about cycles, but total Amp hours delivered. For example, assuming a 1C discharge rate, which never happens, but is the metric used to calculate cycle life, if you have a 200 Ah LFP battery, it will deliver 780,000 Ah if recharged at 70% SOC, 580,000 at 50% SOC and 432,000 Ah at 20% SOC. No clue on the effect of a more realistic discharge rate, but I suspect it changes the numbers, so it's your choice to define optimal SOC.....


Look under: NeverDie BMS Standard Series User Guide https://lithionicsbattery.com/support/


And: https://oceanplanetenergy.com/wp-content/uploads/sites/86/2015/10/WhatIsLifeExpectancy.jpg

I suspect the happy medium would be to recharge somewhere around 50% SOC. Knowing that if the situation requires it I can occasionally discharge to 20% SOC or less without sacrificing overall battery life. In reality I could probably let it discharge down to 5% every time and the battery will outlive me

 

[Delfin]

I suspect the happy medium would be to recharge somewhere around 50% SOC. Knowing that if the situation requires it I can occasionally discharge to 20% SOC or less without sacrificing overall battery life. In reality I could probably let it discharge down to 5% every time and the battery will outlive me

50% is my trigger for re-charging, mostly because that is a day's consumption, and running the genset at anchorage for 90 minutes daily isn't a hardship.

 

[Statistical]

LFP are pretty tolerant of high level of charge. If the boat is in use you could drop the max SoC to 90% or 80% but I wouldn't go below that. The key thing is that if dropping max SoC causes the min state of charge to go significantly lower than 50% you are just making things worse. In other words cycling 50% to 100% is better for longevity than cycling 20% to 70%.

If this is for long term storage where the battery won't be significantly discharged and SoC will be flat most OEMs recommend holding SoC at around 70%.

Honestly though unlike other chemistries LFP is really tolerant of being charged to a high SoC the differences are minimal so I wouldn't reduce it unless you are sure that your usage with lower peak charge won't deeply discharge the battery otherwise you are robbing peter to pay paul.

 

[twistedtree]

In other words cycling 50% to 100% is better for longevity than cycling 20% to 70%.

I don't think this is correct. Happy to see research to the contrary, but all I have seen says it doesn't matter.


As Delfin pointed out, lifespan is based on Ah throughput. Discharge cycles means nothing without also saying the range of SOC that it was discharged over. It's like saying a boat is fuel efficient because it burns 5gph. It means nothing without also knowing how fast it's moving.


In the research on LFP, cycles are typically a full discharge cycle, so from 100% to 0%, and that dependably yields 2500-3000 cycles.

 

[SteveK]

The numbers are all over the place if you read enough. With this new tech ever growing has any of the suggested % been tested and re tested. Or is it all guessing. I read the BMS counts the number of charges, why, does it plan end of life at stated expected charges.
My read has found, yes you can completely discharge, but shoot for 20%. Yes you can completely charge to 100% but shoot for 80%. Never found any real reasons, just suggestions.
Still 20-80 is a 60 spread and more than the 30 spread of the FLA. (discharge FLA to 50% and get a full charge around 80%)

 

[Delta Riverat]

i agree.

LiFePO4 has a very flat charge/discharge curve. You get a lot of capacity for not much voltage change.

 

[Simi 60]

Between 40% and 70% is the norm for us but do put in the effort to get to 100% every month or two
600 days of constant use showing as 174 cycles on the BMS history

 

[Don L]

What I can't determine is what is the optimal SOC to start the charge cycle to prolong the life of the battery.

Are you asking for where to "maintain" your charge, or what range to "cycle" your battery?

If "maintaining", which means stored condition around 60% seems to be the recommendation.

If cycling regularly I don"t think it matters as long as you stay away from the "knees" at both ends of the range. If it were me (and this is how I do it if charging from the battery charger) I would start charging at around 60% SOC if I am expecting to go lower before the next time I feel like running the generator, and stop in the mid 90s SOC. But I mostly charge via solar and lots of times the decision to run the generator is based on the cloud forecast for tomorrow. I will admit that I still suffer from lead acid battery thinking some and my LFPs have only been less that 50% once in 6 months.

 

[Delfin]

Lithionics provides batteries to the Dept of Defense, and I gather their milspec testing standard is cycle life at a 1C discharge rate and 0% to 20% State of Charge. Because these are milspec conforming tests, I'm confident they are accurate. You can use the "analog" projection of cycle life at lesser depths of discharge to see that the total energy delivered by a good LFP battery throughout cycle life increases as DoD is reduced.

Of interest is note 3 on the attached test result graph, because it suggests that memory loss from cycling an LFP battery at shallow depths of discharge happens. For that reason, I manage mine using a typical 50% DoD, with occasional discharges to 85% before re-charging.

 

[BILL GOOD1]

Delfin you bring up good info for a Mil spec battery but we are being sold cells which have failed testing to that standard as suitable for solar. But that as it may & before I knew of this I like a lot of others built my battery. Mostly light loads but microwave etc for short periods. All good- but to allow for the quality of the cells? de-rated capacity to 80%. 280a/h with 50a solar charger & solar cell cap well above this. All good set the discharge to 50%SOC only to realize battery was full early in the day & about the time the sun set was when the charge would start! Good if the sun did rise & no rain the next day. After further reading settled on restart @ 95% SOC which means a much higher charged battery when the sun goes west. It results in a shallow charge through out the day. Now will it outlast me? Time is on the batteries side!! PS when AC power is available, this results in min 95% SOC (using the de-rated 100%)

Regards Bill

 

[Delfin]

Delfin you bring up good info for a Mil spec battery but we are being sold cells which have failed testing to that standard as suitable for solar. But that as it may & before I knew of this I like a lot of others built my battery. Mostly light loads but microwave etc for short periods. All good- but to allow for the quality of the cells? de-rated capacity to 80%. 280a/h with 50a solar charger & solar cell cap well above this. All good set the discharge to 50%SOC only to realize battery was full early in the day & about the time the sun set was when the charge would start! Good if the sun did rise & no rain the next day. After further reading settled on restart @ 95% SOC which means a much higher charged battery when the sun goes west. It results in a shallow charge through out the day. Now will it outlast me? Time is on the batteries side!! PS when AC power is available, this results in min 95% SOC (using the de-rated 100%)

Regards Bill

Thanks Bill. My understanding is that all Lithionics does is buy cells from China, impedance match them, use appropriate cabling (unlike many other manufacturers), and appropriately contain the cells in a plastic box. There may be some magic in their BMS, but I wouldn't know if that's true or not. Their circuitry grounds a cell in the pack when it reaches 3.65 vdc, and that is how they balance the pack. Only had the BMS do something other than balance cells once when I let a low voltage event trigger a disconnect.



My point is, I'm not sure a Lithionics is worth the price, EXCEPT as a guarantee of good engineering practices. That was worth it to me five years ago when I installed them, and I am pretty sure they will outlive me.

 

[Barking Sands]

Delfin you bring up good info for a Mil spec battery but we are being sold cells which have failed testing to that standard as suitable for solar.

Regards Bill

Can you expand on that a bit? Who is being sold failed cells? Its clear that early on this was the case. The cells were tested, grouped, and simply sold at the tested capacity. But it is my understanding many of the top manufacturer's now sell with brand new cells.

 

[twistedtree]

Can you expand on that a bit? Who is being sold failed cells? Its clear that early on this was the case. The cells were tested, grouped, and simply sold at the tested capacity. But it is my understanding many of the top manufacturer's now sell with brand new cells.

I think the quality manufacturers have always use new cells. And there have always been reputable sellers of bare cells that are factory fresh with factory certification testing. And there always have been, and always will be bootleg crap.

 

[Barking Sands]

I think the quality manufacturers have always use new cells. And there have always been reputable sellers of bare cells that are factory fresh with factory certification testing. And there always have been, and always will be bootleg crap.

Yes, I think that is more accurately stated. Just double checking. Bill's statement seemed to imply knowledge otherwise. Even many of the reputable mid-level products seem to be using new cells these days.

 

[BILL GOOD1]

The main question was how are we charge our batteries. What is the best method? The last LIFEPO4 battery was on the previous boat for over 10Yrs & a lot more cells are now on the market & I chose EVE 304A/H cells. After I bought them found any cell that had lasered terminals were in fact "rejected as suitable for EV'S in China". They then have terminals lasered on & sold for Solar. No record of the rejection is offered. These cells could be a few yrs old by the time they are sold to us!! The load on my boat is less than 10A/H & 50A/H solar resulted in a full battery by mid-morning. (No separate load connection- charge or no charge). Initially had selected 50% SOC to restart charging but realized the battery would dis-charge slowly until the sun set & continue to dis-charge overnight. OK so long as the Sun was doing the job next morning! Found it works better if the charge was set to 90% SOC for off & 85% SOC for on resulting in a much higher SOC when the Sun set in my case. The solar seems also to be supplying power to the load longer through the day & all equipment & cells appear to be cooler. Balancing of the cells is good. Regards

 

[Caballero II]

I have a LFP bank that is only charged by an AC powered charger. The BMS outputs the batteries' SOC so I can turn the charger whenever needed. What I can't determine is what is the optimal SOC to start the charge cycle to prolong the life of the battery. The manufacturer just says to expect 3500 to 5000 cycles, the BMS does count charge cycles.

I am assuming that what your are talking about LifePro batteries but do not know for sure. If you are going to start a thread, please don't without a full description/reference of what your abbreviation refers to. I don't know how this discussion relates to my sealed AGM (Absorbent Glass Mat) technology, a lead acid battery.

 

[Don L]

it was clear to me!

 

[Ka_sea_ta]

I am assuming that what your are talking about Lifepro batteries but do not know for sure. If you are going to start a thread, please don't without a full description/reference of what your abbreviation refers to. I don't know how this discussion relates to my sealed AGM (Absorbent Glass Mat) technology, a lead acid battery.

I have no idea either how this discussion relates to your AGM batteries... I was simply asking the question when the is the optimal SOC to recharge LFP or Lithium Iron Phosphate or LiFePo4 batteries. As far as Lifepro , I've no idea what they are, perhaps you could elaborate.

 

[Delfin]

The main question was how are we charge our batteries. What is the best method? The last LIFEPO4 battery was on the previous boat for over 10Yrs & a lot more cells are now on the market & I chose EVE 304A/H cells. After I bought them found any cell that had lasered terminals were in fact "rejected as suitable for EV'S in China". They then have terminals lasered on & sold for Solar. No record of the rejection is offered. These cells could be a few yrs old by the time they are sold to us!! The load on my boat is less than 10A/H & 50A/H solar resulted in a full battery by mid-morning. (No separate load connection- charge or no charge). Initially had selected 50% SOC to restart charging but realized the battery would dis-charge slowly until the sun set & continue to dis-charge overnight. OK so long as the Sun was doing the job next morning! Found it works better if the charge was set to 90% SOC for off & 85% SOC for on resulting in a much higher SOC when the Sun set in my case. The solar seems also to be supplying power to the load longer through the day & all equipment & cells appear to be cooler. Balancing of the cells is good. Regards

LFP cells might not be "suitable for EV's in China" whether new or old, so I'm not sure that means what you have is defective in any way. Ev's generally use a different chemistry because of LFP's low gravimetric energy density relative to other much lighter Lithium ion chemistries. I gather car manufacturers are starting to experiment with LFP because of their safety, longevity, rapid charging rates, but I suspect the weight penalty will limit their use to vehicles where weight is not as important as other factors.