270ah DIY LiFePO4 build

[Kelpie]

I understand that the move from lead acid to LifePO4 requires a shift in mentality about the state of charge when a boat is not in use, and then I twigged that a recent addition DJI made to their drones might be worth considering. In short, when a drone battery is neither in use nor charging, the battery firmware enables some kind of load to gently run down the battery to about 50% at which point it stops. Would that be possible so that idiots like me don't have to think about it?

There are a number of ways you could achieve this. Somebody cleverer than me can probably come up with something smarter, but at the most basic level just have a device with a configurable low voltage disconnect (e.g. an inverter or fridge) and set that to say 13v, then turn it on and go home. It will run the battery roughly half way down and then shut off.
Or you could change the settings on your MPPT and stop charging at a lower voltage, either permanently (with occasional higher voltage charges for balancing) or when leaving the boat for a period of time.
Or just stop worrying about every last detail. These batteries are already cheaper than lead acid, don't deliberately misuse them but don't lose too much sleep over it either.

 

[Poey50]

checked for 8S (24V) works out 108euro/kWh, not bad, not to mention that if I've survived the last 4yrs with 225/2 Ah, why should I go for 0.8*280Ah. 110Ah usable (trojans) vs 224 (lifepo4)
makes you think if one should get a 200Ah package (not that I can find many - fe this seller only does 280 and 100)

BTW, probably asked before but don't remember:

in LifePO4 land, is it better to have the bank going daily from 95-70% and same all over again (charging with solar) or a smaller pack that would go daily 95-50%?
approx diff of using a 280Ah vs a 200Ah pack. Not liveaboard so this applies for 60-90days pa.

V.

I think best longevity would be to work around the midrange so, instead of 95- 50%, go 80-35%. You can then go to 95 if you need a bit more in the tank for a particular trip.

 

[Poey50]

There are a number of ways you could achieve this. Somebody cleverer than me can probably come up with something smarter, but at the most basic level just have a device with a configurable low voltage disconnect (e.g. an inverter or fridge) and set that to say 13v, then turn it on and go home. It will run the battery roughly half way down and then shut off.
Or you could change the settings on your MPPT and stop charging at a lower voltage, either permanently (with occasional higher voltage charges for balancing) or when leaving the boat for a period of time.
Or just stop worrying about every last detail. These batteries are already cheaper than lead acid, don't deliberately misuse them but don't lose too much sleep over it either.

All good suggestions, perhaps mostly the last. The way things are going PBO will be giving cells away free with the magazine. And in 10 years time other battery technologies may sweep LFP away. There certainly seem to be a lot of new ideas knocking on the door ... the successful ones may be as cheap as chips.

 

[geem]

The development of lithium batteries for cars is moving at a pace. It may not be relevant for marine domestic banks but interesting none the less. A pal of mine in the industry, heavily involved, says the stumbling block for cars is the slow recharge times. Fast chargers currently give a reasonable charge after 45mins of charging. The current development is in liquid cooling. Under lab conditions the same charge can be applied in 7 mins instead of 45 with liquid cooled batteries. This technology is five years away for the car industry. I am not sure I would be buying an electric car at the moment knowing that battery technology is going to change a lot soon.

 

[Poey50]

...

 

[TernVI]

The development of lithium batteries for cars is moving at a pace. It may not be relevant for marine domestic banks but interesting none the less. A pal of mine in the industry, heavily involved, says the stumbling block for cars is the slow recharge times. Fast chargers currently give a reasonable charge after 45mins of charging. The current development is in liquid cooling. Under lab conditions the same charge can be applied in 7 mins instead of 45 with liquid cooled batteries. This technology is five years away for the car industry. I am not sure I would be buying an electric car at the moment knowing that battery technology is going to change a lot soon.

To charge a useful size battery in 7 minutes needs a pretty impressive power source.

 

[geem]

To charge a useful size battery in 7 minutes needs a pretty impressive power supply
QUOTE
Correct. There are some major infrastructure requirements needed to meet the huge electrical load. There are also major players getting involved. The likes of BP who have the existing forecourt capacity in what will be a diminishing requirement for petrol and diesel. I understand there may well be very large battery facilities required to store power to reduce peak load at the charging sites. The plan being to charge up these banks at night when wind power may well be more plentiful then release it during the day to charge car batteries at these sites. I think it will be a very interesting future. I am sure plenty of spin of technology gains for marine use

 

[Kelpie]

To continue this slight tangent, there is also the prospect of induction charging. Certain major roads could be equipped with this technology, so that vehicles leave the motorway fully charged. Probably a long way off but that would be a game changer. Rather hard to replicate at sea of course!

 

[vas]

I guess at some point they'll figure out a way to revert the electrolytic action of dissolving boat metals into charging it's batteries, shouldn't be hard
after all floating in brine is the first step ?

 

[Poey50]

Those on the Lithium Batteries on a Boat Facebook Group may also have seen this image. It's not for the faint-hearted but, as with all big problems, there may be things to learn.

This meltdown happened four days into a delivery trip 100 miles offshore of Pensacola, Florida. The solo delivery skipper noticed a sudden loss of power and put on the engine to charge the battery. When he went below later he got the terrible smell and tried to prepare for what he thought would be an inevitable fire. That didn't happen - just the meltdown. He then made it to his delivery destination without power. The recent history of the boat is that it blew off a cradle during a hurricane and had extensive hull repairs before this delivery trip.

There is no report from the owner and it is all a bit of a mess. Nevertheless there are some clues to the cause from the picture and the history and it might be helpful in planning for safety to think though the possibilities.

 

[vas]

looks yummy!
like this coffee and whipped cream looks although tbh cannot make out what exactly was behind this mess... too many bus bars for my liking

 

[Poey50]

looks yummy!
like this coffee and whipped cream looks although tbh cannot make out what exactly was behind this mess... too many bus bars for my liking

The whipped cream is I assume the electrolyte having vented from an overheated cell. The multiple bus bars are because it is a 2P4S configuration. The problem with that is two separate cells are effectively a single cell for the purposes of cell-level monitoring. So a problem in one of the two may be compensated for by the other rather than triggering a catastrophic level BMS disconnect. I don't think that is the only problem here but, I suspect, one contributory factor along a chain.

 

[Poey50]

The other thing to notice are the number of cables which appear to terminate direct to the pack. Not all of these (possibly none) are protected by a cell-level BMS operated relay. My theory is that charging from the alternator had no BMS control since, even if the overall voltage of the faulty combined cell was pulled under the high voltage threshold by the duff cell, the BMS should have disconnected on the basis of high temperature alone. Instead the alternator kept on pumping charge.

Edit: Rod Collins has just chimed in and has spotted the BMS connector - but the BMS had been removed!

 

[Poey50]

Rod Collins has added to his reply. This is his full post ...

"Here's what I can see from just one photo:
(involved in marine fire investigation)
#1 DIY Built Li-Ion Battery (most likely be LiFePO4 chemistry)
#2 What appears to be a typical BMS balancing / cell voltage communication plug not connected to anything! This means the battery had ZERO BMS protection!
#3 Sloppy over-all workmanship including the battery cable terminations.
#4 Multiple wires attached to POS terminals and no Class T fuse/s anywhere in-sight. With four positive wires on B+ this is a near impossibility that any of these wires are actually protected by a contactor based BMS... With the gauge of the negative wires it is pretty easy to see that any "FET based BMS is also not being used as they use significantly smaller GA wire and a FET based BMS is switching negative not positive..
#5 What appears to be melted clear plastic, most likely used to cover the terminals.
#6 The use of plywood end-boards most likely for cell compression. If those end-boards caught fire the boat may have been a total loss.
#7 The reliance of straps that can stretch as the means to keep the prismatic cells compressed. Cell compression should ideally be in the form of non-combustible end-boards and threaded rod or similar metal to keep the cells in compression.
#8 When the battery got hot, and the very thin clear plastic melted, this left a metal buckle directly in contact with battery terminals.
My gut assessment on cause of fire (been doing LiFePO4 since 2008 and am active member of the ABYC Li-Ion battery committee):

#1 No BMS protection (see image below)
#2 Poor quality reject or low grade cells off Aliexpress or similar. The industry produces thousands of reject cells every day and most of these reject cells are pawned off on unsuspecting buyers who are no educated enough to know the difference between a good cell and a bad cell..
#3 A lack of the DIY/marine tech properly top balancing, capacity testing and internal resistance testing before assembly.
#4 It would appear they started out with a BMS then possibly gave up to to disconnects likely due to poor quality cell matching.
#5 With no BMS balancing or proper cell data, and a battery at 11.6V (way to low for a LFP in normal use, with only "pack level data", could have very easily had one cell go to 0.00V and physically reverse its polarity! When you attempt to recharge the battery the result will look exactly like we see here. This is why a BMS, that measures cell voltage, not just pack level (11.6V), is critically important. It is also why when DIYing a pack every cell MUST be tested for cell Ah capacity and internal resistance.
This sort of thing is only getting more common as the prevalence of "you-tubers" who only know enough to be dangerous are fan-boying LFP day-in, day-out...
If you don't actually know what you're doing with LFP, study, study, study......"

 

[TernVI]

Rod Collins has added to his reply. This is his full post ...
......
This sort of thing is only getting more common as the prevalence of "you-tubers" who only know enough to be dangerous are fan-boying LFP day-in, day-out...
.....

My thoughts exactly .
It's similar in the motorbike world.
https://images.s1000r.co.uk/IMG_20191005_165208687_HDR.jpg

 

[Poey50]

My thoughts exactly .
It's similar in the motorbike world.
https://images.s1000r.co.uk/IMG_20191005_165208687_HDR.jpg

That doesn't look like LiFePO4 to me. It's noticeable that the meltdown of the boat battery (above) didn't catch fire. It could however have caused a fire through igniting combustible materials.

 

[adwuk]

I got the impression that the skipper didn't really understand what he was dealing with. His original comments were that the pack voltage went down to 11.2V, and then he revised that to 13.2V. The multiple leads onto the battery +ve means that the install was suspect, and the BMS appears to be missing from the end of its connector. It all adds up to a bit of a horror show. We can all speculate as to why the battery ultimately failed (overheating acrylic lid and a short caused by one of the metal clips on the battery ties has my vote), however it does highlight that these things can go wrong, especially when used by folk who don't understand how they work. So if your yacht is to be used by others then it seems that the systems need to be totally fool/bullet proof, or you should stick with lead acid.

 

[Poey50]

I got the impression that the skipper didn't really understand what he was dealing with. His original comments were that the pack voltage went down to 11.2V, and then he revised that to 13.2V. The multiple leads onto the battery +ve means that the install was suspect, and the BMS appears to be missing from the end of its connector. It all adds up to a bit of a horror show. We can all speculate as to why the battery ultimately failed (overheating acrylic lid and a short caused by one of the metal clips on the battery ties has my vote), however it does highlight that these things can go wrong if used by folk who don't understand how they work. So if your yacht is to be used by others then it seems that the systems need to be totally fool/bullet proof, or you should stick with lead acid.

He was a delivery skipper doing a solo trip and did a remarkable job to get the boat home using his own resources. I feel differently about the owner / battery builder.

 

[TernVI]

That doesn't look like LiFePO4 to me. It's noticeable that the meltdown of the boat battery (above) didn't catch fire. It could however have caused a fire through igniting combustible materials.

It makes little difference what the battery technology is, there is a lot of energy stored and it's all capable of getting very hot very quickly if not treated right. That's beoming increasingly true of lead-acid as people install ever bigger banks of batteries and ever more solar with questionable charging regimes.
You see enough threads on here where people really don't know what systems they have on their boat.


Most of the motorbike 'lithium' batteries seem to be FePO4.

It's not some binary thing 'Lithium Ion Bad, LiFePO4 Good' , there are for sure other flavours of 'Lithium-Ion' which are more dangerous, but no battery capable of whacking out hundreds of amps is exactly 'safe' in amateur hands.

Fundamentally, I have my doubts about mixing Li technology of any flavour with crude automotive alternators and suchlike, which is all based on being just about good enough for lead-acid. We have a century's experience of knowing what is OK with lead acid, tried and tested on a billion vehicles. Then some bloke in a shed with a Youtube account thinks he can do better without even owning an oscilloscope.

 

[adwuk]

He was a delivery skipper doing a solo trip and did a remarkable job to get the boat home using his own resources. I feel differently about the owner / battery builder.

Apologies Poey, I wasn't having a go at the skipper. He had to deal with the cards that he was dealt. It does provide some comfort that the cells didn't burst into flames, but I suspect that we will never know the full story.