as soon as you push them both charge and discharge them heavily, they start to go out of balance, as soon they settle down they will go in balance again Forgot, dear german or austrian Andy in Australia, thanks for your very informative and great entertaining videos from another lifepo4 nerd!😀
Your efforts to plan/buy/build/iterate/video/edit/upload/work/generate capital, your brain power/tenacity/passion beyond amazing, thanks, I will donate!
I am just starting, I have 2 , 12.6 v Chins batteries configured in 24v system. And have just learned learned from you the high voltage cut off limits that doesn't turn my inverter off. I have learned a lot from you. Thanks man!!
Excellent test and information! It also confirms what I've observed with active cell balancing compensating for variations in individual cell chemistry during cycling. When I turned off active balancing in two deep charge cycles the pack balanced nearly perfectly.
Hi Andy! I agree with everything you've said in the video. The active balancers do create a massive imbalance in the flat part of the curve. What I have done with mine is to add a switch that I use to manually turn the balancer on and off during the absorption phase, also set at 56.8V. This was done by desoldering the "RUN" contact and adding the wires and the switch. I had a Daly BMS which always showed the incorrect voltage to the inverter, so I assumed it was faulty. Took that out and replaced with an active balancer and a 125A circuit breaker. The balancer is now turned on during the absorption phase and within minutes every cell is now balanced again. For those with a working BMS like yours, it will keep everything balanced, but adding the active balancer to that will help to balance quicker on the days when one can sit and monitor everything. I do this once a week or so over the weekends. The rest of the time all should be well.
Thank you Herman, that is great advise. A manual switch is certainly the easiest solution and you just flick it from time to time and to a 'maintenance'.
I love to watch your experiences with LiFePO4 batteries. I have learned a bunch!!! I did buy a balancer (12V), but never used it. The BMS's do their job as you described in this video. Thanks for your videos!
Wow this is very interesting how it unbalances! Yes very true Andy, you have to trial and error and do tests to find out why and it is very helpful to us all!
I think they are perfect for maintenance, depending on the deviation when it goes out of spec. Thanks for sharing and educating us. (You said in the beginning this is the off-grid garage from Austria 🤣😂 0:35)
The other option is to put enough active balancers in parallel that they can transfer the energy fast enough at both ends to make up for what happens in the middle. That's effectively the situation I have on all of my LiFePO4 powered cars starting batteries. Balancer can do up to 5A or so, cell capacity 8AH or 16AH, never a problem, goes to full voltage together or goes flat together.
I could be wrong, but I don't really see the value in these Active Ba lancers. I have a 16S Haltec and a 4S Haltec and I can't really say either does much. Maybe for very mismatched cells? What has been priceless is all your hard work and sharing information. Thank you Andy!!
So, only use the active balancer when not charging or discharging and at high enough cell voltages to also have a high state of charge. Thank you for taking the time with the exhaustive testing as usual.
Yeah, as I said, this is nothing new. We just did a detour to get there. The active balancer is not needed if you charge to such high voltages as the BMS can get the job done just fine as shown. At least it's not needed all the time...
This is how i think it works... There are slight differences in the capacities of each cell due to manufacturing tolerances. This is like having a series of tanks of slightly different widths. When you top balance... You fill all tanks to the same level regardless of their size. Since they are all in series, as you discharge, the exact same number of coulombs are taken from each cell. In this tank model, this is like taking the exact same volume of water out of each tank. The smaller tanks end up having a lower water level than the bigger tanks. As you recharge, the exact same amount goes into each cell. The level in the smaller tanks rises faster than the level in the bigger tanks. They all arrive at the same level once more when they return to the original, top ballanced/full level. But if you had rebalanced the level in the tanks when they were at their lowest point, then refilled them... The smallest tank would still fill faster than the biggest one, and it would become overfilled before the biggest tank returned to the top ballance/full state.
Thanks for this lecture. according to this clip: 5:34 - 5:47 . Once the bms turns off, i expect the active balancer to have the time to balance the cells without interference
Thanks Andy for all your work. I learn a lot from your quest to knowledge. For me the best way is NOT TO NEED TO BALANCE. To minimize unbalance I prefer to use multiple banks with their own BMS. Unbalance is created due to the difference in internal resistance on each cell and is increased due to high current. It is not the same to take out 200 amps from a 400 Ah battery bank than take out 25 amps from four 100 Ah ones. Personally I think that BMS are simply not created for those big banks. The current capacity of BMS had increased on the past years, but balance current on them had remain the same. You will notice that the unbalance will be less when you add the other battery banks on your system. This will be a good topic for a video.
Great feedback and comment. That is pretty much the case, yes. The Daly BMS has only 25-30mA, so what can you achieve with that for such a pack, unless you charge really high and stop charging to give the BMS time.
I have very much enjoyed watching you discovering that the out of the box profiles are good and that the BMS manufacturers are designing internal balancers that should do the job!
Hi sir, im not using active balancer for 2 years in my lifepo4 battery and no problem in cell unbalancing😊 I just simply charged my batteries individualy to 3.5 volts and connected them in parallel for 1 week then connected in series with bms,. Till now my batteries have a 20mv difference, and im glad to share my experience with you😊
Thank a lot Andy. I do have to install LiFePO4 in my RV, and nw I'm not sure I'll use the Active Balancer. Depends by the use of the batteries I'll have. I'll keep them under control and let you know. Thank again Andy.
Same here. My 4S HANKZOR active balancer is disconnected for now. But the electronic board has a soldered connection between two pad labeled "ON". I plan to desolder and solder a switch for controlling when to active the balancer. As you noted, balancing (passive BMS or active device) should only occur when charging which would require a constant boost voltage stage where the charging current tapered down to 0A. But as you show, the BMS balancing is fine. The important part is setting up the proper charging profile. Some LI chargers do not have a constant V charging stage.
Yep, I tested this little on/off function in one of my videos. A simple switch would be sufficient and you just turn it on form time to time when the battery is a high SOC and let it run for a while.
That is the reason why the balancer build in a proper BMS will only balance during charging and not during discharge as doing so is counterproductive as you found out. So if you can get a signal when battery is charged and can use that to enable the balancer then all will work great. The lowest capacity cells also usually have the highest internal resistance and those cells will have higher voltage during charge but lower voltage during discharge thus a balancer that has no idea if battery is charging or not will just waste energy by unbalancing then trying to balance your pack.
So what your saying is BMS's have top balance circuits and that is the correct way to balance lithium packs. His active balancer does have an on/off switch, but I don't know how that will help? Probably just throw away the active balancer I guess?
@@BischesseHunting Bad idea. That is why mine has separate paths for charge and discharge. You do not want to absorb and or float charge a Lithium battery if you want long life and so as soon as battery is charged the charging should fully stop and not affect discharging.
@@uhjyuff2095 Yes top balance is what anyone in energy storage industry uses (DIY energy storage will not be counted). Maybe that ON/OFF switch can be used but he will also need a signal from the BMS or whatever measures the battery current that enables balancing when current flows in to battery (charging). I set mine to start cell balancing only when battery charge current is above 300mA. Very small cell balancing currents are needed to maintain a battery in balance most EV's with huge 40 to 100kWh batteries have cell balancing currents below 100mA. With LiFePO4 voltage at the end of charge rises super fast it can take just seconds to get from 3.4V to 3.6V so that is not an indication of the state of charge. You can have a cell at 3.45V and one at 3.55V and delta in state of charge between them can be a very small fraction of a percent thus not an indication of how well the pack is balanced. My house battery (fully off grid) was never disconnected for the past 4 years or so and it is in perfect balance with just over 100mA of cell balancing maybe a few minutes per day in average. The only time the battery was empty was after first year when I did a capacity test to check degradation (less than 1%) and last year when I did the same test and I may do another test next year when there will be 5 years from when it was installed.
I had the balancer connected for ~3 weeks now and could see the battery struggling when it's fully charged. Lots of cells were peaking, even the BMS disconnected the battery twice due one cell reaching the cut-off voltage. You could have the balancer kicking in at say 3.5V to help balancing at this point. There are circuits out there which can do that through a relay. But do you really need it?
Yep i just came to the same conclusion. I've spend near $400 on those deligreen active balance modules, and have noticed the wild out of balance variance during absorbtion. Going back to top balancers again I think.
@@OffGridGarageAustralia I plan to wire in a switch (as discovered in your other video), then balance manually if needed. I *think* it should remove the need to disassemble the pack for maintenance.
When you will fix the relay on your previous BMS, you can also modify it, tapping on all the resistors are burning the extra charge. If the BMS switch the resistors on the low side you need a 16 input AND gate and an extra NOT gate. Connect the gates inputs on the low side of the resistors. If the BMS switch on the high side you need a 16 input OR gate. The output of the gate can drive an optoisolated relay (or a MOSFET) that connects your active balancer. You will need a transistor and a diode if the relay is not optoisolated. Doing this, when just single cell is getting balanced by the BMS the active balancer will kick in. Not before and not below the voltage you have set in the BMS balancing voltage. If everything works you can replace the low resistance burning resistors with high values, so you will not waste any power and all the balancing will be done actively. Basically you will have a BMS with smart active balancing. I hope I was clear. If you are interested I can explain with a deeper level of detail.
@@OffGridGarageAustralia in the while I thought of a even simpler implementation. input side of 16 optocouplers with a current limiting resistor (if not integrated) replacing the load resistors (basically every optocouplers will have his own virtual ground). All the output collectors should be connected to B+ (or a intermediate cell to not exceed the optocoupler max Vce) via a pull up resistor. All the emitters to ground. The common collector can drive a MOSFET will drive the relay (on the low side), to connect the active balancer. A freewheeling diode is needed if the relay is not optoisolated. The MOSFET could also directly drive the active balancer, but I guess it's better to switch on the high side (I don't know exactly how the active balancer chip get the power from) This implementation will make it low-high side balancer switching BMS transparent, reduce the BOM, and most importantly...safer. A shorted transistor/gate chip could lead to a disaster. If the BMS already uses optocouplers to turn on the balancing resistors, only 16 higher value resistors (to replace the load resistors) and a MOSFET and/or the relay are needed.
I'm yellow with envy for your 50A off grid setup. My 20A should be plenty to charge my mower batteries for some afternoon mowing occasionally. I do have some crazy plans for running a small coldroom in the future..
Thank you Mr.Andy.if been wondering about these balancers.i have a small 24v setup in camper. my cells do the same as yours one runaway while one charge is still low. Balanced pack hoping for better results
@@OffGridGarageAustralia I have a 4s balancer like yours. There is a solder blob on the "Activate" pads, just replace it with a switch and Bob's Your Uncle!
@@OffGridGarageAustralia Then you know that you could leave it connected and just switch it off and only use it when needed, or maybe there's an option on your inverter/charge controller to use a "External Balancer" and have it switch it on and off at the appropriate times? Also have you turned it off when it is trying to balance to see what the ACTUAL CELL VOLTAGE IMBALANCES are?
@@mysterytechknowledge3664 I can see the actual cell imbalances under load on the BMS app. I showed this in one video, it was like 70mV at 80A. If the balancer pushes only 1A for two hours, that's enough to create chaos for when the battery is being recharged.
Andy, if you are getting large cell deviations during discharge, you may not have well matched cells. Have you checked the internal resistance of the cells? There is a formula to do that is you don't have a cell monitor that can test it. Also, I have one of those balancers (4s) that I use with my Headway batteries. Something that I noticed is that when charging the pack and monitoring the cell voltages, the balancer will "artificially" increase the cell voltages that are low (and vice-versa) when it is actively balancing. This makes it hard to accurately monitor voltages unless you turn off the balancer. In my case it's easy for me to unplug the single 4s cable. However, you may want to solder a switch and some wires onto the balancer. (inplace of the solder blob that on the "activate" connection.) It would be Very interesting to see what the ACTUAL voltages of your batteries are when charging and discharging by turning on and off the balancer with a switch!
This is why a BMS with active balancing built in usually allows you to set the start voltage when it starts/stops balancing or allows you to only balance on charging.
Hi, same here, since you/he ask, I use the jikong active balancer bms that is on the OffGrid Garage webpage to balance over a set voltage. Works for me and good to selectivly stop charge or discharge. But normaly only needed if the charger or inverter would fail to limit themself normally. Works as overload disconnect aswell that could avoid blowing fuse. Tested when activating when not using the precharge.
Excellent Video ! I have older Daly nonconfigurable BMS ( kicks in over 3.50 V @ 30mA. I don't use this as I normally charge my battery to a SOC value ( 70 - 80% ) and only resync my Victron BMV every 3-4 weeks - error is only around 0.3Ah/DAy on average so super accurate. ( I set Peukert to 1.00 and Charge Efficiency to 99 % ) I manually rebalance every 2- 3 months or so at 3.425V /cell and have worked out the sensitivity of change in Voltage to change in AH that needs to be applied or removed from the imbalanced cell. This sensitively is 0.0020AH/mV / 100AH Cell. So for a 10 mv imbalance you need to correct by .02AH for a 100 AH cell ( at 3.425 V ) . Which is tiny ! This factor will change depending what your absorption voltage is - I like 3.425V.
Wow, that is great information. Thanks for that. I will be in the same boat and rarely get to 100% SOC with my setup so the smartshunt does not reset and keeps coulomb count. But as you said, that's pretty accurate.
Andy, you need to get a Nissan leaf battery and see how they do it. BMS knows the internal resistance of every cell. I think it takes this into account when measuring balance in charge and discharge!
I really wish that most of the BMSs available either had active balancing built-in or had an aux output for controlling an active balancer. The issue with most BMS units is that they turn off charging before turning on resistive balancing so there is no way to have both on with the given settings so they go back and forth between charging and balancing until the battery is full on most BMSs since the low-current balance can't keep up when charging at several amps. So if you need more balancing then having an active balancer and a voltage-controlled relay board to turn it on and of as needed when charging seems to be the way to go.
I have the option to turn off charge-balancing in the BMS which I have. It does not make sense to balance while there is current coming in. Eventually, I will just stay in the flat part of the curve and may not have to balance often, maybe 1-2/year.
The moral of the story is (drum roll, please): Stay away from the "knees". Not does operating your cells within a conservative range dramatically extend the longevity (a.k.a. lifecycles) of the pack, but cells tend to remain within a small Vdelta that the BMS balancer is able to remedy...given proper configuration. To that end, Andy, are you able to post both your BMS configuration and charge controller profile settings? That might prove extremely useful for some. As for active balancers, I build every pack to be active balancer-ready. That way, if and when things go a bit wonky, I connect the active balancer for a cycle to bring everything back within tolerance automagically, rather than resorting to old school manual balancing. So they do have their place. They're just not a panacea.
I'm installing one on my ebike battery and in my case I think it will prevent my bms from cutting off because of 1 or 2 cells being .5v low. But this is interesting to see how large constant loads actually confuse the balancer. With my bike the BMS usually cuts off the second I press the throttle at low SOC. My thoughts are the balancer would buffer that. Thus getting more range out of my battery. So i'll install a switch and see what works better, leaving it on, or just top balancing. Also, my battery is li-ion 14s 3p. It's only 12ah at 52v. I have to wonder if this balancer is more effective on smaller capacity batteries. But this is something very few people make good videos about. Thank You for your amazing videos!!👏
I think in order to get those active balancers to work you need to turn them off when you're below a certain voltage so they're not doing anything normally and only really kick in when you're at 99% or so charged
That is exactly correct! But then, do you actually need one at all? As shown in this video, the BMS's balancer was totally capable of balancing the whole battery to 2mV within ~30min
From most of what I have read a single series pack at these high voltages will always go out of balance. I run a BMS and active balance because I have an iffy cell.
After having viewed other channels and read some on balancing, I tend to agree that active balancing all of the time is not needed and could be a mistake. I did note that Herman Jordaan installed a switch that he uses to control his active balancer. That seems like a good approach. I did note on another channel that periodic use, say once or twice a year, of an active balancer only on one or two charging cycles may be the ticket. That particular channel complained about how the installed BMS, I think that it was a Daly, was not maintaining balance and active balancing was required. His problems reminded me of the decision process that you used to incorporate an active balancer into your system. I had already viewed this video and directed that person to this particular video to help him with the decision process.
I think I'll have to remove my battery from the van and do a proper top balance (the cells are now balanced at a lower state of charge, so I'll parallel them and take them to 3.6v). Perhaps then my problems will be solved and I won't need a new BMS or the flawed neeeeeeey active balancer. Thank you for your new vid, I thought you'd use a boost converter and you confirmed my fear of some draw while inactive.
You can leave the battery in place and charge the whole pack slowly so the balancer has time again to do its job at a higher voltage. Once you're at 3.6V or above with every cell, disconnect the balancer and leave it off. Connect it every here and then for maintenance. The BMS balancer should be OK to keep it under control.
I wouldn't buy one if it's a new system and you just get started. If you have any problems down the track, you can always add one and check it out. It's easy done....
There is a place where you can add a switch or relay to the board just unsolder the bridge “RUN” on your board and then you can add a switch or trigger mechanism to work with the BMS when the BMS starts to balance and should balance the cells quicker and with less “Burnt” energy loss 👌🏾
I have rec bms with 32 lishen 272ah, recbms has 1 amp balance no problems for me, also tampers amps on victron if there is some unbalance in the battery.
@@OffGridGarageAustralia yes i have victron ccgx, bms connecs to ccgx by canbus, on the end of charge the bms lowers max amps to allow bms shunt to bleed the cells. Better than that is that i also have one raspberry pi with home assistant, i made a script to stop charge at any percentage i want, basicaly when bms reports 90% soc i tell ccgx to stop charge.
@@OffGridGarageAustralia Almost out off the box, in ccgx enable DVCC, SVS and STS, in victron inverter i set other canbus BMS, then after that BMS commands everything. You then use bms software to select max charge voltage per cell, max discharge/charge amps etc.
@@piritech aaaah, that's how it works. So you need to set a setting in the SCC as well to listen to the BMS. That part was never clear to me before. Thanks for explaining that.
My set of winston 400ah cells are 8 years old only ever top balanced once when commissioned Never used a balancer of any sort when charging at the upper end of the charge cycle at 3.5v per cell i see a maximum of 40mv difference and once resting voltage difference is about 4mv
@@OffGridGarageAustralia i charge to 3.5v per cell (14v) hold and absorb until charge current drops to 50% output which is 50amps my charger is 100amps The bank wouldn't be at 100% soc but at this point internal resistance is building and I see no reason to charge further. Bms has never had to interfere in 8 years of use
You could have bought a turn key solution like battleborn and that includes matched cells and a BMS but no knowledge transfer, THANK YOU, money tight now, will donate soon!
I think you’ll incorporate the active balancer during the year when you don’t have as much sunshine so your top balancing time will be shorter due to cloud cover. With the voltage triggered relay like you mentioned.
I'll try without for a while now because the cells are now all top balanced. I want to see when they start drifting again without being disturbed by the balancer. Then I can probably develop a 'maintenance schedule'. Eventually, I will stay between 3.1 and 3.45V so I should not se too much drift over time.
It is confusing why the BMS sees these different voltages during the discharge and charge when they are all starting out balanced and the same current is being drawn on them all. I would think since the curves are so flat for this chemistry that they would all stay very close during the "middle" of the SOC. I do agree with your explanation of why the balancer is not helpful during these times. Thanks for all your experimenting...we're all learning a lot!
Thank for your feedback. The balancing at lower SOC does only occur when I pull higher loads and the voltage spreads out over a longer time. This lets the balancer kick in and it starts its job. Can't blame it for that 😏
This is how i think it works... There are slight differences in the capacities of each cell due to manufacturing tolerances. This is like having a series of tanks of slightly different widths. When you top balance... You fill all tanks to the same level regardless of their size. Since they are all in series, as you discharge, the exact same number of coulombs are taken from each cell. In this tank model, this is like taking the exact same volume of water out of each tank. The smaller tanks end up having a lower water level than the bigger tanks. As you recharge, the exact same amount goes into each cell. The level in the smaller tanks rises faster than the level in the bigger tanks. They all arrive at the same level once more when they return to the original, top ballanced/full level. But if you had rebalanced the level in the tanks when they were at their lowest point, then refilled them... The smallest tank would still fill faster than the biggest one, and it would become overfilled before the biggest tank returned to the top ballance/full state.
All these tests and your recommendations led me to buy 2 active balancers. I am on a sail boat and because we are often away from shore connection, we will rarely achieve full charge/SOC. I should have listened to @WillProwse who said that active balancers are a waste of money and don't work. 😒
only connect it when you need it (especially during higher state of charge), and you should be fine. Alternatively if you are stuck at sea with no sun for long and BMS hits low cell cutout due to weak cell you vould connect it and it will shuffle power to allow BMS to turn back on a bit longer.
@Ralph They are not a waste of money. Connect them and have a switch to turn them on or off. If you fully charge your batteries once in a while, turn the balancer on and they will do a great job.
I've been using this different kind of balancer that basically is a line of resistors for each cell that turns on when the cells reach whatever voltage it is and just applies that resistive load to that one cell So the one I have is for sales and it turns on each cell at 3.44 volt so whichever cell reaches that first gets the load applied first and I stopped charging once all four cells have the load applied to them once the voltage dips below 3.44 all the load is removed
I don't know if they have one for a 16s or not Also it would be nice if you could program it and change whatever voltage the resistive load is applied at that would be better I did have to go through three different models of this similar product I had one that wouldn't start the load until about 3.66 volt that's totally unacceptable and then another one was it like 3.6v pretty good but I didn't really want to be charging that high
The top-balancing the BMS does should be sufficient (I've said that before). I really don't like active balancers (I've said that before too)... At worst they mask a bad cell to the point where that bad cell might cause a fire, leak electrolyte, or otherwise damage the rest of the pack. The solution to having a bad cell (as long as you don't have too many bad cells) is to replace it with a new one, simple as that. When throwing new cells into an existing battery pack, the new cell's full energy storage won't be used but it will enable the full energy storage of the remaining cells to be used. Its the easiest solution. Basically everything will be top-balanced, so the new cell won't be drawn down quite as much as the other cells during discharge. It may seem to go 'out of balance' in that respect, but it really isn't... it will re-charge back to approximately the same properly top-balanced voltage. One other facet of this sort of repair is that you might have to adjust the low-voltage shutoff for the whole pack to be a bit higher to avoid constantly hitting individual cell low-voltage cut-offs. The new cells will raise the whole-pack voltage a bit at the point where the rest of the pack is fully discharged. So just keep that in mind. Its less stress on the pack. It would be quite interesting if you did a test like that, actually. Throw in a brand new top-balanced cell and see what happens during the discharge and recharge cycle. -Matt
Thanks Matt, great advice and comment. I have 16 new cells here so... but I don't know which is the 'faulty' cell in my pack (if there is one). After the top balancing and removing of the balancer, it is a different cell every time which peaks out a bit when charging to higher voltage. I think this is good.
If I could, I would say to myself, "Welcome to the club."After doing top balancing as per the book on the 32 302 Ah cells, I build the battery bank, I connected the two active balancers to the two series of batteries connected in parallel,and after a few days , the gap between them reached 120 mV,at 3.47 volts at as I charge the cells,55.5 volts for entire bank..This gap increased progressively over 5-6 days.The balancing option has been permanently activated on the two active balancers. They are devices with bluetooth and Android applications. After watching your video I realized that you are right and that balancing at low SOC is not okay, and maybe it's not okay even at high SOC. So now I disassembled the battery bank, I do top balancing again and I will try again without using the balancing option from the balancers.Or should to enable the option only when the chargers switch to absorption mode, and disable it when they switch to float mode? How has your battery bank evolved, and the gap between cells been kept so low?
Connect the balancer to the relay on the Victron charger (if you have a model that has the relay terminals) and set it to switch on when in float mode, this way you will only balance when the batteries are fully charged and it will switch off once you go back into Bulk mode.
My SCC have only the Virtual Load settings which feed back to the Raspberry and it can trigger a relay somehow I think... But again, do we need the active balancer as it seemed to work fine without. Let's see what happens in a few days when I fully charge again and the battery has been used...
@@OffGridGarageAustralia My system struggles to stay balanced without the active balancer, it end up over time about 300mV delta without and 10mV delta when used in Float mode only. For me I need the balancer.
The question is why you get such a high voltage deviation on discharge. I would assume that the cells are pretty equal at this low cycle count. So as mentioned below, probably the connections are not all equally good. This should be fixed anyway, as this may worsen over time, until you get hot poles, heating the cells and causing more degradation at this cells. Unbalanced cells are just a side effect. At the moment the top balancing of the BMS seems to be good enough to handle this. Another problem seems to be that this balancers start balancing at only few mV, instead of having a higher start level. The ones I use start at 10 mV. Of course, in your environment with very high discharge over several hours, the imbalance introduced is increased. In setups that use lower C-rates you will probably not notice the problem. Having a possibility to turn off active balancing would be helpful sometimes.
The Ri is actually pretty matched across the cells. But when I discharge the battery with 80A, that creates the unbalance due to small mismatches. There is also different resistance across terminal-busbar connection, cable connections and more. So this all makes the cells spread over time. Of course if they would have 100% the same Ri, it would be less, but this is only one part of the problem.
oh, you're in Austria ..I didn't even know that this country is so big that it stretches across several time zones (always thought that with the end of the 'Danube Monarchy' the former empire would be shrunk to a manageable size ..stupid me). Presumably a small, charming, "Freudian slip of the tongue" crept in during your introduction because as an Austrian you can only dream of the many hours of sunshine that are likely to be the norm in your area) .) 'love the scene at 22:16 (terrified scream with half-second time delay ...well, that's how we humans are). I myself have my own problems and doubts when using my "Daly" BMS and with balancing. Glad that I saw your video in time because I was already tempted to get an active balancer. Maybe I just should be a little more tolerant with my slowly working BMS ... thanks for sharing your successes and faults with us ..informative and funny.
LOL Shame on you Andy for taking advantage of the beautiful Mermaid! I laughed at that one! And I imagine she will too. I have two wish list items for you if you will. One i would like a screenshot of your controller voltage settings. I have a similar system as yours (my system is 24V 40kw). I know you discuss the voltage settings but a screen shot I can absorbe better. Second on the wish list is the DVCC setup for a Cerbo GX. Thank you for your in-depth videos!
I will make a whole series of videos here on the channel soon explaining the settings of all my devices. Quite a few people have asked for that. I have also planned to add these info to my website, so it can be easily found. So patience, young Padawan, it will come soon...
Unlce Andy, thanks for another informative video. Probably not logical but would parallel of say 3 or 5 of these active balancers not work on solving balancing issue at any SOC, PS when you building that battery shelf further. Thanks, Wayne scheepers
Thanks Wayne. I don't think this is the way of doing it. It's not about transferring more and more amps. See my pack is now perfectly top balanced and the balancer of the BMS with 150mA is capable of holding it in balance at 3.55V. The balancer just created a mess as it balanced all the time. Imagine having a few balancers doing that, the mess mess would be even larger.
@@OffGridGarageAustralia very true, I got one because some weeks it get cloudy and the SOC gradually drops, and its just that 1 cell that gets to 2.6v and then boom house goes offline.
@@OffGridGarageAustralia Yes, the balancer works all the time, but obviously are the 5 amps of the active balancer not enough to keep them balanced at any state. It would be interesting to see what happens with more balancing power...But I agree it would be better to run the battery by itself without much intervention.
Had the same thoughts about active balancer un balancing when in a lower soc. People kept saying no they work great etc. purchased two of them. still sitting on the shelf while i sit on the fence. I think i am coming to learn that my initial thoughts may be true.
I think, it depends on the setup people have. If they have shallow DoD and have the battery mainly running from 70-100% it may work for them actually. Also, if the have only low discharge currents which does not cause a big voltage deviation, it will work too as the balance does not cause any trouble then. It's really not an easy decision. If you constantly run into mismatched cell voltages when the battery is full, I would probably try them out. If the deviation is not too big, leave it as it is.
I charge my battery up to 54.4V (3.4V / cell) with an active balancer like yours, my cells normally have a difference of 0.010V. I have my BMS configured so that it cuts the load at 3.55V and activates it when it returns to 3.40V, without the balancer my battery does not rise above 54.0V, since the BMS cuts the load due to overvoltage of a cell (normally the 6). The internal balancer of the BMS (280mAh) was not able to balance my cells and I was having trouble using an inverter soldering iron. Since I installed the Active Balancer, my problems are over, I also don't have your problem that the BMS cuts due to overvoltage, since with my configuration the cells stay within the normal range programmed in the BMS. Talking to the seller of my cells (who is a manufacturer), he told me that it is very difficult to control the loading ramp, once they pass 3.35V; In their capacity tests they do not take the cells to the limit, they advise working in the 80-90% load range and this is achieved with 3.35V; They think that raising the voltages higher is unnecessary, due to the difficulty of controlling the cells, for the little benefit of charging that you get. For all this, and watching your videos (which are very educational), I configure my system as follows. Hybrid inverter: absorption charge: 54.4V float charge: 54.3V Gross charge time: 120 minutes BMS: overvoltage protection: 3.55V overvoltage protection disconnection: 3.40V The balancer always on.
Very good experiments. Please continue your tests. Have you measured the [mV] between busbar and pole during high amp discharge? If this is high (like 50mV for some cells) the active balancer will transfer Ah between cells only because of bad connection. In my setup I have this active balancer on all the time, I do regular 60A charge, -50A discharge for several months - overall I see no big problem - at end of charge sometimes one cell goes to 3.46V or so, but then it went down. One big difference from my setup to your is that I have set the floating voltage much lower, its about 3.4V per cell.
Thanks for this Andy very informative. I've been wrestling with the run away voltage on a simple 24V 8s 50A system using a Daly BMS and no active balancer. It has been driving me nuts when one cell quickly gets to maximum voltage and leaves the others waay behind. I don't think the problem is really different battery chemistry as much as these Smart BMS systems are really quite stupid. It just doesn't seem that difficult to take into account different battery chemistry and the typical voltage/capacity function when charging. So as you mentioned it primarily functions as a battery protection switch as using it to balance means endless charging/discharging cycles. Life is too short not to mention wasting electricity and contributing to global warming. I brought a petrol generator for the sake of the planet, humanity and my sanity.
No, they work in the same direction. They both deplete the cell with the highest voltage at the same time. You can add as many balancers as you like, they will all work together.
I recall that the active balancer had a hidden bridge where it could be disabled or enabled, so I wonder if with a relay output from the inverter or charge controller could be used to enable it? Or use a voltage controlled relay so that it only active balances above a certain pack voltage or when the battery is in float?
So it would be wise to switch on the actvice balancer only in adsorption mode. This should be possible if the brigde on the run switch is removed and a little relay is placed there, which is activated by the yellow adsorption LED of the solar charger. Or it could just be voltage triggered.
@@OffGridGarageAustralia Right, a the current state you don't need it. But you have made a top balancing just before. The voltage deviation might increase during time and the balancing current of the BMS is very little. So at a higher deviation it might not be able to finish the balancing before the battery is discharged again. Well, in this case you can switch on the active balancer again. And if you want it to be controlled by the yellow LED, you will just need a few resistors, a capacitor, a foto-transistor in a short black tube that fits on the yellow LED, a zener diode and a little (reed) relay. But I think it would be easier to make just a voltage triggered version, which could be placed on top of the active balancer.
You can, but then you transfer even more amps at a lower SOC as well so this won't fix the problem at all. You would need to make sure, the balancer only turns on at higher voltage so it does not mess up things at lower voltages
You would think the active balancer would be able to adjust the turn on point like the passive balancer in the bms. If a bms had active balancer wouldnt it be much better??
Thanks for your videos! I'm looking to parallel 3 or 4 cells before the series connection. And would love to know if this configuration also helps to keep cells in balance.
I have a 4s battery and saw the same. Active balancer helped a lot getting the cells in balance but after a few cycles one of the cells started reaching 3.65 and resetting the bms. So after it balanced the battery at 3.55v I disconnected the active balancer and also lowered the boost voltage on the charge controller to 3.475v-3.5v per cell and maintain ther for one hour, then float at around 3.375v and now at the top of the curve I see no more than 20mV and I am able to fully charge the battery. I think I will use the active balancer to perform periodical maintance/balance without the need to disassemble the pack and put the cells in parallel.
I'm offgrid and use simple top balancing boards at 3.6V. This seems to keep things within 10mV or so across my entire string of 16x200AH - I measure periodically with my 5/5 digit HP meter. I've held off getting an active balancer so far because it all works well as things are. I do wonder about the accuracy of the active balancer readout - component variations could give significant error on the reported values, even 0.1% component spread will give reading errors of 3.6mV from one cell to another and I doubt that a cheap balancer uses such precise components. Have you compared the actual cell voltages with the reported values?
Thanks Andy. I have not compared the voltages yet. It is hard to determine which voltmeter to believe unless you have calibrated one correctly and know it's the point of truth. If they keep everything within 10mV that's all we need, I guess.
@@OffGridGarageAustralia I guess the voltmeter accuracy isn't critical provided all the batteries read the same value. I don't trust the BMS to report all the cell voltages with the same error because of component spread.
Hallo Andy, ich denke, dass da immer noch ein Widerstand zwischen den Messleitungen und den Polen ist. War bei mir so , aber auch nur bei einer V2A Unterlegscheibe. Dadurch kommt es zu einem Spannungsabfall an dem Widerstand (Unterlegscheibe ) und zu einem Messfehler . Bei meinen Cellen hatte ich bei einem Ladestrom von nur 10- 20 A fast 500 mV Spannungsunterschied, der aber nur ein Messfehler war, da durch den Balancerstrom an den gut 20 Ohm Überganswiderstand diese Spannung abfiel. Seit ich die Anschlüße der BMS und Balancer Leitungen unter die Unterlegscheiben direkt auf die Busbars gepackt habe ist dieses Problem weg. Mess doch bitte mal jede der 17 (34) Ringösen gegen die jeweilige Brücke ( Pol ) , ob sich da irgendwo ein Übergangswiderstand eingeschlichen hat. Grüße aus dem Ruhrpott Guido
Hallo Guido, danke fuer deine Nachricht und Tip. Bei mir sind die Ringoesen seit Installation direkt mit der Sammelschiene verbunden, die Unterlegscheibe ist dann oben drauf. Aber sicherlich hast du generell Recht, Ich habe immer noch ein Widerstandsproblem und Uebergangswiderstaende die mir falsche Messdaten uebermitteln. Sobald die zweite Batterie eingebaut ist, kann ich diese hier auseinanderbauen und mal richtig neu verdrahten, Kontakte reinigen und alles besser machen, was im Moment noch nicht so gut ist...
It depends on the SOC. I can really see that it balances the cells at any time and when I charge back up it's all over the place again. Now, without the balancer, they stay inline. If you balance them at night and they are not full, there is no point to balance in my eyes...
@@OffGridGarageAustralia yeah that is a point it really shows from all the different settings you have done that you cant use none adjustable bms and just forget
Ah, that was just a joke in regards to the last video where I was supposed to say 3.38 and said 3.8V all the time. People commented on that... 😏 The 3.64978V are also not really true. It was mmore like 3.64856V actually 😂
@@OffGridGarageAustralia but amps doesn't show the amount of power being created (Amps multiplied by Volts equals Watts, which is the measurement used to determine the amount of energy.)
Hi Andy been thinking about this for a while your experiment does not simulate the real situation of an offgrid house. Perhaps you can simulate the experiment with a normal consumption of a house . if you go partially offgrid with a residential there are always irregular consumers. a refrigerator, freezer, computers on, off, heating the water, coffee maker, etc. And that all happens in the belanscueve! What would be the end result of the charging cycles in a few days?
That's exactly right and a very good suggestion. I kept an eye on my consumption and made sure I can recharge every day to see the result of balancing, absorption, floating and fine tune that. Now, I have the pump running and also charge the vehicle again from the battery so won't be fully charging for days. It will be interesting to see what happens next weekend when I have planned to fully charge again.
I think the use case is actually more clear than you indicate. The amps capable of being pushed around in the balancer should be equal to the draw on a typical day or input on a typical day divided by the amount of paralleled batteries. 50 amps pushing at 48 volts with no paralleled cells means you would want to be able to push 50 amps around at the worst case. Add your second battery paralleled and you are down to 25 amps. Parallel two of the 5 amp balancer units and they are dealing with 12.5 amps at worst. Split this over the two hours of rest time and it is 6.25 amps per hour. You start to get really close to having protection in the worst case scenarios. As to whether you wish to build for the worst case is a personal question. With new cells like you have you probably could come up with some calculation for the acceptable level of risk and return. Again, the above is dealing with charging, but with discharging you most likely don't pull a 1C pull constantly. You can mix the math above with a discharge typical load. I hesitate to offer a "rule of thumb" response and when asked I say get the active balancer and get a BMS. The worst that can happen is that the BMS disconnects the battery at the low voltage turn off if your video's described situation occurs and at best it will make sure that the batteries age equally with regards to high and low voltage cycles. After all, if your battery is already kicked overvoltage relatively new you can make sure to limit dendrite formations by avoiding the higher voltages completely instead of just reacting when it hits a high voltage. Thank you for all you are doing. This is a journey that you are bringing a lot of folks with you. There is probably a unified theory of best practices, but right now it is hard to really identify it.
Thank you. Yes, that is correct what you're saying. I think the best practise is to keep the battery in the flat area of the curve, so 3.1-3.45V. If you need more capacity, get a second battery. In terms of the balancer. I don't like the fact that it balances all the time, charging discharging, high/low amps in and out. It just causes trouble over time. That's just not how a battery should work. But as I said, I'm experimenting and exploring all these different situations right now to get a better understanding why some settings work better than others.
as soon as you push them both charge and discharge them heavily, they start to go out of balance, as soon they settle down they will go in balance again
Forgot, dear german or austrian Andy in Australia, thanks for your very informative and great entertaining videos from another lifepo4 nerd!😀
Your efforts to plan/buy/build/iterate/video/edit/upload/work/generate capital, your brain power/tenacity/passion beyond amazing, thanks, I will donate!
I am just starting, I have 2 , 12.6 v Chins batteries configured in 24v system. And have just learned learned from you the high voltage cut off limits that doesn't turn my inverter off.
I have learned a lot from you.
Thanks man!!
Thanks a lot, great to get such feedback.
Excellent test and information! It also confirms what I've observed with active cell balancing compensating for variations in individual cell chemistry during cycling. When I turned off active balancing in two deep charge cycles the pack balanced nearly perfectly.
Hi Andy! I agree with everything you've said in the video. The active balancers do create a massive imbalance in the flat part of the curve. What I have done with mine is to add a switch that I use to manually turn the balancer on and off during the absorption phase, also set at 56.8V. This was done by desoldering the "RUN" contact and adding the wires and the switch. I had a Daly BMS which always showed the incorrect voltage to the inverter, so I assumed it was faulty. Took that out and replaced with an active balancer and a 125A circuit breaker.
The balancer is now turned on during the absorption phase and within minutes every cell is now balanced again.
For those with a working BMS like yours, it will keep everything balanced, but adding the active balancer to that will help to balance quicker on the days when one can sit and monitor everything. I do this once a week or so over the weekends. The rest of the time all should be well.
Thank you Herman, that is great advise. A manual switch is certainly the easiest solution and you just flick it from time to time and to a 'maintenance'.
I love to watch your experiences with LiFePO4 batteries. I have learned a bunch!!! I did buy a balancer (12V), but never used it. The BMS's do their job as you described in this video. Thanks for your videos!
Thanks Peter.
Thanks!
Thank you very much!
Happy to be of service!! ^_^ - x_x
Also, 100%, learning is the fun part. Highly recommend to everyone!
Wow this is very interesting how it unbalances! Yes very true Andy, you have to trial and error and do tests to find out why and it is very helpful to us all!
That makes total sense to me. You do a good job of figuring it out and explaining it. Thanks for the work you do it keeps me sane lol
I think they are perfect for maintenance, depending on the deviation when it goes out of spec. Thanks for sharing and educating us. (You said in the beginning this is the off-grid garage from Austria 🤣😂 0:35)
Comment #8 about Austria. I was keen to see how many pick this up. Also, the 3.8V joke from the last video 😁
Andy, finally you have admitted that you are actually here in Austria! The beer is better here anyway 🤣
Haha . I thought I must have a look in the comments before I make a comment. And I found yours 👍
@@nl1cat and it seems that I'm one of many
The alps are getting sunny this time of year
Now I really don't know if I'm watching Off Grid Lederhosen from Austria?
Luckily Andy cleared this up a few minutes later.
Comment #12
Nice, keep them coming 😂
The other option is to put enough active balancers in parallel that they can transfer the energy fast enough at both ends to make up for what happens in the middle. That's effectively the situation I have on all of my LiFePO4 powered cars starting batteries. Balancer can do up to 5A or so, cell capacity 8AH or 16AH, never a problem, goes to full voltage together or goes flat together.
I could be wrong, but I don't really see the value in these Active Ba lancers. I have a 16S Haltec and a 4S Haltec and I can't really say either does much. Maybe for very mismatched cells? What has been priceless is all your hard work and sharing information. Thank you Andy!!
So, only use the active balancer when not charging or discharging and at high enough cell voltages to also have a high state of charge. Thank you for taking the time with the exhaustive testing as usual.
Yeah, as I said, this is nothing new. We just did a detour to get there.
The active balancer is not needed if you charge to such high voltages as the BMS can get the job done just fine as shown. At least it's not needed all the time...
This is how i think it works...
There are slight differences in the capacities of each cell due to manufacturing tolerances.
This is like having a series of tanks of slightly different widths.
When you top balance... You fill all tanks to the same level regardless of their size.
Since they are all in series, as you discharge, the exact same number of coulombs are taken from each cell. In this tank model, this is like taking the exact same volume of water out of each tank.
The smaller tanks end up having a lower water level than the bigger tanks.
As you recharge, the exact same amount goes into each cell.
The level in the smaller tanks rises faster than the level in the bigger tanks.
They all arrive at the same level once more when they return to the original, top ballanced/full level.
But if you had rebalanced the level in the tanks when they were at their lowest point, then refilled them... The smallest tank would still fill faster than the biggest one, and it would become overfilled before the biggest tank returned to the top ballance/full state.
Good video showing the pros and cons of an active balancer.
Thank you.
Thanks for this lecture. according to this clip: 5:34 - 5:47 . Once the bms turns off, i expect the active balancer to have the time to balance the cells without interference
Thanks Andy for all your work. I learn a lot from your quest to knowledge. For me the best way is NOT TO NEED TO BALANCE. To minimize unbalance I prefer to use multiple banks with their own BMS. Unbalance is created due to the difference in internal resistance on each cell and is increased due to high current. It is not the same to take out 200 amps from a 400 Ah battery bank than take out 25 amps from four 100 Ah ones. Personally I think that BMS are simply not created for those big banks. The current capacity of BMS had increased on the past years, but balance current on them had remain the same. You will notice that the unbalance will be less when you add the other battery banks on your system.
This will be a good topic for a video.
Great feedback and comment. That is pretty much the case, yes. The Daly BMS has only 25-30mA, so what can you achieve with that for such a pack, unless you charge really high and stop charging to give the BMS time.
I understand your point but 4x25 = 100 so it literally is not the same as 200 amps
Servus,
0:30 "...austria..." ???
When did you move and when can I come over to visit you? 😁😋
Also heard Austria. :-)
Caption also render it as "Austria". Slip of the tongue I guess. Pretty close to Australia! I am pretty sure his accent is from Austria,
Definitely said Austria. What I could understand was how Australia was annexed or conqured without any fuss getting on the news.
Comment #7 about that 😂
I have very much enjoyed watching you discovering that the out of the box profiles are good and that the BMS manufacturers are designing internal balancers that should do the job!
Yes, hahaha. As I said, I like to go off the track for learning and understanding why things are as they are 😏
Hi sir, im not using active balancer for 2 years in my lifepo4 battery and no problem in cell unbalancing😊
I just simply charged my batteries individualy to 3.5 volts and connected them in parallel for 1 week then connected in series with bms,. Till now my batteries have a 20mv difference, and im glad to share my experience with you😊
Thank you. This would mean you have to disassemble your whole battery once in a while to parallel it. That seems like a tremendous job though!
Andy, I love the bird song you get in Austria 🤣🤣🤣
I was wondering if that was a Freudian slip.😂😂😂🇦🇹
Hey Gary, thanks for comment #14 about this. Hahaha, that was on purpose to see how much comments I get 😂😂😂
@@OffGridGarageAustralia 14???? I must try harder 🤣
Thank a lot Andy. I do have to install LiFePO4 in my RV, and nw I'm not sure I'll use the Active Balancer. Depends by the use of the batteries I'll have. I'll keep them under control and let you know. Thank again Andy.
Same here. My 4S HANKZOR active balancer is disconnected for now. But the electronic board has a soldered connection between two pad labeled "ON". I plan to desolder and solder a switch for controlling when to active the balancer. As you noted, balancing (passive BMS or active device) should only occur when charging which would require a constant boost voltage stage where the charging current tapered down to 0A. But as you show, the BMS balancing is fine. The important part is setting up the proper charging profile. Some LI chargers do not have a constant V charging stage.
Yep, I tested this little on/off function in one of my videos. A simple switch would be sufficient and you just turn it on form time to time when the battery is a high SOC and let it run for a while.
That is the reason why the balancer build in a proper BMS will only balance during charging and not during discharge as doing so is counterproductive as you found out.
So if you can get a signal when battery is charged and can use that to enable the balancer then all will work great.
The lowest capacity cells also usually have the highest internal resistance and those cells will have higher voltage during charge but lower voltage during discharge thus a balancer that has no idea if battery is charging or not will just waste energy by unbalancing then trying to balance your pack.
Thank you Dacian. What do think about common port bms?
So what your saying is BMS's have top balance circuits and that is the correct way to balance lithium packs. His active balancer does have an on/off switch, but I don't know how that will help? Probably just throw away the active balancer I guess?
@@BischesseHunting Bad idea. That is why mine has separate paths for charge and discharge.
You do not want to absorb and or float charge a Lithium battery if you want long life and so as soon as battery is charged the charging should fully stop and not affect discharging.
@@uhjyuff2095 Yes top balance is what anyone in energy storage industry uses (DIY energy storage will not be counted).
Maybe that ON/OFF switch can be used but he will also need a signal from the BMS or whatever measures the battery current that enables balancing when current flows in to battery (charging).
I set mine to start cell balancing only when battery charge current is above 300mA.
Very small cell balancing currents are needed to maintain a battery in balance most EV's with huge 40 to 100kWh batteries have cell balancing currents below 100mA.
With LiFePO4 voltage at the end of charge rises super fast it can take just seconds to get from 3.4V to 3.6V so that is not an indication of the state of charge.
You can have a cell at 3.45V and one at 3.55V and delta in state of charge between them can be a very small fraction of a percent thus not an indication of how well the pack is balanced.
My house battery (fully off grid) was never disconnected for the past 4 years or so and it is in perfect balance with just over 100mA of cell balancing maybe a few minutes per day in average. The only time the battery was empty was after first year when I did a capacity test to check degradation (less than 1%) and last year when I did the same test and I may do another test next year when there will be 5 years from when it was installed.
I had the balancer connected for ~3 weeks now and could see the battery struggling when it's fully charged. Lots of cells were peaking, even the BMS disconnected the battery twice due one cell reaching the cut-off voltage.
You could have the balancer kicking in at say 3.5V to help balancing at this point. There are circuits out there which can do that through a relay. But do you really need it?
Yep i just came to the same conclusion. I've spend near $400 on those deligreen active balance modules, and have noticed the wild out of balance variance during absorbtion. Going back to top balancers again I think.
That's my experience too. Unless I can turn the active balancer only above a certain voltage.
@@OffGridGarageAustralia I plan to wire in a switch (as discovered in your other video), then balance manually if needed. I *think* it should remove the need to disassemble the pack for maintenance.
@@OffGridGarageAustralia after a year of use, what about the jk bms? does it give you the same issues with active balancing?
Excellent Andy.!
When you will fix the relay on your previous BMS, you can also modify it, tapping on all the resistors are burning the extra charge. If the BMS switch the resistors on the low side you need a 16 input AND gate and an extra NOT gate. Connect the gates inputs on the low side of the resistors. If the BMS switch on the high side you need a 16 input OR gate. The output of the gate can drive an optoisolated relay (or a MOSFET) that connects your active balancer. You will need a transistor and a diode if the relay is not optoisolated.
Doing this, when just single cell is getting balanced by the BMS the active balancer will kick in. Not before and not below the voltage you have set in the BMS balancing voltage. If everything works you can replace the low resistance burning resistors with high values, so you will not waste any power and all the balancing will be done actively.
Basically you will have a BMS with smart active balancing. I hope I was clear. If you are interested I can explain with a deeper level of detail.
That is a great idea! The active balancer would just follow the settings of the BMS.
@@OffGridGarageAustralia in the while I thought of a even simpler implementation. input side of 16 optocouplers with a current limiting resistor (if not integrated) replacing the load resistors (basically every optocouplers will have his own virtual ground). All the output collectors should be connected to B+ (or a intermediate cell to not exceed the optocoupler max Vce) via a pull up resistor. All the emitters to ground. The common collector can drive a MOSFET will drive the relay (on the low side), to connect the active balancer. A freewheeling diode is needed if the relay is not optoisolated. The MOSFET could also directly drive the active balancer, but I guess it's better to switch on the high side (I don't know exactly how the active balancer chip get the power from)
This implementation will make it low-high side balancer switching BMS transparent, reduce the BOM, and most importantly...safer. A shorted transistor/gate chip could lead to a disaster.
If the BMS already uses optocouplers to turn on the balancing resistors, only 16 higher value resistors (to replace the load resistors) and a MOSFET and/or the relay are needed.
I'm yellow with envy for your 50A off grid setup.
My 20A should be plenty to charge my mower batteries for some afternoon mowing occasionally. I do have some crazy plans for running a small coldroom in the future..
Thank you Mr.Andy.if been wondering about these balancers.i have a small 24v setup in camper. my cells do the same as yours one runaway while one charge is still low. Balanced pack hoping for better results
If you can turn the balancer on and off with a switch, that's a pretty good solution and you can do that from time to time to top balance the pack.
@@OffGridGarageAustralia I have a 4s balancer like yours. There is a solder blob on the "Activate" pads, just replace it with a switch and Bob's Your Uncle!
@@mysterytechknowledge3664 I know, I did this in the first video about the balancer 😊
@@OffGridGarageAustralia Then you know that you could leave it connected and just switch it off and only use it when needed, or maybe there's an option on your inverter/charge controller to use a "External Balancer" and have it switch it on and off at the appropriate times?
Also have you turned it off when it is trying to balance to see what the ACTUAL CELL VOLTAGE IMBALANCES are?
@@mysterytechknowledge3664 I can see the actual cell imbalances under load on the BMS app. I showed this in one video, it was like 70mV at 80A. If the balancer pushes only 1A for two hours, that's enough to create chaos for when the battery is being recharged.
Andy, if you are getting large cell deviations during discharge, you may not have well matched cells. Have you checked the internal resistance of the cells? There is a formula to do that is you don't have a cell monitor that can test it.
Also, I have one of those balancers (4s) that I use with my Headway batteries. Something that I noticed is that when charging the pack and monitoring the cell voltages, the balancer will "artificially" increase the cell voltages that are low (and vice-versa) when it is actively balancing. This makes it hard to accurately monitor voltages unless you turn off the balancer. In my case it's easy for me to unplug the single 4s cable. However, you may want to solder a switch and some wires onto the balancer. (inplace of the solder blob that on the "activate" connection.) It would be Very interesting to see what the ACTUAL voltages of your batteries are when charging and discharging by turning on and off the balancer with a switch!
This is why a BMS with active balancing built in usually allows you to set the start voltage when it starts/stops balancing or allows you to only balance on charging.
Hi, same here, since you/he ask, I use the jikong active balancer bms that is on the OffGrid Garage webpage to balance over a set voltage. Works for me and good to selectivly stop charge or discharge. But normaly only needed if the charger or inverter would fail to limit themself normally. Works as overload disconnect aswell that could avoid blowing fuse. Tested when activating when not using the precharge.
@johthe My BMS with active balancer was lost ins space, so I have ordered a new one. Great to hear it is working.
Excellent Video ! I have older Daly nonconfigurable BMS ( kicks in over 3.50 V @ 30mA. I don't use this as I normally charge my battery to a SOC value ( 70 - 80% ) and only resync my Victron BMV every 3-4 weeks - error is only around 0.3Ah/DAy on average so super accurate. ( I set Peukert to 1.00 and Charge Efficiency to 99 % ) I manually rebalance every 2- 3 months or so at 3.425V /cell and have worked out the sensitivity of change in Voltage to change in AH that needs to be applied or removed from the imbalanced cell. This sensitively is 0.0020AH/mV / 100AH Cell. So for a 10 mv imbalance you need to correct by .02AH for a 100 AH cell ( at 3.425 V ) . Which is tiny ! This factor will change depending what your absorption voltage is - I like 3.425V.
Wow, that is great information. Thanks for that. I will be in the same boat and rarely get to 100% SOC with my setup so the smartshunt does not reset and keeps coulomb count. But as you said, that's pretty accurate.
Andy, you need to get a Nissan leaf battery and see how they do it. BMS knows the internal resistance of every cell. I think it takes this into account when measuring balance in charge and discharge!
I really wish that most of the BMSs available either had active balancing built-in or had an aux output for controlling an active balancer. The issue with most BMS units is that they turn off charging before turning on resistive balancing so there is no way to have both on with the given settings so they go back and forth between charging and balancing until the battery is full on most BMSs since the low-current balance can't keep up when charging at several amps. So if you need more balancing then having an active balancer and a voltage-controlled relay board to turn it on and of as needed when charging seems to be the way to go.
I have the option to turn off charge-balancing in the BMS which I have. It does not make sense to balance while there is current coming in.
Eventually, I will just stay in the flat part of the curve and may not have to balance often, maybe 1-2/year.
The moral of the story is (drum roll, please): Stay away from the "knees". Not does operating your cells within a conservative range dramatically extend the longevity (a.k.a. lifecycles) of the pack, but cells tend to remain within a small Vdelta that the BMS balancer is able to remedy...given proper configuration.
To that end, Andy, are you able to post both your BMS configuration and charge controller profile settings? That might prove extremely useful for some.
As for active balancers, I build every pack to be active balancer-ready. That way, if and when things go a bit wonky, I connect the active balancer for a cycle to bring everything back within tolerance automagically, rather than resorting to old school manual balancing. So they do have their place. They're just not a panacea.
Beer drinking must come after editing the video Austria? 3.8 volts?
hahaha, great you picked up on these two. #5
I'm installing one on my ebike battery and in my case I think it will prevent my bms from cutting off because of 1 or 2 cells being .5v low. But this is interesting to see how large constant loads actually confuse the balancer. With my bike the BMS usually cuts off the second I press the throttle at low SOC. My thoughts are the balancer would buffer that. Thus getting more range out of my battery. So i'll install a switch and see what works better, leaving it on, or just top balancing. Also, my battery is li-ion 14s 3p. It's only 12ah at 52v. I have to wonder if this balancer is more effective on smaller capacity batteries. But this is something very few people make good videos about. Thank You for your amazing videos!!👏
Hi, did you end up adding it to your ebike? Im thinking about doing it to mine. How did yours work out?
Love all the test!! Still my Favorite Channel!! Thanks Andy!!💥🤘
Thanks Joe, much appreciated!
I think in order to get those active balancers to work you need to turn them off when you're below a certain voltage so they're not doing anything normally and only really kick in when you're at 99% or so charged
That is exactly correct! But then, do you actually need one at all? As shown in this video, the BMS's balancer was totally capable of balancing the whole battery to 2mV within ~30min
Nice! No knowledge is ever learned without mistakes.
Absolutely correct! Thank you.
Love the intro. I’m going to add a switch to the active balancer to turn it on and off.
From most of what I have read a single series pack at these high voltages will always go out of balance. I run a BMS and active balance because I have an iffy cell.
Yes, that is a good reason to run one.
After having viewed other channels and read some on balancing, I tend to agree that active balancing all of the time is not needed and could be a mistake. I did note that Herman Jordaan installed a switch that he uses to control his active balancer. That seems like a good approach. I did note on another channel that periodic use, say once or twice a year, of an active balancer only on one or two charging cycles may be the ticket. That particular channel complained about how the installed BMS, I think that it was a Daly, was not maintaining balance and active balancing was required. His problems reminded me of the decision process that you used to incorporate an active balancer into your system. I had already viewed this video and directed that person to this particular video to help him with the decision process.
Mistakes are mearly an opportunity to learn faster 👍
Love the channel 😎
Thank you. I love exploring and testing. Much more to come 😊
I think I'll have to remove my battery from the van and do a proper top balance (the cells are now balanced at a lower state of charge, so I'll parallel them and take them to 3.6v). Perhaps then my problems will be solved and I won't need a new BMS or the flawed neeeeeeey active balancer. Thank you for your new vid, I thought you'd use a boost converter and you confirmed my fear of some draw while inactive.
You can leave the battery in place and charge the whole pack slowly so the balancer has time again to do its job at a higher voltage. Once you're at 3.6V or above with every cell, disconnect the balancer and leave it off. Connect it every here and then for maintenance. The BMS balancer should be OK to keep it under control.
@@OffGridGarageAustralia thank you 🙂
Thanks Andy, I won’t waste my money on an active balance or when I build my system!
I wouldn't buy one if it's a new system and you just get started. If you have any problems down the track, you can always add one and check it out. It's easy done....
There is a place where you can add a switch or relay to the board just unsolder the bridge “RUN” on your board and then you can add a switch or trigger mechanism to work with the BMS when the BMS starts to balance and should balance the cells quicker and with less “Burnt” energy loss 👌🏾
I had this bridge unsoldered when I tested the balancer and connected a switch to it. The trigger from the BMS would be interesting...
Thank you for yet another informative video. You also always have just enough goofball humor to keep it entertaining lol.
I have rec bms with 32 lishen 272ah, recbms has 1 amp balance no problems for me, also tampers amps on victron if there is some unbalance in the battery.
Have you connected the BMS to the Victron gear?
@@OffGridGarageAustralia yes i have victron ccgx, bms connecs to ccgx by canbus, on the end of charge the bms lowers max amps to allow bms shunt to bleed the cells. Better than that is that i also have one raspberry pi with home assistant, i made a script to stop charge at any percentage i want, basicaly when bms reports 90% soc i tell ccgx to stop charge.
@@piritech nice! Does this work just out of the box or how is the software being setup? You have to enable DVCC in your GX, I guess?
@@OffGridGarageAustralia Almost out off the box, in ccgx enable DVCC, SVS and STS, in victron inverter i set other canbus BMS, then after that BMS commands everything. You then use bms software to select max charge voltage per cell, max discharge/charge amps etc.
@@piritech aaaah, that's how it works. So you need to set a setting in the SCC as well to listen to the BMS. That part was never clear to me before. Thanks for explaining that.
My set of winston 400ah cells are 8 years old only ever top balanced once when commissioned
Never used a balancer of any sort when charging at the upper end of the charge cycle at 3.5v per cell i see a maximum of 40mv difference and once resting voltage difference is about 4mv
And do you even have that at very high charging rates and high voltage? Do you absorb at 3.5V of just stop charging?
@@OffGridGarageAustralia i charge to 3.5v per cell (14v) hold and absorb until charge current drops to 50% output which is 50amps my charger is 100amps
The bank wouldn't be at 100% soc but at this point internal resistance is building and I see no reason to charge further.
Bms has never had to interfere in 8 years of use
@@kurt6738 great, thanks for sharing.
@@OffGridGarageAustraliacheers and thanks for your all the great content on your chanel 👍
You could have bought a turn key solution like battleborn and that includes matched cells and a BMS but no knowledge transfer, THANK YOU, money tight now, will donate soon!
I think you’ll incorporate the active balancer during the year when you don’t have as much sunshine so your top balancing time will be shorter due to cloud cover. With the voltage triggered relay like you mentioned.
I'll try without for a while now because the cells are now all top balanced. I want to see when they start drifting again without being disturbed by the balancer. Then I can probably develop a 'maintenance schedule'. Eventually, I will stay between 3.1 and 3.45V so I should not se too much drift over time.
It is confusing why the BMS sees these different voltages during the discharge and charge when they are all starting out balanced and the same current is being drawn on them all. I would think since the curves are so flat for this chemistry that they would all stay very close during the "middle" of the SOC. I do agree with your explanation of why the balancer is not helpful during these times. Thanks for all your experimenting...we're all learning a lot!
Thank for your feedback. The balancing at lower SOC does only occur when I pull higher loads and the voltage spreads out over a longer time. This lets the balancer kick in and it starts its job. Can't blame it for that 😏
This is how i think it works...
There are slight differences in the capacities of each cell due to manufacturing tolerances.
This is like having a series of tanks of slightly different widths.
When you top balance... You fill all tanks to the same level regardless of their size.
Since they are all in series, as you discharge, the exact same number of coulombs are taken from each cell. In this tank model, this is like taking the exact same volume of water out of each tank.
The smaller tanks end up having a lower water level than the bigger tanks.
As you recharge, the exact same amount goes into each cell.
The level in the smaller tanks rises faster than the level in the bigger tanks.
They all arrive at the same level once more when they return to the original, top ballanced/full level.
But if you had rebalanced the level in the tanks when they were at their lowest point, then refilled them... The smallest tank would still fill faster than the biggest one, and it would become overfilled before the biggest tank returned to the top ballance/full state.
All these tests and your recommendations led me to buy 2 active balancers. I am on a sail boat and because we are often away from shore connection, we will rarely achieve full charge/SOC. I should have listened to @WillProwse who said that active balancers are a waste of money and don't work. 😒
only connect it when you need it (especially during higher state of charge), and you should be fine. Alternatively if you are stuck at sea with no sun for long and BMS hits low cell cutout due to weak cell you vould connect it and it will shuffle power to allow BMS to turn back on a bit longer.
@@jamesbsa6450 Thanks James
@Ralph They are not a waste of money. Connect them and have a switch to turn them on or off. If you fully charge your batteries once in a while, turn the balancer on and they will do a great job.
I've been using this different kind of balancer that basically is a line of resistors for each cell that turns on when the cells reach whatever voltage it is and just applies that resistive load to that one cell
So the one I have is for sales and it turns on each cell at 3.44 volt so whichever cell reaches that first gets the load applied first and I stopped charging once all four cells have the load applied to them once the voltage dips below 3.44 all the load is removed
I don't know if they have one for a 16s or not
Also it would be nice if you could program it and change whatever voltage the resistive load is applied at that would be better
I did have to go through three different models of this similar product I had one that wouldn't start the load until about 3.66 volt that's totally unacceptable and then another one was it like 3.6v pretty good but I didn't really want to be charging that high
Any link to this balancer please?
The top-balancing the BMS does should be sufficient (I've said that before). I really don't like active balancers (I've said that before too)... At worst they mask a bad cell to the point where that bad cell might cause a fire, leak electrolyte, or otherwise damage the rest of the pack. The solution to having a bad cell (as long as you don't have too many bad cells) is to replace it with a new one, simple as that.
When throwing new cells into an existing battery pack, the new cell's full energy storage won't be used but it will enable the full energy storage of the remaining cells to be used. Its the easiest solution. Basically everything will be top-balanced, so the new cell won't be drawn down quite as much as the other cells during discharge. It may seem to go 'out of balance' in that respect, but it really isn't... it will re-charge back to approximately the same properly top-balanced voltage.
One other facet of this sort of repair is that you might have to adjust the low-voltage shutoff for the whole pack to be a bit higher to avoid constantly hitting individual cell low-voltage cut-offs. The new cells will raise the whole-pack voltage a bit at the point where the rest of the pack is fully discharged. So just keep that in mind. Its less stress on the pack.
It would be quite interesting if you did a test like that, actually. Throw in a brand new top-balanced cell and see what happens during the discharge and recharge cycle.
-Matt
Thanks Matt, great advice and comment. I have 16 new cells here so... but I don't know which is the 'faulty' cell in my pack (if there is one). After the top balancing and removing of the balancer, it is a different cell every time which peaks out a bit when charging to higher voltage. I think this is good.
I'm still intrigued with the battery shelf build. Are you waiting on parts to finish that up?
It'll be in the next video 😒
If I could, I would say to myself, "Welcome to the club."After doing top balancing as per the book on the 32 302 Ah cells, I build the battery bank, I connected the two active balancers to the two series of batteries connected in parallel,and after a few days , the gap between them reached 120 mV,at 3.47 volts at as I charge the cells,55.5 volts for entire bank..This gap increased progressively over 5-6 days.The balancing option has been permanently activated on the two active balancers. They are devices with bluetooth and Android applications. After watching your video I realized that you are right and that balancing at low SOC is not okay, and maybe it's not okay even at high SOC. So now I disassembled the battery bank, I do top balancing again and I will try again without using the balancing option from the balancers.Or should to enable the option only when the chargers switch to absorption mode, and disable it when they switch to float mode?
How has your battery bank evolved, and the gap between cells been kept so low?
Very interesting video as usual, thanks Andy, ya legend!
Glad you enjoyed it Thank you😁
Great work, great research 😎
Thanks!
Loved the intro. Andy should invite us in in all the future video's.
Yeah, come over and knock 😁
Thanks sir now i understand. Dont need active balancer for my 280ah 16s3p. My daly 300A BMS is more acurate.
Connect the balancer to the relay on the Victron charger (if you have a model that has the relay terminals) and set it to switch on when in float mode, this way you will only balance when the batteries are fully charged and it will switch off once you go back into Bulk mode.
My SCC have only the Virtual Load settings which feed back to the Raspberry and it can trigger a relay somehow I think...
But again, do we need the active balancer as it seemed to work fine without. Let's see what happens in a few days when I fully charge again and the battery has been used...
@@OffGridGarageAustralia My system struggles to stay balanced without the active balancer, it end up over time about 300mV delta without and 10mV delta when used in Float mode only. For me I need the balancer.
The question is why you get such a high voltage deviation on discharge. I would assume that the cells are pretty equal at this low cycle count. So as mentioned below, probably the connections are not all equally good. This should be fixed anyway, as this may worsen over time, until you get hot poles, heating the cells and causing more degradation at this cells. Unbalanced cells are just a side effect.
At the moment the top balancing of the BMS seems to be good enough to handle this.
Another problem seems to be that this balancers start balancing at only few mV, instead of having a higher start level. The ones I use start at 10 mV.
Of course, in your environment with very high discharge over several hours, the imbalance introduced is increased. In setups that use lower C-rates you will probably not notice the problem.
Having a possibility to turn off active balancing would be helpful sometimes.
Andy, Do you think if the internal resistances were better matched, the cells would perform more uniformly??
thanks, dave
The Ri is actually pretty matched across the cells. But when I discharge the battery with 80A, that creates the unbalance due to small mismatches. There is also different resistance across terminal-busbar connection, cable connections and more. So this all makes the cells spread over time. Of course if they would have 100% the same Ri, it would be less, but this is only one part of the problem.
In Austria. @2:31. Kleiner Versprecher 😂 Grüße aus Kanada 🇨🇦
hahahaa I caught that also. Yes I checked it 3X
Comment #6
(that was on purpose though 😂)
@@OffGridGarageAustralia Yes we are paying attention
oh, you're in Austria ..I didn't even know that this country is so big that it stretches across several time zones (always thought that with the end of the 'Danube Monarchy' the former empire would be shrunk to a manageable size ..stupid me). Presumably a small, charming, "Freudian slip of the tongue" crept in during your introduction because as an Austrian you can only dream of the many hours of sunshine that are likely to be the norm in your area) .)
'love the scene at 22:16 (terrified scream with half-second time delay ...well, that's how we humans are). I myself have my own problems and doubts when using my "Daly" BMS and with balancing. Glad that I saw your video in time because I was already tempted to get an active balancer. Maybe I just should be a little more tolerant with my slowly working BMS ... thanks for sharing your successes and faults with us ..informative and funny.
LOL Shame on you Andy for taking advantage of the beautiful Mermaid! I laughed at that one! And I imagine she will too. I have two wish list items for you if you will. One i would like a screenshot of your controller voltage settings. I have a similar system as yours (my system is 24V 40kw). I know you discuss the voltage settings but a screen shot I can absorbe better. Second on the wish list is the DVCC setup for a Cerbo GX. Thank you for your in-depth videos!
I will make a whole series of videos here on the channel soon explaining the settings of all my devices. Quite a few people have asked for that. I have also planned to add these info to my website, so it can be easily found.
So patience, young Padawan, it will come soon...
Unlce Andy, thanks for another informative video.
Probably not logical but would parallel of say 3 or 5 of these active balancers not work on solving balancing issue at any SOC,
PS when you building that battery shelf further.
Thanks, Wayne scheepers
Glad I'm not the only one with a DC toy box *lol please label it as such * regards.
Thanks Wayne. I don't think this is the way of doing it. It's not about transferring more and more amps. See my pack is now perfectly top balanced and the balancer of the BMS with 150mA is capable of holding it in balance at 3.55V. The balancer just created a mess as it balanced all the time. Imagine having a few balancers doing that, the mess mess would be even larger.
@@OffGridGarageAustralia very true, I got one because some weeks it get cloudy and the SOC gradually drops, and its just that 1 cell that gets to 2.6v and then boom house goes offline.
@@OffGridGarageAustralia Yes, the balancer works all the time, but obviously are the 5 amps of the active balancer not enough to keep them balanced at any state. It would be interesting to see what happens with more balancing power...But I agree it would be better to run the battery by itself without much intervention.
Had the same thoughts about active balancer un balancing when in a lower soc. People kept saying no they work great etc. purchased two of them. still sitting on the shelf while i sit on the fence. I think i am coming to learn that my initial thoughts may be true.
I think, it depends on the setup people have. If they have shallow DoD and have the battery mainly running from 70-100% it may work for them actually. Also, if the have only low discharge currents which does not cause a big voltage deviation, it will work too as the balance does not cause any trouble then.
It's really not an easy decision. If you constantly run into mismatched cell voltages when the battery is full, I would probably try them out. If the deviation is not too big, leave it as it is.
Very interesting, i just received my active lifepo4 cell balancer. So how many kW a day do you get on average.?
0:30 No kangaroos in Austria, we are on the the other side of you.
Comment #9 😂
(just me being silly...)
I charge my battery up to 54.4V (3.4V / cell) with an active balancer like yours, my cells normally have a difference of 0.010V. I have my BMS configured so that it cuts the load at 3.55V and activates it when it returns to 3.40V, without the balancer my battery does not rise above 54.0V, since the BMS cuts the load due to overvoltage of a cell (normally the 6). The internal balancer of the BMS (280mAh) was not able to balance my cells and I was having trouble using an inverter soldering iron.
Since I installed the Active Balancer, my problems are over, I also don't have your problem that the BMS cuts due to overvoltage, since with my configuration the cells stay within the normal range programmed in the BMS.
Talking to the seller of my cells (who is a manufacturer), he told me that it is very difficult to control the loading ramp, once they pass 3.35V; In their capacity tests they do not take the cells to the limit, they advise working in the 80-90% load range and this is achieved with 3.35V; They think that raising the voltages higher is unnecessary, due to the difficulty of controlling the cells, for the little benefit of charging that you get.
For all this, and watching your videos (which are very educational), I configure my system as follows.
Hybrid inverter:
absorption charge: 54.4V
float charge: 54.3V
Gross charge time: 120 minutes
BMS:
overvoltage protection: 3.55V
overvoltage protection disconnection: 3.40V
The balancer always on.
Yes its happened to me .I use 4s active balancer .all my batter all gone. Batter use one time only
Glad, I'm not the only one... 😁
Very good experiments. Please continue your tests. Have you measured the [mV] between busbar and pole during high amp discharge? If this is high (like 50mV for some cells) the active balancer will transfer Ah between cells only because of bad connection. In my setup I have this active balancer on all the time, I do regular 60A charge, -50A discharge for several months - overall I see no big problem - at end of charge sometimes one cell goes to 3.46V or so, but then it went down. One big difference from my setup to your is that I have set the floating voltage much lower, its about 3.4V per cell.
Funny, 0:30 welcome to the off grid garage Austria... No wonder people always mix up Australia with Austria.. ;) :) :)
Comment #17 about that. I keep counting 😂
Can you make a switch on + or - wire to turn on or off the Active balancer without pull the whole WIREs off ? thanks
Thanks for this Andy very informative. I've been wrestling with the run away voltage on a simple 24V 8s 50A system using a Daly BMS and no active balancer. It has been driving me nuts when one cell quickly gets to maximum voltage and leaves the others waay behind.
I don't think the problem is really different battery chemistry as much as these Smart BMS systems are really quite stupid. It just doesn't seem that difficult to take into account different battery chemistry and the typical voltage/capacity function when charging.
So as you mentioned it primarily functions as a battery protection switch as using it to balance means endless charging/discharging cycles. Life is too short not to mention wasting electricity and contributing to global warming. I brought a petrol generator for the sake of the planet, humanity and my sanity.
Hello Andy!Can you tell us how the situation evolved with the battery bank , without active balancer?
Thumbs up and subscribed! This is critical info!
Is it possible that the active balancer and the bms balancers ,are fighting each other?
No, they work in the same direction. They both deplete the cell with the highest voltage at the same time. You can add as many balancers as you like, they will all work together.
My thought is maybe you only turn on the active balancer during charging (at a minimum voltage level above ~3/4 full)
I recall that the active balancer had a hidden bridge where it could be disabled or enabled, so I wonder if with a relay output from the inverter or charge controller could be used to enable it? Or use a voltage controlled relay so that it only active balances above a certain pack voltage or when the battery is in float?
Yes, the balancer can be turned on and off either via external manual switch or a relay from a voltage triggered source.
Yeah, the relay out on a BMV could be used to only turn on the active balancer at a certain voltage of SOC.
So it would be wise to switch on the actvice balancer only in adsorption mode. This should be possible if the brigde on the run switch is removed and a little relay is placed there, which is activated by the yellow adsorption LED of the solar charger. Or it could just be voltage triggered.
Yes, that would be ideal. But then, do you need it actually? As my test showed then BMS small balance current was enough to get to 1mV...
@@OffGridGarageAustralia Right, a the current state you don't need it. But you have made a top balancing just before. The voltage deviation might increase during time and the balancing current of the BMS is very little. So at a higher deviation it might not be able to finish the balancing before the battery is discharged again. Well, in this case you can switch on the active balancer again.
And if you want it to be controlled by the yellow LED, you will just need a few resistors, a capacitor, a foto-transistor in a short black tube that fits on the yellow LED, a zener diode and a little (reed) relay. But I think it would be easier to make just a voltage triggered version, which could be placed on top of the active balancer.
Good morning brother
Good morning Australia 🇦🇺
From Philippines 🇵🇭
Good Morning to you too (one day later) 😁
Yes, I've think you've identified the problem.
It seems so, that last 2 day, no issues any more...
you can install 2 active balancer parallel. so you balance with 10 ah
You can, but then you transfer even more amps at a lower SOC as well so this won't fix the problem at all. You would need to make sure, the balancer only turns on at higher voltage so it does not mess up things at lower voltages
Nicely explained....
u make a point...
Excellent ....
i have to get bms...👍🍺
You would think the active balancer would be able to adjust the turn on point like the passive balancer in the bms. If a bms had active balancer wouldnt it be much better??
Thanks for your videos! I'm looking to parallel 3 or 4 cells before the series connection. And would love to know if this configuration also helps to keep cells in balance.
I have a 4s battery and saw the same. Active balancer helped a lot getting the cells in balance but after a few cycles one of the cells started reaching 3.65 and resetting the bms. So after it balanced the battery at 3.55v I disconnected the active balancer and also lowered the boost voltage on the charge controller to 3.475v-3.5v per cell and maintain ther for one hour, then float at around 3.375v and now at the top of the curve I see no more than 20mV and I am able to fully charge the battery.
I think I will use the active balancer to perform periodical maintance/balance without the need to disassemble the pack and put the cells in parallel.
I'm offgrid and use simple top balancing boards at 3.6V. This seems to keep things within 10mV or so across my entire string of 16x200AH - I measure periodically with my 5/5 digit HP meter. I've held off getting an active balancer so far because it all works well as things are.
I do wonder about the accuracy of the active balancer readout - component variations could give significant error on the reported values, even 0.1% component spread will give reading errors of 3.6mV from one cell to another and I doubt that a cheap balancer uses such precise components. Have you compared the actual cell voltages with the reported values?
Thanks Andy. I have not compared the voltages yet. It is hard to determine which voltmeter to believe unless you have calibrated one correctly and know it's the point of truth. If they keep everything within 10mV that's all we need, I guess.
@@OffGridGarageAustralia I guess the voltmeter accuracy isn't critical provided all the batteries read the same value.
I don't trust the BMS to report all the cell voltages with the same error because of component spread.
Hallo Andy,
ich denke, dass da immer noch ein Widerstand zwischen den Messleitungen und den Polen ist. War bei mir so , aber auch nur bei einer V2A Unterlegscheibe.
Dadurch kommt es zu einem Spannungsabfall an dem Widerstand (Unterlegscheibe ) und zu einem Messfehler .
Bei meinen Cellen hatte ich bei einem Ladestrom von nur 10- 20 A fast 500 mV Spannungsunterschied, der aber nur ein Messfehler war, da durch den Balancerstrom
an den gut 20 Ohm Überganswiderstand diese Spannung abfiel.
Seit ich die Anschlüße der BMS und Balancer Leitungen unter die Unterlegscheiben direkt auf die Busbars gepackt habe ist dieses Problem weg.
Mess doch bitte mal jede der 17 (34) Ringösen gegen die jeweilige Brücke ( Pol ) , ob sich da irgendwo ein Übergangswiderstand eingeschlichen hat.
Grüße aus dem Ruhrpott Guido
Hallo Guido, danke fuer deine Nachricht und Tip. Bei mir sind die Ringoesen seit Installation direkt mit der Sammelschiene verbunden, die Unterlegscheibe ist dann oben drauf. Aber sicherlich hast du generell Recht, Ich habe immer noch ein Widerstandsproblem und Uebergangswiderstaende die mir falsche Messdaten uebermitteln. Sobald die zweite Batterie eingebaut ist, kann ich diese hier auseinanderbauen und mal richtig neu verdrahten, Kontakte reinigen und alles besser machen, was im Moment noch nicht so gut ist...
do you think maybe having the balancer com on at night time to balance when they is no power coming in
It depends on the SOC. I can really see that it balances the cells at any time and when I charge back up it's all over the place again. Now, without the balancer, they stay inline.
If you balance them at night and they are not full, there is no point to balance in my eyes...
@@OffGridGarageAustralia yeah that is a point it really shows from all the different settings you have done that you cant use none adjustable bms and just forget
@@1981dasimpson That's right. The system always needs a bit of tweaking here and there over time.
3.645978 rounded is 3.6? You say 3.8, am i missing something? Love the channel.
Ah, that was just a joke in regards to the last video where I was supposed to say 3.38 and said 3.8V all the time. People commented on that... 😏
The 3.64978V are also not really true. It was mmore like 3.64856V actually 😂
Hi Andy you should be saying how many watts of power from the pannels not amps
Alan, this channel is all about amps 😂
@@OffGridGarageAustralia but amps doesn't show the amount of power being created (Amps multiplied by Volts equals Watts, which is the measurement used to determine the amount of energy.)
What happened to your unplugged ev Chanel? You left us on a cliff hanger lol
Not enough hours per day, I guess... It'll come soon ;)
Hi Andy
been thinking about this for a while
your experiment does not simulate the real situation of an offgrid house.
Perhaps you can simulate the experiment with a normal consumption of a house .
if you go partially offgrid with a residential there are always irregular consumers. a refrigerator, freezer, computers on, off, heating the water, coffee maker, etc.
And that all happens in the belanscueve!
What would be the end result of the charging cycles in a few days?
That's exactly right and a very good suggestion. I kept an eye on my consumption and made sure I can recharge every day to see the result of balancing, absorption, floating and fine tune that. Now, I have the pump running and also charge the vehicle again from the battery so won't be fully charging for days. It will be interesting to see what happens next weekend when I have planned to fully charge again.
I think the use case is actually more clear than you indicate. The amps capable of being pushed around in the balancer should be equal to the draw on a typical day or input on a typical day divided by the amount of paralleled batteries. 50 amps pushing at 48 volts with no paralleled cells means you would want to be able to push 50 amps around at the worst case. Add your second battery paralleled and you are down to 25 amps. Parallel two of the 5 amp balancer units and they are dealing with 12.5 amps at worst. Split this over the two hours of rest time and it is 6.25 amps per hour. You start to get really close to having protection in the worst case scenarios. As to whether you wish to build for the worst case is a personal question.
With new cells like you have you probably could come up with some calculation for the acceptable level of risk and return.
Again, the above is dealing with charging, but with discharging you most likely don't pull a 1C pull constantly. You can mix the math above with a discharge typical load.
I hesitate to offer a "rule of thumb" response and when asked I say get the active balancer and get a BMS. The worst that can happen is that the BMS disconnects the battery at the low voltage turn off if your video's described situation occurs and at best it will make sure that the batteries age equally with regards to high and low voltage cycles. After all, if your battery is already kicked overvoltage relatively new you can make sure to limit dendrite formations by avoiding the higher voltages completely instead of just reacting when it hits a high voltage.
Thank you for all you are doing. This is a journey that you are bringing a lot of folks with you. There is probably a unified theory of best practices, but right now it is hard to really identify it.
Thank you.
Yes, that is correct what you're saying. I think the best practise is to keep the battery in the flat area of the curve, so 3.1-3.45V. If you need more capacity, get a second battery.
In terms of the balancer. I don't like the fact that it balances all the time, charging discharging, high/low amps in and out. It just causes trouble over time. That's just not how a battery should work. But as I said, I'm experimenting and exploring all these different situations right now to get a better understanding why some settings work better than others.