I Like that You Refer back to Your Trusty 24 Volt System to Stay Running - I am Running 4 24 Volt Systems here @ Indian Creek Ranch of Southern California High Desert 🌴 💦 🌵 Aloha ⛩
With a few days of cycling under your belt you can look at the Victron "Trends" graph and see the voltage where there BMS disconnects up (because it won't look like a normal charging curve, the voltage will more or less shoot straight up and then oscillate around the charge target). Unfortunately Victron only keeps a few hours of high-resolution data, but you can still get an idea about its balancing progress. The voltage it gets to before it suddenly shoots up should (theoretically) get better and better each day when it is making progress balancing, until eventually it stops shooting up vertically at the end and just looks like the normal exponential ramp. An oscilloscope is really needed to see the BMS disconnect because the whole thing happens in about half a second... the voltage overshooting and then returning to the charge target. That process is less than a second. But will keep happening if there is an inverter load. Volt meters and the Victron monitoring display average the voltage over a second or two and can't catch it completely. Another way is to look at the "History" tab of the Victron charge controller. It shows the "max" battery voltage each day. In a properly working system that fills up to 100% that voltage will be precisely the charge target. In a system where the BMS disconnects, that voltage will be significantly higher and also vary a lot each day because of the averaging it does. In your video at 7:59 we can (barely) see the Battery "max" and it does seem to be bouncing around a lot. Though that might also be due to you changing settings. -Matt
I would imagine you will still get bms disconnects at 56.8 if it's partly cloudy. It's the way you approach a full charge that matters. If you approach it slowly you'll get a lower voltage and less overshoots and disconnects. For balancing you could lower the current from the charge controller and length the absorption time. I think maybe it's easier for the charge controller to regulate the current. For balancing you want to limit it to around what the BMS balancing current is and just let it sit there until the current tails off. Afterwards I would then set the charge voltage at something like 56 volts if that's as stable. The lower the charge voltage the longer absorption time needs to be to balance the cells. If you are reaching full charge everyday then you can afford to lower the voltage and have slower balancing. If that achieves stability you can work with it. Even at 55.2 you should get some balancing.
I do love solar it is fun for sure.. people have asked me if I'm dooms day prepping LOL. . I say no .I just prefer to pay for my electricity in advance 👍👍💪💪
If the battery is out of balance, I t will take about a week to get it to balance with the internal BMS. It usually has a .05 A or 50 Milliamps of current. Thanks a lot for a 90A cell.
That method will likely work it takes time but the thing is you can't see if the high cell is what shuts you down when the bms sees a single cell overvoltage it will stop the charge until the balancer tries to equalize the other cells sometimes it may take a week on CV depending on the bms and how far apart the cells are and as the battery gets older the spread will more every year. Sooner or later it is just easier to choose a lower voltage balance point. Cheers and have fun
@@kevinroberts781 the BMS maybe working as intended but it can still be causing the problem. As far as I could see it was the charge controller causing the BMS to shutdown that caused the problem. If I look in the log of my shunt I can see the same thing by looking at the max and min voltages recorded which are higher and lower than they should be. I also watched it happening on the victron app. I had to reduce the absorption voltage to the equivalent of 55.6 volts and creep it up from there. I'm still only on 56 volts equivalent I think. At the victron default I was still seeing bms disconnects.
Why don't you charge the battery until it switches off? And then unscrew the battery case cover and measure the individual cell voltages with a multimeter?
If it doesn't work, why not build another 48v bank with the previous 4 batteries you had with the balancer on and try and isolate the problem to the battery, charge controller or inverter?
I’m still unclear why you’re not using the rated charge voltage of 54V for that battery. My 48V batteries reach 98% SOC at 54V, which should be sufficient. If you look at the specifications for large server rack batteries, such as the Pylontech US5000B, they have charge voltages between 52.5V and 53.5V. When the BMS stops charging, it indicates that one or more cells have exceeded 3.65V. An immediate charge disconnect could cause a voltage spike in the inverter.
The inverter doesn't know nor does it care if your batteries are balanced or not. You have bad component in that inverter that is throwing off the voltage measurement.
That was great. So many with great information and ideas . This is a 48V server rack battery !. Take the screws out and open the lead. you should get to the cells and test each one for voltage. use same meter. right down each cell voltage as you go along. Usually when writing face up the top lid comes off and you can meter the cells. All you have to do is test all cells for voltage . POP the LID ! Your multimeter is best tool to have. Also put meter direct on battery terminals to test. test all connections. 1 step at a time. right down each reading. This is not hard but will take some time.
I believe your battery cells are only balanced at the top of the charging cycle. I fitted an active balancer to my 2 systems, both 16 cell 200 amp series wired battery. The active balancer balances all the cells against each other through out the battery voltage range. Where as your bms I believe ony balances at top of charging range. I fitted these balancers 4 years ago and the cells have never been out of balance. They come from 0.5 to 5 amp balancers.
@@noelroberts8082 If this is a balance issue this can be added inside the battery. waiting to see if this is balance issue or inverter problem. we will know with more testing. great point with the active balancer. we could have a bad BMS as well. the balance part . open the lid and test the cells .
Generally speaking there is no point trying to balance the cells below the exponential portion of the charge curve. The reason for this is that minor differences in cell construction willl cause the cells to have slightly different offsets in the flat portion of the charging curve. So attempting to balance there will actually take the cells out of the balance instead of put them into balance. Balancing is usually only done in the exponential part of the charge curve for this reason. Usually above 3.45V/cell. In the exponential part of the charge curve the BMS can tell which cells need to be worked on with complete certainty. Active balancers do generally work because they will correct for imbalances they produce earlier in the curve when they get to the later portion of the curve. And they work quite well. But if the BMS is doing its job an active balancer is unnecessary. The biggest reason for a BMS not doing its job properly is typically the BMS and charge controller configurations not being compatible with each other and leaving too little time for the BMS to balance cells. 99% of the time its human error because people think charging to voltages like 3.50V/cell or 3.55V/cell damages a LiFePO4 battery. In daily-cycling (solar) systems, charging to 3.55V/cell does not damage a LiFePO4 battery. The battery just doesn't stay up at that voltage for very long before dropping to the Float (3.35V to 3.375V/cell). Once in good balance, 30 minutes of hold at the charge target is all you need to keep the battery in perfect balance. (LiFePO4 batteries work very differently than NMC and NCA lithium chemistries). -Matt
Well, Matt, please explain where the exponential is on a liFPO4 since it doesn't have an absorbtion voltage. Exponential is only on lead acid, NiMY and NiCD because battery resistance is quite hight on these types so need an absorbtion voltage to complete the charge. Where as LiPO4 has a low resistance therefore does not need an absorbtion charge so no exponential. This is why an active batter balancer is better for LiPO4 than a fixed voltage balancer. By balancing throughout the voltage range there is less to do at max voltage.
@@noelroberts8082 The absorption setting for LiFePO4 is to give the BMS time to balance the cells, not because the cells need absorption. The cells are fully charged once the current tails off (usually less than 30 minutes upon reaching the target voltage, depending on the C-rate). If you have the current-tail turned on, the battery might not have enough time to balance its cells. So a fixed "absorption" time of between 30 minutes and 2 hours is recommended. The exponential portion of the charge curve occurs roughly above 3.45V/cell when charging at 0.2C. You can readily see the curve go exponential on any charge curve for LiFePO4 so just google it. The same thing happens on the discharge curve once the battery heads south of around 3.1V/cell. The reason BMS's balance batteries in the exponential portion of the charge curve is because slight differences in cell construction get swamped by the curve. Whereas if the BMS tries to balance the cells below that point, minor differences in cell chemistry result in voltage offsets at thet same SOC that cause the BMS to make the wrong balancing choices. -Matt
Its ok that You try balancing like this,but if the BMS`s balancer is slow,it might take even more time,its not ideal that you cannot see all the different cell voltages somehow,with a comunication cable and a laptop,or in some any other way.Becouse than You could see if they are still balanced at 56.8V,at the top end You can see always when they are unbalanced.My bateries had 200-250 milivolt deviation on cells(some at 3.3V some at 3.550 or higher....) and it took 6 days to balance them with a 4amp active balancer.I took my battery apart and lost varanty too,but the varanty guys didnt wanna do anything,and having that imbalance is not just bad becouse i cannot use the battery capacity,but dangerous for the cells long term.Having them in balance is key factor.Watch it when You charge them and they stay on absorb how your charge controler and batery behaves,maybe the leds will show some unusual thing,like turning all off,and just the "working" one led will beep.I dont know power queen batery bms-s,but i wish You good luck,hope it solves the problem,and its not Your inverter.Or maybe the battery is balanced,i think maybe that is even the better scenario,and the inverter is faulty.The battery is more expensive,and harder to deal if it could go in unbalance(weak bms let that happen,good ones never let the battery be imbalanced)and you just get another victron inverter to avoid future problems :) Aloha!
I Like that You Refer back to Your Trusty 24 Volt System to Stay Running - I am Running 4 24 Volt Systems here @ Indian Creek Ranch of Southern California High Desert 🌴 💦 🌵 Aloha ⛩
Good Evening ! Beautiful looking day. TAKE CARE..
With a few days of cycling under your belt you can look at the Victron "Trends" graph and see the voltage where there BMS disconnects up (because it won't look like a normal charging curve, the voltage will more or less shoot straight up and then oscillate around the charge target). Unfortunately Victron only keeps a few hours of high-resolution data, but you can still get an idea about its balancing progress.
The voltage it gets to before it suddenly shoots up should (theoretically) get better and better each day when it is making progress balancing, until eventually it stops shooting up vertically at the end and just looks like the normal exponential ramp.
An oscilloscope is really needed to see the BMS disconnect because the whole thing happens in about half a second... the voltage overshooting and then returning to the charge target. That process is less than a second. But will keep happening if there is an inverter load.
Volt meters and the Victron monitoring display average the voltage over a second or two and can't catch it completely.
Another way is to look at the "History" tab of the Victron charge controller. It shows the "max" battery voltage each day. In a properly working system that fills up to 100% that voltage will be precisely the charge target. In a system where the BMS disconnects, that voltage will be significantly higher and also vary a lot each day because of the averaging it does. In your video at 7:59 we can (barely) see the Battery "max" and it does seem to be bouncing around a lot. Though that might also be due to you changing settings.
-Matt
I agree the bms is only doing what it is programmed to do.
I would imagine you will still get bms disconnects at 56.8 if it's partly cloudy. It's the way you approach a full charge that matters. If you approach it slowly you'll get a lower voltage and less overshoots and disconnects. For balancing you could lower the current from the charge controller and length the absorption time. I think maybe it's easier for the charge controller to regulate the current. For balancing you want to limit it to around what the BMS balancing current is and just let it sit there until the current tails off. Afterwards I would then set the charge voltage at something like 56 volts if that's as stable. The lower the charge voltage the longer absorption time needs to be to balance the cells. If you are reaching full charge everyday then you can afford to lower the voltage and have slower balancing. If that achieves stability you can work with it. Even at 55.2 you should get some balancing.
I do love solar it is fun for sure.. people have asked me if I'm dooms day prepping LOL. . I say no .I just prefer to pay for my electricity in advance 👍👍💪💪
👌🤙
bob i sure hope it works because i have two of them a 5000 and 2500 reliables
If the battery is out of balance, I t will take about a week to get it to balance with the internal BMS. It usually has a .05 A or 50 Milliamps of current. Thanks a lot for a 90A cell.
That method will likely work it takes time but the thing is you can't see if the high cell is what shuts you down when the bms sees a single cell overvoltage it will stop the charge until the balancer tries to equalize the other cells sometimes it may take a week on CV depending on the bms and how far apart the cells are and as the battery gets older the spread will more every year. Sooner or later it is just easier to choose a lower voltage balance point. Cheers and have fun
His BMS isn't the issue. It's clearly in the inverter. His inverter and BMS don't talk to each other
@@kevinroberts781 If it is an inverter problem he could just turn off the solar charging and see if the inverter still gives him the error.
@@kevinroberts781 the BMS maybe working as intended but it can still be causing the problem. As far as I could see it was the charge controller causing the BMS to shutdown that caused the problem. If I look in the log of my shunt I can see the same thing by looking at the max and min voltages recorded which are higher and lower than they should be. I also watched it happening on the victron app. I had to reduce the absorption voltage to the equivalent of 55.6 volts and creep it up from there. I'm still only on 56 volts equivalent I think. At the victron default I was still seeing bms disconnects.
@@matthewwakeham2206 I agree
Why don't you charge the battery until it switches off? And then unscrew the battery case cover and measure the individual cell voltages with a multimeter?
If it doesn't work, why not build another 48v bank with the previous 4 batteries you had with the balancer on and try and isolate the problem to the battery, charge controller or inverter?
Aloha Bob, I hope this works 🙏 wish you had more sun for this test but it's a good place to start.
I would replace the BMS with a jk smart . Then you will have Bluetooth on the battery and can see and set everything
I’m still unclear why you’re not using the rated charge voltage of 54V for that battery. My 48V batteries reach 98% SOC at 54V, which should be sufficient. If you look at the specifications for large server rack batteries, such as the Pylontech US5000B, they have charge voltages between 52.5V and 53.5V. When the BMS stops charging, it indicates that one or more cells have exceeded 3.65V. An immediate charge disconnect could cause a voltage spike in the inverter.
👍😊
The inverter doesn't know nor does it care if your batteries are balanced or not.
You have bad component in that inverter that is throwing off the voltage measurement.
The inverter cares if the input voltage goes above a predetermined value, which is almost certainly what is happening here.
That was great. So many with great information and ideas . This is a 48V server rack battery !. Take the screws out and open the lead. you should get to the cells and test each one for voltage. use same meter. right down each cell voltage as you go along. Usually when writing face up the top lid comes off and you can meter the cells. All you have to do is test all cells for voltage . POP the LID ! Your multimeter is best tool to have. Also put meter direct on battery terminals to test. test all connections. 1 step at a time. right down each reading. This is not hard but will take some time.
I believe your battery cells are only balanced at the top of the charging cycle. I fitted an active balancer to my 2 systems, both 16 cell 200 amp series wired battery. The active balancer balances all the cells against each other through out the battery voltage range. Where as your bms I believe ony balances at top of charging range. I fitted these balancers 4 years ago and the cells have never been out of balance. They come from 0.5 to 5 amp balancers.
@@noelroberts8082 If this is a balance issue this can be added inside the battery. waiting to see if this is balance issue or inverter problem. we will know with more testing. great point with the active balancer. we could have a bad BMS as well. the balance part . open the lid and test the cells .
Generally speaking there is no point trying to balance the cells below the exponential portion of the charge curve. The reason for this is that minor differences in cell construction willl cause the cells to have slightly different offsets in the flat portion of the charging curve. So attempting to balance there will actually take the cells out of the balance instead of put them into balance.
Balancing is usually only done in the exponential part of the charge curve for this reason. Usually above 3.45V/cell. In the exponential part of the charge curve the BMS can tell which cells need to be worked on with complete certainty.
Active balancers do generally work because they will correct for imbalances they produce earlier in the curve when they get to the later portion of the curve. And they work quite well. But if the BMS is doing its job an active balancer is unnecessary.
The biggest reason for a BMS not doing its job properly is typically the BMS and charge controller configurations not being compatible with each other and leaving too little time for the BMS to balance cells. 99% of the time its human error because people think charging to voltages like 3.50V/cell or 3.55V/cell damages a LiFePO4 battery.
In daily-cycling (solar) systems, charging to 3.55V/cell does not damage a LiFePO4 battery. The battery just doesn't stay up at that voltage for very long before dropping to the Float (3.35V to 3.375V/cell). Once in good balance, 30 minutes of hold at the charge target is all you need to keep the battery in perfect balance. (LiFePO4 batteries work very differently than NMC and NCA lithium chemistries).
-Matt
Well, Matt, please explain where the exponential is on a liFPO4 since it doesn't have an absorbtion voltage. Exponential is only on lead acid, NiMY and NiCD because battery resistance is quite hight on these types so need an absorbtion voltage to complete the charge. Where as LiPO4 has a low resistance therefore does not need an absorbtion charge so no exponential. This is why an active batter balancer is better for LiPO4 than a fixed voltage balancer. By balancing throughout the voltage range there is less to do at max voltage.
@@noelroberts8082 The absorption setting for LiFePO4 is to give the BMS time to balance the cells, not because the cells need absorption. The cells are fully charged once the current tails off (usually less than 30 minutes upon reaching the target voltage, depending on the C-rate).
If you have the current-tail turned on, the battery might not have enough time to balance its cells. So a fixed "absorption" time of between 30 minutes and 2 hours is recommended.
The exponential portion of the charge curve occurs roughly above 3.45V/cell when charging at 0.2C. You can readily see the curve go exponential on any charge curve for LiFePO4 so just google it. The same thing happens on the discharge curve once the battery heads south of around 3.1V/cell.
The reason BMS's balance batteries in the exponential portion of the charge curve is because slight differences in cell construction get swamped by the curve. Whereas if the BMS tries to balance the cells below that point, minor differences in cell chemistry result in voltage offsets at thet same SOC that cause the BMS to make the wrong balancing choices.
-Matt
Excited to see what happens. Aloha
I have two of those inverters and both of them do the same thing it's the inverter trust me
set float to 55.2 volt
good job Bob
Its ok that You try balancing like this,but if the BMS`s balancer is slow,it might take even more time,its not ideal that you cannot see all the different cell voltages somehow,with a comunication cable and a laptop,or in some any other way.Becouse than You could see if they are still balanced at 56.8V,at the top end You can see always when they are unbalanced.My bateries had 200-250 milivolt deviation on cells(some at 3.3V some at 3.550 or higher....) and it took 6 days to balance them with a 4amp active balancer.I took my battery apart and lost varanty too,but the varanty guys didnt wanna do anything,and having that imbalance is not just bad becouse i cannot use the battery capacity,but dangerous for the cells long term.Having them in balance is key factor.Watch it when You charge them and they stay on absorb how your charge controler and batery behaves,maybe the leds will show some unusual thing,like turning all off,and just the "working" one led will beep.I dont know power queen batery bms-s,but i wish You good luck,hope it solves the problem,and its not Your inverter.Or maybe the battery is balanced,i think maybe that is even the better scenario,and the inverter is faulty.The battery is more expensive,and harder to deal if it could go in unbalance(weak bms let that happen,good ones never let the battery be imbalanced)and you just get another victron inverter to avoid future problems :)
Aloha!
Aloha Bob awesome video
Aloha 🌺
Need drain battery every two months20 %
I have always wondered why you have multiple systems that charge individual appliances. Why not have one system to power your whole house?