[B]Best practice charging for LFP & LFMP bank?[/B]

[BigBeakie]

Originally Posted by kenbo

To BigBeakie's OP, I'm afraid there is no data on actual side by side charge cycle effects of 13.8v vs 14.4v. Only anecdotal and some non-controlled empirical evidence has been provided to date.

I do not subscribe to the syncophant philosophy that states if "x" person states this is the way to go it must be alpha and omega of battery science.

I have the exact same questions that you do about these drop in batteries and a few more.

I'm concerned about using them in a parallel installation to attain the AH target needed but maybe overly so.

I'm concerned about their rather low amperage discharge limits of anywhere from 40A to 150A depending on battery.

Hopefully we can get some answers.

And finally, at least for me, battery vendor bashing is not helpful especially when it's done on either second hand info or supposition.

Thanks kenbo, that's a good summary of my mindset at this time.

On your point of concern about the amp delivery capability, our application is the batts driving 2x 15kW electric motors at full power. So that sucks alot of juice, and the only 2 brands that I know have been proven to deliver the 2C requirement are Victron and SuperB. So it is an issue with some brands that I'm aware of.

As to this idea that companies like Mastervolt, Victron, SuperB with their "drop-in" batteries are deliberately being misleading on this charging voltage issue, this does seem to tilt towards the conspiracy theory mindset. They have their corporate survival at stake if they can be shown to know that their specifications are misleading. Personally, the Victron claim that they get 10,000 cycles at 70% DoD and retain 70% capacity is pretty attractive to me, if true.

I know they test the hell out of their batteries, but there is no data published that I can find. I was hoping to hear from CFers if anybody had done the reference checking.

I have not dug that deeply yet, or got on the phone with their tech engineers, so it looks like I'll have to do that and ask the hard questions.

[hellosailor]

Originally Posted by senormechanico

I made a small 4 digital voltmeter display under my companionway steps.

Those meters could be deceiving you.

Start with an accuracy of 1%. So with a 4-volt cell, 4.00 could mean it is really 3.96-4.04 volts. Then, there's a small Chinglish line in the specs that mentions a "1-3 cm error" and since I doubt they are having a centimeter problem, I'd say they mean a 1-3 RMD (right most digit) error.

So 3.96 now may show 3.99-3.93 and a real voltage of 4.04 could show 4.07-4.01.

So now you've got indicated voltages of 3.93-4.07 all really showing the same 'real' 4.00 volts on each cell.

I'm not sure you could use those meters to reliably balance the cells, when there is potentially a 4% difference between the cells, all being indicated as being the same voltage.

(Gotta love this digital meter stuff.)

[john61ct]

I am not just paying attention to Maine Sail, and the lower-volt recommendations did not originate with him. Across dozens of forums many long-time users report the same. In Australia especially, heavy caravan use of LFP has been status quo for years now.

Who has been vendor-bashing here?

Originally Posted by kenbo

I have the exact same questions that you do about these drop in batteries and a few more.

I'm concerned about using **them** in a parallel installation to attain the AH target needed but maybe overly so.

I'm concerned about **their** rather low amperage discharge limits of anywhere from 40A to 150A depending on battery.

What cells are referenced by "them" above?

This is a drop-in 12V



This is a "packaged system" product.



The main difference is whether the cells are accessible or the BMS is available for integration with outside monitors, relays etc.

Both charge and discharge rates are far higher for LFP than any lead. 3-4C is no problem.

180AH cells can therefore pump out 300+A long as wiring infrastructure can handle it.

A cheap 12V drop-in internal (inaccessible) BMS may have lower limitations, but that is nothing to do with the chemistry type.


______
I prefer at least two parallel strings for redundancy.

Rule of thumb is to limit strings to three, maybe four as a maximum if necessary,

say the waiting time for 300+AH cells is too long for you, so you choose 180AH CALBs, at 12V 4s3p gives 540AH, a pretty reasonable capacity.

Going to 4s4p 720AH is pushing it but IMO not likely to cause balance issues, maybe just keep an eye, checking cell Vs monthly even if no problems for years, rather than going to a less frequent schedule.

But if you can, keeping the bank at larger cells and fewer strings is best practice.

[john61ct]

Originally Posted by BigBeakie

our application is the batts driving 2x 15kW electric motors at full power. So that sucks alot of juice, and the only 2 brands that I know have been proven to deliver the 2C requirement are Victron and SuperB.

OK, for a EV propulsion use case, all bets are off, never seen anyone talking about lower voltage regime for that, afaik pertains to far gentler House bank usage.

For such uses, definitely consult with engineers and follow vendor specs IMO.

The fact that huge-amp draws are dominant in the industry are a more likely reason to me than any conspiracy theory.

House usage is such a tiny market, just isn't even on the radar of the actual cell makers.

[Paul L]

Originally Posted by hellosailor

Those meters could be deceiving you.

Start with an accuracy of 1%. So with a 4-volt cell, 4.00 could mean it is really 3.96-4.04 volts. ....

Is the spec 1% of reading or 1% of max reading?

[john61ct]

A professional's quality DMM (costs hundreds) is periodically re-calibrated for 2-3 decimal places accuracy.

Get one to check the accuracy - or at least consistency - of your cheapie voltmeters.

[CatNewBee]

"House Battery" usage means nothing, the usage is not necessarily "moderate".

I will have a house battery of four cells with 1.000Ah LiFeYPO4 each, that will run moderate lighting, navigation lights, radar, electronics, but also the water maker, the diving compressor, the hot water maker, the washing machine, microwave oven and eventually a induction cook top - of course not all at the same time. Also possible to run one of the four AC units for 1-2 hours to cool down before sleeping.

This is a storage of 10kWh usable energy.

There will be 2 inverter, 1 x 5.000W Victron Quattro wit 200A charger and one legacy Mastervolt 2.000W Inverter, that can team up using the power assist function of the Victron to up to 7kW - drawing up to 550A. (0.6C)

The battery will be re-charged by 1.600 Wp Solar (approx. 100A) and eventually by using the generator with up to 380A (200A from the Victron Quattro + the on-board chargers 2x40A + 100A Solar), so a total re-charge can be done in around 2 hours. (max. charging with 0.5C)

So there is not only moderate slow discharge all the time, but also significant peeks of 500A and more. This is the main reason to use LiFeYPO4, lead acid batteries would be killed after a short time on this usage pattern.

The battery cells alone will cost around 5.000€, the electronics around including the solenoids will add another 1.000€, On such a setup I would never go without a proper balancing and monitoring system of the single cell health, SOC, temperature etc. Also the system will automatically detect excess energy available from solar and use it to heat up the water or run the water maker or washing machine by a programmed stand-by chain when SOC is above 90% and the solar array delivers excess energy.

I would never monitor such a system manually. You can do this the first weeks or months, but the system is designed to run for years maintenance-free, people tend to get tired an lazy as long as thing do not go wrong. A good monitoring system is necessary for routine surveillance and alarming if something runs out of sync.

[john61ct]

Yes larger systems are more likely to have sophisticated protection. Not just need it, but makes more sense economically.

500A for short periods is only .5C, yes heavy for House usage, but lighter than EV which deals with that and more, every time you pull away from a red light.

Using for propulsion on a boat is a great Science Project too, very jealous.

Those with say 20-40AH daily draws would have different needs.

[CatNewBee]

Well, that's true, but compared to out-of-the-shelf LFP battery packs from Victron, Mastervolt, SuperB or whatever, I get 1.000Ah for the price of a 200 to 300Ah with the protection and balancing system I want to have.

So I prefer to built my battery myself out of Winston ThunderSky cells and chose + understand the protection system, so i can fix it, if necessary, under way while cruising. A spare BMS with active balancers is about 300€, so no problem to have it handy on board - and even without it - one can survive several weeks charging and discharging with manual monitoring too.

[john61ct]

Could you please specify, ideally link to the devices you are using for LVD / OVD ?

Any temp protection?

[senormechanico]

Originally Posted by hellosailor

Those meters could be deceiving you.

Start with an accuracy of 1%. So with a 4-volt cell, 4.00 could mean it is really 3.96-4.04 volts. Then, there's a small Chinglish line in the specs that mentions a "1-3 cm error" and since I doubt they are having a centimeter problem, I'd say they mean a 1-3 RMD (right most digit) error.

So 3.96 now may show 3.99-3.93 and a real voltage of 4.04 could show 4.07-4.01.

So now you've got indicated voltages of 3.93-4.07 all really showing the same 'real' 4.00 volts on each cell.

I'm not sure you could use those meters to reliably balance the cells, when there is potentially a 4% difference between the cells, all being indicated as being the same voltage.

(Gotta love this digital meter stuff.)

I also have a Cellog 8 wired to the same spots on the cell's terminals.
While calibrating, I used my Fluke multimeter which has been calibrated by the factory in the past couple of years.
The whole system is also metered by my BlueSky Pro Remote monitor.
The Blue Sky reads about 0.1 high for pack voltage when lots of charging takes place due to wiring resistance between it and the cells, but agrees at anything under 15 amps or so.

Thanks for the warning, but I think I'm good.

[evm1024]

Originally Posted by CatNewBee

Well, that's true, but compared to out-of-the-shelf LFP battery packs from Victron, Mastervolt, SuperB or whatever, I get 1.000Ah for the price of a 200 to 300Ah with the protection and balancing system I want to have.

So I prefer to built my battery myself out of Winston ThunderSky cells and chose + understand the protection system, so i can fix it, if necessary, under way while cruising. A spare BMS with active balancers is about 300€, so no problem to have it handy on board - and even without it - one can survive several weeks charging and discharging with manual monitoring too.

I keep reading forward tense in your statements about your battery bank.

Just to be sure that it is not a lost in translation kinda thing could I ask you how long you have been using your battery bank?

What experience do you have with building LiFePO4 banks?

Thanks!

[CatNewBee]

Originally Posted by john61ct

Could you please specify, ideally link to the devices you are using for LVD / OVD ?

Any temp protection?

I am investigating at the moment this one, it looks very promising:

REC d.o.o.

It is completely configurable and has active balancing + isolated ports + isolated relay contacts and a computer interface for programming / monitoring.

The more traditional is the ECS BMS with cell modules, but I would prefer an enclosed dedicated central unit. (LiPro1-6)

https://ecs-online.org/BMS:::2.html

Both have temp. protection.

[john61ct]

Well I don't want the balancing, and $500 added cost per 4-cell block is just nuts anyway.

Just looking for OTS adjustable and reliable LVD and OVD

To handle high amps throughput looks like better to source small VSR and thermostat units and use them to control high-capacity relay/contactors.

Anyone with specific suggestions along those lines, would be greatly appreciated.

[john61ct]

And protection against charging in below freezing temps never seems to be addressed in these systems?