Almost lost it! Why just an active balancer is not enough!
ฝัง
- เผยแพร่เมื่อ 24 ก.ย. 2024
- The BMS disconnected several times! Saved the battery. Just because the active balancer was too slow and could not keep up. I share my findings and thoughts with this situation occurred last weekend. I was really happy with the active balancer until this moment. It seems like, it has still trouble to catch up with the demand of pushing amps back and forward in such a high capacity setup. Luckily the BMS was there and disconnected the battery once one cell hit 3.65V.
It was a bit disappointing to see that the active balancer was struggling to prevent the BMS from disconnecting the battery several times. The balance currents just were not large enough to make a difference while still charging the cells at the same time with 20A.
I probably have to give the system more time to top balance now and need to increase the absorption voltage slowly so the balancer can keep up. It's still not an ideal solution... Also, the absorption voltage was set to 3.45V only which means there is still a lot of energy to be 'balanced' across the cells until the voltage moves. We're still in the flatter area of the curve. It is more beneficial if the cell voltage is higher and amps moved have a bigger impact on the cell voltage. I'll keep testing....
Testing the difference of charging to 3.4V and 3.5V with and without Absorption.
• Charging LiFePo4 (LFP)...
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!
**** These are the batteries you are looking for ****
🔋➡EVE LF280 LiFePO4 cells 3.2V/280Ah: off-grid-garag...
🌞➡Solar, 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...
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 (free 1500 SuperCharger km / 1000mi for you and me)
ts.la/stefan96963
![[LIVE] : ONE ลุมพินี 80 | คู่เอก "รักษ์ vs ยอดนำชัย"](http://i.ytimg.com/vi/ckLMRorsMeY/mqdefault.jpg)
Another great video Andy. Anybody can do a tear down/ review videos. But Andy is is doing real world off grid problem solving videos that some of us run into.
Thank you very much for your feedback.
Totally agree.
Andy's channel has solutions to real world issues people face all around the world.
This is also why a BMS with separate charge and discharge control is nice to have. You can have the balance catch up while not charging, but still able to keep powering your loads without a disconnect event. Yes, they are FET based, but there are some really good ones out there and I wouldn't change mine to a relay version...
Which ones do you consider really good and able to handle high amps?
Diybms has had some nice upgrades recently with SOC added.
@@JimmyLLL On the 'lower cost' end, something like a JK/Heltec BMS - I've had those on my packs for almost two years now without issues.
@@upnorthandpersonal I really want to steer clear of cheap chinese BMS's simply because i'm powering a house 24/7 and want reliability.
@@JimmyLLL I do power a house 24/7 with these, and even had a situation where a 225A class T fuse blew which I thought for sure would have taken out the BMS, but it's fine. On the other hand, something like a Batrium and Victron equipment it can talk to are the more expensive option, and they work fine too.
This was very informative helpful and I was actually excited to see someone actually talking of using both types of BMS in harmony. You are a hero. I will be looking into the parameters of turning off the resistive balancing in the passive BMS so it doesn't diminish the capacitance reactance transfer of the Active Equalizing Balancing BMS. I really 🤔 think you are the only one that is aggressively diving into the fullest working of this setup. Also because I am on full solar power all my computers are on UPS battery 🔋 backup, because of things like you just went through.
Thanks so much for your feedback. I'm always not sure about such videos as they can be long and boring. But I always think, I'm not the only person who has these problems and I'm more than happy to share what I experience. So thanks again for your kind words!
@@OffGridGarageAustralia I just hope that the interest of the important details we glean from your testing stays at a sustainable rate for you to receive comfortable compensation for your time. Making content for TH-cam is not as easy as those that have never done it for a sustainable time understand. I also think 🤔 the energy of your enthusiasm helps keep people entertained. I like how you talk to the viewers like they are there with you and you use us and we, making it personal like we are figuring it out together. Electronic circuits, programming logic, RF energy and A/C and D/C applied voltage ⚡ is my field of expertise and I made a lot of money enjoying the work but, that has nothing to do with making Chinese apparatus work with or as intended, in our energy storage applications. 😀
@@xr680r ‘
Love your work, Andy. Your channel is proving invaluable to me as I design my own off grid garages here at home... just an hour or two north of you actually. Thank you so much for all your work.
I like that we get to see the real world uses of this cheapest made in China electronics that we might want to implement in protecting or benefitting our self made power source. In the end I want to build the best low cost package I don't have to worry about. If I can get a more fluid charge to discharge cycle lowering the amount of cycles by adding a $70 active equalizing bms then the investment is worth it to me. In all my searching of TH-cam it seems it's almost religious response to using a active equalizing balancing BMS, as a waist of money unless you are using used miss matched batteries but, here we are seeing real test not just conjecture.
Thank you Bruce. That's good feedback. I think the investment is worth it. The balancer seems to get the job done. Also see the next 2 videos in regards to that.
Some folks have had positive results by bottom balancing. I use that method on various setups. I use my setups in various capacities fractionally on boats and remotely. Jack Rickard produced some good material on the subject.
Thanks for taking the time to share all your hard work and journey.
I know, I have seen Jack's video and never agreed with him on that topic. All gear you buy these days is designed to balance battery at the top. Solar charge controllers, BMS, balancers... I really don't see the point or advantage of bottom balancing. How do you set the solar charge controller for a bottom balancing battery?
Great PES - product experience share - again, looking forward to when you repeat this with 2 batteries and 1 with a FET BMS - see how they handle things differently.
Yeah, that would be a great test!
@@OffGridGarageAustralia least your computer would keep running with FET based BMS ;)
I guess that the one advantage of a hybrid grid tie system is that all the switching and management happen on the fly. Since I started watching your vids Andy i've been paying a lot of attention to what my system is doing at the micro level. Today was a good example here in Scotland, overnight battery discharged to 84% SOC, it reached 100% SOC at 11 am, after that it would charge at 16 watts for 15 minutes then discharge at 10 watts for 5 minutes. I think this the built in BMS/balancer holding the cells at 53.10 volts per 15 cell module × 8, this cycle went on 'till 7.30 pm and the battery began supplying the house load. System has been running almost 3 years non stop. I'm waiting for a cable and software to arrive which will allow me to access the historical data stored in the 8 modules as the system info does not include the total battery cycles, SOC, SOH, max min cell voltages/current and through put. A lot of your vids would be helpful to owners of grid tied hybrid inverters with pre built LiFePO4 batteries in understanding what's going on in the background on a day to day basis. Thanks for all the information and entertainment bro.🙂👍🏴
Thanks so much for sharing these information and you experience.
Is this a Victron system you have installed over there or what is the charge controller's brand? Mine does not do this cycling when the battery is full. It keeps it constant until the solar cannot keep up with demand any more. It could be a setting in your controller, absorption/float close together maybe?
@@OffGridGarageAustralia System as follows, SolaX 3.6 kw hybrid inverter, 2 string MPPT, 50Amp charge/discharge, no float setting ability although it does ramp down charging over the last 30 minutes from 50 Amps gradually 'till it reaches 100% SOC, 18 Qcell 295 W G4 PV panels 5.3 kWp and 8 Pylontech US2000B 48 Volt 15 cell modules.
As I said after some initial teething issues after install in Oct 2018 the system has been running and all i've done is check the App to see that the system is up. On a decent Sumner day in June I get a max 32 kWh, Mid December a really good day is 3 kWh, Annual production is 4.2 MWh.
With the inverter handling the charge/discharge, battery management, grid export and house load management i've not had to do anything apart from switching between self use, where the only import is the small draw the inverter uses to monitor grid frequency/voltage which averages .5 kWh per day, from early February 'till early November when I switch to timed use where I can set timers on the inverter to top up the battery with 2-3 KWh of cheap rate energy (5p per kWh). After watching a few of your vids I got the confidence to actually look at the detailed inverter settings for max/min voltage etc and found that they were set well within the safe limits you've demonstrated, i'm just watching you take your test panels off your solar garage right now. Happy to answer any further questions you have if I can and thank you for taking the time to reply, love your vids mate.
Jimmy🙂👍🏴
Great video Andy! Very interesting, I love these sorts of experiments in your videos! So very interesting to see how it all works 🤔
Thank you very much for your feedback!
Hi from Philippines x Pat from USA Yu have answered lots of my questions thanks always watching and thanks by the way still no power since past Dec but it's coming slowly
100% right both of these work in concert together. also if you have more charge current than the balancer can handle then it WILL overpower it and cause problems. the bms is the safeguard (like the guard at the gate) and the balancer ( cops inside the city) helps it to be better. both are needed.
Yep, agreed. That is correct.
Mr Andy, I hate to say it but I think it’s time to consider a Batrium setup. You seem to be constantly chasing your tail here, and the active balancer doesn’t seem to have the extra balancing capacity we were all hoping for.
Ppl have nothing but praise for batrium cell balancing. Plus I know you love new hardware.
Fantastic channel, very much appreciate the content.
Man, you know, I'm testing, observing and exploring. I know this BMS is far away from being optimal as I am from Iceland.
The last time I did a bottom balancing with the pack and connected the balance. It balanced it over time and now fully charging the battery I was just curious what is going to happen.
Batrium, yeah, but how boring if it just works. It's like making videos about how charging a Tesla works 😬
Just parallel 250pcs of active balancers and you’ll be fine! 😎
Joke aside, the result was as expected and I’m glad you made a video of it. Some people seem to think that active balancers are the solution to all cell problems no matter the size of the battery bank and the charge/discharge currents. That’s obviously not the case. An Active balancer will only work in both ends if it can push on par with the charge/discharge current.
Only 1A of current transfer from the active balancer seems almost trivial compared to the 280Ah battery cells!
I believe that those things can be paralleled but purchase of two or more might not be the most cost effective option... I believe that that balancer would be better suited to your 5.1Ah cells... I'm sure that they can be designed to ramp up the current sooner and higher but the designer probably didn't imagine anyone using them on 280Ah cells, hopefully he's watching this channel and comes up with something that works better for this particular situation...
I can't wait to see the new video for the new balancer....
Great work Andy 👌...
Try with 17s lifepo4 and 57v absorb u hold cap/power (because aditional cell)and cells goes up to 3.35v so u are on the safe side if someting goes .....u still have in each battery room for additional electrons....
I have nearly the same setup as you have and observe exactly the same behavior. From 3.5V on the weakest cell starts running away and even a combination of the BMS' balancing and the balancer can't stop it. I often see 40A charge current. This is way too much to get rid of at just 200mV deviation between the cells. I guess reducing the charging voltage will help, but unfortunately I have one weak cell in the pack that causes the trouble while all the others are still ok. So the overall voltage in this situation is still decent. I already thought about buying a small capacity cell to put in parallel with this weak one in order to remove it as a bottleneck. One of these 5Ah cells you used for testing would be ideal. Maybe you want to try that. Would love to see it.😁
I think we will be OK, once we really top balance the pack and let the balancer do its thing. Charging to 3.4 or 3.45V is just not enough for that balancer to do its job correctly. There is only little difference in voltage and therefore only little balance current flowing AND at this lower voltage, it takes a long time to actually change the voltage.
Good video. I understand electronics very well. I am never an engineer but I can say that active balancer is great. I still wonder about 5 amps capability while 19+ amps charging current was in effect. I am still learning about what I am getting into with Lithium but I feel you either need a larger balancer or some of your cells are just way out of match anymore.
Hi Andy, you are right a BMS will always be required for these Lithium batteries! Active cell balancing circuits can only work while the BMS and chargers do NOT deliberately overwhelm them! If you are charging at 20 amps and the active cell balancing ccts can only transfer 1 - 5 amp the situation is hopeless and active balancing must fail!
I reckon you have the possibility to set your charging voltages and timing parameters for both bulk and absorption so that the chargers, BMS, and active cell balancing, can safely work effectively together. You will need to be kind to your battery by setting lower charge voltages and spend longer in absorption with lower absorption current.
It would be best for battery longevity and safety if you never trigger the BMS cell overvoltage shutdown protection yet still go through the full charging cycle including active balancing late in the absorption charging cycle. Setting lower charge voltages in the chargers and BMS should do the trick without too many adverse side effects.
Hi Andy,
I really love your channel. Want to build a similar battery set-up using 8 EVE 280Ah cells.
Some thoughts/observations:
If I remember correctly, you actually did a bottom balance with your battery, by giving the active balancer a lot lot of time to balance at the low voltage situation.
As you proved by the other experiment with the 5Ah cells, there's actually no balancing in the middle (flat) part of the LiFePO4 curve. So, maybe your battery pack could still be considerate "bottom balanced" the moment it was reaching the almost 100% SOC. So that sounds like double trouble: The battery doesn't get enough time to do the "top-balance", as the charge current coming from the SC is way too high AND it has been bottom balanced before.
The basic problem is of course a capacity mismatch between individual cells in your battery. In that situation you have to choose for either top or bottom balancing.
For Solar, typically top balancing is advised. Than, like other people (including yourself) suggested before, it makes sense to control your active balancer in such a way that it supports the top balancing only.
Hello Andy you are brilliant.. Small suggestion, what if you use battery bank without connecting to BMS & connect it to Victron inverter as a low voltage disconnect(let the inverter to take care of the low voltage related issues & sufficient way of draw out the power from batteries in the event of high voltage cells & enough time to balance the cells) & connect BMS to Charge controllers as a high voltage disconnect, high voltage/unbalance cell etc....You have to separate the Charge controllers & inverter then required to install an additional cable from your future battery rack to inverter...(Currently I'm using this type of setup to connect my CC, BMS, Active balancer, Inverter)
Good to see a real world BMS TEST on large capacity cells.
So far my 14S Leaf packs haven't deviated very much. But I'm being conservative with my upper and lower voltages. The real test will be when I build the 2P14S for the mower, and start pulling 6.5 to 16kW! (well within cell specs)
Just added another three of the $18 330W Trina panels. 2kW on my Victron easysolar now. Slowly but surely.
Thanks Andrew. We need to catch up when (crazy) times permit. I've got tons of questions about your mower project.
I am late to this thread but I use LTO cells and they can take 5c. The reason for using LTO cells is their cold tolerance and very fast charging. This would be a spring,fall and winter battery for my Ebike. I have used that balancer but I found out like you that the balancer can't keep up at high current charging on low AH packs. I considered using a bms with it but to make cleaner wiring, I could use an active bms. LTO cell voltage is lower and seems to affect the balancer current. I can't get the currents you get. This summer another pack will be built and testing at lower currents will be performed.
You're not too late 😉
I have yet to experience the LTO cells myself. But as I don't have any high current application in my setup, I probably wait until we convert the ride-on mower to electric...
@@OffGridGarageAustralia I have about 500 lbs of LTO cells that I bought at 1-2 dollars a lb. I have 4kw of solar panels in the shed to install and I have a Citicar to rebuild and upgrade to lithium. I need to stick to things that work. I have a box of balancers that have issues and bms's with issues. I need to keep weeding out the bad buys and move on! You have done a great job helping me out with your testing. Thanks again!
As some have mentioned you can use the vebus on the charge controller to back off the amps. You need a device maybe an Arduino between the comms on the BMS and the ve bus. Having a victron controller capable of doing this is a great advantage. The cheap Chinese ones I havendont have any data interface. Of course a diybms handles this better. Diybms has had some upgrades recently that make it work even better.
Even I'm working in IT, I would have no idea how to make this work. I know what you're saying and understand the principal of that but how to actually do that and how to does the SCC know to ramp down it's not clear to me.
@@OffGridGarageAustralia I just found out last night that diybms is getting victron integration. Keep an eye on this space. The parameters of the charge controller can be set via vebus. Just read the highest cell voltage and write the command to adjust charge setpoint to suit that moment. If done continually the BMS will steer the charge controller. Easily handled by an Arduino.
Andy, you need a back-up power supply UPS for the computer. Also, I’ve noticed the same tendency for a cell to run away at the top of the curve... ironically, when the voltage is high enough to trigger balancing. Was wondering if your BMS + balancer was able to keep it in check? I think when a certain threshold is met, the charging current needs to be reduced a lot in order to give the balancers a chance to do their job!
Great video. I'm a little confused. As of your research one has to distinguish between BMS with balancers and without it. Are there combined systems on the market? For a real "Batterie Management System" I would expect a balancer build in.
Thanks Peter. Yes, there is one model I know about. It's linked on my website and I have ordered it months ago but it seems to be lost... They have re-send it again last week...
I try to engineer my systems to be inherently fault tolerant, so even at mach charging the battery capacity sufficiently high to regulate it, thanks tot he size of Lithium polymer batteries I acquired that much will hold true still (Bit mind boggling I can dump my entire power systems worth of energy into them and only muster 0.25C charge rate!) but still deff putting on a BMS with generous safety margins!
Thank you! 😊 very much appreciated.
Thanks for your feedback!😉
Thanks Andy. Love your work 👍
You are the best! Just yesterday I charged the battery bank up to 3.5 volts per cell and I watched the voltages on the cells. I mention that I made two series of 16 cells, each with its balancer, then the two series connected in parallel .The balancers are the active,2 amps with bluetooth, I think they are the same as the ones you will test in the next video. The trigger threshold for balance is set at 20 mA, and during the absorption, which lasted about 10 minutes, the deviation was not higher than 100 mV, at a cell average of about 3.5 volts. I think it's ok, or 100 mV is too much?I noticed that the gap between the cells increases with increasing charging current.. I was charge the whole bench with approx. 80 amps when the deviation was 100 mA.
It makes sense that balancer can not do nothing unless cell balancing current equal or higher than charge current in this case about 19A (not reasonable).
As for charging to only 3.4V instead of 3.5 or 3.6 will do nothing to reduce degradation. In fact absorption and float charging to 3.4V will be worse than just charging to 3.6V and then stop the charging until SOC drops enough to re-enable charging.
You can imagine electrons as marbles that will fill some gaps in the Anode and to many of them will produce the damage but you will get the same amount or more if you charge to 3.4V slowly reducing the current (absorption) and then keeping there float than you will just charge with full available current to 3.6V then completely stop the charging.
A BMS like the one you have tat fully disconnects the battery from both charge sources and Loads is not that useful. Ideally the BMS will be able to only turn OFF the charger but allow the load to be powered from battery.
And the Victron MPPT chargers are almost the only ones not damaged from battery being removed while PV panels are still connected most others will not handle that.
Thanks for your feedback. Will Prose made a video a while back and he tried to 'break' several SCC when disconnecting the battery while leaving solar connected. They all survived. Maybe they all are getting better with this.
I 100% agree on the BMS. It's not good to disconnect everything in such situations. The load should stay connected to help discharge the battery if one cells goes over.
How's my 48V Electrodacus project going. Keen to get one... 😉
@@OffGridGarageAustralia Maybe it depends on how long the PV cables are and what sort of protection they have on the input. Main problem will be regulation on some SSC can not react fast enough when battery is disconnected.
There will be no 48V SBMS version as I only do projects in compliance with the ultra low voltage standard usually 60V to 75V DC depending on where in the world.
While 48V battery is below that the PV array will require at least 2x 60 cell solar panels in series and that is already >80V DC open circuit so above even European 75V standard and well above the 60V in many other places.
Thank you Andy!
Thanks for watching
Setting to 3.4 for a longer time would have been better. Having a diversion load set point of 3.4 is what I have decided on. Using a solid-state relay would get rid of that click and be more reliable in the future. You have a lot of panels, unfortunately, they get shaded. Maybe the match of panels to the balancer isn't right. I came here to learn about using an active balancer. This does make me consider using a BMS. interesting video.
Hi Ron, you always want to use a BMS nit just a balancer. These are totally different things. A BMS is a must as it protects your battery by monitoring each cell individually and disconnects the battery if one goes out of specs. A balancer will not protect your battery at all and can cause a fatal run-away of one or more cells.
@@OffGridGarageAustralia Thanks, I got that from watching your video. It makes you wonder how many batteries are ruined by not having a BMS. I did watch a video on YT of a guy with a really large battery bank who never used a BMS. I was thinking that an active balancer and the charge controller would be enough, but I guess not.
Ever thought of a slight tilt to the North for your panels, for better winter performance, as you probably don't need the max performance in the summer.
It's a lot of effort and not really worth it as I have shading in winter anyway. My production has increased by 120% in the last 5 weeks since we have past solstice. It's only in the 2-3 months late autumn and winter when production is down.
8:18 "... higher voltage more stress ..."
I agree. There is no sense squeezing 1% more, just to lessen their lifespan by more than 1%. not worth it.
It's a tradeoff, either you get a balancer that can draw more amps, or you lower the charge voltage.
The balancer actually only works with higher charge voltage. Everything under 3.6V, the balancer really takes too long. It just cannot handle it and react fast enough.
Looks like you put in a micro “ flux capacitor “ at the very end of the video . This is a great video .🤙☝🏾✌️👍🏿
Thank you very much. Yeah, just some little coolers for the Pi to help him cool down.
@@OffGridGarageAustralia don't forget that in vertical position they eventually may fall off after a couple of months/years (my experience)
Hi, Andy. I am enjoying your videos from Japan! I am also using this BMS. In my opinion, if the input current is too large for the active balancer, how about adding several balancers parallelly? Because one can manage 0.9A, 5 balancers can manage 4.5A!
Hello and thanks for watching from Japan!
You can parallel several balancers, that's not a problem form a technical point of view. But, is it necessary? This one balancer can already handle 5A just fine and has no trouble balancing the 280A battery pack.
I mean, if your battery pack needs several of these balancers to stop cells from running away, there is maybe something else wrong?
Maybe that is why the default absorption time is very long like 2 hours, so during this long time the voltage is kept very low and very close to the actual battery voltage, and the current gets smaller and smaller to give you the time and chance to balancing?
Hi, I think again about what I commented some videos ago. Could we use absortion at 3.40v and equalisation at 3.50v. In equalisation mode we can choose maximum current, so we can configure a very low charge current , so bms or balancer can make its job
With 25a charge current ot is impossible.
Note: i am not doing equalisation to a lifepo4, i am saying to use this setting to use as our convenience to give some hours to bms/balancer to balance the cells. Even we could choose not to do " this fake equalisation" every time battery reach 100%.
We need to low charge current when battery is above 3.4 or 3.45v
Lowering the voltage will probably just delay the spike up from a high charged cell. So if you set it to 3.4v per cell a couple cells will still hit 3.65v just at a later time like the next day instead of today. If you have excess solar the batteries will always get to 100 percent charge.
but the current power depends on the sun hours per day, so you can set it to max in the winter and reduce it in the summer, or hang more consumers on it, such as a swimming pool pump andy ;)
@@longdistancerider691 normally inverters have an option to configure max charge current during equalisation phase.
Yes, that would be an option. Just not sure if it is enough to do this balancing at higher voltage once a month or if it is already too late then as other say. I guess, I will find out. 😉
@@OffGridGarageAustralia once a month, once a week, ... who knows ? 🤣
But we know that we can use balancing at top voltage with high charge current. It will be better that bms tell to inverter... "hi, please, reduce charge current i have a cells above 3.5v and I need to balance it". But I dont know if that is possible with diy batteries.
Good morning! I didn't catch which balancer you have coming in. This is what I'm coming up with as well. As I've been scaling up my design, I am finding that I will need higher and higher active balancing currents. Maybe a bank of MOSFETS operating with a micro controller. It just seems that one active balancer does not fit all.
Hello, I've shown the balancer in pervious videos and tested it many times.
The problem here is that the charge voltage is not high enough for the balancer to get the job done. The voltage difference is too low. Even the balancer could do 5A, it will only push 0.5A to the other cells. And this just takes far too long for these high capacity cells.
@@OffGridGarageAustralia Ah. I thought you said you had a new one comig in, with bluetooth. I get what you're saying about the differential voltage sensitivity correlating with the amps. I'll look in to my bag of tricks and see if I can cobb something to test with more sensitivity. Thank you Andy.
This is why you should get a BMS like the Orion or REC so it can communicate with the Victron MPPTs. You can slow down your charge current if there is an imbalance till it has been rectified.
Will it work with the Pi though? I'm not keen to buy a GX device on top of that...
@@OffGridGarageAustralia If you can get CAN Bus to work yes, probably not worth the hassle. Just get a Cerbo, worth the investment as it has 2 CAN circuits and 3 VE. direct ports. 2 Relay's for automation.
LOL, whenever to decide to charge the cells into the Knee above 3.500 you can expect HVD's to occur, most especially with Bulk Commodity Cells they can start to run above 3.400. To blame it on anything else is quite silly Honestly. Remember that above 3.500 and below 2.850 are the upper & lower knees, while Standard LFP Working Voltage IS 3.000-3.400 (which is the 100% Working Capacity)
For me it is quite normal, that an active balancer with max. 5A can never prevent a cell from being overcharged, if the charging current is about 20A. Even if it works with its max. 5A (which it can only do by higher voltage deviation), there are still about 15A going into this cell. The only way the balancer could prevent a cell from beeing overcharged is when the ingoing charging current is lower than the (real not the max.) balancing current. But the cutoff if a cell goes up in voltage can never be done by a balancer. That's why I will never use a Lithium - Ion / FePO4 Battery without a BMS.
And about the BMS with the relay. Well, it does its job, but I don't like it. I'm looking forward to see the new BMS and how it works.
That is exactly right, Gunther. I have shown this in previous videos and explained that as well.
As long as we charge that high, it will charge all cells with the same current and the 0.5A you discharge one cell with, don't make a difference at all.
I don't like the BMS either. If the new one would actually come...
Yes, your pack is not easily balanced, because the cells are not from the same batch. The active balancer seems to work fine, but can not really cope with this pack. If you charge the battery to e.g. 55V, let it absorb for an hour, then go to e.g. 55.5V, let it absorb for an hour and then finally go to 56V. That would maybe give the active balancer some time to properly do some kind of top balancing.
9:28 "... far away from the 5A the balancers can do..."
The current is proportional to the voltage difference between adjacent cells; and the one you are using is rated at ~1A per 0.1V. You get the 5A if the difference is >0.5V but only between adjacent cells.
14:11 I am not sure if it is a good idea to rearrange the cells, alternate strong and weak cells. This way, adjacent cells will always be the ones with the highest voltage deviation to make the Balancer work faster. (In your case swap #4 with #6)
The ultimate goal of any BMS is to manage the state of charge (SOC). But the only tool you have is cell voltage. SOC can be ESTIMATED with cell voltage. But there are several factors that make it only an estimate. Are we charging or discharging? How long has the cell been idle? Is the temperature elevated etc.?
No, the ultimate goal of a BMS is to be the last line of defense when something goes wrong with the individual cells (over/under voltage, temperature, etc.). Accurate SOC measurement is done with a shunt.
The BMS should be your last point of disconnection if everything else fails (solar charge controllers, balancer,...). And that what this one does. Not more though!
Hi Andy
Been catch up on you channel, IV just ordered my first lifepo4 cells to make a 12v 600ah unit. Looks lll be looking after a nuclear reactor not a battery bank........ha ha,
Love the Chanel really like you film your ups and downs.
Great work many thanks
Marc up top......lol
Thank you Marc. Much appreciated. Yeah, sharing a lot of content and findings here. Most channels do reviews or building a battery in a 7min video. Not much to learn from that. It's just a different style, I guess.
@@OffGridGarageAustralia
Thanks for your reply, I guess it's a big thing for us all.....
We all learn by our mistakes but also learn as we make progress.
Really like the way you show both your mistakes but also your wins, it's also great that so many people write in to which will all learn off too.
Nice work, top channel ⭐⭐⭐⭐⭐
Intéressant !
Super video!! I was thinking is there not a function in the charger that will lower the charge current as you reach max voltage?
Thank you. Yes, there is a function in each solar charge controller. The problem here is with the BMS and its relay. Once the BMS turns off it disconnects the battery from the controller so it has no idea about the battery voltage any more. So this BMS is less than ideal...
Hi Andy interesting again. I agree on having a bms always. But I have built a lifepo4 starter battery for my camper van. It does not have a bms, because I am afraid that someday it could fail, and I cannot start the car any more. I have fitted a 500ma passive balancer. Regards, Rijk from Holland.
Good morning brother from Philippines 🇵🇭
Hello there!
You could add a second heltec, I cannot think of anyway it could cause problems and would double the balance amps?
I was thinking the same. Julian Ilett did a video trial on this recently.
The active balancers transfer power to the next cell in line, so it is better to connect the weakest cell (5) next to the strongest cell(6) in your pack.
NB: Stronger cell should be closer to the pack's negative terminal
Today max charging current was?
Dont panic
185A
I hope I never suggested a balancer alone is OK. Obviously it is not.
As far as it keeping up with those 280 cells, once you get back to cycling them into your car, I think they will eventually match up at the top. Unless there is a problem with the cell itself. :)
And this is the problem, the balancer is not fast enough to keep these large capacity cell in balance unless you go beyond 3.55-3.6V. The it makes a difference.
My batts should arrive soon looking forward to not having lead acid batts anymore and the mess they make
Hey great, you will love all the tinkering with them 😂
@@OffGridGarageAustralia Just bought another 8 280 ah so I'll have 16 minimum for my 24V Victron Quattro 8 KvA inverter
If you have a BMS and a balancer together would they’ll be some kind of a conflict between both because the BMS should balance your cells but then your active balancer also is balancing the cells why do you need an active balancer if the BMS does the same function or was I wrong on the bms
Thanks for great videos
They work perfectly together without any problem. Both will work in the same direction of lowering the voltages of higher cells. Some people even use several active balancers in parallel.
Andy, did you get the RS485 to usb converter with the bms. This bms can connect to the Victron Venus unit. We got guys this side running this bms with the Venus setup.
I'm just thinking out aloud here. It sounds/looks like that the balancer needs to have 125% capability of BMS. This would allow the balancer to to remove the charge amps from the over charged cell and also remove amps and place into another cell. How about putting 2, 3 or 4 balancers in parallel if you can't locate a higher capacity balancer.
Date: Sat Aug 14, 21
Place: San Francisco
I loved your vids and would love to tell you of some of my solar encounters including the old voltage eyebrows. awaiting your reply.
John
Thanks John. Just share your stories here in the comment section. 😉
I first got into Solar way back in 1963 when I was in Middle school.
Back then it was 1500-dollars a watt.
And my only source of Solar was International Rectifier who sold small solar panels (4.5-Volt & 9-Volt) for small transistor radios.
later on I got my cells from a army/navy Surplus Store.
These cells were from Hoffman solar who made small helmet mounted solar panels for some of our troops in Vietnam.
Of course the light would bounce off the panels and give away their position.
And so those now useless cells made their way into Army Navy Surplus stores.
Four 32-ma .4-volt sold for 1-dollar...great deal for a kid like me.
More to my story if you want to hear it.
John
This is getting way too confusing. From what I understand a BMS will not balance a battery's cells until the 12V battery voltage reaches 14.4 volts (3.6 volts/cell). So, from what you say we should charge a cell at 3.45, which if multiplied by four, is a charge of 13.80 for a 12V battery. But, this voltage will not activate the balancer in the BMS. So, should we set the charger to 13.8 which will charge the battery at however many amps the battery can receive? The battery is fully charged when the amps drop to .1 or 0 ? However, the cells may be unbalance with some cells exceeding the maximum recommended charge? By the way, how do we really know that charging a battery at a voltage higher than 4.o volts will actually harm it?
Well, that totally depends on your balancer or BMS you're using. Most devices can be programmed and you can change the balance start voltage accordingly to your needs. Some older, cheaper ones have a fixed start voltage for balancing. Then you have to surpass this voltage to actually start balancing.
Yes, you're right, the pack is totally unbalanced at the moment due to the bottom balancing I did when I installed the balancer a few weeks back. I top balanced it again last night. Hope it works now and the balancer can keep up with much smaller differences.
Nice video as usual. Apparently these 280AH cells are too large for this balancer to regulate at high voltage. I learnt these balancers can be paralleled though and it will be nice if you can test that as well. Perhaps having relatively weak cells in parallel with stronger cells (say three 100AH cells) to form 3P16S pack of 48V300AH will be better for balancing than this huge 1P16S 48V280AH system. Cost and ease of assembly/monitoring is another matter though.
Maybe he needs a couple more active balancers. Problem solved!
I think his running voltage settings 🤔 are much lower then these testing voltage ⚡ settings. So, in his real world setup he can set his per cell maximum voltage release time much longer, giving time for more transfer of Equalizing Balancing to happen. He had a hard wall because of his aggressive charging voltage for a quicker charge as he mentioned in the beginning.
@@xr680r What full pack voltage do you recommend to set that will eliminate a cell from drifting to 3.65v and tripping the relay?
@@uhjyuff2095 I am running a Outback charge controller and am going to set it to 55 volts for a beginning setup. My BMS to 3.45 per cell and play around with swapping batteries between the two banks till I figure out what makes them happy. The only thing that my setup is going to make this more difficult is each box 📦 for each bank is climate controlled by a ducts and compression. So, I have to figure out if I want to fight all that 🤔 and for how long before I lose interest. I have both a Smart 300a BMS and the same active Equalizing Balancer he has in this video for each bank. My goal is to build this and have fun with it till I forget about it and it just works magically for twenty years. It's replacing two banks of 8 Trojan 105ah batteries that have been running for six years. 😳 I don't want to maintain lead acid anymore. Oh, my setup in in a 42 foot Fifth wheel trailer with a verified 3600 watts of output from the 4000 watts of panels. I have tested to a constant 60 amps of charge into the lead acid batteries for over 7 hours in one sunny 🌞 southern California day 😎, running both my house and a neighbors during the fires for an extended extreme testing. Actually discoloration in the panels around the sodler connections!
I think it's the other way around: the voltage is not high enough for the balancer to make a difference. At 3.4V the capacity difference between similar cells is a lot larger than at 3.55V. So at higher voltage the balancer has an easier job to balance successfully.
A separate port bms will cutoff charge when a cell voltage is to high and continue to allow discharge at the same time.
I wonder what the capacity difference is between the high and low cells. Hopefully not more than 5aH
We will find out once the other pack is build and in place.
Excellent video what bms are you using and where did you buy it
I think you need a lower Recover Voltage for the high Voltage disconect.
It is set to 3.45V in the BMS. The problem (or feature) is though as soon as the BMS detects load it reconnects the relay again which also starts the charging process. Stupid to develop such a BMS!
@@OffGridGarageAustralia BMS has to do that. Battery is overvoltage disconnected and suddenly some load comes on. BMS has to close relay to allow discharging. If it does not, you will have whole house powered off due to battery beeing full untill you manually re-enable BMS
Hi Andy, Love your channel. is your solar controller absorption set the same as the BMS or just below so you don't get the over voltage and shutting it down
Thank Glen. The BMS kicks in above the set absorption voltage of the SCC. The balancer works all the time though.
Hi Andy, nice video as always. In your video I saw on your Victron display a DC load from battery (not going to the inverter). How do you measure that? What instrument is used for that?
I have heard that stop discharging and start charging at 3.0V and stop charging at 3.4V
It is supposed to increase life and supposedly no need of balancer OR BMS. However,
cells need to be bottom balanced. Can you try it and show us. How do you bottom balance accurately?
Hello If i understood correctly that balancer will switch off charging and by that it will switch off/disconnect battery from system..If that is the case i would be worried..Because first of all those chargers don't like to be done that way..If i may suggest switch off charging or inverter they usually have that one contact that will switch them off ..That is the proper way how to switch them off and not by disconnecting power from DC side ..
Hello, I showed what happens when the battery is disconnected from the SCC.
@@OffGridGarageAustralia Yes my friend I know . It does work .but i do suggest not to do it like that it wasn't designed like that even it would work
Interesting Andy, do you think a 5A or 10A balancer would be enough to not get a renegade cell?
It's not about the amps, Jonas. As we could see from this test, the balancer is pushing only 0.3-1A across the cells at this voltage difference. Over time it will balance. The problem with mine is, that I rarely fully charge it. It was like 3 months since I fully charged it the last time. That's why I had a greater drift across the cells (and because of the bottom balancing I did before). Now after fully charging it over some days, all is in balance now...
@@OffGridGarageAustralia Hi Andy, thank you for the reply, what you mention is in the end when pack already is close to full, but during the charging when battery is 50, 60 70 and 80% full your active balancer probably can shuffle 2A between cells, and if that re-shuffle would have been 10A (before it is to late) the pack maybe would have been a little more balanced before reaching the charing-level where the balancr cant move more than +0.5A between cells? I'm just asking and have no idea if it is true, but I hope it might be :)
Great video & info as usual Andy. Again, I would like to see what a Big Cap Bank would do across ur busbars, could this help in a number of ways with absorbing/filtering surges & spikes. Also I wonder if the controllers would like this extra dump for ramping up & down as your balancer switches on & off the charging load, + it would also keep ur inverter &/or computer going during shortfall times, while ur Mppt controllers can still be pumping fairly full amps to caps for a short duration, maybe they will ramp down & then back up as inverter uses excess & power ready for when balancer kicks back in. Cheers n thanks, really good info.
Still loving your video's
Putting all cells in parallel and top balance letting them equalize is not so good way to balance I think. Its almost the same as using an active cell balancer. Unless you let the cells sit for a very long time in high voltage the cells will be badly balanced.
I think best way to balance is to charge each cell separately while it is already assembled in 16s configuration. Using bench supply charge one cell at a time to 3.6 volt and let it absorb until x amount of amps then stop. X amps absorb current being equal for all cells.
That does not make a difference. Parallel balancing them at 3.6V does not take long, maybe a few hours, that's enough and you can connect them in series afterwards. The balancer of the BMS witll do the rest if necessary. If you charge them individually, they loose voltage again after disconnecting the power supply.
If the trees shading your PV setup are yours, why don't you have them topped. That would be good for the trees and would make them bush out lower, and your PV set up would have more sun. I wouldn't take out the trees completely. I would trim them down low enough to help with passive solar and allow more sun on the panels. This, of course, depends on the type of climate you live in. Here in North America, we position deciduous to shade in the summer and let the sun in during the winter. It's all a balance you know.
Because they are 40m high... If you ant to come over and take off some selected branches, I've got some cold beer for you 🍺
Shading makes it interesting. Like cheap BMS' or batteries. If you have everything optimal who would watch that? Everyone know it just works.
hello thank you 😃
Active balancing is the worst. Just take care lifepo4 (as example) have the same voltage above 3.4V. No cell should run away to 4V and start the cooking process
If I was the Balancer I would be a bit cross and sue for defamation and cruelty. I mean you used it to bottom balance the batteries which will completely throw the top balance. You then have a cutoff in the flat part of the charge curve that gives it no opportunity to try and balance and you then give them loads of amps and get them into the steeper part of the curve and then wonder why the poor little balancer can’t keep up. I mean how can it when you are feeding 10s of amps and the time it gets to balance is only a tiny fraction of capacity and it can only balance at a limited current which has to decrease as they get closer together. It is designed to maintain the balance, not perform it. To do it it needs to be allowed to regularly see the difference so it can pull the voltages together again. These batteries are not perfectly matched; at the price you wouldn’t expect them to be, and so the balance will drift and the longer you let it drift without allowing the balancer to have a go the harder it will have to work.
I have the same cells and the same balancer and I set the cutoff voltage to 3.5v/cell as that takes it far enough into the ‘knee’ for the balancer to do its job. As I don’t run them low this works well as it doesn’t try and bottom balance and they stay within about 10mv but the balancer does have to work. The first full charge of the day will often see a bit of drift but by the time a second might happen they are all back together and friends again. I think only having the one relay is a mistake for reasons that are obvious when your computer gave up.
Great comment Phil. That was exactly what I wanted to see. I did the initial bottom balancing with the active balancer connected and it did a really good job. Then there was no activity in the middle part and at the top now, there is no time for the balancer to do it's job due tot he high capacity cells and not enough current. I have top balanced the pack again last night and re-connected the balancer. Let's see what happens over the next few days when it fully charges again.
3.5V is probably the way to go for the balancer to make a difference. 3.4V what I have set is still too low and the difference in capacity is too large at this voltage while the difference in voltage is not enough for the balancer to work in time.
I also advise you to use the BMS balancing function in tandem to the active cell balancer.
I did, makes no difference either... just not enough amps for these high capacity cells...
yes always use a BMS, something may fail and you dont want things to get damaged.
I wonder if you could put multple balancers on there? Not sure that would work...I got one of those same balancers since I had one parallel pack of A123 cells charging faster than the others, and the active balancer is working perfect, but it is only a 60ah battery though so the 5A balancer works fine. Keeps the battery ballanced to under 10mv delta all the way up to 3.6v per cell pack.
Yes, you can and it would work better but still not optimal. The balancer needs time with these larger cells. At 3.4V it is just not enough difference in voltage to push some Ah from cell to cell. I'll try and charge to higher voltage and see if it works better.
Hey Andy, I came across something surprising today... And if possible I'd like you to run a test.
I had a Google mini plugged in to a 4 way surge protector extension, via my victron inverter. I also had a mains watt meter plugged in the socket. I noticed the mini's power factor was absolutely awful, PF0.04. I also noticed though the meter was showing 4/5watts only, the amps did not add up 0.245a at 245v 🤔 60watts.
I decided to unplug the surge protecting extension and put the mini direct to the meter. Now a PF of 0.8, and 0.015a at 245v.
It would appear the surge protection was wasting 60watts!!!! I'm now suspicious I've been wasting a lot of energy 😔
Can you run the same test but also use a shunt/DC meter to prove this missing 60watts/consumption
Wouldn't have that problem if the BMS could reduce the PPTs' output voltage to whatever is needed to keep the highest cell voltage under control. Then the inverter would always have uninterrupted access to total solar+battery power since individual cells would never be allowed to reach the trip point and force a disconnect.
Rec bms does that, it talks to victron and lower the limits to keep voltage lower than setpoint
Yes, it highlights again the the QUCC BMS with relay is very sub-optimal. Even it has an RS485 port, I'm nto sure if it could communicate with the SCC to slow down charging.
@@OffGridGarageAustralia Even if you did have full voltage control, you'd still want a relay to disconnect the battery as a last resort in case something fails and lets the (dis)charging go out of control. Even if you had a FET-based BMS, you would probably still want a relay as an OH-SHIT backup in case the FETs short out.
Yep. BMS prevents cells from “running amuck”! 😂
I have a Tesla module that has a delta of 0.42V between highest and lowest cells. Do you think this balancer would bring these back in line? Many thanks
All of my diy lithium ion battery now have Ballance leads as well as bms so I can check them out and manual balance if needed but since these 23ah packs are so small there is no Bluetooth connection for them so thinking about making a jumper so I can add one of my large 100 amp bms with Bluetooth so I can get a faster balance and see what's going on anytime till the balance is done otherwise I have to separate one row at a time and charge to equal the highest cell row and that's somewhat time consuming . Do you think this is a reasonable way to get the job done
I'm starting to think that a balancer should not just kick in at the top end of SOC... but also at the bottom end. I'm taking advantage of all your testing and purchasing only JK brand BMSs for my batteries... so top end balancing will be configured. But this has me thinking I should buy a separate balancer to kick in at night for low end balancing if the SOC gets low. Unless there a way to configure a single JK BMS to both top AND bottom balance the battery. Does anyone know if that is configurable?
I have questions about the size of the active balancer… that is if a 5 amp balancer is the right size of a 100 AH set of batteries than what size balancer do you need to have for a 280 AH battery set… how about as the amp hour ratings climb to say 450 amp hours??
3500 cycles once per day is nearly 12 years...... I dont think its worth stressing over the "stress" a higher voltage causes. We need to balance in a voltage area where there is only 1ah difference in SOC, then the balancer + BMS can deal with things within an hour or so....... And hopefully hit "float" on an almost daily basis to keep the gap/difference small, so that the small balance currents can keep on top of things.
Im ready to buy 16 cells, and I think vs all of your hassle of BMS on, off, on off hitting overvoltage, I would rather power down the inverter and give things an hour to balance.... Worst case scenario of course!
Yes absolutely correct. The balancer will not work at lower voltages and need this 3.55/3.6V for balancing.
A few videos ago, I bottom balanced the pack with the balancer and now back at the top, it has messed things up.
Also, the BMS works sub-optimal with the relay. It's just not good.
Battery hoarders support group are on the phone!
Hahaha, yeah, I know!
I dont use BMS only active balancer....but probably i will get one daly is on my list still but i dont know is expensive....
Allowing the balancer to run PERMANENTLY will only make a mess of the balance. It should run ONLY during the CV phase (let's call it Absorption), for one or two hours.
The CV voltage should be at least 3.5 V * 16.
It is not necessary to charge to this higher voltage EVERY DAY...maybe only once per month.
Perhaps you could use the Equalization settings of the chargers for this purpose.
it is nice to see that i am not the only one who cqannot throw old computers away. all my old towers i have dumped windows and i am running linux in preparation for the end of the world .lol haha.
Hahaha, the apocalypse is coming. The old DELL, is actually a descend computer with i5, 8GB RAM and 510SSD, 12 USB ports and uses only 50W when running. Ideal for my workshop and the EB Tester Software.
Maybe use 2-3 active balancers? :)
Hah!....I was going to suggest 100 but that would $6K. Guess we're all stuck with Ohms Law. ;)
boss please can you do a video on how to charge battery with BMS and bleed them down to equal voltage without using bench battery charger?
What would happen if the BMS balancer was on too would it fight the one another.
That's scary when one runs away from the rest. I sit along side worried my investment will be damaged even though I know the bms will disconnect and thing will be just fine AND the fact I can see everything on my phone as well just the fact is enough to make me want to be right there, just to be sure haha. Guess that makes me a battery nut, oh well
@@peterking1134 I'm not reporting any issues and it's never to the point of damaging but cells with different internal resistance will usually act differently than the rest. Not all of us can afford cells from the same manufacturing group, those differences will often show when pulling capacity tests. Not many circumstances are the same as others.
And I too use a balancer along side a bms and it works fine. Just relating to what off grid garage was commenting on
I think that was because I did a full bottom balancing a few weeks back with the balancer connected. Cells were all over the place. I fixed that last night again and have restored the top balancing again.
@@OffGridGarageAustralia your prob right, I learned it totally depends on the situation. Top balanced the lithium in the car and bottom balanced the lithium in the garage. Bottom balancing in the car was a lesson learned for sure. No one's an expert, some know more than others but were learning more every day. Thanks for the great, informative and positive content.
Should not the batteries be hooked up in parallel to do the balance job?
V would remain a 3.x volts, and will they not balance automatically, equalizing without any sort of balancing gadget?
At the beginning before building a battery, yes. But now, it a system in use.
@@OffGridGarageAustralia Wonderful that you took the time to answer and are helping me tremendously to cement my understanding of a solar system... Pardon my ignorance, I am a complete noob, I am looking to understand everything theory wise before I even plunge my money in this potential abyss.
ok, I agree, I suspected that part on the 'system in use', the series setup...
So, in a system in use in series, what if a smart BMS/charger/discharger(discharger should not be needed) with the proper wiring and diodes(or-what-not), would isolate the culprit batteries only on the charging end where current is allowed to flow out but not in, not cutting off the rest of the system!?
...and I am scratching my head how that would be accomplished electronically .
Would that not be a better approach? ...am I pushing too far and want my cake and eat it too!?
Though my knowledge of electrical dynamics is only a basic understanding of Ohm's law, I am also thinking such configuration already exists in the electronic world, just a matter to apply it here.
hmmm! after pondering this, I am wondering if that is exactly what the balancer does? ...thus that is the circuit that already exists, although not exactly at the same point in the flow!
No usable Power from a full battery proves the bms is of no help. Which modell will be the new bms?
Is it possible to connect more than once active balancers ?