LiFePo4 Power -- Controlling Alternator and Inverter/Charger

[Maine Sail]

Originally Posted by nebster

In real life, the "surface charge" on LFP sticks around so long that it biases the measured voltage toward either the charge or discharge curve for so long that it is impractical to get a reading.

I suspect "in real life", meaning practical experience with actual LFP cells, is important info to consider..

The statement regarding surface charge "sticks around so long" is accurate as well. I currently have some LFP test cells that have been sitting on the test bench, after being top-balanced to 3.60V, that are still resting at 3.4032V after 21 days. This at an average room temp of about 72F. Unlike LA batteries, the surface charge, of an unloaded LFP battery, takes a long time to dissipate..

This fragile high resting OCV will drop quite drastically even with just a small load because there really is no quantifiable capacity at that voltage.

[john61ct]

Gibbo discusses a quiet different usage of the term here

http://smartgauge.co.uk/surf_chg.html

but I imagine not very relevant for lithium chemistries.

[nebster]

Originally Posted by john61ct

The idea of "surface charge" afaic is only at full SoC, above 3.37V after resting isolated 24 hours

Okay, you can use whatever name you want for it. (I did put in in quotation marks for a reason.) I don't have another, official word for it.

But what I am talking about is the difference between the voltage under current versus open circuit, and your assertions about working with the two betray your inexperience, again.

[nebster]

Originally Posted by john61ct

It stands to reason, when the 100% SoC definition has shifted down by 10% or more, precise measurements will reveal slightly different chemistry behaviour, including the V vs SoC curve shifting.

I mean, do you have any evidence of this you can share with us?

Because it certainly does not stand to reason.

The slope I wrote about above (~1.7mV/SOC% per cell) is a function of SOC, which itself is already (and always) in proportion to capacity... whatever that capacity may be, regardless of how much it "shifts down."

When the capacity goes down over time, the slope stays the same. SOC is unitless.

If you are suggesting that the characteristic shape of the discharge curve under load changes in older packs, I encourage you to present your evidence. Mine shows that the curve looks pretty much just the same, even after years and years of cycling, for well taken care of cells. (Certainly good enough for eyeball estimating of SOC from voltage or for using voltage-based float as a way to hold a stable SOC, which are the practical things those of us using these batteries are talking about.)

Yikes.

[nebster]

Originally Posted by Delfin

That said, I guess I am still not convinced that holding the pack at some nominal voltage with constant small loads won't have undesirable consequences.

Yeah, it's certainly possible. The regulation is not perfect; the charger cannot respond instantaneously to the load, so there is always some buffering happening and then catch-up.

I doubt anyone's going to produce a study on this, but I do think it's at least nice that we, collectively, know about it and can try to look for it in our own "experiments" over time.

I have a suspicion we're not going to find much, though.

[nebster]

Originally Posted by Maine Sail

The statement regarding surface charge "sticks around so long" is accurate as well. I currently have some LFP test cells that have been sitting on the test bench, after being top-balanced to 3.60V, that are still resting at 3.4032V after 21 days. This at an average room temp of about 72F. Unlike LA batteries, the surface charge, of an unloaded LFP battery, takes a long time to dissipate..

It's a good thing we have patient Maine Sail to sit there and watch that counter tick down for three straight weeks.

[Inkog]

Originally Posted by john61ct

By aging I do not mean in calendar terms but functionally, by performance and capacity dropping.

With quality LFP cells well-coddled that does take many years to even start.

It stands to reason, when the 100% SoC definition has shifted down by 10% or more, precise measurements will reveal slightly different chemistry behaviour, including the V vs SoC curve shifting.

No, LiFePO4 cell voltage doesn't fundamentally change in relation to (diminished) capacity. The phase change of Li and FePo4 dictates cell voltage which is ~3.4V. Capacity is a function of how much Li the cell contains. Capacity is lost when free Lithium in the electrolyte is destroyed or bound in other compounds. Over heating (and overcharging) can can destroy free Li and bind Li to the graphite anode (while charging) forming Lithium Carbonate (Li2CO3). This is similar to lead acid sulfation, but you can't reverse it. This is why LFP has such a narrow operating temperature range and why you should current limit your charging if needed. Electrolyte breakdown can occur as low as 120F depending on the specific chemistry. Its a similar principle when cold. Check your battery's recommended charge and discharge parameters.

[john61ct]

Originally Posted by nebster

what I am talking about is the difference between the voltage under current versus open circuit

afaik that's the way to say it, just clarify not resting

voltage sag for the difference, loaded voltage

and charging voltage the other way.

I thought you were actually referring to surface voltage, as in a higher V that does not represent any useful delta in stored energy.

Thanks for clarifying.

[john61ct]

ESR / internal resistance is closely related to V vs SoC correspondence both loaded and charging.

That is easy to measure, and changes pretty drastically as LI banks age, i.e. lose capacity.

In an EV / propulsion context, this can happen in a relatively short time, certainly over just a few years.

When sorting through a large batch of mixed-health cells, IR is an important factor, separately from capacity testing, in assembling packs where the cells' behaviour in use will match as closely as possible.

Those dealing with only top quality cells still in their healthy stage of life and low C-rate use cases, may not have experienced these issues. As they get past (what should be) EoL, that V vs SoC correspondence and other aspects like thermal behaviour, is dramatically different from newer healthier cells.

[tanglewood]

Originally Posted by Delfin

Not so much. Mine will read 26.2 volts from about 80% SoC to around 30% under modest load. There is no way for me to know what the SoC is from voltage, except when full or below 30%.

What are you using to measure/read the voltage? I wonder how accurate it is.

[Delfin]

Originally Posted by tanglewood

What are you using to measure/read the voltage? I wonder how accurate it is.

A Blue Sea digital voltmeter installed at the bank, an SG200 and a Link 20, cross checked with a Fluke multi meter. All agree.

[Gregr]

For what it's worth we had the same sort of boat as you for 14 years (just sold it). The original battery banks were totally inadequate, so at the time we replaced both banks with 2v flooded motive lead acid batteries. Each bank 325 AH. Problems went away, charged via gen set or motor and batteries were still ok when I sold it after 13 years use.


I like your current solution however, would just take a deep breath on the cost difference.

[Pete7]

Interesting conundrum.

Dockhead, is the primary aim:

a. To find a suitable replacement battery system to last 2-3 years before selling the boat?

b. As a science project to learn more about LFP before designing a system for the new boat?

c. Both.

If A: then dropping £10k into the boat doesn't really make financial sense. Most of the LFP adopters on CF presumably intend to keep their boats long term and therefore want LFP for the major advantage of 3000 - 5000 cycles. However, if your plan is to sell in 2 years you won't benefit from this but you will be paying the up front costs.

Also worth thinking about future buyers. Will they look upon a home brew LFP system with glee or horror? will they have enough understanding of the chemistry to be able to continue to manage systems, or will it end in tears in short order? Replacing the LFP because they 'cocked it up' shortly after taking delivery would leave a bitter taste for them.

If B: well that's different and could be interesting. The long term float issues are clearly a problem. They need solving ready for the next boat; having been thoroughly tested on the current boat.

Is there a solution with a mix of battery chemistry? You currently have two banks inter-connected providing 24v. If the bank under the saloon floor which I think is 6 or 8 GP31 in size were converted to LFP, you ought to get 400AH at 24v in there. I note Mainesail clamped the LFP cells in tightly so you will probably need a new box to do likewise.

The second bank at the stern convert to carbon / lead batteries. 4 x 100AH should give a usable 100AH @ 24v for £1000 in a GP27 size battery and a claimed 2000 cycles

https://www.alpha-batteries.co.uk/12...isure-battery/

How does this work in a marina? well assuming shore power is available, the LFP bank is disconnected and sat at say 50% SOC. Can't see any advantage to using LFP sat in a marina.

The pure-lead-carbon (PLH+C) are fed by the multiplus and in the event of a sudden pontoon supply failure, which isn't unknown, the PLH+C take over supplying short term low power items on board. For example lights and hydronic heating, kettle for tea until such times as the pontoon supply is restored. Also emergency option if LFP system goes pear shaped.

How does this work at sea or anchor? Connect LFP in and remove PLH+C from system. Charge up LFP before leaving port or with engine / genny at anchor. Maintain LFP once a day with short charge from genny which also creates hot water etc and runs washing machine. Turn off genny when reaching 100% SOC and run LFP supply through multiplus to feed mains throughout the vessel.

In 2 years time buyer sees two systems, one quite conventional with a 5 year warranty on the batteries. The other bank (LFP) new technology with major advantages plus up front cost already paid, but PLH+C option to fall back on in the event of a problem.

Hint; if you subsequently decide you don't like the PLH+C put me down for two

[Pete7]

Interesting YT video and on topic for this forum. Nice to see real life testing and demonstrating problems rather than just internet hearsay.

Pete

https://www.youtube.com/watch?v=jgoIocPgOug

[Dockhead]

Originally Posted by Pete7

Interesting conundrum.

Dockhead, is the primary aim:

a. To find a suitable replacement battery system to last 2-3 years before selling the boat?

b. As a science project to learn more about LFP before designing a system for the new boat?

c. Both.

If A: then dropping £10k into the boat doesn't really make financial sense. Most of the LFP adopters on CF presumably intend to keep their boats long term and therefore want LFP for the major advantage of 3000 - 5000 cycles. However, if your plan is to sell in 2 years you won't benefit from this but you will be paying the up front costs.

Also worth thinking about future buyers. Will they look upon a home brew LFP system with glee or horror? will they have enough understanding of the chemistry to be able to continue to manage systems, or will it end in tears in short order? Replacing the LFP because they 'cocked it up' shortly after taking delivery would leave a bitter taste for them.

If B: well that's different and could be interesting. The long term float issues are clearly a problem. They need solving ready for the next boat; having been thoroughly tested on the current boat.

Is there a solution with a mix of battery chemistry? You currently have two banks inter-connected providing 24v. If the bank under the saloon floor which I think is 6 or 8 GP31 in size were converted to LFP, you ought to get 400AH at 24v in there. I note Mainesail clamped the LFP cells in tightly so you will probably need a new box to do likewise.

The second bank at the stern convert to carbon / lead batteries. 4 x 100AH should give a usable 100AH @ 24v for £1000 in a GP27 size battery and a claimed 2000 cycles

https://www.alpha-batteries.co.uk/12...isure-battery/

How does this work in a marina? well assuming shore power is available, the LFP bank is disconnected and sat at say 50% SOC. Can't see any advantage to using LFP sat in a marina.

The pure-lead-carbon (PLH+C) are fed by the multiplus and in the event of a sudden pontoon supply failure, which isn't unknown, the PLH+C take over supplying short term low power items on board. For example lights and hydronic heating, kettle for tea until such times as the pontoon supply is restored. Also emergency option if LFP system goes pear shaped.

How does this work at sea or anchor? Connect LFP in and remove PLH+C from system. Charge up LFP before leaving port or with engine / genny at anchor. Maintain LFP once a day with short charge from genny which also creates hot water etc and runs washing machine. Turn off genny when reaching 100% SOC and run LFP supply through multiplus to feed mains throughout the vessel.

In 2 years time buyer sees two systems, one quite conventional with a 5 year warranty on the batteries. The other bank (LFP) new technology with major advantages plus up front cost already paid, but PLH+C option to fall back on in the event of a problem.

Hint; if you subsequently decide you don't like the PLH+C put me down for two

The main reason for this is to gain experience. It is possible I will be able to start the new build next year which means this boat could be on the block within two years, so of course such an installation won't pay for itself, but it looks like £5k or £6k or so will be an adequate budget so it's worth it just for the experience.


And who's to say the boat won't sell better with it. We see more and more lithium installations among sophisticated cruisers in Europe, and a well done custom installation with real engineering drawings which demonstrably performs well should be a plus not a minus. Besides that, my current installation is a mess (as you've seen) and will probably reduce the value of the boat.


The middle way would be to order new battery boxes and put in T105's, but I would still have a crappy split bank with all of its drawbacks. To eliminate that would be really hard to do as there is no place in the boat where I can get all the batteries I would need in one place.



I suppose there is some way to manage and balance split banks, but I doubt it would be cheap or easy. So on my boat even lead will be expensive to do properly.


Therefore, lithium seems like an elegant solution.


And yes, if you saw the other thread with the drawings, it would have a lead reserve bank -- basically I would leave in place the lead bank under my bunk, charge it from the lithium with a B2B charger, and feed the house loads through it. This would increase total capacity somewhat and simplify the wiring. So in the end I would have maybe 300 amp hours of lithium and 210 amp hours of lead for usable capacity of maybe 345 amp hours at 24v, which is more than 150% of what I have now.