Victron smart 12.8v smart lithium battery

2

Comments

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Great to see some progress!

    All the cells being at 3.6V shows that the battery is balanced. This shows the balancing circuitry working, looks like it works at 3.6V and not 3.55. So what has happened is that when the first cells hit 3.6V the balancing circuitry switched on and kept the voltage of the cells at 3.6v from going any higher which allows the other cells to catch up. If the charge current had been too high it would have overwhelmed the balancing circuit and the voltage of the cells at 3.6V would have continued to go higher.

    24 hours at 2A is 48Ah which is 48% of your batteries capacity. The batteries should have been shipped at less than 50% SOC so it is possible that 24 hours is not enough time.

    I agree that doing all the sixteen batteries like this is going to be a chore. You could do two at a time in parallel at 7A which would speed things up.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Hi Simon

    Thank you for your input. The manual says the balancing starts after 3.4vdc which is weird!

    The funny thing is the cells were balanced after the charger finished its charge cycle (it may have took 26 hours, life is a bit of a blur right now since we have a newborn :) So I believe the batteries were charged.

    The other weird thing is that the charger ran on bulk setting the whole way through. But I never saw the cells reach a higher voltage than 3.35vdc. When using the chargers lithium setting.

    İ wasn't content so I put it on reconditioning mode and that's when the voltage went higher. But if the batteries were charged then why did it charge for 8 or 12 more hours on reconditioning mode? I'm missing something here.
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Yerf said:
    The funny thing is the cells were balanced after the charger finished its charge cycle (it may have took 26 hours, life is a bit of a blur right now since we have a newborn :) So I believe the batteries were charged.

    weird thing is that the charger ran on bulk setting the whole way through. But I never saw the cells reach a higher voltage than 3.35vdc. When using the chargers lithium setting.
    Congratulations on the new baby.

    What made you think that the cells were balanced?

    From what you have said here I think that the charger had not fully charged the battery.

    From my reading of the manual When the bulk and lion indicators are on it is supposed to be charging normally. If so it should be charging at 7 amps. Do you have a means of measuring the current output from the charger to confirm this.

    Simon


    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Thank you.

    No I cannot measure it.

    Right now I'm charging one of the batteries from scratch and it's charging at 7 amps 13.7 Vdc. Been at bulk stage for about 5 hours.

    I contacted victron they said that the battery should be reaching at least 14.2 Vdc.

    I haven't seen that under normal charging yet. Only if I put the charger in normal reconditioning mode.
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    It might be worth getting a multimeter, I have a two, a UNI-T UT61E for high accuracy work and a UNI-T UT210E DC/AC clamp meter which allows me to read AC and DC current without breaking the circuit. They are cheap and cheerful but do the job well. If you only wanted to get one I would get the clamp meter or maybe get a clamp meter with a higher current range, the 210E only goes up to 100A.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    edited December 2017 #37
    What is balancing the cells exactly? Is it about filling the batteries to 100% or is it about getting the cells to equal voltage?

    For example if all my cells were 3.33vdc after a balance charge, could I then connect in series parallel or do they have to be 100% charged before connecting series parallel ?
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    edited December 2017 #38
    To be precise getting the cells balanced is getting then all at the same % SOC. If we have two cells with the same capacity, one at 50% and one at 51% SOC the voltage difference between them would be in the order of 0.001V. If we add 45% charge to both cells connected in series we get individual SOCs of 95% and 96% and the voltage difference will be around 0.050V. If we add another 3.5%SOC to 98.5% and 99.5% the cell voltages will be ~3.5V and ~3.6V, a difference of 0.100V. If we add another 1% of charge we get one cell outside its safe operating zone and one at ~3.6V.

    This is the reason for having to balance the cells. With LFP batteries it is only practicable to balance at an SOC close to 100% or close to 0% SOC because this is the only time the individual cell voltages are far enough apart from each other to accurately measure the difference in SOC.

    As far as I can tell your batteries use what is known as passive balancing. Passive balancing is where current is bled off through a resistor when the cell voltage gets to a preprogrammed voltage, in your case 3.6V (which is very close to 100%SOC) so that rather than overcharging the cell the charge current heats the resistor up. All the cells that haven't reached 3.6V continue to be charged until they too reach 3.6V. At the end of the balancing you get all the cells at the same SOC. If you then remove the charge current the voltage will naturally relax back down to ~3.35V.

    You should only connect cells is series if they are all at the same SOC.

    Balancing is further complicated by the fact that the individual cells might not all have the same capacity. If we have two cells, one with half the capacity of the other and we balance them both at 100% (this is known as top balancing) what happens when we discharge the battery by 60% of the stronger cell? The stronger cell ends up at 40% and the weaker cell at -10% which will damage it beyond repair. This is why you have to have a BMS that will detect that the voltage of the week cell is outside its safe operating region when it gets close to 0% SOC and will take action to stop the cell being damaged and potentially becoming dangerous.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    I am surprised in this thread that Victron is not of more help in a battery with their name on it?
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    edited December 2017 #40
    +1  and explain the high voltage referred to... which is up in the FLA battery range...

    I contacted victron they said that the battery should be reaching at least 14.2 Vdc.
     
    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
  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    > @westbranch said:
    > +1  and explain the high voltage referred to... which is up in the FLA battery range...
    >
    >
    > I contacted victron they said that the battery should be reaching at least 14.2 Vdc.

    So basically I have to make sure they all reach 14.2vdc and then baby sit them until all cells are the same voltage?
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    edited December 2017 #42
    @Yerf
    That's the thing with lithium batteries. The initial top balance is critical and time consuming, but once done, maintaining balance is easy and lasts a long time.

    Rick
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    edited December 2017 #43
    Yerf said:
    So basically I have to make sure they all reach 14.2vdc and then baby sit them until all cells are the same voltage?
    Yes that is right.

    Is the Victron charger charging to 14.2V?

    It is likely that on the Li-ion setting that the Victron charger only does a very short absorb period or maybe none at all. You would have to check with Victron.

    You could use the Lead Acid normal setting which will charge at 14.4V and will give you some absorb time before it switches to float.

    Ideally you want a 5 amp power supply that will charge at 14.2 volts and leave it connected to the battery until all the cells have all reached 3.55V and the charge current has dropped to below C/100 (for your batteries 1A)

    Simon

    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Thank you so much guys. Your input means alot. I feel very grateful towards you. I want you to know that. I'm learning so much.

    Yes you are right about the liion setting doing a short absorb.

    I have to put it on normal charge + recondition mode to get it to go into long and slow absorption. But it's a bit scary because I fell asleep and woke up to a high battery Vdc. For some reason the battery charger was charging at 14.4vdc and then suddenly switched itself to 14.6vdc. Luckily I caught it just in time.

    What I did was put the battery on charge at 2 amps on liion setting. Waited over 24 hours to finish and then charged it on normal + recondition setting. Then I started seeing the voltage climb to 3.6vdc per cell, it was scary because one was lagging behind at 3.49vdc. It took around 5 hours to get the other cell to 3.6vdc which then I had to turn off the charger manually.

    I've only done that with one of the batteries so far. Should I do the rest in the same way? I've already charged all of them fully on the liion setting.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    It is too bad that there is not a built-in "crow bar" that will limit this overvoltage. This what the Schneider bridge does and will start turning off chargers if the BMS is not doing what it is suppose to. This  situation is very different as the LG is a bank of batteries manufactured as a bank and tested as a bank at the factory. Since you are the factory, you need to be careful !

    I am not going to guess for you and will let one of the guys that are use to this kind of thing.

    As I said before I am really surprised that Victron does not actively support this better on a battery with their name on it. Have you called them and asked for support or if there is a blog. I know they have a forum from when they were using the LG 48V Lion.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Yerf said:
    I have to put it on normal charge + recondition mode to get it to go into long and slow absorption. But it's a bit scary because I fell asleep and woke up to a high battery Vdc. For some reason the battery charger was charging at 14.4vdc and then suddenly switched itself to 14.6vdc. Luckily I caught it just in time.

    What I did was put the battery on charge at 2 amps on liion setting. Waited over 24 hours to finish and then charged it on normal + recondition setting. Then I started seeing the voltage climb to 3.6vdc per cell, it was scary because one was lagging behind at 3.49vdc. It took around 5 hours to get the other cell to 3.6vdc which then I had to turn off the charger manually.

    I've only done that with one of the batteries so far. Should I do the rest in the same way? I've already charged all of them fully on the liion setting.
    Do not use the Recondition function. It will charge up to a voltage of 16.2V and could reach this voltage in a short period of time after the battery is full which could easily damage your batteries.

    Although not ideal what you could do is charge each battery using Normal mode only. In Normal mode the charger manual says that it will do a minimum absorb time of 1/2 an hour then drop to a float voltage of 13.8V. If the charger has switched to float when you are about to take the individual voltage readings you could start a new Normal charge cycle and check the individual cell voltages around the 25 minute mark while the charger is still in the absorb phase at 14.4V to see if they are all the same, if they aren't run another normal cycle.


    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    > @karrak said:
    >
    >
    > Although not ideal what you could do is charge each battery using Normal mode only. In Normal mode the charger manual says that it will do a minimum absorb time of 1/2 an hour then drop to a float voltage of 13.8V. If the charger has switched to float when you are about to take the individual voltage readings you could start a new Normal charge cycle and check the individual cell voltages around the 25 minute mark while the charger is still in the absorb phase at 14.4V to see if they are all the same, if they aren't run another normal cycle.
    >
    >
    > Simon

    I actually got the reconditioning mode idea from Victron. But what you say makes more sense Simon.

    I will do that top charge in normal mode. Once all cells reach 3.6vdc I will consider them done and switch it off manually?

    What would be the ideal way?

    Regards
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    You have to laugh about the situation when I make the statement in bold in my previous post and you come back with but the manufacturer said ...

    I did think about using the Recondition mode as it is current limited to 8% of the charger output, in your case 0.56A. If all the cells in the battery being charged are full and balanced and not accepting any more current the balancing resistors will have to bleed all the current supplied by the charger to keep the battery voltage at 14.4V. In this situation they will be dissipating ~8W as heat. It is possible that they are capable of doing this but I didn't think it is worth the risk.

    Hopefully the person you have been corresponding with in Victron is aware of these technical details, one of my favourite sayings is "the devil is in the detail".

    Now that I am on my soapbox, one thing that I find really frustrating these days is the lack of technical details supplied with products like your batteries. It the good old days (there was no such thing, but it sounds good) there would usually be a page of very detailed technical specifications supplied with something like your battery. Information that would have made life allot easier would have been
    • What type of balancing does the battery employ (passive or active)
    • What voltage does the balancing start at
    • What is the maximum current that the balancing circuitry can bypass
    along with all the other information on maximum charge and discharge current, working temperature range etc. etc. all in one place so you don't have to wade through the whole manual trying to find the information. Enough of my rant.

    Ideally you would have something like a laboratory power supply where you would dial in the output voltage and current and hook it up to the battery and wait for the charge current to reduce to less than C/100. Then you would check the individual cells voltages. When they were all the same you could disconnect the charger.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Yerf said:
    I will do that top charge in normal mode. Once all cells reach 3.6vdc I will consider them done and switch it off manually?

    You could switch it off manually or wait for the charger to terminate the charge itself, I don't think it would make much difference.
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    Yerf is very lucky for your help Simon as he has said!  I hope when he gets this sorted out he can explain how Victron has let this get so borderline dangerous. At least the cells are some of the safest out there.

    The thing I have to laugh about is how "wild west" a major company like Victron is about a battery with their name on it. If I am wrong about this I would like to know.
    I have seen their gear over the years and since none of the major solar dealers I use has ever carried Victron in North America I really do not know how they have historically dealt with situations like this.
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Hey guys.

    Just for your info.

    I tried charging in normal mode and the battery went up to 14.4vdc. But the cells weren't equal yet, so left it on. And then I noticed the charger went up to 14.6vdc and was at 1.6 amps. Checked the cells, they were equal so shut it off.

    I'm just gonna do this to the rest of em and hope for the best.

    Regards
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Thanks for the update, it is always good to know how things are progressing.

    I don't think you need to hope for the best. What you are doing will balance the cells correctly.

    Regards
    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Thanks again Simon. All the best. Happy holidays to you all.

    Regards
  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Hey all

    Hope you had a good time over the holidays.

    I know I had a good time installing these batteries :smile:

    After finally balancing them all I installed the 4 banks.

    While the batteries were charging (the bank was at 56.8vdc absorption ) I noticed 2 of the batteries had a high cell voltage and one had a low cell voltage compared to the others.

    Nevertheless, the batteries charged to 100% without any errors, so I just left it like that for the time being.

    The bank went below 50% at one point and I thought I'd try the generator to see how they charge. They were charging at 400 amps! So I shut it down. I'm thinking of setting it at 200 amps. What do you guys think?
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Yerf said:

    While the batteries were charging (the bank was at 56.8vdc absorption ) I noticed 2 of the batteries had a high cell voltage and one had a low cell voltage compared to the others.
    It looks like there is a seven minute time difference between those two sets of readings. Were you using solar to charge the battery at the time and had the charger switched to float with a lower voltage when the second screen shot was taken?

    Are you using the Victron VE.bus BMS?
    The bank went below 50% at one point and I thought I'd try the generator to see how they charge. They were charging at 400 amps! So I shut it down. I'm thinking of setting it at 200 amps. What do you guys think?
    I agree with you. As you have four banks of 100Ah cells, charging at 200A is C/2 for your battery which matches the recommended charge current in the Victron manual. Faster charging will decrease the lifespan of the batteries. If you are charging from the generator I would use the Victron BMV700 to turn the generator on at an SOC of around 20%-30% and turn off at an SOC of around 40%-50% to limit the use of the generator.

    Simon

    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Sorry wrong screenshots. Here are the correct ones taken at the same time.

    Yes I am using the victron BMS.

    Thanks for the generator tip.
  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    My bmv just auto synced to a 100% from 40%?!

    The settings were at default except I changed the battery efficiency to 99 and pukherts to 1.05 like it said in the manual.

    Any ideas?
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Have you set up the Battery Capacity, Charged Voltage, Tail Current and Charged detection time on the BMV? See page 19 of the manual.

    The difference in battery voltages means there is a discrepancy between the SOC of the different batteries. The difference should get smaller over time as the batteries balance themselves.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Yerf
    Yerf Registered Users Posts: 65 ✭✭
    Yes I have set the battery capacity to 400 amps but didn't touch the other settings you mentioned as it says in the manual that default settings apply for lifepo4 and to only change pukherts and battery efficiency settings.

    DO you recommend changing the settings you mentioned? What would values be in your opinion?
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    The values are dependant on what you have set your charge controller to.

    The Charged Voltage needs to be slightly less say 0.1V-0.2V less that your Bulk/CV/Absorb voltage.
    The Tail Current needs to be set slightly higher than the charge current when it has tapered down just before the controller switches to float.
    I would set the Charged Detection Time to the minimum 1 minute.

    The SOC meter will reset to 100% when the battery voltage is still higher than the Charged Voltage and the charge current is less than the Tail current and these conditions must be met for at least the Charged Detection Time.

    What are your controller settings at the moment?

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    Just lurking... How did you get 400A @ 57V ?  A 25KW generator?  I would think the VICTRON BMS would limit overcharge but I guess it is not tied into the CC or charger on the Inv/Chg ?
    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net