270ah DIY LiFePO4 build

[Poey50]

We still experimenting we found whats works best at the moment , abs 14v float 13.5v seen to be about right dropping the abs on13.8v on sunny days , soon we be using less induction hob and mico waves the electric rad as already been put away so we won't be using so much then and I will set Abs to 13.8v no point charging higher if I don't need to.
once a month I increase charging to Abs 14.1v .

At 14.1v the different in cells are nothing , as it get close to 14.2 cells 3 slowly start to drift 14.3v would see cells 3 hitting 3.65 while the others are within 0.01 of each of the other cells
Whats strange and it may be normal is cell 3 is always .01/.02 behind all the other cells until it reaches 14.2v
I not tested at 14.4v as I know cell 3 will be pass 3.65v
See photos at 14v

Our 8 cells are in a 2P4S so we can use just one bms in reality each module it reading 2 cells as one, now if thats causing the different in reading as the cell get close to 3.65 dont know .

I wouldn't say I'm dissatisfied they seen to be working OK but for the money its more of a diy kit then a complete bms , Module are very easy to damage and as you know you have to go to the expense of adding relays to protect the batteries where many bms have protection build in .
If you add the full cost of the 123 plus relays you can buy a much better quality like the orion, which is what i would go for if I did it again.

Personally I don't thing 123smart are suitable for boats , on land sure even a RV .

Vic,

A few points ...

Keep in mind that the quoted accuracy of voltage measurement for the 123SmartBMS is + or - 20mV so that 0.01 or 0.02 difference you are seeing may or may not be significant. What is the largest difference between cells when the first reaches 3.65 volts?

As far as modules go, if you remove the outside case of the top of the range Victron lithium battery you will find pretty much the same thing (see below). The boards just need to be protected from any external damage. I have a cover over my pack so they can't get damaged. Integrated units also have thin wires leading to the terminals that need protecting - in the case of the Electrodacus, very thin indeed, judging by the Will Prowse video.

Only cheaper BMSs have current passing through them. Orion and REC both use external contactors. It is the best way for higher charge and discharge.

Soldering on wires to the boards does irritate some people. For me it is a basic skill and it means I could choose the size of wire to best fit the ring terminal.

That said, if doing it again, I would go for something that would control my charging devices directly (although that has its own disadvantage of relying on the BMS entirely) but that comes at a higher cost than I was willing to pay. The app and ease of configuring remotely via Bluetooth are all of course excellent features of the 123Smart.

 

[sailaboutvic]

D agree the app good i think I said this else where and easy to use altho why you can't change Vbal within the app and have to play around with tho tiny switches god know.
But hey once done unless you want to change them there no need to go back there .

I sure many electric have similar module but there protect because they are so delicate,
I also got mine covered there a grp which cover the whole lot but if your working in there you can quite easy damage them.

Hold my hands up I didn't know there a 20 mv +- so in reality if I wanted to charge the pack to 14.4v which I dont , I won't have a chance are one cell will show that its pass that point when it could be much less . Unless I had a meter on each cell.

Moving on , see still no reply to your question on Alu cells

 

[Poey50]

D agree the app good i think I said this else where and easy to use altho why you can't change Vbal within the app and have to play around with tho tiny switches god know.
But hey once done unless you want to change them there no need to go back there .

I sure many electric have similar module but there protect because they are so delicate,
I also got mine covered there a grp which cover the whole lot but if your working in there you can quite easy damage them.

Hold my hands up I didn't know there a 20 mv +- so in reality if I wanted to charge the pack to 14.4v which I dont , I won't have a chance are one cell will show that its pass that point when it could be much less . Unless I had a meter on each cell.

Moving on , see still no reply to your question on Alu cells

Yep that dip switch is minute! A bright light, magnifying glass and needle job. I won't be changing that in a hurry.

20 mv is literally nothing to be concerned about when charging to full. 3.65 volts is a good endpoint for a cell but nothing terrible happens if you exceed that by 0.02 volts.

 

[vas]

Raspberry Pi/LiFePo4 sounds like a match made in monitoring heaven, save mV readings and temperature of every cell once a second with alarms etc. I've a board slowly coming along which should do all that with headers to mount a Pi zero and rj45 sockets for a breakout boat at the batteries .

Anyone else doing this?

Easier than you might think and a Pi Zero has very little power consumption so can be left on 24/7.

the more I read about all these BMSs the more I think that I should do one myself and fully integrate it with all the other crap I have onboard...
will wait and see what you come up first though as I'm planning on lifepo4 for next season
I guess yours wont be N2K ready and wont integrate/talk with my other kit, signalk maybe

Q to all knowledgeable ppl:
is there any decent quality recommended BMS that DOESNT try to be a solar charger or some other things at the same time?
I wouldn't pay for Electrodacus for a screen I don't need and for a solar controller I don't need- makes you wonder what is actually left behind then...

I guess, I'd only need a "monitoring" device for all 8cells (24V bank) and most likely passive balancing (if it is really needed!) but then cue on GHA, a couple of ADS1115 would do that more accurately
Starting to question that, maybe best to get a warning that things are shifting, and get your bench power supply out and do a top balance.
Do we have any data on how often the passive balance is actually "activated" on lifepo4 banks running on boats?
Vic, Poey, are there any data on the bms on that?

V.

 

[sailaboutvic]

Q to all knowledgeable ppl:


V.

Sorry mate your on the wrong forum hahaha

 

[sailaboutvic]

sorry Vas couldn't resist that.
Data has in what?

 

[Poey50]

Do we have any data on how often the passive balance is actually "activated" on lifepo4 banks running on boats?
Vic, Poey, are there any data on the bms on that?
V.

That depends on how you use them. Chris if I remember correctly has never balanced his cells since he first installed them. For anyone limiting their maximum state of charge to 85-90% may not need to at all or often. 123SmartBMS advice is to top balance around once per month. I haven't been able to use mine often enough to comment from experience.

 

[sailaboutvic]

That depends on how you use them. Chris if I remember correctly has never balanced his cells since he first installed them. For anyone limiting their maximum state of charge to 85-90% may not need to at all or often. 123SmartBMS advice is to top balance around once per month. I haven't been able to use mine often enough to comment from experience.

Really once a month top balance, there no way I be pulling mine apart every month , for sure some thing would break .
Maybe in a couple of years if I see any big different around the 13.8v stage.

OK as said I had mine working now for three months , and yes voltages dont tell you much when it comes to lithium compare to LA .
But wait for this . I reckon I could set my charging to 13.8v and set an alarm to say 12v and get away without a BMS , we monitor our quite a lot , before any large load we check the cells ,
the victron mppt I have to say in three years of using then not once have they gone pass voltage they been set at .
So some one come up with a cheap devise to monitor the cells with an alarm I be temped if I needed to replace the bms .

 

[Alan S]

I have been lurking here and all over the net for a while as I am planning a 48V LiFePo4 propulsion system for a small inland/coastal cruiser.
Vas asked

" is there any decent quality recommended BMS that DOESNT try to be a solar charger or some other things at the same time?"

Maybe this would do the job...

GWL Cell Performance Monitor | shop.GWL.eu

... although in might be seen as overkil for a 12V system as it can handle up to 16 cells.

It doesn't do cell balancing but it shows the volts of your lowest and highest cells so you can see at a glance if manual balancing is required.
Various outputs for alarms and/or load and charge disconnect relays. (no connection with gwl by the way)

Rasberry Pi idea sounds great but I think the adc volts measuring interface might be tricky. Very interested to see how everyone's project progresses.

 

[mitiempo]

sailaboutvic

You should not use either absorption or float. Charge to a voltage and stop. 14.2 if a high current charge source or about 13.8 if using a low current charge source.

Electrodacus doesn't come with solar controller - it is an option. I will be using Victron MPPT. And Electrodacus is not expensive, about 8- pounds.

A bank properly top balanced originally, that is kept under 14.2 volts without absorption and float shouldn't ever drift. Pulling apart tp bance should not ever be necessary, if you wish pull it apart every year or two but you probably do not need to. The passive balancing in the BMS will take care of any small differences.

Regardless of the BMS used it is should not need to be depended on. All charge sources should be properly set. Max voltage 13.8 to 14.2 depending on current, no absorb (or minimum absorb or lower voltage if adjustable), and no float or float at 13.2 at which the batteries accept no current.

 

[sailaboutvic]

Hi Mitiempo
in theory agree you don't float lithium , in practice you have to set the mppt to some thing And 13.5v seen an acceptable voltage by most people you talk to .
Even smart agree to this so to victron .

As for Abs again it as to be set to some thing , in our case if overcase I want to keep the voltage up so it sets to 14v these days with very sunny days here in the Med its set to 13.8v . Abs time which is more important is 5 mins, you don't want it on Abs for long periods.
You have to take into account we full time liveaboard using power ever second, not weekend sailors .
Almost all my charging is done by solar , shore charger not use and if I had to it be throught the b2b .

B2b judging by our only cruiser this year ,( Next one in about three weeks) we found during day time we didn't need the b2b if motoring and it was turned off , panels do the job nicely, only on the night sail when motoring did we turn it on and we could had easily done without it but the following day was going to be very overcast.
as you see by what i write we monitor and manage our batteries, there not just fitted and forgoten relying on the bms and relys.

I was only joking with the last posting , but joking apart I could had and if need be manage without a bms actually we did for three weeks when our end board packed up.
Agree we run the risk of an mppt shooting off but with a good alarm on each cell I could easily managed without ,
As to what to set end voltage, there lots of option and they are just that, some insis 14.4v others 14.2v I'm happy with 14v when need be and when I dont need to 13.8v
Who right and who's wrong will not come to light the day our batteries dies , I guess the 13.8 to 14v will come out on top .
So come on GHA get to work .

 

[Poey50]

Really once a month top balance, there no way I be pulling mine apart every month , for sure some thing would break .
Maybe in a couple of years if I see any big different around the 13.8v stage.

I think I've misled you. I don't mean take them apart and top balance them in parallel. I mean deliberately hold them above 3.4v per cell when charging as usual to let passive balancing do its business. I should have said 'balance' or 'rebalance'.

 

[mitiempo]

Hi sailaboutvic

With a conventional BMS you are correct - the absorption and float have to be set at something. If the absorb is set as short as possible and float low enough that the batteries are not accepting current - eg 13.2 which effectively shuts the charger off - you are fine.

The Electrodacus is different. It turns off charge sources when any cell reaches a specified voltage so the charger doesn't go into absorb or float. Factory settings are 3.55 (14.2) charging and 2.8 (11.2) discharge. As I will be using solar only (except when powering) I will probably lower that to 3.45 or 3.5.

You could set your charge voltage at 15 volts with the Electrodacus if you were crazy or sloppy and it would still shut the charging off when the first cell hits 3.55 or whatever it is set for.
That to me is one of the advantages of the Electrodacus. It also has independent control over charging and loads so when the charger is shut off the loads are still active.

I am not advocating that charge voltages are not important. I will be setting my charge sources to voltages that can not damage the cells even if the BMS dies. I look at a BMS as a last resort failsafe but the added convenience of the Electridacus turning off charging when one cell hits a given threshold is a huge plus as far as I can tell.

 

[vas]

I have been lurking here and all over the net for a while as I am planning a 48V LiFePo4 propulsion system for a small inland/coastal cruiser.
Vas asked

" is there any decent quality recommended BMS that DOESNT try to be a solar charger or some other things at the same time?"

Maybe this would do the job...

GWL Cell Performance Monitor | shop.GWL.eu

... although in might be seen as overkil for a 12V system as it can handle up to 16 cells.

It doesn't do cell balancing but it shows the volts of your lowest and highest cells so you can see at a glance if manual balancing is required.
Various outputs for alarms and/or load and charge disconnect relays. (no connection with gwl by the way)

Rasberry Pi idea sounds great but I think the adc volts measuring interface might be tricky. Very interested to see how everyone's project progresses.

thanks Alan, actually that's exactly my point ADS1115 A2D converter is probably more accurate (or the same) with what such a board is using. Having a relay output or three is nothing, so a pzero with the right bits that GHA is contemplating (or a teensy similarly equipped that I'll probably do) will do nicely as far as monitoring voltages per cell and setting simple or complex rules regarding course of action if something goes wrong.
That's 30quid in h/w and a few hours in s/w (and more fun than getting something ready made for some of us)

I think I've misled you. I don't mean take them apart and top balance them in parallel. I mean deliberately hold them above 3.4v per cell when charging as usual to let passive balancing do its business. I should have said 'balance' or 'rebalance'.

Poey, looks like I've forgotten some of the things I read at the beginning of the last 5m (and counting) lockdown we had.
So passive balancing is not done with some type of resistive loads or something like that, just pushing all cells up to 3.4V for a certain period of time and monitor carefully how they match?

Meaning that during discharing, BMS is only monitoring for discrepancies and based on the rules set get the system offline if something goes way off, correct?
if so, a custom BMS is even more simple to implement, but tbh I vaguely remember reading some more complex issues that a decent bms deals with (and I don't remember them now!)

cheers

V.

PS. GHA, ordered the boards or not yet?

 

[sailaboutvic]

Mitiempo , I just notice we talking and comparing different things, the bms I'm using and poey 50 come to that 123smart have no Abs or float setting ,
it has a cell min and max voltages which in turn ingage external relays , other bms as I understand it will switch off the batteries with internal relays .
I not seen The Electridacus so it be wrong of me to common on it .
But I see you say (You could set your charge voltage at 15 volts with the Electrodacus if you were crazy or sloppy and it would still shut the charging off when the first cell hits 3.55 or whatever it is set for.)
I think you can do that with most bms i know if i set my mppt to 20v once one cell hits 3.6v which is my safety setting the bms tell the relay to open.

All I will say is until you buy some thing and use it you don't know how good it is ,
I think we all brought stuff thinking it do the job then find out it crap.
Good luck in trying it out .
Anyway

My charging setting are at my mppt ( victron) and all the switching is done there the same apply to my victron b2b , as for my shore charger that only has gel or LA setting 14.4 v ABS now that's also Regulated throught my B2B so although the charger is set to 14.4v ,
when the b2b see 13.8v it switches off .
The bms only act as a last port of call .
As I wrote before we full time liveaboad so what even setting we have as float doesn't matter the voltage isn't there for long,
but thats my own option after may months of research and more importantly talking to over cruiser face to face who been using lithium for some time is 13.5v if about right..

 

[mitiempo]

sailaboutvic

You state " ...I know if I set my mppt to 20v once one cell hits 3.6v which is my safety setting the bms tells the relay to open."

Does that mean that the charging stops but the loads are still active with your bms?

I am a full time liveaboard as well, heading offshore in the near future, covid depending.

 

[sailaboutvic]

sailaboutvic


You state " ...I know if I set my mppt to 20v once one cell hits 3.6v which is my safety setting the bms tells the relay to open."

Does that mean that the charging stops but the loads are still active with your bms?

I am a full time liveaboard as well, heading offshore in the near future, covid depending.

That depend if you using one or two relays , with the 123smart It has a load and charging switching .
It can be use to open which one its connected too .
There you go poey , I found some thing I do like about the 123smart haha

 

[sailaboutvic]

if some one comes up with some way to switch off a relay once any cells gets to X voltage
Add a cheap balancing module and you have a ready made bms for next to nothing .

 

[Poey50]

That depend if you using one or two relays , with the 123smart It has a load and charging switching .
It can be use to open which one its connected too .
There you go poey , I found some thing I do like about the 123smart haha

Finally!

 

[Poey50]

thanks Alan, actually that's exactly my point ADS1115 A2D converter is probably more accurate (or the same) with what such a board is using. Having a relay output or three is nothing, so a pzero with the right bits that GHA is contemplating (or a teensy similarly equipped that I'll probably do) will do nicely as far as monitoring voltages per cell and setting simple or complex rules regarding course of action if something goes wrong.
That's 30quid in h/w and a few hours in s/w (and more fun than getting something ready made for some of us)


Poey, looks like I've forgotten some of the things I read at the beginning of the last 5m (and counting) lockdown we had.
So passive balancing is not done with some type of resistive loads or something like that, just pushing all cells up to 3.4V for a certain period of time and monitor carefully how they match?

Meaning that during discharing, BMS is only monitoring for discrepancies and based on the rules set get the system offline if something goes way off, correct?
if so, a custom BMS is even more simple to implement, but tbh I vaguely remember reading some more complex issues that a decent bms deals with (and I don't remember them now!)

Vas, I'll cover more than you ask for as others may be interested in a summary.

I think a Battery Management System is, on the whole, misnamed - it's really more a cell management system preventing overcharging, over-discharging, overheating, and low temperature operation at cellular level. Basic BMSs isolate the whole pack when one of these charge or temperature parameters are reached by any cell. At a higher level of sophistication (but in my view important for marine installations), charge and load circuit relays are controlled independently. Even more sophisticated BMSs can operate warning alarms prior to disconnect on any of those parameters and at the top of the range can directly control charging sources. Some can progressively switch off loads.

All BMSs have a capacity to balance cells in a pack which does deserve the term 'battery management'. Active balancing is the redistribution of power from a higher charged cell to a lower charged cell. This is found in high-end BMSs. Passive balancing is more common and works by burning off power through a resistive load when any cell reaches a voltage threshold. 3.4 volts is a common setting. This holds back the more highly charged cells allowing others to catch up but no power is transferred between cells.

Some BMSs provide state of charge information by way of Coulomb counting, a few have data logging built in and there are a variety of means of displaying information with Bluetooth becoming more common. A good BMS allows all parameters to be easily user-configured.