Practical Questions about LiFePo4 Power Systems

[Dockhead]

I understanding from reading about BMS systems, that the BMS needs a way to disconnect the charging source in case of overvoltage.


That will fry the diodes of an alternator, if it happens.



There must be a solution to this -- the BMS123 Smart diagrams show a starting battery connected in parallel through a diode isolator -- a neat solution. But not really applicable for my boat since my start batteries are different voltage and on completely separated systems. I'm sure this is a 101 level question, but how is this dealt with?


The same document (https://files.i4wifi.cz/inc/_doc/att...S123-Smart.pdf) talks about automatic disconnection of the loads, by the BMS.




This makes me a little nervous -- I realize that it would happen only in a critical error situation, but do I want a computer deciding to shut down automatically my autopilot, depth sounder, etc.? Does anyone keep a separate small lead acid bank for critical electronics? GMDSS incidentally requires a separate battery for radios and emergency comms, so doing it this way has other advantages, but wishing to avoid complexity . .






Another unrelated question -- is anyone using the Victron Multiplus charger/inverters with LiFePo banks? Victron claims compatibility -- do they work ok?

[nebster]

Originally Posted by Dockhead

Another unrelated question -- is anyone using the Victron Multiplus charger/inverters with LiFePo banks? Victron claims compatibility -- do they work ok?

Yes, I am using them. They work okay.

They used to have a charging algorithm that terminated CV (absorption) at a programmable fraction of the baseline charge current. For some reason, they removed that from the firmware a year or so ago. It would have been the "best" way to charge a simple bank with no BMS providing charge commands.

They also for some reason do not allow you to disable CV entirely, nor float, when using the built-in algorithms. This is inconvenient but not a showstopper; creative values plugged in here and there can achieve almost the same thing.

If you use a BMS that has charge authority, the above is moot. The Victron charger can then simply be toggled on and off via a pair of potential-free contacts.

[Dockhead]

Originally Posted by nebster

Yes, I am using them. They work okay.

They used to have a charging algorithm that terminated CV (absorption) at a programmable fraction of the baseline charge current. For some reason, they removed that from the firmware a year or so ago. It would have been the "best" way to charge a simple bank with no BMS providing charge commands.

They also for some reason do not allow you to disable CV entirely, nor float, when using the built-in algorithms. This is inconvenient but not a showstopper; creative values plugged in here and there can achieve almost the same thing.

If you use a BMS that has charge authority, the above is moot. The Victron charger can then simply be toggled on and off via a pair of potential-free contacts.

OK, and I gather all the commercial BMS systems have "charge authority"? Over here most people seem to use the "BMS123 Smart" sold by EVPower in the Czech Republic.

And what about the alternator? Can this BMS switch that as well?

[john61ct]

If there is a lead batt on the circuit as buffer, the BMS can just isolate the batt.

If you use a relay to terminate the charge at source,

like cutting field current to the alt,

then loads must start depleting the bank,

rather than carried by the source.

All depends on what you want.

[john61ct]

All a charge source needs for decent LFP compatibility is the ability to let the user adjust the voltage setpoint.

Ideally also allow setting both Absorb Hold Time and Float voltage to "none", just stop when the setpoint voltage is reached,

when this is what you want (see above).

But those are less common.

[nebster]

Originally Posted by Dockhead

OK, and I gather all the commercial BMS systems have "charge authority"? Over here most people seem to use the "BMS123 Smart" sold by EVPower in the Czech Republic.

Yeah, I think most devices sold as a standalone BMS can issue a charge enable/disable command. Apart from an emergency disconnect command, it's the most important thing they can do.

And what about the alternator? Can this BMS switch that as well?

I haven't had to deal with any alternator-based charging, so I can't help you there. I only have solar and a dedicated diesel generator.

[Dockhead]

Originally Posted by nebster

. . . I haven't had to deal with any alternator-based charging, so I can't help you there. I only have solar and a dedicated diesel generator.

I bet that the Balmar regulator can switch off the alternator. I wonder if it can do this in a way which doesn't require a lead battery in the circuit.


I think I can switch off my alternator now by simply putting a relay into the exciter wire.

[john61ct]

Cutting the field current circuit is the standard way to stop an alt generating power.

No special gear is needed.

Having a manual switch for that at the helm is SOP for any setup where a big amp / high CAR bank can steal too many horses from the main propulsion function, in the event.

[Dockhead]

Originally Posted by john61ct

Cutting the field current circuit is the standard way to stop an alt generating power.

No special gear is needed.

Having a manual switch for that at the helm is SOP for any setup where a big amp / high CAR bank can steal too many horses from the main propulsion function, in the event.

Well, with 100hp I don't have the manual switch at the helm, but in any case -- what is SOP for LiFePo banks? Wire the BMS to the field current?

[CharlieJ]

Balmar recommends turning off the power to the regulator instead of turning off the field current. Easier on the regulator.

To the OP original question: I have always put an override switch (old Navy: Battle Short) in the BMS to allow the operator to make the decision, on receiving a LVE or HVE alarm, to override the BMS protection algorithm because of the operational environment at the time of the alarm. Obviously, pushing the "Battle Short" switch cannot be taken lightly as destruction of the LFP bank may occur.

[Tornadosailing]

It might also be wise to put a big electrolytic capacitor on the output of the alternator to absorb any voltage spike between the BMS cutting off any further charging, and the collapse of the magnetic field at the alternator. It's only a few milliseconds, but things happen fast with electrons.

[Dockhead]

Originally Posted by CharlieJ

Balmar recommends turning off the power to the regulator instead of turning off the field current. Easier on the regulator.

To the OP original question: I have always put an override switch (old Navy: Battle Short) in the BMS to allow the operator to make the decision, on receiving a LVE or HVE alarm, to override the BMS protection algorithm because of the operational environment at the time of the alarm. Obviously, pushing the "Battle Short" switch cannot be taken lightly as destruction of the LFP bank may occur.

Well, I can well understand that. I am a little nervous about the idea that I could lose all power based on a decision by the BMS. Some critical systems (e.g. navigation, depth sounder, comms) run off the main battery bank.

[a64pilot]

I did a little research myself on a Lifepo system, and based on a little experience with LiPo, I wanted similar protection strategies, which of course means a full BMS.
I eventually decided, not yet, yes of course it can be done, but as your discovering to cover all the bases and not leave weaknesses, it gets complex.
Sort of the nuclear powered wrist watch analogy, some love elegant, complex systems. I don’t, I like simple things, so I’m still waiting.

I believe one way is dual banks and BMS’s, unlikely to lose both.

[Dockhead]

Originally Posted by a64pilot

I did a little research myself on a Lifepo system, and based on a little experience with LiPo, I wanted similar protection strategies, which of course means a full BMS.
I eventually decided, not yet, yes of course it can be done, but as your discovering to cover all the bases and not leave weaknesses, it gets complex.
Sort of the nuclear powered wrist watch analogy, some love elegant, complex systems. I don’t, I like simple things, so I’m still waiting.

I believe one way is dual banks and BMS’s, unlikely to lose both.

At the heart of all of this are systems architecture and integration questions. Funny enough, what is drawing me back to LiFePo is not the main and obvious advantages of them, but a problem I have with my existing system which I can't really solve without batteries with a lot more power density. A banal packaging question.


My bank consists of two banks of 4x 12v (not 6v) FLA batteries which have been combined. The boat was designed for God knows what reason (considering how elegant and high quality the rest of the electrical system is) with 4x 110AH "leisure" batteries as the house bank, and another similar bank powering thruster, winches, and windlass. I combined the two banks and it works somewhat, but is suboptimal in many ways. The battery boxes are not high enough for golf cart batteries.


I could change the boxes and get golf cart batts in at least one of them, but not enough for the kind of capacity I need, and I don't want to make any major alterations to the system without solving the issue of the split bank and multiple parallel batteries.


So LiFePo is the only way I can get all the capacity I need in one bank.


And then of course there are the several immense advantages for one who does not have solar (no point to LiFePo if you have a decent solar installation in my opinion).



I put 200 hours on my generator while I was in Greenland and I'm tired of that. So the wheels are turning in my mind about a winter project involving LiFePo.


It's not really all that expensive in Europe compared to lead (lead batteries for some reason are quite a bit more expensive here).



The idea of having a BMS doesn't bother me at all, but the idea of unexpected power interruptions does. I guess I could put critical systems on a separate lead acid bank -- and I'll have a battery box all ready for that, very convenient -- but that would re-introduce complexity I'm trying to get rid of.



I'll be grateful for all tips from those on here who have gained experience with these, especially with the Smart123 BMS which is commonly used over here.

[rgleason]

I believe the Victron Overcharge Disconnect has a Field disconnect included.
What are the interruptions from? Sorry haven't read the whole post.

Perhaps the lead acid battery could also have 24hr loads such as Bilge and other essential loads that might run the Lithium down if a float gets stuck or something.
That is one disadvantage to the proposed wiring for my boat.