Testing LiFePo4 capacity after 3 years of use. Are 4000 cycles realistic?

Testing capacity of 3 years old LFP batteries. Pack built with grade A cells. Really gentle use with low charge/discharge amperage.
Am I going to get advertised 4k cycles?
Battery build video: • Building 23.5kWh batte...

[@boatelectricaldiy]

I install profesionally. A fixture is mandatory for these cells. In larger instalations like yours the expansion can stress the terminals and break the posts. You can't compress the cells now, but i sugest you lossen and then re-tighten the bus bars to take some of the stress off the posts. Drill the holes a bit bigger on the bus bar if you have to. You could also replace with a flexible bus bar. This might require a full disasembly of the pack if things are really tight.

[@SolarEngineering]

thanks for the advise, I did loose and re-tight nuts on top 3 cells

[@boatelectricaldiy]

@@SolarEngineering I'd sugest doing the rest too. They proably have not expanded as much due to the weight on them, but there was no fixture so they will have expanded too.

[@ssoffshore5111]

I personally would disassemble the entire pack, set the cells to a lowish state of charge, then fit in a fixture with springs (valve springs are cheap and work well).

[@boatelectricaldiy]

@@ssoffshore5111 this was my thought the first time I came accross this problem many years ago. The problem that arises is that the fixture ends up cocking the cells back and forth, and makes more stress on indvidual terminals. Also, if the fixture is too tight, the pressure seal on the cells breaks. An ideaI haven't tried this yet, is adding a fixture that is coupled with some sort of calking between the swelling gaps of the cells that would stop the further movement of the cells.

[@jerbear7952]

What about pouring epoxy between the cells

[@kchortu]

I have really come to believe that the 16s1p, 8s1p,4s1p configurations are the correct way to go considering the cost of a good bms isnt that much more than the cost of getting one cell sent to you. I think andys off grid garage is a good channel he has riding the compression non compression debate pretty hard. I also think the more modern (considering where the tech has come in 3 years too) cells are looking like the production techniques are spot on and that the technology is really taking its stride. I just bought some eve mb30 and it was a very easy streamlined process and they seem nice time will tell.

[@Raphael_Hofmann]

Flexible busbars would make a real difference. Also, it is too late for that, but there are two orientations that your cells should not have. On of them is your orientations. On the side, the gas bubbles from the first cycles can not escape and reduce your capacity. Also Fixation of the cells would help with this issue during the first cycles at least.

[@gandalfstormcrow9605]

I second this. The 4,000 cycles is with compression (at least for the first several cycles until gas has escaped). Without compression I thought it full cycle life was more like 2,500.

[@ProspectorTripp]

On my 107,520 watt system I charge/discharge my 280ah lifepo4 cells about 70% from 3.5v down to 2.8v or 56v~44.8 (using only 75,264 watts of the battery potential) Having Eight DIY 48v 280ah batteries in parallel they never get pushed and stay balanced primarily, because I run them at about 70% capacity I feel. This light duty cycle rate has been very stable and over the last two years performed very well. I have one 48v battery that needs to have cell 16 charged up a bit to fully balance the pack so it doesn’t kick off a tiny bit early. I may replace this cell if it doesn’t hold its balancing act 😂 I have the cells lightly ratchet strapped together but not compressed like so many others have done. The only swelling I have seen on any cell was because I was manually charging this one cell and forgot about it for two hours too long and over charged it to over 4 volts. I of course replaced the ballooned cell. This damaged cell sat outside for the last two years in the elements along with one other cell that was slightly weak. (about 5% below spec) I recently took these two cells and installed them in a 1952 Farmall Super C tractor that has a 6v system. I didn’t charge or check them and they kicked that old tractor to life super easily! Amazing LFP cells! I am very happy with my system and have No Cell Swelling Whatsoever, except as explained. Of course they are being run very lightly, exactly as I planned. Thanks for your informative video Merry Christmas Peace Prospector Tripp

[@SolarEngineering]

thank you for details about your system

[@user-sl4ms1oj9o]

Hello friend Can I contact you through your Email? Expertise consultation on assembled batteries

[@captainhappy]

I believe 4:25 someone doing these same kind of tests said that the capacity loss does not happen linear, in the total cycles of battery the capacity loss would be higher at start, then very little capacity loss long time, and then capacity loss would get again higher when battery life is about to end. You probably could confirm or bust this information by doing some tests with shorter time period between them and compare the results if it's linear or is it more likely the bigger capacity changes happen only at start and end of the battery life.

[@MrMandala1111]

Such a shame about the bloating. I heard a battery specialist not long ago talking about compression of cells and the importance of it at the beginning of a cells service life. He explained how tiny bubbles need to be released inside the cell and this only happens right at the start which is why some people compress their cells for the initial top balancing. He said that compressing later on does nothing as it's too late, it supposedly does reduce the life of the battery by missing this step.

[@Eraknelo]

I'm by no means an expert, but from everything I've seen, the degradation of these type of cells is not linear. They drop most of their capacity in the first 4 - 5 years (around 10%), but then they take another 10 years or so to drop another 4 - 10%, so they may very well still hit their 80% target or even exceed it (in a positive manner). That being said, I'd imagine that if part of the cells are much more heavily bloating like in your top row, they might drag down the capacity of the entire pack heavily. Maybe you could even regain full pack capacity if you replace the worst cells, to prevent the BMS from triggering a low voltage cutoff on the worst cells. Could you maybe see which cells are the lowest voltage at low voltage protection levels?

[@SolarEngineering]

I haven't checked what cells heating the low-voltage disconnect (in video's screenshot it is cell#1 which is top cell in pack). Hopefully, you are right about not linear degradation.

[@Eraknelo]

@@SolarEngineering Ah yeah @4:03 , so then maybe I'm right about the pack being severly handicapped by the swollen cells reaching lower voltages much sooner? Looks like that #1 cell is MASSIVELY lower voltage than the rest of the pack (about 0.32v, which is a massive part of a LiFePo4 voltage curve). If you replace and re-balance a new cell, it could maybe reverse a large part of the degradation.

[@houseofancients]

I choose the middle ground between diy and premade. Use exclusively the seplos mason 280 diy kit. This enclosure compresses the cells, bms talks to my inverter.. All in all really happy with them, and more importantly, after 3,5 continues use, the is not capacity difference between my newest pack (3,5 years old) consisting of grade a lf304 cells, and my newest pack containing the same cells

[@houseofancients]

Btw, hate to say it out loud, but bsl doesn't sell automotive grade cells, only b grade.. Good news is though, that cell got bigger, and a lot cheaper... Plus there are now us sellers that carry them. 18650 battery Ezeal Battery hookup to name a few

[@SolarEngineering]

did you measure capacity on 3.5 yo?

[@houseofancients]

@@SolarEngineering yes I did.. every time I add a pack I do a capacity test before I add it..

[@matthewknight5641]

I built a 840 ah 24 volt battery. Its 3p8s. I use a jk BMS. Its only used when Im home a couple days a month. So far so good but it's only got 4 complete cycles.

[@garygowanoffgridsolar]

I bought 150 ah cells from bls battery in 2019. They were made by higee i think in late 2018, steady use down 9.5% capacity and 2 to 3 mil bloating on ever cell. They stay balanced at 3.5 per cell higher than that they spread out alot.

[@SolarEngineering]

thanks for the info

[@bwilson948]

The cells attached directly to the positive cable and the negative cable will take the full blunt of any charging or discharging putting you cables to the packs on the bottom would give you compression from the entire stack,. But keep in mind if you have a failure in charging our discharge those cells should blow first. In you current setup a pair of bolt cutters and fresh air the primary cells could cut and removed very quickly. Having two metal buckets 1/2 fill of sand handy may be useful one to put the fume cell into the other pour of it. You may want to think over move you +/- from the top or cells 1&16 to the bottom of the pack. And look at some large 100 VDC capacitors. The company that made you invertor and/or chare controls put in caps to protect them.

[@ricardomarcelino8388]

Hi. I am a LiFePO4 user and hobbyist. My idea is that not charging to full capacity (not 100%), is actually doing more harm than charging them all the way 3,45v and absorb there for an hour. It is my understanding that leaving the batteries at a lower state of charge (80%), is more likely to hit a low state of charge mire often. And my guess is that a being at lower voltage for longer periods of time is what kills LiFePO4 prematurely. Will Prowse has mentioned it recently, that all the batteries he left empty, were done in few weeks, whilst the ones he left fully charged had no degradation whatsoever (not in intensive use). If your batteries are going to degrade that fast, I would say it makes sense to fully charge them to 56V (3,5v per cell), absorb at 3,45V and float at 3,35V. It is still pretty conservative, and giving the opportunity for the pack to balance over 3,45V every day for a little time. I would definitely also consider some CALB or EVE cells, renowned providers. Also have to say I am jealous of that workshop and space... I live in an apartment and have no south facing space to fit panels... Nice short video. Cheers

[@SolarEngineering]

thank you for the info and kind feedback.

[@nik-lc3ob]

I suggest you do a proper top balance of all cells and then re-test capacity

[@SolarEngineering]

do you think that 33mV at 3.6 is a big imbalance?

[@nik-lc3ob]

No, not at all. I missed that part I guess@@SolarEngineering

[@USA-GreedyMenOfNoIntegrity]

2 yrs diy 230 kWh of LFP. No problems. I have separate Jk bms for every 280aH battery. You have too large of a capacity for 2A balancer and lower quality cells contributing to your issues.

[@dc1544]

Set charging voltage to 55.2V have balancer turn on at 3.45V I did this and my cells are coming way closer to each other. if you balance before 3.45 volts in a cell it causes more imbalances. Also split your pack into 4 packs. you are not doing .15C those cells are small and large amounts of amps is too much. example is a 25ah cell in a Battery pack still is 1C at 25 amps. Doesn't matter if you have 4p. with a 25ah cell 5 amps is 20% or .2C. Your cells will not last as long as you think since you are really doing 3C or 4C. I have 280ah cells I do not charge or discharge any pack (4 packs of14.3kw) over .2 which is 56amps. That is why I have 4 packs and soon will have 6 when I add more solar and another inverter so I stay low C rating. I have 1.2 years on these cells. Just now getting them top balanced.

[@SolarEngineering]

thanks for the advise

[@lexxreker]

Похоже взлулись из за конвекции горячего воздуха вверх. Стягивать бандажом по сути нужно . А123 пакеты на заводе, стягивали. Моей батареи 11 лет. Мощность снимаю небольшую 1.5 килловат. За время работы старение сказалось на время балансировки. При старости фосфат баланс становится более длительный. Старые батареи только заряжать используя более продвинутые алгоритмы.

[@rcboosted]

Thank you for the update video on capacity! I wonder if the difference in accuracy of the shunts used from 1st test to the Victron would account for some of those capacity differences?

[@SolarEngineering]

yeah, potentially it's possible

[@largepimping]

@@SolarEngineering That was exactly my thought - 5% drop in that short of a timespan sounds drastic, but diff's in measurement technique could explain a lot of that. I also have heard it claimed that they lose "a lot of capacity" in the first several cycles, but that loss over time flattens out. How many times did you cycle them initially? IOW, is that initial baseline even valid, irrespective of the differences in the shunts being used? Not really a criticism, just wondering...

[@SiBex_ovh]

Still the DIY storage with JK-PB2A16S-20P si 33% cheaper then buy china one who is cheaper from big brands. Compression with proper BMS & ActiveBalancer is a key, and every mo do a Balance.

[@RJ-cc1fz]

Love these type of videos. Looking forward to the battery rebuild if you decide to do it. However the cells will never again be flat? Very strange how your cells stay so far out of balance. I have had my smaller (17kwh) bank for over a year. My CATL cells stay very well balanced till the cells get to 3.5v per cell, then they start to drift a bit. But the jk handles that easily. Unfortunately I’m thinking you got bad b grade cells to begin with. If you decide to get rid of them keep me in mind

[@SolarEngineering]

thanks. I think just going to leave battery alone, will see if bloating will continue, then will replace bad cells

[@techmaster_sree]

I had an issue with balancing that is solved using additional neey active balancer. I think it will be much better if you are using compression on cells. Top row cells not getting any compression and this cell orenation not good so chances of heigher degradation . I think vertical arrangement with proper compression will be good for longer cells life Thanks for sharing your experience.

[@SolarEngineering]

I tried to add another active balancer, but 2 balancers were balancing each other sometimes :) so removed one.

[@techmaster_sree]

@@SolarEngineering in my case in built balancer is only 200ma (BMS) and in neey Bluetooth balancer I can configure parameters and it's 4A

[@dc1544]

Also any connection that shows more than .02 resistance difference should have connector redone. look at the cell that had .102 while others had .075

[@arebear4797]

I have concern wit how you arrange the cell bottom up. wit that said, my concern is that the bottom cell will have a different pressure compare to the top. That probably efect the fasyer degradation of your cell overtime. Did you check single cell capcacity between you bottom cell and your top cell?

[@SolarEngineering]

no, haven't run test on individual cells

[@andrewrobb633]

Sorry if this is a dumb question. So you only need to balance cells in series not parallel? For example, I could parallel 4 of your power walls and not balance them other than what is done in the individual power walls?

[@SolarEngineering]

if you connect 4 PW in parallel than it's going to balance voltage across PWs, but not cells in each PW. Still need some kind of balancer in each PW (either active or passive)

[@babylonfive]

Thanks for the information. I'd point out that the way the upper batteries don't 'line up' is not an indication of swelling of upper cells only, but simply some cells in the stack. You won't know which cells are swollen until you take it apart, and also the differences are an AVERAGE of the likely swelling, so there could be swelling in the lower stack AND the higher stack together, showing more swelling than you might originally think. Just MHO.

[@SolarEngineering]

visually it's only top cells, but you might be right that other cells are swelling as well (most likely they are)

[@tau9632]

What was the new BMS that you put in? I couldnt quite hear

[@SolarEngineering]

JK Smart Active Balance BMS

[@tau9632]

@@SolarEngineering Thanks mate!

[@termikina]

Las celdas LiFePO4 prismáticas no se pueden instalar en posición vertical, la posición horizontal provoca mucha degradación, el número de ciclos puede llegar a ser hasta 400% inferior a la posición vertical que es la correcta. Quiero añadir también que el balance activo ha de activarse solo cuando se llegue a 3.45v, por debajo de ese voltaje no es efectivo.

[@TylerTheCompiler]

Great vid as always thank you. How about a video on Growatt TL MIN-XH-US installed to grid??? Nobody shows Growatt installs and they are very reasonably priced inverter.

[@SolarEngineering]

thanks, sorry not planning to do growatt reviews yet

[@Tumbleweed5150]

I am not sure about the model you mentioned, but David Poz does a few videos on the Growatt SPF5000ES. After watching his videos, that is the model I decided to go with, even though I had to get a Mid-Point Transformer in order to get 120V, as they put out 240V. I still feel that they are one of the best inverters for the price. I've seen a few, including the EG4 6000XP Off-Grid Inverter, (Which is also sold by Signature Solar), but I wasn't able to afford two of those at $1,399 each, while I was able to get the 5000ES at $900 each , or $2,258 for two plus the Transformer. They are now lower priced at $746 each, (plus, of course, the cost of the Tranformer). On the other hand, DIY Solar with Will Prowse has totally gotten away from Growatts and gone with the EG4 brand.

[@TylerTheCompiler]

@@Tumbleweed5150 I really like eg4 6000 but you need two and at least one battery. That’s three times the price of Growatt.

[@URackADisciprine]

Are you licensed to work in Nevada...I live in Las Vegas? I would like to get a solar system installed but nobody will install the EG4 18Kpv here since it isn't Enphase or SolarEdge. I found one company that will install a Sol-Ark 15K but only if they supply it and they want to charge a fortune for the unit even though the prices have come down a lot since the release of the EG4. I do have a great supplier (CSE Solar) here in town so I could get anything and have it on site.

[@SolarEngineering]

no, only in CA

[@URackADisciprine]

@@SolarEngineering OK thanks, its a shame I cannot find anyone here that will do it.

[@gg-gn3re]

too late to clamp, as said by manufacturers this is for the initial charging process when they're gassing. The expanding of batteries doesn't really matter

[@blackbumble9988]

Hey SE, your panels you put on your roof 10 months ago. Shouldn't you have a metal screen around the perimeter of the panel to prevent small animals birds etc from making nests under.??

[@SolarEngineering]

installed about 2 years back, I don't have issue with birds/squirrels. but installing metal mesh for some customers if trees too close to the roof.

[@Electronzap]

Nice tour of your system. I don't think I have any good tips to give, sounds like you have a good grasp on things. As time goes by, better options get cheaper, so there's no use comparing an older system to newer ones.

[@joshkelly3743]

It is to late to compress the cells. The damage is done

[@SolarEngineering]

yeah, unfortunately

[@slydog7131]

I just bought my first battery, so I don't have experience, but I've been learning a lot. This is what I have gathered: There is a difference between an "A" grade cell and an EV grade cell, an EV grade being suitable for use in Electric cars and which must pass the most stringent criteria specified by the Chinese government. Grade A cells are those which fail to pass EV cell tests and are sold for less. They have no government standard and it is up to the seller to decide what to call them. Were your cells identified as EV grade and did they come with the original spec sheet? If they were EV grade then they should have. The spec sheets are often faked. If you got a spec sheet that was all black and white, it is fake. It should be colored. At least this is what I have learned (if it is indeed correct.) Non-EV grade cells will have lesser life or performance characteristics. Feel free to set me straight if I'm wrong about this.

[@MrGarethrn]

Great video mate and hello from the UK. I'm not sure if you've seen the device linked here: kzbin.info/www/bejne/rabQaIVuic6kgMU but it's an interface that will allow your JK BMS to communicate with your inverter (depending on protocol). Hopefully it will help. Keep up the good work!

[@SolarEngineering]

thanks for the link, I placed an order. Will see how it works.

[@dc1544]

your cells are out of balance is why you are loosing part of the 5%. I see now you have 16s3p I do mine at 16s1p and have 4 of them. sure I had to buy 4 JK BMS. But after 1 year I am getting a bit more. I never top balanced all my cells. I got 278ah average from all my cells and now I am getting 281 average. I did what Off Grid Garage said to do. he has not lost anything after 2 years.

[@mlg779]

Did you just deleted my comment? Or you tube did it?

[@SolarEngineering]

nope, I'm not removing any comments

[@mlg779]

@@SolarEngineering probably youtube

[@stream5967]

Зайшов подивитись станцію, і чую в твоїй англійській українську мову. Ти з України?

[@SolarEngineering]

из Беларуси

[@stream5967]

@@SolarEngineering Супер.

[@stream5967]

@@SolarEngineering Я декілька років тому для однієї з компаній трошки в USA пробував проекти для сонячних станції робити.