Changing to LiFePo Batteries

The friendliest place on the web for anyone who enjoys boating.
If you have answers, please help by responding to the unanswered posts.

Bob Cofer

Guru
Joined
Aug 27, 2013
Messages
1,522
Location
PNW
Vessel Name
Puffin
Vessel Make
Willard Vega 30
The time has come to replace my AGM house bank and I have decided to go with a LiFePo bank. Additionally I am changing my starter battery to an LiFePo. Fortunately my system is set up with an AC charger and alternator/regulator that can accommodate this. The only issue is the ACR (Blue Sea 7618) that is installed does not have the proper voltage sensing for the new batteries. I will be removing the ACR for now until they come out with their new version later this year. In the meantime I will be going back to the old fashioned way of actually having to use a selector switch.

Attached is a drawing of my new system, overcurrent protection is not shown for clarity. This is not the optimal solution but will but will work for now. The batteries are all from the same manufacturer.
 

Attachments

  • Puffin LiFePo DC Distribution.pdf
    1.4 MB · Views: 96
Are you sure an Li battery can supply the current required to start your engine.

Also have you considered the load that an Li house bank will put on your engine’s alternator?

David
 
Bob - does your Willard still have a Perkins 4.107/4.108? What start battery are you going with? I see two brands in the market - Noco and Antigravity. I bought an Antigravity for my 20hp Tohatsu and love it. When I eventually make it back to Florida, I might go with an LFP start battery for my 4.236. I'm very curious to see others doing it too. I forget what thread I replied to, but engine room heat in lower latitudes is killing my engine start battery so hoping the higher operating temp of LFP might do the trick. And of course can get rid of the DC-DC charger and associated cabling.

Inquiring minds want to know.....

Peter
 
What start battery have you picked? It's good to see them starting to be available as start batteries.

Personally I would replace the ACR with a DC/DC charger for the start battery. It won't take much to recharge it, and it would allow you to get rid of the need to operate the big selector switch which to me is a problem waiting to happen due to operator error.

Will you be able to locate the batteries away from engine heat?

Do your batteries provide some sort of disconnect warning that can be used to turn off the alternator?
 
I didn’t think it was a good idea to connect those type of battery when they’re at significantly different levels of charge. Might want to rethink the 1-2-all switch?
 
Bob - does your Willard still have a Perkins 4.107/4.108? What start battery are you going with? I see two brands in the market - Noco and Antigravity. I bought an Antigravity for my 20hp Tohatsu and love it. When I eventually make it back to Florida, I might go with an LFP start battery for my 4.236. I'm very curious to see others doing it too. I forget what thread I replied to, but engine room heat in lower latitudes is killing my engine start battery so hoping the higher operating temp of LFP might do the trick. And of course can get rid of the DC-DC charger and associated cabling.

Inquiring minds want to know.....

Peter
Puffin has been repowered with an Isuzu C240.

This is the start battery I will be using, 1000cca, 105ah: Amazon.com

Interested to see how it works, keeping my good old Optima Yellow Top on standby.
 
Will Prowse reviewed Weize regular LFP battery - bit of a mixed bag. One of the Amazon reviews states the battery was unable to start his Toyota Tacoma truck but Weize made good on it including shipping. Might want to buy it at last minute so you can return to Amazon within 30-day period.

Good luck Bob - Peter
 
I wish I could help you with this. I simply am not up to speed on LFP starting batteries. My question for you would be what are the issues with mixing LFP starting with LFP house. I realize that you can have the same dual purpose LFP in both but not sure of the issues charging both at mixed levels of charge.

Using Dc to Dc chargers would probably solve this problem. I look forward to seeing your experience.
 
I didn’t think it was a good idea to connect those type of battery when they’re at significantly different levels of charge. Might want to rethink the 1-2-all switch?
Yes, good point, and another reason not to have a switch other than an emergency/maintenance disconnect.
 
How is the alt protected from sudden charge shutdown when battys reach full charge?
My understanding is others use DC-DC charger but they retain FLA or AGM as start battys that taper the charge as they reach 100% SOC.
 
How is the alt protected from sudden charge shutdown when battys reach full charge?
My understanding is others use DC-DC charger but they retain FLA or AGM as start batteries that taper the charge as they reach 100% SOC.
Just to keep this in perspective, the only time you will get a sudden shutdown or disconnect of the batteries is in some sort of fault condition. So it's wise to protect against it, but in a properly operating system it never happens.

To me, the preferred way to deal with this is with a signal from the batteries warning of an imminent disconnect, and that signal is used to simply turn off the alternator. The signal from the battery can be an actual disconnect warning, or even just an Allow to Charge (ATC) signal. The trouble is that many (maybe most) drop in batteries do not provide any such signal, so you are running blind.
 
Installed the batteries yesterday and it went pretty well. I removed the ACR as it's not compatible with Li batteries and will replace it with one that is, in the meantime it will be manually switched when needed. The start battery worked very well, ran the glow plugs for 15 seconds then started the engine with zero issues.

Attached is the information provided with the batteries. The overvoltage disconnect was my biggest concern. It is 15.4v so I set the regulator to stop charging at 15v to protect the alternator. I am also installing an alternator protection device as a secondary safety. Time will tell how this system holds up. The experimentation is interesting to me.
 

Attachments

  • Battery Photo.jpg
    Battery Photo.jpg
    157.7 KB · Views: 46
  • BMS Parameters.pdf
    560.8 KB · Views: 50
Bob, if your set up was FLA I would say you wired your batteries incorrectly. If it was FLA I would recommend you move your house positive to the opposite end of the bank.

I am not sure if it is as big an issue with LFP. Just curious what others think.

I look forward to hearing about your experience with this new arrangement.
 
Bob, if your set up was FLA I would say you wired your batteries incorrectly. If it was FLA I would recommend you move your house positive to the opposite end of the bank.

I am not sure if it is as big an issue with LFP. Just curious what others think.

I look forward to hearing about your experience with this new arrangement.
Went with the manufacturers recommendation on wiring configuration, time will tell.
 
Was that verbal or a document I can look up? Just curious.

Eco-worthy wanted me to do equal lengths to a bus bar but showed a 4’ bus bar and didn’t care about unequal attachment on the bus bar. Their second recommendation was for equal length path among all batteries.

This is not meant to be a commentary on how you did things. I am just noting that there appears to be a lack of consistency between battery suppliers.
 
Was that verbal or a document I can look up? Just curious.

Eco-worthy wanted me to do equal lengths to a bus bar but showed a 4’ bus bar and didn’t care about unequal attachment on the bus bar. Their second recommendation was for equal length path among all batteries.

This is not meant to be a commentary on how you did things. I am just noting that there appears to be a lack of consistency between battery suppliers.
It was actually on a decal attached to the top of the battery. I didn't take a photo of the schematic prior to removal.

As an industry standard/best practices installation on large scale systems:
- The load connections are at one end of the bank and the charge connections are at the other end.
- Positive terminals are connected via equal length bussing as are the negative terminals.
- Equal length cabling is used on both charge and load connections.

As with all things there are compromises.

No offense taken! The whole point is discussion and education.

Regards,
Bob
 
Good discusion. Looking forward to replacing my start batteries with LifePo. I am very impressed with my house bank LifePo.
 
Here is why I put positive at one end and negative at the other. There are more complicated ones but this works well enough. Look at what happens to charging when +/- are at one end of the bank.
 

Attachments

  • Battery balancedcharging.pdf
    517.1 KB · Views: 56
Here is why I put positive at one end and negative at the other. There are more complicated ones but this works well enough. Look at what happens to charging when +/- are at one end of the bank.
This is how I have always wired battery banks. With FLA the document is completely correct. However, while still correct for LFP, I am not sure if the difference has any relevance to the life of an LFP battery.

It is the belief of some that we will all suffer BMP failures long before the actual batteries degrade in any meaningful way.
 
If you want to connect your existing alternators to LiFePO4 batteries you are going to need some form of controller that can be programmed.
You will need it for two reasons:

1. A LiFePO4 battery can take a max capacity charge of the alternator until it is full and then it will shut down all of a sudden. Charging at full capacity can mean the alternator can overheat, so you will have to limit the max charge of the alternator to around 80 %. That way it won't overheat and your battery will charge at a high level.

2. When the LiFePO4 batteries are full the alternator needs to stop charging, which means it needs to be cut out of the loop. If you don't do that you will get a power surge in your electrical system, potentially frying a lot of equipment. There are regulators such as Wakespeed WS500, but there are also others.

Reason for the cables having to be at exactly the same length is that otherwise the battery with the longer cables is not going to be used as much or even at all. The max difference in cable length is less than 1 cm.

Also, before you connect the terminals, clean them all very well, make sure there is no grease or sweat on them. That would influence the connection and cause major problems later on.

Another important item is the battery balancer. Depending on which BMS you have (and whether the batteries have an internal balancer) you will need to allow time for balancing.
A LiFePO4 battery will first charge, but the individual cells won't be balanced. After the battery is full the balancing will start. That can take quite a while and if you would start discharging at that time the battery cannot deliver max capacity.
A battery with an internal BMS will have less problems with balancing than a set of batteries which is balanced by one BMS only.

Changing to LiFePO4 does have a lot of consequences, but in the end it is definitely worth it.
 
I don't worry about this stuff with LiFePO4. It doesn't really matter if all the batteries are the same, they enjoy "life in the middle"

Set up charge bulk less than max and set float less than max. Makes them last longer. They don't like the highs or the lows. They aren't going to sulfate either.

I have 2 in parallel on House 2 and another on House 1. If I set helm switch to all they are all parallel. Vastly different cable lengths, but the charge current is pretty close and the charger only sees the net at charger and is set well below max.

It's different than lead acid where stagnation and sulfation are concerns.
 
My LFP battery manual says to connect opposite ends, I just assumed others say as well.
Eco-worthy says connecting opposite ends is acceptable but not preferred. It recommends equal lengths to a bus bar.

While it might be technically best to use the bus bar method, I wonder if there is any real world noticeable loss by doing it the way Bob did or they way SteveK and I have done it. Certainly if in doubt choose the best method but in Bob’s case he felt other spacing issues were of concern. He is willing to risk battery life issues. I’m not certain that he will ever experience a noticeable degradation in his battery life.
 
According to my smart batteries, the opposite end sends a balance at start, then each battery distributes to each cell to get a balanced charge, then since all batteries are ethernet connected they assist each other by directing the charge. BMS at work. Will Prowse explains it better.
 
If you want to connect your existing alternators to LiFePO4 batteries you are going to need some form of controller that can be programmed.
You will need it for two reasons:

1. A LiFePO4 battery can take a max capacity charge of the alternator until it is full and then it will shut down all of a sudden. Charging at full capacity can mean the alternator can overheat, so you will have to limit the max charge of the alternator to around 80 %. That way it won't overheat and your battery will charge at a high level.

2. When the LiFePO4 batteries are full the alternator needs to stop charging, which means it needs to be cut out of the loop. If you don't do that you will get a power surge in your electrical system, potentially frying a lot of equipment. There are regulators such as Wakespeed WS500, but there are also others.

Reason for the cables having to be at exactly the same length is that otherwise the battery with the longer cables is not going to be used as much or even at all. The max difference in cable length is less than 1 cm.

Also, before you connect the terminals, clean them all very well, make sure there is no grease or sweat on them. That would influence the connection and cause major problems later on.

Another important item is the battery balancer. Depending on which BMS you have (and whether the batteries have an internal balancer) you will need to allow time for balancing.
A LiFePO4 battery will first charge, but the individual cells won't be balanced. After the battery is full the balancing will start. That can take quite a while and if you would start discharging at that time the battery cannot deliver max capacity.
A battery with an internal BMS will have less problems with balancing than a set of batteries which is balanced by one BMS only.

Changing to LiFePO4 does have a lot of consequences, but in the end it is definitely worth it.
Item 1 is not really a real world concern. On my boat there is never a time when 12v isn’t being used so even if the BMS were to shut off there would never be a power spike to the alternator.

Item 2 is a real world concern for anyone trying to charge a well depleted LFP bank with alternators. This is why we recommend alternators being directed to a FLA or AGM start bank and DC to DC chargers to pass excess power to the LFP bank.
 
If you want to connect your existing alternators to LiFePO4 batteries you are going to need some form of controller that can be programmed.
You will need it for two reasons:

There are a bunch of good points here, but also potentially misleading without more context

1. A LiFePO4 battery can take a max capacity charge of the alternator until it is full and then it will shut down all of a sudden.
This needs clarification. When LFP batteries reach full charge the voltage rises rapidly and you need to stop charging current before the cells go over voltage. So in that sense the alternator output will reduce/stop quickly.

It’s also true that if charging drives the batteries over voltage, the BMS should disconnect the batteries. If this sort of a “shutdown” is occurring, then there is a problem with the system, and the charging parameters in particular. A BMS shutdown is a fault condition, and should never happen in normal operation.

Charging at full capacity can mean the alternator can overheat, so you will have to limit the max charge of the alternator to around 80 %. That way it won't overheat and your battery will charge at a high level.
That’s one way to do it. Alternately, most good external regulators have alternator temp sensors, and the regulator will limit output to not exceed a safe operating temp for the alternator.

2. When the LiFePO4 batteries are full the alternator needs to stop charging, which means it needs to be cut out of the loop. If you don't do that you will get a power surge in your electrical system, potentially frying a lot of equipment. There are regulators such as Wakespeed WS500, but there are also others.
When the batteries are full, then regulator need to reduce current, and in time switch to a lower float voltage. There is no need to disconnect it, if that’s what’s meant by “cutting it out of the loop”. If fact, an abrupt disconnect is exactly what creates a possible surge. Any regulator will reduce current in a controlled manner to maintain the programmed voltages. It’s what they do in life.

Reason for the cables having to be at exactly the same length is that otherwise the battery with the longer cables is not going to be used as much or even at all. The max difference in cable length is less than 1 cm.

Also, before you connect the terminals, clean them all very well, make sure there is no grease or sweat on them. That would influence the connection and cause major problems later on.

Another important item is the battery balancer. Depending on which BMS you have (and whether the batteries have an internal balancer) you will need to allow time for balancing.
A LiFePO4 battery will first charge, but the individual cells won't be balanced. After the battery is full the balancing will start. That can take quite a while and if you would start discharging at that time the battery cannot deliver max capacity.
In what way wouldn’t the batteries be able to deliver full capacity? They could certainly deliver full current, and do so for all or nearly all of the SOC range. I suspect what you mean is that with one or more cells out of balance, the battery may appear full discharged when the lowest cells go flat even though there is charge left in other cells. This is true, but unless the cells are massively out of balance, it’s immaterial.

A battery with an internal BMS will have less problems with balancing than a set of batteries which is balanced by one BMS only.
There are lots of different balancing circuits with varying levels of performance, but I don’t think there is any inherent performance advantage in internal vs external. So i don’t think it’s a good basis for judging the performance of a balancer.


Changing to LiFePO4 does have a lot of consequences, but in the end it is definitely worth it.
Totally agree.
 
In what way wouldn’t the batteries be able to deliver full capacity? They could certainly deliver full current, and do so for all or nearly all of the SOC range. I suspect what you mean is that with one or more cells out of balance, the battery may appear full discharged when the lowest cells go flat even though there is charge left in other cells. This is true, but unless the cells are massively out of balance, it’s immaterial.
What I was told by Victron and which you can also find in many Youtube videos on LiFePO4 cables, batteries etc, is that in a multiple battery set up the batteries that have cables which are too long or which don't make good contact, will be less discharged. In other words, the system starts favoring the battery with the least resistance up to the point where you can have 1 battery empty while another one is still almost full. The BMS will then shut everything down apparently.
Now, I have not tried this myself, so no idea if these claims are true or false, but since it does not cost that much to clean the connections before connecting we simply cleaned them.

As for the battery not being balanced that is something I can see on my app as well. When I did the first charge of the battery I took a look at the voltages and temperatures of the cells. They were far off and it took quite some time before each battery was fully balanced. I was warned for this and told that I could influence the lifespan of the battery if I did not balance the battery completely. On top of that, if the battery would not balance I would have to send it back to the manufacturer.

So all I do now is check the battery status, check the voltage and temperature of the individual batteries (7 in total) and make sure that it shows balanced.
I let the BMS inside the batteries and the main BMS plus the Wakespeed regulators for the alternators (1 for each alternator) make sure that all goes well. How they communicate I cannot tell you, am not a software engineer, but all I can say it works well. (LOL).
 
I have mentioned my set up several times in past posts. But for the new comers, this may or may not help.

To protect the ALT I wired it the AGM starting bank. Than used a DC2DC charger (60A total) to charge the LifeP04 house bank. I used 2 bus bars Neg and Pos with 3' cables to each battery. I can Bluetooth into each battery to check Voltage, Amps, Temp and more.

This solved a few problems, No need to change out the ALT or regulator, ALT protection, if the house bank totally died I can run the house off the starting bank. This is the 4th season and more confident that the BMS will not shut down.

Later, when LifeP04 become a thing used in starting. I will upgrade the regulator and wire the Alt back to the house bank. We shall see...................
 
To protect the ALT I wired it the AGM starting bank. Than used a DC2DC charger (60A total) to charge the LifeP04 house bank.
I've seen this solution a lot, and it makes sense, but I usually see it in the context of a single engine, hence single alternator, configuration.

I've got two engines. One has a higher-capacity, externally-regulated alternator to charge the house bank, while the other has a stock alternator and charges the starting bank. It's not really clear to me if a DC-DC charger would benefit me at all.

Each engine has one wire from the battery selector switch to both the starter and alternator. And since I can (usually) start fine off the house bank, there's no need to flip switches when starting or stopping the engines.

My current thought is that my first step toward a LFP migration would be to run separate wires for charging and starting. I'm going to have to draw those circuits out and figure out the best way to accomplish that.

Thoughts?
 
Back
Top Bottom