Announcement

Collapse
No announcement yet.

lithium batteries, bulk float absorb time? end or return amps?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • lithium batteries, bulk float absorb time? end or return amps?

    Hi, we are off grid. We have 16 cells 48v 300ah , 2000w solar, Outback flexpower one pack. We have evpower BMS. Bulk charge to 56, float 54., and I am looking to use absorb setting too. (?) We have now got a fridge & freezer and I think we should set some time for the absorb but I don't know what to enter. We have been told 1hr for every 200 ah we use, per cycle. We use 50 - 80 ah per day, cc can be set in increments of .1 hour so I assume I should set accordingly, but I am nervous about changing things. At the moment is is on zero absorb time. We charge to 58.4 once a week to keep them happy, but I think this may be why the BMS fuel gauge ah counter shows 70% and the soc on the unit show 100% based on voltage, it's getting there on total voltage but two cells are .03 and .06 out of balance and dont seem to be balancing as there is no absorb time to do so.?? Am i correct in this? I could really do with some advise! Please help me. The battery supplier did not want to know us after the sale. I am obviously a newbie, haha!

  • #2
    from my understanding, Lithium batteries do not need much absorb time, (Just a couple - 5 minutes) is all.

    What is the brand/model of battery ? Does the BMS top limit, lower limit, or what ?

    You do not want to put too much power into them, as soon as the BMS reports the first cell is full, you are pretty well fully charged.

    Some folks here may be able to coach you on how to manually charge the lower voltage cells. They are the authority, not me.

    Check out this thread
    http://www.solarpaneltalk.com/showth...on-Phosphate-) Lots of posts and info.
    Last edited by Mike90250; 01-23-2015, 10:47 PM.
    Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister

    Comment


    • #3
      thank you for helping me

      They are sinopoly lfp300ah I think. Thanks, i will check your link. BMS top limits and lower limits each cell, it resets the ah counter thing and balances all cells at 58.4. I guess what I,m wondering, is when it gets to 58.4, it resets & balances, but as absorb is not set to any time at all it has no time to shunt the stuff to the other cells. It stops charging as soon as it reaches the limit and cells stay where they are, slightly different to each other.

      Comment


      • #4
        Originally posted by zuessdoggg View Post
        two cells are .03 and .06 out of balance and dont seem to be balancing as there is no absorb time to do so.?? Am i correct in this?
        Pull those cells and charge them individually aka Bottom Balance.

        FWIW using LFP for solar really requires you know exactly what you are doing and what corrective action to take. All it takes to turn your $10K batteries into a brick is one simple mistake. Fortunately LFP is forgiving to over charges, but like all lithium does not tolerate being fully discharged and thus require a LVD and cell monitoring.

        Lastly there is no need to charge LFP to 100% SOC. In fact they will last 2 to 3 times longer if you only charge them to 90%, and never let them go below 10-20%.

        How does you BMS balance? Do you have a passive balance board on each cell that goes into bypass when the cell is fully charged?
        MSEE, PE

        Comment


        • #5
          Leave it as-is and do not mess around with the absorb time. Your voltages for bulk and float are fine.

          Normally one does not need or desire a float with lifepo4 from a technical sense, but in a real-world sense your float voltage is low enough to be benign, and serves as a parasitic-load trap for any minor unforeseen loads to take the bank down when unattended and not in use.

          Your settings are just fine right now. Do not sweat out the .06v difference between cells. Just keep an eye on them, and if they go past 0.10v, then that may be an indicator of a failing cell, or cell-interconnects going high-resistance. Have you PROVEN that there is truly a .06v difference with a reliable instrument like a Fluke voltmeter just as added reassurance that what is displayed in the Outback is really going on? I'm sure it is, but it is nice to have that reassurance.

          Just keep on monitoring them. Note that a BMS can also be a source of outright failure, or fail very slowly with a parasitic load of it's own that is not balanced! With expensive cells like that, just monitor for the red-flag of going beyond 0.100v difference near the high end of charge. It can be the cells, high-resistance interconnects, or the bms itself aging.

          The one thing that takes some getting used to when switching to lifepo4 is changing our mindset in regards to battery operations as compared to 100 years of lead acid. Namely, you don't NEED to fully charge them all the time. As long as your recharge provides you enough capacity to get the job done without going beyond 80% DOD, you are good to go.

          Charging to 100% SOC is a hard habit to overcome after dealing with lead for so long. Charging to only 90-95% is just fine and the cells will live longer. One thing to look for in a pre-canned setup from others, is if they are compensating and calling 90% SOC the "new" 100% SOC as an aid to those not familiar with lifepo4 characteristics.

          Also note that unless you severely under-designed your battery capacity, most solar operations are in what we call a "Sub-C" application. That is, we use far less than 1C of current during discharge, and when lightly loaded like this, once the battery is balanced - it stays balanced! Thus, one doesn't need to obsess over balance on each and every charge cycle, forcing the battery to actually reach the higher bleed-off voltage for very little gained in overall capacity.

          If you strive for ZERO volt differences between cells, you are fooling yourself that you are actually balancing anything. Each cell has a slightly different overall capacity and internal resistance, which means there will be slight variations. Balqon might even supply that information to you. The problem with obsessing over balance is that you spend a lot of TIME at the higher voltages doing bleed-off, and time spent at higher voltages is what contributes to parasitic reactions in the battery. Thus by seeking zero volt differences, you may actually be harming the cells by the time of exposure spent doing so!

          I think this is truly a case of if it ain't broke, don't fix it.

          Comment


          • #6
            Originally posted by PNjunction View Post
            Leave it as-is and do not mess around with the absorb time. Your voltages for bulk and float are fine.
            I agree with that

            Originally posted by PNjunction View Post
            Your settings are just fine right now. Do not sweat out the .06v difference between cells. Just keep an eye on them, and if they go past 0.10v, then that may be an indicator of a failing cell, or cell-interconnects going high-resistance. Have you PROVEN that there is truly a .06v difference with a reliable instrument like a Fluke voltmeter just as added reassurance that what is displayed in the Outback is really going on? I'm sure it is, but it is nice to have that reassurance.
            I take a little bit of an issue with that. Lithium discharge curves are very flat which is a good thing. 3.6 to 2.6 is the range, and almost linear except at the ends of the curves. Anything between 10 and 90% SOC is linear with 1% = .01 volts. So .06 volts is roughly 6% difference in SOC. That could cause problems if you by chance full around 0 or 100% SOC. It would really be a problem below 10% SOC, where you could push that cell over the cliff and turn it into a brick. .1 volts is way too much IMHO
            MSEE, PE

            Comment


            • #7
              Definitely - due to the flat curve, one NEVER balances in the middle of that curve since the cells could be quite unbalanced, and if measured in the middle could be absolutely spot-on perfect in voltage with each other leading to a false sense of security.

              I think in our application where top-balance is the norm, too much emphasis (and TIME at higher voltages!) is spent on trying to achieve perfect voltage balancing which doesn't balance any capacity at all (since individual cells vary in manufactured capacity and internal resistance).

              So I'm comfortable with <.10v difference between cells as long as they are measured at the top in a sub-c application where you don't plan on driving these things below 80% DOD on average. EV and other vehicular projects - by all means balance at the bottom and get those voltages in line with each other - no question there as there is no headroom at the bottom to be making a mistake.

              Comment


              • #8
                thank you for helping me

                I will have a read over the info you have provided.

                Comment


                • #9
                  As someone who asked a bunch of questions here and spent some time researching LFP and BMS, this is a great thread. Always helpful to see real world experience - especially when the seller of the batteries has stopped returning messages.

                  Comment


                  • #10
                    Well it is a rainy bone chilling 58 degrees here today and have some time to revisit this post and see if I can clarify some things for the OP and ease his/her mind.

                    BMS

                    I am familiar with your EV Power Australian Limited BMS. That is what you have correct? It is a Passive BMS system and works like most other Passive BMS systems. You should have 16 Balance Boards, one for each cell group right? Not sure which Balance Boards you have or which of the two monitors as it makes no difference because they all models do the exact same thing. the exact same way.

                    Let's start with the Balance Boards and how they work. They are Passive Bleeder's aka Vampires with Monitor. There are a very simple circuit board that bolts directly onto the battery term post. They perform two functions:

                    1. Turn on and Bleed or Shunt current around the battery when the battery reaches a Voltage Set Point of around 3.65 volts. The value of the current depends on which model you have of either 800 ma or 1 amp. Very simple operation when the battery reaches 100% SOC the shunt turns on to route up to 1 amp around the cell to pass onto lower SOC cells, or burn off power as heat if all cells are charged up. It has a GREEN and RED indicator LED. Green means you are below 100$ SOC and is monitoring. RED indicates 100$ SOC and in BYPASS.

                    2. Second function is to monitor cell voltage and temperature. It uses a 2 or 3 wire data loop in series to send data back to the BCU.

                    The other piece you likely have is one of the two BCU's they offer. I assume you have the NEC-16C a 48 volt, 16 cell control unit. The BCU monitors the voltage and temperature of each cell. If one or more cells falls outside the pre-programmed parameters will operate either the Low Voltage Disconnect or High Voltage Disconnect relay which are external and you have add-on. It also has a basic signal to tell the charger when the first cell reaches 100% SOC so the charger can reduce current to 1 amp. In addition it also has a crude Coulomb Counter Fuel gauge.

                    OK now we gt that out of the way let's look at the big picture of how the two peices fit together and how they are intended to operate, and why you don't have to worry about a lot of it because your equipment is not compatible with it and how to work around the issues.

                    I am going to start with the so called Fuel Gage and real simple IGNORE IT. It is a Coulomb Counter and it counts Amp Hours In and Amp Hours Out. It would be worth little if you could tell the thing what size batteries you have and could calibrate it. However you can do neither with it. Even if you could Amp Hours are very inaccurate because AMP HOURS going in at at higher voltages than going out. To be useful a Coulomb Counter should be counting Watt Hours, not Amp Hours because Amp Hours is meaningless with out a voltage. The way the Fuel Gage works is ever changing. When you first turn it on you discharge say 50 amp hours on day one, and next day you put back in 50 Amp Hours and it thinks you are back to 100%. Next day you discharge 75 amp hours and recharge 75 amp hours and it thinks you now have a 75 amp hour battery and re-calibrates itself. It does not know what size battery you have. Only way for you to tell it is to turn off the solar and let it run until you completely discharge the batteries and the LVD operates. Even then since Amp Hours In does not equal Amp Hours Out means after just a couple of cycles is already out of calibration. So just IGNORE IT, and pay attention to cell voltages.

                    OK now let's talk about the Balance Boards and charger because that is what you want to know. The Balance Boards are just PASSIVE BLEEDERS that turn on when the cell reaches full charge. During BULK if that is what you want to call it you CC is pumping as much current as the panels can supply which is what we want to get charged up as fast as possible. Issue is your Solar Charge controller is made for Pb batteries, not LFP. BULK is set to terminate when the battery voltage reaches a set point of say 57 volts. At that point Absorb stage is triggered and the controller goes to a Constant Voltage of around 58 to 60 volts and stays there until it times out or the sun sets. If you reach the end of the ABSORB phase the controller now switches to FLOAT which is just another Constant voltage of a bit lower voltage of 54 volts. Well that is great for PB batteries, but does not work for LFP.

                    So what do you do for LFP. For LFP the ideal charge is to charge at Constant Current which is the BULK stage of your CC of as much current as the panels can give until the very first cell reaches 100% SOC. At that point you stay in CONSTANT CURRENT mode but you limit the current to the amount the BALANCE BOARD can shunt of 1 amp. Your CC cannot do that because it is a voltage set point. If you continue to pump say 30 amps, the Balance boards can only shunt 1 amp around the fully charged cell and th eother 29 amps still flow through a fully charged cell until all lower cells equalize. That is a huge No-No on LFP batteries.

                    So how do you work around that problem. Real damn simple, there is no reason to charge LFP batteries to 100% SOC. They are not Pb batteries and can operate in PSOC modes where Pb cannot. In fact you wil prolong the bateries cycle life if you only go to 80 to 90%. So set your CC BULK = Absorb = Float = 80 to 90% SOC, and set the Absorb time to as long as possible until after the sun sets.

                    So what do you do to Balance the cells? Well real simple. Any off grid system requires a genny. Get you a LFP AC charger that is compatible with your BMS BCU and run it off the genny. Does not have to be a large genny, no larger than your panel wattage, because when the first cell reaches 100% current falls back to 1 amp or rough 60 watts. That means it could take several hours to get the lowest cell back up to 100% At 60 watts is a long time and a trickle.

                    Lastly the one must have add-on for your LFP system is a LVD. LFP batteries cannot and will not tolerate one single over discharge. Just one time and you turn that expensive cell into a Boat Anchor. So if you do not have one, GET ONE

                    Hope that helps.
                    MSEE, PE

                    Comment


                    • #11
                      Originally posted by Sunking View Post
                      {snip}
                      OK now let's talk about the Balance Boards and charger because that is what you want to know. The Balance Boards are just PASSIVE BLEEDERS that turn on when the cell reaches full charge. During BULK if that is what you want to call it you CC is pumping as much current as the panels can supply which is what we want to get charged up as fast as possible. Issue is your Solar Charge controller is made for Pb batteries, not LFP. BULK is set to terminate when the battery voltage reaches a set point of say 57 volts. At that point Absorb stage is triggered and the controller goes to a Constant Voltage of around 58 to 60 volts and stays there until it times out or the sun sets. If you reach the end of the ABSORB phase the controller now switches to FLOAT which is just another Constant voltage of a bit lower voltage of 54 volts. Well that is great for PB batteries, but does not work for LFP.

                      So what do you do for LFP. For LFP the ideal charge is to charge at Constant Current which is the BULK stage of your CC of as much current as the panels can give until the very first cell reaches 100% SOC. At that point you stay in CONSTANT CURRENT mode but you limit the current to the amount the BALANCE BOARD can shunt of 1 amp. Your CC cannot do that because it is a voltage set point. If you continue to pump say 30 amps, the Balance boards can only shunt 1 amp around the fully charged cell and th eother 29 amps still flow through a fully charged cell until all lower cells equalize. That is a huge No-No on LFP batteries.

                      So how do you work around that problem. Real damn simple, there is no reason to charge LFP batteries to 100% SOC. They are not Pb batteries and can operate in PSOC modes where Pb cannot. In fact you wil prolong the bateries cycle life if you only go to 80 to 90%. So set your CC BULK = Absorb = Float = 80 to 90% SOC, and set the Absorb time to as long as possible until after the sun sets.

                      So what do you do to Balance the cells? Well real simple. Any off grid system requires a genny. Get you a LFP AC charger that is compatible with your BMS BCU and run it off the genny. Does not have to be a large genny, no larger than your panel wattage, because when the first cell reaches 100% current falls back to 1 amp or rough 60 watts. That means it could take several hours to get the lowest cell back up to 100% At 60 watts is a long time and a trickle.

                      Lastly the one must have add-on for your LFP system is a LVD. LFP batteries cannot and will not tolerate one single over discharge. Just one time and you turn that expensive cell into a Boat Anchor. So if you do not have one, GET ONE

                      Hope that helps.
                      I'm a little confused. I was considering LFP with a similar BMS, with cell mounted boards. Is it the case that if I plan to stay between 80% and 10% DOD with LFP to avoid the knees, that I would not normally reach 100% SOC on any cell, unless the batteries were quite badly out of balance? That shunting would not normally be triggered on any cells on a daily basis if I stopped at 90% SOC? I believe someone here suggested that the cells would not normally get that far out of balance. The individual cell voltage monitoring that comes along with the BMS would be used constantly for the LVD and HVD function.

                      If this is true, it brings me back to a question I asked a month or two ago. How would balancing occur? Even though trying to avoid 100% SOC, would you let the batteries get charged to 100% only occasionally (when you see one or more cells off by 0.1V), and the shunting would occur at this time? And would the low current charger only be used during this infrequent procedure so as to avoid the condition described - 1A shunting and 29A through the batteries?
                      Last edited by Living Large; 02-01-2015, 05:37 PM. Reason: Typed FLP not LFP

                      Comment


                      • #12
                        Originally posted by Living Large View Post
                        I'm a little confused. I was considering FLP with a similar BMS, with cell mounted boards. Is it the case that if I plan to stay between 80% and 10% DOD with LFP to avoid the knees, that I would not normally reach 100% SOC on any cell, unless the batteries were quite badly out of balance? That shunting would not normally be triggered on any cells on a daily basis if I stopped at 90% SOC? I believe someone here suggested that the cells would not normally get that far out of balance. The individual cell voltage monitoring that comes along with the BMS would be used constantly for the LVD and HVD function.

                        If this is true, it brings me back to a question I asked a month or two ago. How would balancing occur? Even though trying to avoid 100% SOC, would you let the batteries get charged to 100% only occasionally (when you see one or more cells off by 0.1V), and the shunting would occur at this time? And would the low current charger only be used during this infrequent procedure so as to avoid the condition described - 1A shunting and 29A through the batteries?
                        It looks like you are correct in your thinking. I am sure others will come along and will comment with their thoughts. You now understand why there are people that say a BMS is useless, but LVD and HVD are required. Those people just manually balance their cells as required or use the method described above. Most BMS products out there are for the EV installation were max power from the cell is the goal and are set to more than 3.5v for balancing. A Voltage programmable BMS seem way to expensive for our use in Off grid battery banks. Off grid battery banks are about longevity more and seldom hit that high a voltage that start the balance circuits on the average BMS.

                        -YS

                        Comment


                        • #13
                          Originally posted by Living Large View Post

                          If this is true, it brings me back to a question I asked a month or two ago. How would balancing occur? Even though trying to avoid 100% SOC, would you let the batteries get charged to 100% only occasionally (when you see one or more cells off by 0.1V), and the shunting would occur at this time? And would the low current charger only be used during this infrequent procedure so as to avoid the condition described - 1A shunting and 29A through the batteries?
                          All correct. As we discussed earlier there is no reason to go to 100% in normal operation because LFP last longer if you operate between 10 to 90%. Only EV's require going to 100%, and a lot of the guys like me don't even go to 100% just for the fact we do not need that extra mile or two and want our packs to last longer.

                          There are two ways to balance lithium cells: Top Balance and Bottom Balance. The inexpensive easy way is Top Balance using a Passive Bleeder as is being discussed here. But to use them with Solar we have a compatibility issue with Solar Charge Controllers and the very nature of how solar systems operate. You can charge at very high rates initially which is no problem with solar. It is when the first cell reaches 100% that we have two problems: 1 We need to limit charge current to 1 amp. The BCU on the BMS generates a signal (contact closure) when the first cell reaches 100%, but the Solar Charge Controllers on the market today do not have the ability to use that signal and limit the current. 2. At 1 amp if there is any significant difference in capacity, there is likely not enough sun hours in a day running 1 amp to EQ all the cells.

                          However that is no big deal and the work around is easy. If you have and Off-Grid Battery system you had better have a Generator and AC Charger. So all you gotta do is buy an AC Charger made for Lithium Batteries. Any DIY EV shop has or where ever you buy your LFP batteries will have them and compatible with the BCU. So when the times comes to Balance you run the genny. Problem solved.

                          Not going to discuss bottom balance as their is no point at this time.
                          MSEE, PE

                          Comment


                          • #14
                            Originally posted by yewsuck View Post
                            You now understand why there are people that say a BMS is useless, but LVD and HVD are required. Those people just manually balance their cells as required or use the method described above.
                            Hold the bus Sir, that is an Oxymoron. To do any one of those functions requires a BMS. Well at least the LVD and HVD

                            LVD and HVD for LFP work differently than PB. For Pb batteries you monitor the pack voltage. So if you have a Pb 48 volt system you set LVD at 42 volts, and HVD at say 60 volts. For LFP you monitor every cell and if any one single cell hits it set point operates the associated disconnect. The only thing in common is the location of the disconnects. Th eHVD is between charge and battery, and LVD is between battery and load.

                            As for balancing there are a few different ways to go about it and can be done manually, but is a real PIA to do manually and could take a few days.

                            The issue right now battery management for Lithium batteries is still in development. There is NO DIFFINITION of what a BMS is, thus no standard protocol. Each manufacture is fighting right now saying their way should be the standard. In the EV world some common standards are coming about called CANBUS which is a data bus protocol so all the parts of an EV can communicate with each other like the motor controller, pumps for battery cooling and hydraulic braking, steering motors, and BMS. Bu twhen you get to the BMS there is no common method. Commercial manufactures use Active Balancing that is a Bottom Balance method which involves taking power from the higher SOC cells to the lower SOC cells. They also incorporate Thermal Management.

                            Botton line is for Solar it just needs to be a KISS system using passive management. Right now that comes from the DIY EV product line. There is really nothing useful for Solar right now because there is NO MARKET using LFP batteries for solar. Today lithium is not ready and cannot compete with Pb systems. I think that will change, but not today. Right now the only thing out there is toy stuff like GenSun made for a boat 12 volt marine LFP battery.

                            2 things have to change for solar. 1 prices of LFP needs to come down. Unfortunately they have bottomed out and starting to rise like solar panels. 2 Is the most important thing, test of time. So far the first two generations failed that test. 3rd generation has just been released, so it will be another 10 years to pass the test of time. If 3rd generation fails, gotta wait even longer to prove 4th generation. I think you see where I am going with this. Personally I think th eLFP market will die, and something like Lithium Manganese will take over. All I can say for now is we are not there yet.
                            MSEE, PE

                            Comment


                            • #15
                              Originally posted by Sunking View Post
                              All correct. As we discussed earlier there is no reason to go to 100% in normal operation because LFP last longer if you operate between 10 to 90%. Only EV's require going to 100%, and a lot of the guys like me don't even go to 100% just for the fact we do not need that extra mile or two and want our packs to last longer.

                              There are two ways to balance lithium cells: Top Balance and Bottom Balance. The inexpensive easy way is Top Balance using a Passive Bleeder as is being discussed here. But to use them with Solar we have a compatibility issue with Solar Charge Controllers and the very nature of how solar systems operate. You can charge at very high rates initially which is no problem with solar. It is when the first cell reaches 100% that we have two problems: 1 We need to limit charge current to 1 amp. The BCU on the BMS generates a signal (contact closure) when the first cell reaches 100%, but the Solar Charge Controllers on the market today do not have the ability to use that signal and limit the current. 2. At 1 amp if there is any significant difference in capacity, there is likely not enough sun hours in a day running 1 amp to EQ all the cells.

                              However that is no big deal and the work around is easy. If you have and Off-Grid Battery system you had better have a Generator and AC Charger. So all you gotta do is buy an AC Charger made for Lithium Batteries. Any DIY EV shop has or where ever you buy your LFP batteries will have them and compatible with the BCU. So when the times comes to Balance you run the genny. Problem solved.

                              Not going to discuss bottom balance as their is no point at this time.
                              Alright - I'm glad this is being revisited. Now, a few last questions. I was looking at off-grid, 4900W PV, 6000W (max) generator, maybe 500Ah @ 48V LFP for example, for normal everyday operation. I'll need a backup generator for the main generator of some sort, which I have not yet considered - in case of failure. But...

                              Let's say I only needed 1A @ 58V for balancing (58W). Is a generator overkill? How long might be needed for this balancing procedure? Might a storage battery suffice to power the LFP charger? Just asking.

                              Comment

                              Working...
                              X