270ah DIY LiFePO4 build

[halcyon]

That seems like quite a small question but in fact is potentially massive. Do you mean 12 volt drop-in batteries with an internal battery management system? These are the simplest approach to installing LFP on a boat but, in general, the least suitable.

Thanks, never ask small questions, I am assuming drop in at the moment, waiting to hear from customer.

My current plan is two service batteries LiFiPo, then FLA engine battery, bow battery customer choice, when advise when I learn more. I can monitor loading on both services batteries, warn if one shuts down on charge, if one shuts down on load we can switch out high load equipment to maintain remaining bank under light load till fault corrected. On charge as the engine is always connected so no problem with on charge drop out.

Brian

 

[sailaboutvic]

Thanks, never ask small questions, I am assuming drop in at the moment, waiting to hear from customer.

My current plan is two service batteries LiFiPo, then FLA engine battery, bow battery customer choice, when advise when I learn more. I can monitor loading on both services batteries, warn if one shuts down on charge, if one shuts down on load we can switch out high load equipment to maintain remaining bank under light load till fault corrected. On charge as the engine is always connected so no problem with on charge drop out.

Brian

If its any help my plan is to re wire and have the bow T wired to the starter LA battery as it will only be used when the engine working .

 

[halcyon]

I'll try to answer this part of the question as I think it goes to the heart of the difference between lead acid cells and LFP cells.

For all their considerable advantages LFP cells need constant supervision. Unlike lead acid an LFP cell will keep taking charge until it swells up and over time will go into thermal runaway. It's like those breeds of dogs that never know when they are full and will eat themselves sick. On discharge the cells will not slowly dim the lights but suddenly dip too low and this can permanently damage the cell causing what is sometimes called a 'time bomb' effect. At high temperature the cell capacity may be seriously damaged, the same applies if they are charged below zero degrees. The effects of all these things are either to create long term damage (loss of capacity) or can destroy a cell from the inside. So a battery management system is not looking out for a faulty cell as one might do with lead acid but is a permanent life-support for the cell - it simply cannot function without it. Some people do without a BMS by becoming their own battery manager but that takes a lot of care, understanding and not ever getting distracted when charging or discharging - something that I'm personally incapable of. All this management has to be done at cell level, not at battery pack level as cells can get out of balance. The voltage of a pack under charge may be looking OK at 14.4 volts for example but three of the four cells in series may be at a safe 3.53 volts but the fourth at a damaging 3.81 volts.

A reasonable basic BMS will therefore isolate the circuit if any cell is threatened with over-charge, over-discharge, over-temperature and under-temperature. In addition it should also have some ability to balance the unevenness between the cells (another big subject in its own right).

On a 12 volt drop-in battery all these functions are undertaken by a BMS housed inside the case. But if there is any form of disconnect then the battery is in a non-communicating state neither providing power nor accepting charge - not the kind of thing you want on a boat. If it shuts down with the alternator running then the voltage spike may take out the alternator diodes as well as expensive marine electronics. So the whole question of safe design is how this is all to be managed on a boat - it can be done but not easily or satisfactorily with drop-ins.

The Nordkyn Design articles on my post #1 go into design in depth but the single big question to ask of any LFP battery is what are the compensation methods when it closes down to protect itself. As I've said several times in various threads people properly like the idea of a battery management system but don't always realise that the BMS of a drop-in battery is only concerned with protecting the battery - it has no interest in protecting the user or her or his boat. The lights may go out, all coms lost, the alternator blown, the chart plotter destroyed, the nav lost but the drop-in will be absolutely fine!

Okay, bottom line , we are looking at cell imbalance due to a faulty cell, that is construction or cell voltage imbalance due to charge system. By monitoring each cell you can identify a faulty cell before it becomes a danger or problem.

Couple of things that come to mind, why fit a voltage monitor on each cell, why not just monitor the first cell, current and 4 voltage sense leads to the cell ? This way the module can be potted, giving a much more serviceable marine item. Charging,why has no one built a charger for LiFePo batteries, is there a technical reason I do not see, as there are better ways for this type of cell.

Brian

 

[Poey50]

If I may add as a novice who be researching these batteries for over six weeks day and night, yes I need to get a life,
I was going buy a couple of dropping not only you don't need to know all the stuff about building them or how they work but there can be cheaper but the more I looked into it the more I started to accept how unsuitable are they for boat using heavy loads unless you go for something like master volts or Victron and then your looking at big bucks .
Problem with dropping I found is
1 you don't know how or with what cells they made up with .
2 the BMS are inside and if for any reason that goes pop you probably not going to be able to do much with it.
3 like LA you have v+v- post and every thing goes through the BMS and if the BMS isn't rated high enough putting a ligh load for say a bow T will just cut the BMS out , now for most people with a boat just running the run of the mill stuff that's probably not going to be a problem but for tho running inverter with heavy load it's going to be .

With cells you can over ride the BMS and put in relays , leaving the BMS to look after the cells and the relay to take the big loads why the BMS keeping an eye on the cells and shutting down only when the cells get to low .
But before anyone start to take that route of building your own bank you really need to do a lot of research depend how you plain to use them .
Every day I finding some thing new , why I can't do this or that and having to change my plain .
@Poey50 correct me if I wrote something that's not quite right .
As said still a novice .

Rod Collins - author of the most quoted writing on LFP seems to know more than anyone I know about the subject and refuses to accept that he is an expert. There is so much still to be known. Researching and reading day and night is, I think the only way. You already know more than me about how to manage heavy loads Vic because I haven't needed to get my head round that problem. This is the value of a forum discussion to share knowledge and test ideas.

 

[Poey50]

the last four posts nicely demonstrate why IMHO hybrid systems (as in both LA and LiFePO4) are so much more user friendly and safe
pity (and correct me if I'm wrong) no BMS built for that in mind and user has to do all sorts of custom work with contactors, signalling, etc
OK, hybrid concept is drafted in a couple of docs and an experimental system is running on a boat for over a year now somewhere in Caribbean.
Again IMHO, problem is not having some solar and a decent alternator regulator running LiFePO4, but adding a 2-3KVA inverter in the mix.
Hybrid BMS is my next season's microcontroller project just hope I can get 8X200-280Ah cells for under a grand by then (in EU tax paid)

cheers

V.

That's cutting edge stuff, Vas, but based on a sound foundation which is that with a hybrid system no matter what is happening with the LFP you never have the sudden removal of a battery from the system with all the problems of the destructive nature of a voltage spike.

 

[Poey50]

Couple of things that come to mind, why fit a voltage monitor on each cell, why not just monitor the first cell, current and 4 voltage sense leads to the cell ? This way the module can be potted, giving a much more serviceable marine item.

That one's beyond me. Chris visits here sometimes and may be able to help with that.

Charging,why has no one built a charger for LiFePo batteries, is there a technical reason I do not see, as there are better ways for this type of cell.

There are LFP specialist chargers available which typically charge a 12 volt nominal battery to 14.6 volts and then stop. I have a cheap one from AliExpress that I use at home. It's all bulk charging - no absorption, no float and certainly no equalising. They tend to be used with drop-ins with the recommendation of regular charging which then keeps the cells balanced. Battleborn recommend this with their batteries. 14.6v is now generally seen as too high a voltage target for a variety of reasons covered earlier in this thread so most people use the user-configurable settings on Victron chargers for mains and solar and fully configurable external regulators for the alternator such as the Balmar MC614 (which I have) or the Wakespeed 500.

 

[halcyon]

That one's beyond me. Chris visits here sometimes and may be able to help with that.



There are LFP specialist chargers available which typically charge a 12 volt nominal battery to 14.6 volts and then stop. I have a cheap one from AliExpress that I use at home. It's all bulk charging - no absorption, no float and certainly no equalising. They tend to be used with drop-ins with the recommendation of regular charging which then keeps the cells balanced. Battleborn recommend this with their batteries. 14.6v is now generally seen as too high a voltage target for a variety of reasons covered earlier in this thread so most people use the user-configurable settings on Victron chargers for mains and solar and fully configurable external regulators for the alternator such as the Balmar MC614 (which I have) or the Wakespeed 500.

Thanks, I look at it from design / manufacture point of view, tried to avoid getting into this new tech and semi retire, this is now getting me into.

Looked at TN Power (random choice ) for initial guide on drop in batteries, supprised to find there recommended charge setting basically covered by the mains charger I built in the 90's. Back then we looked at split bank charging 24 volt banks, this allowed all batteries to be charged indipendently. I would have thought that the same principle could be employed to charge each cell, allowing constant cell balancing under charge.

Brian

 

[Poey50]

Thanks, I look at it from design / manufacture point of view, tried to avoid getting into this new tech and semi retire, this is now getting me into.

Looked at TN Power (random choice ) for initial guide on drop in batteries, supprised to find there recommended charge setting basically covered by the mains charger I built in the 90's. Back then we looked at split bank charging 24 volt banks, this allowed all batteries to be charged indipendently. I would have thought that the same principle could be employed to charge each cell, allowing constant cell balancing under charge.

Brian

Go for it .. you might make yourself a fortune!

 

[vas]

Go for it .. you might make yourself a fortune!

...and never retire, nah!

 

[halcyon]

Go for it .. you might make yourself a fortune!

Trying to retire and go sailing, instead at 74 I have started converting all my stuff over to micro control, you are now getting me back to mains chargers I gave them up 20 years ago

Brian

 

[Kelpie]

Some people do without a BMS by becoming their own battery manager but that takes a lot of care, understanding and not ever getting distracted when charging or discharging

I've yet to actually start buying the kit (going to wait until the dust settles after 1st January) but this is to some extent the approach I'm going to take.
A set of four mini digital voltmeters, one for each cell, and a thermometer. Mounted somewhere easily visible for real-time monitoring. That's the first line of defence against over-discharge.

I will use a BMS, but it will be a cheap Daly unit. Primarily there to balance the cells and to act as a second line of defence.
I'm not going to charge directly from the alternator, it sounds too risky without appropriate safeguards in place. So my only charging will be MPPT and B2B, both of which are configured specifically for the lithium battery and should stop charging at the appropriate voltage.
I won't run anything critical (instruments, autopilot) from the lithium battery, because I'm using a basic BMS that could cut the power without warning.
And the main loads I am running from the lithium battery have their own low-voltage disconnect, which is the fist automated line of defence.

It's an unashamedly cheap-skates approach- basically I'm buying cells, cheap BMS, and the B2B (which is optional, if you want to try relying entirely on solar). I'm expecting a total cost of under £750 for 280Ah (not including the MPPT). But it will obviously have limitations and may need a degree of active management. Definitely not suitable for leaving unattended on a boat that is going to sit unused for months at a time.

 

[Poey50]

I've yet to actually start buying the kit (going to wait until the dust settles after 1st January) but this is to some extent the approach I'm going to take.
A set of four mini digital voltmeters, one for each cell, and a thermometer. Mounted somewhere easily visible for real-time monitoring. That's the first line of defence against over-discharge.

I will use a BMS, but it will be a cheap Daly unit. Primarily there to balance the cells and to act as a second line of defence.
I'm not going to charge directly from the alternator, it sounds too risky without appropriate safeguards in place. So my only charging will be MPPT and B2B, both of which are configured specifically for the lithium battery and should stop charging at the appropriate voltage.
I won't run anything critical (instruments, autopilot) from the lithium battery, because I'm using a basic BMS that could cut the power without warning.
And the main loads I am running from the lithium battery have their own low-voltage disconnect, which is the fist automated line of defence.

It's an unashamedly cheap-skates approach- basically I'm buying cells, cheap BMS, and the B2B (which is optional, if you want to try relying entirely on solar). I'm expecting a total cost of under £750 for 280Ah (not including the MPPT). But it will obviously have limitations and may need a degree of active management. Definitely not suitable for leaving unattended on a boat that is going to sit unused for months at a time.

I'm selling a Sterling 60 amp B2B if you are interested. Can be used at 30amps. I bought it on clearance but checked and re-warrantied by Sterling. This was when I was planning to charge the LFP from the start battery via a B2B. That plan changed when my alternator regulator died and I couldn't get a replacement so went straight for the upgrade to Balmar alternator and external regulator. The B2B was then overkill for charging the engine battery from the LFP. PM me if interested ... it won't be expensive and its in top condition.

 

[sailaboutvic]

I've yet to actually start buying the kit (going to wait until the dust settles after 1st January) but this is to some extent the approach I'm going to take.
A set of four mini digital voltmeters, one for each cell, and a thermometer. Mounted somewhere easily visible for real-time monitoring. That's the first line of defence against over-discharge.

I will use a BMS, but it will be a cheap Daly unit. Primarily there to balance the cells and to act as a second line of defence.
I'm not going to charge directly from the alternator, it sounds too risky without appropriate safeguards in place. So my only charging will be MPPT and B2B, both of which are configured specifically for the lithium battery and should stop charging at the appropriate voltage.
I won't run anything critical (instruments, autopilot) from the lithium battery, because I'm using a basic BMS that could cut the power without warning.
And the main loads I am running from the lithium battery have their own low-voltage disconnect, which is the fist automated line of defence.

It's an unashamedly cheap-skates approach- basically I'm buying cells, cheap BMS, and the B2B (which is optional, if you want to try relying entirely on solar). I'm expecting a total cost of under £750 for 280Ah (not including the MPPT). But it will obviously have limitations and may need a degree of active management. Definitely not suitable for leaving unattended on a boat that is going to sit unused for months at a time.

I wouldn't worry about running the auto pilot off a cheap BMS even a 20A BMS will deal with that and normally equipment without a problem , it's went you get into 150A plus things get a bit more difficult my biggest one off load going to be 210/230A which we will use for 20 mins ( water maker) at a time when I have power to waste and time to replace most of it on sunny Med days ,that going to take some managing .
Replace our gas top with an induction hob 110A @12v which will be a bigger problem as it will be running for much longer periods, this all mean we going to have to replace what we take out the next day .its all going to be a balancing act .
How to mange charging on cloudy days is going to be more of a challenge but then it would mean starting the Gen and using the Victron charge at the same time use stuff like the water maker ,
It's seen a lot of work but we been working around batteries that had lost most of their capacity for the last year so we use to that .
And for time when it's not possible to replenish what we taken out it mean going back to using the Gen .
Some may ask why as we got a Gen am I going to run heavy loads of the lithium,
well first of all I can't see any point spend over 2k by the time I finish and running the Gen and wasting fuel , polluting the air plus the noise,
Secondly if I am onLy going to run nav / fridge and Freezer and LED light , with 750w of solar add days when we have to use the engine the lithium are going to hang around 90/95 SOC But what they like is to be cycled.

Personlly if you thinking of just running small load like fridge, light, nav stuff I be looking around 200ah mostly because these batteries do like to be cycled and not sit up on Float .
But then again if that's all I wanted to run I would just stick to cheap LA .
How is all this going to work ? ask me mid season and I let you know .

 

[Kelpie]

I wouldn't worry about running the auto pilot off a cheap BMS even a 20A BMS will deal with that and normally equipment without a problem , it's went you get into 150A plus things get a bit more difficult my biggest one off load going to be 210/230A which we will use for 20 mins ( water maker) at a time when I have power to waste and time to replace most of it on sunny Med days ,that going to take some managing .
Replace our gas top with an induction hob 110A @12v which will be a bigger problem as it will be running for much longer periods, this all mean we going to have to replace what we take out the next day .its all going to be a balancing act .
How to mange charging on cloudy days is going to be more of a challenge but then it would mean starting the Gen and using the Victron charge at the same time use stuff like the water maker ,
It's seen a lot of work but we been working around batteries that had lost most of their capacity for the last year so we use to that .
And for time when it's not possible to replenish what we taken out it mean going back to using the Gen .
Some may ask why as we got a Gen am I going to run heavy loads of the lithium,
well first of all I can't see any point spend over 2k by the time I finish and running the Gen and wasting fuel , polluting the air plus the noise,
Secondly if I am onLy going to run nav / fridge and Freezer and LED light , with 750w of solar add days when we have to use the engine the lithium are going to hang around 90/95 SOC But what they like is to be cycled.

Personlly if you thinking of just running small load like fridge, light, nav stuff I be looking around 200ah mostly because these batteries do like to be cycled and not sit up on Float .
But then again if that's all I wanted to run I would just stick to cheap LA .
How is all this going to work ? ask me mid season and I let you know .

I'm fitting a pretty large (3kw) inverter to power a few high current devices. None of these would be run unattended so in theory they could bypass the BMS, but for simplicity I'm just going to fit a large Daly (250-300A).
I've got a few reasons for not running the autopilot off the lithium battery. Firstly, we have a Hydrovane. The AP will likely be used when we are motoring, in which case the SLA bank will be charging off the alternator. Secondly, the 'dumb' BMS could cut off power with no warning, which could be a bit too exciting. And thirdly, it means leaving the existing systems untouched.

I'm planning on 280Ah, these cells are the best value for money at the moment. When I have abundant power I'll get to cook with induction, and when I don't I'll burn some gas.

 

[crisjones]

Hi,
Not checked this thread for a while but see there has been quite a bit of activity. A few comments on what has been discussed:-

Initial balancing has come up in a few posts – my advice would always to do as comprehensive a top-balance as you possibly can when you get your cells. Take the cell voltage up to the highest recommended voltage for your cells. A very good detailed walk-through on the process on MarineHowTo. Also discussions on how to do it without a power supply although I would definitely recommend using a decent bench top power supply – it makes the process much easier, more accurate and more controllable. The cost of a suitable power supply is pretty small in the overall cost of a LiFePo4 system and it is surprising how useful it is once you have it.

When you have got your cells properly balanced then connect in their final configuration and then discharge to about 50%. They can then sit around for days or weeks without worries prior to final install. If you start with properly balanced cells then routine or automatic balancing is not really necessary. My Winston cells have been in constant liveaboard use for almost 2 years and cell voltages are still within less than 5 mV at all states of charge and loading – they have not been balanced at all since install. Maybe previously used cells or lesser quality new cells may benefit from routine balancing – top quality new cells should not need it after a proper initial top-balance.

BMS choice – plenty out there of course, all with pro’s and con’s. For boat installs you really must have a BMS that controls a separate main contactor, this means you can choose a contactor that is large enough to deal with any loads you may draw from the battery bank (inverter / bow thrusters etc). It is also essential that any BMS monitors individual cell voltages and reacts to over voltage or under voltage events at cell voltage level – DO NOT consider a BMS that cannot meet this criteria. Any BMS that has a max current limit of, say 100A, is useless and will only cause you grief in the future since it will always rely on some manual intervention if you want to run high loads.

The BMS units that control a separate main contactor can usually deal with systems that can cope with loads up to 500A – this should be more than enough for any 12V boat system even with high inverter loads. I would suggest that you design a boat system with 500A as the max current limit – it leaves enough overhead in all the components to give long life expectancy and scope for coping with any unplanned overloads on the system – and these overloads will inevitably happen no matter how you think you have eliminated the possibility (Sod’s Law applies here).

Our system has a 500A main contactor with a discharge limit of 300A and a charge limit of 400A, these two limits are programmed into the BMS and could be changed up to the 500A limit of the contactor if I wished.

Try and choose a BMS that can also give signals to the charging or load sources in your system – this means that the BMS can send a signal to turn off these sources BEFORE the BMS activates the main contactor, in this way the main contactor becomes the “last line of defence” and in a properly designed system it should never operate in anything but exceptional or dangerous circumstances. This should also mean that you have almost minimal risk of being left without any power for navigation purposes and also minimises risk of damage to alternators due to an unplanned disconnect. Nordkyn Design goes into the design of such a system in great detail and it should be read two or three times until you fully understand the implications of your design decisions.

The BMS should just do it’s job without any user intervention – almost all now allow constant monitoring by phone app or computer – that is fine and good to see what is going on, however you should have total confidence in your system such that you never look at anything other than SOC for weeks on end.

I will repeat the most important aspect – design your system so that the main contactor is the last line of defence and should never operate in anything but exceptional or dangerous circumstances.

I have mentioned it before but I will say again that I have the Orion Jr BMS, it is not the cheapest option but it is extremely good with many signal output relays to give excellent control over all other parts of the system. The monitoring and configuration software is very good and it also comes in CanBus versions if required. Technical support from Ewert Energy has been prompt and comprehensive whenever I have emailed any questions.

At the moment there are no suitable LiFePo4 cells with integral BMS that could be considered to be “Drop-In” replacements for lead acid batteries without significant changes to the rest of the boat’s electrical system to ensure long life and reliability. They are also very expensive for what they offer.

Quite a few mentions of hybrid systems in the posts – I have a pure LiFePo4 system so I have no experience of the hybrid system. My small amount of research indicates that it is a viable solution with some advantages but also disadvantages. Personally I prefer to have only one type of battery in the system (ignoring start battery). If someone was designing a complete system from scratch they would be unlikely to choose a hybrid system, it would be pure LiFePo4, hence it is my belief that whenever possible that is the best choice for retrofit also.

I am sure the hybrid system is useful in some circumstances and can avoid re-configuration of other parts of the system in an effort to make the install simpler and that is OK if you have properly understood the pro’s and con’s. I think it would be a mistake to believe that a hybrid system can turn LiFePo4 cells into a “Drop-In” type of replacement – it is not.

Main Contactor – some mention in the posts of using latching type main contactors or flip-flop type relays. These do give the advantage of zero power consumption once they are switched however they cannot be considered “fail-safe” and are probably not the best choice for the last line of defence in a LiFePo4 system. Nordkyn Design makes it clear that he believes any system should be designed to be “Fail-Safe” and I certainly agree with him.

With a latching relay it will require some type of signal to switch off – what happens if the signal wire is broken or disconnected or the switching transistor in the BMS fails? In these circumstances the latching relay will not switch off and you could potentially end up with a ruined battery pack – is that what you want?

High Current main contactors are available with dual coils or other control circuitry that means the current draw to hold the contacts closed is much less than that required to initially operate the contactor. I have a Gigavac GV140 main contactor and it draws around 0.1A at 12V to keep the contacts closed – so just 2.5Ah per day.

Obviously it is sensible to consider these “parasitic” current draws when designing the system and there may also be other relays in the system that are “always on” so it is possible that it can rise to a noticeable level in smaller systems, however it is important to decide if saving those few Ah is worth the trade off for not having a fail-safe system. On a liveaboard boat with suitable solar charging a parasitic load of less than 5Ah per day should not cause any concern and you have the benefit of a fully fail-safe system if you do your design work properly.

The above is from my 2 years full time liveaboard experience and countless hours of research, like Rod Collins I do not consider myself to be any sort of “expert” in LiFePo4 battery systems, I just hope that my experience is of some use to some of you.

 

[Poey50]

Hi,
Not checked this thread for a while but see there has been quite a bit of activity. A few comments on what has been discussed:-

Initial balancing has come up in a few posts – my advice would always to do as comprehensive a top-balance as you possibly can when you get your cells. Take the cell voltage up to the highest recommended voltage for your cells. A very good detailed walk-through on the process on MarineHowTo. Also discussions on how to do it without a power supply although I would definitely recommend using a decent bench top power supply – it makes the process much easier, more accurate and more controllable. The cost of a suitable power supply is pretty small in the overall cost of a LiFePo4 system and it is surprising how useful it is once you have it.

When you have got your cells properly balanced then connect in their final configuration and then discharge to about 50%. They can then sit around for days or weeks without worries prior to final install. If you start with properly balanced cells then routine or automatic balancing is not really necessary. My Winston cells have been in constant liveaboard use for almost 2 years and cell voltages are still within less than 5 mV at all states of charge and loading – they have not been balanced at all since install. Maybe previously used cells or lesser quality new cells may benefit from routine balancing – top quality new cells should not need it after a proper initial top-balance.

BMS choice – plenty out there of course, all with pro’s and con’s. For boat installs you really must have a BMS that controls a separate main contactor, this means you can choose a contactor that is large enough to deal with any loads you may draw from the battery bank (inverter / bow thrusters etc). It is also essential that any BMS monitors individual cell voltages and reacts to over voltage or under voltage events at cell voltage level – DO NOT consider a BMS that cannot meet this criteria. Any BMS that has a max current limit of, say 100A, is useless and will only cause you grief in the future since it will always rely on some manual intervention if you want to run high loads.

The BMS units that control a separate main contactor can usually deal with systems that can cope with loads up to 500A – this should be more than enough for any 12V boat system even with high inverter loads. I would suggest that you design a boat system with 500A as the max current limit – it leaves enough overhead in all the components to give long life expectancy and scope for coping with any unplanned overloads on the system – and these overloads will inevitably happen no matter how you think you have eliminated the possibility (Sod’s Law applies here).

Our system has a 500A main contactor with a discharge limit of 300A and a charge limit of 400A, these two limits are programmed into the BMS and could be changed up to the 500A limit of the contactor if I wished.

Try and choose a BMS that can also give signals to the charging or load sources in your system – this means that the BMS can send a signal to turn off these sources BEFORE the BMS activates the main contactor, in this way the main contactor becomes the “last line of defence” and in a properly designed system it should never operate in anything but exceptional or dangerous circumstances. This should also mean that you have almost minimal risk of being left without any power for navigation purposes and also minimises risk of damage to alternators due to an unplanned disconnect. Nordkyn Design goes into the design of such a system in great detail and it should be read two or three times until you fully understand the implications of your design decisions.

The BMS should just do it’s job without any user intervention – almost all now allow constant monitoring by phone app or computer – that is fine and good to see what is going on, however you should have total confidence in your system such that you never look at anything other than SOC for weeks on end.

I will repeat the most important aspect – design your system so that the main contactor is the last line of defence and should never operate in anything but exceptional or dangerous circumstances.

I have mentioned it before but I will say again that I have the Orion Jr BMS, it is not the cheapest option but it is extremely good with many signal output relays to give excellent control over all other parts of the system. The monitoring and configuration software is very good and it also comes in CanBus versions if required. Technical support from Ewert Energy has been prompt and comprehensive whenever I have emailed any questions.

At the moment there are no suitable LiFePo4 cells with integral BMS that could be considered to be “Drop-In” replacements for lead acid batteries without significant changes to the rest of the boat’s electrical system to ensure long life and reliability. They are also very expensive for what they offer.

Quite a few mentions of hybrid systems in the posts – I have a pure LiFePo4 system so I have no experience of the hybrid system. My small amount of research indicates that it is a viable solution with some advantages but also disadvantages. Personally I prefer to have only one type of battery in the system (ignoring start battery). If someone was designing a complete system from scratch they would be unlikely to choose a hybrid system, it would be pure LiFePo4, hence it is my belief that whenever possible that is the best choice for retrofit also.

I am sure the hybrid system is useful in some circumstances and can avoid re-configuration of other parts of the system in an effort to make the install simpler and that is OK if you have properly understood the pro’s and con’s. I think it would be a mistake to believe that a hybrid system can turn LiFePo4 cells into a “Drop-In” type of replacement – it is not.

Main Contactor – some mention in the posts of using latching type main contactors or flip-flop type relays. These do give the advantage of zero power consumption once they are switched however they cannot be considered “fail-safe” and are probably not the best choice for the last line of defence in a LiFePo4 system. Nordkyn Design makes it clear that he believes any system should be designed to be “Fail-Safe” and I certainly agree with him.

With a latching relay it will require some type of signal to switch off – what happens if the signal wire is broken or disconnected or the switching transistor in the BMS fails? In these circumstances the latching relay will not switch off and you could potentially end up with a ruined battery pack – is that what you want?

High Current main contactors are available with dual coils or other control circuitry that means the current draw to hold the contacts closed is much less than that required to initially operate the contactor. I have a Gigavac GV140 main contactor and it draws around 0.1A at 12V to keep the contacts closed – so just 2.5Ah per day.

Obviously it is sensible to consider these “parasitic” current draws when designing the system and there may also be other relays in the system that are “always on” so it is possible that it can rise to a noticeable level in smaller systems, however it is important to decide if saving those few Ah is worth the trade off for not having a fail-safe system. On a liveaboard boat with suitable solar charging a parasitic load of less than 5Ah per day should not cause any concern and you have the benefit of a fully fail-safe system if you do your design work properly.

The above is from my 2 years full time liveaboard experience and countless hours of research, like Rod Collins I do not consider myself to be any sort of “expert” in LiFePo4 battery systems, I just hope that my experience is of some use to some of you.

Excellent contribution, Chris. Thanks.

 

[Kelpie]

For boat installs you really must have a BMS that controls a separate main contactor,

This has been the main issue that I have been trying to make a decision on.
I'm leaning towards the relatively cheap and simple option of a high current BMS, e.g. a Daly 300A.
The inverter can in theory pull 250A but I would try to keep it below 200A, so that there is some headroom in the system.

The alternative would be a lower rated BMS and a separate contactor. I haven't ruled that out, but it was starting to make my head hurt choosing the contactor, and it seems as though I'd still be using a Daly or similar BMS unless I wanted to triple the budget on that component.

 

[sailaboutvic]

I personally know Chris for some years now and we did cruisers to gather for a time, a man with lots of experience when it come to lithium, not only he set his own boat but have helped other to do the same .
So Chris good you popped in again with your advise .
Your first par i sure is aimed at me so you be glade to hear i ordered one this morning

 

[steveallan]

I'm selling a Sterling 60 amp B2B if you are interested. Can be used at 30amps. I bought it on clearance but checked and re-warrantied by Sterling. This was when I was planning to charge the LFP from the start battery via a B2B. That plan changed when my alternator regulator died and I couldn't get a replacement so went straight for the upgrade to Balmar alternator and external regulator. The B2B was then overkill for charging the engine battery from the LFP. PM me if interested ... it won't be expensive and its in top condition.

PM sent

 

[Pete7]

My current plan is two service batteries LiFiPo, then FLA engine battery, bow battery customer choice, when advise when I learn more. I can monitor loading on both services batteries, warn if one shuts down on charge, if one shuts down on load we can switch out high load equipment to maintain remaining bank under light load till fault corrected. On charge as the engine is always connected so no problem with on charge drop out.
Brian

That's why a hybrid system appeals to me. Something like a gel battery in parallel with the LFP. BMS shuts down the FLP battery and the gel takes the load. Sure the kettle, hair dryer and windlass won't work. However, if you are at sea, then VHF chart plotter and AP should continue for sometime. If the engine or solar also continues to charge what's left of the house bank, it may not be a complete snafu.