I top balance. In either case, your battery only has the usable capacity of the smallest cell. There are 3 differences between cells, even if the cells have consecutive serial numbers, having had come off the assembly line, one right after the next. The three differences are, Capacity, internal resistance, and self-discharge rate. Top balancing addresses all 3, if you ONLY balance, during absorption/float charge. When the battery is discharged, the balancing should be OFF. Once discharged, and you begin to charge, leave the balancer OFF. Disabling the balancer during discharge and bulk recharge addresses differences in internal resistance by not falsely sensing different states of charge, due to differences in voltage drop caused by load/charge current. Waiting until it's charged enough for the current to roll off, reduces those differences. Capacity is also addressed, exactly as you described. Once you have used that hundred amp hours, you put back the hundred amp hours, and once the current rolls off, you will find that the cells are almost nearly equal in voltage again. The ONLY reason for voltage balancing, is to compensate for self-discharge rate, and should be done when the battery is in a "standby" condition, with little to no current across its terminals. Self-discharge is a time based thing, having surprisingly little to do with charge/discharge cycles, or for the most part, even depth of discharge. Self-discharge happens, whether the battery is used or not. This becomes even more critical, as the battery ages, whether by cycles, or calendar life. Differences in self-discharge rates become greater, as the battery ages. I personally, like to maintain my batteries at ~3.5 volts per cell, and, voltage balancing ON, when not in use. Using voltage balancing ONLY during float/absorption charging helps a battery to be as reliable as a single cell.
LOL... Thanks for the plug Andy, great to see the Bus Bars going into use again. I can't really comment much on the balancing issue this time, you covered just about all of it. The important thing for people to remember, is that there's no "one size fits all" solution. Each person and each setup will have different priorities and needs, so people really need to weigh up the pro's and con's, and make their own decision. That's where videos like yours are so good, because they nearly always show all the variants and explain them enough for people to make that decision. Nice work.
This is a great topic and there is definitely a solution needed for this. But even more important is what happens if your charge controller does fail, your bms opens, does that fry your inverter from high voltage too? I think we are looking at this the wrong way. We should be cutting off the solar input to the mppt on high voltage cell detection....not just disconnecting batteries. We need a smart bms rated to shunt solar charge current(10-50A per cell)with some relay outputs for charging, bulk inverter, and critical loads.
Yep, I 100% agree. What BMS could do that though. Also, would both charge controller fail at the same time? If the BMS disconnects, the SCC supply power to the inverter (tested multiple times). At least with Victron gear that seems to work fine.
@@OffGridGarageAustralia If the charge controller failed, then it would also trip the high voltage of the battery as long as it put out over the 200mA. See my point? It doesn't protect against a failing charge controller, it only would stop the batteries from getting overcharged....which might be worse than destroying an inverter...but both can be avoided with a smarter BMS. It still seems odd that all the BMSes use an integrated relay. They should just be the smart part and have a separate relay for over/under voltage. And you wouldn't want to put a relay on the DC side of an inverter either since when it activates again it would get hit by high currents from the capacitors. Better to switch the AC side of the inverter too.
VERY SIMPLE: Top balancing: your cells remain in an area of SOC where the internal resistance is lower. Lower internal resistance = better performances and less heat generated as well as less aging. Just check the internal resistance curve vs SOC on LiFePO4.. That is your answer. When bottom balancing, your average cell voltage ( also called NOMINAL voltage) is a bit lower than on top balancing. You get more Wh per Ah when top balancing because the total pack voltage is higher. Also ask yourself.. do the cells prefer remaining long period at 100% SOC during balancing or if they prefer remaining long period at 05 soc during balancing? At low voltage (bottom balancing) , cells get their copper anode disssolved.. At high voltage, ( top balancing) you get lithium plating..
Thank you for your comment. Internal resistance is the lowest between 25%-75%. Sometime it increases already at 50% SOC, depending on the method been used to measure that.
I have tried various strategies. One thing I know that helps is having a 2s configuration. I have an active balancer and I monitor the high cells and low cells over a period of time. I then split the 2s groups and swap the high and low cells to minimise the difference between the pairs of cells. This works. I have now set my balancer to only start balancing when the voltage difference is 0.03 volts so the balancing only operates at the top and bottom of the charge and discharge cycles. My BMS balancing is turned off because it fights against the active balancer.
@@PowerPaulAu all I did was watch the balancer and note which cells were lowest and which ones were highest and lowest and swapped one cell from each pair, and I repeated this a month later with very good results after only the second swap.
The answer is YES! All the more reason to design a system to operate in the 80/20 range or 90/20 if your feeling confident. Nothing wrong with a periodic combination of practices. Call it a health checkup. Too balancing once a month by means of an active balancer then via BMS, a low limit disconnect would give you a look at the overall health. Of course with a data log. As long as you mamage the extremes, you're good. This is why bottom balance wins. This is probably one of the best explanations I've seem on the subject. I remember when Jehu built his bus without a bms. He took some flack over that.
@@OffGridGarageAustralia yes, sorry I wasn't completely clear. Also about Jehu's van, remember they were l'ion 18650's. As in all applications, keep in mind the chemistry. I'm an old hack from the lead acid era. The inverter chargers typically have a monthly top charge of around 15+ volts to help keep down the dendrites. On any lithium, too high of a charge voltage will help create dendrites (bad). I've ordered my 100kw 85/20 bank last week. Need to start deciding what I'm going with on this management. Bottom balance at the 85/20 point with the previously mentioned scenario. It is for my business, wanting to beat down my $1100/mo electric bill. Keep checking for my obit to find out if it worked or not. Take care Andy. Alex
You bottom balance when your cells are often at low state of charge (think EVs). You top balance when they are often at high state of charge (think solar). In either case, your battery is only as good as your weakest cell. The thing an active balancer does when there is an imbalance between cells is prevent the BMS to trip while it's compensating for this weak cell, and through this compensation you can get overall more capacity out of the pack - the downside being the stress placed on the weak cell in doing this. However, all of this is rather moot since the difference between cells in general is a couple Ah, not 180Ah as per your example, so an active balancer won't put real stress on the cell, and in addition, knowing you have a weak cell, lower the Ah of the total pack to the Ah rating of the lowest cell - no more need for an active balancer, accept some 'capacity loss', and just charge to the Ah of the weakest cell with some safety charge/discharge margin (like 80/20, or 90/10). Need more power? Add a battery.
@@OffGridGarageAustralia the way I add my battery power is to have another system . I will let the main battery to be charged untill full while using the subsidiary battery .
@@OffGridGarageAustralia I would (and I do) top balance. I have two (soon 4) sets of 16 280Ah cell batteries. You just top balance the new one, make sure that both are within reasonable voltage of one another, and put in parallel. Let them 'talk' to each other, it doesn't matter, it will settle. You will always have some cell drift at the top and bottom and if you have a balancer (like yours, or active one I have in my BMS, a JK/HELTEC one) this will help. Just accept you will have some capacity loss (even 250Ah or less) per battery (your lowest cell in essence) and stay between 80%-20% or 90%-10% of that capacity. Whatever cell drift between these margins should be small and shouldn't matter.
Since the total Ah for the system is determined by that one weak cell, if it's a significant difference, then wouldn't replacing that one cell with a new one would raise the system capability to the next lowest? It may be a bit of a waste having a new cell among all the old ones, but would it be better than having the limiting bad cell in the mix or replacing them all?
Great video. The problem with balancing is voltage DOES NOT equal capacity 1:1. Every battery is unique. Balance the pack once, then disable balancing, If it looses its TOP/BOTTOM balance then you have another problem you need to fix first.
As for the second part in your demonstration there are active balancers that you can add a relay to that's what I've done to mine and then I'm using my victron equipment to switch the relay on only when I'm at a high enough voltage so they are active balancing but they're only doing it when I say so and I'm only doing it on a higher voltage that prevents them from charging the lowest battery when the pack is low and then you charge it and then it has to immediately take all that power back out of the lower capacity cell
Top or Bottom balance, your pack will still become unbalanced over time. It's not just capacity of each cell in play. Internal resistance variation is a problem, and this is a major contributor to unbalancing at high load, or high charge current. My suggestion if they are good/new cells, top balance, if they are crap mixed capacity cells made from old laptop batteries then bottom balance. Now pay attention this is key, DO NOT cycle between 2.5, and 3.6, set it up to cycle between something like 2.7, and 3.5. Also I think I might eventually recommend the heltec active balancer, I've been playing with it, and will do a video soon.
Thank you. I cycle between 3.1V and 3.45V which gives me 90% capacity. I'm still not sure about the balancer, if this is good or bad. I've got some here and have tested also. Seems to work but this constant amps back and forth... don't know...
@@OffGridGarageAustralia Right on. With the 3.1 cutoff that's fine until you have a heavy load. Then it might cut off sooner than your 90% calculation because of voltage sag. A brief heavy load such as a power saw may sag the voltage under 3.1 then your BMS cut out, but you still really have over 90% capacity. Something to consider.
@@BradCagle Oh, no no, 3.1V is not the cut off but that's usually what I have seen to be the lowest when I discharge the battery under normal usage. The inverter cut-off is set to 3V under load and to 2.8V under light load. The inverter adjusts automatically dependent on the load at this time
I've always liked the idea of bottom balancing. The BMS should be there as a redundant safety device if your charger malfunctions. Don't depend on it alone to protect from an over voltage. No one wants to start a fire.
@@jasondevine6014 Thanks for linking that video. Interesting failure. Mosfets can fail closed too. Look into an "emergency/combined disconnect" by Albright International. SD series switch. I got a used one. This acts like an emergency stop pushbutton and a high current DC contactor all in one. I'll post a video of mine when I get it hooked up.
@@OffGridGarageAustralia True what is good for setup purposes might not be good for longevity. And how much of the batteries life does it really sacrifice? Say you had to replace your pack 1 year earlier would that really matter? Maybe the convenience is worth it?
Andy, I was conflicted about this too and about adding a second pack. I've done this twice now and its easy. I use grade B 105ah cells. I ran across some server racks and chassis recently. I gutted the servers and there is room for 8s 3.2v in each chassis. Tie two chassis together for a 48v 16s. Overkill bms. I test each cell by charging to 3.55v then discharge to 2.8 for a capacity test (thats my 80% operating system range for me). Parallel 16 cells for a week then install 8s in each server rack. Two chassis together with a 48v 16s bms. I wait for a cloudy or poor solar charge day. I discharge my system 48v to 2.8/cell and hook up my new 48v 2.8v/cell pack. I have an active balancer flat cable to each pack but not hooked up. Bms balancer off. Basically bottom and top balancing once a month/pack. Then Each month I also charge the packs full 3.55v/cell and take one pack offline for a 24hr balancing hooking up the active balancer (cables already hooked up and leaving them hooked up.) Next day after my system voltage is equal to my balanced online pack, I bring everything back online together monitoring for any inrush of large current. Hope that helps. Almost ready to add additional 4.8kw. About $400/48v per pack with bms. Cells are cheap. Enjoy your laughing birds. In So California we have birds that chirp as a flock only at night....all night, Crazy.
To be pedantic about balancing... 1) check voltage to group in sets of four that are close (ideally within 1/10v) 2) with 4s bms wire in series 3) charge each 4s with 12v until BMS shuts off 4) unwire battery(s) 5) check voltage of each cell and any that are 1/10v or more different charge to close enough to others 5) rewire cells all in parallel 6) charge to 3.6v until current is negligible 7) unwire battery, rewire in desired configuration The balancer size needed is not determined by the size of your battery but by the size of your charge and discharge currents.
What you are missing is cells with different capacity and impedance will have different charge and discharge efficiency for a given current level. The higher the %C rate of charge or discharge current for a given battery, the worse the efficiencies. So putting in 100 AH's does not mean all cells are filled by 100 AH's. The difference in yield will accumulate over time and will diverge their relative state of charge AH capacity. Round trip charge/discharge efficiency at 25 degs C for 0.1C rate current for LFP cells are about 99%, 0.5C is about 95%, 1C is about 91%. These numbers get a little better for warm ambient temps and much worse at colder temps. Cell leakage is about 2% of capacity /mo at 25C, 1%/mo. or less at cold temps, 5%/mo at 40C. Bottom balancing (or top balancing) will not do anything to avoid this. The closer cells are matched the less the divergence.
Roind trip charging efficiency consist of 2 parts: coulombic efficiency (how much you get from each Coulomb you out in) and voltage efficiency (how high is discharge voltage compared to charge voltage) Charging / discharging at high rate ONLY worsens voltage efficiency (I^2*r losses), but NOT coulombic one. Coulombic efficiency (and variations of it beetwen cells) is one thing, responsible for cells staying within balance. You don't put / take same amount of energy into series connected cells, but same amount of charge
I do top balancing Bottom balancing is in my mind flawed for solar, since you rely on the pack always have the same voltage distribution when the weakest cell hits a predetermined point (say 3.6v). The sum of all voltages is your charge voltage. In theory this is great, but the cells will drift over time and cause a different voltage distribution. Then, In the case where the weakest cell peaks earlier, the bms will disconnect the charge. All good to save the pack, BUT, that might fry your inverter or charger when the rail voltage suddenly jumps up. Depending on the inverter/controller hardware it might survive a “battery disconnect” even if you have a high solar voltage. In my case the inverter throws an error and shut down, but I’ve have fried a cheap controller by disconnecting the battery (with the solar still connected).
That's exactly what I was referring to in the last two videos. I think it's crazy to bottom balance when you use a BMS as single voltages in your pack may drift over time. The overall voltage of the pack may seem fine but when voltages shift, it may trigger the BMS for a single cell overvoltage. Not good.
Very good or an excellence detail how to balance your battery, think bottom balance cells win and I think BMS should stay in with the cells for just in case.
I like your analogy of balancing buckets of water. That is the same one I have been using. It makes it clear that you are fighting a losing battle with trying to keep the level of a small-capacity cell at the same level as other larger-capacity cells during charge-discharge cycles. As you have identified, it is a tradeoff between limiting the battery capacity to the smallest cell capacity or moving current to and from the smallest capacity cell to keep it within voltage limits. While not an expert, I think that overvoltage in a small-capacity cell is going to be much more detrimental than the additional charge cycle to keep that cell from overvoltage. That said, I will point out that there is in fact typically not going to be a current reversal for the lowest capacity cell as it charges until all cells start to drop. Let's say you are charging with 10 amps, and the balancer is removing 1 amp. You still have net 9 amps into the smallest capacity cell. It never starts to discharge. To my mind, the balancing allows you to recover capacity that would otherwise be limited by the smallest cell. The reality is that a 1 amp balancer with a 200 amp-hour battery will do very little. I took some data on a 200 amp-h 4S battery with a 5 amp balancer and it appeared it was going to take 5 hours to balance out 70 mV. Finally, I would like to suggest a better alternative to agonizing over top/bottom balancing is to actually improve the cell capacity matching. Parallel Cell Capacity Balancing (PCCB) Procedure. diysolarforum.com/threads/parallel-cell-capacity-balancing-pccb-procedure.26301/#post-312069
What if the odd cell was a higher capacity than the rest in the extreme example? Is that going to cause all the other cells to go over voltage because the one cell remains at a lower voltage due to the higher capacity?
Glad you see the light a active cell balancer removes a ton of headache / hassle as I suggested several times its the way to go for simplicity and theyre power draw is almost a mute point the thing is it protects your cells from imbalance of cells dipping to low while others are in the good this can and will prolong the lifespan and use ability of your lithium cells its a win win for literally pennies compared to the rest of the hardware and such so where is the negative there is only positive in installing / using a active cell balancer.
Another option I am contemplating is doing an active top balance with batteries in series with the charger (powerlab 8) with balance leads...no pressure on the weakest cell etc.
In addition to that the c rating of the charge is is slightly higher on the lower capacity cell so you're actually charging it at a faster rate in relation to its actual capacity
Top balance if you are want to use mostly the top part of your battery (i.e. full charge more often than you full discharge). Bottom balance if you want to use mostly the bottom part of your battery, (i.e. frequently run your battery all the way out, or a full discharge more often than full charge). Either one can work. It's easier for automated solution to maintain top balance which is why so many BMS include a top balance function. However if you do not fully charge and hold it there with some current but not too much for the balancer, the simple top balancer cannot do its job.
I would have loved to see your wireless BMS that you say don't exist. People don't believe me either when I tell them I have nuclear fusion power at my place until I point to my roof.
FYI, a BMS that has feature of cell impedance readout and external current shunt will likely have greater cell voltage reading accuracy for balancing. This should allow more successful balancing at lower cell voltage giving more balancing time before highest state of charge cell over-voltage trips BMS off.
What do you need to do is do a test bottom balance load test with active BMS. Then top balance load test with active BMS and I bet you’ll find a top balance gives you more power.
I'm bidding on a 29kw hour battery, so all this is going to come in handy, after I deconstruct it and re-assemble I'll need to buy six matching cells for a 40KwH battery!!!! Talk about life changing for a 100% off grid solar guy! I can run an entire week off it if the battery is good!
I would like to see if you make a video about, bottom balance 4x 280ah cells, charge them series at 3.6v, and do a capacity test. That will prove what is the correct way. top or bottom. :)
Questions about this: - so you want me to stop charging once the first cell hits 3.6V? - how to maintain the balancing over time? - how can I have two of such batteries in parallel?
@@OffGridGarageAustralia stop charging at 3.45 volts, and do a capacity test of each individual cell to check is charging without a bms is a good idea. i think it isn't a good idea and that the capacity of each cell is not the same. and with a bms it should be the same. don't use bottom balans in your system. just make a 2p with proper bms.
Bottom balance method makes the most sense to me. However you still must understand that all of these problems come from the fact that a cell has an internal resistance and that resistance is different between the cells. By its very nature resistance consumes power. Whether you are charging or discharging, part of the energy is being dissipated as heat, IE is a loss. However, because each battery has a different internal resistance, the amount of that loss is different between the cells. This implies that regardless of what you do you will have drift between the cells over time. This implies that eventually you will need to rebalance.
For my money all this top balance / bottom balance / active balance / passive balance is just theater. The objective of a battery is to deliver current. Current (electrons) does all the work. When you hit the knee you are within a few percent of done charging or discharging. STOP! if you need more current get a bigger battery. You can spend your life chasing the final couple of percent capacity or you can enjoy your limited time on this earth. Buy a bigger battery.
My cells are charged and discharged up to 80 amps. The active balancer compensates with a maximum of 5,5 amps. This means that the cells are simply charged and discharged with a maximum difference of 5,5 amps. I don't think this stresses the cells. Only with high voltage differences does the active balancer have high equalisation currents. I do not switch off my house completely at night. I have a base load of about 150 watts, which is already over 3 apere at 48v. Li-ion/Lifepo4 Capacitive Balancer Board Balance start voltage : 3.0V-4.2V Balance termination voltage : 2.9V (into undervoltage sleep state) Maximum Balance current : The whole group accuracy is 0.1V1A ; 0.5V3.5A; 1.0V5.5A So in my opinion there is no additional charging or discharging with the balacer. I have one active blancer and one BMS (not balancing) per battery pack. China cells are all used and relabelled in my opinion. So they need all the care they can get. Best regards from Austria to Australia
It is a conundrum because as Mr. Hairless legs said, no two batteries are the same. So, I guess it would depend on one's individual circumstance. Still, there is a lot of food for thought here, especially given what Doctorbass posted about dissolving anodes and lithium plating.
Looks like really nice bus bars. I guess Al need around 50-60% larger cross area to get same resistance as Cu and the thickness of these looks like 2x the one that follow batteries, so it should be quite a lot better. :)
The best way to look after a solar battery, is to bottom balance the battery. The reason for this is that inside cells are plates that will discharge at different rates. So if we bottom balance, we can set out cut off. We can then charge the cells and use the power to the peak voltage we want. 3.5v then we don't over discharge and the batteries should balance, the deviation in the top is less of a issue.
But all BMSes only support top balancing. So do charge controllers. I think bottom balancing is something from the past, when not many good electronics existed.
@@OffGridGarageAustralia it's not that. If you have series cells and parallel then you can't monitor the cells in parallel This way you get a weak cell and it drags down the pack and then reverse charges in the worst cases. 1p batteries should not have the issue with the BMS but you can work to max power, in a top balance, A disregard for cell capacity. A bottom balance is battery life. Which finds the empty point of the cells to avoid it. But charge to the cells equal point. (Not a ideal way to explain without sleep, hope it makes some sense)
Now i am confusedererer... But i learned something. Lower voltages are on lower positions. I never noticed this before. This bottom balancing seems to be a good approach, but i would not use it without BMS, as you mentioned, for safety reasons. And i would also not bet, that it stays forever like that. You put the same amount of Ah in every cell, but internally the chemistry may do something slightly different in every cell? Also the cell degradation may vary slightly, so i think it will shift over time... But this are just my thoughts, no experience yet... And yesterday my first panels were delivered, so i had a chance to try this elejoy tester :-) Feel free to not upload a video every day ;-)
Thanks for the great video. In your opinion, if I am running an active balancer as well as my daily BMS do I need to top balance my batteries when I get them or can I just let the active balancer do that over a couple of cycles?
The balancer should be fine and will do the top balancing for you. You can leave the battery charged to 3.55V/3.6V and the balancer will work it out over night. Remember to only balance when cells are above 3.45V ("top balance").
Bottom balance all the cells in parallel once before building the battery, next build the battery pack with a BMS to Top balance from time to time, during the whole life cycle of the battery. The BMS will bring you safety and peace of mind . . . Active balancer is the preferred option, but not available on the market!
Thanks for the great videos. it will be possible to mount the second battery with down balancing. with both systems at the same time. after a few weeks you would compare the two balancing methods.
That is my problem though. I would assume the two battery strings (16 cells in series each) will not have the same overall voltage when the first cell hits say 3.45V. When I parallel them, they will balance and... destroy the bottom balancing.
@@OffGridGarageAustralia Hello, I put both batteries with a BMS. What would happen is that when the battery with bottom balancing was at 100%, it would disconnect from the system. when the battery with the highest voltage dropped to the lowest level, both returned to work. I only see it happen, but for a little while, when voltage are at the top or the bottom. if the system was programmed to work between 5% and 95% the problem was solved. What do you think??
I have tested several now and they all work more or less. You always want to make sure to not leave it connected all the time and just balance the top of your charging cycle. At the moment, I'm not using an active balancer any more but will use the built-in one in the JK-BMS once it is installed. That seemed to work perfectly fine.
If your use case is running on empty most of the time do bottom balancing. If usually running with greater then 50% SOC do top balancing. It is also more likely to damage cell from overcharging then overdischarging.
Aeh, I think it's the other way around, cells damage very quickly at under 2.5V while overvoltage can go up to 4.2V with these cells. 3.65V is the max recommended voltage. My battery rarely gets under 40% so I'm staying rather in this 2nd half as well.
@@OffGridGarageAustralia You can take a cell down to 1v if you recharge it soon after. If left in over-discharged state for long time (days) it will grow lithium metal dendrites that can short out cell. This is why you should initially charge from over-discharge at low rate. If it has dendrites, the cell voltage will not rise and the low current charge will avoid overheating on cell. Electrolyte will breakdown below 0.2v (or above 4.3v) assuming cell is using most common electrolyte. (lithium hexafluorophosphate LiPF6 ‘salt’ , with Ethylene Carbonate C3H4O3 'solvent'), Most damaged cells on the forums are the result of over-charging where the electrolyte breaks down and puffs up the cell with gases released from decomposed electrolyte. The other by-products from electrolyte breakdown coats the anode and cathode surfaces, clogging them up and reducing ion flow in to and out of the electrodes increasing cell impedance. The 2.5v spec is based on no significant capacity is achieved below 2.5v and if left at or above 2.5v for long time it will not grow damaging dendrites.
Top balancing focuses on the SOC (100%). Voltage is used as an indirect measurement. Bottom balancing works for very well matched cells (internal resistance) with equal cooling. This is needed to ensure that the cells do not drift with each cycle. If the cells drift with cycling, bottom balancing will work only for a limited time. Top balancing has the benefit that it can be re-done automatically at the end of each charge cycle (or at some convenient time interval, e.g. every 30 days) to eliminate cell drift added during cycling. In your previous video, one of your cells reached high voltage because the BMS caused imbalance trying to balance at 3.35V. It was NOT due to the small balancing current of the BMS. The BMS balancing current and the "absorption" duration must be big enough to allow the BMS to rectify the imbalance added since the last balancing occurred. As I've read on another forum, bottom balancing makes sense for an EV. You may be low on charge and far from a charging station. You want every possible mile of range, without risking of running the weakest cell into reverse polarity (beyond 0% SOC). For a PV storage system, top balancing makes more sense. An active balancer that runs PERMANENTLY must be able to transfer a current on the same order of magnitude as the usual charge-discharge current. But you can also get an active balancer that can run on-demand, only when needed, at the end of a charge cycle.
Yeah, I know about the reason for top or bottom balance while both don't seems ideal. My Tesla does balance all the time for example, not just while charging or driving. They have a patent for that system. The ~2900 cells are all in absolute balance all the time. It's crazy.
@@OffGridGarageAustralia For Tesla Model 3 with LFP batteries (Made in China), Tesla recommends charging to 100% once per week to balance the cells and to "calibrate the range" (i.e. to synchronize the SOC to 100%). For the USA variant with NCM cells, I will do some research. But NCM does have a steeper voltage curve, not like the flat middle curve of LFP.
I use the technique described by Jack Rickard and recommended by @Helder Pereira. Before anything, I bottom balance the cells to 2.75 by discharging them down to 2.65, and let them set overnight. Because the voltage climbs back up a little, I discharge them 1 more time and they settle between 2.70 and 2.75 Volts. Now the cells can be put in series and are ready to be charged.
@@OffGridGarageAustralia I turn off the balancing in the BMS and just use it for protection. I like to charge up to 80% so the charge controller is set with an absorption of 53.2V and slightly less for float.
The main reason for bottom balancing in EVs is to protect the weakest cell from going into reverse current when the battery is low and you go full throttle.
@@Doctorbasss the bottom balancing people did not use a BMS, they just used a low voltage cut off... These people were living for physics at that time🐸
@@mjp0815 So, what happens when one cell drops way below the low voltage knee BEFORE the pack Voltage drops below the cut-off voltage? Are you saying that this does not, or can not, ever happen? Bottom Balancing, then running without a BMS, is not normally done in an off-grid situation.
@@MrSummitville when one cell drops to zero before the pack voltage is at the low voltage cut off really bad things happen as this cell is driven into "reverse polarity" instant destruction. Bottom balancing allows you to use the entire pack safely in high current situations like EVs without BMS, just using voltage monitor without running into this dangerous and expensive issue. Also, the bottom knee of the kurve is nicely visible in bottom balanced packs. Another reason why early DIY EV owners liked to bottom balance their packs.... They could tell that they were empty by simply looking at the pack voltage. The charging process was a low amp thing, lots of time to catch high voltage cells.... History... Today's BMSs top balance the packs now, and monitor each cell for low voltage... Drbass rightly said.
Thanks, Nathan. 10A charge will add 10Ah/h. If you have all batteries in series: 200Ah divided by 10A = 20h If you have all batteries in parallel: 32 x 200Ah divided by 10A = 640h
Thumbs up - thank you for sharing all your experiences and explaining the "religious" topics from a scientific and mostly neutral perspective! btw: the active balancer was in fact a boost Konverter, wasn't it? 😉 Und auch ein großes Dankeschön für die tägliche Seifenoper! Bin langsam schon ganz neidisch auf Eure Flächenverhältnisse in AUS...
Danke Dir, Dirk. Ja, das war ein Booster 😉 Es ist alles sehr viel groesser hier. I'm moment im Lockdown aber was macht das schon bei 30.000qm land. Guck mal auf Living Down Under fals du das noch nicht kennst ;)
thanks again Andy...If i could ask a random question please the wifi set up do you use wifi extender to the house? i have my shed about 55mts from my house router and just looking for a suitable solution to stay connected TIA..ps... i finally received my 280ah batteries today 12 weeks...stay safe..stay charged
I'm using one of these to make the 'bridge' from the access point inside the house to the carport (shortest distance). Form there a 10m cable runs to an old router (I use it as an AP) inside the garage which provides Wifi and LAN. www.tp-link.com/au/home-networking/wifi-router/tl-wr902ac/
Hello from America Andy I have a tip I think will help you, get some magnetic tool bars to hang your tools on your wall. I think I know witch cells I'm going to order they are 420 ah life4o cells in a 4s2p configuration but I still don't know what bms I'm going to get any recommendations for a 4s 500 ah bms. Thank you for sharing your experience and ideas with us, keep up the good work and stay safe
New cells with ~ same capacity and ~ same chemistry/resistance so... Leave 5% buffer on top and bottom. Use 90% or a little less. Every 3 to 6 months ( depending on the loads) top balance 3,5 -3,6 V low amps and let the BMS balancers do it's job. Repeat balancing if necessary. It should work. But as you said it's a religious thing... so...
Balancing cells with the bars is fine as long as there charge is very equal. If voltage is very different the current is way to high when connecting them.
Why not middle balance the cells? I know, no available equipment that do that and it is more complex when you can't just measure the voltage. But apart from that it ought to be best for the LiFePO4 cells if I'm correct?
This would only work with a precise coulomb count of Ah in and out of the battery. It is difficult to find a common spot for all cells where they have a known state to compare to. But good thinking!
@@OffGridGarageAustralia Yes. If you don't use balancing you could use your testers to find and charge all cells to their middle point of charge and then just use the bms to stop charging/discharging when the weakest cell goes high/low. Preferably at really safe voltages, this way no cell is stressed and the pack should in theory not get unbalanced.
Good video, well I'm going to do a bottom balance on 12v 200ah lifepo4 blue carbon battery, two of the cell are already 2.5v, im going to connect a 12v fan until stop working. Hopefully all the cells gets 2.5v, then I will connect smart charger lifepo4 to the battery until gets full. I don't have Active balancer. I hope this work.... Any advice?
Does the battery has a BMS? These days, all chargers, MPPTs and BMS are designed for top balancing. Nobody really does bottom balancing any more these days due to the fact that we now have BMS available which work reliable and also balance the cells at the top.
@@OffGridGarageAustralia yes, the battery has bms but when they were installed in series never were balanced and one of the two batteries started showing lower voltage, I opened and I noticed two cells with 2.7v.
Even on my 4.2v chemistry's (on ebikes usually) I bottom balance, but I also have a bms set to disconnect the pack if any cell gets above 4.1 or below 3.1. I prefer bottom balancing as at high current drains with a low SOC you have less risk of pushing a weak cell into a reverse polarity situation if it goes over the cliff.. You mention VW samba's with battery packs without a bms, well there was two... was being the important word.... apparently it was being stored in a field, and the panels were still connected but covered by a tarp, until they were not covered by a tarp..... I wont mention who fitted that 'system'.....
Yeah, very different scenario in e-bikes and other high current equipment. Also, using Li-ion is very different as the charge and discharge curves are linear to the voltage, unlike LiFePO4.
@@OffGridGarageAustralia agreed, but both have similar issues if you get to single percent soc and then something in your home current spikes like a heater/shower turns on. But I think the time has come to finally pull the trigger on a pack of batteries for the camper. I wanted to use lots of 100ah cells for an L shaped pack but looks like 300ah will be more cost effective which means going to 600ah instead of 500
Irrefutable fact: It doesn't matter if you top balance and shut off discharge on weakest, or bottom balance and shut off charge on weakest, or capacitor balance and get significantly more usable capacity out of your pack by doing both, your weakest cell *ALWAYS* gets the most wear and tear on it.
I think I would agree. If you bottom balance, all your cells get the same amount of cycles, but your weakest cell always experiences the highest charge voltage. If you top balance, you're putting extra discharge cycles on the weakest cell, and it experiences the lowest charge voltage. If you active balance, you're putting micro cycles on everything, and significantly more micro charge cycles on the weakest cell(s). Active capacitor balancing will always get you the highest yield, but lowest longevity. Unless somebody makes a highly sophisticated BMS that monitors capacity and modulates current draw on a per-cell basis, then there will always be a weakest link. Such a BMS could distribute the wear evenly. The best way to maximize longevity is to keep the DoD as low, and the SoC as close to 50%, as possible. E.x. Low voltage shutoff 3.0v, high voltage shutoff 3.3v, bottom balance at setup, no balancing during operation. This ensures same # of cycles for all cells, but leaves enough SoC buffer that cell life is greatly increased, and any normal deviation minimally impacts cell degradation. All that is likely required is a pack balance every hundred cycles or so. Obvious drawback is greatly reduced total capacity. So at the end of the day, unless you have a magic do-it-all BMS, a set it and forget it system will always have a weakest link that degrades exponentially.
@@MicroplaysMC I recently contacted loads of Chines LFP companies and the 5 that replied all said the same... keep the DOD between 20 - 80% for best compromise on cycle life and 2 of them further said 50 - 80% DOD for max cycles. CATL claim >8000 cycles at 50/80 you just need a larger battery due to only using 30% of what is available.
be kind to your live audience outside the window. i will bet if you put a like button out there they would press it. what i have found out with connection problems (resistance ,heat ,loss ) it to use copper never seize on all of the connections. i have not had any problems since.
I have one of my 16 cells that will not charge above 3.4 volts - as in its never seen more than 3.3xxx volts. I end up with a voltage difference of 0.125 v when my battery is charged (I charge to around 55.4 volts). I have raised this with the seller and that have said initially that this was "a dynamic voltage so you shouldn't worry about it". After shutting down the battery and checking it with a multi-meter, I am still seeing the same difference after an hour of being turned off. Seller says "within tolerance" and it wont be a problem. I call BS and think it will become an issue in time; probably reducing the life of my battery. Thoughts?
Can you manually charge just this one battery? I would charge all cells as usually to 55.4V and then connect only this one cell and charge it as well to 3.46V. So it is inline with the others.
@@OffGridGarageAustralia I ended up charging it to 3.5 volts - a little more than the other cells. It took the charge at 25 amps - took around half an hour to get there. I have installed it back into the battery bank and powered everything up. Immediately and as expected the BMS started doing its thing and taking power from that cell to distribute to the others. As of right now, the cells are all sitting 3.324 (53.18) volts - within .001v of each other - around about 88% SOC. Everything looks okay, but I wont know for sure until tomorrow when the solar pumps back up to top the batteries up. I see this behavior normally and then when I go to charge, it wont go above the 3.4 as described. Fair to say that the cell can charge, but isn't doing so in the Battery. Also a few days ago, I did terminate my balance lead again to that particular cell and even tried moving it to another position to rule out the BMS as the issue. I'll update again when I get another charge of sunlight!
@@nonicknamesremain Thanks for the update. The single charged cell may now cause the opposite as before and could trigger OVP in the BMS. Can you leave the battery on a power supply over night and keep it a55.4V to give the balancer time to do its job? That should bring it in line with the other ones again.
I intend to charge to 3.4 (54.4) then float at 3.4 (54.4) for 30 minutes. As you pointed out in a previous episode, this puts all cells at max capacity. I would also stop the charge if any cell hit 4.45v. Is there a problem with this? I would turn of all balancing. Do you think this would cause plating problems?
I top end balance, higher the voltage the better, as the voltage increases quickly so does the charging current go down, that way you get the best out of your BMS balancing current
@@MrSummitville I run my banks at 56.4 volts, and have done for over 2 years, it is a dual battery setup with Lifepo4 and Li- ion battery running together
"You are constantly transferring energy in and out of the weakest cell" - well, yes and no, yes you are, but during discharge and charge it's better to think of the balancer as *shunting current* PAST the weak cell, ie, the capacitor balance board is partially powering your load in the discharge low end state and partially charging your other cells in the charge high end state.
How do people feel about extending pv cables by soldered splice and marine shrink wrap? I ask, as I need to extend some existing cables, and there is little room in the conduit for anything else. Otherwise I have to spend £100 for new cable. Would the soldered join act as a resistor? The cable is 4mm2, 1000v . I intend to run around 1kw at @98v down these wires.
I am a total novice on electricity and all that stuff. So now I have ordered a set of 4 3.6V 280 ah batteries and bms. I do not have a bench charger. Only solar panels, a charge controller, and the bms. Will I need to find a way to balance the cells or can I simply put the parts together and go from there? Pros? Cons?
If the cells are new and you don't need the state of charge you should top balance them. If you don't have a charger or power supply, you can put the battery in series and set the charge controller to a lower value, say 3.35V. Keep measuring the voltage of the cells until they all reach that 3.35V. Slowly increase the voltage of the solar charge controller by 0.2V and monitor the cell voltages. Keep going until you hit 3.6V with all cells. It depends on the BMS how it will handle the balancing. Got a DALY BMS? Forget all the above as it won't work with these BMSs. Alternatively you can use an active balancer (~$25 for a 12V battery) and also increase the voltage slowly as above to give the balancer time to... balance. I'll make a video about how to do all this once the new batteries are in place.
In a top balanced system where each night you consume significant capacity (ie not a standby system) the weak cells will spend significant time at a lower SOC than the stronger cells. The stronger cells should therefore age faster and over years may help stabilise capacity deviation. If bottom balanced then it is the weaker cell that may age faster than the strong cells.
You are going to have two batteries, right? let's call them A and B. You could charge the batteries one after the other, so charge A, until balanced, and then charge battery B, until balanced, go back to battery A etc. when a battery is being charged, us the other one to drive the inverter, unless it is empty. This way you could hit max charge regularly and use a regular BMS. If you do not reach a full state of charge when charging battery A (resp. B), because you consume electricity from it, charge battery A (resp. B) again till full charge. There is a bit of mechanics & computer around that to organize the automatic connecting & disconnecting of batteries to the panels and the inverter. (cheers from Waterloo,BE by the way, we have good beers ;-) ).
Thanks for your comment. That would be ideal to treat the batteries separately but it's a lot of work and I would really preffer having them both connected all the time.
I used that 5 amp active balancer to bottom balance my 8s pack while still hooked up in series with BMS. I did n't dare redo the nuts on these cheap flimsy cell terminals again. I could feel these cheap terminals moving when I put it together the last times. the buss bars and holding the =/- cables while tightening the nuts prevents movement. runit down to low cell shut down, then wait for balance, then run it down till a low cell cutoff again and waited for balance again. each time the cell difference got less . till they all run out the same time.
I need help, we are assembling batteries in bulk, cant afford to have individual bunch of 15 cells Top or bottom charge as we have to assemble around 50 units per day. We are facing issues due to these imbalances although cells we use are all A grade of the same batch. My question is if we don't do the initial balancing and rely on the balancer to eventually balance it after 3-4 complete charge/discharge cycle. Is my approach correct, if not if there is a system where I can do initial balancing in bulk (let's say 150 cells in one go)
Bottom balanceing is ideal but there aren't any bms's that do anything but top balance so I don't know how useful it is and as far as active balancers they are not quite that smart they can only move energy over 1 cell I think what would work better is taking a high voltage cell and boosting to the whole pack voltage and dumping it there and taking power from the whole pack to charge an undervoltage cell and just do that for one cell at a time I think would be fine but I don't think they over cycle the battery because it is mostly doing it during charging so instead of like 10 amps it's getting like 9
Well, if you bottom balance, you would not need a BMS with balancing capabilities at all then. This would just destroy your initial bottom balance. Same for a balancer.
@@OffGridGarageAustralia but I mean cells are probably going to drift but if you are not planning on having any balancing what difference does it really make if you had them top balanced or bottom balanced they are going to drift just the same
@@MrSummitville Shenzhen Luyuan Technology Co., Ltd on Alibaba They are brand new matched grade A EVE cells and come with a test report on each cell. Mine tested over 290Ah just as the report stated. They cost more, but are well worth it. Also they come with welded studs which is a plus.
with 1 active balancer you charge and discharge your weak cell with 5 A only, does it count as charge cycles? Of course with 100Ah against 280Ah the difference is huge, but with more equal cells, the weakest cell wouldn't suffer that much. If we charge with 50A , shouldn't we need 10 of those 5A balancer in order to shift all the charge as fast as possible, so all voltages remain equal at all times? And we don't need balancing at rest.
I think the balancer will only have to work once there is a voltage difference across the cells, so at the bottom or top of the curve when cells start drifting. It should be very easy to balance such high capacity cells with 5A at this stage. There is no need to have multiple balancers, I actually think they would counter each other and start fighting.
Bottom balance without balancer assumes cell properties do not change but thats just wrong. Over multiple cycles capacity as well as internal resistance changes. The worst call will change more then others and when the cells get old this will get a fire hazard.
Aluminum bus bars are dangerous, at least in buildings. Aluminium develops a coating of oxidation that eventually causes arcing at the connections. The building where I had a condo had a huge fire because it had big aluminum bus bars instead of copper wiring. Such bus bars are not unlawful under the building code.
I have used aluminium busbars for 5 years in an RV without any corrosion problems. It might be more of a risk in damp buildings. To reduce risk of shorts I have recently started coating them in heatshrink and cutting out circles around the terminal points to enable good conduction. This would probably help minimise any risk of corrosion.
@@OffGridGarageAustralia Yes, that is what I have heard too, that Aluminum is the metal that expands and contracts the most of all metals, and when you add corrosion into the mix, the arcing occurs at the connection points.
@@JR-kk6ce Theory is often in conflict with practice, sometimes more limiting sometimes less. In this case, unless you have some extreme temperature and humidity conditions, the practice wins. In the case of humidity you will probably see the external effects of corrosion on the terminal busbar junction long before it becomes an issue. Expansion due to temperature around a terminal over the range of temperatures that lithium cells can be used isn't going to be an issue.
@@nigelcharles511 I would be careful with aluminum busbars. It is not the humidity that you have to worry about, it is the the material of the post, washer, and bolt that you are going to use. Aluminum is very reactive with other metals and will galvanically corrode. The corrosion at the connection will create hot spots and possibly even a fire. I know this because I worked many years doing thermal imaging of aluminum busbars in buildings that were built in the 70s. Once you have a hot spot at a connection, the heat expands the aluminum and when the current is taken away, it cools and contracts. Eventually, the connection becomes loose and greatly increases the risk of fire.
Can this chemistry of battery be fast charged? Can they handle a lot power to charge a bank? Also can they handle a high power draw never to be drawn down below 45%?
@@OffGridGarageAustralia Thank you I've been looking for the chemistry of battery to conduct an experiment with and this seems like the one I should use. Thanks, now my research begins on this battery. But one more question; how many charge cycles does batteries generally have? Thanks again!
You bet the weakest cell gets stressed by the active balancer! That's why I said previously that it isn't the best solution. :) The best solution is more closely matched cells.
I don't know about balancing by capacity it seems to me that you're relying on your ability to measure how much capacity you took out and how much capacity you put back in the shunts are pretty accurate but I don't know if they're quite accurate enough
All balancers take the cell voltages as measurement. As long as there is a difference, they keep either burning off energy or shuffling into other cells. A coulomb count through a shunt may work but is not 100% accurate either as it would need several other parameters as well (current, temperature, SOC) to make an accurate decision. Seems like it, we cannot really find the perfect solution.
That's wasn't a balancer at all. At this point of making the video I didn't have one so took this buck booster as a place holder. But I have balancers on my website. They work great: off-grid-garage.com/battery-management-systems-bms/
@@OffGridGarageAustralia Thank you. I was a bit confused on the way it looked. (Big toroidal Transformer etc.) Great videos . Thanks again and congrats on hitting 10k !!
I top balance. In either case, your battery only has the usable capacity of the smallest cell. There are 3 differences between cells, even if the cells have consecutive serial numbers, having had come off the assembly line, one right after the next. The three differences are, Capacity, internal resistance, and self-discharge rate. Top balancing addresses all 3, if you ONLY balance, during absorption/float charge. When the battery is discharged, the balancing should be OFF. Once discharged, and you begin to charge, leave the balancer OFF. Disabling the balancer during discharge and bulk recharge addresses differences in internal resistance by not falsely sensing different states of charge, due to differences in voltage drop caused by load/charge current. Waiting until it's charged enough for the current to roll off, reduces those differences. Capacity is also addressed, exactly as you described. Once you have used that hundred amp hours, you put back the hundred amp hours, and once the current rolls off, you will find that the cells are almost nearly equal in voltage again. The ONLY reason for voltage balancing, is to compensate for self-discharge rate, and should be done when the battery is in a "standby" condition, with little to no current across its terminals. Self-discharge is a time based thing, having surprisingly little to do with charge/discharge cycles, or for the most part, even depth of discharge. Self-discharge happens, whether the battery is used or not. This becomes even more critical, as the battery ages, whether by cycles, or calendar life. Differences in self-discharge rates become greater, as the battery ages. I personally, like to maintain my batteries at ~3.5 volts per cell, and, voltage balancing ON, when not in use. Using voltage balancing ONLY during float/absorption charging helps a battery to be as reliable as a single cell.
I think this is one of the best videos explaining the purpose and different kinds of balancing. Very well done! Thanks for your work.
Awesome, thank you!
@@OffGridGarageAustraliaYa! 👈👁️🐽👁️👍
LOL... Thanks for the plug Andy, great to see the Bus Bars going into use again.
I can't really comment much on the balancing issue this time, you covered just about all of it. The important thing for people to remember, is that there's no "one size fits all" solution. Each person and each setup will have different priorities and needs, so people really need to weigh up the pro's and con's, and make their own decision. That's where videos like yours are so good, because they nearly always show all the variants and explain them enough for people to make that decision. Nice work.
Thank you very much Paul!
This is a great topic and there is definitely a solution needed for this. But even more important is what happens if your charge controller does fail, your bms opens, does that fry your inverter from high voltage too? I think we are looking at this the wrong way. We should be cutting off the solar input to the mppt on high voltage cell detection....not just disconnecting batteries. We need a smart bms rated to shunt solar charge current(10-50A per cell)with some relay outputs for charging, bulk inverter, and critical loads.
Yep, I 100% agree. What BMS could do that though.
Also, would both charge controller fail at the same time? If the BMS disconnects, the SCC supply power to the inverter (tested multiple times). At least with Victron gear that seems to work fine.
@@OffGridGarageAustralia If the charge controller failed, then it would also trip the high voltage of the battery as long as it put out over the 200mA. See my point? It doesn't protect against a failing charge controller, it only would stop the batteries from getting overcharged....which might be worse than destroying an inverter...but both can be avoided with a smarter BMS. It still seems odd that all the BMSes use an integrated relay. They should just be the smart part and have a separate relay for over/under voltage. And you wouldn't want to put a relay on the DC side of an inverter either since when it activates again it would get hit by high currents from the capacitors. Better to switch the AC side of the inverter too.
VERY SIMPLE: Top balancing: your cells remain in an area of SOC where the internal resistance is lower. Lower internal resistance = better performances and less heat generated as well as less aging. Just check the internal resistance curve vs SOC on LiFePO4.. That is your answer. When bottom balancing, your average cell voltage ( also called NOMINAL voltage) is a bit lower than on top balancing. You get more Wh per Ah when top balancing because the total pack voltage is higher. Also ask yourself.. do the cells prefer remaining long period at 100% SOC during balancing or if they prefer remaining long period at 05 soc during balancing? At low voltage (bottom balancing) , cells get their copper anode disssolved.. At high voltage, ( top balancing) you get lithium plating..
Thank you for your comment.
Internal resistance is the lowest between 25%-75%. Sometime it increases already at 50% SOC, depending on the method been used to measure that.
Very informative video explaining the pros and cons of top and bottom leveling as well as when to use a bms. Thank you!
I have tried various strategies. One thing I know that helps is having a 2s configuration. I have an active balancer and I monitor the high cells and low cells over a period of time. I then split the 2s groups and swap the high and low cells to minimise the difference between the pairs of cells. This works.
I have now set my balancer to only start balancing when the voltage difference is 0.03 volts so the balancing only operates at the top and bottom of the charge and discharge cycles. My BMS balancing is turned off because it fights against the active balancer.
That is great, thank you so much for sharing, Colin.
Yeh I've considered that too. Rate every cell (which could take a long time), and then mix the low and high ones to create an average.
@@PowerPaulAu all I did was watch the balancer and note which cells were lowest and which ones were highest and lowest and swapped one cell from each pair, and I repeated this a month later with very good results after only the second swap.
The answer is YES!
All the more reason to design a system to operate in the 80/20 range or 90/20 if your feeling confident. Nothing wrong with a periodic combination of practices. Call it a health checkup.
Too balancing once a month by means of an active balancer then via BMS, a low limit disconnect would give you a look at the overall health. Of course with a data log. As long as you mamage the extremes, you're good. This is why bottom balance wins.
This is probably one of the best explanations I've seem on the subject. I remember when Jehu built his bus without a bms. He took some flack over that.
Thank you for your feedback and comment.
So bottom balance and the BMS without balancing option?
@@OffGridGarageAustralia yes, sorry I wasn't completely clear. Also about Jehu's van, remember they were l'ion 18650's. As in all applications, keep in mind the chemistry.
I'm an old hack from the lead acid era. The inverter chargers typically have a monthly top charge of around 15+ volts to help keep down the dendrites. On any lithium, too high of a charge voltage will help create dendrites (bad).
I've ordered my 100kw 85/20 bank last week. Need to start deciding what I'm going with on this management. Bottom balance at the 85/20 point with the previously mentioned scenario. It is for my business, wanting to beat down my $1100/mo electric bill. Keep checking for my obit to find out if it worked or not. Take care Andy. Alex
You bottom balance when your cells are often at low state of charge (think EVs). You top balance when they are often at high state of charge (think solar). In either case, your battery is only as good as your weakest cell. The thing an active balancer does when there is an imbalance between cells is prevent the BMS to trip while it's compensating for this weak cell, and through this compensation you can get overall more capacity out of the pack - the downside being the stress placed on the weak cell in doing this. However, all of this is rather moot since the difference between cells in general is a couple Ah, not 180Ah as per your example, so an active balancer won't put real stress on the cell, and in addition, knowing you have a weak cell, lower the Ah of the total pack to the Ah rating of the lowest cell - no more need for an active balancer, accept some 'capacity loss', and just charge to the Ah of the weakest cell with some safety charge/discharge margin (like 80/20, or 90/10). Need more power? Add a battery.
OK, so bottom balance... but how do I add a second battery then? They won't have the same overall voltage and will start balancing...
@@OffGridGarageAustralia the way I add my battery power is to have another system . I will let the main battery to be charged untill full while using the subsidiary battery .
@@OffGridGarageAustralia I would (and I do) top balance. I have two (soon 4) sets of 16 280Ah cell batteries. You just top balance the new one, make sure that both are within reasonable voltage of one another, and put in parallel. Let them 'talk' to each other, it doesn't matter, it will settle. You will always have some cell drift at the top and bottom and if you have a balancer (like yours, or active one I have in my BMS, a JK/HELTEC one) this will help. Just accept you will have some capacity loss (even 250Ah or less) per battery (your lowest cell in essence) and stay between 80%-20% or 90%-10% of that capacity. Whatever cell drift between these margins should be small and shouldn't matter.
Since the total Ah for the system is determined by that one weak cell, if it's a significant difference, then wouldn't replacing that one cell with a new one would raise the system capability to the next lowest? It may be a bit of a waste having a new cell among all the old ones, but would it be better than having the limiting bad cell in the mix or replacing them all?
@@wingerrrrrrrrr
Hi Andy.. I am recommending you to discuss about low frequency and high frequency inverters including low and high voltages connections.
Great video. The problem with balancing is voltage DOES NOT equal capacity 1:1. Every battery is unique. Balance the pack once, then disable balancing, If it looses its TOP/BOTTOM balance then you have another problem you need to fix first.
What is that other problem?
As for the second part in your demonstration there are active balancers that you can add a relay to that's what I've done to mine and then I'm using my victron equipment to switch the relay on only when I'm at a high enough voltage so they are active balancing but they're only doing it when I say so and I'm only doing it on a higher voltage that prevents them from charging the lowest battery when the pack is low and then you charge it and then it has to immediately take all that power back out of the lower capacity cell
Thanks for this info.
I did the same with a JK-BMS (it has an external power supply pin that can be used to turn it on during charging, when DC voltage > 16*3.45V)
Top or Bottom balance, your pack will still become unbalanced over time. It's not just capacity of each cell in play. Internal resistance variation is a problem, and this is a major contributor to unbalancing at high load, or high charge current. My suggestion if they are good/new cells, top balance, if they are crap mixed capacity cells made from old laptop batteries then bottom balance. Now pay attention this is key, DO NOT cycle between 2.5, and 3.6, set it up to cycle between something like 2.7, and 3.5. Also I think I might eventually recommend the heltec active balancer, I've been playing with it, and will do a video soon.
Thank you. I cycle between 3.1V and 3.45V which gives me 90% capacity. I'm still not sure about the balancer, if this is good or bad. I've got some here and have tested also. Seems to work but this constant amps back and forth... don't know...
@@OffGridGarageAustralia Right on. With the 3.1 cutoff that's fine until you have a heavy load. Then it might cut off sooner than your 90% calculation because of voltage sag. A brief heavy load such as a power saw may sag the voltage under 3.1 then your BMS cut out, but you still really have over 90% capacity. Something to consider.
@@BradCagle Oh, no no, 3.1V is not the cut off but that's usually what I have seen to be the lowest when I discharge the battery under normal usage. The inverter cut-off is set to 3V under load and to 2.8V under light load. The inverter adjusts automatically dependent on the load at this time
@@OffGridGarageAustralia Got ya! Check out the vid I just did on the Heltec ruclips.net/video/yM3v1-K1q4w/видео.html
I've always liked the idea of bottom balancing. The BMS should be there as a redundant safety device if your charger malfunctions. Don't depend on it alone to protect from an over voltage. No one wants to start a fire.
Especially not a Daly...ruclips.net/video/JbnFyU79Nf0/видео.html
It seems to make more sense to bottom balance but it's harder to rectify again over time.
Schneider NSX DC mccb has over/ under voltage disconnect as precautionery measures
@@jasondevine6014 Thanks for linking that video. Interesting failure. Mosfets can fail closed too. Look into an "emergency/combined disconnect" by Albright International. SD series switch. I got a used one. This acts like an emergency stop pushbutton and a high current DC contactor all in one. I'll post a video of mine when I get it hooked up.
@@OffGridGarageAustralia True what is good for setup purposes might not be good for longevity. And how much of the batteries life does it really sacrifice? Say you had to replace your pack 1 year earlier would that really matter? Maybe the convenience is worth it?
You Gotta have the PERRRRT Sounds!!!😆🌬💥🤘
Joe how is your Oscilloscope coming a long mate. Morning from Victoria mate. My new batteries are on it's way.
Good to see you here Joe, Looks like Andy picked up some good tips from you. lol
Hope it works!!!! 😁
Andy, I was conflicted about this too and about adding a second pack. I've done this twice now and its easy. I use grade B 105ah cells. I ran across some server racks and chassis recently. I gutted the servers and there is room for 8s 3.2v in each chassis. Tie two chassis together for a 48v 16s. Overkill bms. I test each cell by charging to 3.55v then discharge to 2.8 for a capacity test (thats my 80% operating system range for me). Parallel 16 cells for a week then install 8s in each server rack. Two chassis together with a 48v 16s bms. I wait for a cloudy or poor solar charge day. I discharge my system 48v to 2.8/cell and hook up my new 48v 2.8v/cell pack. I have an active balancer flat cable to each pack but not hooked up. Bms balancer off. Basically bottom and top balancing once a month/pack. Then Each month I also charge the packs full 3.55v/cell and take one pack offline for a 24hr balancing hooking up the active balancer (cables already hooked up and leaving them hooked up.) Next day after my system voltage is equal to my balanced online pack, I bring everything back online together monitoring for any inrush of large current. Hope that helps. Almost ready to add additional 4.8kw. About $400/48v per pack with bms. Cells are cheap. Enjoy your laughing birds. In So California we have birds that chirp as a flock only at night....all night, Crazy.
To be pedantic about balancing...
1) check voltage to group in sets of four that are close (ideally within 1/10v)
2) with 4s bms wire in series
3) charge each 4s with 12v until BMS shuts off
4) unwire battery(s)
5) check voltage of each cell and any that are 1/10v or more different charge to close enough to others
5) rewire cells all in parallel
6) charge to 3.6v until current is negligible
7) unwire battery, rewire in desired configuration
The balancer size needed is not determined by the size of your battery but by the size of your charge and discharge currents.
What you are missing is cells with different capacity and impedance will have different charge and discharge efficiency for a given current level. The higher the %C rate of charge or discharge current for a given battery, the worse the efficiencies. So putting in 100 AH's does not mean all cells are filled by 100 AH's. The difference in yield will accumulate over time and will diverge their relative state of charge AH capacity.
Round trip charge/discharge efficiency at 25 degs C for 0.1C rate current for LFP cells are about 99%, 0.5C is about 95%, 1C is about 91%. These numbers get a little better for warm ambient temps and much worse at colder temps. Cell leakage is about 2% of capacity /mo at 25C, 1%/mo. or less at cold temps, 5%/mo at 40C.
Bottom balancing (or top balancing) will not do anything to avoid this. The closer cells are matched the less the divergence.
Roind trip charging efficiency consist of 2 parts: coulombic efficiency (how much you get from each Coulomb you out in) and voltage efficiency (how high is discharge voltage compared to charge voltage)
Charging / discharging at high rate ONLY worsens voltage efficiency (I^2*r losses), but NOT coulombic one. Coulombic efficiency (and variations of it beetwen cells) is one thing, responsible for cells staying within balance. You don't put / take same amount of energy into series connected cells, but same amount of charge
I do top balancing
Bottom balancing is in my mind flawed for solar, since you rely on the pack always have the same voltage distribution when the weakest cell hits a predetermined point (say 3.6v). The sum of all voltages is your charge voltage. In theory this is great, but the cells will drift over time and cause a different voltage distribution. Then, In the case where the weakest cell peaks earlier, the bms will disconnect the charge. All good to save the pack, BUT, that might fry your inverter or charger when the rail voltage suddenly jumps up. Depending on the inverter/controller hardware it might survive a “battery disconnect” even if you have a high solar voltage. In my case the inverter throws an error and shut down, but I’ve have fried a cheap controller by disconnecting the battery (with the solar still connected).
That's exactly what I was referring to in the last two videos. I think it's crazy to bottom balance when you use a BMS as single voltages in your pack may drift over time. The overall voltage of the pack may seem fine but when voltages shift, it may trigger the BMS for a single cell overvoltage. Not good.
Very good or an excellence detail how to balance your battery, think bottom balance cells win and I think BMS should stay in with the cells for just in case.
Seems like the logic method....
I like your analogy of balancing buckets of water. That is the same one I have been using. It makes it clear that you are fighting a losing battle with trying to keep the level of a small-capacity cell at the same level as other larger-capacity cells during charge-discharge cycles. As you have identified, it is a tradeoff between limiting the battery capacity to the smallest cell capacity or moving current to and from the smallest capacity cell to keep it within voltage limits. While not an expert, I think that overvoltage in a small-capacity cell is going to be much more detrimental than the additional charge cycle to keep that cell from overvoltage.
That said, I will point out that there is in fact typically not going to be a current reversal for the lowest capacity cell as it charges until all cells start to drop. Let's say you are charging with 10 amps, and the balancer is removing 1 amp. You still have net 9 amps into the smallest capacity cell. It never starts to discharge. To my mind, the balancing allows you to recover capacity that would otherwise be limited by the smallest cell. The reality is that a 1 amp balancer with a 200 amp-hour battery will do very little. I took some data on a 200 amp-h 4S battery with a 5 amp balancer and it appeared it was going to take 5 hours to balance out 70 mV.
Finally, I would like to suggest a better alternative to agonizing over top/bottom balancing is to actually improve the cell capacity matching.
Parallel Cell Capacity Balancing (PCCB) Procedure.
diysolarforum.com/threads/parallel-cell-capacity-balancing-pccb-procedure.26301/#post-312069
What if the odd cell was a higher capacity than the rest in the extreme example? Is that going to cause all the other cells to go over voltage because the one cell remains at a lower voltage due to the higher capacity?
Glad you see the light a active cell balancer removes a ton of headache / hassle as I suggested several times its the way to go for simplicity and theyre power draw is almost a mute point the thing is it protects your cells from imbalance of cells dipping to low while others are in the good this can and will prolong the lifespan and use ability of your lithium cells its a win win for literally pennies compared to the rest of the hardware and such so where is the negative there is only positive in installing / using a active cell balancer.
Another option I am contemplating is doing an active top balance with batteries in series with the charger (powerlab 8) with balance leads...no pressure on the weakest cell etc.
You remind me of my 6 grade math teacher, except he was clean shaved.
Good Job. I have more idea on top, bottom and balancer. Thanks so much.
Thank s for your feedback!
In addition to that the c rating of the charge is is slightly higher on the lower capacity cell so you're actually charging it at a faster rate in relation to its actual capacity
Yes, that is true, more wear and tear on the weak one.
Top balance if you are want to use mostly the top part of your battery (i.e. full charge more often than you full discharge).
Bottom balance if you want to use mostly the bottom part of your battery, (i.e. frequently run your battery all the way out, or a full discharge more often than full charge).
Either one can work. It's easier for automated solution to maintain top balance which is why so many BMS include a top balance function. However if you do not fully charge and hold it there with some current but not too much for the balancer, the simple top balancer cannot do its job.
I would have loved to see your wireless BMS that you say don't exist. People don't believe me either when I tell them I have nuclear fusion power at my place until I point to my roof.
FYI, a BMS that has feature of cell impedance readout and external current shunt will likely have greater cell voltage reading accuracy for balancing. This should allow more successful balancing at lower cell voltage giving more balancing time before highest state of charge cell over-voltage trips BMS off.
What do you need to do is do a test bottom balance load test with active BMS. Then top balance load test with active BMS and I bet you’ll find a top balance gives you more power.
I'm bidding on a 29kw hour battery, so all this is going to come in handy, after I deconstruct it and re-assemble I'll need to buy six matching cells for a 40KwH battery!!!! Talk about life changing for a 100% off grid solar guy! I can run an entire week off it if the battery is good!
Failing to post an episode yesterday has already been marked down in your permanent record.
Oh, Sh*t!!! 😑
Good explanation. My diybms is 1a balance and so far my 2p 280ah cells seem to keep up.
The DIYBMS does not do active balancing, it just burns off the excess energy, right?
@@OffGridGarageAustralia yep
What an entertaining video! Was very informative and fun to watch! Subbed!
Thanks a lot!
Andy you are doing a great job
Thank you very much!
I would like to see if you make a video about, bottom balance 4x 280ah cells, charge them series at 3.6v, and do a capacity test. That will prove what is the correct way. top or bottom. :)
Questions about this:
- so you want me to stop charging once the first cell hits 3.6V?
- how to maintain the balancing over time?
- how can I have two of such batteries in parallel?
@@OffGridGarageAustralia stop charging at 3.45 volts, and do a capacity test of each individual cell to check is charging without a bms is a good idea. i think it isn't a good idea and that the capacity of each cell is not the same. and with a bms it should be the same.
don't use bottom balans in your system. just make a 2p with proper bms.
Bottom balance method makes the most sense to me. However you still must understand that all of these problems come from the fact that a cell has an internal resistance and that resistance is different between the cells. By its very nature resistance consumes power. Whether you are charging or discharging, part of the energy is being dissipated as heat, IE is a loss. However, because each battery has a different internal resistance, the amount of that loss is different between the cells. This implies that regardless of what you do you will have drift between the cells over time. This implies that eventually you will need to rebalance.
For my money all this top balance / bottom balance / active balance / passive balance is just theater. The objective of a battery is to deliver current. Current (electrons) does all the work. When you hit the knee you are within a few percent of done charging or discharging. STOP! if you need more current get a bigger battery.
You can spend your life chasing the final couple of percent capacity or you can enjoy your limited time on this earth. Buy a bigger battery.
My cells are charged and discharged up to 80 amps.
The active balancer compensates with a maximum of 5,5 amps.
This means that the cells are simply charged and discharged with a maximum difference of 5,5 amps. I don't think this stresses the cells. Only with high voltage differences does the active balancer have high equalisation currents. I do not switch off my house completely at night. I have a base load of about 150 watts, which is already over 3 apere at 48v.
Li-ion/Lifepo4 Capacitive Balancer Board
Balance start voltage : 3.0V-4.2V
Balance termination voltage : 2.9V (into undervoltage sleep state)
Maximum Balance current : The whole group accuracy is 0.1V1A ; 0.5V3.5A; 1.0V5.5A
So in my opinion there is no additional charging or discharging with the balacer.
I have one active blancer and one BMS (not balancing) per battery pack.
China cells are all used and relabelled in my opinion.
So they need all the care they can get.
Best regards from Austria to Australia
Thanks for sharing, great comment.
Another edge of your seat saga of As the Electron Flows…
Yeah, it looks like it... there is always more to discover and thing about.
Thanks for your massive and multiple beer donations 🍺🍺🍺🍺🍺🍺🍺🍺🍺
It is a conundrum because as Mr. Hairless legs said, no two batteries are the same. So, I guess it would depend on one's individual circumstance. Still, there is a lot of food for thought here, especially given what Doctorbass posted about dissolving anodes and lithium plating.
Looks like really nice bus bars. I guess Al need around 50-60% larger cross area to get same resistance as Cu and the thickness of these looks like 2x the one that follow batteries, so it should be quite a lot better. :)
Yes, you are right.
The best way to look after a solar battery, is to bottom balance the battery.
The reason for this is that inside cells are plates that will discharge at different rates.
So if we bottom balance, we can set out cut off. We can then charge the cells and use the power to the peak voltage we want.
3.5v then we don't over discharge and the batteries should balance, the deviation in the top is less of a issue.
But all BMSes only support top balancing. So do charge controllers.
I think bottom balancing is something from the past, when not many good electronics existed.
@@OffGridGarageAustralia it's not that.
If you have series cells and parallel then you can't monitor the cells in parallel
This way you get a weak cell and it drags down the pack and then reverse charges in the worst cases.
1p batteries should not have the issue with the BMS but you can work to max power, in a top balance,
A disregard for cell capacity.
A bottom balance is battery life.
Which finds the empty point of the cells to avoid it.
But charge to the cells equal point.
(Not a ideal way to explain without sleep, hope it makes some sense)
Now i am confusedererer...
But i learned something. Lower voltages are on lower positions. I never noticed this before.
This bottom balancing seems to be a good approach, but i would not use it without BMS, as you mentioned, for safety reasons. And i would also not bet, that it stays forever like that.
You put the same amount of Ah in every cell, but internally the chemistry may do something slightly different in every cell? Also the cell degradation may vary slightly, so i think it will shift over time...
But this are just my thoughts, no experience yet...
And yesterday my first panels were delivered, so i had a chance to try this elejoy tester :-)
Feel free to not upload a video every day ;-)
I would keep the BMS but would turn balancing off while still monitoring the single voltages.
Thanks for the great video. In your opinion, if I am running an active balancer as well as my daily BMS do I need to top balance my batteries when I get them or can I just let the active balancer do that over a couple of cycles?
The balancer should be fine and will do the top balancing for you. You can leave the battery charged to 3.55V/3.6V and the balancer will work it out over night. Remember to only balance when cells are above 3.45V ("top balance").
very gut
I like the term, very gut. I may use this in some of my videos!
Bottom balance all the cells in parallel once before building the battery, next build the battery pack with a BMS to Top balance from time to time, during the whole life cycle of the battery. The BMS will bring you safety and peace of mind . . . Active balancer is the preferred option, but not available on the market!
Thanks for the great videos. it will be possible to mount the second battery with down balancing. with both systems at the same time. after a few weeks you would compare the two balancing methods.
That is my problem though. I would assume the two battery strings (16 cells in series each) will not have the same overall voltage when the first cell hits say 3.45V. When I parallel them, they will balance and... destroy the bottom balancing.
@@OffGridGarageAustralia Hello, I put both batteries with a BMS. What would happen is that when the battery with bottom balancing was at 100%, it would disconnect from the system. when the battery with the highest voltage dropped to the lowest level, both returned to work. I only see it happen, but for a little while, when voltage are at the top or the bottom. if the system was programmed to work between 5% and 95% the problem was solved. What do you think??
Up to 11.2k subs now, congrats! Thumbs up and subscribed and 🍺$! Nice wildlife!
I sold all the 24v stuff and went 12v, no problem no more 🍺🍺🍺 Happy Amps !
Andy What active balancer are you using? The red one with the coil and caps
I have tested several now and they all work more or less. You always want to make sure to not leave it connected all the time and just balance the top of your charging cycle.
At the moment, I'm not using an active balancer any more but will use the built-in one in the JK-BMS once it is installed. That seemed to work perfectly fine.
If your use case is running on empty most of the time do bottom balancing. If usually running with greater then 50% SOC do top balancing. It is also more likely to damage cell from overcharging then overdischarging.
Aeh, I think it's the other way around, cells damage very quickly at under 2.5V while overvoltage can go up to 4.2V with these cells. 3.65V is the max recommended voltage.
My battery rarely gets under 40% so I'm staying rather in this 2nd half as well.
@@OffGridGarageAustralia
You can take a cell down to 1v if you recharge it soon after. If left in over-discharged state for long time (days) it will grow lithium metal dendrites that can short out cell. This is why you should initially charge from over-discharge at low rate. If it has dendrites, the cell voltage will not rise and the low current charge will avoid overheating on cell. Electrolyte will breakdown below 0.2v (or above 4.3v) assuming cell is using most common electrolyte. (lithium hexafluorophosphate LiPF6 ‘salt’ , with Ethylene Carbonate C3H4O3 'solvent'),
Most damaged cells on the forums are the result of over-charging where the electrolyte breaks down and puffs up the cell with gases released from decomposed electrolyte. The other by-products from electrolyte breakdown coats the anode and cathode surfaces, clogging them up and reducing ion flow in to and out of the electrodes increasing cell impedance.
The 2.5v spec is based on no significant capacity is achieved below 2.5v and if left at or above 2.5v for long time it will not grow damaging dendrites.
Top balancing focuses on the SOC (100%). Voltage is used as an indirect measurement.
Bottom balancing works for very well matched cells (internal resistance) with equal cooling. This is needed to ensure that the cells do not drift with each cycle.
If the cells drift with cycling, bottom balancing will work only for a limited time.
Top balancing has the benefit that it can be re-done automatically at the end of each charge cycle (or at some convenient time interval, e.g. every 30 days) to eliminate cell drift added during cycling.
In your previous video, one of your cells reached high voltage because the BMS caused imbalance trying to balance at 3.35V. It was NOT due to the small balancing current of the BMS.
The BMS balancing current and the "absorption" duration must be big enough to allow the BMS to rectify the imbalance added since the last balancing occurred.
As I've read on another forum, bottom balancing makes sense for an EV. You may be low on charge and far from a charging station. You want every possible mile of range, without risking of running the weakest cell into reverse polarity (beyond 0% SOC).
For a PV storage system, top balancing makes more sense.
An active balancer that runs PERMANENTLY must be able to transfer a current on the same order of magnitude as the usual charge-discharge current. But you can also get an active balancer that can run on-demand, only when needed, at the end of a charge cycle.
Yeah, I know about the reason for top or bottom balance while both don't seems ideal. My Tesla does balance all the time for example, not just while charging or driving. They have a patent for that system. The ~2900 cells are all in absolute balance all the time. It's crazy.
@@OffGridGarageAustralia For Tesla Model 3 with LFP batteries (Made in China), Tesla recommends charging to 100% once per week to balance the cells and to "calibrate the range" (i.e. to synchronize the SOC to 100%). For the USA variant with NCM cells, I will do some research. But NCM does have a steeper voltage curve, not like the flat middle curve of LFP.
I use the technique described by Jack Rickard and recommended by @Helder Pereira. Before anything, I bottom balance the cells to 2.75 by discharging them down to 2.65, and let them set overnight. Because the voltage climbs back up a little, I discharge them 1 more time and they settle between 2.70 and 2.75 Volts. Now the cells can be put in series and are ready to be charged.
What about the BMS and charge control settings?
@@OffGridGarageAustralia I turn off the balancing in the BMS and just use it for protection. I like to charge up to 80% so the charge controller is set with an absorption of 53.2V and slightly less for float.
@@SpeakerKevin Thank you.
The main reason for bottom balancing in EVs is to protect the weakest cell from going into reverse current when the battery is low and you go full throttle.
That is the job of a GOOD BMS.. balanced cells dont protect them... BMS does.
@@Doctorbasss the bottom balancing people did not use a BMS, they just used a low voltage cut off... These people were living for physics at that time🐸
@@mjp0815 So, what happens when one cell drops way below the low voltage knee BEFORE the pack Voltage drops below the cut-off voltage? Are you saying that this does not, or can not, ever happen? Bottom Balancing, then running without a BMS, is not normally done in an off-grid situation.
@@MrSummitville when one cell drops to zero before the pack voltage is at the low voltage cut off really bad things happen as this cell is driven into "reverse polarity" instant destruction. Bottom balancing allows you to use the entire pack safely in high current situations like EVs without BMS, just using voltage monitor without running into this dangerous and expensive issue. Also, the bottom knee of the kurve is nicely visible in bottom balanced packs. Another reason why early DIY EV owners liked to bottom balance their packs.... They could tell that they were empty by simply looking at the pack voltage. The charging process was a low amp thing, lots of time to catch high voltage cells.... History... Today's BMSs top balance the packs now, and monitor each cell for low voltage... Drbass rightly said.
your a good teacher please how many hours do i need to charge 32 cells of 200Ah with DC supply of 10A?
Thanks, Nathan.
10A charge will add 10Ah/h.
If you have all batteries in series: 200Ah divided by 10A = 20h
If you have all batteries in parallel: 32 x 200Ah divided by 10A = 640h
OK thank you very much
please is there any way for me to use traditional 12 battery charger to balance them up
Thumbs up - thank you for sharing all your experiences and explaining the "religious" topics from a scientific and mostly neutral perspective! btw: the active balancer was in fact a boost Konverter, wasn't it? 😉
Und auch ein großes Dankeschön für die tägliche Seifenoper! Bin langsam schon ganz neidisch auf Eure Flächenverhältnisse in AUS...
Danke Dir, Dirk. Ja, das war ein Booster 😉
Es ist alles sehr viel groesser hier. I'm moment im Lockdown aber was macht das schon bei 30.000qm land. Guck mal auf Living Down Under fals du das noch nicht kennst ;)
thanks again Andy...If i could ask a random question please
the wifi set up do you use wifi extender to the house?
i have my shed about 55mts from my house router and just looking for a suitable solution to stay connected TIA..ps... i finally received my 280ah batteries today 12 weeks...stay safe..stay charged
I'm using one of these to make the 'bridge' from the access point inside the house to the carport (shortest distance). Form there a 10m cable runs to an old router (I use it as an AP) inside the garage which provides Wifi and LAN.
www.tp-link.com/au/home-networking/wifi-router/tl-wr902ac/
@@OffGridGarageAustralia thank you so much Andy..your a champion
Hello from America Andy I have a tip I think will help you, get some magnetic tool bars to hang your tools on your wall. I think I know witch cells I'm going to order they are 420 ah life4o cells in a 4s2p configuration but I still don't know what bms I'm going to get any recommendations for a 4s 500 ah bms. Thank you for sharing your experience and ideas with us, keep up the good work and stay safe
Thank you, James. Magnet tool bars?
Question, when bottom balancing, when you drain the batteries to 2.5v do them individually or in parralel?
All cells need to be at 2.5V. You can do this by parallel them or on an individual base, it does not matter.
we like everything, meat , fish if a reason for a new video! science is a great religion
New cells with ~ same capacity and ~ same chemistry/resistance so...
Leave 5% buffer on top and bottom. Use 90% or a little less.
Every 3 to 6 months ( depending on the loads) top balance 3,5 -3,6 V low amps and let the BMS balancers do it's job. Repeat balancing if necessary.
It should work.
But as you said it's a religious thing... so...
Balancing cells with the bars is fine as long as there charge is very equal. If voltage is very different the current is way to high when connecting them.
How much current will there be if the voltage is different? Stay tuned.
Why not middle balance the cells? I know, no available equipment that do that and it is more complex when you can't just measure the voltage. But apart from that it ought to be best for the LiFePO4 cells if I'm correct?
This would only work with a precise coulomb count of Ah in and out of the battery. It is difficult to find a common spot for all cells where they have a known state to compare to. But good thinking!
@@OffGridGarageAustralia Yes. If you don't use balancing you could use your testers to find and charge all cells to their middle point of charge and then just use the bms to stop charging/discharging when the weakest cell goes high/low. Preferably at really safe voltages, this way no cell is stressed and the pack should in theory not get unbalanced.
Good video, well I'm going to do a bottom balance on 12v 200ah lifepo4 blue carbon battery, two of the cell are already 2.5v, im going to connect a 12v fan until stop working. Hopefully all the cells gets 2.5v, then I will connect smart charger lifepo4 to the battery until gets full. I don't have Active balancer. I hope this work.... Any advice?
Does the battery has a BMS? These days, all chargers, MPPTs and BMS are designed for top balancing. Nobody really does bottom balancing any more these days due to the fact that we now have BMS available which work reliable and also balance the cells at the top.
@@OffGridGarageAustralia yes, the battery has bms but when they were installed in series never were balanced and one of the two batteries started showing lower voltage, I opened and I noticed two cells with 2.7v.
Hahaha left a 👍🏼just for the fart noise. Shoutout Nacho Libre
It works!
Even on my 4.2v chemistry's (on ebikes usually) I bottom balance, but I also have a bms set to disconnect the pack if any cell gets above 4.1 or below 3.1. I prefer bottom balancing as at high current drains with a low SOC you have less risk of pushing a weak cell into a reverse polarity situation if it goes over the cliff..
You mention VW samba's with battery packs without a bms, well there was two... was being the important word.... apparently it was being stored in a field, and the panels were still connected but covered by a tarp, until they were not covered by a tarp..... I wont mention who fitted that 'system'.....
Yeah, very different scenario in e-bikes and other high current equipment. Also, using Li-ion is very different as the charge and discharge curves are linear to the voltage, unlike LiFePO4.
@@OffGridGarageAustralia agreed, but both have similar issues if you get to single percent soc and then something in your home current spikes like a heater/shower turns on. But I think the time has come to finally pull the trigger on a pack of batteries for the camper. I wanted to use lots of 100ah cells for an L shaped pack but looks like 300ah will be more cost effective which means going to 600ah instead of 500
I've never heard that keeping a cell at 3.6V degrades the cell. Are you sure?
Irrefutable fact: It doesn't matter if you top balance and shut off discharge on weakest, or bottom balance and shut off charge on weakest, or capacitor balance and get significantly more usable capacity out of your pack by doing both, your weakest cell *ALWAYS* gets the most wear and tear on it.
I think I would agree.
If you bottom balance, all your cells get the same amount of cycles, but your weakest cell always experiences the highest charge voltage.
If you top balance, you're putting extra discharge cycles on the weakest cell, and it experiences the lowest charge voltage.
If you active balance, you're putting micro cycles on everything, and significantly more micro charge cycles on the weakest cell(s).
Active capacitor balancing will always get you the highest yield, but lowest longevity.
Unless somebody makes a highly sophisticated BMS that monitors capacity and modulates current draw on a per-cell basis, then there will always be a weakest link. Such a BMS could distribute the wear evenly.
The best way to maximize longevity is to keep the DoD as low, and the SoC as close to 50%, as possible.
E.x. Low voltage shutoff 3.0v, high voltage shutoff 3.3v, bottom balance at setup, no balancing during operation.
This ensures same # of cycles for all cells, but leaves enough SoC buffer that cell life is greatly increased, and any normal deviation minimally impacts cell degradation. All that is likely required is a pack balance every hundred cycles or so. Obvious drawback is greatly reduced total capacity.
So at the end of the day, unless you have a magic do-it-all BMS, a set it and forget it system will always have a weakest link that degrades exponentially.
@@MicroplaysMC I recently contacted loads of Chines LFP companies and the 5 that replied all said the same... keep the DOD between 20 - 80% for best compromise on cycle life and 2 of them further said 50 - 80% DOD for max cycles. CATL claim >8000 cycles at 50/80 you just need a larger battery due to only using 30% of what is available.
be kind to your live audience outside the window. i will bet if you put a like button out there they would press it. what i have found out with connection problems (resistance ,heat ,loss ) it to use copper never seize on all of the connections. i have not had any problems since.
I have one of my 16 cells that will not charge above 3.4 volts - as in its never seen more than 3.3xxx volts. I end up with a voltage difference of 0.125 v when my battery is charged (I charge to around 55.4 volts). I have raised this with the seller and that have said initially that this was "a dynamic voltage so you shouldn't worry about it". After shutting down the battery and checking it with a multi-meter, I am still seeing the same difference after an hour of being turned off. Seller says "within tolerance" and it wont be a problem. I call BS and think it will become an issue in time; probably reducing the life of my battery. Thoughts?
Can you manually charge just this one battery?
I would charge all cells as usually to 55.4V and then connect only this one cell and charge it as well to 3.46V. So it is inline with the others.
@@OffGridGarageAustralia Thanks for your response. I'll try the single cell charge this evening. Watch this space!
@@OffGridGarageAustralia I ended up charging it to 3.5 volts - a little more than the other cells. It took the charge at 25 amps - took around half an hour to get there. I have installed it back into the battery bank and powered everything up. Immediately and as expected the BMS started doing its thing and taking power from that cell to distribute to the others. As of right now, the cells are all sitting 3.324 (53.18) volts - within .001v of each other - around about 88% SOC. Everything looks okay, but I wont know for sure until tomorrow when the solar pumps back up to top the batteries up. I see this behavior normally and then when I go to charge, it wont go above the 3.4 as described. Fair to say that the cell can charge, but isn't doing so in the Battery. Also a few days ago, I did terminate my balance lead again to that particular cell and even tried moving it to another position to rule out the BMS as the issue. I'll update again when I get another charge of sunlight!
@@nonicknamesremain Thanks for the update. The single charged cell may now cause the opposite as before and could trigger OVP in the BMS. Can you leave the battery on a power supply over night and keep it a55.4V to give the balancer time to do its job? That should bring it in line with the other ones again.
I intend to charge to 3.4 (54.4) then float at 3.4 (54.4) for 30 minutes. As you pointed out in a previous episode, this puts all cells at max capacity. I would also stop the charge if any cell hit 4.45v. Is there a problem with this? I would turn of all balancing. Do you think this would cause plating problems?
I don't think your settings will cause any problems. Are your cells not going out of sync after w while?
@@OffGridGarageAustralia still in planning stage. Thanks for your work.
Negative of bottom balance is also that the weakest cell will be the 1st to be 3,6V, therefore weakening it even more
I top end balance, higher the voltage the better, as the voltage increases quickly so does the charging current go down, that way you get the best out of your BMS balancing current
You wrote, "... higher the voltage the better". At some "high" voltage your are degrading the cells ...
@@MrSummitville I run my banks at 56.4 volts, and have done for over 2 years, it is a dual battery setup with Lifepo4 and Li- ion battery running together
"You are constantly transferring energy in and out of the weakest cell" - well, yes and no, yes you are, but during discharge and charge it's better to think of the balancer as *shunting current* PAST the weak cell, ie, the capacitor balance board is partially powering your load in the discharge low end state and partially charging your other cells in the charge high end state.
Is there a possibility that the lesser capacity cell will be overcharged by the charger during top balancing?
How do people feel about extending pv cables by soldered splice and marine shrink wrap? I ask, as I need to extend some existing cables, and there is little room in the conduit for anything else. Otherwise I have to spend £100 for new cable. Would the soldered join act as a resistor? The cable is 4mm2, 1000v . I intend to run around 1kw at @98v down these wires.
So long as the splice solder is thicker than the wire then no issue.
Hey, Chris, there is no problem with that. I have done it numerous times. If the solder connection is solid, no problem.
@@OffGridGarageAustralia thanks all, this is todays job.
Is this relevant for the exam ?
I am a total novice on electricity and all that stuff. So now I have ordered a set of 4 3.6V 280 ah batteries and bms. I do not have a bench charger. Only solar panels, a charge controller, and the bms. Will I need to find a way to balance the cells or can I simply put the parts together and go from there?
Pros? Cons?
If the cells are new and you don't need the state of charge you should top balance them. If you don't have a charger or power supply, you can put the battery in series and set the charge controller to a lower value, say 3.35V. Keep measuring the voltage of the cells until they all reach that 3.35V. Slowly increase the voltage of the solar charge controller by 0.2V and monitor the cell voltages. Keep going until you hit 3.6V with all cells. It depends on the BMS how it will handle the balancing. Got a DALY BMS? Forget all the above as it won't work with these BMSs.
Alternatively you can use an active balancer (~$25 for a 12V battery) and also increase the voltage slowly as above to give the balancer time to... balance.
I'll make a video about how to do all this once the new batteries are in place.
i always do a bottom balance, that's what works for me
And balance in the BMS turned off?
Thanks Andy
Any time!
In a top balanced system where each night you consume significant capacity (ie not a standby system) the weak cells will spend significant time at a lower SOC than the stronger cells. The stronger cells should therefore age faster and over years may help stabilise capacity deviation. If bottom balanced then it is the weaker cell that may age faster than the strong cells.
But it's not SOC which kills your cells, it's the voltage. Either low or high.
You are going to have two batteries, right? let's call them A and B. You could charge the batteries one after the other, so charge A, until balanced, and then charge battery B, until balanced, go back to battery A etc. when a battery is being charged, us the other one to drive the inverter, unless it is empty. This way you could hit max charge regularly and use a regular BMS. If you do not reach a full state of charge when charging battery A (resp. B), because you consume electricity from it, charge battery A (resp. B) again till full charge. There is a bit of mechanics & computer around that to organize the automatic connecting & disconnecting of batteries to the panels and the inverter. (cheers from Waterloo,BE by the way, we have good beers ;-) ).
Thanks for your comment. That would be ideal to treat the batteries separately but it's a lot of work and I would really preffer having them both connected all the time.
I used that 5 amp active balancer to bottom balance my 8s pack while still hooked up in series with BMS. I did n't dare redo the nuts on these cheap flimsy cell terminals again. I could feel these cheap terminals moving when I put it together the last times. the buss bars and holding the =/- cables while tightening the nuts prevents movement. runit down to low cell shut down, then wait for balance, then run it down till a low cell cutoff again and waited for balance again. each time the cell difference got less . till they all run out the same time.
What if my 8 cells I just got in are all reading exactly the same voltage 3.29v all 8 cells do I need to still do a top balance ?
Love the intro!!!
Real Scottish weather here!!!
The terminal you have shown in the video, cell terminal- screw type. What is the actual name, size of that thing? Please let me know.
You can provide link also
You can find the bus bars here: off-grid-garage.com/cables-and-connectors/
I need help, we are assembling batteries in bulk, cant afford to have individual bunch of 15 cells Top or bottom charge as we have to assemble around 50 units per day. We are facing issues due to these imbalances although cells we use are all A grade of the same batch.
My question is if we don't do the initial balancing and rely on the balancer to eventually balance it after 3-4 complete charge/discharge cycle. Is my approach correct, if not if there is a system where I can do initial balancing in bulk (let's say 150 cells in one go)
Bottom balanceing is ideal but there aren't any bms's that do anything but top balance so I don't know how useful it is and as far as active balancers they are not quite that smart they can only move energy over 1 cell I think what would work better is taking a high voltage cell and boosting to the whole pack voltage and dumping it there and taking power from the whole pack to charge an undervoltage cell and just do that for one cell at a time I think would be fine but I don't think they over cycle the battery because it is mostly doing it during charging so instead of like 10 amps it's getting like 9
Well, if you bottom balance, you would not need a BMS with balancing capabilities at all then. This would just destroy your initial bottom balance. Same for a balancer.
@@OffGridGarageAustralia but I mean cells are probably going to drift but if you are not planning on having any balancing what difference does it really make if you had them top balanced or bottom balanced they are going to drift just the same
Buy matched cells, turn off the balancer in the bms, bottom balance the cells one time, charge to 80%, and check back in a year or so.
Where do you purchase your guaranteed matched cells?
@@MrSummitville Shenzhen Luyuan Technology Co., Ltd on Alibaba They are brand new matched grade A EVE cells and come with a test report on each cell. Mine tested over 290Ah just as the report stated. They cost more, but are well worth it. Also they come with welded studs which is a plus.
with 1 active balancer you charge and discharge your weak cell with 5 A only, does it count as charge cycles?
Of course with 100Ah against 280Ah the difference is huge, but with more equal cells, the weakest cell wouldn't suffer that much.
If we charge with 50A , shouldn't we need 10 of those 5A balancer in order to shift all the charge as fast as possible, so all voltages remain equal at all times? And we don't need balancing at rest.
I think the balancer will only have to work once there is a voltage difference across the cells, so at the bottom or top of the curve when cells start drifting. It should be very easy to balance such high capacity cells with 5A at this stage. There is no need to have multiple balancers, I actually think they would counter each other and start fighting.
Top balance in parallel to 3,6V should do the trick.
Tops are good but bottoms are better!
Bottom with no balancing?
@@OffGridGarageAustralia Bottoms not bottom
Bottom balance without balancer assumes cell properties do not change but thats just wrong. Over multiple cycles capacity as well as internal resistance changes. The worst call will change more then others and when the cells get old this will get a fire hazard.
That's why bottom balancing the pack does not make sense in my eyes...
Aluminum bus bars are dangerous, at least in buildings. Aluminium develops a coating of oxidation that eventually causes arcing at the connections. The building where I had a condo had a huge fire because it had big aluminum bus bars instead of copper wiring. Such bus bars are not unlawful under the building code.
I have used aluminium busbars for 5 years in an RV without any corrosion problems. It might be more of a risk in damp buildings. To reduce risk of shorts I have recently started coating them in heatshrink and cutting out circles around the terminal points to enable good conduction. This would probably help minimise any risk of corrosion.
Yeah, but the corrosion is only dangerous at the points of connection, not on the outside.
@@OffGridGarageAustralia Yes, that is what I have heard too, that Aluminum is the metal that expands and contracts the most of all metals, and when you add corrosion into the mix, the arcing occurs at the connection points.
@@JR-kk6ce Theory is often in conflict with practice, sometimes more limiting sometimes less. In this case, unless you have some extreme temperature and humidity conditions, the practice wins. In the case of humidity you will probably see the external effects of corrosion on the terminal busbar junction long before it becomes an issue. Expansion due to temperature around a terminal over the range of temperatures that lithium cells can be used isn't going to be an issue.
@@nigelcharles511 I would be careful with aluminum busbars. It is not the humidity that you have to worry about, it is the the material of the post, washer, and bolt that you are going to use. Aluminum is very reactive with other metals and will galvanically corrode. The corrosion at the connection will create hot spots and possibly even a fire. I know this because I worked many years doing thermal imaging of aluminum busbars in buildings that were built in the 70s. Once you have a hot spot at a connection, the heat expands the aluminum and when the current is taken away, it cools and contracts. Eventually, the connection becomes loose and greatly increases the risk of fire.
Can this chemistry of battery be fast charged? Can they handle a lot power to charge a bank? Also can they handle a high power draw never to be drawn down below 45%?
A 'Yes' to all your questions.
@@OffGridGarageAustralia Thank you I've been looking for the chemistry of battery to conduct an experiment with and this seems like the one I should use. Thanks, now my research begins on this battery. But one more question; how many charge cycles does batteries generally have? Thanks again!
You bet the weakest cell gets stressed by the active balancer! That's why I said previously that it isn't the best solution. :) The best solution is more closely matched cells.
That intro!!!
🙂
Thanks!
no problem
I don't know about balancing by capacity it seems to me that you're relying on your ability to measure how much capacity you took out and how much capacity you put back in the shunts are pretty accurate but I don't know if they're quite accurate enough
All balancers take the cell voltages as measurement. As long as there is a difference, they keep either burning off energy or shuffling into other cells.
A coulomb count through a shunt may work but is not 100% accurate either as it would need several other parameters as well (current, temperature, SOC) to make an accurate decision.
Seems like it, we cannot really find the perfect solution.
Do you have a link to the balancer you showed in your video?
That's wasn't a balancer at all. At this point of making the video I didn't have one so took this buck booster as a place holder. But I have balancers on my website. They work great: off-grid-garage.com/battery-management-systems-bms/
@@OffGridGarageAustralia Thank you. I was a bit confused on the way it looked. (Big toroidal Transformer etc.) Great videos . Thanks again and congrats on hitting 10k !!
@@RemScanProducts Thank you!