Top-Balancing different capacity batteries and full capacity test of the Frankenstein Battery.
HTML-код
- Опубликовано: 4 окт 2024
- The first top balancing of the Frankenstein Battery took me while. I had underestimated that some of the cells had a far lower SOC than others, even initially charged to the same 3.45V. It took me several hours to reach an acceptable result and balanced all 16 cells above 3.6V.
I explain why this battery is actually no different to any other battery we have built before and what we can expect, if we try to balance the battery at the bottom to make use of the remaining capacity of the 304Ah cells.
This out of the way, it is time for the first capacity test of the Frankenstein Battery. Which cell will be the weakest one and determine the overall capacity? The result was certainly surprising for me...
Gobel Power RN150/RN200 BMS. Best BMS so far?
off-grid-garag...
Xijia Inverter, 12-48V, 1-6kW:
off-grid-garag...
Xijia voltage regulation repair:
• CNSwipower Xijia 48V I...
Still 5% on the Jakiper "Jaki" Battery:
off-grid-garag...
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...
The Sunfunkit DIY battery. Great product to build your own 12V battery. Use OGGA at checkout to get 6% off. Video Review: • The Sun Fun Kit Experi...
Best prices for Victron gear in Australia: Auto Parts Co has you covered.
www.autopartsc...
(they have 4WD parts and accessories too)
Take a live sneak peek at the Off-Grid-Garage in the Victron VRM World (link at top of the page):
off-grid-garag...
Here is my Tesla referral link, if you want to buy any Tesla accessories or even order a vehicle. This is a great way to support my channel. Thanks a lot to everyone who have used it so far 🙏 www.tesla.com/...
My other (German) RUclips channels:
/ offgridgarageaustralia
Please don't leave any links in your reply. RUclips will not publish your comment in this case.
If you decide to sent me an email, I really appreciate your time and effort for doing this. However, please consider leaving a comment under my videos instead as I may not be able to reply to your email due to the sheer amount of messages I'm receiving.
as amazing as the information in these videos is, the editing is also amazing.
Thanks a lot, Aaron!
Hello Andy., Thanks for all your amazing videos...
Thanks to your shared information and knowledge I purchased the NEEEEEY,,, Active Balancer and have been using it a lot. One of the things I do for a living is design , Install and commision Renewable systems and also distribute , service and sell Batteries,, including DIY ones with EVE LiFePo4 cells.. I just wanted to add a little information that I have confirmed about the NEEY way of balancing , that actually balances more than a cell at a time, to add to the information you share. You say in many videos that the NEEEY active balancer only balances a cell at a time, and I have confirmed that is not Totally true. I will try to share a video I recorded with a monitoring graph of how the NEEY active Balancer performs the balancing process, where you can see and confirm that the balancer really balances several cells in each circle by getting energy from the Upper voltage cells and pushing to the lower ones.. You will see how it starts with one and as it balances it continues impacting many cells at the same time distributing the energy from the highest voltages ones to the ones with lower voltage.
I already send you the video to your email.. and you could share it if you see it of value...
thank you Andy for all the hours you put in to share such great information.
My pleasure. I'm learning with you.
Awesome video Andy! Looks like the 4s I did at the Jimmy's house (the guy with the cell that released the magic smoke). We tested 4 different brand cells but only did the high drain test. We came up with the same results. All worked great and battery with the lowest AH dictates the whole show!💥
Yeah, actually a very straight forward battery built. It simulates a maybe 8-10 year old battery where some cells have degraded faster than others. Now, will this mix run? Yes, of course it will... and not much balancing necessary.
Incredible video Andy. Explains so much about LiFePO4. Thank You!
Thank you, Nathan.
Good explanation of the issues at the top and bottom with the different capacity cells. Easy to understand with your drawing. Other than the mystery shutdown very close to the bottom it ran like a champ.
The mystery shutdown has been solved... there is yet another setting. Gobel watched the video and pointed me in the direction.
@@OffGridGarageAustralia Good to hear.
👍💪👏 Andy i had exactly the same problem one sell closed my bms .i must balance my cell .i have seplos bms ...that what help me was that i reduce the difference connected the one sell with another parallel for some hours .i repeat with another one with high v than the cell i had problem....i still have 0.40 difference from 0.150.Thanks for the help Andy that you offer to all of us that we are not specialist.
Awesome Andy, really great stuff and great descriptions- this is working out exactly as I guessed, but it’s so good to see it in real world testing. Cheers from very chilly Western NSW.
Three weeks away in the van so far, and still haven’t used ANY LP gas except for the oven.
Have a great holiday down there. Stay charged!
I bought two old 15S 4P 200ah battery packs to refurbish. The information I got out of this video is invaluable to me, thank you SO much!
Fantastic video Andy! Just all sorts of useful information. Thank you for all you do.
Thanks a lot, Philip!
Thanks Andy
Thanks Wayne!
Learning a lot with your testing skills
Thank you. Yeah, me too, actually...😊
Cool! Performance was pretty much as we expected. Looks like replacing those two lowest "Certified" cells with 304ah cells would be worthwhile.
These are all the cells I have though. We will keep an eye on the REPT cells...
@@OffGridGarageAustralia I love these tests Andy!
Here in sunny denmark it was also 7 degrees C last night - we call it summer ;-)
I remember these times! 😁
Great stuff as always Andy. This is what suspected but wasn’t able to articulate it very well on my comment when you said you’d try out the Frankenstein battery. Love it. Thanks.
It all works pretty well, doesn't it. Just wait for the next videos...
@@OffGridGarageAustralia ok will do. Thanks.
Balancing current doesn't come free, more current needs larger voltage differences. If you want to remove more current, the voltage measured at the same time will show a larger voltage drop. The low cell would require even higher voltages to force more current into it. Any thought of 0.3C or higher balance current would be impractical. Max 0.04C balancing seems rare but practical. Aiming for 10% to 90% SOC cycles sounds great but you've confirmed it's hard to calculate SOC accurately by voltage alone (large % of flat voltage curve).
The Frankenstein battery... ⚡IT'S ALIVE !⚡😊
Usually, the balancer either measures the voltage of a cell or balances this circuit, but not at the same time. That's why we see the balancing staring and stopping with some balancers. Others are a bit smarter and can calculate the voltage drop while balancing and take this into account.
We all love Beer ... especially german beer!, Thanks Andy
I have to thank you for the donation!
@@OffGridGarageAustralia Sorry Andy I have to confess it was not me, I have bought you a beer in the past though 🙂
Brilliant analysis.
Thank you.
Love this testing it’s awesome 👏
I love it too!. Happy to share...
Andy, try a high current discharge next. 100mV difference at high current creates some pretty serious heat on the battery. (Got to the end of the video where you have this planned already yay!)
coming up next...
Andy, I know its sounds a bit Bizare but put your strongest cells to the outer edge of the pack progressively at both ends and your weakest in the middle - It always seems to me that the outer edges of the pack get hit harder with any drain and charge quicker - is this because there is more internal resistance the deeper you go into the pack?
I've noticed this on many packs. And would love to know the answer. I did have a theory on this. The first and last packs are closets to your inverter and the heat conducts down the cables. If you see Andys thermal shots of Pack. Cells 8 & 9 are the coolest.
My theory is its the heat that degrades the cells closest to inverter. Might be completely wrong but I have seen this phenomenon.
It's almost like every year you need to physically move your cells around .
@@GarethJones-dk9yp It's odd I've built over a dozen packs mostly for Eco cars, Campers & e-bikes and they all seem to share this phenomenon although I'm still unclear why
Outside cells have one side against another cell and one side against the environment (ie inside of battery enclosure)
Inside cells have both sides against another cell.
Pulling current from, or pushing current into the battery causes internal heating due to the internal resistance, that heating means internal cells are hotter (on average) than the ones on the outside. (note you probably can't measure this with basic external cell temp sensing, you'd need temp sensors IN the cells themselves). In most typical environments of between say 5 and 30 degC, heating a cell speeds up it's chemistry, ie it takes less time to turn electrons in to an ionic difference. As a result you tend to see a larger voltage swing on external cells, which being colder also have a higher DC resistance, and those cells also have a longer time constant so take longer to reach OCV stability.
The temperature difference was very minimal. It always looks worse in the FLIR but actual difference was around 1°C.
The order should make no difference unless we pull high c-rating and the temperature difference is significant.
Maybe in a couple of years when your cell have been in service for a while longer. Capacity test all 16 and you'll probably see cell 1 and 16 with the most Capacity loss. More so cell 16.
Great info!
I know how much you love frogs
but
at about 9:15 you are doing a great frog imitation
But Seriously - thanks for another great vid
Nice ending…Cheers mate 🍻
I had some of them already. For a good reason though!
After the shot of all of the chargers a proper mad scientist laugh would have been appropriate.
Fantastic!
Hi Andy from Zuid Afrika. That is why you can run it without a balancer .
Okay this went to expert in layman's terms, thx
Love it,
👍👍👍
Servus aus Bayern. Great video as always and a really interesting projekt. I have been watching you for a while and haven't seen you test balanced vs unbalanced packs. I would love to see if there is any waste in capacity when balancing. My (very uninformed) theory is if one cell is at 3.45v and for example at 90% then another could be at the same voltage with a different SOC due to internal resistance or other manufacturing tolerances. Then when we are balancing the cells we are potentially loosing some capacity just to make the diff mv low. I can't see (within reason) what that brings?
I would love to hear your thoughts and see any testing on this.
When there is an extreme delta multiple balancers will compliment each other however as the delta gets smaller the two balancers will confuse each other. One balancer pulls down a cell, releases measures the results but at the same time the other balancer see the momentary actions taken and thinks it must correct. Stalemate.
I wouldn't use several balancers either. If you need multiple, there is clearly something wrong with your design or even battery pack.
@@OffGridGarageAustralia
Absolutely correct. I tried just for giggles and ……
Curious if it will top balance, faster. Nope!
Most thermally stable battery - can we thank @TheDigitalMermaid for the busbars then?
the BMS said it would turn off at 0% SOC - maybe you needed to spec 300ah to make sure rather than 280ah.
These busbars are great and can push 200A easily without any problems. We will test this (not quite 200A though).
We can only use 280Ah of these cells, even there is still some energy left in the 304Ah cells at this point of time.
With a normal battery you would probably say, it is discharged at around 3.0V cell voltage. Here, it is a bit tricky, because the 304 cells are still on a higher voltage at this point of time. So, I think we can really only go after the discharged Ah here to determine the best point.
Andy great test! It would be nice if could do a test with the chinese balancer connected during the full test, to see how many Ah are collected from the cells with higher capacity.
Your argument against using a balancer to transfer excess capacity from other cells is right, if you do a full discharge test with high load straight down to 0 %. But in a typical use case as home battery high loads are seldom and not very long. It may take several days until you reach 0 % on all cells. The balancer has many hours to shuffle energy around between cells. Even it moves only 1 A per cell, within 24 h it can move about 24 Ah between cells. Of course there will be some losses.
So it may be interesting to see how many Ah more you get if you do a discharge over 24 h with the chinese balancer constantly connected and activated.
That will not work. Like in the scenario on my whiteboard, the balancer has only 4 A to take from one battery and adding it to another. Then it does the next battery and finally the third. The load must be very very small to prevent any of the cells actually hitting 2.5V. Or, in other words, the balancer is not fast enough.
@@OffGridGarageAustralia Do not use the Neey, but use the other chineese balancer that does balancing on all cells concurrently.
You have to change cell no. 1 & 2 with your best cells no. 8 & 11 for your next high current discharge test. Maybe its better to have high capacity cells on place 1 and 2 and the temperature also will be better than.
The temperature difference was less than 1°C between the cells. The FLIR always makes this look dramatic.
You can’t move at the speed of light, but you can move at the speed of the cell and it’s one at a time 😂
That's what I expected for the discharged test (lowest Ah cell would be the bottleneck). What I'm most interested in though is how in-balance the cells will be when you charge it back up to 100%? Staying tuned...
Exactly my question as well. But I guess we will find out pretty soon!
You can live watch the progress on the SPAT Calibration Centre VRM
vrm.victronenergy.com/installation/219373/share/59a762f8
@@OffGridGarageAustralia but no individual cell voltages on there 😭
Awesome. I'm curious how many amps are being provided from the larger cells well under load. And, what the capacity is left in the 304 ah batteries. I've done smaller scale experiments at one time with lead acid batteries. With a 250 ah, and 100 ah in parallel. 0:00 The smaller battery 100ah provided less amps Amps than the larger battery in order to keep them equalized. Of the total 350ah I was able to get able 165-170 ah from the batteries. Close to capacity of @50% dod.
I love these tests. Is it possible to measure current being delivered from each battery under load? And how much capacity is left, in each battery? Would a different configuration of battery, ( where cells are placed in parallel have a difference in how many overall ah are delivered? I thought the 304 ah batteries would provide more current. Just some ideas, and thanks again Andy!!!!
If you parallel cells with the same chemistry, you should be able to use 100% of the sum of the single capacities. Paralleling a 250Ah with a 100Ah cell, the 250Ah cell will deliver more current while discharging and also takes more current while charging.
The current is the same in each cells of my battery as it is a 16s installation.
You're amazing with a white board! How do you draw such straight lines? Thanks Andy!
Hahaha, that was maybe after calibrating a S.P.A.T. for the next day.😂
Nice video as always. You are my insperation Andy.
Question for the community, Is anyone else concerned that the only real protection we have against a serious battery failure is is "the single point of falure" of the 1 BMS. If there is a fault in the BMS it could allow possible "runaway" of the charging, resulting in a posible FIRE. does anyone think it's possible to connect 2 identical models of BMS in series. So if either fail's it has a Buddy to look after any issues! Any thoughts would be greatfully appreciated. Thank again all in the community!
You ment to type "paralel" for BMS?
Interesting. Though, what kind of BMS fault would you protect against? You d also need to fuse all balance leads and the BMS itself to protect them from one another s failure? Adding a little resistance, could that hurt?
My concerns are the same though. Like Fawlty Towers:"Sir, but what if there is a fire?"
I keep mine outside of the living area.
But people build their own for boats too. I would like that too. But there will be water...
I think they meant to parallel the balance leads but otherwise series connect the current limited pole.
@@lua-nya Yes my mistake. Parallel the balance leads but conect the main charge/discharge leads in series. So if one fails to protect the second is there to keep things safe! Don't know how I would test this setup, so just a thought.
@@camielkotte Thanks for taking an interest.Yes you are right parallel the balance leads. As to "the fault" I dont know. Any fault that would, say, allow the cells to continue charging even if one cell goes over 3.7V and is not noticed due to a sensing fault. For instance you would be "very" adveturous to fly across the atlantic in a single engine aircraft (1 BMS). So most that do those route's have 2 or more engines (2 BMS's). Just a thought!
There is also a fuse or breaker in each circuit. The BMS is just one measure to protect.
7:25 I never top-balanced my cells that way either. its a huge waste of time IMO.
I installed them all in series with a JK and use it. After several charging cycles, JK balanced them all.
If you use Li-ion instead of LFP, that is acceptable. If you use the factory cells (either without capacity checking or after doing exactly the same capacity check to all of those) it is acceptable. Here cells were supposed to be close to 100% before building pack this time that have not happened. Furthermore LFP has very flat curve at 20-95% so differenciating between a 95% charged cell and 80% charged cell is hard.
everything is correct 👍
It's seems far better than expected.
Freezing 7 degrees, really that's summer in the UK 🤣🤣🤣
I know, right?! Everything under 20°C is COLD 🥶
You could use Node-RED to reduce the charge/discharge current when at low/high SOC. This would allow for top and bottom balancing to fully use the +24*6Ah of your pack.
I can't do this in an off-grid installation as explained. Not possible.
For those of us living in Australia, can you share where we can pick up a pair of those safety boots you are wearing in this video please so we can try to pick some up for ourselves? They look very comfy. Thanks.
Hi Andy, You always Top balance and write off bottom balancing but in this one case, wouldn't it be better to ensure maximum use of the worst cell. Im wondering if bottom balancing would cater for the worst common denominator. With top balancing there could be a case where you would stop charging possibly before the worst cell is at its complete absorption. As a future test, Id love to see the different total power between top and bottom balanced batteries to put this one to bed.
If you bottom balance, you can also only use 280Ah. The 280Ah cells would be full, the 304Ah cells would have some 'space' left. The scenario would be the same.
Also, there is no equipment which does bottom balancing for you (like a bottom balancer electronic), so very hard to maintain. As I said, it's a bit of a thing from the past.
But maybe we can run such a test and see what the differences are.
You had alarm status CUV alarm (Cell Under Voltage) on top right corner of the screen at 27:50, even when you didn't have anything on "protect status" . Maybe that stopped the discharging? One of the cells, probably cell 1 or cell 2 just went under voltage for a short period of time. When bms cut the discharging, it went a bit up again. That is my guess.
That was just an alarm. There is actually another setting, I wasn't aware off, which shut down the BMS.
BTW, if you wanted to use "all" of your capacity, you'd run 2S8P with each parallel string being 5 of the 280 cells and 3 of the 304 cells, but of course this would onyl furnish a much lower battery terminal voltage compared to your 16S. Without even numbers of each cell, where every 280 could be parallelled with a 304, this is the best you can do. With 10 and 6 cells, you ar elimited on your common demoninators to get an even noumber of the different cells into parallel strings (5/3 is in fact the only one)
That would not run with my 48V inverter 😁
I also don't like paralleling cells. Far less control and if one cells has an issue... not good!
You would need individual fusing or protection for each battery, so hence I always recommend a series connection with a BMS and then parallel the banks.
Great one again.
Andy when I put the batteries in to my battery cupboard which is made up with wood on a metal frame when I keep my multimeter on the negative and on the metal I can see voltage is this normal. Ohh I havent earth the cupboard yet.
Put a light bulb in between which can handle the voltage. If it comes one, something is wrong, if the voltage just collapses, it's all good.
Yeah, definitely ground your frame.
you might need a neey 4 frankenstein edition - but for sure no one of us wants to get such frankenstein collection of lfp cells that are so naughty, old and suffering from deviation.
I assume you were being humorous when you said it was freezing cold last night, 9 or 10 deg C. Garage door open, no sign of breath, T-shirt, lol.
It's the Australian winter. So dead serious... 😉
Ich hätte ja sowas von Lust einfach neben dir zu wohnen und am Abend bei nem Oettinger zu labern :D
We could talk all night long. I even would share an extension cable over to your place...
Heya, for a frankenstein battery it's not bad at all
but andy, would the greater cell not "put" another energy to the system? Like if you unload 100A, the greater cells would be delivering more A as the "little" ones? Dont you agree? I would suggest to measure this while a load is on....So usually the greater cells are not lowering exactly as the little ones.....
(And iam commenting by looking only at 12:30 at the video, as you make the scematic of the batterys)
Yes, absolutely, You would take a lot of energy and higher currents from the higher voltage cells.
I am just starting to watch, but my pre-impressions are it’s impossible to top balance batteries with such largely differing capacities. When the lowest capacity 280ah cells are peaking, the 304ah cells won’t be in the knee.
Well... you have watched it now. What do you say?
yes I also did not top balance all my cells in my 4 packs. after months they are getting better but looks like it will take me a year with JK bms 2 amp. It seems about every month I keep getting lower cell voltage differences. They will just have to keep doing what they do until they are all more even. I have .001-.004 at 3.4 average. when I get to 3.43 average is where I still get .05-.08. way better than 3 months ago. Maybe by end of summer in USA they will be balanced lmao.
Hi Andy,
Just fired-up a Gobel Rn150 BMS for the first time and cannot modify the settings using the "login password" (123456)
I have downloaded the software from your Google Drive and noticed you are using a newer / different version of the software
Your Google Drive: 1.01.02_11
Your Version: 1.01.02_17
Can you please update your Dropbox? (or paste another link)
Thanks again, Loving your work!
Ian
Super
Kann es sein das die Zellspannung untereinander über 500 millivolt Unterschied war? Ich glaube eine Einstellung im BMS schaltet dieses dann auch ab weil es davon ausgeht das eine Zelle defekt ist.
I know what you but was unable to find such a setting in the software. I'll ask Gobel if there is such a protection.
Good point!
Andy, why don't you try to give them a natural swing, that is, connect the Frankenstein batteries in parallel and leave them for at least 7 days, then put them in series and start the tests again.
I greet you, have a pleasant day
I never did this method and always balanced them in series. It is a bit more effort but faster.
What do you think about this: since the most damage to electrodes is when the cell is fully charged, wouldn't it make sense to bottom balance the cells. Then only one cell would reach the maximum cell voltage setting and the other cells would get less damage. Your opinion would be appreciated thanks.
Can you do another video and data log the voltage of the cells over time and integrate the power output of each cell. The larger capacity cells have a higher voltage over time so put out more power the the two weaker cells. This should show your assumption that the pack only outputs the capacity of the lowest is incorrect. It should be some where in between. It would be very interesting.
What about to to build the super Frankenstein battery? Is it possible to mix pouch and prismatic cells? Is the discharge behavior of pouch and prismatic the same? Is it possible to parallel use a lipo pouch 12V and a lifo prismatic 12V battery? Beste Grüße aus Hessen
Wait for it...
I'm also from Hessen. Nordhessen 💪
@@OffGridGarageAustralia perfect! I ve made a test run. Internal resistance of pouch cell battery seems to be lower, as the current is higher on the pouch cell side ( 10%).
Thanks for all your really good videos!
*I have a strange situation. I have a 13s 48V 32ah li-ion battery, and when the battery is in standby I have a deviation of 0.002V, but when I try to charge with 15A (for example), that deviation jumps to 0.047V. The voltage always jumps on cell group no. 6 and group 13 has the lowest voltage. Why is this happening? This is a new battery*
Could be a problem with the connections/bus bars. Higher resistance increases the voltage drop at higher current.
Yeah!
and you look like the first! ;-)
🥇
A Generic one:
- You know how much in percent is 0.001V (1 mV) from 3.56V ?
- What is the precission of the analog inputs of the Charger and or balancer ?
- The tools you use to messaure voltage, what is the precission of these ones ?
- You know that you have to calculate the total precissions to have the real value ?
- How you guys will tell me how you messaure 0.001 V (1 mV) for deviation ?
I am from the laboratory field to messaure millivolts in the satalite range. But the toole we use are something like Panasonic, Hameg 8000 series and Pillips and some other brands. And the precission of them is 0.1% or better at all.
So come on guys.....1 mV ? ? ?
What brand and model is your thermal imaging camera?
super frankenstien- how about adding 30ah of cells in parallel to the 280ah’s ?
How large is the opening for the display. I have the jk bms with 4.3 inch display and would like to use this enclosure .. ty
Hope you allow the batteries to cool down before recharge for chemical process to settle down.
There is no heat building up in these cells.
But I don't understand, why my cells drift away between 60mV deviation at 55,5v despite they are top balanced at 3,65v...? Should I do top balancing again, or is a little bit of deviation "normal live of the cells"? I have 16 cells lf280k eve from gobelpower
You should take 32 cels and match them up in parallel. This will automatically increase the whole pack life. You get pairs of a stronger cell supporting a weaker one. So 16s 2p with capacity matched pairs. Otherwise the weakest cells wil decrease even faster and bring the pack even more down.
I have only 16 cells left. I don't like paralleling cells as there is no way to monitor them individually. I would always recommend to parallel strings of batteries where each bank has its own BMS.
Andy if I had a heated battery in a battery bank do all batteries need to be heated
Isn't that an ACIR meter? DCIR will be a bit different, won't it?
Personally i think people spend far too much time worryiing about balancing cells when it really doesn't matter unless you absolutely need to leverage 100% of the packs actual SoC! For a home energy pack (rather than say one used in a electric car) realistically i'd suggest you operate in a useable window of around 80% of the actual packs SoC. In that window, having cells imbalanced is frankly irrelevant!
Also, there is a huge difference between a new built pack being out of balance and a pack that becomes unbalanced over time as the cells SoH changes.
For a new pack, built from individual cells, then gross SoC imbalances can be present and yes, these probably need fixing. IME, it's far better to simply parallel balance all the cells before you build them into a pack, ie connect every cell in parallel to every other cell, and just let the charge redistribute itself with every cell at the same state (age/temperature). This is incredibly cheap and easy to do, just some wires with croc clips and a suitable current limiting resistor just in case the starting imbalance is massive (1 ohm's about right).
Once you have a pack in balance, then really the balancing should not actually do much from them on. You charge stops when the first cell hits Vmax, and the discharge stops when the first cell hits Vmin. If you work out the loss of capacity due to the imbalance at that point, chances are, it's a) absolutely tiny and b) why are you goign right to the ends of the S0C window anyway?
As cells age and loose capacity, the lowest capacity cell is absolutely experience the largest voltage change per unit charge, so the cell that hits Vmax first will probably be the one that hits Vmin first. At that point you're done, you could balance the pack, but your "tidal" capacity is absolutely set by the capacity of your smallest series cell(s).
It's also worth noting that paralleling at a cell level is a very good idea from a purely probability perspective. The more cells you parallel, the more "average" that parallel string will become, ie any low capacity cell is more likely to be parallelled by a high capacity one.
The problem here is that you cannot really charge LiFePO4 to 80% only. And it would not make sense as we have tested many times. It's a different story with Li-ion...
Once the voltage rises in LifePO4, you're already far over 90%SOC. Absorbing at 3.45V for example, will leverage over 99% of the capacity. That is also the time to balance. But I agree with the principal, that balancing is not too important for home solar storage with LiFePO4.
@@OffGridGarageAustralia I wasn't suggesting charging to 80% SoC, but simply limiting your useable SoC window to 80% If it makes sense to charge to a known voltage (whatever S0C that relates too) to effectively use the tail current to determine when "charging has ended" then discharging from there by 80% means you typicalyl don't care about individual cell voltages (for a low current pack, ie one where your discharge rate is low enough to avoid dcIR effects) 🙂
@@maxtorque2277 Yes, just use the window between 3,00 V and 3,40 V. Don't use SOC, but use voltage for control.
Stop charging when reaching 3,40 V, restart charging when going below 3,33 V, stop discharging when reaching 3,00 V.
I use active balancers doing balancing at a difference > 10 mV all the time. So in long-term all cells are always within 10 mV. Balancers can do work 24/365.
This works perfectly since 2016 now without any intervention.
@@maxtorque2277 Ah, right, misunderstood that. If your battery capacity is relatively small and you do a full cycle every day, 80% DoD is a good number. Mostly people tend to overbuilt their capacity and during normal usage, the DoD is far smaller, so not a problem. I usually use mine between 50-70% with daily cycling so individual cell voltage are not of any concern, as you said.
And what happened if group the same capacities in groups and then parallel all the groups?
I think the battery shut down a bit early because it wants a bride! At the end, I thought the locals had come with their pitch forks.
They are at the front gate already!!! 🔫
35:15 Rest in Peace 🏴➕✝
I'm feeling reborn!
For normal use it is not quite correct that you need a ballancer that can handle the full current to at least use more out of the battery than the weakest cell can supply. The usage in a house is rarely constant in real life. If you have intermittant use of 100A a smaller ballancer that is running constant will help. 100A for 5 minutes = 8.333A for 60 minutes.
A 100A peak could be enough to bring your lower voltage cells to hit 2.5V.
Anyway, it is not feasible to bottom and top balance one battery. The gain will be very small.
Morning Andy, I'm starting to get frosty now down here. I miss Qld now days.😢
Pure sunshine during the day but cold in the night. Sun is not good for my solar production. I need clouds ⛅
@@OffGridGarageAustralia lol clouds? Why clouds? The system making to much now you have upgraded.
Moin aus Ostfriesland 🙋♂️
Hi 👋
i dont beleve, the top bal is still intact, because cell one and two did burn engery due to heating up. the balancer has to correct that amount of heat/energy in the next top cell balace event. my frankensten battery does it in this way, too. :-) best regards from hannover/germany !
It will be off slightly due to the higher internal resistance but probably not much at reasonable charge/discharge rates
I guess, it won't be worse than with any other battery. It is still charging but should be back to 100% tomorrow at some stage. Watch the S.P.A.T. Calibration Centre for all details.
Andy, you're missing an important virtue... patience 🙂
I'm running capacity tests with an iCharger X12 on 20 Blocks 12s/48V/22...27Ah/LiPo of a Think City BEV Battery Pack CHG (4.1V/10A) -> DISCHG (2.8V/3A) -> CHG (4.1V/10A) -> Storage (3.8V/3A), with every Pack taking about 24 hours to complete and all Data of the cycles stored. This takes now roughly a month, because I also try to align it with the energy prices (Tibber) while using my home storage as the source for charging and the sink for discharging. Energy is usually cheap during noon and expensive around 6-8 pm and cheap again in the very early morning.
A plastic lid would be in order instead of metal
There is a plastic film under the lid but yeah, a (clear) plastic cover would b great.
can you retrofit the BMS/screen/ports in to the Seplos case?
With some modification it should be possible. It's basically the same components.
the bms daly program well not work on the phone iwhen start up it just quite i have reload it still thr same other people als have same proble, what you thought on this
I haven't been able to find what settings you prefer to set the NEEY to, am I missing it somewhere?
There is no dedicated video about it. I have shown it a few times in different videos. I'll make one next time, I have it connected...
@OffGridGarageAustralia sounds good, was hoping you could post something that shows your preference for settings
Yeah.....the Frankenstein needs some more external power to fire it up...!
Where are the arcs? Still missing.... ;-)
I'm trying to avoid sparks!!!!
"Dsg cut-off volt" sollte ja nur den Mosfet öffnen und nicht das BMS abschalten. Man hätte ja dann nicht die Möglichkeit dass sich die Batterie wieder automatisch laden würde wenn die Sonne scheint. 😏
Misteriös...👽
Yeah, it is strange. I watched all my video footage but could not find a trigger for the BMS to actually shut down. Gobel will hopefully find out soon...
Freezing cold at 7 or 8 😂 guess you don't visit northern Europe a lot, during winter 😉
Not any more, but I can remember temperatures at -35°C. Now, we're climatised and everything under 22° is cold.
@@OffGridGarageAustralia haha 😆
Thank you for your knowledge. I've been charging 280ah 16 cells at 3.65v trying to top balance them. I've only charged to 3.31v from 3.29v. Can I charge one cell at a time?
What you can do: charge every cell to 3.6V one after another. Then, they are all on the same level.
@OffGridGarageAustralia I bought one of those cheap dc power supplies from Amazon. It does 10 amps. I have them all in parallel. I'm not getting much of a charge.
@@OffGridGarageAustralia how long would it take to charge one 280ah cell?
@@briarbenchhomestead946 Ah, yeah, that will work. It can take a very long time though.
280Ah x 16 = 4480Ah
10A charging makes it 448h until full.
So, be patient...
@OffGridGarageAustralia Good day! Do you have a recommendation for a faster dc charger? This is taking way too long. I'm thinking about putting my cells in series and charging them with my lithium charger. Also, what size arethe balance wires are the jk bms?
So as I said on previous video.. you needed 10amp balancer at minimum.. where you get one.. is another question
Well, turns out, I don't need any more balancing than with any other battery. Once top balanced, it is all good.
@@OffGridGarageAustralia have had similar 4s battery banks.. and it keeps on drifting faster than normal cells.. combined 105ah, 100ah, 155ah. after top balance within 2 months of use same thing happens. though my cells are rejects I could not return to China due to slight damage..
using daly active balancer...
4:45 isn't it you have that inductor based balancer? from my understanding, that type takes power from all the cells, and dump it on the weakest cell. perhaps that is a better one for this kind of situation?
I have the 10A Heltec inductor based balancer. We have tested it and it and the result was... underwhelming. It is only a 4s version I have.
@@OffGridGarageAustralia Indeed that balancer was sadly meh as well :(
Draining the battery will unbalance the battery. Charging will unbalance it too. Mixing 280Ah and 304Ah will keep a voltage drainage from 280Ah to the 304Ah battery's while charging and vice versa when discharging.
LFP cells do have very high colombous efficiency (A*h DSC /A*h CHG). Pull 276 A*h out of fully charged 304A*h cell and it will take the same 276A*h back to full. Pull 276A*h out of fully charged 280A*h cell and it will take 276 A*h back to full.
Charging will drive all of those cells back into balance.
That is why Andy had to charge some battery's separately.
No, that was not the reason for charging them separately. These are all cells used in different projects and I never had them in parallel. As I explained (twice?), I charged them to roughly 3.45V with more and less absorption. This was all per purpose.
I'm recharging the pack at the moment and we will see how it comes back together once we come close to 100% again...
with the 35/80% ratio of usage voltage will stay close to each cell and need less balancing !!! change my mind
How do you determine 35%/80% though?
@@OffGridGarageAustralia i will try some old LFP for my storage and from all i learned from all nice guys like you and some other youtubers . i could relly after testing and matching (on older Cells )
..but i will choose to be far away from edges... close to 35 %and 75 % even if i have to oversize my capacity !! this might keep it more balanced if loads are too low.. thinking of 48 v
@@panospapadimitriou3498 I would not worry too much if it is LFP chemistry. Charging them close to 100% is totally fine even if they are older. If you keep them at 3.45V max, it will be very save and you still get almost full capacity.
🐸🐸🐸
And that's the reason, why the car manufacturers want their cells to have exactly the same capacity and internal resistance. Since in cars you have a lot higher currents, it wouldn't be possible to balance the cells
Yes, exactly. They are perfectly matched. I have never seen more than 10mV of deviation in my Tesla battery consisting of ~2700 single cells.
@@OffGridGarageAustralia which kind of battery do you have in your Tesla. I think NMC and NCA cells are easier to balance since they have a more linear voltage curve. So it is possible to balance earlier while with LiFePo4 it is only useful above 3.4 volt
Correction: It's the weakest cell + whatever you can balance. And you're going to go continually out of balance because of the pressure (voltage) on the batteries being different with each and the state of charge defines the amount of back pressure (i.e. resistance to charge at a given charge voltage)) And yes the balance will be off on a full discharge and recharge if you don't have a balancer at all and it will get worse every cycle because the back pressure will shift the balance constantly. This is why you need a top balance all the time and it gets worse the older the cells because they're no longer matched. In your case it's absolutely MASSIVE delta every cycle.
There is a working solution: do only top balancing in Lithium based chemistries.
Colombous efficiency (A*hours DSC / A*hours CHG) is very high for all Lithium-ion and LFP (99,5% - bad, 99.8% - better, 99,9% - half way decent cell) battery cells and difference of those in similar battery cells witin 1 battery pack will be miniscule. Therefore top balancing is some (albeit small) amount of work to do and whole battery is easy to keep top balanced (for example turn balancing on at 3.4V per cell for LFP). Any cell with capacity above lowest capacity one will have that extra capacity unusable. This sacrafice is worthy for me as cells should be close in capacity anyway.
I would say balancing during bulk of charging/discharging is counter-productive. Real difference in internal resistance will create voltage imbalance when cells are at same true SOC creating more future work for balancer.
@@volodumurkalunyak4651 As I said. That isn't the issue. The issue is that your top balance is always going to get erased with every cycle because of the pressure difference that you can't see because of the flat part of the curve but is still there.
@@jameshancock Top balance will NOT get erased as voltage difference (at lower state of charge) doesnt matter. Balancer should only balance at >3.4V for LFP - only balance very close to 100% SOC - keep cells top balanced.
Lithium-ion and LFP batteries have extremely high colombous efficiency (A*hours DSC / A*hours CHG) so those generate very small amount of disbalance themself.
Different cells does have those A*h at different voltages and therefore different W*h - that doesnt matter, those come back at 100% to 3.4V (LFP)
I assume the top balancing will be still intact when we fully charge the battery again. That is because the balancer is off below 3.4V and we have discharged (and charged) the same Ah into each cell. Of course due to the difference in internal resistance, we will see some variation, but that is just the same with any other battery as well.
@@OffGridGarageAustralia it isn’t just the internal resistance, it’s the pressure (voltage) difference.
It’s why a cell at 3.4v takes less amperage than a cell at 3.2. Same as a pressure tank for a well pump. If you have 2 of different sizes in series you’ll get imbalances that will form over time because of the difference in pressure at a given volume of water.
Same here.
Only your ocd is unbalanced 😂. An remember that frank was the bruva of Albert
Deine Adresse ist ja auch nicht schwer zu finden :)
Send me something nice!