Active Balancer: always on or only above 3.45V?
ฝัง
- เผยแพร่เมื่อ 29 ส.ค. 2023
- I know this is a controversial topic but I think we all have something to say here and have made different experiences. So please share!
Gobelpower has built-in an active balancer into their battery which is great. It is turned on and active all the time though. From many tests we have done here on the channel, this causes an imbalance in the battery pack. It shifts energy between cells and when you fully charge again, the balancer does not have enough time to restore the initial top balancing. This especially happen during longer periods of no full charge (eg. winter or bad weather). Most likely, one of the cells will be peaking and causing the BSM to disconnect even before your reach you charge voltage.
So, here is the test:.
1. Fully charge and top balance the battery (we go with 58.2V and ~30mV deviation)
2. Fully discharge until one cell hits 2.5V
3. Recharge the battery again from solar over 2-3 days.
4. Compare the deviation to the initial top balancing achieved under 1.
And we are doing the exact same test twice, first one with the active balancer connected all the time, second cycle without any active balancer connected at all.
What will happen? Will we see a higher deviation with the balancer active all the time or does it not matter and it actually helps balancing the battery at the top and deviation is getting smaller?
Keen to read your thoughts about this experiment. I know, it's only one cycle but it still should give us an indication and direction where the journey will go.
Fantastic breakdown on how these balancers work from @StuartPittaway
• How Does An Active Bal...
12% discount on the Gobelpower Battery for a limited time! All info here on my website:
off-grid-garage.com/li-ion-ba...
Gobelpower BMS Test and Review:
off-grid-garage.com/battery-m...
Pace BMS Test and Review:
off-grid-garage.com/battery-m...
ZKE-Tech Battery Tester with PC software:
off-grid-garage.com/measureme...
Please subscribe and join me on my journey of the Off-Grid Garage!
For more information, please check out my website with links to all the batteries, materials, devices and products I have used in my videos: off-grid-garage.com
The 'Buy me a Beer🍻'-link is here (on the website, scroll all the way down):
off-grid-garage.com Thank you for your support!
You can also click on the Thanks button under the video to support the channel. Thank you, if you have already!
T-Shirts, caps, mugs and more in our merch-shop:
off-grid-garage.com/merch
**** These are the batteries you are looking for ****
🔋➡EVE LF280/304 LiFePO4 cells 3.2V/280+304Ah: off-grid-garage.com/batteries/
🌞➡Solar charge Controllers, Inverters and accessories: off-grid-garage.com/solar-cha...
🔧➡Tools, Analysers and Testers: off-grid-garage.com/equipment/
📈➡ Battery Tests and Results: off-grid-garage.com/battery-d...
🔌➡Batteries, charger and accessories: off-grid-garage.com/batteries-3/
⚡➡Cables, Inverters and Electronics: off-grid-garage.com/electronics/
The Sunfunkit DIY battery. Great product to build your own 12V battery. Use OGGA at checkout to get 6% off. More info here: off-grid-garage.com/diy-kits/
Video review: • The Sun Fun Kit Experi...
Best prices for Victron gear in Australia: Auto Parts Co has you covered.
www.autopartsco.com.au/?pd_pr...
(they have 4WD parts and accessories too)
Take a live sneak peek at the Off-Grid-Garage in the Victron VRM World (link at top of the page):
off-grid-garage.com/victron-c...
Here is my Tesla referral link, if you want to buy any Tesla accessories or even order a vehicle. This is a great way to support my channel. Thanks a lot to everyone who have used it so far 🙏 www.tesla.com/en_au/referral/...
My other (German) TH-cam channels:
/ offgridgarageaustralia
Please don't leave any links in your reply. TH-cam will not publish your comment in this case.
If you decide to send me an email, I really appreciate your time and effort for doing this. However, please consider leaving a comment under my videos instead as I may not be able to reply to your email due to the sheer number of messages I'm receiving. - วิทยาศาสตร์และเทคโนโลยี
![[LIVE] : ONE ลุมพินี 68 | คู่เอก "พระจันทร์ฉาย vs โจนาธาน"](http://i.ytimg.com/vi/F3IM3w-XchI/mqdefault.jpg)
Hello Andy, your videos are a wealth of information for solar enthusiasts. Thanks for your very informative videos!
Thanks Andy 👍👍
I’m definitely with you, in the camp of only balancing at the top. I believe constant balancing will mess with the balance. (I had exactly this previously in my old caravan set up before I added a JK BMS. The JK is set up now so it only balances above 3.44v, and my balance has been perfect since then, helped of course that it’s only a 4S/ 12v system.)
Thanks for sharing, Dave. The JK is great. always was and always will be 😄
Thanks Andy,
I have DIY 16-S packs and when I built them, I let the balancer run all the time, that was not ideal. Now just when the cells come up above 3.35 seems best in my system. A good top balance starting point helps a lot, since the cells at 280Ah will take ages to self top balance. Thanks again.
In the "flat" part of the discharge curve, the cells will not show significant voltage differences anyway, even if they were 20% difference in SOC. They only show significant differences in voltages near the top and bottom of the curve. Set the voltage delta to 20mv and use balance "on" at 3.2 to 3.35 volt. If you use too high a set point - like 3.45 but one cell is lagging, the BMS will never balance the pack before cutting off charging.
With 16s (48v) packs, the chances of one low cell is significant.
I have also noticed, a new pack may require a lower start point for balancing for a number of cycles, then as the pack becomes better top balanced, you can change the start balancing voltage higher if you like, since the cells remain closer together in top voltage as the BMS balancer does it's job.
Hi Andy, you are on the right track. My battery banks show higher variances in certain SOC ranges. One bank is in the 30% range and in 60% up to 25mV divergence. If now an active balancer permanently balances and the battery stays overnight in this SOC, the TOP balacing is gone.
Yep, exactly what I have tested and experienced many times. It through me off when I installed the active balancer in the Battery 1.0 and with the next full charge, the cells were all over the place 😄
My house is off-grid with several 230ah packs. I record each cell voltage into a database every 5 and graph into Grafana. I have run several test since some packs are of the same age. A top balance with no balancer is only good for about 4 months of daily cycles down to 70% SOC (my daily usage). After that 4 months, the top balanced pack with no balancer had the same above 3.45+v cell drift as a pack that I had not top balanced at all, and I just connected up the balancer and put it into service. As for deep cycling, I have drained the packs to 3v per cell (48v) and the inverters are configured to turn off then. No cells fell under the 2.50v LVD, but all started to drift. When charged back up, all cells drifted a lot above 3.45v, but the balancers fixed that after a few hours. Based on the off-grid house application, where I am fully charged, or in the flat part of the curve for 99% of the time, I am not seeing an issue with having the balancers on all the time. If I am loosing capacity, it's only showing up in the bottom 10% of SOC, and for the few times a year when I hit that, I am ok with it. I don't have a way yet have a way to turn off and on the balancers based on voltage (it's in the works), so I can't test your theory. You might be right in a pack that is deep cycled all the time and spends more of it's life near 10% SOC.
That is very interesting. Thanks a lot for sharing your results and experience!
Thank you for helping me to understand the differences between products and voltages with effective habits. 🤔. ❤❤❤😮
You will like the next video then...
2x 5A active balancers (capacitive) in parallel. Always on. For over 2 years now without any issues.
- 12V in-car system
- 13.8V charged voltage
- SoC ranging from 20-100%
- 240W PV charge
- 400W DC/DC charge
Works a treat.
Hi Andy, did all this tests a couple of years ago and you are spot on. Cells have a deviation in inner resistances, this vary with SOC. The cell difference you see highly depends on the current and direction of the current while charging / discharging. The same cell may be lower than the others in the flats or while drawing energy and may run off while chharging. Regardless of balancing you can have a high cell at 80% while charging, that is overtaken later by the lower cell at 90% SOC that becomes the high cell then. This cells are crazy, it makes no sense to balance them early, only on the last 3-5% and only while charging. As soon as you turn off the current they seem balanced (after draining the surface charge with a small load for a short time).
Thank you. Yeah, from my testing, balancing LiFePO4 only makes sense above 3.45V. As you said, below that it's useless...
@Off-grid Garage, Thanks for your many intuitive, educational, technical & engineering videos. Kindly also carry out extensive tests with focus on cycle life at the specified DoD, product price, performance including calendar life, durability, charge rate, discharge rate, operating temperature range, trade-offs/compromises or competitive disadvantages, etc, and do a Cost Benefit Analysis of the S-volt LFP cells as compared to the more popular Eve & CATL cells. Endeavor to acknowledge this comment.
Thank you but I'm not a laboratory and don't have the appropriate test equipment and environment to carry out such long term tests. It's just me in my garage 🤷♂️
yes, Andy, I think you are absolutely right. it may be possible to leave it as an active balancer all the time if there are real conditions for it. for example, the battery stays fully charged for a long time. however, the battery is constantly used as soon as it is charged and immediately discharged, the next day it may not even have time to charge and it is discharged again, it is really likely that the top balance will be bad. I'm almost sure that your video will prove it.
Thank you for your thoughts. Video is coming tomorrow!
I’ve seen those comments about balance always on works for them. Well, it depends on the user’s pattern of use whether or not this works. However, if you balance only above the knee, it works without a doubt for any pattern of use. No worries
I agree, with only 2 amps if one were to use the batteries daily it would have very little effect in my opinion.
I only use my batteries for a backup generator and after several days of not using them they automatically balance themselves at the level they settle at which is around 3.3 volts
New batteries that are super well matched and consistent with each other, its probably its not a big deal. But older batteries with different/larger varying internal resistances it makes more sense to only balance at the top where it's needed.
Yep, that's a good point. Cell quality and how well they are matched plays a big role. But also, how well will they they still be matched in 5 years time?
@@OffGridGarageAustralia Good to see you also think about the future, most just "revieuw" as new and tend to forget how it works over time👍
No, it doesn't work at the lower end. You can see an all-time on balancer with voltage difference activation as an universal solution. In summer when voltage near top it does top balancing. In winter when voltage is mostly at the lower end it does bottom balancing.
This all helps to get most out of the cells and also helps to age the cells more equal.
As you observed at 80% SOC there was 2mV deviation so the active balancer should be doing nothing. Will follow with interest as a learning opportunity. I like the look of the Gobel Power battery so want to learn as much as I can about it before taking the plunge in my own povo pack offgrid shed (Im on 12V at the moment lol)
Hi Anvoltage the stop when done!!
NO Issues!!dy
That Is Why I Like The JK BMS
Balancer is set to balance at a certain
I suspect Andy, that with the Active Balancer left ON during the discharge and recharge cycle, the amount your top balance will be effected will depend on how well the discharge curves of your cells match, and how fast you discharge and recharge. That being: If you discharge at a rate higher than the active balancer can compensate for imbalances through the cycle, then you will end up with cell imbalance. However if you discharge and recharge at lower rates at which than the balancer can compensate for, then you will not experience premature low or high voltage cut-offs due to cell imbalance, and having the balancer ON all the time might result in a slight increase in overall capacity as it's effectively both top and bottom balancing.
Correct.
yes this i true
Correct
No charging of Teslas allowed then all will be good.
We do the test twice, exact same discharge and charge method but one test with balancer on all the time, the other test with no Ab at all.
Ah, yes, that comment about internal resistance at 7:15 needs a little extra explanation. The internal resistance will effect the measured voltage of the cell under load. This is what the balancer uses to determine whether balancing is needed, but it is getting faked out because the number of electrons being moved is still the same for all cells regardless of the internal resistance. The internal resistance just drops the energy of the electron contribution from the cell, it does not change the number of electrons (amp-hours) going into or out of the cells.
So what should happen is that IF the balancer kicks in during charging or discharging and decides to move charge between cells, that should take the battery out of top balance. But the question then becomes... how quickly and by how much? Obviously if the battery is ever charged to full again the balancer will re-top-balance the cells, and it MIGHT be able to do so in one charge cycle if the intervening cycles didn't cause the cells to get too out of balance.
This is why I think a test like this is going to be very, very sensitive to charge and discharge current and also requires that the batteries not be fully charged (which would re-top-balance the cells and mess up the test).
* I suspect that at low C rates the balancer just won't kick in all that often, and at higher C rates it will. So the test really has to record just how often the balancer kicks in.
* And I suspect that if the balancer is active while the battery is being discharged and the SOC is below around 15%, it will mess up the top balance even worse as it gets into the exponential curve on the empty side of the battery. The internal resistance will have an outsized effect on measured voltage under load when the SOC is below 15% or above 85%.
-Matt
Thanks, Matt. As always great deep dive and explanation. Appreciate it.
I have no idea but the conclusion will no doubt be interesting.
Tomorrow!
Coming from the world of lead/acid into NMC then LIPO, I've seen many folks get a bit confused. A full time AB may be more applicable to NMC chemistry because of the charge/discharge curve characteristics. My experience dictates just top balancing is best. Also helps identify problem cells in the mid level range where a full time balancer may mask issues.
What balancer do you recommend for NMC? Gobel?
Yes, absolutely, the AB will work fine all the time with NMC batteries. Linear voltage/SOC curve.
100% correct, especially for the bad cell detection.
I use a 12 x 12v RollinCart battery 24v system with the HA01 type of balancer in each pair.
They balance all the time, but only ever balance at 10% or 90%.
I have not noticed any issues with the top balance, all the balancers stop flashing within 1/2 an hour of hitting full charge, all the batteries are within very close voltages.
I think it may cause a big issue with batteries that sit at a very low SOC for ages, but because the balancers have such a small amperage, the deviation doesnt get too bad in normal operation.
Just my observations for the last 3 years.
Like you said it depend on the conditions and all cells are slightly different, all my BMS are JK with 2A balance and last months I have done tests with 3.2 - 3.4V and just started with 3.45V and all my batteries have grade A matched cells and all have become unbalanced. First reason, I have an Deye inverter and charge in lead acid battery mode and I cant program absorption time so at 3.4 it will only balance a few minutes until it reach my absorption of 3.45V, because worst cells have higher voltage on charge this means the bad will reach balance voltage first and will discharge to charge good ones, with 3.2V balance in one month the pack lost about 15Ah, so with your 3.45V it won't balance but won't discharge the bad. Second reason for the unbalance since my cells are true grade A, I charged 490Ah with 110-90A and the higher the A the bigger the power differential and balance ports don't measure the exact voltage and I had one port with a difference of 22mV, the cell was good but because read higher voltage won't fully charge. Increasing Con. Wire. Res. Settings allow to drop the voltage of the cell and put the correct value and now I have all cells fully charged. This difference in value can be many things, BMS, bad contact on bus bar, the terminals are bad so can be the also there, oxidation, bad wire, wrong torque, etc so first fix those and after adjust the parameter. The third thing I have done was lower the charge to 60A and probably will go to 45A, in summer we have many sun hours so we can charge slower, in winter I will increase the value. Forth the equalization parameter has a time so I adjust it to 1 day and 2h/2 and 55.2V, it only work even days but at least now at least do some balance, everyone talk about absorption time that I didn't realize that what I need was the literal meaning of the parameter.
I have my Active Balancers always on. And if anything it has improved my top balancing. Over time.
so i guess we need to confirm it with a last test.
Works for me too
That's good that it works for you. When I was starting out with reclaimed cells and built battery packs, some the same lot, some mixed, the always on worked well. When I set up my fully off grid and bought all from the same lot(100+kwh) the always on was giving me an erratic deviation whenever I would fully charge. Turned out, I had a cell that had issues I could not identify until the balancer was off. Both modes are probably ok but for diagnostics at a glance, I leave only on top balance. Cheers
@@babaluto Well I dont put much faith into reclaimed cells . Like they say , Its like a box chocolates, You never know what you will get. But sometimes you get good stuff also.
Have tried just about everything since going lifepo4 in 2014. Always-On active balancers threw my balance out and created a bigger challenge to deal with at the top. I went back to ev power passive balance boards. About to try active balancing again with the voltage relay control.
Thanks for sharing.
The voltage control relay works a treat. Since we installed this in the QSO battery, no problem any more. Perfectly balanced down to 10mV
Hi Andy i think its more simple than it looks, if the batteries quality is high after a top balancing they will keep balanced on all the states of charge, if they top or bottom umbalance some cells have higher internal resistance or slighty different capacity
Yes, cell quality and matching play a role as well as usage pattern with large or small currents.
In the short term:
The underlying assumption to NOT use a balancer below, say 3.45V, is that, IF we view each cell independently, it should ideally precisely trace the discharge/charge curve several cycles. However, balancing all the cells would force (in the pack situation) each cell to move another (unnatural) charge curve. Thus, it would prematurely ruin the top balancing and the small delta voltage. I agree with your arguments.
In the long term:
You can expect each cell to change with time; thus, using a capable balancer above 3.45 V makes sense to keep the pack at top capacity over its life since discharge/charge traces will change over time. Cheers.
Thanks for your comment and explanation. Aging of these cells as also a good point. Even they are perfectly matched today, they may not be in 5 years and hence the balancer makes sense.
my 600ah 12.8v home built I have my balancer on a switched circuit running off my victron mppt 75/15 charge controller at 13.7v on after Andy said this could be better. I have over 120amps of charge controllers going to the thing. 2 100/30's, 1 100/50 and the 75/15
I get about .02 volt deviation once it hits my 14.10v "full charge" voltage. With it always on I was observing about .04 volt. It's generally cycled from 50% to full most days.
Great, thanks for sharing.
Rain this afternoon, Andy down here. 1 degrees this morning. Done say it, I know how didI end up back here after living in Queensland. I'm having fun talking to Growatt here in Australia about my 2 x SPF5000ES. Oh, I did give you the full information.
thanks for the info
Hi Andy,
I really enjoy your videos and all the valuable information that you provide to this community, Thank you. 😀
I have a battery with 96 groups of cells in series with 4 cells connected in parallel for each group for a total of 396 volts, give or take a few volts. 2 of the 96 groups has only 25% capacity left after 4 years in service. What would work best in providing the most amp hours available?
Would top balancing and then turning off the balancer improve total amp hours available?
Or
Would leaving the balancer running increase total amp hours available?
Before you answer I would like you to refer to your Frankenstein battery as it also has mixed capacity cells.
Looking forward to your response, Willem
Thanks Willem.
That is a strange battery constellation you have. If 2 groups have only 25% capacity left, then your whole battery must show only 25%. Like the Frankenstein battery (and frankly all batteries), it all depends on this weakest cell in your pack.
I have a 16S 304 Ah configuration with JK BMS and Growatt as inverter/controller , All cells looked perfectly balanced with only 1 mV deviation at 3.35 volt and after 6 months I did a top balance and the deviation went over 150mV. I think is best to start the balancing process not before the keen part of the curve, at least with this confuguration.
Need a bit more detail about pack voltage and JKBms active balancer settings.. Sounds as though your system was not pushed to 56V pack for long enough frequently enough to allow incremental top balancing each day, then when you did push you pack voltage up you discovered your pack was unbalanced.
Yeah, 3.35V is still in the flat area of the curve, so it won't show any deviation. You need to be at least at 3.42V -3.45V to see the spread. Then the balancer should be active as well.
I DK. kan lav temperaturen på virke cellerne forskelligt, såsom max opladning + temperaturen og kvaliteten af batteriet , jeg synes at 3,4-3,450 volt for opstart af balance punktet, er ok, og burde og burde være passiv , ved fra 3,00 til 3,45 volt tak for dine gode videoer . Henrik
Hi, there's no need for none.
Rezone been you charge and discharge the battery probably every other day.
That is enough for balancing situations.
Thank you Sir.
What about the 4 months of winter where most of us are not able to fully charge and use the battery between 10-50%?
@OffGridGarageAustralia yes, that's a good tough.
This beautiful earth has two major elements.
SUN, and Wind.
This is one solution, Wind Generator
I've set mine at always balancing and whenever the cells are totally balanced, it does not work so it's still good actually. This way, there would a perfect balance most of the time for discharge and charging. Active balancers are not also overworked if the voltage difference is maintained because they would not have to much to catch up or nothing to work on even if you set it to always balancing.
Just because it is set to always balancing, it does not mean that it would always be working. It is just there as a setting so that whenever there will be enough voltage difference, the active balancer will fix it.
I've had an issue before when I had it on 3.45v where I was using a battery pack to power an electric bike. Did a long uphill trip and one cell group was way too discharged that it hit my UVP but the other cells were still really good. However, it is no longer balancing because everyone is below 3.45 so that's a problem.
I no longer encounter this problem when I set the balancing voltage at 2.6 so it's always balancing. Of course, I am also worried that the smart bms might be overworked and be broken overtime.
However, I noticed that even if it was set to always balancing, whenever the cells are balanced the smart bms do not work to balance the cells any longer. That means that it is totally better to have them on all the time than to leave them at the top.
You cant realy compare a E-Bike battery to a solar battery that is about 10 times the capacity.
Because if the balancer courses an unbalance of like 5% on a 20AH Ebike battery it can quickly be corrected on the voltage drop at a lower stage of charge on a 300AH the balancer has not enough power.
Same for charging an E bike battery that is charged from the grid has like 10hours of time for balancing, a solar battery maybe just 1 hour with the ramaining bit of sunshine.
The differents of the application are way to big to compare them.
My balance ON starts at 3.1v (and above) per cell. We need balance during discharge as well.
Thanks to Andy for suggesting JK, it works perfectly.
Why?
My balancer is always on, however I don't draw large loads, typically only 1000w, the battery is always balanced when I check.
I totally agree with you Andy.👍 The balance should only work in the upper range. The only thing I can imagine if you set the difference at 50 millivolts . In that sense, the balance in the middle would never work....🤔
Yeah, I think this is what Seplos tries to do now with their active balancer and having a 50mV setting. It starts balancing above 3V.
But then you can never get better than 50mV ever. Is that good enough? Probably...
@@OffGridGarageAustralia Yes, that could be the idea behind it. Of course, the battery is then not so well balanced in the end. But the Chinese often have different requirements than we do 😅
PS But when a Chinese electronic part giving smoke, is called a dragon Smoke.
Just as my interpretation 😅😜
Is smart active balancer the better choice? The 1a DALY smart balancer allows for start voltage and Delta parameters to be set.
Andy, Thanks for testing this!!
Suggestion: how about turning OFF the BMS balancer for the test, it balances by shunting (wasting) energy. The active balancer redistributes it within the pack (aside from a few milliamps consumed for switching). The only thing the on-board balancer can do when the active balancer is connected is to remove energy and interfere with the active balancer's mission.
After seeing how these capacitive balances work, IMO the 24/7 connected active balancer doesn't appear capable of messing up the balance as you've described in your notes. The only power it has to work with is the cell's differential voltage, not enough to push the state of balance during a charge/discharge cycle. Also per your notes, if the battery was left awhile at a lower SOC, say 35%, with the ActBal connected, wouldn't the cells be balanced at that 35% SOC? Is that not good to be balanced at the start of a recharge?
BTW, There's a really good video on TH-cam that describes/diagrams how these Heltec-style capacitive balancers work. (cant post link but exact title is: "Active Balancers - How Do They Work?" by Brad Cagle). You only need the first 2-3 minutes to get the picture.
So far, I could not see that the BMS balancer interferes with the active balancer. I had them always running together. The BMS balance current is just tiny in comparison to the AB current.
I always have the problem when charging up after winter and the battery did not get charged to a higher level for several months.
Even in the middle part, the delta can be 20-50mv with high currents coming from the battery. If I charge the vehicle for several hours on 2.4kW it pulls around 50A from the battery and the balancer will use the delta to transfer energy. Over time, this can be a lot of energy and as you said, it will balance the pack at this 35%. This is not good though as we top balanced the battery at 98% already. You basically can have only one SOC where the batteries are balanced.
Good old Brad. He's here on the channel as well and leaves comments sometimes.
Also watch the video from Stuart in the description.
Always balancing will indeed destroy the balancing when operating in the flat area and then repair the right balancing in the top area. So it will mainly just work for nothing. Indeed during the charge phase in the flat area, the voltage is not so flat because of the accumulation/absorption phases. So it will never stop to exchange energy randomly, function of which one absorb and which one accumulate at the current time.
So yes the result at the end will be the same but your balancer will work most of the time for nothing.
Fortunately into the flat area, the exchanged current will be very low because this current is proportional to the voltage deviation between cells but it will be waste of work.
The risk of large unbalancing will happen when the battery stays always in the flat area for long and is never fully charged or discharged regularly.
Sums it up perfectly!
Before balancers, I hooked 4s LiFePO4 up to a fully charged AGM battery to ballance them. Would be pretty cool to see you try that for the DIY community Andy. Drop it old school. No BMS, 3.6V, 100mV ripple, AGM charger and a 4 lead oscilloscope to keep things straight. After you charge the AGM and LiFePO4 in parallel, let it sit for a few days. Charge it again, then let it sit again. It takes more time and repetition, but it's neat to watch how it works.
4s is easy to balance, 16s would be impossible with such a setup. Far too many variations.
@@OffGridGarageAustralia 16s is not impossible my friend, it would take longer though. AGM can be put in series too. LiFePO4 wants to be balanced and a constant DC power source like an AGM helps that happen when it's sitting at the top end of it's charging range.
There isn't going to be much voltage difference and therefore balancing going on for most of the capacity range, so as long as the battery isn't almost empty most of the time (or at least longer than it is full) it should still work fine.
OK, but even discharging the battery with say 50A will cause a certain deviation (20-30mV) which the balancer will use to balance. And especially in winter and using the battery only from 10-50%SOC, this causes a lot of imbalance.
Naturally the balancing in the lower voltage level is not a good idea, because the system know nothing about the real load-condition. Every cell is a bit different and with the top balancing we get the best possible solution.
So we can load the cell up to the maximum voltage (around 3.45V is enough) and if we deplete the battery to 0%, then we know that some cells will have still 3.3V but other cells are at 3.0V so that we have to stop the energy consumption. There we have at the lower voltage area the voltage drift, but if we want to store the energy very fast, then we can do it without any problems. We can fill up the battery without any time delay to 3.45V each cell.
Great, love it.
Thanks!
Thanks a lot for your ongoing support, Sree. Much appreciated!🙏
Ich habe 2 Batteriepacks a 24 Volt in meinem Inselsystem. Die tollen Hitiumzellen (ja, ich weiß) gepaart mit je einem Daly-BMS (Ja, ich weiß);
Ein Pack lief von Anfang an ohne Probleme, bei dem ich kein Top-Balancing gemacht habe und der Andere, bei dem ich das vorher tat machte von Anfang an zicken. Eine Zelle driftete weg, war schon auf 3,65 Volt, da waren die anderen noch nicht auf 3,5.....Ladung wurde abgeschaltet, da ich den maximalen Drift auf 0,2 Volt festgelegt hatte... Ich hatte ein wenig Panik und habe mir einen aktiven Balancer 5 A zusätzlich drangehängt.
Der läuft jetzt seid einem Monat durch. Geladen werden die Zellen auf 3,5 Volt, entladen bis auf 2,6 Volt, was wohl erst im Winter passieren dürfte.
Jetzt hält sich der Zellendrift in Grenzen und ich bin zufrieden.
Bei guten, gleichen Zellen ist das zusätzliche Balancing vielleicht nicht nötig, aber bei dem billigen Schrott anscheinend schon.
Ich könnte ihn mal wieder abklemmen und schauen, ob die Zellen wieder auseinander driften.
Or you could try using such a voltage detection relay to turn on the balancer at 3.45V and see if that improves the deviation even more.
Thanks for sharing. I like the Hithium cells. They work well in my Frankenstein battery.
I think everything depends how good are cells and how long you don’t use battery with active balancer on. Like example if you daily charge and discharge battery active balancer shouldn’t do anything wrong just keep cells in balance. I think would be bad to leave active balancer on 24/7 when you don’t use your battery
Yes, very good and valid point. If you have well matched cells, it may just work with the balancer being on all the time.
Depends on the time spend at low soc(0-10%). In winter most batteries will be on the low SOC most of the time wich will destroy the top balance with an active balancer active. In summer when not reaching low soc, the balance current will be low at the flat part of the curve and will not be a problem(at normal charge rate 0.2c). I agree using a balance voltagepoint above 3.4-3.45 will always work. Secondly, make sure your battery can reach 100% occasionally, even in winter and not leaving at at low soc for months(as i seen often).
Leaving it a a few days cycle between 0-30% wich is happening in winter in our country ( The Netherlands) for example. actually this will go on for months.. It will destroy top balance. Most inverters can not easely be set to seasonal "lowest soc" without going through a lot of settings so people tend not to do so. In summer batteries will be at 100% soc for a few hours a day so then the active ballancer does great.. This is the avarage from what i see happening. Every system and the capabilities for good maintanance is different. When systems are not maintained and sitting at low SOC a lot in winter then the active balancer is actually a good thing.
true, I rarely see the battery manufactureres also state that the batteries should be charged to 100% ca. monthly to calibrate SOC etc. Deye ESS at least states that for theirs, so people have a point to follow (if they are smart)
Well, we do the same discharge test with both option (with and without AB), so it should not matter for the sake of comparing both models.
I'll test tonight and video will come tomorrow morning. Then we know.
I think: it depends on the charge and load current respectively how it´s charged/discharged during the day. High charge cuurent and high discharge vs high charge and low discharge vs low charge and high discharge vs .... and so on. it depends. The best way to achieve best balance while activeBalancer is always on is to charge very slowly so the aBalancer will have the most time to balance while charging.
Yes, if you charge slowly, it will work.
Having the balancer running all the time would work fine, as long as your battery spends a significant amount of time at >3.5V, where it can properly balance.
If you try to run the battery between 20 and 80% and never fully charge it, the balance will get worse and worse, and the battery will never properly balance.
Yes, correct. The balancer needs time...
If you don't fully charge the battery for months during winter and have the balancer running, it will take a while to fix this 'damage' again.
Liked I tried to message
The JK BMS is the Only One that Balances the Cells Properly
The only BMS, yes.
I believe your theory is correct. I am uncertain if your test will reveal anything, though. One cycle might not change the balance so much and the balancer might be able to even that out quickly.
I could imagine that in winter time when the battery oscillates between 0% and 30% your top balance will be completely destroyed over time. When balancing in the bottom steep area.
Yeah, winter is always problematic with a permanent balancer. It has so much time to do 'damage'.
We will see what is going to happen, video is coming out tomorrow morning at the usual time.
I have seen many of your videos talking about being out of balance and having weaker cells. It seems logical to have an active balancer on all the time. If anything have a scheduled monthly balance day. Then see what happens. My Schneider c-40 has a setting for a monthly auto higher charging cycle to help with balancing. I haven't used this yet, my system is not built.
A lot depends on the state of the batteries, old? problems? and the settings you have decided on.
I get it about not balancing on the flat part of the charge/discharge curve, but just as you can put more capacity into the charging by stopping the cells reaching their upper voltage limit, surely the same must apply to discharging, whereby it can stop the discharging stopping because one cell has fallen below the cut-off voltage, I'd like to think that both top and bottom balancing would give you the best capacity and also the balancing current would have to be proportional to the current being put in or drained out of the battery, surely this isn't a hard algorithm to apply
Very hard to do. As you said bottom and top balance the battery will only work if the load current is very small and the balancer has time to move huge amounts of energy across all cells. Almost impossible.
And if you really need all the capacity for your loads, get a second or bigger battery.
I think when a battery has an active balancer that is actively pulling it literally out of balance and you don't reach full often in other words you don't top balance often then that one sunny day you do get that drives you up to full quickly can cause a major imbalance and literally a safety shutdown situation Because of a high cell that goes over 3.65 volts. Next question is with regular lithium batteries like from the Nissan leaf since they don't have a super flat voltage curve can we use active balancers on all the time with those batteries? I say more likely yes but not necessarily yes because the same principles will still come into play as they near top of the charge.
There is 1s LifePo4 balancer. It's a rare item and very expensive. It's monitor individual cell only since it 1s.
Wow, great. I need to get one!🧐
My theory is when you discharge aggressivly ( ie 70A into a Tesla) and then charge aggressively you create battery cell imbalances that dont represent the true cell imbalances seen at the top when the current thru the pack is minimal. Therefore the cell balancer may actively unbalance the pack. I have a JK standalone active balancer that has no setting to turn on/off according to cell voltage. On a 16s 15kWhr LiFePo4 system it has been left connected continuously for 2yrs. But deltaV setting switches on active balancer when > then 20mV. After 2 yrs deltaV is 20mV at 56.0V. I have a washing machine, dishwasher and 2 water pumps but no Tesla! Switching on the active balancer above 5mV however will likely create a much larger "imbalance" if left connected 24/7. If the charging is too agressive at the top end and there is minimal time spent above 3.4v when the true state of the balance states directs the active balancer then the pack doesnt get balanced properly.
Yep, great.
Discharging the battery with 50A is a bit less than 0.2C for the battery, so I would not consider this as aggressive.
Setting the voltage deviation/delta correct is a good point if the balancer runs all the time. I guess this is also the Seplos strategy with their new active balancer which has a 50mV setting and 'runs' all the time above 3V.
Hey Andy, it looks like you could use some solar powered walkway lights for your late nite trips to the garage.
That wouldn't make good scenes for my video though. The torch is always a bit more mysterious 😂
I think the diffrence would not be much.
Why?
I track all battery telemetry all the time in HA. I only can see significant cell voltage deviation during absorbtion phase. Not during charge, neither during discharge. So even if the balancer is permanent on, it would do nothing exept arround absorbtion phase.
Some info about the battery:
Two packs paralell, each 16s lf280k jkbms.
Charging to 55,4 V, absorbtion at 55,4 V, after that bulk at 55,2V.
Balancer starts at 3,4 V.
Deviation is 30-50 mV, goes down to 5 mv within 15-20 minutes balancing with 2A. Then further down during the day without balancing.
During discharge deviation does not exeed 5 mV for longer then some minutes.
After running this for 24 full cycles (around two months) i changed the balancer treshhold today to 3,45 V (as Andy recomnents it) to see if it makes a diffrence.
Interesting settings. So, you basically keep the battery at 55.2-55.4V for the rest of the day. That will give the balancer a lot of time to do its job.
Yeah, let us know if the change in settings makes a difference. I don't think it will. Settings are too close together.
Hi Andy, and it’s interesting, as always!
My opinion: That the balancer is permanently, will mean that the battery will be balanced correctly, and sincerely I don't see why it won't be balanced?
But that the balancer is in function is useless, the system for example of the Neeey is much better, or as you did with the
voltage detection relay.
I will be very surprised, if the battery does not balance, with the balancer always still On?
we will see?? Thank again Andy!
Thank you, Video will be online tomorrow...
it hacks me off when i see a quote from a person who's clearly not analytically disciplined enough to understand their own words ughhhh - great vid and thanks for slapping the village idiots from time to time as requiredddd
I think it will depend on how fast / slow its charged and discharged
also how long it stays at a high / low state of charge
if the battery never went above say ~50% SOC for months on end over winter
I have a feeling the balance would be completely distroyed
Well, we will pull the exact same current from the battery and charge the Tesla with it. Exact same test conditions just one with and one test without balancer.
Andy, can you test how much deviation there should be before the active balancer starts to work. If I'm not miistaking it's more than the 5mV you mention. This will determin if it's okay to keep it running 24/7
I think it is around 5mV. It will still work below that but efficiency is very very low.
You probably can leave it running with 50mV deviation as you won't reach that in the flat part and it only balances if you get more than that.
IMO, the unbalancing will happen when the cells are around 3.1V~2.5V if you intend to top balance.
(Or around 3.4~3.6V if you intend to bottom balance).
I am in the camp of balancing only when cells are >3.4v, and not always.
I’m kind of expecting it to look fully balanced at around 60% SOC, but go totally out of whack by the time it hits 100%… but who knows!
Good guess!
Heya, active balancing is good for high voltische not the hole time
So I have two packs, both with 5a balancers. One has the voltage controller/relay that turns balancer on at 3.45v and off at 3.43v the other is always on. I will say under max current charge, the one without the controller almost always has a cell hitting 3.65v which disables charging for the pack. I never have this problem with the pack with voltage controller/relay.
That matches my observation and testing as well. Great test, thanks for sharing.
Remember the Angels take their share of power out of any electronics running. (Like they do with the brewing of the Whisky!).
Keep it to a minimum regardles of if a balancer is ok on all the time or not.
Balancing up top but upto 3.55v my Daly isn't too bad at controlling wee 105Ah cells above 3.55v it can't cope.
Down low is shockingly bad with a masive delta at aprrox 3.15v highest cellI I have a low voltage disconect at 2.65v for the lowest.
Andy Charge the Goooooble with as many amps as it will take.
Really Clump it one!!!
Luv Ya
Gary
Hahaha, thanks Gary!
Would be nice to have a third test using the Neeeeey 4a balancing between 3.4-3.45V and then another without an active balancer to see if that maintains the balance better.
Na ... this test is useless, because I think he had done this in an other video.
If the batters was top balanced, then you do not need to re-balance for a while. There should nothing happen, only if one cell is losing more capacity as the other cells, then it would be necessary.
@@micropower8 I mainly suggested it because it shows the difference (potentially) between having an active balancer on all the time and only when top balancing. It's still worthwhile with these allegedly well paired cells even if it's been done before, in my opinion.
I could do that as well, why not...
Hopefully Gobelpower made this test also and therefore came to the conclusion that its a good thing to leave the active balancer on "working" 24/7. Why otherwise they should install it in that way? We will see because of our testing hero Andy.
Thanks, hahaha.
Doing the last test today and will edit the video tonight, so ready for you Friday evening to watch.
I can imagine, they just throw the balancer in there hoping it will work. It's a cost factor to either use a NEEY as standard or a voltage control relay. Or even a simple switch to turn it on or off...
For my experience the big Frankie rept and lishen cells battery, even starting balance at 3.375v it destroyed my balance by up to 25AH for one cell.
Oh wow, that is a lot!
@OffGridGarageAustralia when I replaced the bad cells, I discharged the new cells, I think too much that even after a few months, they seem didn't balance or the balancer made the issue worse, so I charged them using a power supply. It seems much better now, I used to set the jk balance, start at 3.37, then at 3.4, and now 3.45.
I am not so familiar with lithium but I came here to learn. Certainly with flooded lead acid you only EQ (balance) the battery when SG readings deviate by more than 0.03 points or more. eg most of the cells read 1.280 (fully charged) but a couple read 1.250. In this instance I would extend the absorption time by 30 mins for a month and check again. I NEVER use EQ as this puts a lot of strain on the cells that are already performing well and essentially over charges them. Lithium is a different animal because there is no SG to check so balancing has to be done by software. When you consider leaving this software running 24/7 it is constantly cycling the batteries by robbing higher voltage cells to top up lower voltage cells no matter the SOC. This is probably okay when the PV is producing surplus power as the energy required to balance comes from PV rather than robbing other cells but this can't be good when the PV is producing less than required where the SOC drops off. In my opinion you are just cycling the cells more than necessary which over time, will shorten their life. I believe the balancer should only operate when there is surplus charging and after all it's just a chemical reaction inside the cells so the less this chemical reaction takes place (charge/ discharge) the longer the cells will last.
Cool, I learned something about LA today. Thanks for sharing.
Interesting that you never equalise...
For me there is also a question about charge/discharge speed in relation to being able to reach 100% SOC and full top balance. If one babies the batteries around with 0.1-0.2C charge/discharge it is more unlikely to have the cells become unbalanced and SOC reported wrongly (expereince from Seplos BMS and 280Ah 16s x5 packs in parallel). I understand you, Andy, also top balance/charge the batteries with the Wanptek 21A charger at only 21A... this is less than 0.1C, no? So very little stress on the cells, in real life for me at least - I charge/discharge with sometimes 200A (40A per battery still is very low) and when playing around with single packs and then reconnecting, I sometimes see also ~150A per pack peaks.
Not directly to the topic, but could be useful....
I have a Neeey gen4 installed in all my 5 Seplos packs and the cells were top balanced to 3.6V, also full battery cluster charged to 100%. I did not know at the time, Seplos calculates the 100% oddly, so figured this was fine. I had Neeey running all the time - 3.1 to 3.55V if I remember correctly (as you said "if you want the balancer to balance, let it balance - remember :)? ). Now my SOC% is 7% different from max to min pack and thus cannot fully discharge the packs to full potential. Is it the active balancers' fault or just stupid Seplos, my inexcperience and too low V "top balance" or all combined, I cannot say for sure :)
I hope you can test the Seplos parallel system soon too (V2 BMS) as currently I am testing the idea of having the SOC "calibrated" - all packs to 100% and also top-balanced from time to time from a 21A Wanptek charger too. My inverter sees BMS average SOC as 100% although some packs might still be 92-95%... it then disables charge and then it is only the possibility for me I think to keep charging with another source until all packs reach the 100% and desired cell voltages (and get balanced too). And I guess, I will also try to top balance them to the max 58.2-58.4V first with the help of the Wanptek charger.
Great comment and yes, I agree, it also depends on the current comming through he battery how far they go out of balance.
The NEEY should not run all the time. That's what the settings are for to start (try 3.45V) and stop (try 3.4V). This will fully top balance your batteries again.
Yes, definitely more Seplos magic coming. I have a few 10E BMS here which I would like to test. Unfortunately, the 10E BMS needs a cell or pack OVP to reset to 100%.
Luckily it shows only full numbers in my Victron so 99.5% shows as 100%. That could be a sneaky work around with the correct settings. We will see...
Given that if you hook 16 cells in parallel at 3.2v they wont' balance at all, I would guess one of 2 things:
1. It does nothing below 3.4+ish simply because there is nothing to balance and no pressure and thus the capacitors wont' actually discharge and after the first pull won't charge either.
2. It completely screws it up because the amperage still flows despite the voltage being the same because there is still some latent pressure to transfer in the capacitors. If this occurs it's probably because the cells are mismatched and the total capacity is different by a large amount between cells.
Think about different internal resistance of the cells and the different voltage they will have therefore under load. The AB will take this voltage and starts working...
Mine 16S capacitor type active balancer would turn on (led light on) at 2.48v per cell or 45.5v at 16s.
Yeah, not great...
I'd be interested in the energy lost when running the balancer all the time. It must waste some electricity when balancing?
For sure it is burning some energy in the process. Interesting thought...
What’s you’re idea - my battery 3.2v 280AH will not charge over 3.37v ? At different times I am able to get the voltage up to 3.43v stop changing and the surface change goes away back to 3.37v ? “ please what’s your expert idea please ?
It's incredible how many people fail to understand that active balancing is almost pointless!
For a series string of cells, the individual average voltage swing is simply set by the effective capacity of those cells. if you put a 10Ah cell in series with a 5Ah, then clearly, whatever the volts per Ah slope, the 5Ah battery will be the limiting factor, ie will charge to vmax first and discharge to vmin first. But this doens't matter. As long as your BMS can stop charging when the first cell hits Vmax, and stop discharging when the first cell hits Vmin, you are, for the most part done.
Pointlessly equalising each cell voltage is frankly a waste of time, energy and money! These days, most cells are pretty well matched, so make sure you do a good inital top balance to get all the cells to 100% SoC, then just use your battery. I don't even run with ANY balance circuit on my battery and after 18 months all the cells still hit Vmax within about 15mV of each other, so the absolute useable capacity i have "lost" by not getting all the cells to exactly vmax is minimal (it works out to be 0.3% of the nominal capacity). Even when you get all your cells to vmax match perfectly, you battery useable capacity is STILL limited by the weakest cell (ie the cell with the lowest useable capacity), that cell will always experience the largest voltage change per unit charge.
For a typical domestic or off grid solar battery, you should be NOT using anywhere near 100% of available capacity to maximise cell life (if you do the maths, it's better to buy more capacity and not use it, because the cycle life goes up by a greater proportion than the cost of the extra capacity), and if you are not using all the capacity, do you care if you are using say 80% or 82% of it? no? thought not
Spending money on an expensive balancer is a false economy, it's far better to simply spend the money on more installed capacity, enabling a longer overall life because your avergae discharge ratio is smaller!
Thanks for your comment.
I will test this with the Seplos batteries soon and see if it will work that way. I have tested many, many batteries and have never seen it working as you describe. You must have very high quality and well matched cells. However, this is not standard and most cells are not so well matched and if they age, it gets even worse.
Talking about aging, I think it is more economical to fully charge a battery every single day and use the full capacity (90%DoD), than buying a second one and use only 60%DOD. It's dead capital and assets you have spend money on don't use it. Only if you're after the faster ROI.
LiFePO4 can be charged to almost 100% on a fairly low voltage, so it does not stress the battery and causes any increased degradation. And with thousands of cycles, calendar age will kill them before cycles will show a negative impact.
@@OffGridGarageAustralia My system uses Envision Gen4 cells that are from a Nissan Leaf, so are obviously full automotive quality certified, and yes, that makes them very well balanced. They are also not a LiPo chemistry so experience a large deltaV per unit charge.
The important point is really that the only time you actually care about small voltage variations between your cells is at the top and bottom of the capacity range, and really for a off-grid or home storage application, you really shouldn't be going there! For an automotive application where people might need the full range from 100% SoC then yes, precision balancing can be effective at freeing up that full range, but even here when you do the maths on what useable capacity you actually loose with say a 50mV delta, you find it isn't actually that much.
For a practical battery, where all the cells are operating under similar conditions (ambient temp and series current) then the weakest cell is ALWAYS the weakest cell, you still can't use any more capacity in practice than contained in that cell (unless you either have an entirely parallel cell architecture (unlikely) or have some sortof massive, complex andhugely expensive BMS that can somehow shunt the ENTIRE series current around a cell (and if you have this, you're a bit silly, because for less money you could have simply put in more static capacity and avoided the need to do it in the first place....)
Hi andy im really learning alot from your videos. And i sometimes argue with people who thinks AB should be 24/7 always on.
Can i ask your opinion since i have a 16s i have 3 rouge cell that quickly goes too high/low during charge/discharge.. i have set my balancer(capacitive with a30 controller) to start at 55v and was thinking of shutting down at 51.2v(my daily DOD CYCLE)
Im really having issues during discharge cause one cell of 3 rouge cell is really too low hitting LVD too early (at 52v cellpack)
Thanks and cheers
Sorry Andy can't find the message from @thegreenthing7603 going to reply here and hope he get's it. For the BMS I use Andy parameters and for Deye inverter I have this Batt-V Mode, Float V : 54.5, Absorption V 55.8V, Equalization V 55.8, Equalization Cycle 1 and Equalization Operating Time: 3 h/2. The inverter first charge to the parameters you have and when consider the battery full stop the charge and only some minutes after will activate equalization, you should use a long time, mine is for 2A JK BMS, the long time is because the worst cell will reach the higher voltage and will discharge to reach the voltage of the others and only when they are all equal will resume the charge of the bad ones. Hope this helps.
@@CantFindInTH-cam hi thank you for finding time to find and reply to me, really appreciate it.
My battery pack has oem bms(dont know the brand) your parameters are ok but may i ask does the equalization parameters work on battery with bms since equalization is meant to be use for lead acid directly without bms.. i kinda notice that when my battery is full, the bms cuts any charging how will equalization feature would work with bms power pack.. is my assumption correct?
If your BMS cannot talk with the battery you need to setup the inverter for lead and it works very well. My first winter I had only 1 hour absorption and equalization off and after 3 months had a cell with less 10Ah. This inverter don't have absorption time so you do it in the equalization. The BMS is to protect the battery if it cuts the charge you need to revise your configuration. The cut of the charge is done by the inverter. For the inverter be able to balance you should setup in the bms the balance voltage at 55.2V and the cell OVP lower than 58,41V, I use 57.6V. If your bms don't allow to setup values you need to know them and adjust the inverter according. @TheGreenThing
@@CantFindInTH-cam my battery/bms can communicate via can with the deye 5k. unfortunately, i have no access to the bms setting parameters & i do not know what type/brand is installed it just displays "WELCOME TO SMART USE BMS"
Though ive seen the voltage pack charging upto 56v only then it cuts the charging.. ive also seen the deye inverter sending 55.xxV with -13w(afer reaching 56V) but when i check the display on the battery case it just stays 53.6v with zero current hence i do not know if the bms/battery is in absorption time.
Should i just use lithium mode or BattV mode with your parameters and equalization mode ?
@thegreenthing7603 If the CAN is working the parameters you put in the inverter should work. You can try to program the inverter with same values of Andy. To use has BattV mode you must have control of the configuration of the BMS or at least know the parameters of the BMS. I read the initial post you wrote to Andy, do you have a multimeter to check the voltages of the cells in red? Can you very if the cells and bms cable have good contact? Don't trust 100% in voltage of the bms balancer or even in the inverter V. In my case my DEYE 1 measure -0,3V than the actual voltage and DEYE 2 +0,13V and I have to adjust the values of 2 of my JKBMS to read the correct voltage of the cells, after it I confirmed that my cells were OK BMS was reading wrong value and unbalancing the cells. Use your voltimeter has your reference and configure your devices based on it. Any way if your system is working and you don't notice loss in capacity or high temperatures ignore the values and enjoy it, just try to fully charge the battery at least 2-4 times a year.
Perdona mi inglés es muy bajo ,pero veo que la gran cuenta pendiente que tienen las baterías de litio ya sean de autoproyecto o comerciales es tener un buen balanceador
Yes, unless, the cells are well matched and of high quality, meaning, they stay that way when they age.
But having a balancer is not a bad thing
Sorry to put my question here...But....
Help please...!! Is it possible to build a battery with the cells arranged in a different way than we usually see? I am planning to build a battery in which the cells are lying on their side...is this a bad idea...?
How are you getting 58.2 volts with active balancer?
Here is why I am asking, perhaps I have something misconfigured. My bms when active balancer volts reach 3.45 it never goes higher than that alone per cell. With your 58.2/16= 3.63 per cell. You indicated that your active balancer is set to start at 3.45v. Can you explain why I am seeing different results.
im not sure and im not familiar with special balancer characteristics - but maybe the yellow LED just shows the balancer is powered, and this is not leaving a statement bout the working status?
It will turn off below ~2.8V, so it's a bit of both. If it runs, it balances if there is a voltage difference.
@@OffGridGarageAustralia k.
Remember, balancing cost allways energy, so you should as well measure thr losses with allways balancing and with only balancing on the top.
In my opimion, allways balancig will cost a lot more of losses as well.
Yes, good point, Juergen.
If the cells are being charged or discharged at a current that the balancer has no hope of catching, why would it be on? Is voltage alone the correct way to enable/disable a balancer? The cell differences will often exceed what can be balanced while under heavy charge/discharge so why not disable the balancer until the cells are 'quiet' and the balancer can do it's work uninterrupted - as it were? The balance threshold would be a voltage and a current.
Just musing.
Yes, exactly. I balance at absorption voltage when the current is low or even off.
I expect a higher imbalance. But It is not insurmountable IF you have at least a 5A NEEY or the like when the battery is at 55.2 or higher.
I use 4X Dali BMS 250A these use 250 milliamp passive balancing and 32 EVE 280 AH matched batts been is service 24/7 over two years and a Victron Quattro 8KVA 24V
That's a massive balance current for a BMS! My QUCC BMS had 200mA.
So you're not using any active balancer at all?
I think you should drain it to maybe 10% (below the voltage flat spot) and bottom balance it for a few days, then charge it to see if the top balance was messed up.
Of course the top balance will be messed up if you bottom balance it. You do only one. And bottom balancing is outdated in my eyes. That's a thing from the past, where we didn't have BMS and balancers.
Once top-balanced, isn’t it enough to connect an active balancer once a year?
But it depends how consistent your cells and connections really are.
I think you need to run the discharge over a long period not just dumping into the Tesla, say over a couple of days. Give the active balancer a good change to f-up the battery. Otherwise the time for the balancer being exposed to the flat area of the curve is not as long as it would be under a normal condition.
Well, wee will see. Finishing the test today, video will come out tomorrow.
I tried an active balancer last winter and it destroyed the top balance. Even the state of charge was not right after the whole winter time. But may be there is another solution, which can work. Usually even a high load does not cause a big imbalance at the flat area of the the cells. There will only be a deviation at the top and at the bottom. So maybe you can set the active balancer to work at around 20 mV deviation. This would lead to more at capacity at the bottom under 3V. Can you try this?
These balancers cannot be set in any way. They just work as soon as they are connected.
But yeah, this is something we look into with the NEEY and program it to say 30mV delta, so it won't turn on of the deviation is below that. It may be able to run all the time. Or say at least above 3V.
Mine is always on for 1 and a half year. 8kw litium ion battery. Never charge or discharge at max
I wonder how much loss the balancer introduces by being on all the time. Balancing the cells can't be free. Maybe it is so little it does not matter or is too hard to measure because ambient temperature is a much bigger factor.
On the other hand, having a balancer in the lower state of charge might just help if a few cells are not as good as the rest of them. Let's say in the future you have one not so good cell that only goes to 250 Ah and other cells go up to 300 Ah. Can the BMS still top off those good cells? And if so, will the battery last longer if the balancer still recharges the one bad cell from all those good cells in a lower voltage range, thereby not taking down the entire battery much too soon when all of the other cells still have 50 Ah each?
Just some thoughts of the top of my head about the cons and pro's of balancing if you just want to maximize the kWh you can get from a full charge. I may be completely wrong on how this works.
You're not completely wrong, but spot on!
Balancing in the lower SOC could be problematic though as the balancer would only work, if your load current is smaller than the balance current. Otherwise it wouldn't have a chance.
The already weak cells may get a harder workout if they get constantly charged and discharged by the balancer, so it could actually make the situation worse over time...
@@OffGridGarageAustralia So maybe keeping the balancer running continuously is a good way to keep up on some batteries. But on LFP cells with a very flat voltage curve it might guess wrong and start overcompensating for a hard to measure the actual SoC difference. I wonder if that range could have a higher relative difference between cells to start balancing. I agree that it may be hard to keep up with the discharge and bad cells will only get worse. But without the right skills and knowledge, servicing these individual cell is not an option. It would be so much better if these were in hot-swappable brackets.
Rare full charge for me today - 90mv difference at 3.575 average/cell. 50ma passive balancer on BMS, 12kwh battery, coming upto 4 year battery life.
Why all this worry about balancers? To answer your question, I think it will be more stable without the active.
Yeah, depends on your usage and how good and matched your cells are.
@@OffGridGarageAustralia my 24 cells must be awesomely matched, although 6P4S makes for a good average for each cellpack of 6 cells verus your single cells. And yes an average cycle is a week not a day - 182 cycles in 3.5 years according to the smart shunt.
My friend never listens to me on this matter and i laughed my ass off when he told me it's not fully charging anymore and sometimes it is charging, when i checked on the bms it showed that 3 cells are already at 3.65V and the other cells were at 3.4, 3.42, 3.35, and one was at 3.3V his active balancer messed everything up, as soon as i modified the balancer(i.e making a voltage switch that makes the balancer turn on at 3.45V per cell) the cells balanced flawlessly with the highest at 3.65 and the lowest at 3.62V
also i might've made a joke about using a horse to balance it, he asked me what it was and i neighed at him, he laughed and is now part of the NEEEY fam
I have always ON balancer. Now its one year. Chinnese battery from aliexpress 16S.(Litokala)
Charging voltage is 55V discharge voltage 50.8V delta almost every day is 200mv.
IF Charging voltage is 54.4V discharge voltage 50.8V delta almost is 80mv.
Balancer after charge took about 20min to take 200mv to 30/20mv.
I think turning OFF balancer while discharging bring better solution becouse if I have bad cell whitch I have. Capacity is about 55 - 70Ah of "Declared 90Ah"
At 50.8V I have delta about 30-40mv and ballancer doing downbalancing all night long. Next day TOP is again 200mv.
Compared with other battery pack with 2A JK-BMS I have max delta only 30mv max when do TOP charge and after 10min = 5mv.
So I know answer now but I am lazy right now to rebuild system becouse anyway planning sell this pack and build new.
Yes, great, very good point. Thanks for sharing. A bad cell can make the balancing worse if the AB runs all the time.
I'm having an active balancer all the time and everything is fine. I don't have unbalanced cells anymore.
But before, when i didn't had an active balancer, my cells lose balancing and i have a huge deviation when the cells discharge...
Interesting. You're using the same capacitive balancer?
@@OffGridGarageAustralia I'm using the chinese Hankzor 3-4S balancer. The same you had in your videos.
I always had one cell that discharge over time much faster than the others and because of it i lose the entire 12v capacity cell pack, because that cell will trigger the BMS.
But now, i think it still lose capacity but at least the active cell balancer is draining the current from the other cells and therefore keeps the whole pack in balance. That's why, now my pack lasts much longer.
Active or passive balancing below 3.4 V creates problems.
I want to take the time again to say, an active balancer is only necessary when a pack is first made.
If you need one after a pack has been made, you have a bad cell that needs to be replaced, or you've hooked something up incorrectly. Finely tuned passive balancing is all you need in a good battery. After the initial active balance, all the batteries I make settle in after a week or two and have no more then a 10 mV deviation. Years later they get better and sit with 0 to 5 mV difference because of how amazing low current passive balancing is. It's why all BMS's on the market use passive balancing, it's efficient and cost effective.
Love this investigation Andy, keep up the good work!
Thank you.
Well, I never had any battery pack (built or bought) which could just cope with the tiny BMS balancer. These large capacity cells (100Ah and above) are drifting a lot. This might not be the case with a 12V setup and using 4 cells only. But 16s is a completely different animal and deviation is real.
@@OffGridGarageAustralia You also don't charge your batteries to at least 3.6 volts per cell though. You may find they balance better when you do. All my batteries are set to 3.6. They won't balance at all in the Canadian cold weather if I use 3.45 at 5 degrees C.
@@boatelectricaldiy yes, absolutely. I'll start exactly this experiment today. Thanks for sharing.
@@boatelectricaldiywhat are your settings exactly you are using?
You've done this before and i suspect the 3,45V is the way to go.
OK, let's see 😉
I am always curious about leaving the active balancer on all the time. My BMS has passive balancing and I can set that parameter. the Heltec active balancer is added..it is what it is.. but I am Li-ion chemistry. The pack will stay at .001 dev. when not in use. No matter what SOC. I really never look at SOC as I am more concerned with total pack voltage.(& not having it to the top when it is not in use for a few days..) The closer the cells are... the more I can get.. I watch all your videos but again.. I am Li-ion chemistry . This is a motorcycle racing application . 20s8p Molicel p42s
It will work perfectly with Li-ion batteries due to the linear curve and relation between voltage and SOC. And that's where these balancers are coming from. The same with the small BMS balancers. With Li-ion, you can basically balance all the time and the 50mA will get it done over time.
The bigger delta I have.. the more volts it will use to bring it back.. 2v on the passive and 5v on the active... (max) Im sure you know all this.. LOL It all works very well for me..Learning a lot from you... but having to translate it to LI-ion chemistry ... @@OffGridGarageAustralia
Instead of turning on the active balancer at certain voltages (where the difference is greater), wouldn't it be sufficient ta have the balancer turn on when the difference in voltage is a bit higher, like 0,05V difference?
I was thinking about the same.
Could it be possible to to find an optimum between detlaV of the balancer to kick in and Vbat before one cell Hits 3.65V
That's the approach Seplos is taking with their new AB. It only runs all the time but is only active if cells are over 50mV apart.
So, 50mV would be the best we get then. Would that be OK? I guess yes.
Or you give it a try and keep the original fectory setting 30mV? Finaly it is an optimization problem. I sugest between 30mV and 50mV could be fine.@@OffGridGarageAustralia
Those active balancers have wrecked about $1400 of my batteries. The last seller I contacted said it stops balancing at less then 2.0V. I tested it below 2.0V and it still was balancing. Those cheap Chinese balancers are dangerous junk. They do not work according to design and will discharge your cells too low if not carefully monitored.
My 24v battery with .6a jkbms balancer works just fine when the battery is only used in every other weekend to 40% charge min.
I starten balance at 3.40 and charge to 3.45 for 20 minuten and later changed the absorbe time to 45minutes . No difference. No tail current btw.
So it had 4 to 5 days a week to balance. No problem and up to 1mv difference only for a year.
I then changed the bms for a 200amp jk with 4a balance. Guess what?! Differences are 3 to 6mv with same setting . So, heavy usage you need a big balancer, little usage you need a small balance current?
Great test and conclusion. Thanks for sharing.
You remember lesson about balance? Lesson not just lithium only.Lesson for whole life!