LFP charging - charging to 100% detrimental to life?

[Wdeertz]

I have 12 24V 50Ahr battle born batteries wired in parallel as depicted in the attached file (3x4) with all interconnecting cables between batteries/bus bars equal to provide equal resistance. My charger is setup for bulk voltage of 28.8v with a 2.5 hr absorb phase (this is a misnomer used by battle born as the batteries are 100% SOC but this holds the voltage at 28.8v to allow each bms to balance any cells needing it). After this the charger goes into silent mode (as LFP don't like float) and then rebulks once the voltage hits 26v (after about 3 days). While on anchor its my practice to cycle the batteries between 95% SOC to 5-10% SOC. I have 200A of charging so generally less than 2 hours/day of generator run time is sufficient.

I specifically asked battle born how long I could go without running any absorb phase (cell balancing) and they suggest I run the 2.5 hr absorb at least every 2 weeks. While reading an article from an "industry expert" on another forum he recommended discontinuing charging at less than 100%. I followed up with the author and in his reply he accepted LFP can be charged to 100% SOC but said its not ideal for battery longevity. This was the first I've heard that charging to 100% SOC is detrimental to LFP cycle life.

Anyone have any views on this matter?

 

[DavidM]

Some say charge to 70% for best life. Why don’t you ask Battleborn this question?

David

 

[DDW]

The industry concensus seems to be that cycling between 30 and 70% or 20 and 80% will produce the best battery life. There has been a difference shown in testing but it isn't dramatic. There is also a reduction in life charging at higher voltage, some pundits recommend 14.2 (or 28.4) as the limit. Again a difference in longevity but not dramatic. The extra capacity gained by going the 14.4 is quite small.

 

[sbman]

Yes even with cel phones some manufacturers have allowed settings that keep the battery in the 20/80 range for extended lifespan at the small cost of some available capacity. To store lithiums, it is recommended to drain them to a lower level of charge before storing them.

Unfortunately issues specific to various battery chemistries get burned into people's brains from past experiences and they then assume that it applies to the newer stuff as well. Some people still hang on to battery 'memory' that was prevalent with NiCd for example on their smart phones, which does not apply to Lithium chemistries. Lead acid users have been taught to top them up 100% and even overcharge them for good health which is good practive for those but overcharge a lithium and you destroy it, 100% all the time isn't great for them either.

 

[catalinajack]

The industry concensus seems to be that cycling between 30 and 70% or 20 and 80% will produce the best battery life. There has been a difference shown in testing but it isn't dramatic. There is also a reduction in life charging at higher voltage, some pundits recommend 14.2 (or 28.4) as the limit. Again a difference in longevity but not dramatic. The extra capacity gained by going the 14.4 is quite small.

I have been following this steady march to lithium for some time but I am a bit of a skeptic as to its long-term efficacy and return on investment. I am not wholly averse to lithium. One of the advantages cited by many is the significant increase in amp-hour capacity vs flooded batteries. Now it seems that, in order to extract the number of cycles needed to justify their cost over the long term, use must be kept in a 40 - 60 per cent range. In terms of useable amp-hours, how is this any different than the widely-accepted 50% depletion standard for flooded batteries? So, three or four times the cost of flooded batteries for no more usable amp-hours. What am I missing? There must be something. Forgive my ignorance.

 

[gsholz]

For best longevity lithium batteries require some rethinking. E.g. if connected to shore power and you are not using the boat, don't float them, keep them between 50-70%.

Tesla does not recommend regularly charging to 100%, only when necessary before long trips that then start immediately. I'd expect marine LFP inverter/chargers eventually be connected to the Internet and controlled via an app similar to Tesla's in the future. Then you'd keep your battery at 50% for storage and charge to 90% while you drive to the boat.

Vendors: Even $100 game cameras now have cellular connectivity. There is no reason your $1000 inverter/charger cannot provide this for a small monthly fee. Add some alarm and boat monitoring features and you have a real business opportunity with recurring income. You are welcome.

 

[sbman]

I have been following this steady march to lithium for some time but I am a bit of a skeptic as to its long-term efficacy and return on investment. I am not wholly averse to lithium. One of the advantages cited by many is the significant increase in amp-hour capacity vs flooded batteries. Now it seems that, in order to extract the number of cycles needed to justify their cost over the long term, use must be kept in a 40 - 60 per cent range. In terms of useable amp-hours, how is this any different than the widely-accepted 50% depletion standard for flooded batteries? So, three or four times the cost of flooded batteries for no more usable amp-hours. What am I missing? There must be something. Forgive my ignorance.

It's a fairly minor effect with lithiums, and only slightly improves their longetivity to restrict their range. Not nearly as dramatic as sulfating of lead acids.

For those that are off-grid it's actually quite a boon that they don't need to be 100% full on a regular basis, they'll be perfectly happy to work in their mid range for thousands of cycles.

 

[Simi 60]

For best longevity lithium batteries require some rethinking. E.g. if connected to shore power and you are not using the boat, don't float them, keep them between 50-70%.

Tesla does not recommend regularly charging to 100%, only when necessary before long trips that then start immediately. I'd expect marine LFP inverter/chargers eventually be connected to the Internet and controlled via an app similar to Tesla's in the future. Then you'd keep your battery at 50% for storage and charge to 90% while you drive to the boat.

Vendors: Even $100 game cameras now have cellular connectivity. There is no reason your $1000 inverter/charger cannot provide this for a small monthly fee. Add some alarm and boat monitoring features and you have a real business opportunity with recurring income. You are welcome.

Tesla is totally different chemistry to LFP

And, why would you need your batteries connected to the internet for a fee?
Our LFP batts are connected with Bluetooth app to tablet and controllable for free
I can do the same with inverter charger, no fees for either

 

[sbman]

Vendors: Even $100 game cameras now have cellular connectivity. There is no reason your $1000 inverter/charger cannot provide this for a small monthly fee. Add some alarm and boat monitoring features and you have a real business opportunity with recurring income. You are welcome.

Victron has a couple of different monitor boxes that do this and you only have to have one connection for all of their devices on your boat. They provide the data all on a free portal on their website.

 

[Simi 60]

Victron has a couple of different monitor boxes that do this and you only have to have one connection for all of their devices on your boat. They provide the data all on a free portal on their website.

The victron data is also available via Bluetooth app
No connecting to the internet required

 

[sbman]

The victron data is also available via Bluetooth app
No connecting to the internet required

That is true, and it's device by device and requires physical presence on the boat. Their monitoring products aggregate the data (sum up multiple solar charge controllers for example), show it on a local screen if desired as a system overview and optionally upload it to their web portal so you can access it from anywhere in the world. They also allow you to configure and view data from most of their devices that don't have bluetooth built in, and even do that remotely.

My VictronConnect app can now see all of my devices and configure them just like I was on board no matter where I'm at as long as the boat has connectivity.

It's pretty fun stuff and the charting/graphing is very helpful for troubleshooting problems.

My point was that some vendors are making meaningful web access as an option for reasonable pricing even without putting a 4G modem in every single device.

 

[twistedtree]

Regarding cycle life, there are dozens, if not hundreds of studies showing that you can get 2000 cycles out of LFP cells under the most stressful use conditions. That would be discharging to the lowest allowed voltage, charging to the highest allowed voltage, and with charge/discharge rates that greatly exceed anything you would ever experience with a house bank. So you can pretty confidently count of 2000 cycles. That's 200 nights at anchor for 10 years.


Consensus is that reducing the daily charge/discharge voltages a bit keeps you safely away from the limits that start to cause battery damage, and sacrifices very little in capacity to do so. Charging to 3.45 volts per cell is commonly used, vs the max allowed of 3.60V for LFP. But the number of Ah stored between 3.45 and 3.60 vpc is very little. Like single digit percentages. I did a balance on my bank to re-level cells that had drifted apart by about that difference of 150mv. It only took 1% of the cells AH rating to bring them back in line, so that gives you an idea how much capacity is lost. Answer, not much. I think it's easy to still utilize 80% or more of the rated capacity even with the most conservative operation.


The whole issue of float and storage is much less of an issue than people make it out to be. Float is your friend, not your enemy. You just need to use it correctly. The confusion arises because with lead you set the float voltage ABOVE the resting voltage of the fully charged battery. That causes a continuous trickle charge current that is good for lead batteries.



LFP will be damaged if you maintain an ongoing charge current one that are fully charged, even a small one, so you need to set the float voltage BELOW the resting voltage of a fully charged LFP. That causes the batteries to initially carry any loads, and slowly lower their SOC until the voltage matches the charger float voltage. At that point, the charger picks up the loads, and maintains the battery at that SOC level. Setting the float voltage at a 70% or so SOC also accomplishes the goal of long term storage. 3.35 vpc works very well for this.

 

[catalinajack]

Regarding cycle life, there are dozens, if not hundreds of studies showing that you can get 2000 cycles out of LFP cells under the most stressful use conditions. That would be discharging to the lowest allowed voltage, charging to the highest allowed voltage, and with charge/discharge rates that greatly exceed anything you would ever experience with a house bank. So you can pretty confidently count of 2000 cycles. That's 200 nights at anchor for 10 years.


Consensus is that reducing the daily charge/discharge voltages a bit keeps you safely away from the limits that start to cause battery damage, and sacrifices very little in capacity to do so. Charging to 3.45 volts per cell is commonly used, vs the max allowed of 3.60V for LFP. But the number of Ah stored between 3.45 and 3.60 vpc is very little. Like single digit percentages. I did a balance on my bank to re-level cells that had drifted apart by about that difference of 150mv. It only took 1% of the cells AH rating to bring them back in line, so that gives you an idea how much capacity is lost. Answer, not much. I think it's easy to still utilize 80% or more of the rated capacity even with the most conservative operation.


The whole issue of float and storage is much less of an issue than people make it out to be. Float is your friend, not your enemy. You just need to use it correctly. The confusion arises because with lead you set the float voltage ABOVE the resting voltage of the fully charged battery. That causes a continuous trickle charge current that is good for lead batteries.



LFP will be damaged if you maintain an ongoing charge current one that are fully charged, even a small one, so you need to set the float voltage BELOW the resting voltage of a fully charged LFP. That causes the batteries to initially carry any loads, and slowly lower their SOC until the voltage matches the charger float voltage. At that point, the charger picks up the loads, and maintains the battery at that SOC level. Setting the float voltage at a 70% or so SOC also accomplishes the goal of long term storage. 3.35 vpc works very well for this.

So, TwistedTree, for my simple mind, discharging to the lowest allowed voltage and charging to the highest allowed, what is the number ofusable amp-hours for a 100 amp LFP battery to reach 2,000 cycles? Won't it be somewhat less than 100? By the way, relative to what I had to say in a related post, I apologize for my inexcusable ignorance as to the fire danger and, to a slightly lesser degree, for my ignorance on the efficacy of a LFP system. You did mention 80% using conservatively but what is your opinion as to getting 2,000 cycles? Can it be more than 80%?

 

[Wdeertz]

Regarding cycle life, there are dozens, if not hundreds of studies showing that you can get 2000 cycles out of LFP cells under the most stressful use conditions. That would be discharging to the lowest allowed voltage, charging to the highest allowed voltage, and with charge/discharge rates that greatly exceed anything you would ever experience with a house bank. So you can pretty confidently count of 2000 cycles. That's 200 nights at anchor for 10 years.


Consensus is that reducing the daily charge/discharge voltages a bit keeps you safely away from the limits that start to cause battery damage, and sacrifices very little in capacity to do so. Charging to 3.45 volts per cell is commonly used, vs the max allowed of 3.60V for LFP. But the number of Ah stored between 3.45 and 3.60 vpc is very little. Like single digit percentages. I did a balance on my bank to re-level cells that had drifted apart by about that difference of 150mv. It only took 1% of the cells AH rating to bring them back in line, so that gives you an idea how much capacity is lost. Answer, not much. I think it's easy to still utilize 80% or more of the rated capacity even with the most conservative operation.


The whole issue of float and storage is much less of an issue than people make it out to be. Float is your friend, not your enemy. You just need to use it correctly. The confusion arises because with lead you set the float voltage ABOVE the resting voltage of the fully charged battery. That causes a continuous trickle charge current that is good for lead batteries.



LFP will be damaged if you maintain an ongoing charge current one that are fully charged, even a small one, so you need to set the float voltage BELOW the resting voltage of a fully charged LFP. That causes the batteries to initially carry any loads, and slowly lower their SOC until the voltage matches the charger float voltage. At that point, the charger picks up the loads, and maintains the battery at that SOC level. Setting the float voltage at a 70% or so SOC also accomplishes the goal of long term storage. 3.35 vpc works very well for this.

Twisted tree, thanks you pretty much confirmed what I had previously been led to believe. I contacted battleborn technical support and they said their “drop in” LFP batteries are top balancing so they should be charged to 100% every few weeks to allow any cell imbalances to be equalized. I like your comment on setting the float voltage at 70% SOC voltage for long term storage. Do you think this approach is preferable over having the charger go into silent mode and rebulk at a set voltage? While on shore power I bounce between 100% to 30% over a 3 day period.

 

[twistedtree]

First, Sorry for being snippy in my earlier post....

So, TwistedTree, for my simple mind, discharging to the lowest allowed voltage and charging to the highest allowed, what is the number ofusable amp-hours for a 100 amp LFP battery to reach 2,000 cycles? Won't it be somewhat less than 100?

It would be the full 100Ah. Those are the standard "rated conditions". That said, I've seen a lot of cells that test out above the rated capacity when new.

You did mention 80% using conservatively but what is your opinion as to getting 2,000 cycles? Can it be more than 80%?

2000 cycles is basically hammering the batteries as hard as you can, pushing all limits. So I think that's a very safe and conservative minimum one can expect, especially considering that in use as a house bank, batteries get operated much less harshly.


For perspective, these studies that get 2000 cycles are typically charging and discharging at anywhere from 1C to 5C, which is doing a full discharge or full charge in as little as 12 minutes, and no more than 1hr! That's harsh, and completely unheard of for a house bank. In contrast, batteries made for use as a house bank typically call for charging in no less than 3-4 hours. And no useful house bank would be discharged over any less than 12 hrs, with 18 - 24 hrs being more reasonable.


As far as charge/discharge voltages, you can quickly damage an LFP battery by exceeding the allowed limits, so the common solution is simply "don't do that", and provide some margin for error. So LFP max charge is 3.6V per cell (vpc), and most house batteries call for charging not to exceed 3.45 - 3.50 vpc. Similar margins are used at the discharge limits. It's like driving on a road where there is a big shoulder, and a solid guard rail.



The big questions is how much longer will cells last if you operate them more gently as in a house battery bank, and I don't really think anyone knows for sure. At least not yet because it takes so darn long to test it. If an example house bank usage pattern is 24 hrs with 20hrs to discharge, then 4 hrs to charge, you can do one cycle per day. 2000 cycles will take 5.5 years. So honestly, everything above 3000-4000 cycles is based on modeling. And nobody really knows if the number will be 3000 cycles, or 10,000 cycles.



Getting back to the capacity question, and sticking with the driving-on-a-road analogy, it turns out that there is very little capacity forfeited by having shoulders on the road. This is because when you charge and discharge LFP, the voltage rise (or drop) at the end of the cycle is quite abrupt and steep. So there is no mistaking when you are approaching the end of the cycle, so it's easy to stop. It's kind of like having rumble strips at the beginning of the shoulder. It's just one example, but I needed to balance a couple of the cells in my battery bank about a year into use. The correction needed to get them rebalanced was just about equal to the spread in voltage between a conservative full charge of 3.45 vpc and a max possible full charge of 3.60 vpc. All it took was 1% of the battery's capacity in extra charge to catch up the lagging cell. So that suggests that very little actual capacity is being left on the table. Rod Collins (MarineHowTo) did a good controlled study of this and concluded there was something around 5% of capacity given up. So I think assuming that you can actually use 80% of an LFP's rated capacity is really quite conservative.

 

[twistedtree]

Twisted tree, thanks you pretty much confirmed what I had previously been led to believe. I contacted battleborn technical support and they said their “drop in” LFP batteries are top balancing so they should be charged to 100% every few weeks to allow any cell imbalances to be equalized. I like your comment on setting the float voltage at 70% SOC voltage for long term storage. Do you think this approach is preferable over having the charger go into silent mode and rebulk at a set voltage? While on shore power I bounce between 100% to 30% over a 3 day period.

Some batteries, Battleborn being one of them I think, only activate their balancing circuits while the batteries are at or near full charge. So if you never fully charge them, they will never balance. This is why it's always important to follow any specific directions from the battery manufacturer.


Regarding float, I think some early guidance was misinterpreted and the whole thing skipped the rails and took on a life of it's own. Unwinding it is now a very difficult task.


It really comes down to what you mean by a "float charge". With lead, it's an elevated charge voltage that causes a continuous, ongoing charge current. You definitely don't want to do that with LFP. Once they are charged, you need to stop applying current. This led to a "don't float LFP" Mantra, without really understanding that it's more specifically a continuous charger current that's bad.



With that Mantra, people just disabled the float feature in their chargers, leading to what I believe is gratuitous and wasteful cycling of your batteries. People do exactly as you have which is to charge to full, then shut off the charger and run on the batteries until they drop to some low level, then recharge the again to full.


This is particularly wasteful when you are charging with solar. When the sun comes up, your batteries start charging and let's say they are full by 14:00. If you just stop charging, then all your afternoon loads drain the batteries even though you have solar power right there for the taking. Then when morning comes around again, you have not only drained the batteries with your night loads, but also the afternoon loads from the previous day. And the same is true if you are on dock power. Why cycle your batteries when dock power is continuously available.


If instead you make use of the float function in your chargers, it solves multiple problems. First, set the flat voltage so it doesn't create a continuous charge current when the batteries are full since we know that's bad. Instead set the float voltage to match the battery's resting voltage somewhere in the 50-80% SOC range. Note that because in this range then resting voltage of LFP is pretty flat across a wide SOC range, this will NOT be a precise setting. What's important is that it's below about 80% if you expect to leave the batteries on the charger for an extended period of time. That solves the long-term storage issue.


Now what happens is your charger initially does a full charge cycle and brings the batteries to 100%. Then it switches to float, which is less than the battery's resting voltage. In this situation, the charger doesn't deliver any current, and the loads run completely off the batteries. At this point it's just like turning the charger off completely. However the charger is on and waiting for the battery SOC to drop to the point where the battery voltage matches the charger's float voltage. When the voltages match, the charger starts to pickup to loads rather than the batteries, and it all settles out in an equilibrium where the batteries are held at a reduced SOC rather than cycling, and the charger is carrying the loads. This system can stay like this indefinitely.


The only down side is that if you unplug and motor away, you are getting underway with less than 100% SOC. But as you are motoring, you will also (presumably) be charging the batteries back towards 100% SOC, ideally reaching it before you drop the anchor for a nice quiet night. So I think it actually all works out pretty well.

 

[dutch-barge]

The victron data is also available via Bluetooth app
No connecting to the internet required

But if you have a Venus or Gerbo. you can get it for free on the internet..You can even then add your bilge pumps to the control unit and get a signal when they go on, or when shore power goes off. All for free! ( no need to pay a $30- $100/mo monitoring fee)

 

[Simi 60]

But if you have a Venus or Gerbo. you can get it for free on the internet..You can even then add your bilge pumps to the control unit and get a signal when they go on, or when shore power goes off. All for free! ( no need to pay a $30- $100/mo monitoring fee)

Totally useless for the full time cruiser.

 

[dutch-barge]

Totally useless for the full time cruiser.

Ah yes as a full time cruiser you never set foot on shore, you never go out to dinner on land and you never have to travel on land to see something interesting ...




Yet the ones who do set foot on shore, go see interesting things on land, rent the occasional car to make a trip to something not visible from the boat, can use it.

 

[Simi 60]

Ah yes as a full time cruiser you never set foot on shore, you never go out to dinner on land and you never have to travel on land to see something interesting ...




Yet the ones who do set foot on shore, go see interesting things on land, rent the occasional car to make a trip to something not visible from the boat, can use it.

Gee, I have enough faith in the systems on board that I am comfortable enough going the day without connecting to the interwebs to see if things are ok

I remember cruising when the interwebs didn't even exist, how did we ever manage?

Also, it might surprise you that often cruisers are in areas where there are no interwebs to connect to even if they wanted to.

 

[Jmarsh203]

From what I’ve been studying on lithium iron phosphate batteries twisted tree is spot on. When treated properly they can go well beyond what manufacture specs are with some studies that I have read suggesting 2-300% increase in life expectancy while retaining over 75% Also improper storage temperature being one of the largest degradation of capacity over actual cycle life has been noted.

 

[sbman]

Gee, I have enough faith in the systems on board that I am comfortable enough going the day without connecting to the interwebs to see if things are ok

I remember cruising when the interwebs didn't even exist, how did we ever manage?

Also, it might surprise you that often cruisers are in areas where there are no interwebs to connect to even if they wanted to.

It's a convenience thing for sure, with good LTE equipment you can get signals in a lot of places. Yesterday I couldn't remember if I turned off un-needed stuff when I arrived at shore. Easy enough to look at my phone and see. I rowed back to the boat to turn things off. Whether it's useful or not is up to each individual.

This charting is what I find most useful about the Cerbo GX

There is a ton of information available with that to tune the system and get the equipment doing exactly what you want it to do, as well as compare over time how things are performing.

 

[Simi 60]

This charting is what I find most useful about the Cerbo GX

There is a ton of information available with that to tune the system and get the equipment doing exactly what you want it to do, as well as compare over time how things are performing

And I get the exact same charting via Bluetooth,
no internet connection or cerbo cx required
But I rarely look at it

 

[sbman]

And I get the exact same charting via Bluetooth,
no internet connection or cerbo cx required

I get that you don't see the utility of the internet connection or the Cerbo GX and it isn't for you. The Bluetooth logging is very limited only recently did they add the ability to log when your phone isn't actively connected and it is one device at a time.

The online data has a much longer retention period (years?), and provides a combined view of your entire system. You can also get the same data on SD card locally or capture it with other data capture tools (InfluxDb, Grafana, etc) to create your own local data displays and analysis. For myself, I appreciate the ability to dial in the system based on data collected and available however I choose, online or off.

This is a bit off topic... it is however a single method of evaluating your Lithium Iron battery system over time.