Balance Start Voltage for LiFePO4 explained. Why you should not start too early... 🔋⚡
ฝัง
- เผยแพร่เมื่อ 30 ต.ค. 2023
- Still no word from JK about the new BMS... what better could we do than a quick tutorial about how and when to balance LiFePO4 batteries. I often see that users are setting the balance start voltage too low. They think it's better to start earlier to give the balancer more time to do its job. Especially when using the passive balancer of the BMS, we tend to panic and balance early.
However, this will achieve the exact opposite of the desired outcome of balancing: already low voltage cells are getting discharged even further.
I have my whiteboard out and explain quickly what is happening. Only until the JK developer team has the new firmware ready. Who knows, how long this will take though...
This video is by no means an comprehensive guide on how to balance. but it gives you a good starting point. There is a lot more too it, like current, deviation, voltage, system design and including all these factors would not made it an easy to understand tutorial. Also, all systems are different and there is no settings which fits all, while the balance start voltage will work for all balancers.
MASSIVE 15% discount of the Gobelpower Standard battery. Only in November!
off-grid-garage.com/li-ion-ba...
Test and review of one of the best pre-assempled batteries on the market:
• THIS is it. The Gobelp...
and • Capacity testing the G...
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
![[Teaser EP.2] “MasterChef Junior Thailand Season 3” วันอาทิตย์ที่ 9 มิ.ย. นี้ 6 โมงเย็น ทางช่อง 7HD](http://i.ytimg.com/vi/KojBzrifbWg/mqdefault.jpg)
Andy, most BMSes look at cell voltage deviation as well as balance threshold voltage. The lowest cell will never be pulled down because it's used for the deviation calculation.
My thoughts! Start @ 3.45V ans Charge to 3.55V with Float set to 3.4V. Works perfectly for me. Starting at the top knee an then falling back to the start again after balancing.
Yes that is correct. I didn't go down that rabbit hole as it makes this all more complicated and does not really change the start voltage.
@@OffGridGarageAustralia So if you leave a Passive Balancer connected and you are not charging will it run all the cells down ?
First and foremost, I want to extend my congratulations for your captivating videos; they are undeniably engrossing. In general, I find myself in agreement with you regarding the technical aspects you present. My concern stems from what I perceive as inaccuracies in your explanations concerning battery balancing.
To begin, let's revisit the fundamental operation of a BMS balancer, whether it's of the active or passive type. Its fundamental purpose is to ensure the equitable distribution of the charge among the cells of a battery. In the case of passive systems, their role is primarily to discharge the cells with the highest charge. On the other hand, active systems go a step further by not only discharging the highly charged cells but also redistributing their energy to the less charged ones. However, the core principle remains unchanged.
In your description, you seem to equate the **starting voltage** with the **target voltage** that cells must achieve. In reality, these two factors are entirely unrelated, and the balancing system doesn't concern itself with the target voltage, as it is solely determined by the chargers (mppt & inverter).
Let's reconsider the example of the unbalanced system you showcased. You suggest that, since all four batteries exceed 3.4 V, they will all undergo discharge until they reach this voltage. However, this is unequivocally erroneous. As previously mentioned, the primary role of a balancing system is to strive for equilibrium among the cells. In the case of a passive system, as in your example, it would simply discharge the cell with the highest charge (3.48 V) into a resistor. Yet, it certainly wouldn't discharge all the other cells as you implied. If it were to do so, it would cease to be a "smart balancer" and instead resemble a "dummy discharger." The sole purpose of the startup voltage is to prevent the balancing system from initiating the balancing process while the cells are still in the linear zone where charge is not directly proportional to voltage. It is, therefore, imperative to set this voltage beyond the inflection point of the curve. However, it's essential to recognize that this has no bearing on the final battery voltage.
I hope that these clarifications will assist in refining and enhancing the accuracy of your explanations.
Thank you for your comment.
I was aware to open a can of worm here and the start balance voltage is just the beginning. I will make a part II of this video and go into more depth. As I explained in the video description, this is just to give people a guide where to start balancing... of course, there is far more to it...
THANKS Andy, as always, a big 👍👍. Right on the money.
Years ago I had a slow balancer on my 400A LFP caravan battery, and couldn’t set a start voltage so it was always “balancing”, and can absolutely confirm your findings.
Since I fitted the JK in the van, with balancing set to start at 3.45v, it’s all been PERFECT. I typically see just a few millivolts difference between cells.
This all only applies to passive balancing BMS, not the active type like the JK's, which support the weakest cell and instead wears the strongest cell at all times.
Active balancers like this in theory even allow for the FULL capacity of each cell to be used (even if they differ), surpassing top-balancing, as the BMS is supporting weaker cells with stronger cells during charging and discharging.. this naturally needs a BMS whose balancing current is at least as large as what the maximum dis/charge-current one has got in the pack times the difference between the worst cells. Say the difference is 30% between best and worst cell and the max current is to be 10A, then the BMS active balancer needs to be able to balance 3A.. which means the strongest cell is delivering 13A while the weakest delivers 7A.
The overall concept is then heading into the direction of not needing to top-balance a pack at all once the BMS balancer is strong enough to deal with any mismatched cells - it's simply balancing all the time anytime. Economically the whole thing lives from the price difference of a strong active balancing BMS vs what a well matched battery pack costs compared to a not so well matched pack.
In IT they use the term JBOD for Just A Bunch Of Disks to create a larger volume.. maybe once strong active balancing BMS become widespread and economical enough we'll be talking about JBOB in the future and don't care about matched cells for a battery pack anymore.. because the balancer is able to cope with very unmatched packs due to being able to let the strong cells work for the weak ones.
Simply having a stronger balancer might cover some serious defects... Another useful future BMS function could be intelligent sensors that notice weird behaviours and flag this to the user... Like changes in cell drifting behaviour...
Fully agree, at active balancers which only discharge the strongest cell to load the equalization capacitors it is better to start earlier to keep the cells in balance. When it comes to passive balancers Andy is 100% right.
@ kirkdis
I think Andy is just not realizing that cells with all the same voltages at the top is just a way to make passive balancing work (and it looks "tidy").
Active balancing doesn't care about looks, it just tries to get the job done of getting a bit more out of the battery (driving individual cells differently), which naturally must lead to the top balance being off when it comes back up on recharge - because it purposefully balanced it further down and anywhere in between.
Active balancers simply do not keep the "tidy" top balance intact.. there is no need for it really. They re-balance all the time anyway.
And if they are strong enough and the balancing current is capable to divert enough energy (the diff between the strongest vs the weakest cells(s)) the battery will actually be in (soc) balance 100% of the time, not just at the top.
As much as I love Andy and his channel, in my situation I found this not to be the best advice. Using JK BMS and solar charged batteries, I found 2 things that makes my situation maybe different:
1) There are days (especially in winter, or during rainy days) when my battery does not reach full charge. It could be many days in a row, when my battery reaches anywhere between 50-80% charge and then discharges (down to 10%). In this situation I would be many days/cycles without any balancing.
2) From my observation my JK BMS does not behave like mentioned in the video. The JK BMS takes the highest voltage cell to charge the lowest voltage cell. Does not touch any other. Furthermore, the deviation settings help, so if the batteries are within 5mV it does not kicks in anyway.
@@tomsydney2430Same here except I use 10mV delta to start the balancing. While the cells at the flat curve (10-90% SoC) there is so little voltage difference between them the balancing do not happens.
There's some confusion in this explanation. If your target is 14.2v you don't want to start balancing at that point. You do want to start before that. The first explanation was kind of correct because @ 13.8v (3.45v/cell) is around the point the cells start running away from each other (they are starting to climb out of the flat). I've watched this closely with my 12v set ups and it's pretty consistent that when the first cell reaches 3.43v the deviation starts to increase quite quickly. 3.43 or 3.45 there's very little difference. But if the target voltage is 3.55, you certainly want to start balancing @ 3.43/3.45 because this is the voltage range where you can clearly identify that the one cell (which is actually the weakest almost always- not the strongest) is becoming close to fully charged, You want to start discharging that cell as the others increase. There's no benefit in waiting until one cell is @3.55 volts to start discharging it. If you have cell #1 @3.45v and the other 3 cells are at 3.33 for example, you can be sure that cell #1 is running away and can start balancing that cell. To make that assumption below 3.45v has no benefits which is where Andy is correct.
If you got an active balancer that is strong enough to carry the capacity discrepancy of the weakest vs the strongest cell as a diversion energy transfer during both charging and discharging it will effectively be able to divert the energy that is either too much or missing at the weakest cell and get it from or push it to the strongest cell(s). This should in theory wear the stronger cell(s) faster (as their currents are higher) and spare the weaker cell.. over time bringing all cells to a more common level and remove the discrepancy.
@@joansparky4439 If you have an active balancer like the Neey that is strong. Set the start voltage to 3.45v and forget about it. You won't even be watching this video probably. But If you are charging with 100amps and set the balancer to only start at the target voltage you're going to go into a start stop scenario immediately.
@ AussieAustrian
Yeah, if the cell difference (weak/strong) requires the energy diversion to exceed what the active balancer is capable of diverting the weak cell will hit the boundaries quickly, leading to start/stop cycles (just like in a passive balancer or without a balancer at all).
But I think if the strong active balancer technology becomes economical enough we won't be talking so much about matched cells anymore, which is a good thing IMHO as it allows for cheaper packs to be built.
Right now the JK's use just one capacitor to run around and move energy between the cells. If they could build those with two or 4 of those energy movers for the bigger BMS this would be good I think.
I bought the voltage relay and heltek 5a balancers for my 16s packs with JBD BMS. Best thing I ever did. Start balancer at 3.45v, been running this way since two weeks after your video on the voltage relay..several months at this point. Within each pack they is never more than a .009 Delta. Amazing.
Fantastic! Thanks a lot for sharing!
Thank for you explanation.
It was REALLY helpful for me!
Thanks for the feedback.
Great video and easy simple explanation.
Thank you.
Thank you for sharing another well explained video.
Well done Andy, thanks for comprehensive recap of cell balancing, Something I've noticed on some cheap no-name and Daly low-dollar passive balancers with fixed parameters commonly used in ebike packs, is that they start at 3.50 volts per cell, but they don't discharge only the cells at a higher charge, but simply discharge all cells to 3.55V or 56.8V on a 16S pack, and cut of all charging current if any cell goes over 3.7 volts. Therefore to balance the cells I pretty much have to charge to 58.4v and disconnect charge voltage until the entire pack discharges to 56.8v and then charge again.
That also means that if you keep a constant charge voltage applied the pack will never fully balance as the passive discharge current is continuously replenished, and balancing only occurs when charge power is manually removed. If load is applied after charging, without leaving any time for all cells to discharge to 3.55v, then cells won't balance either, and I suspect why cheap eBike packs that are left on charge and then disconnected and immediately ridden can develop major imbalances
Pretty crude balancing but functional way of if you understand how it works and charge accordingly
I would not use these cheap BMS where all the settings are fixed. You really don't know what they are doing and have no control over it.
Smart BMS are not expensive any more and you can get them from $40 already with Bluetooth app.
@@OffGridGarageAustralia I would ever use hem on a large storage system either.. Just FYI for people who may have them in eBike or portable power systems as they seem to be common.
For small cheap, potable packs I find them a cheap, compact and adequate for over and under voltage protection, and balancing as long as you understand how they work.
Good, easy and obvious explanation 🙂 This means the balance-start-voltage depends on the aimed battery voltage alltogether! It doesn't make any sense to start balancing earlier!
1. some BMS do have active balancer built in.
2. most active balancer works with cell difference/deviation and then you can have a low start voltage for balancing and the balance only starts when the cells have a deviation > x Volt . Then it's better to have a low start voltage for balancing, specially if you never balance before, it can take over a month for 1A to balance it.
Good explanation. I still keep the balance atart voltage at 3.41 because to smoothen the highest cells down early and omit high peaking od the cells. With this I can keep the maximum cell voltage below
If it is an active balancer, you can start at 3.41V.
It's just a guideline what I showed in the video.
THANK YOU! I needed this.😁
Thank you.
Dall'Italia,sei numero 1,anche con i sottotitoli.
Practically genius. Thank you!
Loving your content two thumbs up
Thank you.
You are a legend dragging us out of the "balancing is not necessary" dark ages. You have done this by initially trial and error but now we have all begun to understand how it all actually works thanks to your testing. "All" includes the manufacturers who begun to add features thanks to your documentation. I'd be interested to see how in day to day real life the Neey system performs with voltage on/off set at 2.7V and deltaV = 20mV. Is the flatness of the discharge curve and 20mV enough to prevent unbalancing?
Hello Andy i live in Montréal canada thank you so much for you good expérience i change all my daly bms to jk 200am on my système now i have using this for 280 day it very good bms. Thank you
Wow, that's a big upgrade. I'm not sure if I would have done it or just add some active balancers to the Dalys.
However, JK is definitely the better BMS...
Andy,
Great video again. One thing I would like to point out that if you are using a ‘smart’ balancer like found on your typical BMS, they
1. Have an additional setting that turns on the balancing - the cell differential threshold (called ‘balance open diff voltage on Daly’ called whatever on others) that will turn on only once differential crosses a boundary.
2. Will not drain the lowest cell for passive balancer or it will charge the low cell for active balancer (i.e. you will see this in the monitoring app on your phone- the lowest cell will not be pushed lower.) This is the ‘smart’ part of the BMS. (Related to this at 6:35 in the video you mention ‘it will discharge all of them’, hence you recommend not to use 3.4V. It should only discharge all but the lowest.)
In practice I found this means you can set the start voltage to 3.45V even if you are charging to 3.55V. The primary requirement for the ‘start’ voltage is that it must be in the knee (above 3.4V or so). If you take measurements you will find if 3.45V start is used for 3.55V charge, it will start balancing a tiny bit earlier than the 3.55V start case. From seconds to minutes - since the high cell voltage ramp time from 3.45V to 3.55V for an out of balance battery is often quite short, depending on the current.
Please note I am not pushing for 3.4V, but simply stating in practice, 3.45V start works great with 3.45V charge, 3.55V charge, or 3.6V charge, or you can set them the same and loose a tiny bit of balance time.
Yes, thanks for the further details. I didn't mention the deviation setting in the balancer. It complicates things and does not really change the balance start voltage any way.
The examples were mostly for the internal passive balancer of the BMS. Active balancers act a bit different as you correctly outlined above. But again, the basic start voltage of the balancer can still be set by that simple rule.
Thanks for the response.
As some people point out, at high currents, you may have to bump up the start voltage since the knee may start above 3.4V. Practically I never found that to be an issue.
I originally came across this problem looking for a low cost balancer for a bunch of crappy LIFEPO4 batteries that required the occasional manual balance I wanted to automate without any complexities.... good luck. Tried pretty well every lower cost balancer off ali-express almost none activate at 3.4V or higher. Those that do are passive per cell balancers (no difference detection) that generate a large amount of heat. Still looking.
Since you love to experiment I bet you would have a field day with some type of data logger that you runs whenever you try another experiment. I did and suddenly things become quite clear. You can go back and analyze at your leisure. This is especially important whenever you get a new battery or balancer and indicate that it does not seem to be working too well.
For example I found my JK BMS (about 1A balance current) may take 30 min to balance a battery. Same imbalance on my Daly would take 160hours. I can see it moving on the data logger but not by staring at the voltage short term.
Since your batteries are networked, maybe someone has a clever data logger script you can run in the background? Time, cell voltage, current. I am using a 4S Daly BMS with sinowealth software on PC. Works like a charm.
@@OffGridGarageAustralia
Because of the variance between batteries, I balance all the time, both during discharge and charge. After 7 years of fulltime use and a delta of 6 milliamps and 95 percent of beginning capacity I'll take my method over yours when you seem to have a constant problem with delta drift. My BMS is original to the system with zero problems.
That totally depends on your charge voltage and target. If you don't charge high enough, you will never see a delta with your cells. Also chemistry is important as I explained.
Since my (short by methodically analyzed) experience is more like yours, please share with us the exact precise settings in your inverter AND in your BMS - talking here about voltage settings. Also, geographical are would be great. More important would be to know the setup in terms of inverter, batteries and panels to understand at what SOC your battery stay during winter/summer time. For example, if your battery will charge most of the time at 80%, that would keep all the cells at the same voltage (but not necessarily at the same real energy reserve stored in every cell). Also, what batteries are, in therms of brand and source. Are they certified?
As a side note, my chinese uncertified cells have the tendency to self-balance, without BMS and equalization, which is weird.
But true.
thaks andy, again, very helpfull.
Thank you.
Thank you
Good points today.
Thank you, very basic stuff. We will dig deeper in part two...
Back to basics hey, good, had enough of all the communication stuff. I prefer to work on the K.I.S.S principle. Don't know if I can wait for the new project, nothing sort of popped up in the background that I saw or did it? Keep up the great and informative work, thanks.
We will do a lot of small projects next year. I have some interesting and cool stuff coming up. All very basic... but super useful.
Hi Andy,
Thanks again for another great video.
I know ages ago you did a BMS parameter explanation.
It would be great if you could do one on your prefered settings.
Also one on battery % voltages across the range 0,10,20,30,40,50,60,70,80,90 and 100%
Thanks again Andy!
There is already one if you search the video's
We cannot use voltage to determine the SOC of LiFePO4 batteries. That is impossible. At 3.25V the cell could be at 40% or 80%, no way to tell.
The settings will be different from system to system, so my settings will not fit your design. Very hard to do... Let me think about it...
The problem is that in my case (the boat and solar panels) I am charging my banks with relatively low current - like, 0.1C. And at 0.1C the charge curve is drastically different from 1C and even 0.5C. That is, the battery is fully charged at much lower voltage, something like 3.4 volts per cell (i.e. we get 100% SOC much earlier). In this case I will bever get to the balancing voltage of 3.5 volts simple because at these currents it will mean overcharging the battery. Maybe in this case, setting it to something like 3.5 volts will make sense?
Thank you Andy, always a pleasure to watch your videos. I am trying to find the optimum settings for my new svolt battery / inverter with built in pace bms. One of the cells constantly reach 3.65v while others remain fairly equal within a few mv. . Is it bad for a cell to reach 3.65v often during bulk charge, or can I utilize the function of "forcing" the bms to stay into balance mode as long as no current is drawn. (It is difficult in a production system to not draw current, which gets worse in my country with electricity so called load shedding - e. g. today we will be without electricity for another 8 hours BUT with lots of sun) My supplier suggests 3.5v bulk charge and 3.4v float. Should balance threshold be ok at 3.45v?
-Chris
Thanks for your question Chris.
If you have access to the batteries, the fastest way is to temporary connect an active balancer to the battery and let it do a perfect top balance.
3.5V as charging Voltage is fine if you let the cells absorb for maybe 20-30min afterwards. That will fully charge them. Under such normal charging, none of the cells should reach 3.65V and trigger the BMS to turn off.
Float, I would set a bit lower to 3.35V. 3.4V is a bit to high.
You can start balancing at 3.45V, no problem.
Andy love your video's and have learned al lot.I have a 12v 100amp lifepo4 battery, which has a build in bms . Buit it will only charge to 13.6v. Could it be that my charger cuts of to soon? And voltage drops?( charches with 14.4v) or do i have bad cells.
May it help if i charge the battery with a power supply on like 14.2v and low amps? I use three of these batteries at my campsite for off grid
Mart
Hey Mart, measure the charger voltage without battery. to this voltage it will charge your battery. If the battery stops charging before that voltage, most likely on cell is out of balance and the BMS protects it from overcharging.
If you don't have access to the BMS via Bluetooth, you can use a power supply and charge close to the voltage when it cuts off. Leave it there overnight in hope the balancer will do its job. Increase the voltage very slowly in the morning and see if you can already charge a bit higher. Some BMS don't balance if there is no charge current.
Hi Andy, My Daly 1A Balancer is doing fantastic, but with my Poorly Grade F Litokala cells I need a wider spread on the start woltage to stop a runner. Charge to 3.55 balance start at 3.4 I have tried higher but no good with these cells.
Only 52 Days to the Solstice!!! By Christ it is dark in 🏴
Shine on mate
Gaz
Once you top balance to 3.55V, can you go higher to 3.6V and balance the pack there? Turn off the balancer at this stage and do a cycle. The batteries should stay fairly balanced when they come back to 3.55V. Not 100% but it should work at least to the point that none of the cells shoots over 3.65V.
@@OffGridGarageAustralia Hi mate, I have already tweeked it up a bit. Start voltage is now 3.45v charging to 3.6v and all settled and running well with no disconnects.
I will try your higher start voltage but with small poorly cells the upper voltages rise very quickly. I am not sure how these will act once I charge at higher amps.
Just need more sun lol, the Scottish weather has been very dark and unreliable with a handful of brighter days.
Dave 🐔sends his love, he dreams of roaming your garden in Hot and Sunny Australia .
Cheers Gaz
Hello Andy, thank you for the clear explanation, now i got it for sure. Just 1 question, when you say " your desire voltage" does it mean the bulk voltage or floating voltage, because in our inverter, when reach the bulk voltage the inverter immediately switch to floating voltage which is far lower than 55.2v.
Bulk charging does not have a voltage, bulk has a max charge current.
In addition, LiFePo4 do not use bulk, absorption and float, so asking for float voltage is also irrelevant..
LiFePo4 uses CC/CV charge profile. Constant Current = Bulk, Constant Voltage = absorption.
After CV is done, the charger is supposed to turn off, Victron has solved this my setting the float charge voltage bellow the rest voltage of the LiFePo4 battery..
So the answer to what "your desired voltage", it is the voltage defined in CV part of the charge, or Absorption Voltage for Lead Acid...
Desired voltage = Bulk/Absorption voltage.
If your inverter does not have any absorption time, maybe set the float voltage close to the Bulk voltage to allow the balancer to work. Otherwise it will never balance.
What happens when you have four 100ah batteries with each battery having its own BMS? I am enjoying your channel. Thanks much for your contribution with all the knowledge shared with your subs.
4s configuration to be clear with aEG4 3000W EHV. All in one inverter solar charge controller.
Each 12V set up will drift away by time. Probably not so quickly. Check them every now and then.
Ive wondered about starting passive balancing on the high cells very early, but only if at least one cell is still below the target threshold. Basically, relate the balancing of high cells to the voltage delta. The bigger the delta, the earlier you could start the balancing.
If all cells are in the flat part, delta is small, no early balancing.
As soon as a cell starts growing the pack voltage delta, hit it with balance current. And maybe back off the CCL a little. That should let the balance resistors do more compared to input current.
If you have a cell at 3.400 and another at 3.600, it won’t take much current to raise the high cell over the protection limit of 3.650. So the CCL would probably have to come down to about what the balance resistors can bleed off.
My feeling is that as soon as any cell is balancing, charge current should be limited in order to allow more time for balancing to happen.
But this messes with making use of available solar power by inverters directly, which isn’t an issue in my setup yet. It would be nice to have an onboard battery charge limiter separate from the mppt controllers.
Yeah, charge current definitely matters. If we could fine tune the system to accommodate for that, it would certainly help. I guess with some scripting in Node Red, you could really achieve that and limit the charge current to the balance current.
I see alot of post's about batteries, balancing etc.. but I rarely hear the one most important thing of a battery pack, not so much for lead acid batteries but most other types, resistance matching, if you can, match the cell resistance as close as possible if in parallel and then for any in series, resistance match those too, I have seen some put slighty higher resistance cells in the + and - positions as they get hit first. Resistance matching also leads to much less requirement for balancing and will add to cell life. Obviously, this is more important in larger packs made from 18650 but is also relevant with large prism cells too IE: 400 amp hour cells in a 12v 4 cell pack. Putting power into a cell only to bleed it off is just a very expensive heater.... I got these robot battery packs from the USA 21700 cells 36 v 10s 11p and they are so well matched I barely see a deviation of 0.05 volts while charging and discharging. My application is ebike and related so the cells are 14.5 amps constant 4850 mah, 36v @ 150 amps max discharge per pack 52.7 amp hours each,I have 4 of them heh. So whoever built them, did a damn good job matching, of course when you have access to thousands of cells, it's much easier to match. So if you buy in bulk, or buy used and can check the cells before you buy.... it might be something you want to keep in mind, like if you build something like the guy in Aus with thousands of cells, or a Tesla wall battery thingy ^^
Off topic: I would like to ask if the video about Victron integration into Home Assistant will be coming soon? Thank you
It will come, yes.... Soon, I don't know. I guess, I could... It's very specific though.
I'd add another point to Andy's logical explanation:
Before the chargecurrent doesn't decrease to a very low value close to the balancers capabilities, balancing just doesn't have any big sense.
On the other hand: When coming close to the end of the chargecurve - say 3,55V, it makes sense to lower the chargecurrent to an amount close to the balancers current. In the case of a 2A JK lets say 3-4A, later 2A. Otherwise giving it more, one cell might skyrocket until at 3,65V the BMS shuts down over and over again.
When the chargecurrent is not to much higher than the balancers current, the latter one can catch up and do its job in balancing the cells before the BMS reached OVP.
Perfect, thank you, I 100% agree!
I am in the need to get a few BMS( smart ones) for a off-grid build and was waiting for the new one that Andy just showed and tested but it’s look that i need to wait longer to fix the issues with it
On the other and i need a simpel liable smart BMS where you can set the charge current on different level where you talk about it,what BMS can i buy or what kind of BMS could you recommend it’s for 48v
@@marcvandaele1878 i've got an offgrid-system, too. Very happy with my 2A JK-BMS. I'd recommend going for JK. It's got all the bells and whistles necessary especially a capable active balancer fully configurable out of the box.
One thing to keep in mind, though, is that each BMS-battery-combo acts standalone. So the Solarcharger can't know if there are big differences between the batteries. It'll do the job.
Andy's new JK on the other hand might be a gamechanger in that particular regard.
I think it works the opposite way. The longer into the charging process you wait with the balancing, the higher the risk is for an OVP, because there is no mechanism to lower the highest cell. The balance process itself doesn't increase the risk of an OVP, instead it does the total opposite, it lowers this risk by constantly discharging the highest cell. This is true no matter if you start it early or late. So I'd say it is a good idea to start the balancing as early as possible, as soon as you hit the steep part of the voltage curve where the voltage is proportional to the SOC.
I agree with you that a small balance current of 2 A might not make much of a difference to a high charge current. So the actual effect of early balancing might be very small. But it is still better than nothing. My main point is that starting balancing early has no negative effect, and can only have positive effects.
If you have problems with the BMS balancing your cells, I think that setting the absorption voltage a little lower than normal has a much bigger effect than changing the Balancing Start Voltage in the BMS. Then you get a much smaller charge current at the end of the charge cycle, like you mentioned, and this will make the job of the BMS much easier.
Andy, if BMS-only (no active balancer), I am seeing high .050 Delta voltage (and higher) around 3.3v. I was seeing even higher Delta over .15v last week. The battery bank is about 2 years old (48, 100ah CALB cells in a 6P8S configuration with a DALY 500amp BMS). Having started balancing at 3.3v at .2 Delta, the pack reaches about 3.4v with .25v Delta during Equalization. I have ordered a 1a smart active balancer to add to the system. I may boost balancing start voltage now that the cells are reaching lower Delta.
I have my BMS's and inverters set just like this. Of my 6 packs only 1 is 280AH all the rest of em are all over the place after a year in use. 210AH, 157AH, 215AH, 227AH, 241AH. Ive lost some serious capacity using an overkill BMS. I recently bought 64 305AH and am switching to an active balancer, JK on those. Hopeful the BMS doesnt let the packs get woefully out of balance
On a JBD BMS, allowing it to passive balance during discharging also works wonders, given a month (yeah, I tried) you can get a perfectly (within 5mV) balanced pack even if you started out with a not entirely top balanced pack. Takes a few over-voltage disconnects on the charge end, though.
Thanks!
Thanks a lot for your ongoing support, Sree! 🙏
Another major point is that the charge current will push up the cell voltage and effect the balance start. If the balance start voltage is set too low, high charging current can cause the balancer to start way too early (before the cells are really past 90% SOC).
This is why, by default, the balance start voltage should be at least 3.50V (14.0V in an s4 battery). In otherwords, well into the exponential part of the charge curve. Because the device needs to work properly across the full current and temperature range of the system.
But if you know for absolutely sure that you won't be charging at high currents... say you never charge at more than 0.2C (very common for solar + battery systems). Then you can safely reduce the balance start voltage to 3.45V.
Another factor is the bulk/absorb target that people set. The charge target should always be at least 0.05V (per cell) higher than the balance start voltage. For many reasons not the least of which being measurement error between devices and ensuring a full charge across the whole working temperature range (colder temperatures need higher bulk target voltages, for example).
When we combine all of this together we get a balance start voltage of 3.50V/cell and a charge target of 3.55V/cell as a really good default. And if someone really wants to optimize for low charging currents, a balance start voltage of 3.45V/cell and a charge target of 3.50V/cell is reasonable. I would not go below those parameters, though.
I know a lot of people do and swear by it, but going lower really begins to restrict the operating temperature and current ranges for proper charging operation without really giving much benefit in return. The cells won't really last any longer with lower targets, particularly in cyclical applications such as solar, RV, and boat setups. As long as the charge controller properly goes into a float stage 1-2 hours after absorb with 3.35V to 3.375V/cell for the float stage, the battery will have a great life-span. Even if it is being charged to full every day.
-Matt
Thanks for your comprehensive comment, Matt.
The topic is far deeper than I pointed out in the video. More settings, different setups and designs, charge current, target voltage, absorption time and much more to consider....
Andy, as per your video from a year ago on Daly balancing, you for the Daly active balancer you found the lowest cell had to be above the set starting voltage of 3.3V for balancing to start, that is, all cells must be over 3.3V default to start active balancing. And for the Daly passive balancer in the BMS it is set to start at 3.2V Daly default, but some cells might be higher e.g. 3.4V, and I'm not sure if all cells or highest cells need to be over 3.2V. So what pack charge (absorption) voltage would you target for the Daly, say, 24V system? I am thinking keeping my setting at 3.45V*8= 27.6, or moving the setting up to 28.4, the higher number due to your recent testing videos of getting your passive balancing to work better for battery shelf. If today's video is correct, raising the setting on the Daly from 3.3V to 3.45V may mean that balancing never starts at all😮 because I've notice once absorption voltage is reached the charge current basically tapers rapidly to zero (within one hour) and the voltage is held steady at the target.
That could depend on what Daly BMS/balancer you are using. I have seen them acting differently. But you can test this and the app will show you if the balancer is turned on, so you can check if on cell or all cells need to be above the threshold.
27.6V seems to be a good absorption voltage. As you said, the current will go down quickly and the balancer should start if ONE cell goes over the 3.4V you have set. If the balancer waits for all cells, it is total garbage anyway. At 27.6V you goal is 3.45V for every cell, The balancer catches the ones which are above that, nothing else. and if you cells are all at 3.45V, the balancer will not start at all, correct.
Thank you for taking the time to share your experience in practice. You mention top balancing due to the cell charge curve, and if I follow Stuart's circuit logic correctly, an active capacitive balancer will equalise by averaging the eg 4 cells by switching them from series to momentary parallel connection. You mentioned in your video that your video today applies to both active and passive balancing. I don't know how passive balancing works but I might guess it is resistive burning away of cell capacity by choosing the highest one or two cell at a time. This will result in loss of charge over time and the balancer will stay on until the set balancer voltage is reached, this all cells are slowly brought down even if it is one at a time. Perhaps in practice only the highest cells will come down before the battery pack is discharged via usage anyway. Whereas the active balancer will top balance by averaging, bring cells above the resultant average down and the cells below up to average. Actually, it is more precisely an RC capacitive resistance (switching) circuit so resistive losses too will bring all down all cells somewhat. So active balancing may be able to occur slightly below the highest target charge voltage if it's still high enough in the charge S curve. Also, the RC capacitive balancing logic will have its own curve which would work faster at higher cell deviations. Without knowing all the circuit vakues I would like you have suggested simply have to test and see. Thanks again
Hi Andy! I want to thank You very much for Your videos! I learned a lot from Your channel and finally made my own battery 8s with daly bms and neey balancer. Im very happy of it! )) Longlusting backup power supply for winter blackouts! Fantastic!
I have a question about the charging and floating voltage in my solar inverter. I have a problem with charging (( Charge current 50A, charge v 28,6. Everithing was fine untill voltege increased up to 28,2. The amperage was still at 50A and Neey started at 3,45 to balance at 4A , but in few minutes one cell started to rise and came up to 3,652 and BMS disconnected charge( battery foult on inverter). I set the current to 40 A and voltage to 28,2 from 28,6 and in few minutes bms started again and charge continued untill 28,2 and current started to go down to 0 Amps in the end. Is that foult and bms on/off normal? I understand, that battery is not at 100% SOH becose I need 29,2v for this? Maybe I need to reduсe current manually from max to 20A when voltage will come up to 28,2 to slowly get to 29,2v(once a month for equalization balancing at 3,65)? What should be the charge and floating voltage for save battery from overcharge everyday use?
Thank You!
You have a massively unbalanced battery and need to top balance it first. First charge till the highest cell disconnects. Then cut charge and load. Dismantle the battery and connect all cells in parallel and connect a fixed power supply at 3,65V. Now you have to watch the current, if it Hits 1A. Disconnect the power supply and let them sit for min. 6h better overnight. Then again switch on power supply and wait till you hit 1a again. Now you bank is top balanced. If you cannot dismantle or have a power supply put absorption and float to 29.2 and charger to 10a and charge till it Disconnect and cut the load so the battery stays high and balancer evens out the cells. If that happens apply a load to battery lowest cell is at 3,4V. Then start charging again. Do that cycle till all cells are within 40mV and hit 29.2V Then it's time for normal operation, set absorption to 29.2V, absorption time 0s and float to 27V. Lithium doesn't need absorption, yes you loose 1-2% capacity but you can screw much more up then you can gain when you absorb. So in normal operation the neey has now enough time to do the rest and get your cells as close as possible, how close depends if they where matched cells or not.
For everyday use put HVC at 3,55V, end of charge at 3,53V absorption at 28,25V (8×3,53V) and float at 25,4V (8x3,175V). Reason float so low is your battery is kept on a lower voltage level which is healthier. Remember in discharge you have a voltage drop so battery is at around 95% SOC then.
I'm confused with your 38.6/38.2V what voltage is that? Do you have a 24V system or 32V system???
@@OffGridGarageAustralia sorry, I mean 28,2/ 28,6
Andy, might you be able to mention the Miner, and how much wealth it's producing?
At some stage I will...
@@OffGridGarageAustralia Thanks.
So, balancing is primarily taking place during the absorption phase of the charge cycle. Starting just before the CC goes into absorption and then continuing thru absorption, correct?
That begs another question: if our cells drift a bit more than we are comfortable with and we have a relatively "weak" passive balancer (Overkill) should we not increase our absorption time so the balancer has more time to try and bring the voltages closer together?
Yes, correct. Once we hit the CV phase, we start balancing as this is out of the flat part and cells start showing their true voltage.
By increasing the absorption time, you definitely give the balancer more time. That's exactly the test I'm doing right now with the battery shelf...
If you have a few cells getting too high, you can try increasing the time and see how it goes. I had no success with that method as the balance current is just too small in these balancers.
Hi Andy, should I switch of the BMS balancer of the Pace BMS (by setting a high start voltage) when I have a 4A Neey balancer in parallel?
The BMS balancer can stay on, it does not matter, it is so small, no impact.
The active balancer will discharge the highest cell and charge lowest or even split that to all lower cells for the simple heltec active balancers. With them it makes sense to start balancing in the upper knee of 3,4V as here the devations appear and its a relativly low amount of capacity that needs to be shifted from highest to lower cells. And here it also makes sense if the active balancer even discharge the 2 higest cells and charge the 2 lowest. Here the time counts, the longer you are in this area the more capacity you can shift. You done that yourself in of the last videos about "if the passive balancer keeps up the balanced battery" and to balance the massivly unbalanced battery put the hankzor or heltec simple 5.5A active balancer and setting of charger to 14,2V Absorption and float to enable to balanace the bank. You can even run the active balancer also while discharging till 3.4V as it will still shift capacity from the highest to the lower cell correctly giving the active balancer more time and chance to fully balance the battery. It doesn't matter with active balancer if your cells get balanced to 3,45V, 3,48V 3,50 or 3,55V in this case, thats why you don't need a target voltage here with this active balancer itself, important is they are as close as possible because you will never reach all at the same voltage eg 3,55V, there is always a weakest and a strongest cell. To achieve that you just need to make sure the active balancer works in the upper knee.
Andy, ... I have four 100 aH sealed 12V batteries (LiFePO4) ... connected in parallel ... can I top balance these or reach top balance? None of the batteries has/have a bluetooth to the BMS and I had them way before I started watching your show. I haven't made my own yet. Great video.
Very difficult as you don't know what the single cell voltages are. You can only charge them to a voltage until the BMS turns off. You know then, this is you maximum charge voltage for this battery. If they still balance at this point, you don't know.
That makes sense,,, thank you,.pat@@OffGridGarageAustralia
Hi Andy, please check how the JK BMS does its active balancing.
You set a balance start voltage (highest cell needs to hit that voltage) and a maximum difference between highest and lowest cell.
You can set the balance start voltage to 3.45V on these, and the difference to 5mV.
What happens if we set the charger to 14V (4 times 3.5V) then?
Once the first cell hits 3.45V, the JKBMS discharges only the highest cell (!!) and charges only the lowest cell (!!). This means no unnecessary discharging of the cells „in the middle“. We are already in the area where we can balance LiFePO4 (as you said at 11:55). So it is a good idea to start balancing before we hit the final voltage per cell, as long as we have a strategy like the active balancer in the JKBMS has. Believe me, I have watched mine for hours already. I start balancing at 3.425V in my 4s battery.
Greetings from Germany :)
Yes, tested and shown here many times. There is also a video explaining all the settings: th-cam.com/video/xTKiOvJ_ejg/w-d-xo.html
@@OffGridGarageAustralia
Thanks for the reply. I just wanted to emphasize it is in fact a good idea to start balancing a little bit earlier on the JK BMS, since it will only deplete the highest and only charge the lowest cell. As long as you only do that in the steep part of the charging curve before the max voltage you should be fine. You can start balancing then even if the full pack did not reach the final voltage yet.
@@OffGridGarageAustralia
@@OffGridGarageAustralia
@@OffGridGarageAustralia
I guess this would come down to the implementation.
In case of a passive balancer it might just connect all of the batteries for balancing.
But if the start voltage only turns on the active balancer so it can balance only the cells with enough deviation, then starting just a little earlier should not be a problem as long as you avoid the flat part of the voltage curve.
I left the deviation setting out here as it complicates things, but you're right, the balancer may not run with the examples I brought up.
I would still not set the start balance earlier.
@@OffGridGarageAustralia I'm doing all of this reasoning in theory, but don't you run the risk of the balancer comstantly turning off and on as the charger backs off or even shuts down completely?
I suspect that at these margins the charger may effect cell voltage measurements just a little to much,, especially as balancing also loses some energy as heat.
Or is this all taken care of by the other BMS settings and margins in the balancer itself?
On my two amp JK balance BMS I raised the balance trigger voltage too 0.020 because my cells seem to be quite a bit different size in amp hour. Even though they're perfectly balanced at the top if I start charging quickly the lower amp hour cell starts to raise higher in voltage than the other cells. This would cause the balancer to create imbalances if I did a tighter threshold.
Thank You. That makes since. I changed mine again. I am starting winter and I know I will have to use the grid. I am setting my inverters to save enough for 24 hours of power grid being down in reserve. 20kw of 56kw. I think 40% isn't enough since I do not go lower than 48 volts and I do not go over 56 volts. I will try 45% which is 52.1 volts rested. does that sound right? Winter where I am sucks for Solar. 11.2kw of panels is about 1/2 of what I need and I could also double my batteries to get me close to off grid.
Good idea to keep the battery at a certain level to cater for these days when you need it.
I don't think 45% will have a certain voltage though. That is impossible to pinpoint with LiFePO4 batteries. You have to rely on the BMS or shunt for this purpose.
@@OffGridGarageAustralia I been watching and its not perfect but I have a chart and its real close. The BMS isn't connected to inverters so I have to tell them when the batteries get to a certain voltage switch to grid. I have taking them down to 46 volts so the BMS is more accurate. When doing so I have about 20kw from 52.1 to 48 volts. I double check with BMS which I have 4 JK's
One design change that would make sense for passive balancers...... Note, I said "would", is to have ALL the resistors in a parallel series configuration to maximise the current without burning up or causing deamination to the PCB, and then have the Fets switch in the cells as needed to burn off the most energy possible, This would mean that each cell can be balanced individually, I do suppose it would make the BMS's more expensive, and would require a software rewrite..... But, hey ho...... I suppose that is just me that thinks in this way...... Thanks Andy.
Yeah, making the passive balancer stronger would solve a lot of issues. I don't mind burning off the energy at this point. The battery is full anyway at this stage and energy is free of course...
It may cause other issues though if the heat of these passive balancer becomes a problem. Tight spaces and enclosures for example.
So active balancing makes kind of sense.
Ooooh man you opened a can of worms.Getting ready to test that with 4 Enerkey 10 amp active balancers on my channel
Yes, I know. There were so many emails and questions around this so I thought a simple rule to when starting balancing will help...
I guess, there will be a part 2 😁
New video lets goooooooooooooooo
00:09:30 but the balancer will Kick in only if your battery is above a certain cell deviation, right?
Yes, that is a second trigger. For this video, I wanted to keep it simple. There is far more to it. I may make a 2nd part...
Hi Andy
Are you planning to get and test the sodium batteries?
BigT
I've got 4 here in the garage for 2 months now. Just no time to test them...
You have also to stop balancing when you discharge, weaker cells with higher resistance overshoot when charging with high currents and drop fast with discharge currents, what would cause imbalancing them again if you don't stop balancing on discharge.
That's another point of not balancing too early. With the simple rule I explained in this video, the balancer would stop as soon as the discharge process kicks in.
I am using those affordable BMS'es from Ali Express (for LiFePO4 of course). Do you know if those balance at the top and 'the right way'?
Target voltage/cells in series makes sense for higher target voltages. Does the same formula apply for lower target voltages? Say your target voltage was 54 for a 16s pack, is 3.38V what you would recommend for balance start threshold?
What charge voltage you use from inverter for target 55.2v (3.45 x 16)? On my Growatt Inverters I getting 3.35v on my fully charged batteries from charge voltage of 52v.
Yeah, that's normal because of the voltage drop and the two individual measurements the inverter and BMS take.
You may have to set the inverter to 55.3V/55.4V and see how this goes...
I have a JK BMS B2A8S20P and I have two questions as I practice these things.
1) I can only find bluetooth control on the device - and it keeps me very firmly at home, even though the battery is in a fire safe place. All my other devices are in my real remote control via the wifi hotspot, such as the EcoFlow power station and various smart sockets. How can I also connect this bms device to my wifi router so that I can control it when I have to leave the house?
2) I have a few solar panels in the yard and a small 320W MPPT controller. If I want to charge this LiFePO4 battery with the current produced by the solar panels, is it necessary to find the exact settings for the charge controller software, or is it enough that the BMS device knows how to control the cells and the MPPT controller does not let the voltage get too high?
Thank you!
Thanks for your questions.
1) You can connect you JK-BMS to Home assistant with this project and control it form anywhere. I will show and test this on my channel in January. github.com/Uksa007/esphome-jk-bms-can#readme
2) You need to set the charging parameters in the MPPT and cannot rely on the BMS for that reason. The BMS is a safety device which triggers if the MPPT is faulty. MPPT charges, BMS protects.
It's a bit like making a short and trip the breaker every time you want to turn off your lights.
@@OffGridGarageAustralia I looked at the documentation you linked, but I don't really understand everything. That's why I'm waiting for January like a child lollipops :) Thanks for your thoughtfulness, Andy!
Hi Andy, I agree with what you say but it occurs to me that for LIFE04 it will always be 3.45 volts per cell, no matter the pack size, so why don't they just leave it fixed in the BMS?
You can start at 3.45V, no problem.
All historically. Most BMS have a very small balance current coming from Li-ion batteries where we could start balancing at 3.3V. Manufacturers have just kept these tiny currents for LiFePO4 as well but we cannot start that early.
what about discharging the highest cell and put the energy to the lowest cell?
That's active balancing not passive balancing as shown in Andy's example.
@@vonasi2 My English not is the best, so, maybe I misunderstood something.
@@vonasi2could you be so kind and explain how passive balancing works? And how that is “discharging” the most discharged cell, cause that makes no sense at all, that means the cells will never be balanced :-))) right? I mean, discharging ALL the cells in the same time, that means discharging, not balancing :-)))
@@NoiseSHomePassive balancing works by discharging the cell it is balancing. It removes excess charge on cells which are approaching the top of the curve earlier than other cells in the pack) This is a left-over convention from the use of say packs made from 18650 cells where capacity was maybe 1500mah and balancing works at 0.1c (1/10th of battery potential current flow) versus these bigger cells of 300,000mAh (200 times larger current capacity)
In Andy's example, the stated balance threshold of 3.40v, would mean that all of the cells are higher and will trigger balancing. So, yes all cells will begin "discharge" balancing.
If you're charging with say 17 Amps at that point and balance current is a measley 70mA, then I guess this will counteract this early beginning. However if balancing continues after 100% charge (let's say 3.45v) and is still set to 3.40v, it will discharge those cells back down under the desired charge voltage of 3.45v or whatever. If some sort of over voltage protect is set to 3.45v it may take until 3.40v or under to "recover" and remove that protection state and begin charging again.
The point is to set your balance trigger to meet two states:
1) Is somewhere in the steep end portion of the charge curve
2) Is not lower than the lowest "fully charged" voltage level (let's use float voltage).
I think Andy may have shot himself in the foot by over-simplifying his example (by not discussing cell differential and Bulk/Absorb/Float voltage levels and so on)
@@NoiseSHome Passive balancing has resistors that shunt a certain amount of current . So if the bms turn on the passive resistor for a specific cell , it shunts ( heats up) the small amount of current that most passive balance bms can do - like 80 to 150 ma. So the passive balancer only works well with smaller charge currents . Your question is how do the low cells come up ? Ideally the bms only turns on the passive balance resistors for he high cells, and then this lets the low cell continue to charge. So if the cell balancing is only at the ma value of it's resistor then that cell does not charge. The cells that do not have the balance resistor turned on continue to charge.
Andy, what if my batteries are around 3.3v each but out of balance. would it be ok to balance at the 3.3v and once balanced charge to a higher voltage and rebalance again? The reason I ask is that I only have a 5 watt power supply and trying to get all cells up to the 3.45v each is going to take awhile. 280ah batteries
I guess you mixed up some terms. If we talking about balancing on the charge phase, there is no such thing as a discharge (unless your balancer allows you to bypass ALL charging current for a dedicated battery). So if the balancing current is less than the charging current (which is true for almost all cases) all that balancer can do - is distribute a small part of the charging current between the batteries. So if the charging current is 20A and the max balancer current (delta) is 5A you may have as a result, for 4 cells: 25A (for battery with lowest voltage), 15A (for battery with highest voltage) , 20A, 20A. So there is no such thing as "discharge", just one battery with the highest voltage would have less charging current.
No, I didn't mix that up. Balancing while charging is pointless unless you charging current is lower than your balance current. Just as you said.
@ OffGridGarageAustralia
Balancing while charging removes charging energy from the weaker (faster raising) cell and pushes it to the stronger (slower raising) cell.
As long as the cell capacity/capability difference is within the reach of what the active balancer is able to divert from the weaker to the stronger cells, will this ensure that all cells are being loaded appropriately to their capabilities.. the weaker ones are being loaded less while the stronger ones are loaded more (same applies to discharing while active balancing is on). Over time this should get the cells within the pack on equal footing as the wear on the stronger cells is higher.
All this "top-balancing" wisdom is only useful for passive balancers.. active balancers who can divert the energy difference between weak and strong cells make this approach moot and gives us access to 100% capacity of all cells.. not just the 100% of the weakest cell applied to the whole pack that passive balancing is capable off.
I agree with you. Ok, 2 A balancing current is not much compared to 20 A charge current. But it is still better than nothing, and will slightly prevent an OVP. Starting the balancing early (90%+ SOC) will in worst case accomplish nothing, and best case prevent an OVP. My main point is that it has no negative side effects, so why not do it? It will not artificially lower all the cells like Andy claims. I think he is right about many things, but in this case he is wrong.
Whats the goal of balancing? Having all cells as close as possible. With passive balancer i agree to some degree. If your goal is balancing to 14,2V you have the end of charge voltage to 14.4V because no cell will hit the 3.55V with 14.2V end of charge as as soon one cell hit the 3.55V the charge shuts of and voltage drops so no balancing ever happens. With active balancer its completely wrong as active balancer will deplete the highest cell into
Andy, balance start voltage is not the same as target voltage, and the BMS has no target voltage for the cells. In your example you say that if we set the balance start voltage to 3.4 V, then the BMS will drain all the cells down to 3.4 V. No, that is not what will happen. The BMS will be in balance mode as long as one or more of the cells are 3.4 V (or higher) AND the delta is more than 10 mV (or whatever you set it to). It takes from the cell with the highest voltage and inserts into the cell with the lowest voltage, and repeats. There is no target voltage the BMS tries to reach. If the delta is 9 mV and all cells are above 3.4 V it will still stop.
That's not quite what I said. In my example, we are charging to 13.8V which is 3.45V per cell if full balanced. Setting the balancer to 3.4V, makes it discharging/balancing any cell above that 3.4V, even you try to charge to 3.45V.
@@OffGridGarageAustralia No, this will not happen, because the BMS only balances two cells at a time, and it stops balancing as soon as the delta is lower than 10 mV, no matter if the cell voltages have not yet come down below the starting voltage. An active BMS doesn't eat up any charge from the cells, it just moves the charge between cells. So if you say that it will lower the voltage of all cells, you need to explain where all this charge ends up. As heat? No, this is not how it works. And active BMS doesn't have a power resistor it can use to burn up energy.
Here is what the BMS does when it balances: The cell with the highest voltage is discharged into a supercapacitor, and that energy is charged into the cell with the lowest voltage. Then the BMS measures the voltages of all the cells again, and repeats the procedure. It could be the same two cells this time, or two different cells. So in your video at 6:20, the 3.45 V cells will be left completely untouched by the BMS, because they are neither the lowest or the highest cells, and will never be. The BMS only does balancing as long as two conditions are true at the same time: A) one or more cells are above the starting voltage, and B) the difference between the lowest and highest cell is more than 10 mV. In your example, condition B will happen before the BMS has any reason to touch the 3.45 V cells.
This is also why starting early with the balancing (setting a low starting voltage) does not have any negative effect on the balancing. It will end up exactly the same as starting late, except it will get the pack balanced slightly earlier, because it would have had more time to do it. There is probably very little time difference though, because the charge current is so much higher than the balance current.
Ok this makes sense to me if I was using individual cells that I could check the voltage of each cell. But I am using sealed Drop-in Lifepo4 batteries (2 x 12.8v 100 ah )and have no way to check the voltage of each cell . How do I balance the cells in each of these batteries??
Any comments on the Yoshino Solid state battery???
no.
@Off-Grid Garage
Thanks a lot for Your great videos! Thankfull I have learned a lot.
But this is the first time, I didn't understand, or You are wrong. It's about the explanation beginning at 6:30
You say, the balancer would discharge every cell, which is above the balance-starting voltage. So the balancer would just produce heat.
But an active balancer takes energy from a higher cell, transfers it in his buffer storage (capacitor) and transfers it to a lower cell (to charge it).
In your example, the balancer would discharge cell 2 (3,48V) and charge cell 1( 3,42V).
I´m quite sure, that cell 1 would not be discharged here.
Could you explain, please ;-)
Hi I have a 280AH battery pack 16s which JKBMS do you recommend me must take into account that I also have the NEEEEEEEY 4th and a 5kw voltronic inverter Thank you but it is imperative that it must communicate with inverter ! Thank you your videos as well as being constructive, they are also funny ;o)
You have a JK and a NEEY??? Are your cells that much out of balance that one AB cannot handle it?
You can make any JK BMS talk to your inverter...
Hi Andy, I don't know how else to ask you a question :
I have a Growatt SPH10000TL3-BH-UP inverter with lots of power, but one challenge : this type of inverter has maximum 25A battery load and minimum operational voltage of 100V. If I connect at least 34 cells in series, I get to that voltage, but can I build multiple 16s battery packs each with a bms and daisy chain them?
That totally depends on the BMS in your 16s batteries if they can handle a series connection.
100VDC is dangerous and not recommended for DIY work.
I think jk BMS default of 3V is much smarter, and a cell deviation of .005. Cause what will happen is you will trigger over voltage protect just because one cell hit it when others are lower, thus battery never gets fully charged quickly then
Other issue is let's say you discharge to 10%, could take days or weeks to get battery back to 100% if many cloudy days, it should be balancing overnight so we don't start next day again with to much deviation
If you were starting from the beginning now, would you chose to buy single cells and a BMS or ready made batteries with internal BMS?
I think he discussed this a bit in a recent video. If you have the time to build your own that is an option. But with the price that some of the batteries like Jakiper are selling for , it gets tough to get all the components and case separate without coming close to the pre-made battery price. The only difference is you could these days probably get a better bms than some of the built in ones use. Besides the single cells and bms - you need connectors between each cell, breaker, wires, some sort of cell compression, and terminals.
It's good to have the options. Price wise, it's not much of a difference and both options have pros and cons.
I personally would still build my own battery like I did in the battery shelf.
Andy, can you explain why bottom balancing doesn't make sense? For example: if one cell is depleted prior to the others, wouldn't it make sense to top up the empty cell and continue to run the battery back down?
Cheers mate - love the videos.
Not possible with passive BMS.. u can do that with strong active BMS that are able to get the energy the weak cell(s) hasn't got taking it from the strong cell(s).
But if the active BMS isn't strong enough.. say weak cell is 70% of strong cell and we draw 10A from the pack while the BMS active balancer can only transfer 1A.. it won't be enough, it needs 3A (or maybe even 6A in the case if JK BMS by how it works) to divert enough power from the strong cell to the weak cell for the whole pack to sustain 10A without discharging the weak cell faster than the other cells.
You want to get as much energy INTO the cells as possible. Top balancing means you can do that thus gain Amphours. I may be wrong, but AFAIK bottom balancing was a concept that made provision for not damaging a weak cell in higher internal resistance cells such as lead-acid, as that would increase the one cell's internal resistance out of proportion with the others, resulting in slower charges and imbalanced charge voltage distribution across the cells. With low and pretty consistent internal resistence across the voltage range, that's not an issue. I think ....
@ henniemouton229
'top-balancing' as the term suggests is equalizing the voltage of all cells at the top - regardless of their respective individual "fill"-level (SOC).
They then get all discharged and the first (weakest) cell reaching the UVP will trigger the BMS or the battery will be charged back up, whatever comes first. If the cells are somewhat equal they will all wind up at the top-balance voltage without a problem.. *but the whole batteries performance will be determined by the WEAKEST cell this way.*
Active balancing that is on 24/7 at all states will balance the cells at all times and thus allow stronger cells to support weaker cells by moving energy between those (during charge or discharge). This gives the weaker cell some support and allows the use of storage capacity of stronger cells, making the weaker cells appear "stronger", thus increasing the performance of the battery as it's not the weakest cells performance is what determines the performance of the pack.
If the balancer is strong enough to cover the discrepancy between the weakest vs the strongest cell this will allows of the use of 100% capacity/performance of ALL cells in the pack even..
What guage wires are people using for active balancers? I've used 28awg cable and I think this provides too much resistance over the 2m leads and gives false voltages at the balancer. The best I can do is 70mV deviation across 16S at ~3.55v each.
It depends a bit what kind of active balancer you're using. Usually, it does not matter as the balancer measures and balances at different times.
The balancer cables need to carry the maximum balancing current the balancer is capable of delivering and should be sized accordingly.
Say 2A balancing current, 2m (why so long?) the resistance over AWG 28 (~ 0.3mm^2) copper cable will lose ~400mV (x3.4V ~ 1.3W) over the cable.
A rule of thumb is 1mm^2 per 10A of current for free hanging cable when it comes to heat dissipation being done via convection. Buried in soil, clumped together or within conduits it's less, due to the heat building up faster.
The other point is how the connectors look like at the BMS, no sense in running AWG10 (2.5 mm^2) for the balancing cable and then only having tiny 1A crimp on terminals that wont fit.
First thing I would try to do is get the BMS closer to the batteries.. those long balancing runs carry the full voltage of the cells/banks and if not routed safely can start fires IMHO.
Are you sure the ‘balance start voltage’ is the voltage over which all cells will be discharged and not the voltage over which ONLY the highest cells will be discharged?
This is correct for active balancers!
@@ClausMallorcaso you say that “passive balancing” is just a stupid all cells linear fix current discharging?!? This makes no sense at all. On the other hand, on software we see that at a time, the BMS is balancing 2 cells: the strongest and the weakest.
All the time. And this is not done by discharging both, that would be nonsense and we clearly see that the weakest is charging when the strongest is discharging…
@@NoiseSHome Most bms have a differential setting that has to be met to start balancing. So for example if 20mv differential is set in bms, no balancing will happen at all until the cell deviation is above 20mv. If one cell is 20 mv above all the other cells - the high cell will balance ( bypass resistor turns on ) but the other cells will keep charging to whatever the charge setpoint is . If the low cells raise voltage, and the high cell drops , the deviation turns off balancing below 20mv differential.
Some BMS boards also require all cells to be above the trigger level. It comes down to the manufacturer's implementation. Simple balance board or BMS boards will simply have an analogue circuit (ie not smart ["computer" controlled], and component variance means that the trigger levels will differ slightly, and generally fades in or out instead of a sharp turn on or turn off) for each cell which has a set voltage level where a limit is over ridden (or exceeded) and the discharge path is enabled. This is the most basic of balance mechanisms.
My example was charging to 13.8V but starting balancing at 3.4V. This would then discharge all cells with a passive balancer.
true that is the correct way to balance,
but as far as i know (correct me if I am outdated);
1. the NEEY smart balancer starts balancing when the average is above the set value, so the fewer rogue cells you have, the lower the average, and sometimes balancing will never start.
2. capacitor balancers rely on the value of the first cell to start or stop balancing, so if the first cell is lower than the other cells, it could lead to the balancer never balancing.
Thanks for your questions:
1. The NEEY will turn on balancing if ONE of the cells hits the set threshold and turn off once ALL of the cells are under the stop voltage. This is with the latest gen4 hardware and software on the balancer. It is not using the average voltage for that as in previous version of the NEEY. Watch this video here where I tested this in all details: th-cam.com/video/jpG3siEzll8/w-d-xo.html
2. Capacitive balancers balance all the time, so there is no start or stop voltage. However, they have a low voltage disconnect when cell1 goes under 2.7V to stop further discharging the battery.
More about this in the next video.
@@OffGridGarageAustralia
1. ah yes the newest 4th Gen NEEY doesn't rely on the average anymore, I have couple of dozen 3rd gen, that is why I set my balancing a bit lower, I personally like the the JK Active balancers more, you set the config once, and store it, when it is connected again, it is ready, no need to change anything or activate it, the NEEY on the other hand, turns off when disconnected and you need to manually enable it again. that is why i use NEEY for my batteries since I can intervene, and i use JK Active balancers for remote batteries and client batteries.
2. as for the capacitive balancers I ordered a big quantity last year and customized the starting voltage to be higher, but was a bit disappointed after receiving them and noticing that the first cell's voltage is used to start and stop
Is balancing really only that you discharge the cells over a specified voltage? I though you took current from high cells and pushed that as extra into low cells.
You described active balancing and not the passive balancing as shown in Andy's example.
Both, active and passive balancing should start at this voltage. This was the main point of the video.
@ OffGridGarageAustralia
Passive balancing gives the weakest cell the leading role, thus late to balancing is preferable as that is the only point in the charge curve where it's easy to decide.
Active balancing allows for the weakest cell(s) to be supported by the strongest cell(s) and thus allows for higher capacity in less well matched battery banks. Early balancing is preferable for those applications as it spares the weaker cells and wears the stronger cells more, bringing the cells in the bank to a closer match. But this requires the balancer to be able to divert as much energy as the difference is between the weakest and the strongest cells during dis/charge to work - otherwise we're back to passive balancing.
Hey Andy. Whats the best bms for 12v system?
The JK for 4S batteries: off-grid-garage.com/battery-management-systems-bms/
The BMS that I am using 250A Muller Energy 4s. I have everything in the bms configuration correct and still its having issues with cell one. I thought active balance would help with OVC OCC
And that's even after a full top balance and bringing cell 1 in line with the others at say 3.6V?
Hi Andy, why cant i set my calibrating voltage to more than 13.52v? I am using the jk bms on a x4 cell, 60Ah battery.
I don't know how your balancer works, but mine doesn't start balancing until both the "Start Voltage" & "Delta Voltage" parameters are met. First, a cell's voltage must be above the "Start Voltage" and then the highest cell voltage must be more than the lowest cell voltage by the "Delta Voltage". Then and only then will the balancer balance any cell who's voltage exceeds both these settings and only the cells that do exceed these parameters will be balanced. Your way will only balance when your cells reach their maximum voltage. Due to the flat voltage curve it doesn't make sense to start balancing until the cells reach the knee and the voltage starts to rise quickly.
Hi Andy, How do you set the voltages on the charge controller is it below the cutoff voltage of the BMS or equal or higher? Gruß aus DE
below.. the cutoff of the BMS is the catch all worst case that shuts off the battery pack to prevent damage. The charger is supposed to only deliver up to max charge voltage and not more.. if it fails and goes above that the BMS is there to protect the battery, not to do the job of the charger to stick to it's maximum output voltage.
@@joansparky4439 yes, that's how I set it up for me too, the BMS is the last line of defense. let me ask the question another way 🙂The 3.45 volts is the absorption voltage at the charge controller is this still enough for balancing? Or does the Charge Controller and BMS play Pingpong at 3,45V
@ frankk3873
This depends on the different cells performances, the capability of the balancer vs what is being demanded of the bank and lastly what BMS it is.
Passive balancing is there to get all cells to the same ceiling (pulling down the weakest as those reach it first) and then letting the weakest cell determine the performance of the whole bank - so well matched cells are preferable and the balancing voltage set how it's shown in the video makes sense.
Active balancing OTOH is able to divert energy from/to the weakest cell to/from the strongest cell and allows for less well matched cells. If strong enough it gives access to 100% capacity of all cells, but that depends on what demand the battery is seeing (if that overwhelms what the balancer can divert, it can't compensate). Here balancing should be running all the time, to make that possible.
@@joansparky4439 Ok your second answer makes more sense. I'll have to see what the Victron does when the cell voltage goes down again due to the active balancing, whether it changes from float to bulk again. Balancing is certainly not lossless
@@joansparky4439 I have a 100ah, 16S battery pack with the 2A jk bms active balancer. the charger is a voltronic clone hybrid inverter with only bulk and float settings, no absorption time. i use 3.45v as bulk voltage and 3.375v as float, max charging current set at 50A. A friend advised I set the BMS start balance voltage to 2.9v & 0.005v trigger voltage difference. his theory is that it will provide an even discharge from all cells rather than the weakest cell hitting UVP while the rest cells still have juice. He says that if batteries do not reach full charge everyday (3.45v), then the cells will not get balanced. based on Stuart's video, this theory seems okay. what do you advise? my questions are:
1. do I change the balance start voltage to 2.9 or keep it at 3.45v if batteries will not be regularly fully charged.
2. should I set the bulk and float voltage to the same value, as my inverter has no way to set absorption time?
p.s. the inverter system is basically my main source of power, so its loaded 24/7.
thank you for your time.
Andy, make us a graphical flow chart of when to use what type of balancing start and stop voltages. Seems to me this is not cut and dry with one method.
The start voltage should be very straight forward. Divide your charge voltage by the amount of cells in series.
Pozdrawiam z Poland
TY, i in winter have clouds about 4+ months, i (pavyziu(jealous), sun), ( we are on different sides of the earth)...☀
Yes, we are...
Well, im pretty new to this all. I just got my Victron system running and installed with a JK bms active balancer.
I still need to charge the cells to 3.5 ( victron said that charging higher then 3.45 is pretty pointless, because not much power is absorbed above that ) , but ive set the start balance voltage at 3.4. It seems to work pretty good, it takes power from the highest cell to the lowest, just how you would expect.
So how is this not good ?
I understand that balanceing at a lower voltage is pointless, but 3.4v as a start, again, im a total beginner at this, doesnt seem wrong to me.
I think I have explained that in this video.
If your target voltage is 3.5V why would you balance cells which are at 3.44V for example?
@@OffGridGarageAustralia Well, I would say, that if you charge them to 3.5v and also start balancing them at 3.5v , they would drop quickly below 3.5v ( when battery is in use ) , and thus would stop the balancing program. So I imagine you NEED to start balancing below your set charge point.
But I agree, balancing as high as possible seems to make the most sense.
Ideally you should start balancing after 1hour of rest after charging, but in the real world this probably does not work.
@@edwinbruckner4752The charger will keep the battery at constant voltage at this set target voltage (CV or absorption phase), so the battery cell voltages will not really drop as they get recharged right away again until they are all equal and the charge voltage is the same as the battery voltage.
@@OffGridGarageAustralia
The weakest cell will reach the highest voltages fastest.. to spare that cell it would be wise to start balancing early to keep it's voltage lower and divert the energy it would be trying to store to cells that can easily take it (the ones who are laggards and don't raise as fast with their voltage). This increases the wear on those stronger cells and fills them up faster.
So early balancing with active balancers actually is preferable as it saves weak cells and wears strong cells more and ideally (if the balancer is strong enough) should even enable more capacity in the bank as the weakest cell is supported by the strongest cells now.
For passive balancing with it's minuscule energy burning capabilities in those tiny resistors this is not working and late top balancing is the only viable way, while having to live with the weakest cell determining the overall battery performance.
Active balancers simply allow for less well matched banks to work as good as better matched banks.
Like# 6 - put a bounce back filter on your addy
You made it, Paul 🥇
Woohoooooooooooo 😊
I'm wondering if there is any real capacity difference between charging to 3.45V and 3.55 or 3.65 if they're all followed by absorption and float (at a lower voltage).
Andy did a video on what the capacity differences are at the different voltages. It isn't a lot of difference at the voltages you talk about.
That's the video: th-cam.com/video/pijPu7t-akM/w-d-xo.html
Thanks, Larry.
👍👍
anyone knows is it enough to use 1A balancer with 48v18a LFP? using for ebike 35-45amps, daily charged.
Can you parallel multiple MPPT chargers to one battery?
And when yes, can you also connect to the same solar panels to increase charging current? Or only different panels to one battery?
You can parallel as many MPPTs as you like and connect them all to your battery. Each MPPT can only have dedicated solar panels connected to it. The panels cannot be shared amongst the MPPTs.
3.42V is the ideal balance start voltage. You have to start early to not overshoot 3.65V
Ahm, no... It's not that easy.
Thanks a bunch. This is why I am subscribed. You hit the point and very educational. I have it balancer at 3.45. why would someone want to set the balancer at 3.55? Capacity? Use Application?
In the last few videos on passive balancing, Andy discussed that this higher start point gives the balancer a better voltage differential between cells and therefore will result in balance current being higher. At this point in the curve you see bigger differences in discrete cell levels, and this is where balancing is 1) required and 2) most effective (ie higher balancing current). But he has shown that the passive balancing schemes continue to struggle to bring a pack into balance when the charging current is larger by orders of magnitude than the balancing current.
@@CollinBaillie If solar charge controllers could limit their charging current to a smaller amount at a setpoint voltage that would be great for balancing. But it doesn't seem like any can do that.
Yes, there is much more to it: the deviation setting of the balancer as well as the actual charge current at time of balancing. Charging with 100A and the JK balancer can do 2A balancing, does make much sense and difference.
Unfortunately, most systems cannot be fine tuned that the actual MPPT is slowing down when hitting CV.
With seplos bms the neey should start at the overvoltage warning value
Depends on how you set the warning levels...
@@OffGridGarageAustralia with victron in offgrid like yours you have totaly other settings. But most users have an on grid System and then there is the best solution to let All settings like Standard just put the Pack overvoltage protection a little bit lower.
56.6V (3.55V) i think is best
Overvoltage warning at 56V (3.5)
And here there should be the Start voltage. So the neey starts balancing at 3.5V when the battery is Charged with only 10A to reach the Charge voltage. Just a user experience
💙👊😎
Hm i think you are wrong. Neey starts at this voltage but does Not discharge any of the low cells.
He always discharge the high and Charge the low. Never the less How big the voltage is over the Start voltage. And this is a good thing
Andy was mostly talking about the passive balancers in most bms , not active balancers like he Neey.
The Neey is an active balancer that works across all cells. It opens the charging circuitry (of the balancer, not the battery pack!) To all cells to charge the capacitor(s). Then it disconnects the charge circuitry and opens the discharge circuitry to all cells to charge lower level cells. So, it effectively does what you say. The Neey also considers voltage differential between cells I believe, so it won't activate until there is a significant difference between cells.
There are "active" balancers which move charge between neighbouring cells until it reaches the lowest cell. These work differently in that they will try to move charge from a higher cell to a lower cell. If your trigger level is too low, then all cells could be considered to be high cells (because they are all above that trigger). These are usually inductive balancing in nature vs capacitive.
Then there is passive balancing. Which just discharges cells which are above the trigger level. Usually cells have individual detection and discharge circuitry, so they are never in an All-Discharging state because of a single cell exceeding the trigger. But in Andy's example of 3.40v all of the cells do exceed that trigger so they will all have discharge circuit enabled.
The examples may have been over-simplified for the discussion. It gets pretty complex, and I think Andy may have been trying to avoid that complexity.
@@LarryKapp1 okay i haven't noticed this
Yes, the NEEY or other active balancer will not discharge the low voltage cells, correct. This was more to demonstrate when to start balancing. And this does not change if passive or active balancing.
i am also developing BMS and as far as i can tell, if there is no active balancing i go with history-based balancing at LFP.
Also balancing here is not from a certain cell voltage threshold. We only balance the highest. 216s is the biggest LFP setup which runs fine.