Charge controller for non-AGM/FLA's?

McGuyver wrote: »
Good day friends, I have 6 x 68W Unisolar panels in parallel controlled by an Outback Flexmax 80. I am replacing my FLA batteries with 800Ah LiFePO4. The panels are rated at 16.5v Vpm but according to the controller, I get an average of 22V in from the panels. I have a space issue on my RV and would like to add another 300/400watts. Can I add mono or poly panels into the array? What Vpm should I get - 16.5v or 22v?:confused:


Hi there,

This is exactly what I am working on:
- I ordered a 6 Uni-Solar PVL68 and will be putting them on my roof along with a PVL33 that I already have
- I am recycling a 48V 78Ah LiFePO4 16 cells Mottcell battery kit (Prius retrofit) with balancers, BMS cutting off charge at any cell above 3.8V
- I have purchased 2 Sun Power 250G 250W each (fried) and now a Sun Power 500G 500W (22-60V) and a 300G 300W (10.8 - 30V)
- I am planning on using part of the pannels to fid power into the grid all day long and the other part to store electricity into my battery bank, then reuse it wherever cooking, ironing, dishwasher, water-heater are used

My question is : what kind of controller can I use to fill up my battery bank ? (Outback Flexmax / Tristar / ...) since they are all made for FLA/AGM ?
Thanks

Christophe
Paris

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Charge controller for non-AGM/FLA's?

    Welcome to the forum.

    I'm moving this to its own thread because it's more about charging LiFePO4 batteries than mixing solar panel types.

    Off the top of my head I'd say any of the programmable MPPT controllers should be able to accommodate them providing you have the cell balancers. However, the charge profile required may be different from what a standard controller can provide in terms of Voltage levels and duration. I'm not familiar with the particular batteries in question, but I bet someone here is.
  • McGuyver
    McGuyver Registered Users Posts: 14
    Re: Charge controller for non-AGM/FLA's?

    Hi Christofphe, I thought it good to post my reply to your email so that we may get some answers from the pros. So here it is:

    Good day Christophe.
    ThunderSky (lifepo4 supplier) told me that the Flexmax will work on lifepo4. You can set the voltage to match that of your lifepo but you cannot get rid of the bulk or float cycles. Due to the simple charge cycle of Lifepo4, it will gladly accept what comes from the flexmax. It will work, but it is not the ultimate answer because you will lose solar power on the float cycle. At a guess I'd say that a Lifepo controller should be much simpler and cheaper than a Flexmax. I will probably sell the FlexMAx as soon as my Lifepo arrives on July 18th due to a delay in shipping.

    I have a small buck boost DC/DC converter. To my knowledge, it should work better as a charge controller if you can get a big one to fit your system. I also spoke to "Electric1" from Clean Power Auto LLC. He recons that an MPPT controller gets more out of your solar panels than a DC converter. However, I'm waiting for an answer as to how it will match up with the power loss on the float cycle.

    I hope some of the gurus can give some answers...
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Charge controller for non-AGM/FLA's?

    I wonder what they "want" during a float cycle... "...because you will lose solar power on the float cycle."

    Most high end MPPT charge controllers can have the float voltage programmed to another value--What do the LiFePo4 guys want/Need? Higher voltage to keep the cells fully "charged" and not lose stored energy, or lower voltage so as not to waste power through the cell equalization circuits?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Charge controller for non-AGM/FLA's?

    Further to what Bill said, I believe the Xantrex MPPT units can have the "Float" stage dropped completely?
    I might be getting confused on that; it's been a weekend of too much work. :blush:
  • Les Nagy
    Les Nagy Solar Expert Posts: 121 ✭✭
    Re: Charge controller for non-AGM/FLA's?

    Yes, the Xantrex MPPT-60-150 can do 2 Stage charging without float. But as already said this would waste a lot of energy from the panels.

    If the brilliant minds at Xantrex/Schneider would stop ignoring our requests, we could have the AUX output fire when the charger hits float so it could send the output to another load. Not having the AUX output have the charge state as one of it's trigger seems ridiculous to me.
  • System2
    System2 Posts: 6,290 admin
    Re: Charge controller for non-AGM/FLA's?

    @ McGuyver: thanks for posting your email to the forum, it is a good idea

    What LiFePO4 charge controller did you order ? :confused:
    Thanks

    Christophe
  • McGuyver
    McGuyver Registered Users Posts: 14
    Re: Charge controller for non-AGM/FLA's?

    AFAIK lifepo likes to be charged at constant max volts. Amps can vary as long as the pack does not get hot - usually 0.5 to 1C. When the SOC reaches the charger output volt, the charger should cut off completely. Once the SOC drops to a certain level, say 90%, the charger should kick in and put back the used 10%.
    That is how I understand it from what I've heard and read thus far. Now it is likely easier said than done. Would like to get your inputs....
  • McGuyver
    McGuyver Registered Users Posts: 14
    Re: Charge controller for non-AGM/FLA's?
    chris75sf wrote: »
    @ McGuyver: thanks for posting your email to the forum, it is a good idea

    What LiFePO4 charge controller did you order ? :confused:
    Thanks

    Christophe
    I haven't yet. Still checking what everyone is commenting before I make up my mind. I am currently looking at some DC/DC converters to find how efficient they are. If I can find one at 90%+ efficiency it might be an option. Meanwhile the Flexmax will have to do. ;)

    I got the meanwell charger which is basically a power supply. It does not have any "cycles" like the ones for lead acid. That's why I recon a dumb controller will do the job. It only needs to limit the volts to the max acceptable for your lifepo pack. An external cut-off can possibly be constructed to kick in at 100% SOC which is actually what my BMS will do. I am still not sure what to do with any power generated by the panels when the lifepos are filled up. Where can you go with it? Any ideas? :confused:
  • Joe_B
    Joe_B Solar Expert Posts: 318 ✭✭✭✭✭✭
    Re: Charge controller for non-AGM/FLA's?
    Les Nagy wrote: »
    Yes, the Xantrex MPPT-60-150 can do 2 Stage charging without float. But as already said this would waste a lot of energy from the panels.

    If the brilliant minds at Xantrex/Schneider would stop ignoring our requests, we could have the AUX output fire when the charger hits float so it could send the output to another load. Not having the AUX output have the charge state as one of it's trigger seems ridiculous to me.

    I have been thinking about this for a while and I believe the XW AGS can be used for what you want. I run a 1KW dump load and its a problem to control the relay with the AUX output. The AGS can control its output based on battery STATE. I am looking into hacking the AGS to do what X refuses to provide.
  • McGuyver
    McGuyver Registered Users Posts: 14
    Re: Charge controller for non-AGM/FLA's?

    I received this bit of valuable info from "Electric1" from CleanPowerAuto LLC:
    Float is just keeping battery topped off, which doesn’t even apply to LiFePO4 chemistry. LiFePO4 bank does not care about cycles or float, it simply takes all charge until voltage goes up. As voltage goes up, current goes down and charge is finished. If you hold the voltage at certain level ( float level ) then current will get to zero. If solar panel is attached to the bank which is full, then solar energy is not being produced/stored anymore since current has nowhere to flow. Its simple physics, regardless of what equipment you have.

    So this means that FlexMax can stay and will do a pretty good job. That saves me a lot of drama. :D
  • McGuyver
    McGuyver Registered Users Posts: 14
    Re: Charge controller for non-AGM/FLA's?
    chris75sf wrote: »
    @ McGuyver: thanks for posting your email to the forum, it is a good idea

    What LiFePO4 charge controller did you order ? :confused:
    Thanks

    Christophe

    Hi Christophe, I found this LiFePO4 controller and I think it is very reasonably priced
    :GV-10-Custom $190
    Made by Genasun.
    You may want to contact them and find a price for a custom controller for your application?
    http://www.genasun.com/gv-10.shtml
  • john p
    john p Solar Expert Posts: 814 ✭✭✭
    Re: Charge controller for non-AGM/FLA's?

    The controller from genasun is only rated for a 140w panel. Not a lot, it could not give much gain over a cheap PWM controller
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: Charge controller for non-AGM/FLA's?

    Your charge controller needs to limit charging current when any of Li-Fe cells get to 3.6 Volts. I don't know of any controllers on the market today that have external signal input for triggering charging current limiting. They all rely on total battery pack voltage preset, and don't care about individual cell voltages. That may work for Lead-acid but is bad for Lithium based battery, due to it's greater sensitivity to overcharge and non-linear state of charge to internal resistance relation (when approaching fully charged state).

    I think simple and elegant way to solve this issue, is to feed voltage limit signal from BMS directly to PWM controlled resistive dump load. It will gradually burn off excess charging current without disconnecting battery pack or disturbing charge controller in any way. Load resistor has to be large enough to absorb all current from charge controller. In this setup, charge controller can actually be removed and solar panels can be connected directly to battery pack. But you will not have any MPPT and will have less protection against overcharge in case of dump load failure.

    If you can hack your BMS and get voltage limit signal and interface it to trigger solid state relay that would control dump load resistor, then you are good to go, but hard switching of full dumping current may screw up power point tracking, if your controller has that feature. Gradual PWM current burn-off is better for MPPT to be left undisturbed.
    I am working (slowly) on very similar circuit for my 22 cell Li-Fe lawnmower battery pack. It can work for your application if you cannot find anything else.
  • System2
    System2 Posts: 6,290 admin
    Re: Charge controller for non-AGM/FLA's?

    @ McGuyver
    Thanks a lot for the valuable info on LiFePO4 technology and the link (I actually found it few days ago too...)

    Since yesterday, I changed my mind about my project:
    I will use all the PVL panels in series, to give a total of 105V at 4A (441W) and send it all to a Mastervolt Soladin 600, which is taking 45-145V and will feed back into my grid during day time: it will cover the fridge and small appliances consumption (150W & 70W)
    At the same time I will charge slowly my battery bank with a 48V 2A Pingbattery LiFePO4 Charger, whenever daytime load is under (100W) and other batteries
    I will then use my Sun Power 500G 500W GTI whenever a big load is in use (iron, cooker, water heater, etc)
    During winter/spirng/fall the daytime PV production will cover part of my heating Daikin Air/Air heating system (I will see how much...) and I will maybe not be able to charge my LiFePO4 battery bank at the same time, but we will see about that
  • john p
    john p Solar Expert Posts: 814 ✭✭✭
    Re: Charge controller for non-AGM/FLA's?

    Antron X cells can be taken to 4.4v but there is very little power available from them above 3.8.. And that is a good stop figure as it gives 15.2 v the upper limit of most 12v inverters.
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: Charge controller for non-AGM/FLA's?
    john p wrote: »
    Antron X cells can be taken to 4.4v but there is very little power available from them above 3.8..

    Don't know what variation of Li-FePO4 chemistry you have, but cells made by A123, Thundersky, Calb, K2 all specify charging voltage limit of 3.6 Volts. 4.4V sounds like absolute limit before catastrophic failure (cell electrolyte venting or fire). If you want your cells (well at least cell brands I mentioned) to last longer than few cycles, don't charge them above 3.6 V per cell. Also don't discharge below 2.9 - 3 Volts.
  • McGuyver
    McGuyver Registered Users Posts: 14
    Re: Charge controller for non-AGM/FLA's?
    AntronX wrote: »
    Don't know what variation of Li-FePO4 chemistry you have, but cells made by A123, Thundersky, Calb, K2 all specify charging voltage limit of 3.6 Volts. 4.4V sounds like absolute limit before catastrophic failure (cell electrolyte venting or fire). If you want your cells (well at least cell brands I mentioned) to last longer than few cycles, don't charge them above 3.6 V per cell. Also don't discharge below 2.9 - 3 Volts.

    I recon you're spot on. Here's the spec on LiFePO4 from ThunderSky: http://www.thunder-sky.com/pdf/2011219213120.pdf
    2.8V to 4.0V
    I will be charging the 12V pack to 15V. LV is determined by BMS and set at 2.6V
    Is that not too low?
  • AntronX
    AntronX Solar Expert Posts: 462 ✭✭
    Re: Charge controller for non-AGM/FLA's?

    2.6V is pretty much 100% DoD. Ok for abolute limit setting but not so for everyday cycling. In that datasheet you linked, look at first graph on page 2. Right after 90% DoD mark, voltage starts to fall of a cliff. For longer cycle life it is recommended to only take your Li-Fe cells down to 80% DoD. So you want to stop discharging right before voltage drops off a cliff. With steady load current, estimating voltage cutoff is not hard, but with varying levels of load using voltage alone is not enough. Better to use comulative Ah or Wh meter with % SoC display.