Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

drraptor
drraptor Solar Expert Posts: 218 ✭✭
Is anyone using Lithium iron phosphate battery (LiFePO 4) batteries in Solar or Wind applications ? Varta does produces Battery banks for Solar applications http://varta-storage.de/en/downloads.html
«13

Comments

  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    drraptor wrote: »
    Is anyone using Lithium iron phosphate battery (LiFePO 4) batteries in Solar or Wind applications ? Varta does produces Battery banks for Solar applications http://varta-storage.de/en/downloads.html
    Some people are using them, but not many. At this point the cost per kWh is still high enough that without space or weight limitations on the design it just does not make sense for a stationary system.
    That may well change.....
    SMA SB 3000, old BP panels.
  • drraptor
    drraptor Solar Expert Posts: 218 ✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    do these have any other advantages beside occupying less space and weighing less? Varta list it's life as 8000 cycles at RT which is 10x times that of a FLA. http://varta-storage.de/fileadmin/media/files/downloads/140512_technisches_Datenblatt_Home_EN.pdf
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    One of the concerns with various other chemistries has been cell to cell balance (equalization) and preventing over discharge (one over discharge will "kill" many of the Li-ion chemistry rechargeable batteries). That usually means some sort of battery management system (adds cost/complexity--Although, some vendors cell their Li batteries without BMS saying it is not needed--their cells remain in balance).

    It is complex subject (ask Boeing about their 777 batteries)--We have a poster or so here that believe that LiFePO4 batteries for off grid use are competitive (if not better) now--because of their longer life and ability to surge/charge quickly (as I recall).

    Anyway, that is my (probably wrong) 50,000 foot view--I hope some of the people that like LiFePO4 will come by.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • drraptor
    drraptor Solar Expert Posts: 218 ✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    According to Wikipedia, Boeing used Lithium cobalt oxide http://en.wikipedia.org/wiki/Boeing_787_Dreamliner#Battery problems; while "LiFePO 4" is more recent technology.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    I'm doing that right now to a 4S set of GBS LiFMnP04's with my Xantrex pwm controller and Sharp 80w panel. There are a few vendors like Elite Power Solutions, Batteryspace, etc to obtain them.

    The scoop is that prices are coming down to near what you'd pay for quality AGM's, and given the added cycle life, they may be nearing par. Still, an upfront cost to consider, but not as bad as it was just a few years ago.

    Lifepo4 (and it's near cousins), come in high-energy vs high-power versions. Ie, there is a difference between an SLI based lifepo4 like a powersports Shorai and Antigravity, and then there are those that are more suited to house banks, like the GBS cells. CALB is also another good choice for solar house banks.

    Since we are not doing electric-vehicle duty, most do a top balance, or rely upon the quality of the manufacturer to get them close prior to shipment. Once balanced, they tend to STAY balanced, unless you do something weird to them or damage them with a total discharge that an LVC didn't catch.

    Top balancing can be achieved in a variety of ways. You can charge the cells individually with a bench power supply that has adjustable voltage and current. Typically this is about 3.5v. When you reach C/20 current, you are done. Move to the next cell. If you don't know what you are doing, get the pack pre-assembled. Those that do know what they are doing will typically discharge each cell to a very low value (but not low enough to kill it), put them in parallel, wait a few days, and charge them all up to 3.5v and then reassemble in series and check balance again.

    Because cells that vary in different initial balance levels can produce large amounts of current when you assemble them, it is best to either get the pack pre-assembled, or again, know what you are doing.

    Or, you can use shunt-resistor boards that the vendors supply that will do this for you. However, that is another possible point of failure. You can either leave them on (and hope they don't fail) or do it once, and take them off. These passive shunt boards consume a little power of their own, and if left on a bank in storage too long, can drain them. Or, in some cases they can make balance worse! Try to use quality components.

    Or, you can enlist the aid of a hobby charger, like Revolectrix CellPro8 and balance-charge each cell individually and simultaneously like a huge RC model plane setup.

    For small scale use, I find a 5A Tecmate-Optimate TM291 lithium (lifepo4 only) charger useful. While it doesn't do individual cell-balancing per se, it does balance by stopping and starting the charge process near the end. This way, when one cell is fully charged, and stops the current from flowing to the other cells, during the ramp up time when the charge is starting again, the lower cells catch a little more current than they did before the full cell stops it again. This is repeated many many times. It might take several iterations of charging to get the cells to migrate closer to each other. Very useful on simple 4S packs, or even on powersports lifepo4s. Antigravity uses them. There is more to this charger than I want to cover here, but at least this is a simple way to do it.

    For solar use, top-balance is the way to go. There are plenty of threads on this, but those found here are especially helpful:
    http://www.cruisersforum.com/forums/f14/lifepo4-batteries-discussion-thread-for-those-using-them-as-house-banks-65069.html

    It seems like a lot, but not any more difficult than taking care of a flooded battery properly. Less so once you get the balance right (or verify from the vendor), and use a bit of monitoring. When setting voltages with lifepo4, you'll want good dc accuracy, so I'd recommend something like a Fluke 87V as a *minimum*. Cell to cell balance can be checked with a good voltmeter, but for setting precise voltages, you'll need the accuracy of a higher-quality meter.

    I totally cringe when I see video of guys with $Kilobuck lifepo4 banks and a $10 meter. Egads man!

    You'll also see discussions on using an HVC and LVC to protect from over-charge and over-discharge. The absorb set point on my controller is the HVC. The LVC for me is good monitoring, equipment that shuts itself off after an hour of disuse etc. For some, it makes sense to set the LVC on the weakest cell, instead of measuring pack voltage.

    Lifepo4 is a battery chemistry you can actually SIT on, as under your motorcycle seat. I have no problem with that. In fact, the GBS cells, which have manganese added, *may* even be safer, because they have slightly less energy density than pure Lifepo4.

    All of this is predicated upon just doing things right. You can abuse a flooded cell and throw battery-acid and hydrogen everywhere, but that is passe' these days. Doesn't make a good dramatic video, so not much coverage of that.

    What makes it seem worse than it is, is posts like this one. Once you wrap your head around a few simple rules, you'll probably ask yourself why you didn't make the move sooner. :)
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    So, as is always the case with early adoption of new tech, we have a deficit of information. The old school say the new tech is too expensive, the new school say its actually cheaper. So can we have some numbers? Im willing to crunch them, but i need current actual retail prices for LiFePo4. If the BMS is essential to achieve the cycle life, then i need those costs as well. Cycle depth and rated cycled life. Guarantees. Doesnt matter what their AH rating is so long as they are off grid scale.

    I need real numbers, what does it actually cost people who have done it. If theyre imported, as they appear to mostly be, i need frieght, duty/customs charges etc. Delivered to the door costs.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    Here is the latest quote I got on a small bank. I'm pretty sure it works out to less $ than lead acid on a usable watt hour and cycles basis, even more so in higher temperatures (my application). Lithium BMSs are becoming more available (I ordered the one on kickstarter). On the other hand, there is certainly less track record and who knows how close they will perform to specs; call it an experiment.

    item :lifepo4 A123 20ah battery cell
    FoB unit price is US$15.4
    10Pcs is US$154
    Shipping fee:US$94.2 (delivery term by DHL)
    Total amount :US$248.2

    I am available for custom hardware/firmware development

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    zoneblue wrote: »
    So, as is always the case with early adoption of new tech, we have a deficit of information.

    That link to the Cruisersforum has about 4 years worth of info still valid today, so check that out. Plenty of happy campers from about 5 years ago , and the thread is STILL active. Some of those members may even be active here.

    Spec-sheets for Lifepo4 are available at the vendor or retailers sites. If you want to start small, Batteryspace has plenty of docs, as well as Elite Power Solutions. Big time users that want plug-n-play - see Genasun.

    The info is out there, but yes, there is a lot of distracting FUD too. Hopefully my post isn't one of them. :)
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    jonr wrote: »
    Lithium BMSs are becoming more available (I ordered the one on kickstarter).
    Not my first choice.

    The whole BMS / non-bms issue can eat up your entire life and has produced more endless threads (usually from an EV perspective of bottom-balancing), locked threads, and forum drama.

    Remember, we are NOT like the typical EV user, so our needs are different. Simply put, make sure your cells are balanced. Then stay out of the severe charge / discharge knees. That means running from about 14.1v to no less than about 12.5v for a nominal 12v battery. Since there is no real power in the high charge knee, there is no reason to go there, except for possibly an initial balance. Employ an LVC or just be sure to do good monitoring to avoid going below 12.5v. Below that there is very little power, and the discharge knee is steep.

    Since we are doing very low-C charge and discharge rates (unlike EV'ers who hammer the pack) , the use of a bms *permanantly* on our banks adds additional points of failure that one may not desire. I certainly don't. A revisit now and then is prudent, especially if you are using off-brands. Major players like GBS and CALB are pretty popular. High end yachts may prefer Mastervolt or Genasun.

    It really is that simple!
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    zoneblue wrote: »
    I need real numbers, what does it actually cost people who have done it. If theyre imported, as they appear to mostly be, i need frieght, duty/customs charges etc. Delivered to the door costs.

    How about these guys:
    http://www.ev-power.com.au/-Sky-Energy-Batteries-.html

    They should be able to get you down to the penny. :) Just remember that even though they come from an EV source, our requirements may be relaxed because we doing very low-C duty. How far you want to take that is up to you. Impressive is that these guys openly discuss what happens when you totally abuse batteries, and that fear-factor could lose sales. I suppose it is CYA, but shows responsibility - or to avoid possible litigation like any business would.

    Funny, you don't see that with any of the Pb battery vendors. Shots of battery acid all over the place, hydrogen igniting and so forth just don't amuse the public anymore. At least all the Pb chargers have about 5 pages or warnings that nobody reads. But this NEW stuff - now that's drama!

    Lurkers - I'm being facetious and not advocating being stupid - stay safe with ALL batteries.

    On that site you'll also find the users-manual for the CALB cells in pdf form. Highly useful.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    Cost isn't everything.

    Perhaps I can put this in a non-hyped fashion as to why I'm so enamored with Lifepo4 for solar, without it being a bullet-point brochure of buzzwords. Quite frankly, I'm VERY happy with my properly sized pure-lead AGM's in solar service. So what else is there?

    Cost - debated to death. Prices coming down, yet still an upfront issue for many. Look into this from time to time if cost is the only issue. Remember to compare a lead acid at 50% DOD vs Lifepo4 at 80% DOD. If you haven't, then they are immediately 30% cheaper than before.

    Safety - Lifepo4 is a specific chemistry of the Li-ion family, the safest so far. Yes, you can abuse them like any battery, but they are NOT the same chemistry as the unstable Lico02. The tradeoff is energy-density, so Lifepo4 is bigger. One reason you don't find lifepo4 in cellphones. The lower the energy density, the safer it is usually. This is also the reason you find them under the seats of motorcycles and powersports vehicles.

    Charge Acceptance Rate: Because the internal resistance is even lower than my beloved pure-lead agm's, they charge even faster. More charge, less heat. This makes it ideal for those in low solar-insolation areas where time to recharge is the major enemy.

    Cycle Life - about twice or more than Pb IF you take care of them. Usually one of the first things hyped in brochures. If this was the only advantage, I'd pass. I'm quite content to just buy an AGM twice if need be. But there's more...

    High current charge - these cells like the GBS and CALB's can handle about 3C safely. I seriously doubt any solar enthusiast is going to hit the wall here, except for very small capacity batteries. That could be reached by using a 360 watts nominal 12v panel system to the 20ah lifepo4 battery which is the physical size of about a 12ah agm. :) I would personally cut it down to no more than about 1C or even .5C. Thing is, even at these "low" rates, if you have the panel wattage, you can get your charge done much quicker. Again, those in low solar insolation areas will benefit greatly.

    PSOC - partial state of charge. This is huge. Not only are you NOT required to charge to 100% capacity, most urge you not to do so! There is very little power from 90-100% SOC. Easier on the cells, especially if you are a bit unbalanced. Like me. :) Thus short-cycling is no problem. There is no sulfation to worry about, no floating, no temperature-compensation, and no EQ.

    Tip: While most spec sheets show strict upper and lower voltage ranges, it does NOT mean it is a good idea, even if you can do it without suffering immediate damage. For example, over the years most of the big distributors and manufacturers have been dropping the recommended voltage used for normal use. Yet you may still see old recommendations on the spec sheets. Example - since there is really no power in the steep charge knee, the best voltage to set your controller to is about 3.5v to 3.525v per cell. That would be about 14 to 14.1v on a nominal 12v charge controller for the typical 4-cell batt. You might even luck out with a preset for "gel" or "sealed" on controllers initially designed for Pb/lead. Be sure to check the voltage, and not just trust the silkscreen label!

    Storage - no need to store fully charged. In fact, it is not good to do so. Long term storage is ideal at around 50% SOC, plus or minus. Even here, you don't have to go nuts with exacting percentages like the RC modelers do. Just bringing the SOC down a bit from the top will take you a long way. From a practical standpoint, if you have a solar powered cabin that you visit on the weekends, and at the end of the day before you leave you find that you are not fully charged - even better! Just leave. (Provided of course that you don't have parasitic loads going on, or perhaps a low quality bms or shunting bridge that goes haywire while you are gone -- I've covered my position on that before...)

    HVC and LVC - Low and high voltage cutoffs to prevent damage are not a new requirement. Any SERIOUS Pb/lead installation has these. Yawn.

    Weight and size - about half that of lead or more. For me, that wasn't a big issue since my batteries are pretty much stationary. Still, the lower weight is great for portable use, OR for the average guy to put in a much larger house bank on his own without the need for forklifts and such. I'm getting to the point where an 8D in the basement would end up staying there until the end of time.

    Waiting - not really a pro here. Waiting for the major battery manufacturer's to jump on board to "validate" the technology may mean that you will wait forever. Conspiracies abound as to the reason why, but it doesn't matter. For me, this 15-year old technology has had enough time. I'm going to get on with life and enjoy it, even if the major manufacturers don't. I've lost interest as to why, so it doesn't matter any more.

    These are just a few reasons beyond the upfront cost issue that makes my AGM's jealous. I DIG those, but these Lifepo4's do have some qualities that justified it for me. In those cases where Lifepo4 just doesn't make sense for some reason, I'm more than happy to use pure-lead AGM.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    jonr wrote: »
    On the other hand, there is certainly less track record and who knows how close they will perform to specs; call it an experiment.
    item :lifepo4 A123 20ah battery cell
    FoB unit price is US$15.4
    10Pcs is US$154
    Shipping fee:US$94.2 (delivery term by DHL)
    Total amount :US$248.2

    STOP! There is NO way that 10 pieces of 20ah pouch cells from A123 would cost so little. Not to mention that there is only ONE authorized dealer that will sell A123 to the general public. And that is: buya123batteries dot com

    What you may in fact be seeing is something that is dogging the industry. There are MANY counterfeiters, along with resellers of used A123's pawning them off as new. See any tabs missing or cut in half? Zipper fuses actually open? Most end up blowing up the back end of homebrew E-bikes with these. They can even end up in seemingly legit products on the outside, but the unfortunate corporate small-time purchaser got hoodwinked.

    Then of course the first thing that appears is the dramatic video of it catching fire on the back of the e-bike, and the op warning people about the massive dangers of lifepo4, or bashing A123. Of course in the clouds of smoke, you don't get to see the duct-tape that held it all together, crappy wiring and soldering, and on and on.

    I'm glad you brought this up. If you don't want to be disappointed at the very outset, or endanger your personal safety, DO NOT GO CHEAP. There is NO "bargain" to be had. Buy from a reputable dealer from whom you can trust. Junk-sellers are wringing their hands waiting for the unknowing to open their wallet.

    Related - same thing happened to Optima batteries years back, and to some extent even today. Unauthorized low-lifes like to pretty-up used Optimas, give them a zap, and pawn them off on the unwary consumer who bought at a ridiculously low price. First thing - hit the forums and bash Optima.

    Seriously - Don't buy your lifepo4 batteries from the online equivalent of a guy in a van with some batteries in the back.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    PNjunction wrote: »
    STOP! There is NO way that 10 pieces of 20ah pouch cells from A123 would cost so little. Not to mention that there is only ONE authorized dealer that will sell A123 to the general public. And that is: buya123batteries dot com
    Current price for a carton of ten AMP20 pouches: $700, and currently out of stock. But you get a real deal on a 20-pack instead for only $1400. :)
    SMA SB 3000, old BP panels.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    Heh, total rip, AND you get the pleasure of putting those pouches together yourself.

    How about this: 20ah GBS 4S from Batteryspace. No balance boards. Did it myself easily enough with the Optimate lithium charger. Roughly equiv to a 38ah Pb battery (assuming we don't want to take the Pb past 50% DOD) since you can REALLY take lifepo4 down to 80% DOD at rated cycle life. Mine preassembled and dated 12/2013. No new-old stock. $124

    38ah pure-lead agm, like an Optima D51 yellow-top at Amazon. $165. Actually, the 20ah GBS lifepo4 has a little more capacity. Tested on a West-Mountain Radio battery analyzer.

    Now we're getting somewhere and one can see the gap closing..

    Makes me wonder how many people are comparing capacity prices straight across, not realizing the typical 50% DOD limitation for Pb, and the 80% DOD for LifePo4, which makes it a bit cheaper than originally thought.
  • drraptor
    drraptor Solar Expert Posts: 218 ✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    Impression have got from all this is
    • Lithium iron phosphate battery are prone to failure, if you handle them wrongly evenly slightly, while FLA can sustain a lot of abuse (personal experience)
    • Lithium iron phosphate battery are cheaper than FLA if you compare "KWH" to "KWH".
    • LiFePO 4 don't mind sitting on SOC >50%.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    To be fair--Take either battery below 20% state of charge, even a Lead Acid battery will probably be permanently impaired. Over charge/expose the plates to air in a lead acid battery--will be killed too.

    When operated within their limitations--Either will work fine. Operate outside those limitations--pretty much any rechargeable battery chemistry battery will be killed.

    Lead Acid batteries tend to be cheaper--So killing a bank or two of those in your first off grid system learning experience is less damage to your savings account (why we sometimes call those first 1/2 sets of batteries "training batteries").

    Flooded cell lead acid batteries do have one advantage--You can measure their capacity with a hydrometer relatively accurately. But, measuring specific gravity with a hydrometer is a pain (especially if measuring dozens of cells once a month or so to check balance).

    -Bil
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • drraptor
    drraptor Solar Expert Posts: 218 ✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    Flooded cell lead acid batteries can also be overcharged for 1-2 times, before you realize charging setting needs some adjustments. At least with translucent plastic covers we can check the water level without opening the battery caps
    Attachment not found.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    drraptor wrote: »
    [*]Lithium iron phosphate battery are prone to failure, if you handle them wrongly evenly slightly,
    Um, no.

    Lifepo4, being the SAFEST of the Li-ion family, takes much more abuse, than say the li-ion battery in your cellphone, which is LiCo02. To actually force an event for Lifepo4, you have to push it hard in various extremes. In the middle ground for safety are your powertool batteries, which are LiMn chemistry. (However years ago powertools did actually contain A123 Lifepo4's, but not today). Each trades power density (and size) for stability along the scale. At the low end, you can sit on Lifepo4 in a motorcycle or wheelchair. At the high end of the scale, you have very tiny batteries with a LOT of density, but make a very slight mistake and game over. Lico02 are particularly demanding.

    One has to keep the chemistry right. It isn't much different from someone needing a lead-acid battery, and not knowing if they need to install Flooded, Gel, or AGM. Well, not exactly, but you get the point.
    [*] Lithium iron phosphate battery are cheaper than FLA if you compare "KWH" to "KWH".
    No way. Upfront costs are a LOT more for Lifepo4. However, the other attributes, if honestly evaluated, may make it worth your while. Of course salesman hype these attributes sometimes waaay beyond their importance, and don't account for your real needs in an application.
    [*]LiFePO 4 don't mind sitting on SOC >50%.

    They don't mind sitting at 80% SOC either, BUT despite their low self-discharge, you can get them to a point where they go below this into the deep discharge knee, and if left long enough, undesirable chemical reactions occur and the subsequent recharge at normal current is harmful. That is one reason why you don't "jump start" a dead Lifepo4. You *may* be able to bring it back, but cycle life is vastly compromised. The real way to do it is to use very LOW current, about .01C for many hours until it reaches the normal voltage, and only THEN can you apply the normal charge current. Still, you've taken a hit.

    This also means that you have no sort of LVC or monitoring, which is not good for any battery.

    50% is convenient, but truthfully, unless one is going to go to RC modeling extremes, just not leaving them in a fully charged state for a long time is the way to go. One trades a bit of perfection for practicality.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    drraptor wrote: »
    Flooded cell lead acid batteries can also be overcharged for 1-2 times, before you realize charging setting needs some adjustments. At least with translucent plastic covers we can check the water level without opening the battery caps

    Glad you brought that up. For Lifepo4, you CAN charge to 100% SOC, but there is no need to do so.

    Your best bet is to operate between a 90 - 20% SOC, (or 80-10% DOD if you work in DOD). This allows for a little bit of cell imbalance headroom, and operating from the lower part of the steep charge knee, prolonging cell life.

    The loose analogy here for only bringing Lifepo4 up to 90% SOC, is much like doing microwave popcorn right. Instead of waiting for that very, very last kernel to pop, sacrifice a few kernels. I sacrifice quite a few and stop the cook. Sure, I have less popcorn in the bag, but what's left isn't dry and on the edge of being burnt.

    In this case, while the "spec sheet" for microwave popcorn may take you very close to 100% capacity, the end result is not actually the best in the long run. :) Experienced popcorn makers and Lifepo4 users know that 100% SOC, while achievable, is not the best way to go.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    PNjunction wrote: »
    Glad you brought that up. For Lifepo4, you CAN charge to 100% SOC, but there is no need to do so, as the capacity / power between 95-100 SOC is almost microscopic.

    I keep seeing statements like the above and it drives me crazy!
    The capacity between 95% and 100% SOC is exactly 5% of the total battery capacity, by definition. It will never get bigger or smaller than that!
    What may be an accurate statement is that the relative to the time it takes to bring the battery from 95% to 100% or relative to the cycle life it kills from the batteries service life the amount that it extends the use time is small. But that is not the same as the absolute amount of power being small.

    It is certainly true that being kept in the range above 80% SOC is generally good for Pb chemistry (in terms of cell life) while being kept above 80% SOC when not in use is bad for LiFePO4 chemistry.
    SMA SB 3000, old BP panels.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    inetdog wrote: »
    I keep seeing statements like the above and it drives me crazy!
    The capacity between 95% and 100% SOC is exactly 5% of the total battery capacity, by definition. It will never get bigger or smaller than that!
    What may be an accurate statement is that the relative to the time it takes to bring the battery from 95% to 100% or relative to the cycle life it kills from the batteries service life the amount that it extends the use time is small. But that is not the same as the absolute amount of power being small.

    This is exactly right.

    That last 5% takes more charge energy than the 5% before, and the 5% before that, etc. Charging efficiency drops off as you near 100% SOC, but the battery capacity is defined by percent of the total at a give fixed rate of discharge. Doesn't matter what the chemistry is: 5% of total capacity is 5% of total capacity, whether it's the first 5% or the last.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    inetdog wrote: »
    I keep seeing statements like the above and it drives me crazy!

    You guys are right. I was a bit sloppy as we know that SOC voltage is not a true indicator of capacity. Fixed.

    I think where that reference came from was in the back of my mind from the EV community, where higher voltages can equal more motive power. Perhaps they are saying that if you charge to the top of the charge knee, not only will you NOT feel it and gain no advantage over charging to just past the bottom of the knee, you will also quickly drop like a rock to the flat part of the knee.

    Thanks for the spanking. :) <grin>
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    More of a reiteration to not depend on Voltage as a measurement of SOC.
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    My understanding is that unloaded voltage is effective at determining the ~5% and ~95% SOC points. But at other points (flat part of the curve), you need to also use coulomb counting.

    Luckily, chip manufactures are putting everything needed in low cost ICs.

    I am available for custom hardware/firmware development

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    A very simple test is after a full charge, and perhaps 12 hours of rest, is if your ocv is above 3.45/cell, you overcharged it. If below, you undercharged it, but that is not harmful.

    Trying to determine the SOC in the flat part of the curve is almost impossible with low end metering, BUT with a very high-precision and accurate meter, you can ballpark it. This means something like a Fluke 87V or *better* just to get ballpark figures. Note that high-precision and accuracy do not mean just a large amount of digits past the decimal point, so don't be fooled by $30 specials.

    Thanks to CY on ADVrider, you may be able to get by with calibrating your own meter to someone who does have an accurate meter. Take a CR123 lithium battery to the accurate meter, and compare it to your own. Now you'll have the offset needed. Using a 3v lithium ensures that your meter is doing the measurement in the proper range if it is autoranging, and with the accuracy that range provides.

    http://www.advrider.com/forums/showthread.php?t=757934

    But yes, better methods of monitoring are wise.
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?
    zoneblue wrote: »

    I need real numbers, what does it actually cost people who have done it. If theyre imported, as they appear to mostly be, i need frieght, duty/customs charges etc. Delivered to the door costs.

    I have had my LiFePO4 batteries for 14 months now, here is a link to a costing I did http://forum.solar-electric.com/showthread.php?19079-Alternative-to-Lead-Acid-Batteries&p=178256#post178256
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    One of the important variables in calculating cost is the life in years. I've heard numbers of 5 to 20 years. If it's on the lower end of that, then a very low %DOD to increase cycles might not be beneficial - you will run out of years long before you run out of cycles. Temperature is a significant part of the issue (ie, it limits life in years). Maybe the batteries should be kept in the refrigerator?

    I am available for custom hardware/firmware development

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    I have had my LiFePO4 battery for about 14 months, it replaces a number of ageing flooded NiCd batteries that were out of railway locomotives.

    To date I have been very happy with the new cells.

    A big advantage of the LiFePO4 cells over the NiCds and Pb cells is their efficiency. Coulomb/current efficiency is >98% compared with around 70% for my old NiCds. I can accurately keep track of the energy stored in the LiFePO4 battery with an AH meter. This added efficiency has meant I did not need to buy two more solar panels and upgrade my solar controller.

    Another big advantage is that there is no penalty for not keeping the cells fully charged so I can make better utilisation of my solar panels and inverter. In my case my base load requirements (fridge, lighting, computers, router & server, power to charge electric pushbikes, kettle for boiling water and toaster) use about 2.5kWHrs per day which is the average daily power generated by my solar panels in winter. On sunny days in winter i harvest about 3.5kWHrs and on overcast days it can get as low as 0.5kWHrs. Today is the first day my battery has been "full" (3.45v/cell) for six weeks. During this six week period i would say the average SOC of the battery was around 40% dropping to as low as 15%. This would not be the best way to treat Pb batteries, yet it will extend the life of LiFePO4 batteries to have them around 50% of SOC rather than full. In summer I harvest around 5kwHrs/day and use the extra power to run an efficient electric oven and induction cooktop. This extravagant use of power would be detrimental to Pb based batteries but I don't think will have much impact on the calendar lifespan of LiFePO4 cells.

    With a very gentle charging scheme (3.45v/cell (~95% SOC) for one day a week and 3.375v/cell (~80% SOC) for the other days and keeping the batteries as cool as possible I am hoping to get 10-12 years of use out of the LiFePO4 cells, only time will tell...
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    Do you have any documentation on your system at all? Be interesting to see it.
    karrak wrote: »
    I have had my LiFePO4 battery for about 14 months, it replaces a number of ageing flooded NiCd batteries that were out of railway locomotives.

    To date I have been very happy with the new cells.

    A big advantage of the LiFePO4 cells over the NiCds and Pb cells is their efficiency. Coulomb/current efficiency is >98% compared with around 70% for my old NiCds. I can accurately keep track of the energy stored in the LiFePO4 battery with an AH meter. This added efficiency has meant I did not need to buy two more solar panels and upgrade my solar controller.

    Another big advantage is that there is no penalty for not keeping the cells fully charged so I can make better utilisation of my solar panels and inverter. In my case my base load requirements (fridge, lighting, computers, router & server, power to charge electric pushbikes, kettle for boiling water and toaster) use about 2.5kWHrs per day which is the average daily power generated by my solar panels in winter. On sunny days in winter i harvest about 3.5kWHrs and on overcast days it can get as low as 0.5kWHrs. Today is the first day my battery has been "full" (3.45v/cell) for six weeks. During this six week period i would say the average SOC of the battery was around 40% dropping to as low as 15%. This would not be the best way to treat Pb batteries, yet it will extend the life of LiFePO4 batteries to have them around 50% of SOC rather than full. In summer I harvest around 5kwHrs/day and use the extra power to run an efficient electric oven and induction cooktop. This extravagant use of power would be detrimental to Pb based batteries but I don't think will have much impact on the calendar lifespan of LiFePO4 cells.

    With a very gentle charging scheme (3.45v/cell (~95% SOC) for one day a week and 3.375v/cell (~80% SOC) for the other days and keeping the batteries as cool as possible I am hoping to get 10-12 years of use out of the LiFePO4 cells, only time will tell...
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Re: Lithium iron phosphate battery (LiFePO 4) battery for Solar applications ?

    I think your strategy will get you close, and is much the same thing I'm doing.

    We're pretty familiar with super high energy-dense cellphone batteries (LiCo02) having an aging shelf life of 3 years whether you use them or not. Depending on where you read the data, lower energy-dense LiFePo4 like our Winston, GBS, or CALB large prismatic cells are predicted to have at least double that, so i think 6 years is not unreasonable. However as Prof J. Dahn points out, merely running fast repetetive cycles in a lab is not the only thing to consider.

    Problem is, we aren't 10 years out yet. However, many of the mariners that are following this example by not charging and discharging into the extremes, seem to be just fine at 5 years or so of regular usage.

    One thing they learned about the manufacturer's spec sheets for voltages, is that yes, there is a max voltage associated with immediate damage. Then there is the "recommended voltage", which for many of us seem high. My feeling is that this is to help get across the point of operating lower than the damage point, and not use that as an HVC limit.

    The recommended HVC voltage on the spec sheets, along with 80% DOD nets you the typical 2000 cycle rating. My gut feeling is that the recommended hvc voltage is to help adoption by using existing Pb charging paramaters (3.65v/cell typical). So while 2K cycles is great on it's own with these voltages, by dropping the voltage even further, more than 2K is achievable. That, along with reducing the DOD to merely 75% instead of 80 can be very beneficial.

    The problem I'm running into is that I'm having trouble finding Pb chargers that actually DO have lower voltages. One that I'm looking at now is a Samlex SEC series because I can dip-switch them down to 14.v absorb. I'm not sure how it will react to the extreme low internal resistance of lifepo4 however. I wonder if Iota's can be adjusted down, and if they won't see a lifepo4 as a big short... hmmmm.

    Maybe these Pb charger manufacturers could get the hint and tidy things up for Lifepo4 by merely dropping the voltage, losing temp-comp, (not temp overload!), and doing only a simple CC/CV two stage with sanity timeouts. My Optimate-lithium is great, but only 5A.

    Anyway, this is just me trying to be clairvoyant. With only 10 deep cycles under my belt with my GBS cells, I have a long way to go. :)