Parallel 48V battery banks high voltage/SOC test. Passive vs active cell balancing!

Two 48V battery banks ins parallel.... two different BMS... and today we will fully charge them and observe how the balancing works in the QUCC and JK-BMS. I'm also curious to find out what happens when we hit the 3.45V, which is my normal absorption voltage.
We have the QUCC BMS with is passive 0.2A balancer connected to one battery and the JK-BMS with its active 2A balancer on the other battery bank. Will the QUCC BMS disconnect our battery again prematurely and can the JK-BMS's active balancer keep the cells under control when we fully charge for the very first time?
The QUCC and JK-BMS:
off-grid-garag...
EVE LF280 battery cells from reliable sources:
off-grid-garag...
Finding 20% and 80% SOC with LiFePO4. Why charging current and charging speed matters!
• Finding 20% and 80% SO...
Please subscribe and join me on my journey of the Off-Grid Garage!
For more information, please check out my website with links to all the batteries, materials, devices and products I have used in my videos: off-grid-garag...
The 'buy me a beer 🍻' link is here (on the website, scroll all the way down):
off-grid-garag... Thank you for your support!
You can also click on the Thanks button under the video to support the channel. Thank you, if you have already!
T-Shirts, caps, mugs and more in our merch-shop:
off-grid-garag...
**** These are the batteries you are looking for ****
🔋➡EVE LF280/304 LiFePO4 cells 3.2V/280+304Ah: off-grid-garag...
🌞➡Solar charge Controllers, Inverters and accessories: off-grid-garag...
🔧➡Tools, Analysers and Testers: off-grid-garag...
📈➡ Battery Tests and Results: off-grid-garag...
🔌➡Batteries, charger and accessories: off-grid-garag...
⚡➡Cables, Inverters and Electronics: off-grid-garag...
Get 5% off when buying Bluetti all-in-one Power Storage Products!
off-grid-garag...
Great "Next Gen" AOLithium Batteries (4000 cycles, Bluetooth, CALB cells)
aolithium.com/...
New to AliExpress? Use my referral code below to register and earn US$19 store credit. What a great start! a.aliexpress.c...
Take a live sneak peak at the Off-Grid-Garage in the Victron VRM World:
vrm.victronene...
My Tesla referral code: ts.la/stefan96963
My other KZbin channels:
/ offgridgarageaustralia

[@andrewradford3953]

That zoom in when you were rubbing your neck.. Just Gold! Thanks for another great episode.

[@bboyda4399]

Hey Andy, you do know that the Tesla is part of your storage capacity and should be included as such, so don't turn off car charging, it probably should be priority.

[@CollinBaillie]

When the big batteries are full, the Tesla could charge to take up the excess produced. As could an electric water heater and others. Energy lost is energy wasted. I like your thinking. I think Andy turned everything off for the purposes of running the tests on the packs as they approached full charge. Normally he doesn't get that chance because he's constantly discharging the system, so this was a special case, for science!

[@felaxchow207]

One good thing for us solar folks is that we really don't need the cells to be charged up to 3.65 volts / 100% state of charge like the good old lead acid battery days. The roundtrip efficiency is almost 1:1 that on a good sunny day, we can absorb at 3.45 volts per cell and it doesn't really care. Another great thing about lithium battery technology is that leaving it at a partial state of charge does not shorten its cycle life and in some cases it helps increase it. What would be an interesting test is once the bank is brought to 100% state of charge and balanced. Disable the balancing function and let it cycle for about a month or so, then bring it back to 100% again to see how far the cells drifted over that time. With all of the testing going on with your system, I have a feeling this one is in the bucket list.

[@OffGridGarageAustralia]

Thanks Felax, I have done such a test in the past with the 'old' battery and measured the drift I got without any balancing.

[@upnorthandpersonal]

This is one of the reasons you really want independent charge and discharge control: when you have a cell overvoltage situation, you want the battery to be able to discharge over a load to recover from that situation, while not taking any more charge current.

[@sharyncomstock7066]

Thanks Fred north east nh.

[@sharyncomstock7066]

I have enjoyed your knowledge in the comments. I have a problem with vectron bat. Pretect. On my 24 volt lithium bank. It's not disconnecting at the low battery setting about 12.2? Also on wells site some one stated having a appliance with a capacitor effects it's operation. Would greatly appreciate your thoughts on these two issues.thanks Fred north east nh.

[@upnorthandpersonal]

@@sharyncomstock7066 Do you have the battery protect on the DC line going to the inverter? The battery protect is not intended to be used to disconnect the main DC line of an inverter, see point 4 in the manual "4. The short circuit protection of the SBP will be activated if you try to directly connect loads with capacitors on their input (eg inverters). For that use case, please use the SBP to control the remote on/off switch on the inverter, instead of disconnecting the higher power DC line.". Personally, I don't use the battery protect, never have, It's probably the most useless item Victron has since its only fro DC loads - e.g. in a van or something with DC appliances only.

[@Sylvan_dB]

This is why you must top balance. The passive balancers take a LONG time to balance. They are intended for maint only. If top balanced then when fully charged the cells match and can be maintained. The full pack has the capacity of the smallest cell, because it will reach lv disconnect first. If that is drastically different from the other cells so that they are still flat part of discharge curve, your pack is not going to be happy because the inverter, etc. cannot detect low state of charge before the lv disconnect triggers on the weak cell. You can actually parallel a small cell with the weak cell to mitigate the situation but best to replace the weak cell.

[@OffGridGarageAustralia]

I top balanced my pack a hundred times (almost). The problem here is that I rarely fully charge my pack and it sits at around 40-70% most of the time. If I get the chance of a full charge every couple of months, The cells are all over the place. Now, this battery is the result of 3 different orders, so the cells may not be well matched. But again, if they are top balanced you would expect they will still be top balanced after a few weeks once fully charged again.

[@jackoneil3933]

Thanks Andy, remarkable System and very informative segment. The results bring me back to the notion of intermittently using an active balancer. Also, as FYI I recently received a couple of new defective LFE Prismatic cells that were under rated capacity and slowly self-discharging, and while testing them I inadvertently overcharged them to 4.7v using about 2amps, for several hours trying to re-balance them using a 5V power-supply. They got to about 38c and quickly returned to 3.7v after about an hour of removing the charge. I noticed no swelling or obvious symptoms, and after the overcharge I cycled them about three times and found they went from being about 20% below rated capacity to about 8% below capacity. As such, it appears that LFE cells are indeed more tolerant then Li-On that I suspect might have burned is so overcharged. However, if I had hit those cells with say 20amps rather than 2amps I suspect the result might have been different.

[@marcobrian1619]

As always, another great video of of how the different BMS's work and how effective the "active" BMS is. Nice to see you still playing to make your system even better:)

[@kostric7]

Congrats on 25,000 subs I'm currently pumping down spats in ur honor

[@ttttenney]

It'd be interesting for you to switch the BMS of both battery packs and see how they perform.

[@zxrjimmy1]

Thanks for showing the qucc antics, safe to say it will work better if voltage is set lower, like 3.5 per cell

[@northernboy3023]

The JK BMS are great. I've been running mine on my 2p16s battery made from REPT 280ahr cells since October 2021. I absorb for 15mins at 56.4v and float at 54.4v, balancing starts at 3.39v and 0.007v cell div. I didn't top balance just assembled the cells as they came from the supplier. The JK took about a week to balance the battery and it has been balanced since then only going to the most a 0.040v cell div when it first hits the top of the charge curve and quickly gets back to the minimum cell div of 0.007 by the time the absorb is done. I would definitely recommend the JK BMS to anyone looking for a good affordable BMS .

[@SkypowerwithKarl]

Turn your float down to 53.6 (or less) for better battery life. The very fact that you have the balancing power of the JK negates the need to hold it in float to balance. 54.4(edited, thanks) volts is the absolute maximum float that you’d run if you need to keep a balancer on when you have a BMS that has balancing current only in the millivolts. Let your system balance and go to work or relax. I’m betting your delta is.005 or less on the long slope.

[@northernboy3023]

@@SkypowerwithKarl I don't float at 56.4, that is my absorb voltage and it only goes for 15mins. After it drops to 54.4 float voltage . I got those settings from Will Prowse's post on his forum.

[@SkypowerwithKarl]

@@northernboy3023 Yeah, my bad, 54.4 is the absolute maximum is what meant to say. 56.2 is where I actually max/absorb. I did find where Will recommended float at 55 but from what I’ve been reading that’s high. At 53.6 float, your at 99% of capacity.

[@chevrofreak]

I have the new JKBMS 200a unit with the screen and I am super impressed with how good of a BMS it is.

[@showmequick2245]

Where did you order it from please. Debating JK or Seplos thx

[@edwardvanhazendonk]

Thanks, JK is the way to go with these high capacity cells, great find!

[@chuxxsss]

It looks like the JK active balancer wins the race for this one. Have a great day, Andy.

[@daveduncan2748]

Are you really running the whole show just from the JKBMS battery (20:43)? It looks like QUCC is still on for "Discharging" according to the display. And I would guess that's not just for balancing, because there is separate indicator on the right side that shows that "Balancing" is happening. So I'm guessing if you put a load on the batteries, both would still discharge.

[@excillisbank2611]

Très intéressant ! merci ANDY pour cette belle expérience comme d'habitude inattendue.

[@gumpster6]

Great video Andy. That JK BMS seems like it's nearly perfect for what you need (minus the quirky startup procedure). Good comparison showing where the QUCC is lacking.

[@mondotv4216]

Now Andy to complete the test you need to swap the two BMS to see if the JK BMS does a better job with the old battery… easy for me to say :)

[@trevortrevortsr2]

Interesting contrast Andy, between two balancing approaches - I have had no issues with balancing my 310ah cells with my JK 120a BMS and it bleeds just 60ma - one thing I discovered is you can't run two bleed balance system on the same pack they interact - when a high cell is discharging it reads low and that throws out the second balancing regime - when I used my T8 balance charger I disconnected the BMS bleed balance - The T8 active cell balances as it charges 15 watts and is used extensively as a radio controlled drone charger in the field - The T8 is great for doing the initial charge and top balancing in one hit - none of that charging in parallel faffing about, it feathers off nicely on the heal too : )

[@petereaton6740]

Hey Andy at first I was somewhat sceptical about this but again thinking about your setup I was interested to watch your test and then once you added the higher input/output I certainly think active balancing is needed, I just run (atm) 105ah at 12v and my jbd passive BMS does as intended and even though I have a massive 580W solar setup lol it seems that when I add another 200ah of storage with the same BMS I'm sure it will do as required. This is not a high capacity system like yours being only 12v and no inverter so passive balancing seems to work fine. Keep it up you sure push the limits and this is what we as novices hmmm non experts need to see thumbs up to you ;)

[@dennis_m]

Hi Andy, I finally got the Neeeeeeey!!! I sent you an email. Thanks so much. Love from India!!!!

[@twingoman2000]

Great Tests! This helps to understand things much better, thanks!

[@theoff-gridhouseinrome3598]

Hi Andy great video, in regard to the charge voltage I own a Daly BMS and I charge the cells to 3.55 so 14.2volts from my 75/15 and I Haved a settings in the Victron that once a months charges the battery to 3.65 and then voltage comparator turns on the Active Balancer at 14.6volt.

[@extremeacc101]

I like this this is a really good setting

[@ianb4801]

Andy, is it possible buy extra plugs and terminals for those balancers and BMSs? Strikes me that ribbon cable would be just perfect for the job - provided the strands are beefy enough

[@OffGridGarageAustralia]

Yes, the Hankzor store has them separately. Just contact them and ask for a quote and what size you need. Click on the link for the JK-BMS on my website and use the chat function in their store. off-grid-garage.com/battery-management-systems-bms/

[@solargarage]

Nice Update

[@OffGridGarageAustralia]

Thanks

[@CollinBaillie]

I know it's not fiscally efficient, but maybe one of the Deligreen QNBBM Battery Equalisers I mentioned to you some time ago could be added to the QUCC managed bank. It would be another learning video. QNBBM equalisers are active and can shunt up to 3A between cells. Get 2x 8-cell boards (or 4x 4-cell boards) and parallel them for 3A balancing. They also sell the "Battery Queen 10S-24S 4A battery active equalizer", which as the name implies can balance packs from 10s - 24s at 4A. Maybe you can keep the QUCC when you rebuild the 1st pack, for science, and add an external balancer/equaliser as above? Come on! It's for the learning!

[@magicmanspaz]

Daly BMS is only a dismal 30mah balance. Active balancers are the way to go. There have been a few "experts" dismiss active balancers without actually properly testing them but it seems they are now finally coming around and changed their tune now there are so many people using them with success and others have done extensive testing.

[@SkypowerwithKarl]

Those that don’t like the active balancers had tried the older versions,tried something wrong or were unable to turn it off below 3.4vpc. The newer super capacitor types (as long as you have balance start control) are a world of change. Even better when it’s incorporated into the BMS like the JK. I run a 300 amp Daly with a Heltec +voltage monitor relay on a 620ah 48 Volt bank. The Heltec I have is a 5 amp but only if there’s a lot of voltage differential whereas the newer active balancers can throw full balancing amps even at low difference delta. I have older Nee active balancers but I can’t use them because they won’t shut off at the 3.4vpc and I can’t rig them, so they are useless.

[@waynescheepers145]

Another great vid with a nice test. I seriously need to try get one of those jk bms. One day

[@markdavidson6321]

Great comparison and good information, as always. I think that was the Elbow of the curve, not the knee. hahaha

[@marioescalona1640]

I am missing something so have a couple of questions: 1, why do you start balancing at 3.45v such a high and not at a lower voltage like 3.1v so you give the passive BMS more room to work on balancing the pack? How charging the old package (passive BMS?) to 2.45v max per cell avoid tripping the BMS if you'll always get there? Do you trip by having a lover total package voltage? Great job, keep it up!

[@CollinBaillie]

Pretty sure it's tripping when any individual cell hits 3.7x volts. It's over voltage protection for individual cells.

[@BeezyKing99]

I'd have to say passive balancing isn't all the best when used in high AH banks, I'd definitely set up with active balancing if I ever decide to build myself a battery bank.

[@edwinjidat8885]

Bro. I share little trick to full balancing lifepo4 battery. I done with 1mv different. Open ur series circuit. Make them in 1 paralel. Charging all off them until 3.7v. Let them sit at least 24h. After that, charging them until 0a before series and put bms or balancing board. I'm using lifepo4 battery for super fast ebike. My battery are 60v 20s, 60a. My motor drain is 200a. They still balanced till now. It had been almost 2 years

[@Josh-b3c]

Maybe but without proper intercell connections it's kind of chasing his tail around got to have a cells perfectly connected to each other one cell has slightly higher resistance you'll never be able to really get it right I know cuz I had that same problem finally got those all fixed had to repair some terminals sanded everything down perfectly flat and into no oxide now I have as little as possible resistance between cells and they stay balanced even though my BMS only does 50 millivolts balance current I charge right to a 3.63v for sale and the balancer only comes on for maybe one minute and then it's done and it's right at the perfect spot too where sometimes one celebrity high and then after 1 minute of 50 ml discharge that sell is now the lowest

[@CollinBaillie]

Andy already top balanced the cells. For days. Top/bottom balancing cells is about getting the pack into an equalised state, and then sticking the charger on it and letting the natural state of the cells let them cycle in a healthy way. This assumes that the cells are all fairly equally matched. A BMS give insight into and protection of individual cells, but traditionally no balancing. Now with an active balancer thrown into the mix, the cells are micro-managed to ensure differences between cells are kept under control. If you've got a BMS with an active balancer, top/bottom balancing is just... an antiquated practice. And over time the cells will settle at their full capacity (as configured in the BMS) and (given good cells) will mostly have matching characteristics.

[@Dutch_off_grid_homesteading]

Heya oke this test proves that with these big battery cell's you need a active balance with 4-5 amps of balancing power

[@batterynerd8779]

Hey Andy, since you have a Raspi with Venus os, you can buy the Raspi-7inch screen (70$). It makes for an awesome stationary touch solar display. There are geeat tutorials on yt. I wish i knew this before. I dit it myself and it works great! It costs nothing, in comparison to Cerbo GX (~800$). Cheerio!

[@showmequick2245]

Just for the record Raspberry Pi’s are sold out everywhere. The price they are not is a joke if this continues Cerbo might be the better option

[@batterynerd8779]

@@showmequick2245 just for the record in, Andy already has a pi. In Europe, they‘re not sold out, just a littlebit more expensive. The screen is currently 60 swiss francs. And on amazon 61 euros

[@showmequick2245]

Outstanding and informative video. JK is the new kid in the block. I think QUCC needs active balancer as a partner lol. I have 3 JBD /QUCCs I did t even open them up. ☹️

[@SuperBrainAK]

wow I think you should reduce the delay time for overvoltage. maybe 10 seconds is long enough. good thing you are going to completely get rid of that old BMS right?

[@aussiescotty2950]

Hi Andy, Another great show, I like seeing the numbers too. I just thought of something you can do when you've nothing to do - The perspect cover - does it prevent heat escaping the tops of the cells. You could drill about 100 holes in it to help heat escape. Just joking I am sure they will be fine. Cheers. And by the way I really like the JK BMS.

[@marzy2423]

I want to see a video with the jk hooked up to the old battery and watch it balance

[@jimm7165]

Looks like you need another JK BMS

[@danielkirk8571]

Turn playback speed to 1.5x and it sounds like Andy is a commentator at the horse races!🤣🤣👍👍

[@garys-half-baked-offgrid-dream]

My Daly just can't cope at all with grade B LittoKala cells, I can't charge above 3.35v and can't discharge below 3.05v at 10amps or the deviation is massive. The discharge and charge circuits both cut off and it is dead in the water. Still some usable power in the Goldilocks zone but it is a 💩 situation. Binge watching you JK BMS videos, hopefully Fogstar UK will be doing a black Friday sale. Even without a sale from them it is cheeper that I paid for my Daly.

[@TheDarkfairy47]

Guten Morgen Andy

[@Josh-b3c]

So your cell deviation difference between each pack is what would determine how equally it shares current between the past if all of the wire lengths and everything else is the same so that the pack with higher deviation would receive less current

[@CollinBaillie]

So, trying to understand you here, you're saying pack 2, with the JK BMS and a cell deviation of tens of milliamps takes more current that pack 1 with the QUCC BMS, and cell deviation of 100s of milliamps, governed purely by that cell deviation delta between packs, given that all else is equal? So if pack 1 had a cell deviation of 5 milliamps and pack 2 had a cell deviation of 5 milliamps, they would both charge at identical amps, given that all else was equal?

[@diydsolar]

As always great video thanks. It will be good to community to make a short video of your bms configuration with these Lifepo4 cells. Thanks.

[@jonasgranlund4427]

Really nice to see how über the JK BMS are. :) If I had one thing on the wish list it would be to be able to connect several BMS as master/slave so if you have a bank with 2 parallell JK they could communicate and keep track of each others banks individual cells. Something like the Batrium Watchmon Core, that way the banks would never be able to glide away to much from each other. But it seems like the JK is so good at balancing that it might not become a problem if both banks use the JK BMS, or can you see a scenario where they might drift apart?

[@OffGridGarageAustralia]

I really don't see a problem with the JK -BMS at the moment. The 2A balancer can handle the packs well and so far there is no need for any communication (except pure monitoring and information). What would be the benefit if the BMS of battery bank #1 would know about a high voltage cell in pack #2? They are physically totally separate and independent.

[@jonasgranlund4427]

@@OffGridGarageAustralia That sounds good but they are not totally separated, if one BMS kill the charging of one or other reason for one pack and the other one keep charging it might be a big difference when the turned of BMS turns on again and there will be high current flows between batteries right? Regarding the benefit of a BMS to ChargeController communication would be to maybe charge slower when the batteries are to cold (which you barely will suffer from in Australia) :)

[@hommerdalor6301]

Hello. I think the variation in resistance of the bussbars connections determines the voltage drop. If the resistances between each cell busbar of 1 battery are equal, there should be 0mv defference. For one of the batteries the total resistance could be 10mohm and dissipate p=r x i x i = 0,01 x 2500 for 50Amp = 25W The other battery could be 100mohm and dissipate 0,1 x 2500 =250W This power is not getting into the battery and is lost in heat in the connection. The values of resistances used in this exemple are not related to reality, I made them up to illustrate. Now if each cell has unequal resistances (bussbars + internal) , they all charge and discharge differently, and higher the current, higher the difference. For 0,01 ohm 2A x 2 = 0,04W 5A x 5 = 0,25W 10A x 10 = 1W 20A x 20 = 4W 30A x 30 = 9W 50A x 50 = 25W 200A x 200= 400W More heat = more resistance For copper, The resistivity Rho is about 0,0015 Ohm x mm x mm / M at -5C° 0,0016 10 0,0017 20 0,0018 35 0,0020 60 0,0022 80 0,0024 120 0,0026 140 Resistance = rho x lenght / area Let’s say we polished the bussbars, and get 100 square mm of contact area VS we hammered the bussbars, we get 10 sq mm (a few points in contact) The lenght is the depht of the scratches on the bussbars 0,01mm for the best one 0,1mm for the badest so R = rho x 0,00001/ 100 = 0,00000000017 ohm at best, and rho x 0,0001/ 10 = 0,000000017 at worst at 20C° 0,0000000002 0,00000002 60 0,00000000026 0,000000026 140 The resistivity at around 20C° for aluminium 0,0028 ; carbon 0,0040 If there is one scratch 10mm long, and 1mm wide we loose 10 square mm of area, it means 10% of resistance increase! If the aluminium post is scratch, it’s worst than hot copper at 140C° with rho at 0,0026 as rho aluminium is 0,0028 at 20C° and for sure will increase more and faster than copper. That was for new, clean surfaces, with time, corrosion will occure, there will be more oxygen in the 0,1mm gap than in the 0,01 gap, oxydation will be faster. Dissimilar metals like copper bussbars on aluminium posts will react, so if sunny hot Australia was flooded last week, now that humidity is everywhere in the air, It’s nice that Andy uses conductive paste to protect the connections and more or less fill the gaps. 800 grid sand paper is the minimum to use, using thinner grid is more work, but the smaller the scratches the best is the connection, the thinnest I own is 1500grid, but I know of 3000grid I rectify the entire surface of the bussbar instead of just the contact points, with one entire sheet of 1500grid attached to the back of a miror, the miror is placed on a glass table, and a square of bed sheet between them. I want all my battery posts at the same level, before compressing the cells together with 4 x 250kg ratchets, I place them posts down on my glass table, and surround them with bumper tape. This comment is too long :-D Cheers.

[@SabretoothBarnacle]

What ever happened with the East/West roof discrepancy investigation? Was it ever resolved?

[@CollinBaillie]

The rogue amps generated at night? I'm waiting for the outcome of that also.

[@SuperRockmonkey]

Has anyone hear of/found a 4s BMS with active balancing? I realise you can install a 16s jk BMS as a 4s setup but hoping there is a purpose build 4s BMS active ready to go out of the box.

[@jackiedines5229]

JK have a 4S version coming out in April!

[@steve82608]

I'm not sure this is really a valid comparison because Andy has had so many issues with the original battery cells. Seems like they were not a quality batch to begin with. The only way to validate is to switch and put the JK BMS on the old battery and the QUCC BMS on the new one. I'm sure he's not going to go through that much trouble, but... I would be willing to bet that the old battery would still be out of balance enough that the active balancer in the JK couldn't overcome it.

[@KenSentMe-]

I have boat where I've built 50Ah li-ion pack in 4s configuration. The boat can be on the trailer for very long times, which brought me to power saving related question: Can active balancer be turned on/off just by toggling on/off the single minus wire? I assume it cant work without it 😂 Do you see any disadvantages in this?

[@team13thirteen]

what you think about the Daly Parallel Prodection?

[@pbasista]

If you still have the "flying capacitors" active balancer device, it seems reasonable to me to connect it to the old battery specifically during these high voltage situations. It should help the BMS with balancing the cells. Once they are almost balanced, you can disconnect it and let the BMS finish the job.

[@michaelbouckley4455]

Yes, I agree with that. I think Andy blew his 16S active balancer. I bought one, for my 8S, after experiencing one high cell (which I discharged manually with a headlight bulb) The 200mA passive BMS I have had got hot, over 110C still working. I turned off the panels, then SCC’s, then the BMS to cool down, which was quite quick. Have not needed the Active balancer since, the cell is now often the lowest (the pack balances itself to within 0.001v @ 3.4v). I would connect the balancer, if high SOC, and I am away.

[@williamhustonrn6160]

Andy, I have a test request: Can you run a test to see if it takes longer to balance 16 cells (48v) vs 8 cells (24v) vs 4 cells (12v). My reason for asking this is I have been playing around with this on my 16s test battery bank with those cheap china active balancers and im seeing results that im boggled with and need additional test data. If I connect the 4s Active balancer up to 4 random cells after only 1 single charge cycle on the bank, im seeing the 4s active balancer bringing the 4 cells within balance in a fraction of time as the 16s can. I timed myself connecting the 4s active balancer to the 16s battery bank, 4 cells at a time and I can balance it faster than the 16s active balancer can balance all 16 cells. Granted the groups of 4s cells are not balanced like they would be with the 16s balancer, but im seeing it balance the 4 cells super fast. I have ordered 3 more of the 4s active balancers from china, because based on the results im seeing I think its faster and better to balance DIY battery packs in smaller groups of cells. I am thinking if you used 4 x 4s active balancers on a 16s battery bank it would be faster balancing than if you used a single 16s active balancer.

[@CollinBaillie]

You'd probably want to have more than the 4x4s balancers though. You'd need the 4 balancers for the 4 groups of 4 cells, but you'd want to be sure that the 4 groups were also equally balanced, so you'd have to overlap the balancers. Bal 1: 1-4 Bal 2: 4-7 Bal 3: 7-10 Bal 4: 10-13 Bal 5: 13-16 That way you're going to be in good shape across the entire pack. You don't want 4 rogue cells getting out of sync with the rest of the string. It would be a fun test. 5x4s balancers vs 1x16s balancer. How much difference would there be? Of course it makes sense if you're shuffling 2A around 4 cells, vs 2A around 16 cells it would be much faster. So yeah, your results make perfect sense, especially if the balancers work by discharging 1 cell, then charging 1 cell. Because the number of possible cell-to-cell combinations is way lower in a 4 cell module than a 16 cell module. There are 6 combos with 4 cells, I can only imagine how many possibilities there are with 16-cells. I may work it out... 🤣 (it's 120 combinations) And it depends on how those balancers work also. For a 4s balancer, it's probably direct connection from cels to capacitor, but for 16s I should imagine it probably uses intermediary cells to shuffle power from one extreme to another. Example, if cell 1 is high, and cell 16 is low in a simple consideration, maybe the balancer charges the capacitor from cell 1, and discharges to cell 4. Then another capacitor charges from cell 4 and discharges to cell 7, then another capacitor cell 7 to cell 10, then another cell 10 to cell 13, then 13 to 16. Now imagine having that 5x4s setup. You could effectively have 5 balance operations happening in parallel. And if they're all 2A you're potentially shuffling 10A around the pack. What's the difference in price of a 4S balancer, and a 16S balancer?

[@waynescheepers145]

the only issue would be with these cells is that youd need roughly the same "unbalance" in the packs to justify which i faster , but honestly ive thought about this as the china balancers use flying caps which means they transfer the power from one to the one next to it till it gets it where it needs it , hence the larger the number of cells the longer it takes, moving from 16 down to number 1 is 14 cells it has to cross to get to the target, where the 4s moves from 4 to 1 and only has to cross 2 other cells

[@williamhustonrn6160]

@@CollinBaillie at the time i first bought the 16s active balancer it was $72.09 and the 4s active balancer was $5.80. I just order 4 more of the 4s active balancers for $7.30/ea to do some testing as the battery in my sailboat is far larger than Andy's battery in a 96v configuration to power the 96v EV boat motor i replaced the diesel engine with.

[@williamhustonrn6160]

@@waynescheepers145 honestly i wish i knew how to build one DIY with an arduino or something, because im running an unconventioned solar setup on my sailboat, because the EV motor kit requires 96v input power, so i have a 96v battery bank and i origionally order 2 x 16s active balancers, but the china company sent me a 16s and a 4s active balancer by mistake, so instead of sending me another 16s, they just refunded my money. If i knew how to build one myself, i would do a larger 32s active balancer, but run higher quality capacitors and potentially higher than the 2amps...

[@CollinBaillie]

@@williamhustonrn6160 are you savvy in electronics? You can find open source battery balancing systems out there to study. There is also the Electrodacus DIY BMS. Let me find some info for you (including the correct spelling of Electrodacus)

[@keithcress1335]

My QUCC stops all balancing as soon as charging current drops below about 400mA! So, when a "cell protection" occurs which stops charging all balancing stops leaving everything stalled. What's your secret to keep balancing going after charging is triggered off??

[@OffGridGarageAustralia]

In the QUCC app, go into the menu and choose System Features (or similar... somethign with Features). In there turn of Charge Balance! It's the most annoying thing to have this on. It means the BMS will only balance if there is a current flowing and is above the threshold it can measure. Total nonsense...

[@keithcress1335]

Andy!! You nailed it! I've been stuck on this for WEEKS! That did it. Of course, turning OFF Charge Balance makes it start Charge Balancing. Why could I not see that simple logic.. Good grief. Thank you so much.

[@SkypowerwithKarl]

Andy, I don’t remember seeing you top balance the new group of batteries. (Some say you don’t need to on grade A’s, but I did anyway).That may account for the elevated delta at 56 volts. Since the JK BMS had time to work on them (only an hour Lol) you should see a lower delta next time you reach end of charge. I would also expect a very low delta even with moderate loads on the strait part on the voltage curve “between the knees”. I’m betting you’re replacing that older BMS with a JK too? Thanks for all your insight Andy, you’re part of why my system works so well. (Active balancing & never balance below 3.4vpc!)

[@CollinBaillie]

I reckon the idea of top-balance or bottom balance is redundant when you're using an active balancer... since it's purpose is to balance the cells. Top or bottom balancing is about getting the pack into the best state for cycling without a balancer.

[@CollinBaillie]

I'm hoping he keeps the QUCC and ads a Deligreen QNBBM equaliser, for science!

[@SkypowerwithKarl]

@@CollinBaillie It needs to be 16s, they are 8s and pricey. At this time, I just don’t see a better alternative than a JK 200, 2 amp.

[@SkypowerwithKarl]

@@CollinBaillie I’m still in the top balancing camp. I spent over 6K in batteries and the money is hard earned, so I’m going to do what ever I can do get the most out them. Regardless, A grade or not, the batteries are going to get out of whack. The higher the amp hour the more balancing current you’ll need, especially if it not going to spend lots of time in absorption/balancing.

[@CollinBaillie]

@@SkypowerwithKarl The QNBBM balancers are made as single module. They sell the in 4s, 7s, 8s and probably other configuration boards. They (individual balance modules) are made for paralleling, and the boards are also made to be paralleled, so yeah, 2x 8s boards would be required. They also sell a 10-24s active "equaliser" that balances at a range of voltages at up to 4A. I do know they're (QNBBM modules) pretty costly though.

[@stephsoltesz6731]

For a moment sounded like a Horse Race announcer... LOL Starting to root for the cells lmao... I noticed quite different Temps being shown, I know the new pack has temps sensors, does the old one or was it really that much warmer ?

[@USA-GreedyMenOfNoIntegrity]

55.2 absorb, 53.4 float works for my 10 batteries- 132kWh. I have balanced cells from Amy and Haley. No way can I go past 56V without having issues with over voltage or 150mv+ differential. That's okay, I like charging to 90-95%. Forget 57.6V! Quality control obviously sucks!

[@showmequick2245]

Wait. I thought Amy sold only brand new balances and matched cells no?

[@USA-GreedyMenOfNoIntegrity]

@@showmequick2245 If these $125-$148 cells are the best they can produce, I'd sure hate to see Grade B. That's all. No wonder everyone has to add 5A active balancers. I'll just turn the setpoints down like I did for a little margin. That is why my 220aH Trophy battery has 240aH cells installed and shuts down at around 218 aH. Oversize our battery with low quality cells.

[@marktheunitedstatescitezen185]

I have a DAILY 150A 15s 48v BMS that’s so post to balance my bank but is no

[@jasonbroom7147]

So, while charging or discharging, the deviation is virtually identical between the two and you only see the larger discrepancy when the battery is almost entirely charged to maximum capacity. I know you want to have the cells as balanced as possible, which seems to only be obtained by one of the BMS's, and only at a high state of charge. Give that these types of batteries shouldn't be fully discharged or fully charged very often, isn't the individual cell deviation when they are at rest, actively charging, or actively discharging, more important than the imbalance being observed only at a high state of charge? What happens with the cell voltages of the QUCC battery when charging stops and the cells drop to a "resting" state. Does the voltage difference self-moderate?

[@CollinBaillie]

Yeah, they will balance out, because of the passive balancer. But they will drop in voltage as a consequence, as passive balancing is 100% energy loss to dissipation as heat, while active balancing is around 95% efficient energy transfer (only ~5% loss)

[@CollinBaillie]

In fact, the larger discrepancy was observed when Andy raised the Full SOC voltage to be above the 3.45v the 1st pack had been "exercised" at for months, and as a result it started to operate in "unfamiliar waters". While the new pack is still "breaking in" so it's behaviour seems to be more flexible or agile. I reckon that given time, the 1st pack would grow into that higher SOC cap and begin to lope along comfortably like the 2nd pack.

[@jasonbroom7147]

@@CollinBaillie - Is there a way to quantify the net difference in actual ampere hours gained or lost? We're talking millivolts, and only at the very top of the charge cycle. There's no question that the active balance is more efficient, but if these were vehicles with internal combustion engines, would it be a case of one car getting 30mpg while the other gets 30.1mpg? I really don't know and I've not seen anything claiming to put a number on the net difference.

[@timh2870]

I dropped the charge voltage from 54.4 to 54.0. I used to have a few cells that would start to balance at 3.42v, but keep drifting up towards 3.50. Now every cell stays from 3.35v to 3.41v and none go into balance.

[@uhjyuff2095]

Your dual bms battery setup could get dangerous if for some reason one bms went into protection mode all night long and then the other battery was drained to zero percent at night. Then in the morning both batteries wake up again and BOOM 100 percent battery connected to zero percent battery. Good luck.

[@petertulp8806]

Then the over-charge setting of one BMS and/or the over-discharge setting of the other BMS will be triggered. And no current will flow anymore. Nothing dangerous will happen I think.

[@uhjyuff2095]

@@petertulp8806 yeah you right, just the bms will cycle on off on off until the batteries slowly equalize.

[@georgem3264]

Hi Andy ,why not balance from 3volts

[@OffGridGarageAustralia]

That's not going to work with LFP due to the flat curve. Everything under 3.4V is not worth starting the balancer.

[@davidpenfold]

Hmm, my REC BMS for Sunny Island is passive, but has 5 ohm resistance IIRC for cell balancing. I guess if I keep it a 3.45V it shouldn't be a problem.

[@CollinBaillie]

It would be interesting to know if this is a pack specific thing, or a good generalised observation. It may be because Andy's first pack has been exercised at 3.45 and is happy there. It's probably cell dependant, so all 16 cells charge to 3.45v equally but some cells have a higher internal resistance than others so they drag their feet and take longer to hit the 3.65v while other cells with lower internal resistance suck up power and hit the higher voltage more quickly. Or it could be vice versa. Higher resistance might mean it hits the 3.65v faster and the cells with lower internal resistance get there more slowly. I'm not 100% on how internal resistance relates to charge rate at near-full charge. Maybe if they were exercised more at the higher top-voltage this would slowly change over a few cycles. It's definitely related to the cells themselves, not specific to the cell chemistry characteristics. My point is, probably best to test the settings on your own cells than blindly rely on Andy's observations on his own packs. Andy's videos are about helping people learn how to decide what's best for their DIY systems, rather than "use the configuration of my system and yours will work identically to mine". But his settings are a really good starting point, and fine tune from there. He's done all the hard work to narrow down the settings to something decent. Then you tweak it for your own system.

[@USA-GreedyMenOfNoIntegrity]

@@CollinBaillie No, its the same bs I'm seeing on all 10 of my batteries making up my 132kWh bank. I have 6 different BMS brands. Chargery and JK are my two favorites.

[@luisaviles6066]

Why not move the old battery to new rack, and apply the same configuration to the old batteries?

[@CollinBaillie]

That's the plan, long term (moving into the rack). Who knows where he will go with the BMS. This is a learning channel. I'm hoping he will keep the QUCC BMS and add an external balancer/equaliser, for the learning opportunity.

[@TRETT1956]

Great Werner aus Süddeutschland

[@Josh-b3c]

Charging lithium iron phosphate to a higher voltage does not significantly degrade it as it does with lithium ion it is perfectly fine to go to even 3.6v every single cycle it does not significantly degrade your capacity

[@Josh-b3c]

But yes even with that there's very little extra capacity to be gained my charging to the higher voltage

[@CollinBaillie]

@@Josh-b3c And that was Andy's observation. And while the degrading due to higher charge may not be "significant", it is a degradation, and by limiting charge to 3.45v, that effect is still lessened. By how much, would take a test of two matched banks, over a very long time. Probably not in the scope of this channel.

[@_Miner]

Will you replace the QUCC with another JK-BMS?

[@CollinBaillie]

That would make the best sense in a system being built to just run unperturbed. As Andy is doing this to learn and help others learn, he may seek another option. I'm hoping he adds an external balancer/equaliser to assist the QUCC. *Hint hint*

[@bigbeef2654]

throw that Qucc BMS in the trash please.... and film it, it will be in it's natural environment

[@CollinBaillie]

Nah, keep it and add an external active balancer, as a learning experience.

[@MiguelSilva-mb6mb]

I would add a new JKbms to the old battery bank. You would have a bether system

[@kwynton8272]

Hi andy could you do a test with the inverters what they can handel with a startup of an electric motor like an angle grinder?

[@CollinBaillie]

The inverters Andy has will handle that easily. They're 3KVA and 5KVA, and that's continuous capacity. No angle grinder is going to peak that high. Maybe his smaller (non-Victron) 3KVA might see some effect, because I believe it's a "plugin load", but I may be wrong. Do you have an inverter (make and model would help) that's having issues, or is this an academic question?

[@kwynton8272]

I was planning to buy a 3kVA invertor but was wondering if it could handel it

[@CollinBaillie]

@@kwynton8272 I'll let someone else give a definitive answer, but a relevant question is what sort of battery storage do you have?

[@kwynton8272]

24V 200ah lifepo4 so that should not be a problem (cels can do 2c discharge and bms is going to be 300A continue)

[@davidkettell6236]

I guess the QUCC has to go.?

[@giuseppevillella9445]

Ciao amico mio 🤝🤓🔋👏🇮🇹

[@drewhodge3820]

I still don't know the best type of balancer for my 12v Lifepo4 battery?

[@wolfgangpreier9160]

depends how large the battery is. I use simple passive balancers. They take a long time, days, but thats no problem for me.

[@drewhodge3820]

@@wolfgangpreier9160 I've built two 12v 300Ah battery packs for my campervan. Both batteries have their own BMS. Two separate 40A Epever charge controllers. 70A B to B charger and a 70A mains charger. Solar comes from two 400W panels on the roof.

[@wolfgangpreier9160]

@@drewhodge3820 Even the JK BMS would not be enough. I recommend a 5A active balancer. At least.

[@CollinBaillie]

I'm trying to convince Andy to try out some QNBBM balancers, they have 3A and 4A versions. His QUCC BMS offers the perfect opportunity to showcase/test the setup of a supplementary balancer paralleled with the BMS. I believe they can also be controlled by a BMS if it has the facility to switch a balancer on/off.

[@igorkvachun3572]

Yes👍🔋⚡ 💡 🇺🇦✌☀

[@opless]

SEVENTY THIRD :-)

[@BischesseHunting]

A Proof that it is not the best idea to parallel batteries.... 3.7v should never Happen in any setup

[@pederb82]

Not going to argue with you but this clearly shows you are wrong about the cell voltage. 3,45 is far too low. But by all means if you want an unbalanced pack, keep at it - it will only add to the degradation with regards to deviating load on the cells.

[@CollinBaillie]

But it worked well at 3.45v. Once he changed away from that level, the disparity began to show. I think that maybe of the cells were regularly charged to the higher level, this would disappear, because they cells would "grow into" this SOC as the natural order for the pack. It's the change, stepping out of the established bounds, that caused the behaviour to also change. Keep settings consistent , and behaviour will be more consistent.

[@dahuaba9558]

1st 😂

[@bigbeef2654]

2nd