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

[matureee]

Dont try this at home.
Let me make this very clear. I am not advocating an over discharge. Keep your cells inside the manufacturers specs.

Originally Posted by senormechanico

Interesting theory.
A friend of mine recently put in a home grid tie.
72 240 watt panels, three inverters etc.
His backup is 16 700 ah Winstons with a dual BMS (two for 8 cells each)
When he purchased the backup bank, one of the cells had been discharged by a defective BMS which was replaced under warranty.
The cell was also replaced under warranty and the discharged one didn't need to be returned.
Just to see what would happen, he recharged it. It seems to be just fine.
The resting voltage of the cell when removed? 0.5 volts.
When questioned, the supplier (Balqon) said the cell was probably OK because it had been discharged at a low current rate, never being reversed in polarity.
We shall see, but it's an interesting data point nontheless.

BTW, MY home's grid tied 700 aH 24 volt bank from Balqon is doing nicely.
Unfortunately, the power grid doesn't go out like it used to.
Weird way of thinking, I know.

[matureee]

Originally Posted by OceanSeaSpray

Once they have been over-discharged, all bets are off. Note how the supplier didn't want the associated liability...




The main manufacturers seem to be aligned on 2.5V as bottom limit and the absolute voltage limit for the chemistry is 2.0V as far as I know.
A fairly recent study in cell destruction by over-discharge showed that the process caused internal micro short-circuits to appear and the cells ran warmer upon recharge afterwards, but they over-discharged them to set amounts like 110% capacity, not down to a voltage ("Failure Investigation of LiFePO4 Cells in Over-Discharge Conditions", Hao, 2013).

Absolutely agree with the first point. The study used 2.0V as the definition of their 100% DOD and then went beyond that. We regard 2V as an absolute minimum voltage and usually dont want to go anywhere near it.

[spiv]

Well, just to add to the LA replacement discussion, I found this batteries manufactured in China by a QA company: Solar battery, lifepo4 battery, battery packs
They are 110Ah and claim:

• Interchange design of lead acid battery
• Built-in PCB for over-voltage, under voltage and over-temperature
• Series connection up to four batteries (48V)
• Internal cell balancing
• Maintenance-free
• Rugged mechanical design
• Thousands of cycles under normal conditions
• Can be charged with most lead-acid chargers

They are around USD 650ea, I'd like to buy 6 of them, but first I'd like to hear comments....

[matureee]

Originally Posted by spiv

Well, just to add to the LA replacement
They are around USD 650ea, I'd like to buy 6 of them, but first I'd like to hear comments....

Once more see the link in MaineSails posts follow this through "compass marine how to articles" to "lifepo4 batteries thoughts and musings". Read and digest. This is the best summary I have read anywhere of how to use these batteries.
I know NZ is a big country but if OceanSeaSpray is anywhere near you he would be a good resource.

[hellosailor]

"Interchange design of lead acid battery"
So they have found and taken the Holy Grail?


Drop-in replacement and charging with plain wet lead charging systems seems to be a big no-no according to most everyone else. That might mean Mr. Yu is an enterprising salesman, not a battery engineer. Or that he's just wholesaling someone else's batteries.

[goboatingnow]

[*]Interchange design of lead acid battery

Marketing man was let at the spec sheet, note Chinglish

[*]Built-in PCB for over-voltage, under voltage and over-temperature

Engineering were " told" then to make it a " drop-in" , so they added electronics to hide the differences

[*]Series connection up to four batteries (48V)

Only 4, the electronics, internally arnt up to more !!!!.

[*]Internal cell balancing

done by magic , or the marketing man again

[*]Maintenance-free

where have i head that before .....

[*]Rugged mechanical design

as opposed to what, " fragile mechanical design" !!!

[*]Thousands of cycles under normal conditions

sure sure , thousands, a rigorous testing process, carried out by the marketing department

[*]Can be charged with most lead-acid chargers

well until the electronics fries that is.


PS , I have a bridge to sell..

dave

[Jd1]

Originally Posted by goboatingnow

PS , I have a bridge to sell..
dave

How much do you want for the bridge ? Can it go over troubled waters ? Does it come with an anchor on each side to keep it in place? What kind of anchors ? So many questions ......

[toddedger]

Originally Posted by spiv

Well, just to add to the LA replacement discussion, I found this batteries manufactured in China by a QA company: Solar battery, lifepo4 battery, battery packs
They are 110Ah and claim:

• Interchange design of lead acid battery
• Built-in PCB for over-voltage, under voltage and over-temperature
• Series connection up to four batteries (48V)
• Internal cell balancing
• Maintenance-free
• Rugged mechanical design
• Thousands of cycles under normal conditions
• Can be charged with most lead-acid chargers

They are around USD 650ea, I'd like to buy 6 of them, but first I'd like to hear comments....

The problem I see is the internal relays are limited to 30 amp. Thats probably enough for six loaded together, but also means they aren't too robust. So if a cheap relay goes then you're out $650. The other problem is you aren't going to have any way to know what failed or what's going on inside there. If you are going with a 600 amp system, than to most people it would make more sense to have an externally controlled system.

[OceanSeaSpray]

Originally Posted by spiv

Well, just to add to the LA replacement discussion, I found this batteries manufactured in China by a QA company: Solar battery, lifepo4 battery, battery packs
They are 110Ah and claim:

• Interchange design of lead acid battery
• Built-in PCB for over-voltage, under voltage and over-temperature
• Series connection up to four batteries (48V)
• Internal cell balancing
• Maintenance-free
• Rugged mechanical design
• Thousands of cycles under normal conditions
• Can be charged with most lead-acid chargers

They are around USD 650ea, I'd like to buy 6 of them, but first I'd like to hear comments....

The issue is that not all of the engineering issues with LFP are on the battery side. Nothing in the battery is going to do anything about the voltage being too high, for too long etc. The HV protection is always set far too high in order to allow you to connect to any automotive system without tripping it: it just has to be, think about it. It is a "catastrophic failure protection" only, it won't prevent you from ruining the cells gradually. If you trip the HV protection using one of these batteries, it will open-circuit and destroy your alternator: this is the only thing it can do!
Next, if you start paralleling units like those, you end up buying the protection system several times over instead of once. As there is pressure on costs to be competitive and make a profit, you really need to wonder about what is in the box...

Another engineering issue is that you are effectively paralleling strings. In a low battery situation, even current distribution would get lost and you could end up placing the full load onto one battery only, so current capacity needs to be sufficient there without cooking anything. 30A relays are obviously completely inadequate, but they will be cheap alright!

If you build a system from bare cells as proposed in this thread instead, you choose your cells and components and you can make sure that the whole setup is fit for purpose.

When it comes to capacity, you know better than anyone how much energy you use and what charging capability you have, but I have come to think that a lot of LFP banks are made much larger than they actually need to/should be. This approach can in fact deliver less value. There is nothing suggesting that a bank twice as large will last twice as long, you might just end up with twice as many old buggered cells at the same point down the track. Also, with solar systems especially, an oversize bank tends to spend much more time at high SOC, which is detrimental to its ageing. LFPs like: 1/ cycling because it means they don't spend any amount of time near full, or 2/ sitting around, half discharged or even lower.
The bank needs to be large enough so C-rates remain modest in normal operation and large enough to provide the capacity needed between recharges, but beyond that, all what comes out needs to go back in and the size of the battery makes no difference there.
In the old SLA thinking, larger meant less discharging and longer life and there was a justification, but it is almost the exact opposite here.

Just a few things for you to think about.

Eric

[ele2014]

I must fully agree with Eric "OceanSeaSpray"
I have almost two full years Off Grid with LiFePO4 and solar.
I have a small setup with about 80kWh/month of energy use.
I have:
720W PV array (3x240W PV panels)
8 cells GBS LiFePO4 100Ah (24Vdc 100Ah = 2.5kWh)
BMS/charger
2400W inverter used at most one hour per day most stuff works directly on 24V DC

I will do a capacity loss test on the battery this spring probably in a month or so and I expect to not see more than 5% degradation compared with my first capacity test I done about 3 years ago when I purchased the batteries (I have a video with that test so I will try to replicate the test as closely as possible)
My cells where never above 3.55V or below 3V it make no sense to go beyond this limits since there is not much capacity above or below that.
Charging is done as constant current only when the highest cell gets to 3.55V charging is stopped and only restarted when highest cell gets below 3.4V.

If as I mentioned above battery degradation is at 2% / year or below I expect 10 to 12 years of usable life so I will need another battery to last the 25 years life of the panels.
Inverter is good quality and used one hour / day at most so that should also last 25 years and BMS/charger again noting to degrade in there all solid state so 25 years.

So based on this I made a cost calculation as to my cost/kWh over the 25 years life of the system.
Initial investment about $2600 ($1200 battery + $600 PV + $500 inverter + $180 BMS/charger + $120 ground mount some cables and DC breakers)
If battery degradation will be as expected A new $1200 battery will be needed in 12 years so total over 25 years
$2600 + $1200 = 3800$

Total energy use based on my 80kWh /month average is
25 years x 12 months x 80kWh = 24000kWh
$3800 / 24000kWh = 16 cent/kWh

This 16 cent/kWh is an excellent price for OffGrid but you need to keep in mind that I did not include the installation cost I made all that fro free and is not as convenient as grid power since you energy use will need to be based on weather.
It may also be possible that I need 3 sets of batteries over 25 years future increasing the cost but there are better batteries than GBS at this time at least according to spec.
Battery is by far the most expensive part from this equation.

[ericensley]

I bought 4 tesla battery modules. They are 50 lbs each, 5.3KwH 22.2V each, and are about 20x36x3" each. I bought the tesla batteries because they are made in Japan, and they should be 100% quality, I have bought too many cheap Chinese batteries. My total capacity will be 21.2 KwH, and I will be running a 48V system. I may need to add another group of cells to keep me in the range of the inverter.

I have a 16000BTU AC, and want to know what inverter I should buy. I figure the AC will draw about 1600 watts when runnings buy may take over 6000 watts to start. Will a 3500 or 4000 watt inverter start the AC?

Also how do you all deal with corrosion? Do you spray all the terminals with paint, or rubber spray?

I plan to use a small generator 2-3kw to charge the pack during the day. I will also put in solar cells and maybe a wind generator.

Any help would be appreciated.

[CarstenWL]

... if the batteries are really 22.4V it 'might be tough' to do a 48V system.
... with 2 of these batteries in series and than paralleled you will have 10.5KwH capacity I guess.

'But' I am not familiar with the Tesla batteries ...

Regards,

Carsten

[nimblemotors]

You need to be careful with the Tesla Motors cells, as they are nickel-cobalt-aluminum and can catch on fire if overcharged/damaged from thermal runaway.

[offpist]

nimblemotors is right.
If overcharged these batteries can really cause fire so you need to hook it up to an BMS and make it impossible to overcharge.

I would much more recommend looking at the Nissan Leaf battery.
Main reason for this is it can easily be divided in to each cell and you can connect them in parallel/series however you like.

Over the last year i have harvested 11 Nissan Leaf batteries, and helped people build storage for Solar energy.

However i have used Lifepo4 in my boat.
And why is that?
The nissan leaf are not good for 12V setup, as they operate in voltage up til 4,1V and they come connected in pairs. Meaning you have to use 2,4,6 etc.
So you end up with a battery at 16,4V.

But for 48V setup they are perfect.

This article is about my setup in my house.
Its in norwegian.

Selvforsynt med solenergi og batteribank - Norsk elbilforening

[SV THIRD DAY]

Originally Posted by hellosailor

Drop-in replacement and charging with plain wet lead charging systems seems to be a big no-no according to most everyone else. That might mean Mr. Yu is an enterprising salesman, not a battery engineer. Or that he's just wholesaling someone else's batteries.

We just need a little tweaking of that comment in my opinion.
It's true if you just remove your Old Lead Acid and plop in a LiFePO4 bank and people who would do that...well anyone who would do that....well is an idiot, it just needs to be said.

BUT...if your plain wet lead charging system has the ability to set the charging parameters (Voltage) then besides the real risk of overheating and burning up an alternator by keeping it at full output for too long, you can successfully charge your LiFePO4 bank with more standard charge devices.

Examples:

• Both my Xantrex 150A and Sterling 60A battery chargers let me set the charge voltages. They both don't have real LiFePO4 charge profiles, but that's no big deal. I just set the charge voltages for a value that makes me happy and off they go.
• My BlueSky solar controller isn't set up for LiFePO4 batteries, but the same voltage tweak to the controller and the solar becomes a set it and forget it issue at 13.7v for me.
• Same for my alternator reg and wind generator.