LiFePO4 reference diagram, Alternator version

[zurk]

Originally Posted by newhaul

No more actual connections but DIY has more flexibility wrt repairs.

Question is what are you going to do when one of those internal BMS units die an unnatural death? Battleborn even admits this is the reason for 90% of their warranty failure claims

remove one of the house batteries and replace failed one, warranty failed one when able. no different than any other piece of equipment.

[newhaul]

Originally Posted by zurk

remove one of the house batteries and replace failed one, warranty failed one when able. no different than any other piece of equipment.

So junk a battery worth almost a grand because of a 75 dollar part ?
Ok well your boat your choice just wish I was your shipwright .

[Baronkrak]

Originally Posted by zurk

all the batteries have their own BMS and the battery BMSes are all networked to each other via hard line (CAN bus). the start batteries go through a victron dc to dc charger.

And the batteries have integrated charge disconnect and load disconnect "relays"?

[LifesBetterWhenYoureBeating]

Originally Posted by s/v Jedi

Here is my version that makes the alternator a major charge source, meaning that it charges the LFP batteries directly rather than via a dc-dc converter.



The alternator shown can be the oem alternator or an additional alternator. I have left both dc-dc converters as well, so if this is the only alternator, you can simply keep the start batteries in float using the dc-dc converter.



You need an alternator that outputs the voltage of the LFP batteries. You also need an external regulator that has a charge profile for LFP chemistry as well as a temperature sensor on the alternator to prevent overheating the alternator. Balmar makes this, so I chose their images (and it's what I have).



There is an ON/OFF battery switch to disconnect the alternator. This is used when working on the engine, to prevent a short (Rod Collins recommendation).



I show a fuse on the LFP busbar but I am not sure if there needs to be a second fuse at the alternator. Let me know if this is required and I'll add it.



In real life, for the regulator B+ wire, I would use a fuse position in the fusebox that is connected to the busbar, but in the diagram it is all the way down, too far away from the relays for clarity.



Now comes the tricky part: you need to shut down alternator charging before a BMS disconnects the battery. If you don't, the rectifier diodes in the alternators will burn out. The documentation of the alternator regulator will show how to have it shut down the field current, which stops the charging. In the case of Balmar, they recommend the brown ignition wire, but also allow using the red B+ wire for rare cases where it is crucial that the alternator stops charging, so I use the red B+ wire in the diagram.



For each BMS you now need to add a relay with a NC (normally connected) contact (the blue thingies near the BMS's). The coil of the relay is energized by the warning signal from the BMS. The regulator B+ wire starts with a fuse at the busbar and from there has to go through the BMS controlled relays. Now you can choose between two options: I show the relays to be in series, which means that charging stops when any one BMS raises the warning. You can also put them in parallel, which means that charging continues, which will probably trigger a HVC of the battery that triggered the warning before. This doesn't damage the diodes yet because there is a second battery still online, which didn't warn yet. I find my shown method the preferred option because you need to try to prevent a HVC, which it does.



That's it, not difficult but of course you need to run the ICE (internal combustion engine) to charge the battery.

Any chance you could share an editable version of this diagram?

[s/v Jedi]

Originally Posted by gsalgado

Any chance you could share an editable version of this diagram?

I’m sorry, no. I don’t like many edits of it anyway, but if you think something needs changing, please discuss that here, even if the change is only applicable for a smaller group of boats

[newhaul]

Originally Posted by gsalgado

Any chance you could share an editable version of this diagram?

You could always copy it and edit to fit your vessels needs

[zurk]

Originally Posted by newhaul

So junk a battery worth almost a grand because of a 75 dollar part ?
Ok well your boat your choice just wish I was your shipwright .

yeah. i have a 10 year warranty and if it dies after year 10 thats $100/year. im fine with that. i wish other boat parts were that cheap.

i bought a $2500 miele with a 10 year warranty. it died in year 5 because of a $50 part. i got a brand new miele in year 5, now im on year 7. would i do it again ? you bet.

[zurk]

Originally Posted by Baronkrak

And the batteries have integrated charge disconnect and load disconnect "relays"?

not sure what you mean. the chargers are programmed to shut off when the batteries are 80% charged. and the batteries are disconnected from the load when they are < 60% full. the idea is to keep the start batteries between 60-80% only for max life. there is a full bluetooth BMS inside the batteries with a CAN bus.

[s/v Jedi]

Originally Posted by zurk

not sure what you mean. the chargers are programmed to shut off when the batteries are 80% charged. and the batteries are disconnected from the load when they are < 60% full. the idea is to keep the start batteries between 60-80% only for max life. there is a full bluetooth BMS inside the batteries with a CAN bus.

You are using only 20% of available capacity?

[fxykty]

Originally Posted by s/v Jedi

…
the BMS controlled relays. Now you can choose between two options: I show the relays to be in series, which means that charging stops when any one BMS raises the warning. You can also put them in parallel, which means that charging continues, which will probably trigger a HVC of the battery that triggered the warning before. This doesn't damage the diodes yet because there is a second battery still online, which didn't warn yet. I find my shown method the preferred option because you need to try to prevent a HVC, which it does.

...

We currently have a single 4S 12V 700Ah bank using Winston cells that is about 1.5 years since commissioning. We want to double our capacity and Winston factory engineers have confirmed that they can match new cells to our current cells to create a 2P4S 1400Ah battery. That’s the easiest option as it is a simple replacement of the current battery - everything else can stay the same.

However, your reference diagrams have me considering the advantages of redundancy by creating instead a second 4S 700Ah battery and putting it in parallel with the existing battery. We would need to buy a BMS system from TAO for about USD1000 and a t-class fuse for the second battery, otherwise we’ve already got enough components for the second battery. We would still limit the combined 1400Ah system to 500A as we have no need for higher currents.

We currently have separate positive charge and load busses, each with their own cut off contactor controlled by the BMS. To simplify things it looks like we should get rid of the charge and load distinction and just treat these two busses as a single unit. The charge bus contactor becomes the cutoff contactor for one battery and the load bus contactor becomes the cutoff contactor for the other battery. Makes sense?

So the two batteries would each have their own main fuses and cut off contactors on their positives and BMS shunts on their negatives. The positives would join in parallel at the busses, perhaps through a manual 1/2/Both/Off switch? The negatives would join at the SmartShunt, so each BMS reports what it’s battery is doing and the Cerbo reports on the system as a whole.

The remaining big question is what to do with our BMS stop charge relays? We currently have 7 charging components that are controlled by our BMS and are activated to stop charging at a certain SOC. There is also an inverter load stop relay to the Quattro that would behave the same way. Note that our charge stop relays operate daily or every few days with poorer solar charging. We haven’t had a load stop event yet as our regular charge profiles kick in well before then.
1) First option is to put the output of the two BMSs’ charge stop relays in series so that a stop charge activation by either BMS stops the chargers. Is there a danger that the two batteries would be far apart in SOC?
2) Second option is to put the output of the two BMSs’ charge stop relays in parallel so that only a stop charge activation by both BMS would stop the chargers. To avoid the risk of high voltage cutoff of the first battery we could add another relay that also disconnects the battery that first activated the charge stop. When the second battery BMS activated a charge stop then a signal would be sent to the first battery that it can reconnect to the bus. Is this silly and are their issues with disconnecting and connecting batteries at the end of each charge cycle?

BTW, we stop our alternators by cutting the ignition wire to both alternator regulators. That is the way recommended by Balmar and we can’t see any downsides.

[zurk]

Originally Posted by s/v Jedi

You are using only 20% of available capacity?

youre looking at it wrong. think of it as a buffer with a battery instead of just a battery. its a flow through system with storage for when its needed.

[s/v Jedi]

Originally Posted by fxykty

We currently have a single 4S 12V 700Ah bank using Winston cells that is about 1.5 years since commissioning. We want to double our capacity and Winston factory engineers have confirmed that they can match new cells to our current cells to create a 2P4S 1400Ah battery. That’s the easiest option as it is a simple replacement of the current battery - everything else can stay the same.

However, your reference diagrams have me considering the advantages of redundancy by creating instead a second 4S 700Ah battery and putting it in parallel with the existing battery. We would need to buy a BMS system from TAO for about USD1000 and a t-class fuse for the second battery, otherwise we’ve already got enough components for the second battery. We would still limit the combined 1400Ah system to 500A as we have no need for higher currents.

We currently have separate positive charge and load busses, each with their own cut off contactor controlled by the BMS. To simplify things it looks like we should get rid of the charge and load distinction and just treat these two busses as a single unit. The charge bus contactor becomes the cutoff contactor for one battery and the load bus contactor becomes the cutoff contactor for the other battery. Makes sense?

So the two batteries would each have their own main fuses and cut off contactors on their positives and BMS shunts on their negatives. The positives would join in parallel at the busses, perhaps through a manual 1/2/Both/Off switch? The negatives would join at the SmartShunt, so each BMS reports what it’s battery is doing and the Cerbo reports on the system as a whole.

The remaining big question is what to do with our BMS stop charge relays? We currently have 7 charging components that are controlled by our BMS and are activated to stop charging at a certain SOC. There is also an inverter load stop relay to the Quattro that would behave the same way. Note that our charge stop relays operate daily or every few days with poorer solar charging. We haven’t had a load stop event yet as our regular charge profiles kick in well before then.
1) First option is to put the output of the two BMSs’ charge stop relays in series so that a stop charge activation by either BMS stops the chargers. Is there a danger that the two batteries would be far apart in SOC?
2) Second option is to put the output of the two BMSs’ charge stop relays in parallel so that only a stop charge activation by both BMS would stop the chargers. To avoid the risk of high voltage cutoff of the first battery we could add another relay that also disconnects the battery that first activated the charge stop. When the second battery BMS activated a charge stop then a signal would be sent to the first battery that it can reconnect to the bus. Is this silly and are their issues with disconnecting and connecting batteries at the end of each charge cycle?

BTW, we stop our alternators by cutting the ignition wire to both alternator regulators. That is the way recommended by Balmar and we can’t see any downsides.

I think I wrote this to you before, but I’ll repeat it for the other readers:

When you change a 4S 700Ah battery into a 2P4S 1,400Ah battery, you need to buy 4 new cells to add to your existing 4 used cells. It is impossible for the manufacturer to match cells as thus would require for you to do an accurate internal resistance measurement of each cell (for which you need expensive equipment) after which the manufacturer finds matching cells in their inventory, then confirms their internal resistance.
I think it is much better to have them ship you 4 cells that match your current 4 cells specs when they were new (internal resistance as they measured your cells after manufacturing them) and making sure these 4 cells are closely matched to each other.

After receiving the cells, mark them carefully. Now take the old battery apart and make sure thise cells are marked well too.

At this point you need to fully charge each cell. With 700Ah cells, do each cell individually and I recommend Rod Collins’ stepped-balance system: charge each cell to 3.4V first, then each cell to 3.5V next, followed to 3.6V.

After this, take the 4 used cells and create two pairs. If you can measure internal resistance then do this to decide on the pairs to be as equal as possible. Now do the same with the 4 new cells. Now you have 4 pairs of 1,400Ah each, which you can put in series. Do this, verify all the voltages and now discharge it as a complete battery to 60% SOC. Verify voltages again. If all looks good, clearly mark the pairs.

Do not create parallel pairs using a new and old cell. They are not matched good enough and one will charge/discharge more than the other, creating imbalance that is invisible and can’t be controlled.

All the above is a huge reason to create two separate batteries instead. It is far superior and easier in use. Simply fully charge each battery while the other is offline, then combine and they will be balanced forever. Every time you do a deep cycle or a full charge, you automatically also balance the two batteries.
Yes, the catch is the extra cost. You do need an extra BMS and also a fuse, a switch and relays controlling charge sources and loads. Follow my written recommendations for how to connect those relays, keeping them in series.

You have a choice of adding two contactors for the new BMS or make a common bus using one contactor for each BMS. I find the separate charge-load bus is superior but not worth it for our application because the situations where they are superior will hardly occur on boats, if ever.

[Baronkrak]

Originally Posted by zurk

not sure what you mean. the chargers are programmed to shut off when the batteries are 80% charged. and the batteries are disconnected from the load when they are < 60% full. the idea is to keep the start batteries between 60-80% only for max life. there is a full bluetooth BMS inside the batteries with a CAN bus.

So the batteries are not disconnected (protected) in the case of a low cell voltage or in the case of a high cell voltage?

[Baronkrak]

Originally Posted by zurk

not sure what you mean. the chargers are programmed to shut off when the batteries are 80% charged. and the batteries are disconnected from the load when they are < 60% full. the idea is to keep the start batteries between 60-80% only for max life. there is a full bluetooth BMS inside the batteries with a CAN bus.

So the batteries are not disconnected (protected) in the case of a low cell voltage or in the case of a high cell voltage?

[zurk]

Originally Posted by Baronkrak

So the batteries are not disconnected (protected) in the case of a low cell voltage or in the case of a high cell voltage?

why should i care about cell voltage ? the battery BMS will passively balance all internal cells and in the case of a real problem with the cells the battery will go offline for return to manufacturer.

its not raw cells ive assembled, its a manufactured product with 10 year warranty.


the only reason some people seem to care about cells is that they are buying cells instead of batteries. i dont really care what the cell state is. all i care about is if the battery works or doesnt.