1400 Ah Lithium from cells

[CatNewBee]

Some more results from testing...

Yesterday I charged the bank to full (bulk to 14V, minimal absorption 60min, float with no charge for another hour) and the BMV / REC BMS set to 100% SOC.

Then discharged 20Ah.

I let the battery rest for 12h and forced an immediate float to 13.5V. I am a little surprised, that the battery took 15Ah until charge current drops from 80A to 0.1A.

That said, the difference between 13.5V end of charge and 14V end of charge are only 5Ah capacity on a 1000Ah bank (in reality it delivers from full to empty 1300Ah), so this are less then 0.5% SOC difference.

Maybe I will reconsider my bulk / absorption voltage settings and also lower the balancer start voltage + balancer bulk voltage, there is no need to go up to 14V just to balance the cells.

3.375V (13.5V) is almost sufficient charge, the recommended end of charge at 3.65V (14.6V) are not a necessity for an almost full battery, you loose only 1% SOC, so 99% is pretty good and safe for a long life.

I have to repeat this test after a year or so to see, if it applies only to new cells or if this is consistent over a longer period of usage.

[john61ct]

If you care for them properly, there should be little to no change in internal resistance even after 5-7 years, actual reduction in capacity below rated begins some time (probably long) after that.

It is good to see confirmation how pointless, from a capacity POV, it is going past the 3.45V (13.8V) stop point recommended for that sort of longevity.

If you really feel the need for active top balancing and that requires a higher setpoint, do some testing without it for a while, maybe the frequency can be reduced to once a month, or every few months, like an equalization maintenance routine.

[CatNewBee]

Balancing is a non-issue, the REC BMS is capable to be programmed at any
set voltage to start balancing (while charging current flows), also a second parameter allows bulk balancing - when no current flow at all.

So for instance:

It is possible to set start balancing at 3.375V when charge current > 0.5A
and do the final balancing at 3.385V what translates to start balancing
while charging when the first cell reaches this value / stop balancing at 13.5V on float when all cells are even. As long as one cell is higher than 3.385V it will be balanced regardless of the current.

If there is no current and no cell exceeds 3.385V, cells remain as-is,
also if all cells are almost even, no balancing will take place at all.

This is great to have the opportunity to tweak it to the point exactly you want to have it without the need to go high on absorption just to balance from time to time.

I have my charge controller set to bulk / absorption at 14.0V, but according to the new findings it is safe to stop earlier at 13.8V or even 13.5V without sacrificing capacity.

[john61ct]

Which has been clearly stated by dozens of previous users in the sources I referenced.

But the certainty of verifying yourself is a great thing, kudos for at least being open-minded (and technical) enough to test.

[CatNewBee]

Yes, I know, but you have to see / prove it to believe it.

Especially if you have been using lead acid / AGM / GEL batteries for a long time, where charging to the highest possible limit to 100% SOC is essential to fight sulphation and keep the batteries happy over a long time.

I wonder why so many electronics supplier even specialized in the LFP market do not provide their fixed voltage balancer for lower starting voltages and charge programs for LFP with lower voltage settings. Even Victron / Mastervolt use higher default settings for their LFP batteries / BMS / charger profiles.

It would be sufficient to just bulk charge to 13.5...13.6V and stay there, or allow equalization lets say 200mV above this bulk/float for some time for balancing. There are lots of lead acid charger around with a setting for "storage mode" using constant voltages around 13.4-13-6V that would be suitable for LFP out of the box without a special LFP profile.

[Waveda]

Mai I ask which induction hob you plan on using? thanks

[CatNewBee]

Siemens EX275FXB1E iQ700



It is 35cm x 91cm x 4.5cm , because the galley in the L400 is not European kitchen standard 60cm grid depth.

[Waveda]

Thank you. i have a Siemens hob at home and I'll give serious thought to having it on water as well. Best,

[CatNewBee]

The admiral likes it, it is almost totally black, you barely see the dots for the induction areas, very nice.

Also this dimensions are rarely sold because of the usual 60cm grid sizes in common kitchen, chances are you can order one online at some retailer for a reasonable price. I got mine for 740€ with free shipping while the fantasy "recommended list price" from the manufacturer is 2.100€ - some kitchen studios and a few street retailer sell them around 1000€. It's worth looking in price search engines / Amazon / eBay.

[Waveda]

excellent tip thanks. Once you get so much lithium, it makes sense to just get rid of the gas fuelled cooktops. Thanks!

[CatNewBee]

This one is quite nifty, you can go to the settings and limit the maximum power by the power management settings in steps between 1000W, 1500W, 2000W and maximum (3500W with power boost) per side.

Especially if you do not have the big inverter this can make life easier on board.

However, I figured out how to use my old 2kW MassSine with the new 5kW Victron automatically to have enough juice without the power management settings.

I have the big bank of 1000Ah, however this would be doable with a much smaller LFP bank - I would say from 400Ah up. I can cook several days without re-charging, and LFP batteries can have 1C continuous discharge current and up to 3C peek current - that said the capacity and the power should be OK if your solar is able to re-charge what you use.

I am pretty confident I do not need more capacity than the 1000Ah for an all-electric galley on top of the systems on board.

The tread starter (exitstrategy) has even 50% more on board with a huge solar array. I guess he will have a lot of fun with his setup.

[CatNewBee]

Here is my revised lagoon 400 - 220V sources plan.

[john61ct]

Originally Posted by CatNewBee

I wonder why so many electronics supplier even specialized in the LFP market do not provide their fixed voltage balancer for lower starting voltages and charge programs for LFP with lower voltage settings.

Specs are driven by the desire for profits, not the interest of the consumer.

I think the VARs follow the batt manufacturer specs.

Those come out of EV and military gear usage, where range/per-use capacity is everything and expected lifespan is short anyway.

Gentler House bank storage use cases are .0001% of the cell manufacturer's market, so therefore no testing / different specs are developed for that use case; that has been left to the end users, and the more objective/technical vendors like Maine Sail.

And consider that the auto industry could easily have developed utilitarian cars and parts supply chains, so that consumers could routinely drive them for 50-100 years. For purely rational free-market economic reasons, that has never happened.

I bet most people here can connect the dots without calling me a conspiracy nut.

[CatNewBee]

Here the to be changed wiring for the 12V side (charging, BMS and Inverter)
The battery fuses are left out, the thick wires will be double 95mm² (3/0) wires or bare copper connector bars 30x5mm square.



Not too many changes - just disconnect the charger positives from the battery distribution point at the positive input on the master switch to separate charge / discharge path and connect them to the legacy charger solenoid controlled by SOC from the BMV. The whole charge path including the solar controller can be dropped by the (OverVoltageProtecion) OVP relay signal of the BMS, the (UnderVoltageProtection) UVP relays will drop the Inverter and the rest of the 12V circuitry by a separate relay. I will put two separate relays because of the high currents, the 2000W Inverter remains connected via the manual 12V breaker switch with the rest of the boat. The charger and inverter in the Victron Quattro are directly controlled by BMS signals opto-coupler outputs), so there is usually no need for the solenoid. It is just for safety and remote shut down.

So this is the plan. The BMS is able to drop everything, the manual switches only disconnect the legacy circuitry, the solenoids can be switched off and locked manually if needed, the will be just behind the manual panel - also the main switch of the BSM will be wired to the central panel and can shut the whole thing immediately in case of an emergency / fire or whatsoever...

Almost all original wiring will remain untouched, what is a good thing. The installed Mastervolt Smartshunt + Gauge will be removed / replaced by the BMV712. The SmartShunt is simply not able to handle loads beyond 500A and too complicated to program / use, also the installed display is awfull - it will go to eBay after the refit together with the propane oven / cooktop.

[OceanSeaSpray]

Originally Posted by CatNewBee

Some more results from testing...

Yesterday I charged the bank to full (bulk to 14V, minimal absorption 60min, float with no charge for another hour) and the BMV / REC BMS set to 100% SOC.

Then discharged 20Ah.

I let the battery rest for 12h and forced an immediate float to 13.5V. I am a little surprised, that the battery took 15Ah until charge current drops from 80A to 0.1A.

That said, the difference between 13.5V end of charge and 14V end of charge are only 5Ah capacity on a 1000Ah bank (in reality it delivers from full to empty 1300Ah), so this are less then 0.5% SOC difference.

Maybe I will reconsider my bulk / absorption voltage settings and also lower the balancer start voltage + balancer bulk voltage, there is no need to go up to 14V just to balance the cells.

3.375V (13.5V) is almost sufficient charge, the recommended end of charge at 3.65V (14.6V) are not a necessity for an almost full battery, you loose only 1% SOC, so 99% is pretty good and safe for a long life.

I have to repeat this test after a year or so to see, if it applies only to new cells or if this is consistent over a longer period of usage.

Any voltage up from 3.4V/cell will eventually fully charge a LiFePO4 cell in good condition. The relation between end-of-charge voltage and the capacity reached can be found here.

The reason for charging and absorbing at 14.0V is preventing memory effects from setting in. At first, they seem to charge fine at lower voltages, but over time the upper knee starts creeping backwards on the charge curve and capacity gets out of reach.

Charge properly, cycle properly, don't recharge unnecessarily.

The internal resistance of the cells starts increasing slowly from Day 1. If you take a battery and put enough load on it to be able to see the voltage drop, you can see it slowly increasing year after year.
They don't so much "lose capacity", it is just that there is a point where they won't sustain the voltage any more for a given current. For our applications, it looks like it is going to take a very long time before it becomes an issue, but old cells are old cells and never as good as new ones.