Litium battery management systems

cptdondo
cptdondo Solar Expert Posts: 103 ✭✭✭
I know this is a bit off from the typical setup.  I have a Li-ion battery that consists of 4 cells, being charged from several sources, including solar.  The battery has a capacity of some 200  Ah@12VDC.

I am looking for a BMS for the battery itself; something that will keep the cells equalized while undergoing multiple charge and discharge cycles, as well as play nice with the solar charger.  The typical BMS uses a big relay to disconnect the battery at both over charge and excessive discharge, which can damage the controller. 

Any thoughts or ideas on where to look?

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    I have moved your question into the "New" battery chemistry question forum.

    My question to you--Is this intended for an "unattended" off grid power system where you will need to automatically shutdown the battery bank?

    Or would it be OK to simply have an alarm that alerts you to "issues" with the battery bank?

    Also, Li Ion type BMS systems can measure the voltage across each cell--Or you can simply monitor the voltage of the entire battery bank.

    For (at least) LiFePO4 type batteries--Once you have manually balanced the cells--Many folks seem to avoid the "per cell" BMS type installations--And simply just once a month or so measure the per cell voltage with a good quality DMM to make sure all of the cells are still balanced. Is that acceptable to you?

    If you are using other Li Ion chemistries--Many of those type of cells will have possible explosive/fire issues with over/under charging, and a per cell BMS system seems to be almost mandatory.

    Well--That is about the limit of my BMS/Lithium battery knowledge...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • jonr
    jonr Solar Expert Posts: 1,386 ✭✭✭✭
    I agree with measuring how much of a balance issue you have before deciding on a solution.  Perhaps this, but the charge current is low.  I doubt that mosfets are any better than a relay in terms of possible controller damage.

    http://www.ebay.com/itm/14-8V-14-4V-4S-150A-Lithium-ion-Li-ion-Li-Po-LiPo-Polymer-Battery-BMS-PCB-System-/222114823217

    I am available for custom hardware/firmware development

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    cptdondo said:
    > I know this is a bit off from the typical setup.  I have a Li-ion battery that consists of 4 cells, being charged from several sources, > including solar.  The battery has a capacity of some 200  Ah@12VDC.

    That must be a LiFeP04 chemistry.  Perhaps CALB, Winston, GBS or other cells.  Nominal 3.7v types.

    > I am looking for a BMS for the battery itself; something that will keep the cells equalized while undergoing multiple charge and
    > discharge cycles, as well as play nice with the solar charger.  The typical BMS uses a big relay to disconnect the battery at
    > both over charge and excessive discharge, which can damage the controller. 
    > Any thoughts or ideas on where to look?

    Your simplest option would be to stay out of the charge / discharge knees, especially the upper end by limiting your CV to no more than 13.6v.  At 3.4v per cell, this is guaranteed to never fully charge the 4S set of LFP cells, no matter how much time you let it absorb.  NOTE:  At 3.45v you WILL eventually fully charge the cells with a long absorb, and cell imbalance can be an issue at the upper end of the knees, so don't run at 13.8v !!  It is a small amount of voltage increase, but vital to avoid especially if you are not doing daily cycling, and maybe are using it sparingly in a remote location, etc.

    You can use a balancing circuit, but if you are willing to never charge to full (a great idea btw), and stick to no more than 13.6v, then you won't have any issues, AND will increase cycle life.  And at 13.6v, the voltage isn't high enough to plate the cells, in addition to being able to catch parasitic loads early on.  Of course we are assuming you are already balanced.  In our application, it doesn't have to be absolutely perfect.

    No big relays to deal with either.

    The best part is you don't need to have a rat's nest of bms wiring all over your cells, which in themselves are points of failure.

    Of course, you should run an LVD.  Set it relatively high, at 12.8v (3.2v per cell, perhaps 3.19v) before an alarm or disconnect occurs.  Since you are not a high-current EV application, small amounts of imbalance won't shoot any cell dangerously into reversal.

    For initial balance of a simplistic 4S LFP bank like this, consider getting a "single cell charger", designed for LFP, which has a 3.6 to 3.7v CV.  Attach to each cell and let it finish on it's own.  Use no less than .05C, so I'd recommend anywhere from a 10amp to 60 amp single-cell LFP charger.   After this, just run at 13.6v CV from your charge sources.

    Perform this single-cell charge perhaps yearly as a preventative maintenance and sanity-check procedure.  Be mindful of your polarities!

    People tend to make a simple 4S LFP bank waaaaay more complicated than it has to be.

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    CORRECTION:  LiFeP04 are 3.2v "nominal", NOT 3.7v nominal, as I mentioned above.  Anything *other* than LFP is 3.7v nominal, 4.2v max).

    Had my head in a different lithium project.  Sorry about that.  Again, LFP's are 3.2v nominal (3.6v max typically).  Gotta' keep your chemistries straight so you don't make a blunder and fry your LFP's with the wrong settings or gear.

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    Or Fry your home!
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    You got that right!

    Actually, an overcharge of LFP results in swelled cases, vents opening, and electrolyte cookoff, but not outright combustion.  (The iron phosphate is greedy and holds on to oxygen).  This is one reason you see them a few inches under your butt in a motorcycle or custom wheelchair.  Not good to breathe of course.

    NON-lfp lithium chemistries however are a whole different story.  The cells themselves catch fire if slightly overcharged, and of course also set fire to the wiring infrastructure.  Poorly designed ebikes, benchtop hacks made of kapton-tape and kite string,  and so forth commonly using non-lfp laptop recycled cells, industry rejects, second-hand users, or very poor chargers set these off.

    So yeah, be careful out there, but know the difference in chemistry safety. :)

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    Definitely be careful and think about how you would get the beast outside if needed! I like this picture as a reminder!

    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    That pic is a GREAT reminder about something many don't think about:

    Once non-lfp cells go into combustion, you don't have time to move it.  That's why the RC guys charge in fireproof bags, or at least inside the fireplace.  It is a great tragedy to have carelessness harm you, your family, or property, but as this pic demonstrates, what about the whole hillside that is now on fire and the repercussions from that?

    I see a lot of careless hack jobs being done with kids wiring together non-lfp 18650's without knowing or giving safety any thought for themselves or others.

    ANY battery chemistry, even if it isn't lithium, needs the same kind of care not only about yourself, but of *others*.  Great pic!  I hope it becomes obviously apparent how important safety is.

  • solar_dave
    solar_dave Solar Expert Posts: 2,397 ✭✭✭✭
    Love that picture! Says it all doesn't it.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    Yes it does look like the solar panels made it :)


    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    GREAT chart!

    In it, you'll see the curves for safety - which for me means LFP only for a solar-housepower bank.  One will note the specific energy of LFP (LiFeP04) being the *least* dense.  For a housepower bank, that means your bank will be about 2/3 larger than one made of other chemistries like NMC, or ack, LCO.

    As for power, the charts also show it to be the worst compared to the other two.  Which is entirely ok in our relatively low-current application for a housebank with adequate single or multi-day autonomy.  So not a big issue there.

    Note that they mention a lower voltage for LFP.  In this case, LFP is the ONLY lithium chemistry that is 3.2v nominal, 3.6v max.  This actually works in our favor for 12-48v systems, as in the example of a typical 12v LFP bank only needing 4 cells.

    Sadly, over the last 10 years, the "upfront" cost of LFP has not come down, and I do not think it will in the near future.  Thus, for one to choose lithium for any reason, you REALLY have to justify it, not just because it is the "cool" thing to do.  AND, better hope that your needs don't change in the near future, as nobody in their right mind should buy used second-hand lithium batteries!

  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    PNjunction, can you expand on the term 'used'... A year or 2 back a friend came upon an unused, large LiFePo4 battery that had been built for an E vehicle but was never built.  Battery sat for ~3 yrs. and then the assets were sold off... he asked me if I knew anything about them, charging etc....  I think he is still sitting on them... a large paper weight... 
    So what in your mind defines a USED LiFePo4 battery?
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    Read the warranty. A good company will have all kinds of data that will tell you what you can and can't do.
     If there is not a warranty, it should tell you quite a bit :)
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Heh, a "used" lifepo4 battery means any cell(s) where you aren't the original owner.  If found, one needs to know ALL the details of how they were used in the application, charged, and or stored.  Lifepo4 can hide abuse much better than the typical lithium laptop-battery type chemistry, as problems with laptop type chemistry usually present themselves dramatically right away. :)

    I guess the thing I was really trying to say was that if you make a mistake in your capacity needs, or if your capacity needs change, then trying to pawn a used bank to someone else kind of demands that the future owner trusts you AND your gear completely to have been using and charging them correctly.  Otherwise, you could be 100 cycles down the road and off go the lights. 

    It's kind of a catch-22.  If one has to ask, then you are more than likely to be ripped off, since you don't know what you are looking for.  Such as have these 5 year old cells been maintained at all, or just left to discharge to damagingly low voltages?  If they were well taken care of, do the terminals show any signs of over-torque (well maybe not the ones you are talking about).

    Did the original owner accept batteries that were early-models, factory blem's, or even used batteries before HE even got them?  Many EV'ers understandably are on a tight budget, and with the upfront price being so high, are easily taken advantage of by shady lifepo4 outlets dumping old stock, blems or returns which may have already been abused, unknowingly to the next owner.

    Used or even old LFP batteries, unless one knows ALL the factors, are rarely a good deal, unless you have prior experience and can spot the issues right away.  Hence my admonition from a few years ago to get some simple hands on with a 40ah 4S 12v bank - with NEW cells. :) 

  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    edited December 2016 #16
    Thanks for the post, I think it has something for everyone.  I know that the first thing I would  look for are the charge specs. 
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Yes, the charge specs are the basics.  Thing is, with any lithium chemistry, there are no second-chances, and why all the other details are just as important.

    One thread in a marine forum on another galaxy, a guy lost his very expensive LFP housebank rather quickly being used only intermittently, and much time and effort was spent by the fellow users examining the degradation data over the few years of intermittent use.  Unfortunately, in the *very first* post, the cells were identified as being purchased initially as "slightly used", which NOBODY questioned.

    Everything from that point on was mostly a moot point. :)