LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[john61ct]

For a large bank, putting more than 2-3 series strings (as each 12V unit is, 4S) in parallel can lead to shortened lifetimes from balancing issues.

I haven't seen drop-ins with hundreds of AH capacity, so that's pretty limiting.

[Hoopla]

Originally Posted by john61ct

I made no such claim.

Obviously protecting the alt from its inadequate stock VR not designed for a high CAR bank is a basic necessary step, even if not in itself providing the highest possible current rate.

A robust DCDC charger would also work.

Many who want to take advantage of alt charging are not able or willing to change to a specialized high-output alt. The alt may not even be a significant energy source overall, so not worth spending a lot on.

The issue of halting charge when full is a separate issue, again, multiple ways to deal with that.

Yes, moving the diodes off the alt can allow for increased current rates.

Actually your post inferred exactly that claim being external regulation made a choice between small and large case alternators being largely irelevant for LFP. Hence my post.

I haven't questioned anything else you have said and support fully the concept of having LFP needs to be matched to principal charge source and sailboat use. Some people have AC Generators, some DC/DC Generators
with no need for an Alternator and maybe both with lots of fuel tankage. Some rely on Alternators off an auxiliary engine and have limited fuel tankage. Subject to engine size and front end load capability to drive large alternator capacity low tankage vessels are LFP prime candidates.

And yes multiple ways some superior to others of dealing with alternator load dump in the event of a LFP high voltage event cut off including say using LA start battery as a cushion if Alt can't be directly turned off by BMS to prevent Alt damage.

[john61ct]

Originally Posted by Hoopla

Actually your post inferred exactly that claim being external regulation made a choice between small and large case alternators being largely irelevant for LFP.

You inferred, I did not imply. But let's leave the past there.

All a HO alt buys you is shorter runtime, for many use cases that is irrelevant.

My point is that a good VR is required for LFP, regardless of the alt's current output.

[Hoopla]

Originally Posted by jkleins

I do thank many of you for trying to answer my question but I am hearing communication, integration without really understanding what the communication is trying to achieve.
What does the battery need to tell the alternator other then what voltage to provide it? If the alternator can handle the load and is properly regulated to protect itself while keeping the voltage at the set point what other communication is needed or provided by these integrated systems that a smart regulator won’t or can’t provide?
The BMS function that is about protecting the battery from getting too high or too low a voltage at the cell level seems to be handled by the drop in BMS. What are the other functions that the non drop in BMS’s do that make them superior?

Jim

Jim

This thread numbers hundreds of pages and the answer to your questions are contained therein.

However a quick and dirty starting with just the typical "Drop-In" battery packs in the market today.

LFP cells when linked together provide a pack voltage usually 12 volts nominal. Depending on pack capacity that may range between just say 4 cells or more.

The largest manufacturer of LFP cells in the world are from China. The quality assurance from those manufacturers is like chalk and cheese. Read cell's assembled by hand in a tent to high end manufacturing. The high end manufacturers are either owned by or reliant on the Chinese Govt largely for their military production. So even from the most reliable and 1st tier Chinese suppliers like CALB, WINSTON etc the western world gets the left overs after their domestic and military demand.

For instance If you break open today one of the world's premier Factory/Propriety LFP suppliers being the Dutch Masterbolt company you will find WINSTON cells. A few years ago before Genasun departed LFP you would find CALB cells. They departed this space on account of cell QA out of China, despite a superior offering there were problems.

I have broken open a lot of "Drop-Open" battery packs to interrogate the source of their cells. Those even sporting labels of being made in the US or Europe are nonsense. They are mostly Chinese made cells made by 3rd or worse tier manufacturers.

Why is this important, batteries are just batteries surely?

Unfortunately it is not. In the 10 years since LFP went into commercial production first largely for the electric vehicle market, then other uses specs of voltage tolerance have shrunk at both the low (discharge) and upper (charge) end in terms of cell life. There is no reputable LFP cell manufacturer in the world today who supports even with cell balancing (which is horse **** and another subject) a charge voltage in excess 14.4 volts or say 3.6 volts per cell.

The realities are after a pack voltage of 14.0v or 3.5v per cell individual cells start going out of balance and they do that quickly and nothing can restrain them. The cell balancing technology I have no regard for can't cope with that on account it is desparately trying to convert a voltage disparity into a energy or heat transfer. It is like 100 people trying to get out of a small bedroom.

So those voltage specs are from the best of Chinese. Yet "Drop-In" manufacturers use a max charge pack voltage between 14.4 and 16 volts or where after 14.4 volts individual cells are on the brink of death and beyond guaranteed. So why do "Drop-In"vendors not state this? The answer is simple, they no longer have a "Drop-In" for Lead Acid offering.

As for your "why is battery communication to other devices other than via voltage" question so.important with LFP? It is in essence a question about "Drop-In" LFP being acceptable or not for sailboats.

Quite frankly to answer that properly would take me many pages noting the ABYC have been wrestling with this issue for around 5 years now while the elephant in the room being "Drop-In" LFP gets larger by the minute.

So a quick and dirty and so please treat it as that, where every single Drop-In vendor I have noticed world wide simply ignores the following.

1. The energy density in LFP compared to LA is quite frightening unless it is accommodated for by reticulation design including fusing.

2. The selling specs of Drop-In LFP systems are aligned with Lead Acid batteries not actual LFP battery chemistry which leads to them selling expensive door stops.

3. Being a single bus charge/load system battery power is lost in the event of either a high charge or low voltage event without manual and complicated intervention. Think about losing all power suddenly at night in a constricted and or busy seaway.

4. Having no way of communicating to either charge or load sources in the event of either a high or low voltage event (set at rediculous levels) those sources keep doing what they are doing. They can be AC chargers, alternators, wind vane, hydro generator, solar etc. which in most instances say goodbye when the battery or energy storage device is suddenly cut off mid stream. Then there are dual sources of charge and load to deal with like charger/inverter combos to deal with. A "Drop-In" LFP solution simply ignores all of the above.

An Integrated LFP solution deals with all of the above and doesn't just ignore it as opposed to just selling a supposed "Drop-In" battery pack replacing LA and bugger the consequences.

Sorry for the long winded reply.

[Hoopla]

Originally Posted by john61ct

You inferred, I did not imply. But let's leave the past there.

All a HO alt buys you is shorter runtime, for many use cases that is irrelevant.

My point is that a good VR is required for LFP, regardless of the alt's current output.

We are on the same wavelength then. Cheers.

[OceanPlanet]

I hope that readers on this forum realize they are getting access to some very smart and experienced contributors here (John61ct, MaineSail and Hoopla being prime examples). I've been in the Li battery game for a long time and don't know where else (aside from an ABYC Standards committee meeting) where one would find better info.

[jkleins]

Originally Posted by Hoopla

Jim

This thread numbers hundreds of pages and the answer to your questions are contained therein.

However a quick and dirty starting with just the typical "Drop-In" battery packs in the market today.

LFP cells when linked together provide a pack voltage usually 12 volts nominal. Depending on pack capacity that may range between just say 4 cells or more.

The largest manufacturer of LFP cells in the world are from China. The quality assurance from those manufacturers is like chalk and cheese. Read cell's assembled by hand in a tent to high end manufacturing. The high end manufacturers are either owned by or reliant on the Chinese Govt largely for their military production. So even from the most reliable and 1st tier Chinese suppliers like CALB, WINSTON etc the western world gets the left overs after their domestic and military demand.

For instance If you break open today one of the world's premier Factory/Propriety LFP suppliers being the Dutch Masterbolt company you will find WINSTON cells. A few years ago before Genasun departed LFP you would find CALB cells. They departed this space on account of cell QA out of China, despite a superior offering there were problems.

I have broken open a lot of "Drop-Open" battery packs to interrogate the source of their cells. Those even sporting labels of being made in the US or Europe are nonsense. They are mostly Chinese made cells made by 3rd or worse tier manufacturers.

Why is this important, batteries are just batteries surely?

Unfortunately it is not. In the 10 years since LFP went into commercial production first largely for the electric vehicle market, then other uses specs of voltage tolerance have shrunk at both the low (discharge) and upper (charge) end in terms of cell life. There is no reputable LFP cell manufacturer in the world today who supports even with cell balancing (which is horse **** and another subject) a charge voltage in excess 14.4 volts or say 3.6 volts per cell.

The realities are after a pack voltage of 14.0v or 3.5v per cell individual cells start going out of balance and they do that quickly and nothing can restrain them. The cell balancing technology I have no regard for can't cope with that on account it is desparately trying to convert a voltage disparity into a energy or heat transfer. It is like 100 people trying to get out of a small bedroom.

So those voltage specs are from the best of Chinese. Yet "Drop-In" manufacturers use a max charge pack voltage between 14.4 and 16 volts or where after 14.4 volts individual cells are on the brink of death and beyond guaranteed. So why do "Drop-In"vendors not state this? The answer is simple, they no longer have a "Drop-In" for Lead Acid offering.

As for your "why is battery communication to other devices other than via voltage" question so.important with LFP? It is in essence a question about "Drop-In" LFP being acceptable or not for sailboats.

Quite frankly to answer that properly would take me many pages noting the ABYC have been wrestling with this issue for around 5 years now while the elephant in the room being "Drop-In" LFP gets larger by the minute.

So a quick and dirty and so please treat it as that, where every single Drop-In vendor I have noticed world wide simply ignores the following.

1. The energy density in LFP compared to LA is quite frightening unless it is accommodated for by reticulation design including fusing.

2. The selling specs of Drop-In LFP systems are aligned with Lead Acid batteries not actual LFP battery chemistry which leads to them selling expensive door stops.

3. Being a single bus charge/load system battery power is lost in the event of either a high charge or low voltage event without manual and complicated intervention. Think about losing all power suddenly at night in a constricted and or busy seaway.

4. Having no way of communicating to either charge or load sources in the event of either a high or low voltage event (set at rediculous levels) those sources keep doing what they are doing. They can be AC chargers, alternators, wind vane, hydro generator, solar etc. which in most instances say goodbye when the battery or energy storage device is suddenly cut off mid stream. Then there are dual sources of charge and load to deal with like charger/inverter combos to deal with. A "Drop-In" LFP solution simply ignores all of the above.

An Integrated LFP solution deals with all of the above and doesn't just ignore it as opposed to just selling a supposed "Drop-In" battery pack replacing LA and bugger the consequences.

Sorry for the long winded reply.

Thanks. It is clear that you are not impressed with the current crop of drop in batteries. The question that I am asking is not about that but more about the concept of the drop in. It seems that each of the points you make are not related to the concept but the actual implementation. i.e. #1 proper fusing is needed but normally not related to the actual batteries we have now why are they the responsibility of the battery now?
#2 this is marketing, not anything about the actual product. If they were more clear about charge current,etc would the product concept still be bad?
#3 doesn’t the whole concept of the BMS being built in mean that each battery might shut down if there were a BMS malfunction but it is not guaranteed that all the batteries would shut down at once unless the charge source was indeed faulty. Wouldn’t that be a problem related to the charger not the battery? Don’t they come back online as soon as the charge source is removed and a load applied?
#4 this seems to be a real issue but it seems that many just put a lead acid battery in the charge circuit to accommodate this. Is this a bad idea?
I have read this entire thread as well as many related to electric cars and RVs and I am just as confused as you think I am. I am just trying to sort between bad implementation/instruction/marketing that some of the manufacturer are clearly guilty of and bad engineering. It seems less clear to me that a good drop in battery and a good charge system installed correctly with clear settings designed for LFP batteries wouldn’t be an advance rather then a setback.

Jim

[nebster]

Hey Jim,

I think you are pretty much right: it is possible to engineer your way around some/most of the issues with drop-in. It is just that, then it is not "drop-in," is it? The marketing of these often belies the realities of the product, to a more egregious degree than most, because it does so in a way that can introduce substantial new safety risks.

I think the cost-benefit curve quickly reaches a point, with many systems, where a redesign optimized around LFP characteristics makes much more sense.

[northwestsailor]

I just installed the following system in a Sprinter Van to better understand LiPo management and care. Once I get familiar with it I hope to install a similar but larger system on the boat.

[nebster]

Originally Posted by john61ct

For a large bank, putting more than 2-3 series strings (as each 12V unit is, 4S) in parallel can lead to shortened lifetimes from balancing issues.

I haven't seen drop-ins with hundreds of AH capacity, so that's pretty limiting.

Notwithstanding the many other issues with attempting to "drop in" an LFP-for-LA replacement, this may not be important for most of our use cases.

The data that's available (power LFP cells charged and discharged at 4.5C) suggests a fairly mild degradation even with a substantial IR mismatch. Even the worst-case results on individual test packs demonstrated close to 2,000 cycles before reaching 75% of initial capacity. You can check it out here.

If you are going to charge and discharge at fractional C rates, like we mostly all do with an ESS, the effect almost certainly diminishes. (Though they do not present data to support this; they only surmise it.)

So, does it matter? Yes, a little bit. But I don't think that it is a first-order concern. It shouldn't drive the design by itself, and anyway, paralleling strings is the most flexible way for a substantial LFP battery to be built up and sold commercially as standard modules/SKUs.

As an aside, I have a 16s-then-6p pack in use since April. It is too early to spend time measuring string or cell level performance at this point, since there have only been about 50 cycles, but qualitatively things look pretty good. I haven't observed any meaningful "eddy currents" or meaningful current imbalances at the top of the charge cycle. I must be either lucky, good, or we don't have enough data yet.

[john61ct]

Originally Posted by nebster

it is possible to engineer your way around some/most of the issues with drop-in.

Several of the specific disadvantages I mentioned above cannot be worked around.

I'm pretty sure there are more I missed.

The biggest one is unknown quality on the internal cells, CALB Winston etc have built their rep over quite some time.

The drop-in segment is very immature, lots of scamming hucksters out there, and even the likes of Battle Born will take a long time to engender as much confidence.

[john61ct]

Originally Posted by Hoopla

LFP cells when linked together provide a pack voltage usually 12 volts nominal. Depending on pack capacity that may range between just say 4 cells or more.

4S is the only way to get to nominal 12V. 8S is 24V, and so on

> There is no reputable LFP cell manufacturer in the world today who supports even with cell balancing (which is horse **** and another subject) a charge voltage in excess 14.4 volts or say 3.6 volts per cell.

This is not under control of the maker, LFP is LFP. The (to me stupidly) high charging voltages they spec are not designed to maximize longevity.

I agree with avoiding the need for frequent re-balancing, and agree live / active balancing seems so far not worth the added risk from complexity.

> Yet "Drop-In" manufacturers use a max charge pack voltage between 14.4 and 16 volts or where after 14.4 volts individual cells are on the brink of death and beyond guaranteed. So why do "Drop-In"vendors not state this?

Just like the quality prismatic cell makers,

making easy sales in high volume trumps longevity concerns

their concern for lifespan does not extend past the rated cycle lifetime, apparently a fraction of what is possible

and in fact think about it, many industries capable of producing products with a century long lifespan somehow do not do so.

> The energy density in LFP compared to LA is quite frightening unless it is accommodated for by reticulation design including fusing.

Please clarify this, I don't understand.

[Delfin]

Originally Posted by john61ct

I haven't seen drop-ins with hundreds of AH capacity, so that's pretty limiting.

Look some more. I have 2 x 300 amp @ 24vdc Lithionics LIFePO4 batteries, and the BMS manages them as a single bank, so there are no more balancing issues than there would be if it were a single battery.

[Pelagic]

This has been a fascinating technical thread but at 6000 posts is it beyond most cruisers who are just now considering the battery change?

I will start a new thread:
"Lithium Ion for Dummies" and hopefully the fantastic gurus on here would advise the less specialist captains on here with a redesign solution for their own boat's needs?

My own boat is not unlike the Delos 24v system / 8 x 8D 12v batteries connected series- parallel who have just released a video of their electrical changes.

https://youtu.be/GvDCLtN22jY

It would be interesting to hear the critiques from the knowledgeable contributers on here

I will still follow technical one, but maybe the Dummies one can answer more basic user questions.

[john61ct]

Originally Posted by Delfin

Look some more. I have 2 x 300 amp @ 24vdc Lithionics LIFePO4 batteries, and the BMS manages them as a single bank, so there are no more balancing issues than there would be if it were a single battery.

I did not realize Lithionics sold **any** drop-ins, much less that big.