LiFePO4 Batteries - Okay Tear Me Apart ;-)

[s/v Jedi]

Originally Posted by Viking Sailor

Nick,

No confusion here! I don't even have a start battery. What I do have are two 8D AGMs and two nice big rotary switches. One switch selects for the house loads and the other selects for the starter and windless loads. This arrangement allows me to isolate and charge each battery fully while underway. Which, for the longevity of AGMs, appears to be more important then splitting the depth of discharge between two batteries in a single bank. Additionally, if one battery should fail I am but a switch throw from a good battery.

I know that your setup sounds very logical and good to you and I know it works. But you should do some calculations on the number of hours you need to charge each battery from let's say 50% SOC to full. When you have that number, calculate the time for charging both batteries switched parallel from 50% SOC to full and you will find that there are better (more optimized) solutions. Acceptance rate is the issue during the last 20%. Even if you use a not so big alternator, it's probably doing very little for the last 20% and it could do both batteries at the same time.

If you can't charge both to full, I would suggest to upgrade alternator(s) to bigger one(s).

So you are right that fully charging is important for AGM's and you also know that deeper cycles means less life span. But you made a choice for one of those while you can do both... which is better ;-)

ciao!
Nick.

[noelex 77]

Originally Posted by s/v Jedi

I do not agree with the multiple battery setup that Viking Sailor states we have. I don't have that, just a few of the thousands of boats I know have two house banks and switch loads between them, instead of one big bank.

cheers,
Nick.

I have 2 house banks, but agree not many boats do. It has been more useful than I expected having the option to divide the house bank. With the option of lithium I can see it having even more value with the option to isolate different battery chemistries.

[imcbride]

My father always did, so... I also have two house banks and a separate starter bank. There has been more recent movement towards a single house bank. My fear here would be a single cell in the LFP bank failing catastrophically, leaving the house bank with either a complete fault condition, depending on the BMS, or very low voltage (at best 10.8V).

With the option of lithium I can see it having even more value with the option to isolate different battery chemistries

The only problem with different chemistry is different charging profiles. Once we've started tuning our charging systems to get the most out of LFP batteries, they could fry mere mortal lead batteries. For the interim, though, using lead-acid chargers at <14.4V, you should be able to mix LFP and some lead cells. Just be sure of your charging profiles on both your alternator and AC charger.

Iain

[OceanPlanet]

Originally Posted by imcbride

My father always did, so... I also have two house banks and a separate starter bank. There has been more recent movement towards a single house bank. My fear here would be a single cell in the LFP bank failing catastrophically, leaving the house bank with either a complete fault condition, depending on the BMS, or very low voltage (at best 10.8V).



The only problem with different chemistry is different charging profiles. Once we've started tuning our charging systems to get the most out of LFP batteries, they could fry mere mortal lead batteries. For the interim, though, using lead-acid chargers at <14.4V, you should be able to mix LFP and some lead cells. Just be sure of your charging profiles on both your alternator and AC charger.

Iain

This particular discussion brings up another facet of the Genasun system design. On anything other than a very small boat they use two banks in parallel to make (in effect) one bank. It is two banks acting as one. The difference from a normal parallel arrangement is that each of the two banks has it's own BMS master and each is individually protected. So, if for any reason there is a LVC on one of the banks (low/failing cell, whatever) you still have power. Perhaps a bit obsessive of a safeguard, but better safe than sorry at that level of system cost.

[noelex 77]

First congratulations on the posts. I think this type is forum is fantastic at disseminating information about new technology.
I have been using small (2 and 3 cell 1Ahr) LiFePO4 Batteries in a few appliances such as some home (boat) made headtorches for a couple of years.
They are great batteries and I think they will replace lead acid technology in boats over the next decade or so.
However I think there is some exaggeration by battery manufactures on the expected cycle life.
My own batteries are starting to show some loss of performance after 2 years and 400 cycles
I would also draw attention to everyone’s experience with cycle life when using lithium ion batteries in mobile phone and computers. (The cycle here is often 0% to 100%, but extrapolation to 20-80% is a long way short of the claims)This battery chemistry would be unsafe on a boat, but the battery gurus do not expect LiFePO4 to have a much longer cycle life than lithium ion.
My final point would be the damage done to LiFePO4 Batteries when held at a high state of charge. Lead acid batteries cycle life starts to shorten considerably if discharged below 50%. LiFePO4 Batteries are much better and can go down to 20%, but lead acid batteries are very happy at 100% LiFePO4 Batteries like to be kept below 80%. I think this will be a limitation, particularly those on solar / wind power.
Looking forward to hearing some real world boat experience.

[electric1]

Originally Posted by noelex 77

I have been using small (2 and 3 cell 1Ahr) LiFePO4 Batteries in a few appliances such as some home (boat) made headtorches for a couple of years.
They are great batteries and I think they will replace lead acid technology in boats over the next decade or so.
However I think there is some exaggeration by battery manufactures on the expected cycle life.
My own batteries are starting to show some loss of performance after 2 years and 400 cycles

Cycle life largely depends on how well you treat those cells. I wouldn't blame entire technology based on a couple of bad apples. There are many Lithium chemistries and they are often mixed in discussions. I'm not pointing at you specifically, I just see it way too often on various forums. Do you mind sharing specific brand and model of your cells so we can look at their datasheets? I have seen many cases ( including my own ) where cells are taken outside of their comfort zone either by mistake or by believing in overstated manufacturer's claims, resulting in shortened life. In my own pack of 40 cells after a year of ab-use I have a few "weak" cells now. They become weak because I often taken them to 3C discharge by foolishly beliving Thundersky datasheets.
Also, early models of LFP cells had various production issues and some people bought bunch of bad cells from China. Since then quality has improved dramatically and we also learned their true comfort zone.
If you undersize your pack by trying to save few bucks, you will end up with higher C levels and shortened life cycle, and will throw good money after bad. But isn't it the same story with any battery chemistry?
Best thing I lerned about LFP is to always oversize your cells as much as you can afford both financially and space/weight wise.

I would also draw attention to everyone’s experience with cycle life when using lithium ion batteries in mobile phone and computers. (The cycle here is often 0% to 100%, but extrapolation to 20-80% is a long way short of the claims)This battery chemistry would be unsafe on a boat, but the battery gurus do not expect LiFePO4 to have a much longer cycle life than lithium ion.

Its a mistake to compare different Lithium chemistries and draw conclusions based on that. LFP chemistry stands out from the rest by far in both good and bad aspects, good being safety , ease of management and long life, bad being poor energy density by volume and weight ( although still 2 times better than Lead Acid ). Poor density is the reason you don't see LFP in cell phones or laptops or even too many EVs. But they make great replacement for stationary banks and house banks where space and weight aren't too critical.

My final point would be the damage done to LiFePO4 Batteries when held at a high state of charge. Lead acid batteries cycle life starts to shorten considerably if discharged below 50%. LiFePO4 Batteries are much better and can go down to 20%, but lead acid batteries are very happy at 100% LiFePO4 Batteries like to be kept below 80%. I think this will be a limitation, particularly those on solar / wind power.

This is not entirely accurate. The devil is in details. LFP cells hate overcharge, but there are no issues with 90%-99% SOC. The problem is how do people measure SOC? Coming from Lead Acid world people confuse voltage for good SOC measure on LFP cells. The key to understand in LFP technology is that it has very flat curve with very steep ends. As long as you stay away from steep ends and stay below high C rates, the cells will have a long happy life. My own pack has about 400 cycles without any issues ( except some weak cells which was a result of my own learning ) and I know many people in EV world with 2-3 years of daily LFP use without signs of degradation. Sure many cells are killed by many people as they learn, but then again, many Lead Acid cells are killed every day despite it being 100 year old proven technology.
I personally don't believe claims of 5000 cycles, but 2000 cycles seems very realistic if used correctly, which is 3-4 times better than best Lead Acid.

[ctsbillc]

I would be very wary about a cheap source of batteries. Counterfeit parts out of the far east is very common, so even a good name does not ensure quality (I've never heard of counterfeit batteries, but they counterfeit capacitors so it is possible). If the batteries carry an unfamiliar name, I would not touch them with a boat hook (even an insulated boat hook). I recently purchased a bunch of cheap NiMH batteries from HK. They were awful, but that's OK, because they did not cost much, and they were only powerig domestic stuff, but its a different story when you're stuck in the middle of nowhere with no batteries.

Lithium Phosphate is great, the weigh savings are certainly real.

Good to see cost is coming down, but I caution again against buying cheap from a source you are uncertain of.

Bill

[electric1]

Originally Posted by ctsbillc

Lithium Phosphate is great, the weigh savings are certainly real.

Good to see cost is coming down, but I caution again against buying cheap from a source you are uncertain of.

Bill

This is 100% accurate. There have been examples in DIY EV world of people getting poor quality cells from Asia from unverified sources. Many Chinese companies claim to be manufacturers, while in fact they just buy cheapest cells from possibly bad production runs and resell them. There are several credible sources of decent quality cells from China. Largest US cell importer for DIY market is EV Components, and those guys established good relationship with Chinese factories and get good quality cells at good prices, although shipping time is still an issue and demand is so high they can't even keep up with stock.
I am currently working directly with HiPower, which is one of largest manufacturers to verify their quality claims and possibly import those cells into US.
Unfortunately there are no US based manufacturers of LFP cells which are competitive and/or willing to market their products in retail market, so we have to get cells from China.

[ctsbillc]

Damn shame, but as with everything else U.S., the U.S. battery companies are outsourcing battery manufacture to China. Even now, the U.S. is STILL outsourcing everything which made the U.S. great to China.

Mind you, I cannot complain I suppose. Most U.S. cars are made up here in Canada.

Bill

[noelex 77]

Originally Posted by electric1

Cycle life largely depends on how well you treat those cells. I wouldn't blame entire technology based on a couple of bad apples. .

Thanks for your detailed comments on this post and others, I think there is a lot we can learn from the EV gurus.
The cells I have used have come from 2 sources disassembled DeWalt drill packs (new) and a French company that unfortunately I cannot recall, (but am still searching for on my computer).
I think the problem is no-one is really sure of the lifespan of these batteries with the sort of life they will lead as a storage battery on a yacht.
The lifespan of my batteries fits in with the findings of others. The electric RC flyers seem to have the most experience with these batteries as they can use many cycles in a day. They report on average 300 to 500 cycles from them and they are prepared to buy the best cells they can get.
These cycles are typically 100% to 20% with high discharge and charge rates. I am sure the lifespan in a marine application will be higher, but I think the high cycle life manufactures are quoting will probably not be seen in practice.( I will be glad to wrong however 2000 100-20% cycles sounds great to me)
I still think they are the future and applaud all those taking the pioneering steps.

[s/v Jedi]

On anything other than a very small boat they use two banks in parallel to make (in effect) one bank. It is two banks acting as one.

Yes, here we go into the definition of "a bank". When you connect two serie-strings of cells/batteries in parallel with fixed wiring or switches, it becomes one single bank. I have three strings in parallel but it's just one bank.

My fear here would be a single cell in the LFP bank failing catastrophically, leaving the house bank with either a complete fault condition, depending on the BMS, or very low voltage (at best 10.8V).

If you have a string of cells in series and one cell fails catastrophically, it will have an internal short which doesn't hurt the other cells at all. The question is what the BMS will do about it.

With the option of lithium I can see it having even more value with the option to isolate different battery chemistries.

Yes I can see that for the period that we decide to "check it out" and start with an extra bank of LFP besides what we already have.

For the interim, though, using lead-acid chargers at <14.4V, you should be able to mix LFP and some lead cells.

In a single bank? madness, what reason can there be for doing that?

Viking Sailor has AGM's, but with LFP I can see a revival of multiple bank scenarios. Reason is that LFP is just happy with not being fully charged, plus you would want a BMS for each series string anyway (I think).

We should put our brains to work to come up with new scenarios for a boat with LFP. The technology is so different that sticking to the old setups and procedures is probably not the "best way" anymore. For example: if you put 10Ah into a LFP bank and you can get almost all 10Ah out of it again (which I still don't believe, would love to see actual test results) then the whole principle of using solar and wind-gen output to directly power items aboard to avoid the losses of charging&discharging LA batteries can go overboard.

cheers,
Nick.

[OceanPlanet]

Originally Posted by s/v Jedi

Yes, here we go into the definition of "a bank". When you connect two serie-strings of cells/batteries in parallel with fixed wiring or switches, it becomes one single bank. I have three strings in parallel but it's just one bank.
.....
We should put our brains to work to come up with new scenarios for a boat with LFP. The technology is so different that sticking to the old setups and procedures is probably not the "best way" anymore. For example: if you put 10Ah into a LFP bank and you can get almost all 10Ah out of it again (which I still don't believe, would love to see actual test results) then the whole principle of using solar and wind-gen output to directly power items aboard to avoid the losses of charging&discharging LA batteries can go overboard.

cheers,
Nick.

Re a "Bank"...yes, it is one bank, but each of the pair can be completely isolated by the BMS while the other is still connected.

Re amps in-amps out, in Coulumbic terms LiFePO4 can be very nearly 100% efficient (see attached discharge/charge graph). However, there is a voltage drop on discharging so the actual watts out is indeed less. I've heard that total efficiency percentage is in the 90's, but haven't seen documentation taking the voltage drop into account.

[ctsbillc]

There are new emerging Lithium technoplogies, carbon nanotubes etc, but they will be a while coming. For now, the Iron Phosphate is the safe option, the others have a disturbing tendency to burst into flames.

On efficiency, the Genasum graph shows the battery giving 200 A/Hr, but receiving 240 A/hr, an efficiency of 83%. This is VERY good, most batteries are about 65% efficient. Ultracap, BTW, will give about 93%, and Aluminum Electrolytic capacitors about 98%. Both these options last about forever, but take up a lot more space.

Bill

[seandepagnier]

I am interested in eventually upgrading my battery bank, maybe to lithium iron.. but the cost is high, someone mentioned $5 per Ah.

I got my AGM batteries (160AH) at $140 total for 87 cents an amp hour, but 37 cents of that was shipping... lithium iron should be cheaper to ship, but even still makes them 5x more expensive, so I wonder if the price might go down.. Anyone know a cheap source?

[witzgall]

The cheapest, reliable source right now seems to be www.evcomponents.com. They have the cells, you will most likely need to source the BMS elsewhere. They order in group buy fashion, so it is not something you order and get next week. It took me 2 months if I recall, to get my cells.

Chris