Pros & Cons of Lithium-ion batteries for solar application

johnT
johnT Registered Users Posts: 15 ✭✭
Hi all,

I have a friend who has been telling me that lithium-ion batteries are the way to go for an off-grid solar system. His claims are: longer life, very low to no maintenance, less dangerous (likelihood of explosion), not having to worry about DOD (can be cycled deeply with no ill effects) and a few other things I can't remember right now.

All of this sounds great, but I am wondering...if these claims are real, why haven't I seen anything about them so far in my search for info on building an off-grid solar system? Even after he told me about the possibility and I tried to search on the topic I could find very little info on the matter. I have found this forum to be about the best source of reliable info out there, so here I am.

I realize they are a little pricier...but not prohibitively, especially if you get twice the life (or more).

http://balqon.com/store.php#ecwid:category=3262346&mode=product&product=13936727

This is similar in price to a Crown forklift battery of the same capacity on the NAWS store site (about $700-900 more...but again, IF you get longer life...and maintenance free has to be worth something)

That actually brings up another question I have. This is a quote from the NAWS site about the Crown forklift batteries..."These batteries are designed for extremely heavy usage. In a typical solar electric system they will normally last for about 20 years or more." I told this to my friend and he just laughed and said "yeah, right". 20 years or more DOES sound like a lot from any kind of battery. Just wondering if anyone on the forum has knowledge of this actually happening.

Anyway, any thoughts on going lithium-ion for off-grid?

Thanks
«13

Comments

  • ewryter
    ewryter Registered Users Posts: 1
    Re: Pros & Cons of Lithium-ion batteries for solar application

    Seems pretty expensive.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    johnT wrote: »
    Hi all,

    I have a friend who has been telling me that lithium-ion batteries are the way to go for an off-grid solar system. His claims are: longer life, very low to no maintenance, less dangerous (likelihood of explosion), not having to worry about DOD (can be cycled deeply with no ill effects) and a few other things I can't remember right now.

    All of this sounds great, but I am wondering...if these claims are real, why haven't I seen anything about them so far in my search for info on building an off-grid solar system? Even after he told me about the possibility and I tried to search on the topic I could find very little info on the matter. I have found this forum to be about the best source of reliable info out there, so here I am.

    I realize they are a little pricier...but not prohibitively, especially if you get twice the life (or more).

    http://balqon.com/store.php#ecwid:category=3262346&mode=product&product=13936727

    This is similar in price to a Crown forklift battery of the same capacity on the NAWS store site (about $700-900 more...but again, IF you get longer life...and maintenance free has to be worth something)

    That actually brings up another question I have. This is a quote from the NAWS site about the Crown forklift batteries..."These batteries are designed for extremely heavy usage. In a typical solar electric system they will normally last for about 20 years or more." I told this to my friend and he just laughed and said "yeah, right". 20 years or more DOES sound like a lot from any kind of battery. Just wondering if anyone on the forum has knowledge of this actually happening.

    Anyway, any thoughts on going lithium-ion for off-grid?

    Thanks

    Well, the first thing that caught my eye in on the site was the 5 year warrantee. That should make you nervous about 20 year life. Also I don't think that this kind of LithiumIronPhosphate (LiFePO4) cell has been around for 20 years in anything close to this application.
    The electronics in the Battery Management System (BMS) will also be important in the expected life of the cells. And 20 years of daily use is 7000 cycles, which is a lot for any battery.
    I would approach it with caution unless I also needed the light weight and small footprint that a Lithium battery can deliver.
    There is some experience to go from in use of Li batteries in electric cars, and I would look hard in those forums for more information.

    Still, the idea of a 2100A maximum discharge current from a 412# 12 volt battery is intriguing!
    As long as the BMS can handle it, a high charge acceptance rate could also make it attractive for Solar PV, similar to AGM.

    The combination of the 16 volt charging voltage and 11.2 volt discharge voltage means that you will need a CC and inverter which can adjust to that voltage profile. (For example, putting a 12 volt lighting load directly on a battery being charged at 16 volts will be a problem!)
    SMA SB 3000, old BP panels.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Pros & Cons of Lithium-ion batteries for solar application

    The non-standard "12 volts" can be an issue with devices designed for Lead Acid battery banks.

    The cost of the battery plus cell balancers (per cell charge control) and such make them expensive.

    But if you are pressed for weight (portable/RV applications), Lead Acid cannot touch them.

    We have some older threads about these types--But nothing recently that I recall.

    Hit Google with this search string:
    • lithium iron battery site:wind-sun.com

    Or this google search link.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    I think part of the reason the have not yet become more common is simply historical. Lead acid batteries have been used for a long time now in RE applications and it is what people know.

    I see a lot of unjustified fear and mystification of lithium batteries and their battery management systems by those not familiar with them. I have some iimited experience with LiFePO4 batteries in electric bicycles. Currently I'm using 20 AH 48V nominal (16s) LiFePO4 battery and it's ability to deliver high current and accept high charge rates is impressive. LiFePO4 is a much safer chemistry than other Lithium polymers - no need to worry about fires, exploding batteries, etc. It also has a longer lifespan - many more, deeper discharges before failure. It is heavier per AH than other Lithium chemistries but that is not an issue for RE applications - they're still much smaller and lighter than LA.

    One thing the website linked does not say is whether this battery includes a BMS- if so then the price seems pretty good. The BMS is simply a system which monitors each individual cell, balances current across cells, and keeps each cell within a certain voltage range (usually 2.8-3.6 for LiFePO4). Drop the voltage of a lithium battery too far and it is dead.

    The long term reliability and deep discharge performance of these batteries is still yet to be truly tested. If they perform as advertised, I suspect they will become more and more common in RE applications.

    Endless Sphere Forum is a good source of info on lithium batteries - a lot of very knowledgable people - many building their own battery packs and BMS - primarily for ebikes.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    mtdoc wrote: »
    I think part of the reason the have not yet become more common is simply historical. Lead acid batteries have been used for a long time now in RE applications and it is what people know.
    One thing the website linked does not say is whether this battery includes a BMS- if so then the price seems pretty good. The BMS is simply a system which monitors each individual cell, balances current across cells, and keeps each cell within a certain voltage range (usually 2.8-3.6 for LiFePO4). Drop the voltage of a lithium battery too far and it is dead.
    Battery Management System: Individual cell balancing (temperature, voltage), CAN BUS output

    It looks like the internal components of the BMS are there. Does the CAN BUS output need to be monitored by external CC to shut down charging if cell temperature, etc, get too high, or is this purely informational so that the health of the bank can be monitored?
    SMA SB 3000, old BP panels.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    oh - you're right inetdog! - somehow I completely missed that line on the description.
    Does the CAN BUS output need to be monitored by external CC to shut down charging if cell temperature, etc, get too high, or is this purely informational so that the health of the bank can be monitored?

    I would think the Canbus is just for individual cell monitoring- and perhaps it allows you to adjust some of the BMS parameters. I know they sell displays for BMS that allow people to do just that.

    The BMS itself is what shuts down a cell if voltage is out of spec - it wouldn't be a BMS if it didn't do that.
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    mtdoc wrote: »
    The BMS itself is what shuts down a cell if voltage is out of spec - it wouldn't be a BMS if it didn't do that.

    I was just wondering whether it is also necessary to do something to shut down the voltage/current source if, for example, one or more cells short out and so the string voltage is well below the nominal. I understand that the BMS can "bypass" a cell which should not be accepting any more current, while still allowing current to flow through the rest of the string, but wondered whether the CC would also have to play a role in recognizing a different End of Charge voltage, or whether the BMS can force the battery terminal voltage to rise to the float value even with a bad cell present.

    Thanks for the information.
    SMA SB 3000, old BP panels.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    inetdog wrote: »
    I was just wondering whether it is also necessary to do something to shut down the voltage/current source if, for example, one or more cells short out and so the string voltage is well below the nominal. I understand that the BMS can "bypass" a cell which should not be accepting any more current, while still allowing current to flow through the rest of the string, but wondered whether the CC would also have to play a role in recognizing a different End of Charge voltage, or whether the BMS can force the battery terminal voltage to rise to the float value even with a bad cell present.

    Thanks for the information.

    Good question. I think that a BMS does not do this. At least from my limited ebike experience and from what I've seen the ebike Lithium battery gurus describe, if there is a bad cell then the packs voltage simply is lower and that cell is bypassed. The CC parameters would have to be adjusted then. I guess that is where a connection to look at individual cells can help you diagnose problems.

    I don't know much about how the higher capacity prismatic type cells which are likely to be used in RE applications. In ebikes packs are usually constructed of small cylindrical 3.2 V cells like the A123 batteries. Chargers are specifically set up based on how many cells are in series. My charger is designed specifically for a 16s (16 cell series) pack -considered to be 48V nominal - with charge stage voltages based on that configuration.

    The ebike DIYs often build packs with cells in 72V or 96V configurations and construct their own high powered chargers out of power supplies.

    I would think that a customizable Solar CC like the Midnite Classic could easily be set up to provide the appropriate charge profile for the Lithium pack in the OPs link.
  • waynefromnscanada
    waynefromnscanada Solar Expert Posts: 3,009 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    mtdoc wrote: »

    Endless Sphere Forum is a good source of info on lithium batteries - a lot of very knowledgable people - many building their own battery packs and BMS - primarily for ebikes.

    Awesome link, really great information. Thanks a lot :)
  • Photowhit
    Photowhit Solar Expert Posts: 6,002 ✭✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    johnT wrote: »
    That actually brings up another question I have. This is a quote from the NAWS site about the Crown forklift batteries..."These batteries are designed for extremely heavy usage. In a typical solar electric system they will normally last for about 20 years or more." I told this to my friend and he just laughed and said "yeah, right". 20 years or more DOES sound like a lot from any kind of battery. Just wondering if anyone on the forum has knowledge of this actually happening.

    Yes and Yes, first I'd trust NAWS, next I think we only have one long time user here, but other forums and The old Backwoods solar newsletter have seen examples of 30 years in off grid use (not solar specific) I looked pretty hard before deciding a fork lift battery was the most cost effective storage once I got beyond a double string of golfcart batteries.

    I think int Lithium iron have potential to be the next big thing, I think the Leaf car battery carries a 10 year warranty. I can't recall exactly but they use/loose more in part of the charging cycle as I recall...and maybe less in another...

    To be honest I think NAWS lists forklift as a convienience to it's customers, I paid less than $2500 for a 24V 800 Ahr traction battery delievered.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Magnum MS4024, Prosine 1800(now backup) and Exeltech 1100(former backup...lol), 660 ah 24v Forklift battery(now 10 years old). Off grid for 20 years (if I include 8 months on a bicycle).
    - Assorted other systems, pieces and to many panels in the closet to not do more projects.
  • john p
    john p Solar Expert Posts: 814 ✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    MTDOC The BMS itself is what shuts down a cell if voltage is out of spec - it wouldn't be a BMS if it didn't do that.I have never seen a BMS that can do that and have seen many BMS and tested most of them.The reason is simple it would require wires from every cell that can carry the full current that cell could provide and then an electronic switch to "bypass" the dead cell. Not easy with a big battery bank. Would be possible with a very small serup ,say for an Ebike.
    Every BMS I have tested only cutsreduces the voltage available to a cell if its becoming overcharged .And that can be anything upwards of a simple zenner diode and bleed resistor. This is the method I use on small lithium battery packs for solar and a low voltage cut off switch if any cell drops below 2.4 v ..2.1 v is death almost instantly to Lithium ion phosphate cells. And the drop from 2.5v to 2.1 v is a very steep curve.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    john p wrote: »
    MTDOC The BMS itself is what shuts down a cell if voltage is out of spec - it wouldn't be a BMS if it didn't do that.I have never seen a BMS that can do that and have seen many BMS and tested most of them.

    Sound like you are much more knowledgable on this than me. Perhaps you are taking my phrasing too literally. All I meant was that a BMS prevents a cell's voltage from droping below it's LVC and from being drawn up too high with charging. Is that not correct?
    Every BMS I have tested only cutsreduces the voltage available to a cell if its becoming overcharged .And that can be anything upwards of a simple zenner diode and bleed resistor. This is the method I use on small lithium battery packs for solar and a low voltage cut off switch if any cell drops below 2.4 v ..2.1 v is death almost instantly to Lithium ion phosphate cells. .

    Yes, this is what I was getting at. I guess the phrase "shut down the cell" is a bit too specific (and obviously not correct). Sorry for any confusion.:blush:

    What i have seen is a LiFePO4 battery pack for an eBike with a dead cell that continued to function fine but obviously with lower capacity. I know nothing about the specifics of how a BMS handles that but somehow is does.

    Good to see someone here who has a lot of experience with Lithium batteries and BMS's. I suspect that they could end up being the standard battery chemistry for home RE applications.8)
  • john p
    john p Solar Expert Posts: 814 ✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    MTDOC What exactly do you mean by a "dead cell" ? short circuit (no output at all) open circuit (also no output at all) ??
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    The cell had a very low voltage - below recoverable. At least that is what the person who later pulled the pack apart and tested each cell said. It was a cheap Chinese ebike pack.

    What is your opinion of the battery in the OPs post? Do you have a good source for large capacity LiFePO4 packs suitable fo solar systems?
  • john p
    john p Solar Expert Posts: 814 ✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    Just as I suspected,Not rteally dead. just low output. The reason I asked this is if it was dead and open circuit then that string of cells would not function AND no BMS I know of could overcome that.
    Where I work we have probably in excess of half a million dollars worth of batteries of just about every type. Some you have never heard of. And are not for sale to people outside of governments.. Mostly because of very high costs.
    I work for a test and research facility,and get to test just about every new type of device used for emergency and off grid and some for on grid... I sell nothing and have no friends or relatives that sell anything to do with wind or solar energy. Sorry cant help with you buying items for solar.
  • Eric L
    Eric L Solar Expert Posts: 262 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    Here's a d.i.y-er who has been experimenting with LiFeYPO4 batteries with a 60 amp Morningstar MPPT charger controller and solar panels (experiment starts in post #229). It looks like he started with no BMS, but has been progressively adding it as it occurs to him what can go wrong.
  • SCharles
    SCharles Solar Expert Posts: 123 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    johnT wrote: »
    Hi all,



    http://balqon.com/store.php#ecwid:category=3262346&mode=product&product=13936727



    That actually brings up another question I have. This is a quote from the NAWS site about the Crown forklift batteries..."These batteries are designed for extremely heavy usage. In a typical solar electric system they will normally last for about 20 years or more." I told this to my friend and he just laughed and said "yeah, right". 20 years or more DOES sound like a lot from any kind of battery. Just wondering if anyone on the forum has knowledge of this actually happening.



    Thanks


    I have real-life experience with forklift batteries in our PV off-grid home. I bought a brand-new set of 1860 amp hr. C&D forklift batteries [the typical, 2-v. cells, six of them] back in '91. I don't use a generator to charge my batteries, only the pv panels. I sold them last year, after about 20 yr., and they were still working, if not as well as new ones. So, yes, they do, or can, last that long or longer.

    I did not realize that I had too small an array of pv panels to properly charge such a large-capacity bank of batteries. For well over 15 yr. of having those batteries, my PV array was around 900 watts output. At the fifteen yr. point, I did figure out, by recalculating, my array was too small, and I increased the size by more than double to around 2000 w. However, the batteries never showed any problem with the smaller array, and they did not function significantly better after I doubled the size of my array. What did improve was my charging ability during and following any spells of cloudy and/or snowy weather.

    I have gone to a different battery chemistry now, but my experience with the forklift set was stellar.
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    LiFePO4 batteries have great potential for off grid use. As discussed, BMS system is the key. LiIon batteries don't take overcharging so you need the balancing to ensure each battery gets the charge needed without overcharge. As aging progresses the series batteries have a little different capacity meaning they need slight different charging amounts.

    There is some degradation with wide DOD but staying between 25% and 85% SOC should give good longevity. One big advantage for off grid PV based systems, over lead-acid, is the LiIon's are perfectly happy with continual partial charging. They don't have to be topped off. For PV solar you are better off not taking the LiIon battery to full charge every day as the full charging voltage stresses the battery more. LiIon batteries have very good recharging efficiency, in the 92%-95% range, not considering the losses that might be associated with the BMS system.

    I would not trust the BMS in the four cell packaged 12v battery. The BMS is likely just a resistive bleed system that turns excess charge into heat. For large battery it is impractical to rely on a resistive bleed balancing if you need a significant percentage DOD on a daily basis. You can't shunt away that much power by resistive dissipation so balancing is likely low current resistive bleeding, taking a long time to even out the charge state. Resistive balancing is also wasteful. A resistive bleed system on a single 3.2v cell being charged at 50 amps would dissipate over 150 watts per cell that has its voltage capped at 3.4 vdc. If several cells in a 48v system are hitting their voltage cap, before the other cells have caught up, it can be quite a bit of heat to dissipate.

    The balancing system's ability to bleed away excess cell charging current becomes more critical as the batteries age.

    LiIon automotive systems use a charging system that can add or subtract charging on an individual cell basis, yielding a high efficiency BMS. A high efficiency switch mode power supply can throw current down the total series stack, or add or steal current on an individual cell basis to quickly balance the SOC for each cell.

    Attached is an article that covers various BMS systems. The best, in my opinion, is the simplified figure 10 diagram.

    One thing that cannot be overlooked as a disadvantage to the LiIon BMS system is the massively complex wiring matrix that is required for the battery stack BMS system.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    RCinFLA wrote: »

    I would not trust the BMS in the four cell packaged 12v battery. The BMS is likely just a resistive bleed system that turns excess charge into heat. For large battery it is impractical to rely on a resistive bleed balancing if you need a significant percentage DOD on a daily basis. You can't shunt away that much power by resistive dissipation so balancing is likely low current resistive bleeding, taking a long time to even out the charge state. Resistive balancing is also wasteful. A resistive bleed system on a single 3.2v cell being charged at 50 amps would dissipate over 150 watts per cell that has its voltage capped at 3.4 vdc. If several cells in a 48v system are hitting their voltage cap, before the other cells have caught up, it can be quite a bit of heat to dissipate.

    The balancing system's ability to bleed away excess cell charging current becomes more critical as the batteries age.

    Great info - thanks! How can one tell that a BMS is the resistive bleed type?
  • john p
    john p Solar Expert Posts: 814 ✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    MTDOC as RC in FLA has pointed out it takes a very complex wiring system to do individual cells. and every Thing has to be closely matched in the system. Its not something to can go buy some parts and cobble something together .OK if you have a virtually unlimited budget and the skills to do it Yes you can.
    This is the reason just about ALL reasonable priced BMS use the zenner diode resistor bleed or something very similar. This type of system cannot overcome a open circuit cell. And if you are using 200ahr and above cells you will run into heating problems of the balancing circuit.. Nearly all available BMS are really designed to use cells below 10ahr. This is very easy for the BMS to cope with.
  • RCCola
    RCCola Registered Users Posts: 5
    Re: Pros & Cons of Lithium-ion batteries for solar application

    I am also interested in setting up a Lithium bank for solar power. In addition to the solar bodger’s DIY project which is quite interesting you should take a look at the following website http://liionbms.com/php/index.php this has a lot of information about BMS technologies that are available on the market.

    I purchased the book on BMS systems that was written by Davide Andrea (I believe he owns Elithion BMS and is responsible for the website mentioned above) which was quite interesting; he has a section in the book that compares the advantages and disadvantages of different balancing techniques. After reading the book I would have to disagree with the previous individual that stated resistive bleed type balancing is inadequate for a large bank.

    When you start out with a new battery bank you must first charge each battery individually to the same charge, there is also a scheme that the author goes into in the book in which you can hook all the batteries together in parallel to accomplish the same thing before you put the battery bank together to use in a bank in series. I would have to look at it again to remember the details. If you do not start off with the same state of charge for all batteries you are doomed to failure from the beginning. If you do however do this the periodic balancing from the BMS should not require very much and should be fine to use a resistive type bleed BMS, the author claims on pg 79 that a 100ma bleed is sufficient for most battery banks. He does not state the ah size though. I do not think the BMS is the biggest obstacle.

    At his website he compares a lot of BMS systems; there are several that I think have really good potential for solar systems. The Elithion System and the Orion Systems look the best to me from what I can tell in the states. Overseas there are several other BMS’s that are equally as good, I however would prefer something local. After looking at all of these systems for a smaller battery pack (48 vdc 300 ah) I am leaning towards the Orion system. The Orion system has a wire to every single cell in the battery bank, for banks that would be used in a car with high voltage this may get fairly messy with all of the wires however for a 48 vdc solar battery in series you are only looking at 16 cells so this is quite reasonable.

    For a lithium battery BMS system to be successful it needs to be able to communicate and control other equipment in the system. It needs to be able to tell an inverter to stop inverting if the battery soc is too low so you do not kill the battery. It needs to be able to stop the charger from charging if it gets too high (Lithium batteries will start on fire if they get an excessively high charge state).

    After reading the article I think the biggest problem is the solar components. I am running into the following problems.

    1) Most Current BMS systems must communicate with the inverter & chargers. The primary reason to control external equipment is to prevent low or high voltage. They do this in a vehicle through canbus typically. Inverter/chargers used for solar cannot be controlled by outside equipment as far as I can tell. And none of them (that I can find) except for Victron inverter chargers use canbus.
    2) A work around for this is to install a contactor, basically a disconnect from the system if the battery gets to low or too high. I was looking at doing this with a Xantrex XW4548 and contacted Xantrex. They do not support this configuration and stated if the Xantrex is connected to shore power and the battery is disconnected all bets are off at what will happen next. This leaves a lot to be desired and sounds quite dangerous. They seemed to indicate a fire could be possible, not sure though.
    a. Ideally the bms would work well enough that the contactor would never be used, it is only there as a last effort to prevent a serious problem (fire or dead battery)
    b. As long as the voltages stay close enough the internal settings for high and low voltage in the inverter /charger (xantrex, magnum, etc.) should be able to protect the battery bank. A BMS however is not looking at the overall charge it is looking at the charge of every single battery and if any of them are in a dangerous range they will shut-off external equipment to prevent a problem. Standard solar inverter / chargers do not have the ability to monitor individual voltages which key when working with lithium batteries.

    My current idea for a work around is as follows:

    1) Use a Xantrex XW4548 inverter to power house circuits (Disable charge functions), this would be connected to the Battery bank with a contactor so the BMS could disconnect it if necessary.
    2) Purchase an Elcon PFC-3000 Charger that has the ability to communicate through canbus with the BMS system. This is actually ideal for me because I have a Honda EU3000is that only has 120V output that I would like to use in the system. This would not be compatible with the Xantrex so it is not a big loss for me to have to purchase a separate charger.
    3) The Orion BMS has 2 canbus interfaces and can have several analog outputs to control equipment that cannot communicate with canbus. I would use the analog outputs along with a Tristar MPPT (similar to what the solar bodger has done)
    4) I am somewhat paranoid so I would also like to use the contactor as an ultimate fail safe to protect the bank. From overcharge from the MPPT charger, this however is already controlled by the analog signal so it should not ever be necessary.

    Problems

    1) I don’t know if there is a certain method that must be used when wiring a separate charger (elcon) and inverter (xantrex) to the same battery. There is a power point on the web by Genasun “Genasun-2011-SITS” that indicates you don’t want to have the charger feeding directly into the same bus the inverter is on. I am not sure exactly what he means by this. (This is specifically for solar lithium battery banks)
    2) I don’t know if the inverter or mppt will have problems if disconnected from the battery bank. The inverter may have power coming into it from the Mppt controller.
    3) Any other unforeseen problems, this is a lot of money to spend to find out there is a problem

    Hopefully the information above is helpful. I would also appreciate it if anyone out there with a bit more installation experience with these types of components has any feedback on the best way to connect the equipment stated above. And if there are any gapping holes in the logic of what I am trying to do.

    Additional Information

    Victron looks like a good solution, I however cannot afford their battery bank. I contacted their dealer in Colorado to see if the canbus could communicate with Orion BMS. The answer I got was basically if it isn’t Victron you are on your own.
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    I can tell from experience the battery impedance goes up very fast as they get over about 3.4 vdc per cell. With a low bleed capability you would have to terminate charge or manually cut back the main charger, when the first cell reaches greater then 3.8 vdc. Leaving the high rate charge current on above this point will cause that cell to very quickly zoom above 4.0 vdc into cell damaging territory. The systems gets more unstable for 48 vdc system with 16 series stacked cells because one or two cells zooming up in voltage does not have a large impact the overall stack voltage so the charger does not cut back its charge current rate, unless told to do so by the BMS system.

    Being just a couple of percent off in the SOC balance and the system quickly gets into a run away situation. You have to consider what happens to cells over time as they age and diverge in their capacity. If you are charging and discharging more AH's per day then about ten to twenty times your bleed capability in equivalent AH's per day, the bleeder will not be able to keep up with required balancing. 50 to 1 may work for new batteries precharged to full capacity to start out with but it won't hold up over time. Continued bleeding to balance cells after charging is terminated is not possible if there is variable discharge loading on the batteries, which is normally the use case for an off grid system.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    RCCola wrote: »
    For a lithium battery BMS system to be successful it needs to be able to communicate and control other equipment in the system. It needs to be able to tell an inverter to stop inverting if the battery soc is too low so you do not kill the battery. It needs to be able to stop the charger from charging if it gets too high (Lithium batteries will start on fire if they get an excessively high charge state).

    This does not make any sense to me. True that a BMS needs to prevent cell voltages from dropping too low. The basic idea is the same as the low voltage cut off that most inverters have for any battery. My experience with Lithium and BMS's is limited to ebikes but the principles are the same. A BMS has a low voltage disconnect. No need for any special communication with the Inverter - the BMS will prevent drawing the voltage down to dangerous levels. For example on my ebike, if the voltage reaches the Low voltage disconnect the BMS simply disconnects it and I am forced to pedal.... It would do the same to an inverter without any need for Canbus or any other connection or communication protocol. In addition , most higher power, quality inverters have programmable low voltage disconnect levels which would act as a second disconnect method to prevent over discharge.

    2) I don’t know if the inverter or mppt will have problems if disconnected from the battery bank. The inverter may have power coming into it from the Mppt controller.

    Sorry, but this make no sense to me either. A Mppt (or any) controller does not connect directly to the inverter. It does not send power to it. It connects to the battery and regulates charging of the battery. Neither an inverter or controller will have any "problems" if disconnected from the battery other than the fact that they can only serve their intended purpose when connected to the battery bank.

    I think you're making this much more complicated than it needs to be. Many people are now using Lithium battery banks successfully with RE systems without any special communication between BMS, inverter and controller. The details and optimization of BMS electronics not withstanding, the basic principles of using Lithium batteries in RE systems are really no different than systems using lead acid batteries. The main difference is the need for a BMS to keep individual cells balanced and as you say, keep them from being over charged or over discharged. Lithium chemistry does need different charging parameters but nothing that a modern programmable mppt controller such as a Midnite Classic or Outback FM controller cannot do.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Pros & Cons of Lithium-ion batteries for solar application

    Nice summary of the issues with "high tech" battery topologies and where we are with Solar Power systems are today...

    Trying to make something failsafe because it will save you money (i.e,. won't ruin a battery bank) is a lot easier than making it fail safe because the battery bank may catch fire (if over or under discharged). The Lithium Iron Phosphate (LiFePO4) batteries are supposed to be pretty safe--But you are still left with the possibility of expensive damage.
    RCCola wrote: »
    For a lithium battery BMS system to be successful it needs to be able to communicate and control other equipment in the system. It needs to be able to tell an inverter to stop inverting if the battery soc is too low so you do not kill the battery. It needs to be able to stop the charger from charging if it gets too high (Lithium batteries will start on fire if they get an excessively high charge state).

    You can get inverters with simple remote On/Off input signal... And you can add a dump controller (and resistor load bank) as a shunt regulator to help prevent overcharging from a failed charge controller... But these are "bandaid" type solutions and not a well integrated system design (just adding more components that can fail/screw up). And--This is still not a great solution if you need to prevent a fire from over/under voltage on the bank.
    After reading the article I think the biggest problem is the solar components. I am running into the following problems.

    1) Most Current BMS systems must communicate with the inverter & chargers. The primary reason to control external equipment is to prevent low or high voltage. They do this in a vehicle through canbus typically. Inverter/chargers used for solar cannot be controlled by outside equipment as far as I can tell. And none of them (that I can find) except for Victron inverter chargers use canbus.

    Yep, pretty much everything is proprietary--And they would not "certify" that a software command to shut down would be acted upon in all conditions (i.e., fail safe).
    2) A work around for this is to install a contactor, basically a disconnect from the system if the battery gets to low or too high. I was looking at doing this with a Xantrex XW4548 and contacted Xantrex. They do not support this configuration and stated if the Xantrex is connected to shore power and the battery is disconnected all bets are off at what will happen next. This leaves a lot to be desired and sounds quite dangerous. They seemed to indicate a fire could be possible, not sure though.

    I would hope that a XW inverter does not cause a fire if disconnected from the battery bank when in service--A fuse/breaker could pop and do the same thing. It is possible that there may be damaged if battery bank connection is cut (don't know--And I would hope not--But who knows).
    a. Ideally the bms would work well enough that the contactor would never be used, it is only there as a last effort to prevent a serious problem (fire or dead battery)

    A fail safe contactor--Welded contacts, control logic fails, etc... Even my previous suggestion of a shunt controller/dump load--What if the fail safe dump controller actually turned on and drained the bank. Causing the very condition you are trying to prevent.
    b. As long as the voltages stay close enough the internal settings for high and low voltage in the inverter /charger (xantrex, magnum, etc.) should be able to protect the battery bank. A BMS however is not looking at the overall charge it is looking at the charge of every single battery and if any of them are in a dangerous range they will shut-off external equipment to prevent a problem. Standard solar inverter / chargers do not have the ability to monitor individual voltages which key when working with lithium batteries.

    Basically a fail safe bus... And device that "pulls the cord" shuts down all connected electronics. If the bus fails, it also causes a shutdown. etc...
    1) I don’t know if there is a certain method that must be used when wiring a separate charger (elcon) and inverter (xantrex) to the same battery. There is a power point on the web by Genasun “Genasun-2011-SITS” that indicates you don’t want to have the charger feeding directly into the same bus the inverter is on. I am not sure exactly what he means by this. (This is specifically for solar lithium battery banks)

    I am not quite sure what the mean by this... But, in general, you want the "battery bus" to be a very low resistance connection directly (short/heavy cables) to the battery cells.

    Each cable/wire that leaves the battery bus connection points, goes directly to its destination with its own breaker/fuse for protection. I.e., one branch to a charge controller. A second branch to the DC input of an AC Inverter, etc...

    What can cause problems is if you have a run of cable from the battery bank to an AC Inverter and DC charge controller (sharing the same wire run--Daisy Chained)... Single Phase AC Inverters don't take a "smooth" DC Current flow--But a sine wave squared 120 Hz current (zero to max current). The inverter's input stage does not have much energy storage and needs to pull power every 1/120th of a second to make the 1/2 wave power peak (voltage*current).

    2) I don’t know if the inverter or mppt will have problems if disconnected from the battery bank. The inverter may have power coming into it from the Mppt controller.

    There will be less issues if you pop the PV input power to the charge controller... That is normal operation (sun rises, sun sets, birds fly over, etc.). Most charge controllers pull their operational current from the battery bank (micro processors, data logging, etc.).

    There is already a UL approved method for cutting DC array power--You take two breakers, say a 63 amp + 1 amp breaker ganged together (handles tied together). You pull over 1 amp on on breaker, it trips, and pops the other breaker(s) too. Basically using this as a "contactor" that only pull operating current for a few tenths of second ("short" pulled through 1 amp breaker).
    3) Any other unforeseen problems, this is a lot of money to spend to find out there is a problem

    There are lots of issues with creating "fail safe" systems... Frequently the fail safes become overly complex and prone to faults and failures themselves. And how do you know than an "armed" failsafe will actually work or not (do you test the fail safes monthly--for example)?
    Hopefully the information above is helpful. I would also appreciate it if anyone out there with a bit more installation experience with these types of components has any feedback on the best way to connect the equipment stated above. And if there are any gapping holes in the logic of what I am trying to do.

    Magnum Inverter and Midnite Solar (AC inverter design/manufacturer + Solar Charge Controller design/mfg)... Both are up in Washington State area and (I believe) talk to each other--Perhaps you can give them the requirements (Min/Max bank voltage and what action to take in each case, "RED Button", additional alarm inputs like bank temperature/smoke/fire detectors/etc.) and see if they are interested.

    Additional Information

    Victron looks like a good solution, I however cannot afford their battery bank. I contacted their dealer in Colorado to see if the canbus could communicate with Orion BMS. The answer I got was basically if it isn’t Victron you are on your own.[/QUOTE]

    Magnum Energy Inc.

    www.midnitesolar.com

    And Midnite has their own forums too. http://midnitesolar.com/smf_forum

    If you can give them a set of requirements--They may be able to help implement the requirements with their hardware.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application
    BB. wrote: »
    There is already a UL approved method for cutting DC array power--You take two breakers, say a 63 amp + 1 amp breaker ganged together (handles tied together). You pull over 1 amp on on breaker, it trips, and pops the other breaker(s) too. Basically using this as a "contactor" that only pull operating current for a few tenths of second ("short" pulled through 1 amp breaker).

    Danger Will Robinson!
    Not all (not even most) handle-tied separate breakers will trip both units when one of them faults. A dual breaker unit can have a "common trip" which will insure that both will open together, and you may see an external handle tie on such a unit. But two separate breakers with just a handle tie added will cause both of them to be opened when you manually flip the handles, but may not do what you want when one of them trips.
    The other way to get the effect you want is with an industrial spec breaker which incorporates what is called a "shunt trip" contact and essentially gives you a momentar relay contact which will cause the breaker to open even though there is no overload. Unlike a relay, you will have to close the breaker again manually.
    SMA SB 3000, old BP panels.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Pros & Cons of Lithium-ion batteries for solar application

    I agree with that one--I have factory breakers for 2x 240 VAC circuits. The inside pair will both trip, but the outside pair--I would be hard pressed to trust my life to it.

    The ones I linked to are deigned for the purpose with DC ratings... It may be difficult to find shunt trip breakers that have the DC ratings you need.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • RCCola
    RCCola Registered Users Posts: 5
    Re: Pros & Cons of Lithium-ion batteries for solar application

    BB,

    Thanks for the input, I will see what the Guys at Midnit Solar can come up with.

    Just to clarify, from what I can tell Lithium batteries are only prone to fire if the voltage runs to high. They would not start on fire if the voltage was low. Low voltage however could ruin the battery cell and make it no longer usable.

    The idea with using a charger from Elcon with canbus communication would be to iliminate overcharge from a charger either from my generator or from the power line. I am most concerned about overcharge due to a fire risk. EV converters are already using this charger for cars so I don't think there would be much problem with doing this.

    The tristar MPPT could hopefully communicate with the analog signals from the BMS to shut down charge current from solor panels if the battery voltage gets too high.

    For protection from draining the batteries I think the Low voltage cutoff built into the inverter should be fine. To maximize battery life I don't plan on running the batteries past 20% SOC. Due to this if the batteries are all being balanced periodically I don't think there is much risk at any one cell having a voltage that is different from the rest of the pack that would damage the cell. The BMS can also run the contactor so this would be a back up protection.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Pros & Cons of Lithium-ion batteries for solar application

    One of the things that a BMS does - at least the ones I have experience with - is prevent over charging and over discharging. Using a lithium battery bank with BMS in a typical solar installation with quality charge controller and inverter would mean that now both the BMS and the inverter and charger would be preventing over charging and discharging. Redundancy is good.;)

    The other important function of a BMS, as I understand it, is to maintain balance between each individual battery cell. This is something that an inverter or solar charge controller cannot do. The BMS does not need to communicate with any other electronics (inverter or CC) to do it's job.

    Many electric bike enthusiasts run high voltage, high amp, Lithium ion battery packs with no BMS. They use switch mode power supplies , often stacked, to charge their batteries. But they also use equipment to monitor individual cells and balance cells when needed. They closely monitor their battery packs and act as human BMS's if you will.

    Lithium Iron Phosphate chemistry is inherently much safer than other lithium ion chemistries and the risk of fire, though not zero, - is not a real concern in any reasonable RE system with over and undervoltage protection. Thermal runaway and resultant fire is a possibility with traditional lead acid chemistry as well. And of course with FLA batteries there's the dangers that come with Hydrogen gas production...

    I would not be surprised if LiFePO4 batteries become the standard for RE systems in the not distant future. I think the main hurdles right now is higher cost and even more, the fact that most Solar installers, equipment manufacturers and retailers have little or no experience with them.
  • RCCola
    RCCola Registered Users Posts: 5
    Re: Pros & Cons of Lithium-ion batteries for solar application
    mtdoc wrote: »
    One of the things that a BMS does - at least the ones I have experience with - is prevent over charging and over discharging. Using a lithium battery bank with BMS in a typical solar installation with quality charge controller and inverter would mean that now both the BMS and the inverter and charger would be preventing over charging and discharging. Redundancy is good.;)

    The other important function of a BMS, as I understand it, is to maintain balance between each individual battery cell. This is something that an inverter or solar charge controller cannot do. The BMS does not need to communicate with any other electronics (inverter or CC) to do it's job.

    Many electric bike enthusiasts run high voltage, high amp, Lithium ion battery packs with no BMS. They use switch mode power supplies , often stacked, to charge their batteries. But they also use equipment to monitor individual cells and balance cells when needed. They closely monitor their battery packs and act as human BMS's if you will.

    Lithium Iron Phosphate chemistry is inherently much safer than other lithium ion chemistries and the risk of fire, though not zero, - is not a real concern in any reasonable RE system with over and undervoltage protection. Thermal runaway and resultant fire is a possibility with traditional lead acid chemistry as well. And of course with FLA batteries there's the dangers that come with Hydrogen gas production...

    I would not be surprised if LiFePO4 batteries become the standard for RE systems in the not distant future. I think the main hurdles right now is higher cost and even more, the fact that most Solar installers, equipment manufacturers and retailers have little or no experience with them.

    The BMS protects the battery by telling other equipment to either stop producing power (voltage too high) or stop using power (voltage too low). It also balances the cells by either transferring power between cells or bleeds power off of high cells to allow other cells to catch up (bleeding power is the more common approach). The main idea behind the balancing is too balance frequently to prevent cells from ever getting too far appart. BMS's do not have the ability to blead enough power if they are being feed excessive amounts of power. (ie. if all cells are already at 4 volts and charging must stop but the charger keeps producing power the BMS will not be able to keep up.) The typical method of stopping a charger that you cannot communicate with is for the BMS will disconnect the battery bank from the charging source through a contactor. From what I can gather from online and from Xantrex technical service most solar inverter/chargers do not like to be disconnected from the Battery bank. (I don't know why this is just not a supported configuration, power comming into AC1 from line power with no battery to act as a buffer between the house and AC1. Not sure what will happen but it seems that this is definetly not good with Xantrex XW inverters and from what I can gather on line probably other inverter / chargers have a similar problem. If anyone knows more about this that would be great)

    Another possible method to prevent overcharge would be to just have the BMS send everything to a dump load. Not sure if this would cause additional problems or not. Seems like it should work.

    Typically if all the cells are balanced properly the Low or High voltage cut off built into the inverter/charger should make it so the contactor would never be used. With Lithium cells though if you get an unbalanced bank for any reason a pack voltage is not good enough way of measuring voltage for the low or high voltage cut offs. (possible reason for unbalanced bank with BMS: BMS's typically balance at a specific region they do not usually balance at all voltages. If you have a BMS that Balances at 3.7 volts and you rarely get your batteries to this high of a voltage they might get out of balance over time due to the BMS never kicking in the balancing function. If this happens for too long of a period of time (months) if you ever get to 3.7 volts the difference maybe large enough that the BMS cannot handle it.