Float of LFP

[CaptainRivet]

Originally Posted by sailorboy1

we are talking LFP "float" which is no higher than 100% SOC voltage

I got my answer and am moving on as I can see thread drift is about to set in.

Read eric's article, ypu can overcharge a LFP with float at 93%SOC if yo do it long enough...
Its an eye opener.
I charge till 3 53V then shut of and reengage charge at 3,35V. Always have a load of min 5A...
No need or any advantage in floating a LFP..thats lead thinking...

[Captain Bill]

Originally Posted by CaptainRivet

Total voltage is fake security, at 14.2V one cell can have 3,75V while all other 3.3V or even worse.
Charging must be done on cell level, that why its best the BMS manages the battery and charging and not the charge sources.
Flaot at 13.2V makes no sense with a LFP, if it gets into the knee it reduces the current it takes by itself. An LFP needs no float at all,.you can but it is unnecessary.
A friend of mine i charging his LIfepo4 Calb cells since 17 years with a power supply with 13,85V, BMS cuts off of a cell hits 3,5V. Calbs have still 92% of their original capacity.
I would put 13.8 and 14.2V if you only can on total voltage but add a BMV 712 and use the midbank voltage of the 2nd measurement input to check voltage devation of cell 1+2 versus 3+4. So if devation rasises you avoid overcharging a cell.

All great if you have built your own battery and have total control of your BMS. It also helps if you want to spend your life as a battery engineer. I on the other hand prefer to cruise and enjoy things other than battery engineering. I bought drop in batteries with a built in BMS that doesn't give me a lot of options. I still spent less money than buying another set of lifeline AGMs, the last set of which only lasted three years. The first two sets lasted 7 years. If the cheap built in BMS kills a battery after 3 years I'm still about $200 ahead including upgrading my charging system. If your idea of fun is to babysit a bank of batteries please by all means enjoy yourself. I'll be doing other things. FYI the term float is used by the charging system makers to describe the condition where the charging system is carrying the house load instead of the batteries. Since they use it I'll use it. I do know the difference between a lead acid float voltage and what the charging system makers call float with respect to LFP.

[CaptainRivet]

V

Originally Posted by Captain Bill

All great if you have built your own battery and have total control of your BMS. It also helps if you want to spend your life as a battery engineer. I on the other hand prefer to cruise and enjoy things other than battery engineering. I bought drop in batteries with a built in BMS that doesn't give me a lot of options. I still spent less money than buying another set of lifeline AGMs, the last set of which only lasted three years. The first two sets lasted 7 years. If the cheap built in BMS kills a battery after 3 years I'm still about $200 ahead including upgrading my charging system. If your idea of fun is to babysit a bank of batteries please by all means enjoy yourself. I'll be doing other things. FYI the term float is used by the charging system makers to describe the condition where the charging system is carrying the house load instead of the batteries. Since they use it I'll use it. I do know the difference between a lead acid float voltage and what the charging system makers call float with respect to LFP.

You done the right way for you and having not many option you cannot screw up much.


I enjoy sailing too. I have long experiences with LFP and LTO bein long time in car audio competition and racing cars. We illegally scraped a Lithium battery from a crashed military vehicle over 20 years ago when normal people didn't even know lithium battery exists...no bms,no documentation of any kind so we needed to figure it all out on my own....means i just know how thinks work from experience and also what happens if you screw up.
I also use stuff i know that work, even they might be better out there as things evolve faster i can follow because its just one of my hobbies and yes i like to sail and dicover new places, my priorities change here too.

But the electrical system is very important and guarantees me a lot luxury like in a grid house so i make sure it runs properly and safe with backups as i do world circum also to remote places. And be able to run 600A you really need to execute properly and monitor the system. And yes i will go 12 and 24V soon as 300A is much safer, i had reason to do it like this but realized I made a mistake here even its consider to be safe and will correct it...you never stop learning...
Thats why i chose DIY because i know every bit and whats inside the battery,can fix if out in the ocean if something goes wrong and know why.
I help other moving to lithium or correct **** up installation from so called professional installers for the rum kitty and they help me with other stuff where i have to still learn a lot like rigging. Thats how i see a lot different installs...

[tanglewood]

Originally Posted by CaptainRivet

Total voltage is fake security, at 14.2V one cell can have 3,75V while all other 3.3V or even worse.
Charging must be done on cell level, that why its best the BMS manages the battery and charging and not the charge sources.
Flaot at 13.2V makes no sense with a LFP, if it gets into the knee it reduces the current it takes by itself. An LFP needs no float at all,.you can but it is unnecessary.
A friend of mine i charging his LIfepo4 Calb cells since 17 years with a power supply with 13,85V, BMS cuts off of a cell hits 3,5V. Calbs have still 92% of their original capacity.
I would put 13.8 and 14.2V if you only can on total voltage but add a BMV 712 and use the midbank voltage of the 2nd measurement input to check voltage devation of cell 1+2 versus 3+4. So if devation rasises you avoid overcharging a cell.

It's only an issue if the battery's balancer isn't keeping up, which has been known to happen. But with a good battery, it's really not an issue, and the BMS will let you know (disconnect) if there is really a problem.

[sailorboy1]

Originally Posted by CaptainRivet

Best description i know and not from therory from armchair but practical knowledge.

This thread is over really except for drift life.

But I wonder what the difference between armchair and practical knowledge.

I would say there isn't really.

[s/v Jedi]

The confusion is caused by using old terminology for something different today.

For lead acid, after absorption phase ends, the battery isn’t fully charged yet, for which the float charge cycle was designed. It not only finishes the charging process but also keeps the battery fully charged after that by maintaining a set voltage.

Now LFP. The whole charging of LFP is a mystery for most forum members. Let me explain this for those who missed my previous posts:

Charging LFP is dependent on the charge rate, i.e. how much charge current we throw at it. The manufacturers test this and provide users with numbers for doing a constant current charge with a termination voltage that is valid for the specified charge rate.

Example: charge the cell at a rate of 1C up to 3.65V.

Let’s say this is a 280Ah cell. What they mean is that you must charge it at a rate of 1 x 280 = 280A and stop charging when the voltage reaches 3.65V at which time the cell is fully charged.

This is problematic because we are unable to reach such high charge rates. The complication is that when you lower the charge rate, this also lowers the termination voltage but you don’t know how much. So how do we do this? Spoiler alert: not with over engineered BMS systems that some love to sell for even more than the cost of the cells!

No, we need to determine exactly how our battery is being charged with our charger aboard our boat. Then, we need to adjust settings to tune this the way we need it to perform for optimum use of storage capacity and cell lifespan.

For DIY batteries assembled from individual cells (only connect cells in series, not parallel. Buy bigger cells if needed, or build multiple batteries), this starts with some tools we need to buy if we don’t already have them: a good bench top power supply, a good, accurate multimeter and a Victron Smart battery monitor (the one with display, BMV, not the Smart Shunt).
First thing to do is a top balance of the cells. This is very well explained by Rod Collins on his Marine How-To blog. After this, we connect the cells in series, with the shunt of the BMV in the negative hookup and we know the battery is fully charged.

As we start to discharge, the BMV is registering this and you can watch the Ah used and left as well as the SOC% when you configured the BMV correctly. Continue to discharge the battery to around 60% SOC. Now we can test charging. Set the absorption voltage high, to something like 3.55V and start charging while observing the BMV. Note the battery voltage rising and keep a log on paper for 60%, 65%, 70% etc. and do this for each percent after passing 90%. As soon as you hit 100% note the voltage and stop charging.

Let’s say you find the voltage is 3.45V per cell when you hit 100% but it was 3.45V already at 98%. This is perfect, now you change the charger absorption voltage to 3.45V per cell and you set the BMV “charged voltage” setting to just below that as per manual explains.

Now you cycle again and watch if the charge stops automatically at or just before hitting 100% SOC. Well done, but now we finally get to the float charge.

Most will do this charging with solar and you hope to get the battery recharged fully before the sun goes down. When that’s done, you should want to keep using solar energy instead of wasting it. For this you need to “hack” the float charge.

Start by setting the float voltage at 3.4V per cell. You will find that all consumption comes from the battery that starts discharging while no power from the MPPT is used. This is good, we need to take some of the stress caused by the full charge off the battery. As it discharges a bit, you will see the MPPT starting to supply more power and the battery less, until the battery voltage is down to 3.4V at which time the MPPT fully takes over.
At that point note the SOC%. Also when the sun goes down, note the SOC%. Then the next morning, before the sun is up, note SOC%. If the one at night is below 90% or when your SOC is below 20% in the morning, increase float charge setting to 3.41V

You get the idea. If you can’t make it work like described above, you don’t have enough battery capacity. You will either have to accept less than 20% SOC in the morning or staying at fully charged longer in the afternoon, both of which reduce lifespan. Of course you can also get an extra battery.

If you have a drop-in battery instead of DIY, then only the initial balancing / fully charging is different. Simply follow manufacturers instructions to fully charge it.

Hope this helps

[CaptainRivet]

Originally Posted by s/v Jedi

The confusion is caused by using old terminology for something different today.

For lead acid, after absorption phase ends, the battery isn’t fully charged yet, for which the float charge cycle was designed. It not only finishes the charging process but also keeps the battery fully charged after that by maintaining a set voltage.

Now LFP. The whole charging of LFP is a mystery for most forum members. Let me explain this for those who missed my previous posts:

Charging LFP is dependent on the charge rate, i.e. how much charge current we throw at it. The manufacturers test this and provide users with numbers for doing a constant current charge with a termination voltage that is valid for the specified charge rate.

Example: charge the cell at a rate of 1C up to 3.65V.

Let’s say this is a 280Ah cell. What they mean is that you must charge it at a rate of 1 x 280 = 280A and stop charging when the voltage reaches 3.65V at which time the cell is fully charged.

This is problematic because we are unable to reach such high charge rates. The complication is that when you lower the charge rate, this also lowers the termination voltage but you don’t know how much. So how do we do this? Spoiler alert: not with over engineered BMS systems that some love to sell for even more than the cost of the cells!

No, we need to determine exactly how our battery is being charged with our charger aboard our boat. Then, we need to adjust settings to tune this the way we need it to perform for optimum use of storage capacity and cell lifespan.

For DIY batteries assembled from individual cells (only connect cells in series, not parallel. Buy bigger cells if needed, or build multiple batteries), this starts with some tools we need to buy if we don’t already have them: a good bench top power supply, a good, accurate multimeter and a Victron Smart battery monitor (the one with display, BMV, not the Smart Shunt).
First thing to do is a top balance of the cells. This is very well explained by Rod Collins on his Marine How-To blog. After this, we connect the cells in series, with the shunt of the BMV in the negative hookup and we know the battery is fully charged.

As we start to discharge, the BMV is registering this and you can watch the Ah used and left as well as the SOC% when you configured the BMV correctly. Continue to discharge the battery to around 60% SOC. Now we can test charging. Set the absorption voltage high, to something like 3.55V and start charging while observing the BMV. Note the battery voltage rising and keep a log on paper for 60%, 65%, 70% etc. and do this for each percent after passing 90%. As soon as you hit 100% note the voltage and stop charging.

Let’s say you find the voltage is 3.45V per cell when you hit 100% but it was 3.45V already at 98%. This is perfect, now you change the charger absorption voltage to 3.45V per cell and you set the BMV “charged voltage” setting to just below that as per manual explains.

Now you cycle again and watch if the charge stops automatically at or just before hitting 100% SOC. Well done, but now we finally get to the float charge.

Most will do this charging with solar and you hope to get the battery recharged fully before the sun goes down. When that’s done, you should want to keep using solar energy instead of wasting it. For this you need to “hack” the float charge.

Start by setting the float voltage at 3.4V per cell. You will find that all consumption comes from the battery that starts discharging while no power from the MPPT is used. This is good, we need to take some of the stress caused by the full charge off the battery. As it discharges a bit, you will see the MPPT starting to supply more power and the battery less, until the battery voltage is down to 3.4V at which time the MPPT fully takes over.
At that point note the SOC%. Also when the sun goes down, note the SOC%. Then the next morning, before the sun is up, note SOC%. If the one at night is below 90% or when your SOC is below 20% in the morning, increase float charge setting to 3.41V

You get the idea. If you can’t make it work like described above, you don’t have enough battery capacity. You will either have to accept less than 20% SOC in the morning or staying at fully charged longer in the afternoon, both of which reduce lifespan. Of course you can also get an extra battery.

If you have a drop-in battery instead of DIY, then only the initial balancing / fully charging is different. Simply follow manufacturers instructions to fully charge it.

Hope this helps

Great write up.
Three comments:
1) If you parallel cells in DIY or build separate batteries and then parallel that actualy really doesn't matter as it still reduces the amps a single cell gets during charging your whole bank from your charger output.
So after top balancing you actually connect your whole bank of paralleled cells or paralleled seperate batteries as this is lowering the current per cell. And then you do the process described. Or you divide the charge current by number of paralleled batteries or cells and do the process with the 1/paralleled current with the single cells in series.

2) for the process: if your shore charge or Solar are quite different in output you need to take the current from the charge source that charges the bank in majority. Eg shorecharger does 180A but solar only delivers 100A but charges bank in majority. Then i would run the first part of the process with the shorecharger limited to 100A and do it a 2nd time with 180A and compare the voltages you get. If they are close eg 3.47 for shore and 3,46V solar then just use the 3,46A you got with solar. If they differ big like 3,47 for shore and 3,43 solar then i would set the 3,43V and limit the output of the shorepower charger to the 100A of solar.
You can always manually raise the current of shorecharger if you charge up the bank eg with a portable genny and shorepower from 30 to 80%SOC. But if you are in a habour and connected to shore in 99.9% it doesn't matter if you charge with 100 or 180A. Additionally normally the solar kicks in current too if you connected to shore....
3) if you have a good quality BMS (doesn't mean expensive) you don't have to buy a BMV712, you can use the BMS for the SOC and AH read outs. After top balance connect the BMS and set it to 100% SOC. Then confirm with a good voltmeter the accuracy of the cell voltages displayed. On the other hand the BMV 712 is great last line of defence device in case BMS fails so its a good invest to get it and install as such after this process.

[s/v Jedi]

The reason for using multiple batteries instead of parallel cells within a battery is to prevent most of the balance issues. When all the cells are in series, every cells is guaranteed to get equal charge current.

[Pete7]

Jedi,

I have taken the liberty of taking your post and turn it into a stand alone article for the CF document library as I think its worth retaining long term. Also if the question comes up again we can link to it easily.

https://www.cruisersforum.com/forums...hp?do=cat&id=3

Great succinct explanation

[CaptainRivet]

Originally Posted by s/v Jedi

The reason for using multiple batteries instead of parallel cells within a battery is to prevent most of the balance issues. When all the cells are in series, every cells is guaranteed to get equal charge current.

In normal everyday production it doesn't matter if the cells are in parallel inside one battery or split up into 1p4S/8S and then paralleled. It is still reducing the current per cell in your initial process of getting the cell voltages for 100% SOC charge by 1/number of batteries or cells in parallell.

Why you parallel multiple cells and then put in series and using one BMS versus put one cell in series and then having multiple batteries and multiple BMS each acting individually without one bying the master is another story and thread theme which each pro/cons eg multiple BMS can fool each other...EVs have all 1 BMS and multiple cells in parallel and your 200 or 300AH drop in consits internally in most cases of 2 or 3x100AH cells in parallel and then in series.

[CaptainRivet]

Originally Posted by tanglewood

It's only an issue if the battery's balancer isn't keeping up, which has been known to happen. But with a good battery, it's really not an issue, and the BMS will let you know (disconnect) if there is really a problem.

In case of a DIY full access BMS yes but if you have drop-in like sailorboy you only get to know if its already to late and the battery is in massive imbalance and shuts of very early.
To then fix that without cutting the dropin open and loose all warranties is really a challenge.

[s/v Jedi]

The reason for using the Victron BMV instead of Smartshunt is two-way:

- the BMV has a programmable relay. For example you can program the relay to activate when SOC goes above 60% and deactivate when it goes below 50%, then use the relay contacts to switch a battery charger off and on. This can keep a battery in a 50-60% SOC range while you aren’t using the boat.

- for DIY built batteries or 24V or even 48V banks built from batteries that are connected in series, you can use/reprogram the AUX input on the shunt for midpoint monitoring (see manual). This allows you to set alarms in case of the cells going out of balance and those alarms can be used to program the relay. For example, you can turn on an active balancer or you can stop charging etc.

I recommend to have a dedicated BMV for each battery. If you have multiple batteries in parallel and want to know combined capacity, SOC% etc. (Like what you get to see on a CerboGX display) then put an addition Smartshunt between the negative busbar and the BMV shunts as shown in my reference diagrams. For this SmartShunt you can use the AUX input to monitor a start battery voltage.

[nfbr]

I think the word float is the confusing point.
Float is just a way of saying “what voltage should the charger hold after completing other charge stages”

Li NCM batteries don’t like high SOC for longevity. Tesla / Apple / Android all now have settings to stop charge at lower SOC

LiFePO4 are a different kettle of fish, as Tesla found with the early model 3’s and their problems with incorrect renaining range calculations.
LiFePO4 are much much flatter voltage than Li NCM, so you can’t estimate SOC from Voltage.
They also don’t seem to mind being at higher SOC, though we only have 1000x less data points as there are just not that many millions of them out there.
We we think we know (google) is that holding the Voltage (Not SOC) high reduces their life. So, the closer you hold them to their rest voltage, the long their life. 3.34 great. 3.4 ok, 3.6 not good, 3.8 bad, 4 very bad / swell.
So, read away, but be sure to differentiate your chemistry, know that SOC statements are probably wrong, and that we wont really know for another 5-10 years when enough packs have aged.
They last a very long time if you keep the voltage low.

[nfbr]

The BMV requires a charge to 100% approx weekly to maintain SOC calobration. Charging LiFePO4 via SOC will rapidly drift into meaningless
You have to charge by voltage and find the knee (voltage rise) point to calibrate the SoC
Hey, this Jedi, you do good posts, you know all this

[knutmi]

Read
https://www.canbat.com/how-to-charge-lifepo4-batteries/

Cycles:
Lifepo4 more than 2000. 1 cycles are 10-100%.
I do about 10 full cycles pr year.And about 40 1/2 cycles. So my battery will stay OK in my lifetime. I need power now and often quick.
I prefer charging 14,4 V, 90-95% usage of battery capasity.