Tweet
Did you balance the cells individually before constructing the packs? Find any bad cells?
-
-
Yes, I balanced every module perfectly with a BC168 (6 channel, so 3 modules at a time) to 4.05V before constructing the packs. If not there would have been sparks flying
I haveb't found any bad cells. The BMS reports very similar internal resistance for each 10P module and the voltages track very nicely. Before balancing the cells also had nearly identical voltages (but two different levels since the modules come from two different Leaf packs).Comment
-
Warning to newcomers
Don't get your chemistries confused or you will destroy your cells and run in an unsafe environment.
LiNMC (lithium manganese) such as seen in some EV products, RECENT powertools, flashlights, and the like are designed for HIGH CURRENT capability, (regen and so forth) and typically are designed for no more than 4.2v max charge, just like their laptop cousins.
These differ from large prismatic GBS lifepo4 cells, which have a *dash* of manganese in them. Their chemical structure is predominantly Lithium-Iron-Phosphate, but this *dash* of manganese makes them LiFeMnPo4 chemically. They are rated at most for about 3C temporary bursts, whereas the linmc cells are purposely constructed for much higher rates, much of which would be a waste in a solar house power project of any reasonable size.
That means that you do NOT charge GBS cells beyond 3.6v like you normally would with the other types like laptop / flashlight cells.
Ie, the major players in the lifepo4 arena that charge to no more than 3.6v, are GBS, CALB, Winston and a few others. They are all predominantly lifepo4.
Can you make a house-power bank from other lithium chemistries - sure - but you are wasting money and energy (unless you are just cheaping out for some reason) on cells not designed for a house-power "Sub-C" application! Kind of like putting a Chrysler Hemi into a Yugo. Fun, but not the norm.
For example:
If you did this to GBS cells (which admittedly have a *dash* of mangenese in them), they'd be toast.Yes, I balanced every module perfectly with a BC168 (6 channel, so 3 modules at a time) to 4.05V before constructing the packs. If not there would have been sparks flying
PLEASE keep chemistries straight here or someone may get hurt.Comment
-
If those are the true charging specs, then they are NOT LiNmC, but actually lifepo4 with a *dash* of manganese added much like GBS uses. LiFeMNpo4. Still lithium-iron-phosphate at heart, and follows the charging regimen for lifepo4 which means no more than 3.6v typically.
Watch out - many sharks out there to hoodwink the duct-tape-a-pack together crowd with gray market / EV crash / failed warantee reject cells.Comment
-
And do not forget to make a conscious decision as to whether to bottom balance (best for safety) or top balance the packs. Top balance would be more important if you intended to charge the bank close to 100% SOC, which is NOT recommended for longest battery life.SunnyBoy 3000 US, 18 BP Solar 175B panels.Comment
-
Except when yuu can get really cheap cells from wrecked cars. Recycling at it's best and it's nice to see a 53V lithium bank only sag half a volt or so under 6kW load
And since I paid only a little over $100/kWh I doubt you can get LiFePO4 any cheaper.
Of course. The Leaf cells are refular lithium cells with a max voltage of 4.2V. This is the LiMn thread after all?If you did this to GBS cells (which admittedly have a *dash* of mangenese in them), they'd be toast.
PLEASE keep chemistries straight here or someone may get hurt.
Info is readily available on the AESC modules that is used in the Leaf. I know they are popular for building replacement batteries for some electric scooters. They are excellent quality cells that usually never need much balancing, and they are very easy to mount together and build whatever capacity battery one need.Comment
-
Yeah, mine is charged to nowhere near 100%, I stop at approx 80%. 80% is also the recommended long-life charging mode in the Leaf. So top balancing these is out of the question. Better to balance at all times, ie. middle balacing. The AESC modules in the Leaf track each other very closely voltage wise and seems to respond well to this. And since neither the top 20% or bottom 7-10% is used perfect balancing isn't very important either.
Here's a snapshot of the cell voltages taken now: Skjermbilde.JPGComment
-
I guess it is now. It was just trying to protect tasman from mixing and matching different chemistries.
Certainly recycled leaf cells can be used if you know what you are doing. Recycled or not though, the high-current capability of the cell is going to waste for a relatively low-current solar housebank (other than a one-off diy project). Like all battery projects, unless one knows what they are doing, stay away from used batteries like used toothbrushes. Details make the difference.Does any one have any information regarding lithiun maganese battery cells. I'm using lifepo4, was about to order some more when the supplier offered me great prices on lithium maganese cells, half the price of lifepo4. Had a bit of a look around about them and they seem to be much better the lifepo4, other than they don't have the same energy density.
For instance, your leaf cells from Norway will differ a bit from those who came from crashes in Arizona: (leaf cells are discussed among others - one of my fav videos)
This is not unlike EV forums where the main emphasis was on using large prismatic lifepo4, and suddenly is barraged by guys wanting to power their vehicles with used laptop-pulls. Or wheelchair users thinking that large prismatics are the way to go, when in fact higher-current small cylindrical headway lifepo4 cells (typically 10ah each) are really are what is needed with their huge motor current surge demands. Ideally one fits the battery to the application as closely as possible..
Dunno - maybe to avoid confusion a separate LEAF battery thread could be started with a possible warning or two about their charging differences from lifepo4? Just trying to keep the neophyte from confusing chemistries and making big charging errors ...Comment
-
Yeah, the high current capability is not needed. However it is not a drawback either, since the cells have such low internal resistance the charging efficiency is fantastic. I've calculated it to 99.8%, compared to ~94% on my old FLA bank (12V from 3xRolls 4KS25PS).
Also the modules are too big for applications like wheelchairs or ebikes, so they are not very sough after and thus pretty easy to aquire cheaply. They only really work for EVs and stationary storage like solar. To get a 48V (well, really 52.5V nominal battery for a fast ebike one needs to connect 7 modules in series, which will weigh about 55 pounds. Way too heavy and way to high capacity at 66Ah.
Yeah, one should definately be aware of the heat problems with these modules. That said, an Arizona Leaf battery degraded to 70% of original capacity should still be very usable if it's cheap enough. Just add some more modules for the same total capacity.For instance, your leaf cells from Norway will differ a bit from those who came from crashes in Arizona: (leaf cells are discussed among others - one of my fav videos)
I agree, C-rate must be high enought for the application. I do not however see a problem with high C-rate batteries in a low-C environment as long as they have enough energy density for the application and is reasonably priced. For a 35kWH off-grid battery like mine which will never see a load above 0.2C any battery chemistry will do. So it's better to focus on the other parameters like weight, size, cost, efficiency, durability, maintenance need etc. Here the used Leaf-modules excels in most categories, for solar use.This is not unlike EV forums where the main emphasis was on using large prismatic lifepo4, and suddenly is barraged by guys wanting to power their vehicles with used laptop-pulls. Or wheelchair users thinking that large prismatics are the way to go, when in fact higher-current small cylindrical headway lifepo4 cells (typically 10ah each) are really are what is needed with their huge motor current surge demands. Ideally one fits the battery to the application as closely as possible..Comment
-
I don't get the criticism. He said what it cost, and it should work for a number of years. FLA is hardly trouble free and less expensive. Purpose made li-on battery packs, like the powerwall, still look pricy for the next several years.
Especially in Norwegian with its vast EV %, this seems like a great DIY storage solution.Comment
-
I can agree with that. The new 35kWh lithium pack cost the same as my old FLA pack consisting of 3 Rolls 4KS25PS. That is 12V 1900Ah, or around 10kW usable when only using it down to 50% SOC. The new pack has nearly half the weight of the old one and has 2,5x the usable energy for the same cost (well, nearly, I had to buy a $500 BMS too). Also 10 module stacks of Leaf cells weighs 35kg and is very easy to transport and move compared to a 152kg Rolls battery...Comment
-
I wouldn't call what PNjunction posted as criticism. He is trying to let others, (that are non versed in battery technology) that any type of Lithium battery needs to be treated a little different then a standard FLA type which is what most people have hands on experience and are familiar with.
Getting good data on battery banks from those that understand it like you and a few others is very valuable info.
You just have to understand the number of DIY people that come in here can get themselves hurt because they; first believe everything they see on Youtube and second, think they can "safely" build their own system yet they have ZERO knowledge of electrical systems or batteries.
It is better to be a little cautious when providing details of a specialized system likes yours and let others know it could be more difficult then baking a pound cake.
I too feel that the future is a battery chemistry other than FLA. Yet I have not seen a cost effective manufactured system yet being offered. So until then FLA still falls as my first choice for both economics and IMO safety.Comment
-
Yeah - no criticism.
I'm just saying that unless you know *exactly* what you are getting into, recommending used batteries regardless of the chemistry, is starting out newcomers on the wrong foot since you don't really know what kind of life it has led before.
This is what we don't want from a safety standpoint - guys using can openers and machetes to extract used leaf cells:
Or this dude with cells scattered, duct-tape, and a bucket or two of chicken:
The other issue is that with lithium, capacity measurements are not an actual indicator of health. Far from it. "Sudden Death" syndrome is where one day you have full capacity, and the next day the cell is dead without any apparent discharge. This is mainly from oxidation holding the voltage too high, or even milder voltages too long, and of course high heat. The SEI layer finally just clogs not allowing any intercolation to happen, although just hours before everything was hunky dory. There is plenty of capacity left, but you just can reach it because the sei layer is now closed due to prior abuse. Unlike other battery chemistries, you usually don't get any real warning flagged by reduced performance.
It might be a discussion for another thread about top balancing and the problem of going bananas over matching cell voltages which take too long, and keep the cells at elevated voltages too long each cycle ... oxidation then sudden death sei closure.. which some EV'ers may not even be thinking about - oxidation time.
I just think a dedicated leaf battery thread would be more appropriate, and less hijacking - even from myself would keep it on track. Guys that know what they are doing with used cells and prepared to accept the consequences is one thing - but promoting used cells, regardless of chemistry sets off red flags from 3rd party resellers.Comment
-
You guys are probably pretty sick of me by now....
That is true, but there is a safety factor in play when choosing energy density for your application and needs some reviewing - which is not fear mongering.
Out of all the lithium chemistries out there, lifepo4 is the least energy dense and from a safety standpoint, the safest. The key to this is the FePo4 or iron-phosphate, which is a greedy material which does not like to give up oxygen atoms even when abused. However, ALL other lithium chemistries achieve higher energy density without the greedy iron phosphate and you had better have your act together - which you guys seem to do. Naturally, lifepo4 being on the lowest end of the density scale is the largest physically - but still about 2/3 the size of lead.
The difference is venting vs. venting with flame.
No chemistry likes to be abused, but consider the lifepo4 cell with say an upper limit of 3.8v before damage starts to occur. Going higher than this will harm the cell, but it takes about 30v to go catastrophic. But yes, severe damage has happened.
ALL OTHER lithium chemistries have very slim margins for abuse. Take an LiNMC cell which charges to an upper limit of 4.2v. What happens when you charge to 4.5v accidentally ? Event.
Proper and precise battery management techniques mitigate this problem. But the diy guy may not get so precise. Lifepo4 gives them that headroom for error from catastrophic events, even though their entire battery investment may be gone when going beyond the norm.
So in a fixed installation where you can actually afford the space, why not choose the safest lithium option - lifepo4 the least energy dense material. (And no, Boeing did not use lifepo4).Comment
-
That is true. There should be redundant safety systems when dealing with litium batteries of any kind. Also for protectiong one's investment. For my bank the first "line of defense" is the Outback FM80 which is set to charge the bank to 57V and then float. 57V/14S is 4.07V per cell. Way below critical voltage, and such banks should never be charged to 4.2V anyway because the liftetime will be dramatically lower.
The second line of defense is the BMS, which really should be present in ALL stationary storage lithium battery banks. Mine is set to allow a maximum charging voltage of 4.15V for any cell. If the FM80 breaks and continues to charge above that, the BMS will open the charger safety relay which then disconnects the PV array from the FM80.
The BMS also monitors battery temperature and will prohibit charging at high currents when it is too cold and also disconnects loads if a cell voltage dips too low (currently set at 3.4V)
That said, FLA can also have "events" if charged too high. Excessive gassing and a resulting explosion can happen. Car batteries exploding is not unheard of.So a FLA bank should have adequate ventilaton, which is not necessary with a lithium banks. Different safety measures for different types of batteries.Comment
Comment