One thing to remember with these active balancers is that they can only balance up to 5 amps but that amperage goes down as the cells get closer in voltage to each other and also as the leads get longer. So imagine that your voltage difference across the pack is 1V instead of 0.3V, you would likely see a much higher amperage but the charge moving is still moving to the next cell over so it may potentially be moving the charge at a rate where your meter is to slow to capture the full amperage but is averageing the time on and off for each cycle that the board shuffles some charge from one cell to the next in line. It tends to also be a good idea to only have the balancer kick in as your batteries are reaching the top 20-30% of the charge cycle so that it maintains your pack's top balance. I have had pretty decent results with the one I have but it only runs when the batteries are above 80% full and it does seem to help keep them in check but they aren't really needed once the pack is in good balance and the BMS is more than capable of handling it.
Yeah, balancers should only turn on at the last few percent to keep all the cells in check, so the ones with a lower capacity don't get reverse charged and die, or even worse start venting/blowing up
I think the last bit about the top balance really applies to LiFePO4 since it has that flat curve, probably less applicable for a tool battery or a laptop battery as they use NMC and LCO cells which seem to have at least a not totally flat curve.
I think the balancer moves little buckets of charge (3 capacitors full?) at a time, and can do so only to an adjacent cell. And a cell can only perform one move-up OR one move-down operation at any one time. Think 14 one-armed men passing buckets up/down. The balancing algorithm is probably to pass buckets down(/up) the line in a cascading fashion, as long as there is a cell on the left(/right) with below average voltage. In this curious case, cell8 is being the bottle neck, having to split time passing buckets up and down. Cells 7(/9) only has to pass buckets down(/up) the line, so is outpacing the supply from cell8 by a factor of 2. I reckon it will start to make sense as cells start reaching the average voltage.
Seems very plausible, nice, i would agree with that explanation. Maybe the 5A balance is a Collective value across all the cells, so all 5 single armed people have 1A of balance power each. (So it’s not a 5A balancer, its a 1A lol)
Close, most of the capacitive balancers like this work by having a capacitor (or 3) parallel to each series "cell" (or bank of parallel cells), what they then do is have 2 pairs of MOSFETs, one pair wires each capacitor (or 3) to its cell, then the other pair wires them all in parallel to each other. Then there is a low power oscillator that toggles one pair of MOSFETs off and the other on, then switches them. The practical upshot is all capacitors charge off the battery cells to match them, then they will all be disconnected from the battery and wired to each other in parallel, rapidly balancing themselves so they all are the same voltage, then when the MOSFETs are switched again, they reconnect to their cells pulling the low ones up by discharging into them, and the high ones get pulled down by the capacitors charging from them. ALL cells get balanced simultaneously, it's almost like you connected all the cells in parallel without the hassle of disassembling the pack and rewiring in parallel.
Im also an anti-always on active balancer. I just finished an online argument about active balancers (I argued they just throw a pack out of balance when always active balancing 24/7. Well he's convinced me to also try out a heltec active balancer. Off grid garage also didn't like them unless they could be activated only at the end of the charge. Look forward to your next video.
Off grid garage is anti-always active balancers for LFP (or LiFEPO4) chemistry (n/a for li-ion, only uses LFP for his setup). This is a Li-ion chemistry, charge - discharge curve is not flat anymore as LFP would be, but quite lineal. Off grid garage specifically mentiones flat charge - discharge LFP curve against having active balancer always active.
I am using the smart balancers from heltec an am very happy how they are performing. They are very well calibrated. But I always left the cables in original lengthened them by the same amount.
Andy at the Off Grid Garage channel has recently been testing an active balancer on his setup. His feeling is that you would only use it when the battery is pretty much fully charged to make sure that all the cells are balanced at which point you would take it off and just leave your BMS to manage the cells after that. He showed that if you leave the active balancer on while the system is running it will each cell will have a slightly different power profile which means at different points in the batteries charge it will produce more or less power than it's neighbores and balance that. This means that it will completely throw a well balanced battery completely our of wack. I think the conclusion you are going to come to is that an active balancer is useful to get your cells in balance for the first time but after that the BMS (with it's own balancer) should be enough to main a health balance, if it doesn't there is something wrong with your battery. My oponion is an active balancer is a very expensive device that you should only need once.
Thanks for the effort and professionalism. Every other active balancer test I've seen, except one, the rated amount of current transfer can only be reached using a laboratory input to produce an unusual amount of current. The only review I've seen where the balancer produces as advertised is on The Off-Grid Garage using the Neey 4th Generation active balancer. I appreciate your efforts.
Hello, Thanks for sharing your videos. I have been using that same balancer in my battery for two years and I see it as something essential and indispensable in any high-capacity battery to avoid imbalances between the different cells.
@@HBPowerwall Sure. I have a 40 Kwh battery, made up of 18650 batteries. The balancer is the same as the one in your video and the BMS is 400A from the ANT-BMS brand. If you go into my TH-cam account you can see my battery in one of my videos, the battery has two sides where there are batteries on both sides.
Nice system and good job on wiring. In reality few of us can't afford or have the know how to do a nice job as yours. With that says even with less than nice perfect set up it will still work and be usable. It does not have to be fancy complicated and costly set up to have solar power with battery system. Mine is completely built from recycle reduce reuse concept from panel to battery to inverter. At this point i am going with the active balance regardless of any other opinions good or bad. It makes sense and practical to transfer energy from one to another instead of wasting as heat with the passive balancing and it does that in both charge and discharge mode. To me that is more important than the bms to have the cells balance live as much as possible. I had to rebuild many packs because cells are not balanced more often than i like. I don't have the balancer arrives yet for me to make a conclusive determination how well it works or NOT works. I did some reading and understood how active capacitor balancer works and it is simple yet makes sense and effective (in therory). Regardless of how anyone goes there must be some kind of balancing the cells as imbalance short life and reduces capacity of the whole battery set up. The saying that the chain is only strong from its weakest link is true regarding solar battery storage. Batrium bms is nice and useful but man man it is costly and i am balking at buying one for years. With EV and battery become more prevalent the Chinese will no doubt comes out with a nice BMS for enthusiasts. I am waiting for that. Meantime active BMS and my wifi remote 360 degree camera do the job of keeping an eyes on the battery's health
100 % recycled cells, swaped some labour for them so basicly free, case/metal house was recycled, Active balancer was sponsored, Nickel Strip on the cells was sponsored - it's easy to do cheap if you have the time...
You might be having issues with induced voltage I would suggest running balancing cables in separate shielded loom. Induction will always counter capacitance
@@HBPowerwall you could try to twist them all, if you twist them, it counteracts induced currents, used lots in telecommunications with very high frequencies, much higher than these capacitor boards use, could save some rewiring and give you one more thing to document if you want. The right twist might just be all of them twisted together.
Curious: @8:25 ish, .... Can you find a correlation to voltage differences and cable lengths?..and/or "resistance" per the cables routed to the lower-voltage batteries?(I.E. resistance measured including that caused be crimped/soldered connectors and the wire-length)
I tried to do a resistance test but my meter was either acting funny or I was doing it wrong - either way i'll include some more data in next video or two
I think you are right about the cable. The balancer is trying to charge according to the volts it sees at its PCB, not the actual cell voltage - if it puts an amp to or from the battery, there is a voltage loss or gain. With 4mm cable, the cable drop alone will be around 10mV /A from what I can see. If the cable uses copper plated aluminium, even more. So the balancer is working on compromised measurements. What they should have is separate charge/discharge leads and separate voltage detection leads for equipment like this that is designed to work with mV accuracy, or shut current flow off, measure the cell and act on that. I suspect this latter arrangement will require a uP a precise AtoD, voltage reference and a little RAM to run this approach. I think these things have a bad name because of very poor design.
That would be my guess too. Driving 5A over a long wire with such a tiny voltage difference is optimistic. Maybe possible using 2 wires for each cell: 1 to sense voltage at the cell and 2 to carry current. When using a LiPo charger using the balance connector even for a 1S cell, it raises the voltage higher to account for voltage drop so that it can charge at 30A. The voltage at the charger may be 4.5V, but still only 4.20 at the cells. Without the balance sense the charger assumes the cell voltage is what it sees on the charge cable and limits current. Make the cables as short and as fat as reasonable, and it should improve.
BMS needs to be mounted on terminals of the cell, say e.g. M8 or M12, on PCB to have thick traces is a must, U need FETs with 1 or 2 mOhms, so lets say, 3.5 V is maximum for LiFePO4, 3.500 - 3.475 === 0.025 V , 0.025 V / 0.008 ohms === 3.125 AMPS, if rezistance is higher for PBC terminals, traces, FETs, there is o problem. So we can make another solution, make DC-DC non-isolated, and pump electrons this way 14.1 V to 10.5, 10.5 V to 7 and 7 to 3.5 V, we need to make 3 cheap DC-DC 3 amps for 14 - 10, 4 amps for 10 - 7 and 8 amps for 7 - 3.5 V, ADS1115 for measurements, 3 toroids, 6 FETs or 3 FETs + 3 DIODEs. U can also measure cells before mounting, and arrange cell with respect to capacity from 3.5 to 14 V potencial.
These balancers are quite dumb, meaning they don't even measure the voltage. I have a video on my channel that describes in detail how they work. Now you might be on to something with the different cable lengths, as that might impede the balancing of certain cells.
I would use it with a undervoltage - overvoltage relay and only allow it to run at the higher bulk voltage and turn it off at 1v under the bulk voltage, I have a few of these programmable relays coming myself to test this setup
my findings to date are, buy a proper BMS and ditch the extra expense and complexity of an active balancer - they are a solution to a problem that shouldn't exits and if it does, you should fix the root cause of the issue not fix it with more tech! That said the device works and does it well, but MOST of the time and I'm talking 99.9% the balance current is the same as even a cheap BMS but with the advantage it works all the time not just top of charge.
I built an 8s LFP battery for my trailer. My cells go out of balance a little with every cycle. I don't have any particularly weak cells, so I don't know why. Anyway, my Daly BMS has a pretty useless balancing function so I bought one of these and a low Voltage relay to turn it off when the cells are below 3.4V. The Heltec keeps them balanced very well above 3.4V. I'm no expert, but I wonder if your powerwall is just way to massive for one of these balancers. I've also read that the cells have to be way out of balance to get the full current from the Heltec. The closer the cells are, the lower the current from the balancer.
Think i'll leave that loom in place and solder the new balance leads to the legs of the balance ports to remove a plug & a solder joint - that combined with thicker cable should be interesting too see.. If only for educational purposes
Oh wow, I definitely have not seen mine unbalance the pack like this. I have a 8s, and a 4s they balance super well. Now I do have all the cables the same length. Hmm, those two with the lowest cell7, and 9 do they have the longest length of wire (not counting cell 8)?
Yes they were the furthest away from balancer.. It took 3 weeks for the balancer to balance the battery. renders it useless in my opinion when a good bms will do the job much faster with out the false sence of security
Mr hp powerwall. I'm a beginner and nowhere near your level of building powerwalls. I always solder all connections and I over size my wire. try this it may help. thank you for all your videos.
May this be related to differences in internal resistance of each cell? It also looks like the cells with the longest cables to the heltec, have the lowest voltages. (not counting the one you charged to full.)
The balance current up to 5A is ONLY when you have difference of max voltage. Ie min and max. I think by design you get 200-600mA in average of balance current. Yes the wire length will change the outcome to. And depending on balancer they move it between cells. and not from call a to z but from a to b to c to d and so forth. You should not need to do more to those cable. It may have a resistance but over time it should get back to balance. But do not expect 5a. Perhaps 0.5a in average :) If you do 1 cell at 3v or below and one at 4.2v and let them sit close to each other then you might be able to see higher current. Then you can also test the wire size and length to see how it affects it
the small test happily did 5 amps, was actually thinking about going the other way to test - one cell low 13 high.. that might also show it. I've started a two day test with screen capture to see how much it balances.. also turned off SCC to not skew results. thanks for the info buddy
What sombody needs to design is a balancer that uses a separate cell from the system and is charged with a dc/dc converter that hops around charging each cell at 10 or 20 amps at a time for large packs until it is balanced. I think its doable.
I only sell the 80 p ones, but in the description there should be a link to my 'official thread' that'll have all the plans you can copy & modify to your needs
not really, it works and does the job well, just not at all needed on this battery with the Batrium BMS, It's till fitted infact I just released a new video with it still fitted.
What's the difference in cable length? The voltage drop at 48V along 3mm cable over 1m at 5A is about 0.0615V. Looking at your test set up I would guess the cable to be between 1.5 and 3 metres overall length, so 1.5m difference - say 0.09V between highest to lowest reading. If the issue were just the cable length would expect to see a range of differences from nominal, not one flier (unless all the other cables are about the same, with one cable about 10m longer than the rest. What you are seeing points to there being something amiss with the controller as well as a small contribution from cable length.
Interesting video, i recon you'll find that the cable length's being the same matters due to the slight difference in resistance that is amplified as the battery voltages change.. be interesting to see where this goes lol
I am guessing it can only transfer power from the cell adjacent to the cell out of balance so as it goes through the list of cells and finds one that is different than the average voltage but is also different by the programmed amount it will take power from the adjacent cell until finally they are ll within the programmed difference. if you had discharged the single cell for your test it may have balance faster.
Heltec's complete lack of support, after they sent you the balancers, to review, doesn't say anything about their professionalism, or any ability to back their products. To leave you to absolutely guessing, at what you need to do next (to get the maximum balancing efficiency), has turned me completely off, purchasing ANY of their products! Love your work, mate.
They did reply quickly but feel it was more a translation issue, but yes replies to my questions would have been nice - THAT said, I still have it running and the battery is still very well balanced with a 200w load on it 24/7 I should turn it off to see if the battery unbalances
Wire length should not matter all that much for capacitor balancers, they basically connect caps to each cell, then switch so the caps are all connected parallel together, then switch back to being connected to the cells... this makes them level out as if you have all the cells connected in parallel... just slower because of impedance. Wire length probably matters more for the inductive balancer types but they have the own issues, especially when used in large strings since they only transfer bank to bank so if you have a high bank, it will transfer to the low banks next to it, and then to the ones next to that and so on making a ramp as they stop balancing once they reach a certain voltage difference. Now what will matter is wire resistance. The better connection, shorter the wire, and thicker the wire, the lower resistance and the more current you can get, however, the current is proportional to the voltage difference between each cell... and all cells balance at the same time which is a nice feature for larger packs with more series cells or banks... the inductive type is not as good for this but for smaller packs they are nice and tiny. An inductive balancer might sometimes move more current per unit difference in voltage, but I think that would depend on if the chips are designed to compensate like that. The transformer balancers can move a shocking amount of current but are more pricey. These balancers are really meant to be left alone, set and forget, they take very little power (basically only enough to run a small oscillator and switch some MOSFETs), and they do not have any brains that can get confused, they connect and disconnect capacitors so that they move charges from all higher cells to all lower cells at the same time. But I have seen guys using batteries with flat curves... these can screw them up because the voltage difference are not from different state of charge, just different connector resistances, so they only run the balancer when the batteries are charged up so they have a difference in voltage from full to almost full. For the cells I scavenge (LCO, NMC, usually 18650 case), not that big of a deal since the cells have a more linear discharge curve so the balancer wont possibly screw with a top balance, but for LiFePO4, you might want to check if that is the issue, I suggest a little voltage trigger, one guy used a relay board but I hate relays because they're mechanical and take lots of current compared to a MOSFET.
Hmm. I think the current balancer that they have sent you can only do a-b, b-c, c-d eta balanceing, meaning that the balancer is dishing out current from the cell with high voltage, to the adjacent ones, which are then dishing out current to there adjacent ones. The balancer isn't able to provide enough current from the cell with the high voltage to the adjacent cells, so the 2 adjacent cells are discharging. Because these cells are discharging so much, it is throttling the total balancing current. I know that you hate active balancers, but it would be interesting to see you test there "Smart Balancer 10-24S 4A" as even though this is only 4A, it advertises "Point to point energy transfer algorithms" which sounds to me is able to balance to any cell in the pack.
When I think how to measure voltage accurately, the single attribute that comes to my mind is "end to end" resistance. Larger cable cross section when in "audio" frequencies, i.e below 100kHz, including DC, will result in lower resistance, which means lower temperature rise in the cable. If all "end to end" resistances are equal, it does not matter what the resistances are, because they cancel each other out. Based purely on the BMS components I think I saw, I suspect it is same kind of a balancer aka "Flying Capacitors" BMS that for example Julian Ilett is using in this video th-cam.com/video/bMmlgS46occ/w-d-xo.html IIRC Julian has done a a few of videos that contains some mentions of this type of BMS.
Like you said in the description: an ideal battery wouldn't have this issue. To add insult to injury, if you have an issue, you'd rather want to be informed about it instead of sweeping it under the rug by dumping all your battery power into one bad cell!
Yes, it's a tough call because they fill a void where they can keep and older battery alive/useable for much longer but just because you can, doesn't mean you should.. Whats worse is there are a few battery sellers floging them with their batteries..
Unplug the Batriun to let the Active Equalizer work. If the two systems are connected, it will not work correctly and there will be a gap between the series. Greetings.
Batrium does nothing unless its within a range that I've set - it won't even turn on till 4.13v and start balancing till 4.15v - It just sits there drawing =/- 1w of power waiting till the time it needs to 'manage' the battery.
It's OHMs Law. It's not wire length or anything like that. Working with voltage differences between cells in the millivolt range you will be challenged to get the ballancers peak current rating outside of a direct short. I haven't seen anybody get the rated current out of 'em. And yeah, it's a balancer. Not a BMS. My solution is to top ballance each cell every once in a while and be done with it. :)
Yep, with all the rain I just left my house battery off for 10 days, the longest period it's ever been off in about 6 years and its within 0.2v good enough for me..
I bet the charge level on each is directly proportional to the wire length The balancer will see voltage and a longer length of cable will drop more voltage than a shorter one
it is not about cell voltage , the active balancer is not designed to this mush big battery .it is about average voltage after charging the capacitors , you should check how actually it is working so you can understand what i mean
I know how they work, they move energy around with out logging that energy and end up HIDING critical issues that would be caught other wise with a good BMS..
8:48 That's definitely not 3mm diameter conductor and I doubt if that's 3mm2 conductor. I think there is too much chinesium oxide in those components and that manufacturer is out to break your storage system and/or your confidence. If they were a problem in previous version, they'll a problem for every product they make. As Marty would say: th-cam.com/video/v8CK6CLGTjM/w-d-xo.html
Whats the total resistance of the wires you have extended? I don't think balances can boost voltage, I think they are just higher voltage cell carriers - but I may be wrong. You'll find that the voltage drop in the cable will cause less current flow from cell to cell. It's a direct result of low voltage vs high resistance. I have a couple of active balances in some of my early badly matched capacity 18650 batteries. Because these packs are in parallel with many other packs, the constant load is very low, no more than 10 amps at worst and usually only for an hour or two, they mostly sit discharging just a couple of amps. So I find the balancer does a good job keeping them all in line. But if these packs were my only packs and I was cycling them pretty hard, the balancer wouldn't have a hope in hell keeping the pack happy. You've really got to pick and choose when they are suitable for use.
I love watching creators interact with their audience. It's just awesome to know there is actually someone behind the channel. I know its obvious that there is, but seeing it is just awesome.
@@HBPowerwall There is a point on the board that has two terminals soldered. Remove the solder and a switch can be wired in to turn the balance on or off. Then have the balance turn on only when voltage is above a set value. I am planning a rather large build 25 kW PV and 100 kWh battery but would like to be part of an Australian forum to get help if needed and maybe contribute..
Good video. I just don't think the current generation of active balancers are able to handle the massive parallel banks of batteries we're working with. Relatively flat discharge curves, and enormous Ah storage is never going to work without an intimate knowledge of SoC.
I'm new to this, absolutely no experience. Buuut : If the balancer stops the current flow, before measuring the voltage, then cable length does not matter (and because of that, it should do it like that, but maybe it does not). If it does not, and measures voltage while balancing, then you're possible better of having them all the same length (when a current flows, you get a voltage drop because of the resistance in the wire (veeery little, but not zero)). ...... In general, wires/cables are better of just laying messy around with lots of space in between. If you strap'em close together you have started (somehow) to create a transformer, and if you then run AC (alternating) current through one of the "transformer" wires, then you get a induced current in the very very nearby wires. ...... But that's or AC. ....... Here it's DC, and I'm not sure now, but for DC, I think it doesn't matter (no induction). ......... So my guess is that the firmware is measuring the voltage while balancing, and then the cable lengths should be exactly the same, idealy,....... but if it matters ?,....... probably not. ......... The guy who sugested that the balancing was only done with adjacent cells, maybe he's correct, as the bars on your display could match such a process. But I fail to see why a balancer has to be engineered like that, except, it's probably cheaper to produce like that. .... If this guy is correct however : th-cam.com/video/bFOJdvxhJbo/w-d-xo.html ......... then there possibly are kind of AC currents flowing back/forth in the balancing wires. .......... And if he's correct, that neighbour only balancing is not the case (maybe is in the BMS ?).
At .5 to 1.5 amps I don't think a active balancer can work on such high of capacity system. If you put 20 or 30 of them on your system it might work. Otherwise I only see they would work on a system that is ten fold of the balancing current, or about 10 ah per board.
Weeks later and i'm still not convinced they are a good idea.. I mean they work like it says on the box but no much better than a cheap BMS with the balancing current 99.9% of the time being in the mv range not the 5 amp max it 'could' do.. and to be honest that situation would me you have MAJOR issues with your battery that desperately need to be addressed not hidden
wires are too long for non capacitive active balancing. Try their BT version, it has impulsive capacity active balancing with 4A, wire length does not matter for it. It stores energy in capacitors and transfers it to lower cell in pulses.
for that length of cable you will need to have a thicker cable, to let 3.6v travel in 2 meters of cable without loss you will need 10mm2, wire resistance....sorry pete
Hey , in total full loded u got 200ah maby more maby les , that 5 amp active balancer lets say its balancing 2 cells in 1h , u got 2000 cells there maby more maby less , so that means they balance in 1000hrs if they completly unbalanced , the point is if they are not new cells the charge and discharge are not so eficient in egual mode lets say 2 new cells charge and discharge at the same time , but if not new the charge and discharge are diferent betwen them so u balanced them in 1000hrs and u disbalanced them in not more then 1 hr and ther it goes charging frome solar system first to the good cells then harder to bad ones , in paralel they balancing at highest speed betwen them and in series used to a consumer the bad paralels discharges faster and consumes ah , when the good paralels berly consumes , so there we go begging fore that balancer to balance 2000 cells in shorter time butt it cand still same process 1000hr without u charging or discharging it , so in my opinion add 10 more or more active balancers and putt em in paralel , its not balancing the way it shud normaly to smaler pack so that that u got is a huge mega pack so to kepp up with charging , consumers and balancing add good bms and good balancers , more and use them same in paralel
I know how stressed out you got making this video I know your not keen on these active balance boards like I said I was skeptical but I have been presently surprised and someday I will sort out the spiders web lol
I would much rather the balancer to be in the BMS, however, my small pack (3s16p) has its BMS balancer not working, they come with those shitty passive bleeder balancers that just drain the high cells as they go above 4.3 volts (yikes that is way too high for any lithium ion variant), But they won't activate until you get them to go to full charge, so for people never fully charging the pack they never get balanced. An active balancer not only balances way faster than a passive bleeder balancer, but makes less heat and does so at any voltage... this is a downside for LFP and other lithium ion types with flat curves, but for LCO and NMC lithium ion, it's nice as they should have the same voltage per cell all the way from top to bottom, they should all fall at the same rate, and they all have different voltages across the discharge and charge curves. So if your BMS is defective like mine that screwed my balance in 1 cycle (perfectly balanced parallel banks, next charge 2 were at 0 volts and 1 was at 4.3 volts), they're perfect, and if you use something other than LFP (LiFePO4), then they can work fine 24/7... though mine is an inductive type, not the capacitor type (shown here) or the transformer type (the really high current one Heltec sells), mine works fine for smaller strings and has all solid state parts (ceramic caps, not electrolytic caps), downside is that mine can only transfer energy from one cell to the adjacent cell, not across the pack, so a balance has to propagate across the cells... not a problem for 2s or even 3s, possibly 4s is fine... but you reach a lot more series and have one cell on the end with higher or lower capacity or just a bad balance... well each transfer stops at a set voltage difference, so the difference is fine at one cell across, but gets worse and worse as you go along and they are each a set voltage different. The capacitive type (shown here) transfers energy to all cells at once from all cells at once, so all low cells get charged from all high cells at once. Still, I'd rather have the active balancer be built into the BMS so that it can monitor itself and so that it's only one part to fail with one balance lead and one balance wire per cell, not 2, one for the BMS and one for the balancer. Plus, a BMS designed for LFP could also be programmed to only activate the balancer when the cells are at the knee in the graph rather than the flat part which is where the balancer might unbalance the pack instead of balance it, then a BMS made for LCO or NMC like my cells could have the balancer work all the time there is a significant difference, at any state of charge, because they do not have a flat discharge curve.
You are relying on a for profit company to over deliver. If it doesnt say it, it doesnt do it. Maybe try to keep them honest in a different way. The more complicated, the more outs they have.
One thing to remember with these active balancers is that they can only balance up to 5 amps but that amperage goes down as the cells get closer in voltage to each other and also as the leads get longer. So imagine that your voltage difference across the pack is 1V instead of 0.3V, you would likely see a much higher amperage but the charge moving is still moving to the next cell over so it may potentially be moving the charge at a rate where your meter is to slow to capture the full amperage but is averageing the time on and off for each cycle that the board shuffles some charge from one cell to the next in line.
It tends to also be a good idea to only have the balancer kick in as your batteries are reaching the top 20-30% of the charge cycle so that it maintains your pack's top balance. I have had pretty decent results with the one I have but it only runs when the batteries are above 80% full and it does seem to help keep them in check but they aren't really needed once the pack is in good balance and the BMS is more than capable of handling it.
Yeah, balancers should only turn on at the last few percent to keep all the cells in check, so the ones with a lower capacity don't get reverse charged and die, or even worse start venting/blowing up
I think the last bit about the top balance really applies to LiFePO4 since it has that flat curve, probably less applicable for a tool battery or a laptop battery as they use NMC and LCO cells which seem to have at least a not totally flat curve.
I think the balancer moves little buckets of charge (3 capacitors full?) at a time, and can do so only to an adjacent cell. And a cell can only perform one move-up OR one move-down operation at any one time.
Think 14 one-armed men passing buckets up/down.
The balancing algorithm is probably to pass buckets down(/up) the line in a cascading fashion, as long as there is a cell on the left(/right) with below average voltage.
In this curious case, cell8 is being the bottle neck, having to split time passing buckets up and down.
Cells 7(/9) only has to pass buckets down(/up) the line, so is outpacing the supply from cell8 by a factor of 2.
I reckon it will start to make sense as cells start reaching the average voltage.
you explain it well , time will tell - thanks heaps for the input..
Seems very plausible, nice, i would agree with that explanation. Maybe the 5A balance is a Collective value across all the cells, so all 5 single armed people have 1A of balance power each. (So it’s not a 5A balancer, its a 1A lol)
I like this explanation it should be in the documentation 😉
Close, most of the capacitive balancers like this work by having a capacitor (or 3) parallel to each series "cell" (or bank of parallel cells), what they then do is have 2 pairs of MOSFETs, one pair wires each capacitor (or 3) to its cell, then the other pair wires them all in parallel to each other. Then there is a low power oscillator that toggles one pair of MOSFETs off and the other on, then switches them. The practical upshot is all capacitors charge off the battery cells to match them, then they will all be disconnected from the battery and wired to each other in parallel, rapidly balancing themselves so they all are the same voltage, then when the MOSFETs are switched again, they reconnect to their cells pulling the low ones up by discharging into them, and the high ones get pulled down by the capacitors charging from them. ALL cells get balanced simultaneously, it's almost like you connected all the cells in parallel without the hassle of disassembling the pack and rewiring in parallel.
Im also an anti-always on active balancer. I just finished an online argument about active balancers (I argued they just throw a pack out of balance when always active balancing 24/7. Well he's convinced me to also try out a heltec active balancer. Off grid garage also didn't like them unless they could be activated only at the end of the charge. Look forward to your next video.
Off grid garage is anti-always active balancers for LFP (or LiFEPO4) chemistry (n/a for li-ion, only uses LFP for his setup). This is a Li-ion chemistry, charge - discharge curve is not flat anymore as LFP would be, but quite lineal. Off grid garage specifically mentiones flat charge - discharge LFP curve against having active balancer always active.
they did get him an upgraded active balancer wich works on the end of charge of LifePO4 cells , arround 3.45 v
@@volodumurkalunyak4651
I am using the smart balancers from heltec an am very happy how they are performing. They are very well calibrated. But I always left the cables in original lengthened them by the same amount.
Thanks for sharing. Janek - I really hope I can prove you right on that one. Time will tell..
Andy at the Off Grid Garage channel has recently been testing an active balancer on his setup. His feeling is that you would only use it when the battery is pretty much fully charged to make sure that all the cells are balanced at which point you would take it off and just leave your BMS to manage the cells after that. He showed that if you leave the active balancer on while the system is running it will each cell will have a slightly different power profile which means at different points in the batteries charge it will produce more or less power than it's neighbores and balance that. This means that it will completely throw a well balanced battery completely our of wack.
I think the conclusion you are going to come to is that an active balancer is useful to get your cells in balance for the first time but after that the BMS (with it's own balancer) should be enough to main a health balance, if it doesn't there is something wrong with your battery. My oponion is an active balancer is a very expensive device that you should only need once.
Thanks for the effort and professionalism. Every other active balancer test I've seen, except one, the rated amount of current transfer can only be reached using a laboratory input to produce an unusual amount of current. The only review I've seen where the balancer produces as advertised is on The Off-Grid Garage using the Neey 4th Generation active balancer. I appreciate your efforts.
Thank you!
Hello, Thanks for sharing your videos. I have been using that same balancer in my battery for two years and I see it as something essential and indispensable in any high-capacity battery to avoid imbalances between the different cells.
Thanks for sharing, may I ask the quality of your battery & type of BMS?
@@HBPowerwall Sure. I have a 40 Kwh battery, made up of 18650 batteries. The balancer is the same as the one in your video and the BMS is 400A from the ANT-BMS brand. If you go into my TH-cam account you can see my battery in one of my videos, the battery has two sides where there are batteries on both sides.
Nice system and good job on wiring. In reality few of us can't afford or have the know how to do a nice job as yours. With that says even with less than nice perfect set up it will still work and be usable. It does not have to be fancy complicated and costly set up to have solar power with battery system.
Mine is completely built from recycle reduce reuse concept from panel to battery to inverter. At this point i am going with the active balance regardless of any other opinions good or bad. It makes sense and practical to transfer energy from one to another instead of wasting as heat with the passive balancing and it does that in both charge and discharge mode. To me that is more important than the bms to have the cells balance live as much as possible. I had to rebuild many packs because cells are not balanced more often than i like.
I don't have the balancer arrives yet for me to make a conclusive determination how well it works or NOT works. I did some reading and understood how active capacitor balancer works and it is simple yet makes sense and effective (in therory).
Regardless of how anyone goes there must be some kind of balancing the cells as imbalance short life and reduces capacity of the whole battery set up. The saying that the chain is only strong from its weakest link is true regarding solar battery storage.
Batrium bms is nice and useful but man man it is costly and i am balking at buying one for years. With EV and battery become more prevalent the Chinese will no doubt comes out with a nice BMS for enthusiasts. I am waiting for that. Meantime active BMS and my wifi remote 360 degree camera do the job of keeping an eyes on the battery's health
100 % recycled cells, swaped some labour for them so basicly free, case/metal house was recycled, Active balancer was sponsored, Nickel Strip on the cells was sponsored - it's easy to do cheap if you have the time...
You might be having issues with induced voltage I would suggest running balancing cables in separate shielded loom. Induction will always counter capacitance
I didn't consider that but gah dam that would be one hell of a thiccccc loom!
@@HBPowerwall you could try to twist them all, if you twist them, it counteracts induced currents, used lots in telecommunications with very high frequencies, much higher than these capacitor boards use, could save some rewiring and give you one more thing to document if you want. The right twist might just be all of them twisted together.
It is sad to not see that many battery videos from you. I wished you continued
Doing a battery review at the moment - the $$ v Time ratio isn't appealing
Great content as usually 👌. How many panels you need for that kind a wall .what panels best for today.cheers
Curious: @8:25 ish, .... Can you find a correlation to voltage differences and cable lengths?..and/or "resistance" per the cables routed to the lower-voltage batteries?(I.E. resistance measured including that caused be crimped/soldered connectors and the wire-length)
I tried to do a resistance test but my meter was either acting funny or I was doing it wrong - either way i'll include some more data in next video or two
Thanks for the video Peter. I think you set it up fine. I'm not a fan of these balancers either.
I feel the same but desperately trying to do an ethical video due to being given the devices from Heltec - so far for them its not going to plan..
I think you are right about the cable. The balancer is trying to charge according to the volts it sees at its PCB, not the actual cell voltage - if it puts an amp to or from the battery, there is a voltage loss or gain. With 4mm cable, the cable drop alone will be around 10mV /A from what I can see. If the cable uses copper plated aluminium, even more.
So the balancer is working on compromised measurements.
What they should have is separate charge/discharge leads and separate voltage detection leads for equipment like this that is designed to work with mV accuracy, or shut current flow off, measure the cell and act on that. I suspect this latter arrangement will require a uP a precise AtoD, voltage reference and a little RAM to run this approach.
I think these things have a bad name because of very poor design.
That would be my guess too. Driving 5A over a long wire with such a tiny voltage difference is optimistic. Maybe possible using 2 wires for each cell: 1 to sense voltage at the cell and 2 to carry current. When using a LiPo charger using the balance connector even for a 1S cell, it raises the voltage higher to account for voltage drop so that it can charge at 30A. The voltage at the charger may be 4.5V, but still only 4.20 at the cells. Without the balance sense the charger assumes the cell voltage is what it sees on the charge cable and limits current.
Make the cables as short and as fat as reasonable, and it should improve.
BMS needs to be mounted on terminals of the cell, say e.g. M8 or M12, on PCB to have thick traces is a must, U need FETs with 1 or 2 mOhms, so lets say, 3.5 V is maximum for LiFePO4, 3.500 - 3.475 === 0.025 V , 0.025 V / 0.008 ohms === 3.125 AMPS, if rezistance is higher for PBC terminals, traces, FETs, there is o problem. So we can make another solution, make DC-DC non-isolated, and pump electrons this way 14.1 V to 10.5, 10.5 V to 7 and 7 to 3.5 V, we need to make 3 cheap DC-DC 3 amps for 14 - 10, 4 amps for 10 - 7 and 8 amps for 7 - 3.5 V, ADS1115 for measurements, 3 toroids, 6 FETs or 3 FETs + 3 DIODEs. U can also measure cells before mounting, and arrange cell with respect to capacity from 3.5 to 14 V potencial.
@@PavolFilek Some good ideas there Pavol Filek.
These balancers are quite dumb, meaning they don't even measure the voltage. I have a video on my channel that describes in detail how they work. Now you might be on to something with the different cable lengths, as that might impede the balancing of certain cells.
@@BradCagle Nice channel Brad. I subbed. Watched the nickel plating vid - just what the doctor ordered, so thank you.
I would use it with a undervoltage - overvoltage relay and only allow it to run at the higher bulk voltage and turn it off at 1v under the bulk voltage, I have a few of these programmable relays coming myself to test this setup
effectively that's what I've done with the batrium.. Just using what I already had.
Good video... I wanted to know your opinion regarding whether that 5A capacitor balancer is more efficient than the JKbms...thanks
my findings to date are, buy a proper BMS and ditch the extra expense and complexity of an active balancer - they are a solution to a problem that shouldn't exits and if it does, you should fix the root cause of the issue not fix it with more tech! That said the device works and does it well, but MOST of the time and I'm talking 99.9% the balance current is the same as even a cheap BMS but with the advantage it works all the time not just top of charge.
@@HBPowerwall What BMS do you recommend, that is of good quality and technology?...thanks for the help
Hello .. if do not mind , for what do you use these groups of batteries ?
How coul I see the balance action on my P.C ??
Thanks
I built an 8s LFP battery for my trailer. My cells go out of balance a little with every cycle. I don't have any particularly weak cells, so I don't know why. Anyway, my Daly BMS has a pretty useless balancing function so I bought one of these and a low Voltage relay to turn it off when the cells are below 3.4V. The Heltec keeps them balanced very well above 3.4V. I'm no expert, but I wonder if your powerwall is just way to massive for one of these balancers. I've also read that the cells have to be way out of balance to get the full current from the Heltec. The closer the cells are, the lower the current from the balancer.
I have one on my Nissan leaf batteries it work very good I cramp some ring terminal in the end of cable
Think i'll leave that loom in place and solder the new balance leads to the legs of the balance ports to remove a plug & a solder joint - that combined with thicker cable should be interesting too see.. If only for educational purposes
Man you pit a lot into yer videos respect
Hello
I have a question about 18650 lithium batteries, how many milli ohms is required for my old one?
Oh wow, I definitely have not seen mine unbalance the pack like this. I have a 8s, and a 4s they balance super well. Now I do have all the cables the same length. Hmm, those two with the lowest cell7, and 9 do they have the longest length of wire (not counting cell 8)?
Yes they were the furthest away from balancer.. It took 3 weeks for the balancer to balance the battery. renders it useless in my opinion when a good bms will do the job much faster with out the false sence of security
Active means it's taking power and moving it to the lowest battery? Or is it spreading it out among all the batteries
Someone described it as like handing buckets to one and other till they are all the same
Mr hp powerwall. I'm a beginner and nowhere near your level of building powerwalls. I always solder all connections and I over size my wire. try this it may help. thank you for all your videos.
Thanks for sharing Chris - I plan on doing it all over with over sized wire & soldered ring termals see what comes of it.
May this be related to differences in internal resistance of each cell?
It also looks like the cells with the longest cables to the heltec, have the lowest voltages. (not counting the one you charged to full.)
IR is as close as it comes for a DIY Build, but the length of the cables and quality of the joins would all contribute to the performance.
The balance current up to 5A is ONLY when you have difference of max voltage. Ie min and max. I think by design you get 200-600mA in average of balance current. Yes the wire length will change the outcome to. And depending on balancer they move it between cells. and not from call a to z but from a to b to c to d and so forth. You should not need to do more to those cable. It may have a resistance but over time it should get back to balance. But do not expect 5a. Perhaps 0.5a in average :) If you do 1 cell at 3v or below and one at 4.2v and let them sit close to each other then you might be able to see higher current. Then you can also test the wire size and length to see how it affects it
the small test happily did 5 amps, was actually thinking about going the other way to test - one cell low 13 high.. that might also show it. I've started a two day test with screen capture to see how much it balances.. also turned off SCC to not skew results. thanks for the info buddy
What sombody needs to design is a balancer that uses a separate cell from the system and is charged with a dc/dc converter that hops around charging each cell at 10 or 20 amps at a time for large packs until it is balanced. I think its doable.
Where do I buy the 60p cell wall mounts? Been looking everywhere!
I only sell the 80 p ones, but in the description there should be a link to my 'official thread' that'll have all the plans you can copy & modify to your needs
Do.you know how it is price wise?
Honestly didn't look - probably should
Was there ever a second follow up video for the active balancers?
not really, it works and does the job well, just not at all needed on this battery with the Batrium BMS, It's till fitted infact I just released a new video with it still fitted.
@@HBPowerwall I have been waiting for the follow up video explaining what was causing the cell imbalance and what was causing it.
from memory it was me that caused the imbalance I would need to rewatch the video and thats almost never happened - hate the sound of my own voice
What's the difference in cable length?
The voltage drop at 48V along 3mm cable over 1m at 5A is about 0.0615V.
Looking at your test set up I would guess the cable to be between 1.5 and 3 metres overall length, so 1.5m difference - say 0.09V between highest to lowest reading.
If the issue were just the cable length would expect to see a range of differences from nominal, not one flier (unless all the other cables are about the same, with one cable about 10m longer than the rest.
What you are seeing points to there being something amiss with the controller as well as a small contribution from cable length.
Interesting video, i recon you'll find that the cable length's being the same matters due to the slight difference in resistance that is amplified as the battery voltages change.. be interesting to see where this goes lol
Thanks for the info!
I am guessing it can only transfer power from the cell adjacent to the cell out of balance so as it goes through the list of cells and finds one that is different than the average voltage but is also different by the programmed amount it will take power from the adjacent cell until finally they are ll within the programmed difference. if you had discharged the single cell for your test it may have balance faster.
Heltec's complete lack of support, after they sent you the balancers, to review, doesn't say anything about their professionalism, or any ability to back their products. To leave you to absolutely guessing, at what you need to do next (to get the maximum balancing efficiency), has turned me completely off, purchasing ANY of their products!
Love your work, mate.
They did reply quickly but feel it was more a translation issue, but yes replies to my questions would have been nice - THAT said, I still have it running and the battery is still very well balanced with a 200w load on it 24/7 I should turn it off to see if the battery unbalances
Wire length should not matter all that much for capacitor balancers, they basically connect caps to each cell, then switch so the caps are all connected parallel together, then switch back to being connected to the cells... this makes them level out as if you have all the cells connected in parallel... just slower because of impedance. Wire length probably matters more for the inductive balancer types but they have the own issues, especially when used in large strings since they only transfer bank to bank so if you have a high bank, it will transfer to the low banks next to it, and then to the ones next to that and so on making a ramp as they stop balancing once they reach a certain voltage difference.
Now what will matter is wire resistance. The better connection, shorter the wire, and thicker the wire, the lower resistance and the more current you can get, however, the current is proportional to the voltage difference between each cell... and all cells balance at the same time which is a nice feature for larger packs with more series cells or banks... the inductive type is not as good for this but for smaller packs they are nice and tiny. An inductive balancer might sometimes move more current per unit difference in voltage, but I think that would depend on if the chips are designed to compensate like that. The transformer balancers can move a shocking amount of current but are more pricey. These balancers are really meant to be left alone, set and forget, they take very little power (basically only enough to run a small oscillator and switch some MOSFETs), and they do not have any brains that can get confused, they connect and disconnect capacitors so that they move charges from all higher cells to all lower cells at the same time. But I have seen guys using batteries with flat curves... these can screw them up because the voltage difference are not from different state of charge, just different connector resistances, so they only run the balancer when the batteries are charged up so they have a difference in voltage from full to almost full. For the cells I scavenge (LCO, NMC, usually 18650 case), not that big of a deal since the cells have a more linear discharge curve so the balancer wont possibly screw with a top balance, but for LiFePO4, you might want to check if that is the issue, I suggest a little voltage trigger, one guy used a relay board but I hate relays because they're mechanical and take lots of current compared to a MOSFET.
Hmm. I think the current balancer that they have sent you can only do a-b, b-c, c-d eta balanceing, meaning that the balancer is dishing out current from the cell with high voltage, to the adjacent ones, which are then dishing out current to there adjacent ones. The balancer isn't able to provide enough current from the cell with the high voltage to the adjacent cells, so the 2 adjacent cells are discharging. Because these cells are discharging so much, it is throttling the total balancing current. I know that you hate active balancers, but it would be interesting to see you test there "Smart Balancer 10-24S 4A" as even though this is only 4A, it advertises "Point to point energy transfer algorithms" which sounds to me is able to balance to any cell in the pack.
When I think how to measure voltage accurately, the single attribute that comes to my mind is "end to end" resistance. Larger cable cross section when in "audio" frequencies, i.e below 100kHz, including DC, will result in lower resistance, which means lower temperature rise in the cable. If all "end to end" resistances are equal, it does not matter what the resistances are, because they cancel each other out. Based purely on the BMS components I think I saw, I suspect it is same kind of a balancer aka "Flying Capacitors" BMS that for example
Julian Ilett is using in this video th-cam.com/video/bMmlgS46occ/w-d-xo.html IIRC Julian has done a a few of videos that contains some mentions of this type of BMS.
Like you said in the description: an ideal battery wouldn't have this issue. To add insult to injury, if you have an issue, you'd rather want to be informed about it instead of sweeping it under the rug by dumping all your battery power into one bad cell!
Yes, it's a tough call because they fill a void where they can keep and older battery alive/useable for much longer but just because you can, doesn't mean you should.. Whats worse is there are a few battery sellers floging them with their batteries..
what about security for fire hazards, flooding or even thieves ?
That 70kg metal box covers most of that.. and the hill i live ontop of would take care of the flooding...
Unplug the Batriun to let the Active Equalizer work.
If the two systems are connected, it will not work correctly and there will be a gap between the series.
Greetings.
Batrium does nothing unless its within a range that I've set - it won't even turn on till 4.13v and start balancing till 4.15v - It just sits there drawing =/- 1w of power waiting till the time it needs to 'manage' the battery.
@@HBPowerwall Ok mate.
Best regards.
i think lower the voltage of 1 of the banks and check the current flow into that and see if it is only low voltage cell that gets the higher current
It's OHMs Law. It's not wire length or anything like that. Working with voltage differences between cells in the millivolt range you will be challenged to get the ballancers peak current rating outside of a direct short.
I haven't seen anybody get the rated current out of 'em.
And yeah, it's a balancer. Not a BMS.
My solution is to top ballance each cell every once in a while and be done with it. :)
Yep, with all the rain I just left my house battery off for 10 days, the longest period it's ever been off in about 6 years and its within 0.2v good enough for me..
Hi can you suggest a quality bms definitely not china.
Thank you
www.batrium.com - all day long....
@@HBPowerwall thank you I will check it out.
Any other bms which can commercialised
only one I trust is Batrium - have 4 in production
I bet the charge level on each is directly proportional to the wire length
The balancer will see voltage and a longer length of cable will drop more voltage than a shorter one
Sounds Likely
it is not about cell voltage , the active balancer is not designed to this mush big battery .it is about average voltage after charging the capacitors , you should check how actually it is working so you can understand what i mean
I know how they work, they move energy around with out logging that energy and end up HIDING critical issues that would be caught other wise with a good BMS..
8:48 That's definitely not 3mm diameter conductor and I doubt if that's 3mm2 conductor. I think there is too much chinesium oxide in those components and that manufacturer is out to break your storage system and/or your confidence. If they were a problem in previous version, they'll a problem for every product they make. As Marty would say: th-cam.com/video/v8CK6CLGTjM/w-d-xo.html
OMG that in the bin is getting old FAST lol but yes it might be less than 3mm but that's what was on the spool
Whats the total resistance of the wires you have extended? I don't think balances can boost voltage, I think they are just higher voltage cell carriers - but I may be wrong.
You'll find that the voltage drop in the cable will cause less current flow from cell to cell. It's a direct result of low voltage vs high resistance.
I have a couple of active balances in some of my early badly matched capacity 18650 batteries. Because these packs are in parallel with many other packs, the constant load is very low, no more than 10 amps at worst and usually only for an hour or two, they mostly sit discharging just a couple of amps. So I find the balancer does a good job keeping them all in line. But if these packs were my only packs and I was cycling them pretty hard, the balancer wouldn't have a hope in hell keeping the pack happy.
You've really got to pick and choose when they are suitable for use.
hey mate isnt it Heltec? you spelt it Haltec on the video title. just a heads up. loved the video :)
My bad, woops..
Thanks buddy
I love watching creators interact with their audience. It's just awesome to know there is actually someone behind the channel. I know its obvious that there is, but seeing it is just awesome.
it's the best part, i look forward to releasing a video just to have a chat..
Peter, the balance should only be switched on when batteries are near fully charged. Otherwise it will disrupt the balance.
Thats not how these devices work unfortunately,
@@HBPowerwall There is a point on the board that has two terminals soldered. Remove the solder and a switch can be wired in to turn the balance on or off. Then have the balance turn on only when voltage is above a set value. I am planning a rather large build 25 kW PV and 100 kWh battery but would like to be part of an Australian forum to get help if needed and maybe contribute..
Yes, My Batrium BMS is hooked up to that port and controls it to a degree..
Those thick DC Cables must be the same length so the cells stay balanced, even the balancer cables must be in equal lengths
I think for commercial application that is 100% true but for mine, the differences are so tiny it won't really matter.
Good video. I just don't think the current generation of active balancers are able to handle the massive parallel banks of batteries we're working with. Relatively flat discharge curves, and enormous Ah storage is never going to work without an intimate knowledge of SoC.
I'm new to this, absolutely no experience. Buuut : If the balancer stops the current flow, before measuring the voltage, then cable length does not matter (and because of that, it should do it like that, but maybe it does not). If it does not, and measures voltage while balancing, then you're possible better of having them all the same length (when a current flows, you get a voltage drop because of the resistance in the wire (veeery little, but not zero)). ...... In general, wires/cables are better of just laying messy around with lots of space in between. If you strap'em close together you have started (somehow) to create a transformer, and if you then run AC (alternating) current through one of the "transformer" wires, then you get a induced current in the very very nearby wires. ...... But that's or AC. ....... Here it's DC, and I'm not sure now, but for DC, I think it doesn't matter (no induction). ......... So my guess is that the firmware is measuring the voltage while balancing, and then the cable lengths should be exactly the same, idealy,....... but if it matters ?,....... probably not. ......... The guy who sugested that the balancing was only done with adjacent cells, maybe he's correct, as the bars on your display could match such a process. But I fail to see why a balancer has to be engineered like that, except, it's probably cheaper to produce like that. .... If this guy is correct however : th-cam.com/video/bFOJdvxhJbo/w-d-xo.html ......... then there possibly are kind of AC currents flowing back/forth in the balancing wires. .......... And if he's correct, that neighbour only balancing is not the case (maybe is in the BMS ?).
Do you really believe this little balancing board will manage to do his job fast on those beasts battery packs?
No but I would also expect that it would remove energy from cell 7 & 9 either.. that is unnecessarily cycling the batteries even more.
At .5 to 1.5 amps I don't think a active balancer can work on such high of capacity system. If you put 20 or 30 of them on your system it might work. Otherwise I only see they would work on a system that is ten fold of the balancing current, or about 10 ah per board.
I would say this setup should be fine.
It is and it works, but active balancer is default turned off and haven't turned it back on since this video
Not worth your time. But I do appreciate your hard work.
Weeks later and i'm still not convinced they are a good idea.. I mean they work like it says on the box but no much better than a cheap BMS with the balancing current 99.9% of the time being in the mv range not the 5 amp max it 'could' do.. and to be honest that situation would me you have MAJOR issues with your battery that desperately need to be addressed not hidden
wires are too long for non capacitive active balancing. Try their BT version, it has impulsive capacity active balancing with 4A, wire length does not matter for it. It stores energy in capacitors and transfers it to lower cell in pulses.
That’s exactly what this does
@@HBPowerwall he is right. It wont work the same way.
for that length of cable you will need to have a thicker cable, to let 3.6v travel in 2 meters of cable without loss you will need 10mm2, wire resistance....sorry pete
, 👍 sehr gut 👌😎
Hey , in total full loded u got 200ah maby more maby les , that 5 amp active balancer lets say its balancing 2 cells in 1h , u got 2000 cells there maby more maby less , so that means they balance in 1000hrs if they completly unbalanced , the point is if they are not new cells the charge and discharge are not so eficient in egual mode lets say 2 new cells charge and discharge at the same time , but if not new the charge and discharge are diferent betwen them so u balanced them in 1000hrs and u disbalanced them in not more then 1 hr and ther it goes charging frome solar system first to the good cells then harder to bad ones , in paralel they balancing at highest speed betwen them and in series used to a consumer the bad paralels discharges faster and consumes ah , when the good paralels berly consumes , so there we go begging fore that balancer to balance 2000 cells in shorter time butt it cand still same process 1000hr without u charging or discharging it , so in my opinion add 10 more or more active balancers and putt em in paralel , its not balancing the way it shud normaly to smaler pack so that that u got is a huge mega pack so to kepp up with charging , consumers and balancing add good bms and good balancers , more and use them same in paralel
chop chop pete..... solve the riddle
All over it like a two dollar hooker on a five dollar note!
I know how stressed out you got making this video I know your not keen on these active balance boards like I said I was skeptical but I have been presently surprised and someday I will sort out the spiders web lol
Yeh fix ya spider web will ya lol I'll get it sorted one way or the other.
If you have to go through all these stress just to use it, then i don't think it's worth it.
Its how I learn and helps earn a living.. Gota learn something new every day
🤓👍🖖✌️👌😎
⚡😃😂😬😁
First! :D
Congratulations, i'm making you a moderator and giving you powers beyond a normal viewer could ever dream of.. lol
@@HBPowerwall :D
These things are ticking time bombs that just have to go screwy and will then destroy your pack. Everyone obsessed too much about balancing.
I agree they solve a problem that shouldn't exist
well they try to..
I would much rather the balancer to be in the BMS, however, my small pack (3s16p) has its BMS balancer not working, they come with those shitty passive bleeder balancers that just drain the high cells as they go above 4.3 volts (yikes that is way too high for any lithium ion variant), But they won't activate until you get them to go to full charge, so for people never fully charging the pack they never get balanced. An active balancer not only balances way faster than a passive bleeder balancer, but makes less heat and does so at any voltage... this is a downside for LFP and other lithium ion types with flat curves, but for LCO and NMC lithium ion, it's nice as they should have the same voltage per cell all the way from top to bottom, they should all fall at the same rate, and they all have different voltages across the discharge and charge curves.
So if your BMS is defective like mine that screwed my balance in 1 cycle (perfectly balanced parallel banks, next charge 2 were at 0 volts and 1 was at 4.3 volts), they're perfect, and if you use something other than LFP (LiFePO4), then they can work fine 24/7... though mine is an inductive type, not the capacitor type (shown here) or the transformer type (the really high current one Heltec sells), mine works fine for smaller strings and has all solid state parts (ceramic caps, not electrolytic caps), downside is that mine can only transfer energy from one cell to the adjacent cell, not across the pack, so a balance has to propagate across the cells... not a problem for 2s or even 3s, possibly 4s is fine... but you reach a lot more series and have one cell on the end with higher or lower capacity or just a bad balance... well each transfer stops at a set voltage difference, so the difference is fine at one cell across, but gets worse and worse as you go along and they are each a set voltage different. The capacitive type (shown here) transfers energy to all cells at once from all cells at once, so all low cells get charged from all high cells at once.
Still, I'd rather have the active balancer be built into the BMS so that it can monitor itself and so that it's only one part to fail with one balance lead and one balance wire per cell, not 2, one for the BMS and one for the balancer. Plus, a BMS designed for LFP could also be programmed to only activate the balancer when the cells are at the knee in the graph rather than the flat part which is where the balancer might unbalance the pack instead of balance it, then a BMS made for LCO or NMC like my cells could have the balancer work all the time there is a significant difference, at any state of charge, because they do not have a flat discharge curve.
You are relying on a for profit company to over deliver. If it doesnt say it, it doesnt do it. Maybe try to keep them honest in a different way. The more complicated, the more outs they have.
I'll just share my results good and bad and let the public decide.. either way i'm learning something new