What you REALLY charge your battery to every day!
ฝัง
- เผยแพร่เมื่อ 31 ส.ค. 2021
- This is the start to an interesting series of video. It will show us the true state of charge (SOC) of our batteries. The method is called Coulomb Count and... it counts Ah going in and coming out of the battery to determine the state of charge.
For this reason I fully charge the battery to 3.65V/cell and then reset the smart shunt to 100% SOC. Form now on, it will show us the real SOC when we charge/discharge the battery.
You will realise, that charging to 90% or even 95% SOC is impossible with LiFePO4.
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-garage.com
The 'buy me a beer 🍻' link is here (on the website, scroll all the way down):
off-grid-garage.com 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!
**** These are the batteries you are looking for ****
🔋➡EVE LF280 LiFePO4 cells 3.2V/280Ah: off-grid-garage.com/batteries/
🌞➡Solar, Controllers, Inverters and accessories: off-grid-garage.com/solar/
🔧➡Tools, Analysers and Testers: off-grid-garage.com/equipment/
📈➡ Battery Tests and Results: off-grid-garage.com/battery-d...
🔌➡Batteries, charger and accessories: off-grid-garage.com/batteries-3/
⚡➡Cables, Inverters and Electronics: off-grid-garage.com/electronics/
Get 5% off when buying Bluetti all-in-one Power Storage Products!
off-grid-garage.com/bluetti/
New to AliExpress? Use my referral code below to register and earn US$19 store credit. What a great start! a.aliexpress.com/_mrtR9Ux
Take a live sneak peak at the Off-Grid-Garage in the Victron VRM World:
vrm.victronenergy.com/install...
My Tesla referral code (free 1500 SuperCharger km / 1000mi for you and me)
ts.la/stefan96963 - วิทยาศาสตร์และเทคโนโลยี
I think it's important to remind people that there's no point in aiming for a percentage. The only reason people want to stop charging at 80% or 90% is to extend the life of their battery. But it's a VERY common misconception that SoC is what drives battery degradation. However SoC is *not* what drives cell degradation, it's *voltage*.
The "Stop charging at 80/90% rule" comes from the EV world which primarily uses NCA or NMC chemistries and where cell voltage is not usually displayed to the driver. Those chemistries don't have the same "flatness" to their voltage curve so their voltage does rise somewhat as their SoC increases. So for NCA and NMC type cells it makes sense to aim for 80-90% because that's where the voltage starts to get "high" and where degradation starts to accelerate. Since the actual cell voltages are not easily accessible to the driver and since the SoC percentage is the only metric available to an EV driver to have any gauge of an EVs cell voltages, that's where SoC became the basis to guide these decisions to prolong battery life.
However LFP is a different chemistry. Those same rules that apply to NCA/NMC don't work the same way here. With LFP the voltage is almost perfectly flat until around 98% or 99%. There's no significant difference in cell degradation with LFP between 80% and 97% because the voltage is almost exactly the same. It's only when the voltage starts to increase that the degradation of the cell starts to accelerate.
So there's just no point in trying to stop at "90%" with LFP. Coulomb counters drift over time and can only synchronize at very high or very low states of charge with LFP because of it's very flat voltage curve. If you try to stop at 90% at the top and 10% at the bottom using coulomb counting, over time that is going to get very out of sync with the batteries actual state of charge. (You might think you're charged to 90% but you may only be at 70%, for example) It's even worse with the Victron Shunt because it *only* syncs at 100% SoC (It doesn't sync at low voltage) so your SoC would be even more wildly inaccurate over time since the occasional low voltage disconnect won't reset the Victron Shunt back to 0%.
It's much easier to just stop at 3.4v instead. You'll be at 98-99% and will experience no more degradation than if you stopped at 90%.
LifePO4 is the best battery because it has no degradation above 80 percent like NMC cells do!
Thanks for pointing this out again Jeremy. So many comments under my videos still wanting to follow this 80-90% rule. As soon as the voltage starts rising, the cells are already at 99% SOC anyway. That's the point to stop charging then if not earlier already as you said.
Yes, the shunt will be out of sync as it will not recalibrate until I go to 3.65V again. With this and following tests I want to show that it is impossible to hit 90% SOC with LFP cells. Charging to only 3.4V and then play around with the absorption time is probably the way to go.
Indeed, I have found if you let tail current drop under an amp at 3.4v, you are around 98 to 99%. Seriously.
I was chasing the 90% rule with my LiFePo4 pack for the first couple years after watching a lot of these and other TH-cam videos I started experimenting and ended up at 3.55V Bulk / 3.50V Float.... I might end up with a small amount of degradation over time but it is worth it I seem to have picked up about 10% which really helps on the cloudy days 👍 I am also off-grid with 2 teenagers so every watt counts 😋
It's exactly the opposite. LiFePo4 can tolerate up to 4 volts with no degradation. At 4.2 volts, the electrolyte begins to break down.
Charging to a high state of charge causes degradation, even if it's at 3.45V. Just one example, during the charging process, the intercalation of the lithium ions into the graphite causes physical damage over time.
It is true that LiFePo4 is more resilient than NCA or NMC. Coincidentally, the "fully charged" voltage of NCA and NMC is dangerously close to the maximum value that the electrolyte can handle (4.2 V).
Congratulation: 14000 subscribers Andy! Go on with your style, jokes, critical reviews and help for others who need your support.👌
Oh, wow, yes... 🤦♂️ I missed that somehow. Thank you and thanks everyone for subscribing...
Yes! And 2 years later...over 70,000!👍
ANDY!! How is it possible that you have not been offered a contract to do a network TV show with this riveting resetting of parameters?? I do not understand it, lol. I don’t see anywhere else this quality of explanation of how LFP chemistry works in the real world - THANK YOU. Tom, solar guy in US.
Thanks a lot Tom! I think TH-cam is the TV we all get now. My garage is my studio, I'm the director, writer and actor. It can't get cheaper 😂
Honestly, I had so many questions about LFP and could not find answers anywhere. So I started my own experiments and thought why not sharing with others, I cannot be the only one with these questions.
Love the "dry" stuff! That's where I learn the most... :-)
I'm glad there's people out there doing tests like you and sharing them with the few that can understand them. You do a good job at breaking it down for people who otherwise wouldn't have understood what this test purpose is
This video has helped clear things up for me a lot thank you.
It looks like energy management is now an official hobby. I like it!
red red wine make me feel so fine, it makes me charge my Batt cells all of the time. sing to the tune of Red Red wine :O)
🍷🍷🍷
You are my guide to setting my charging parameters mostly
Excellent test Andy, I can't wait to finish my installation so that I can proceed to mimic most of your studies. . . Thank you ❤️ for everything you've been showing us all 🙏
Thanks a lot, Mauro!
Hey Buddy I have had these very same questions on my system . Wow you are good , please keep it up.
"This topic is a bit dry."
Well then, we'll BEER with you!
Ya beat me to it
I have two 8S 100Ah packs (each with their own BMS) in parallel. When I got the batteries in June I measured the cell voltages, one set was within the 10mV resolution of my meter, the other set varied about 30mV total between cells. The more matched set works just fine with the 50mA balance current in the BMS, with about 25mV spread when it reaches the absorption voltage of 3.5V and zero current (100% full). The other set was way up around 250mV, even if I extend absorption to 2 hours from the 30 minutes I ended up with. So I bought an active balancer for that pack and it brought the spread down to about 25mV. I set my float voltage to 3.35V and if it is a sunny day I turn on the inverter until the evening once it has gotten down to 80-90% since this system is mostly a solar charged UPS for the home office. With the help of your videos I think I've got things the way I want them.
Nice to see you again, as always 🍻
Thank you.
Thanks for interesting information
My pleasure
Another great video.. 👍
Another quality video - great, clear scientific method/process Andy!
There is a benefit to going to *higher* voltage is it potentially allows for a larger voltage difference (if unbalanced) - which would speed up the balancing process.
I set my absorption to 3.5v/cell (well below the 3.65v cell max), that way i get a good enough voltage difference. So during the absorption the voltage remains constant & the current is typically lower - which allows my balancer to work hard & equalize everything,
What do you set your absorption time settings to, your float voltage and balance turn on to?
@@ssyoumans Absorb V = 28v, float 27.5v, absorb time 15min.
Checked the history & its doing 4-5min of absorb. So OG Garage is on the money 100% with little absorb time, especially as i'm charging at 50A for 280Ah cells.
Balance V = I'm using an active balancer so always on, on the BMS (OverKill/JD) its set at 3.2v.
Hope that helps
@@nateb3105 Thanks, I think I’ll update my settings as well. Still pretty new with this as I just got my battery assembled 2 weeks ago, 280Ah, JD BMS as well.
@@ssyoumans Yeah i'm barely 3months ahead of you :)
Andy has done a great job of explaining capacity vs voltage. I now know to stop chasing perfect 3.55v balance. Also making sure the charger & BMS were working together - i was using the BMS cutoff too much, setting my charge controller much lower now.
Good luck mate!
Hello from America Andy, how are things going with you there? I'm trying to hang in there, I still haven't been able to replace my dead AGM batteries in my bus yet but the good news is that I will be getting my food stamps in a couple of days, I'm just hoping I can get everything ready for winter. Thank you so much for helping me understand all of this. Keep up the good work and stay safe and charged
Thumbs up and subscribed!
Good sh*t! Thumbs up and subscribed!
Something I notice is this correct the higher the curve the easier the pack to balance due to the less amps going in and due to this the BMS has time to balance
Good subject to cover, What is the max discharge voltage. Did someone say you lived in south Africa?
Those birds are keeping you honest ... ;-)
Interested to know what you use to calibrate your voltmeter(s) given how critical small changes are.
Sunny hot Aruba. :)
If you install a Victron BMV-712 instead of the SmartShunt, I think you could automate charging to 90% using the BMV relay to control the Remote On/Off input of the MPPTs. But this relies on Coulomb counting and it is very susceptible to accumulated drift errors. So you also have to automate the system to periodically (e.g. once per week) charge to max. voltage to re-synchronize the shunt to 100% (automate the process that you did manually in this video).
Heya, after seeing some much test of yours I think I would go with 3.55 soc espesially here in Holland winter with less sun hours
The SOC as a number based on voltage will differ if that voltage is measured under a constant discharge current or when the battery has been resting without any load for a few hours.
Du könntest den Kanal mal komplett auf Deutsch synchronisieren. ..... Danke sagen wollte ich auch mal! Schön zu sehen wie du dich in das Thema eingearbeitet hast. Gute Arbeit nochmal Danke dafür.
I have a DIY 22 KWh LiFePO4 pak that is recharged daily by 3 separate arrays/charge controllers, we are fully off-grid in Southern Nevada....I started with very conservative charge settings 3.40 bulk/ 3.35 float... What would happen is that my system would fully charge by 11:00 a.m. and then my diversion loads would kick in but because the outback charge controllers had flipped over to float and had a delay in them The battery would cycle back down fairly low in the late afternoon and by the time the sun went down I was usually around 75 to 80% SOC.. After watching some of your previous videos I reset everything and now 3.55 bulk / 3.5 Float... This way when the diversion loads turn on in the afternoon the system holds the bulk charge much longer and if it does float by the time the sun goes down I still have a fully charged approximately 95% full pack... The more conservative settings you're using just didn't work for my situation. FYI I built the pack two and a half years ago and as an extra caution put in 7amp active balancers between each cell... Because the pack is so Big the most I can charge is 0.25 C so I get very little drift at the top I was worried about this but I've monitored it and don't have a problem 👍
The active balancers put more electricity into lower capacity cells, which makes your pack larger, but if you removed the active balancer the pack would still stay completely balanced just the lower capacity cell would be the discharge cut off. Cell balancing should only be done at the top and not during the discharge cycle because then your pack is actively being unbalanced.
@@uhjyuff2095 I agree, especially for smaller sub 10kwh packs, I have turned off the active balancers before and I do get cell drift at the top after several hours of floating... But when you have a really large pack... 22Kwh / 25.6V nominal / 860Ah I never see over .25C charge rate even in the winter. My existing 3 arrays only put out 5 kw total. What I have noticed is that the pack can sit at 3.5 volts for 4 to 5 hours and when I turn it off it immediately drops back down to 3.45 volts... These large packs have an extreme resistance to charging up that last 5% because the curve is so steep and the pack is so large... By running the active balancers I can hold everything at 3.5V and run my diversion loads then as soon as the sun gets low voltage drops down to 3.45 volts and I am basically at 95%. Without the higher current active balancers it is difficult to hold the voltage at 3.5 for the last 3 to 4 hours of sunlight because one cell gets unhappy ... When you don't have the grid as a backup you become very aware of your battery pack's status... It really sucks to have to run the genny 👍 especially with fuel costs on the rise. Holding the pack at 3.5 volts I doubt that I am losing much capacity especially with his lower charge rate as I'm dealing with... I will probably lose my capacity to calendar date before I lose it to charge rate??? Even so if I can get 10 years out of this pack the way battery development is going I should be able to get more capacity even cheaper at that time 👍 FYI I designed and built my own pack out of pouch cells 3 years ago for $5000 😉 I am way further ahead and happier then I would have been with another lead acid bank.
@@kcjones3368 Excellent. Love to hear from experienced folks as yourself.
@@uhjyuff2095Definitely not an "Expert" but my Dad and I have been keeping the lights On While Off-grid between two properties for the last 30Years 😋
I would like to see how much each cell is expanding at these higher voltages. The total pack expansion would be fine also.
this is something i am interested too. Andy i think we got your next couple of video ideas.
How low do you discharge? Average, but also what is the lowest you went to (how often vs average)?
Just wondering if even one low discharge or one every week/month is what causes the/more deviation in cell voltages.
Thank you... "Sunny HOT"
When you talk about cell voltage with respect to SoC you should always mention if you mean OCV or voltage under load (or charge current). This makes a big difference. An OCV voltage of 3.65V per cell is already massively overcharging, meaning higher degradation! But if you mean 3.65V when charging via 1C this is completely o.k. for the cell. It would result in an OCV of 3.35...3.4 V when charging stops as you reach 0.15C when charging. This is recommended in most data sheets with LiFePO.
At first, I was thinking this is a foolish exercise. We know to keep the voltage for each cell below a certain threshold so that we avoid dendrites and extend the life of the battery. Even if this keeps the SOC a little less and the total available capacity is decreased, it's OK because the battery will last so much longer. However, I am like you in that I'm curious to really KNOW what the capacity of my cells is, instead of just the estimate you get from the shunt measuring input and output. It is smart to leave the settings in the BMS to charge to a safe voltage, but it's also good to know you're getting good amp hours of capacity from your battery cells.
How about Angola this week.. ?😃
Totally up to you 😂
I managed to hit that '80%' rule by implementing a 'stop charge script' using the modbus tcp API of the cerbo GX. This script (javascript / iobroker) grabs the SOC from the cerbo and writes a 'DVCC max charge current = 0A' to the cerbo using the same modbus tcp interface. While discharging the battery the same script recovers the 'DVCC max charge current' to the previous value after the SOC drops below 80% minus hysteresis. In my case I use a hysteresis of 3% SOC. In order to calibrate the coulomb counter inside the BYD-BMS my script allows the battery to be fully charged from time to time.
This is the way I go...
That's a clever way of doing it. And Cloulomb count is the only way of doing it with LFP. Even it is not necessary to charge only to 80% SOC with LiFePO4, I appreciate you sharing this here. That is very kind. I will look into these things in the future and show everyone how to do that if they want to. Very interesting.
For Charging: You reported a 90% SoC of at a 52.8V--which is ~3.3 V/cell in your 16 cells batt--so, why not just bulk charge to 3.35 V/cell & then consider the batt pack charged to 90% SoC. Then set the absorption & float charge limits at 3.3 V/cell? That way your batt should never exceed your user-defined 90% SoC limit, right?
For Discharging: Find out where the discharge curve starts to plummet & then set that V as the discharge cut-off limit.
It's may be lower than the desired 10% SoC, but that shouldn't matter because you won't be stressing the batt pack because your system will have already cut-off discharging before it falls deep into the curve plummet region--where the batt pack stresses are likely to be highest if you allowed it to continue to discharge. If you want to be more precise than using a discharge V cut-off, then set your discharge cut-off using your Coulomb counter % to whatever % that you want the cut-off to occur.
The only other thing you may want to install is a manually resetable trip-breaker, so that it will trip if too much current is ever drawn from your batt. That will protect your batt from whatever over-current limit that you set for your batt pack.
Nice work & explanation....
Why modifiying the balance start voltage? You can set it to 3,40v and let it stay there.
Why could we not build a system with an active balancer kicked in at 3,39V and Bms limited cells to 3,45v. This would be perfect! Andy, we know some guys build that, but to be honest i wuold struggle to plan that - use a relay? Maybe a project for you... 🙃
As soon as that 90 percent load throw on the the charger a observe the charge voltage
Voltage over all doesn't change much in capacity, it does change how fast it can charge, but voltage level is important to the BMS, as you increase the voltage per cell the steepest of the curve comes into play for the balancer to to its job between the highest and lowest cell while in the higher bulk mode, absorption mode voltage will allow the pack to settle down from bulk voltage, by this time the BMS has finished the balancing for that day, I set my cell voltage in bulk mode to 3.525v (56.4v on a 16S ), as I send my cell voltage data to a MQTT sever and plot them in 10 sec intervals 24/7, history shows that the pack voltage is maintained very well with not exceeding 3.55v on any one day
It’s in combination with the SCC. On Victron SCC’s Bulk=Absorption volage
Andy I have an idea for a video, alternatives to the Victron smart shunt which is just to expensive for most.
Most BMS units have their own built in shunt and are more accurate than the Victron Smart Shunt. So whatever BMS you're using should be able to tell you battery state of charge without needing the Victron Shunt.
@@JeremyAkersInAustin it’s only Smart BMS can do that. They also have a limited Bluetooth range. The Victron software, and connectivity is better; especially with the Venus products (including Raspberry Pi) and the free VRM portal. I have one Victron Smart SCC 20A, getting a Raspberry pi next month, wired on Ethernet, to WiFi router. Have a 60A Epever 6415AN SCC (cost 1/8 of a Victron 60A) Epever software and connectivity is poor, but getting an eLOG01 to analyse data on PC, later a RS458 to Ethernet adaptor eTCP02. I want a Victron Smart shunt, but will be next year before I can afford it! - I bought a SNADI 24v 3Kw toroidal inverter (better for aircon) for the same price as the Shunt!
@@michaelbouckley4455 Please show me a lithium BMS that doesn't do SoC calculation.
@@JeremyAkersInAustin Please show me a Victron shunt that's not accurate .
My system is set to limit at 54.35v (3.39v per cell) is this considered a good stop voltage.
It gets this voltage around mid day and stays there till 5 pm, Its usually is on 52.65v when it start to charge in the morning. What % range am I using here
Your voltage is fine. The higher voltage setting is for balancing the cells together and also for reaching 100 percent charge sooner, but if you don't mind waiting longer to reach 100 percent and your cells are already balanced then you are doing everything fine.
If you keep the battery at 3.39V and absorb there for a few hours, you will be at 99%SOC at 5pm. That is totally fine and won't stress the battery at all. Perfect setting.
I have an early version of the bluetti ac200p and I dont know if it has a shunt at all or uses the electronic wattmeters to calculate SoC. In any case it forgets its percentage when I switch it off. New record was yesterday, reporting 16% on switching off and 96% when switching it on (before it typically went down). Reported cell volltage @ 96%=3.40V. Strange 🤷
The Bluetti uses LFP cells already, right? If they just use the voltage as indicator of SOC, they fail. It's absolutely inaccurate.
@@OffGridGarageAustralia Yes the AC 200P uses LFP, the AC200 uses NMC. I guess the switch to LFP was done quite hastily given the AC 200 was crowdfunded and I havent seen it on sale. But I wouldnt have bought NMC anyways
@@HolgerNestmann Dang that sucks, I hope someone can fix it with a firmware update for you!
@@uhjyuff2095 Thanks. At least it‘s no self service and I dont wanna ship that bad boy around if I dont have to. But maybe I can find someone in town with the equipment.
If you noted what the highest cell voltage was when the pack rest v came down to 90 percent according to shunt, that’s where I’d float charge it at. Maybe 3.32 or whatever. Set the float time - to however long it took you to about half the time it took you manually run them up to 3.65 v per cell. Your video editing makes that hard for us estimate? 1 hour or maybe two? No over charging and enough easy balance time. Everything is a compromise, right?
There is usually no time setting for float voltage. It will stay there forever until the voltage drops to a certain (set) threshold and a new cycle begins.
You have all the battery data on my website to download for free. The curves, the data and also the software: off-grid-garage.com/battery-data/
👍👍👍👍👍
This is just me , I like to see the Smart shunt say 100% when the batteries are actually at 90%. The way I achieved this is in the shunts "Battery settings" I set the Charged voltage to 13.7V, so when the battery reaches that voltage it resets the Shunt to 100%. If I am totally wrong doing this , I am open to comments.
I agree also but I'm new to this.. Set a meter to the high and low ends and just worry about full safe usable battery capacity. No need to worry about remembering cutoff percentages.
How do you get to 90% SOC? Based on my ruff calculations for 280 AH cells -- set the tail current cutoff on your Victron MPPT controllers to 16 amps at a charge voltage of 13.45 V/cell. Try it! You may need to bracket a bit above or below 16 amps, but that should get you very close.
It's not impossible. You control cutoff SOC by setting the tail current on the MPPTs.
The path to charging up to 90% is having your charge controller pay attention to percentage and cut the charge when percentage reaches 90%. Then have a monthly balance cycle bring it up slower to a top balancing voltage while paying attention to cell voltage and being careful to slowly raise the voltage. My BMS communicates through the can bus to a 10-year-old Brusa smart charger which does exactly this. The problem is it's only an AC charger so it needs the generator. My solar charge controller is not as smart as a 10-year-old AC charger? WTF!
You get what you pay for with SCC’s
If the topic is dry add a SPAT
2 percent. If you let tail current drop below one amp at 3.45, you are at 98%. I keep my balance set at 3.4v, and don't change it. I usually get about 2 amp hours from 3.4 to 3.65, and leave balancing turned on, don't have to babysit it like that.
You need a bms that calculates the SOC correct via a shunt and then it turns of the charger at 90% SOC. But this is rare to find. Can the Victron bms do it this way? Not sure.....
Allmost all BMS do that though. They all have a shunt and do coulomb count.
@@OffGridGarageAustralia and turn off the charger depending on state-of-charge not voltage?
I like the Electrodacus combination BMS/charge controller...It does have some limitations though... 47v solar input so All panels are in parallel... Also only works on 24 volt systems... But it is a great charge controller and BMS if those stipulations are okay
@@BischesseHunting Even the SOC is calculated on voltage. On Smart ANT BMS, you can calibrate the SOC, my 100% is 3.55v (0% is 2.5v) the SOC now initialises to 100% at 95%, after reaching 100% when it was set at 90%, with cell voltages of 3.39v
@@kcjones3368 I'm trying to convince him to make me a 48V system so I can test it out 😋
Because of a limitation in my hardware I'm only able to charge up to 3.397v . What state of charge do you think I'm at?
Yesterday on my system, according to Smart ANT BMS was 100% SOC (for first time) at that cell voltage, which is about the maximum I charge to. On BMS have 3.55v as 100%, and SOC initialise at 90% so I’m quite happy with that. Rarely reach 3.4v the default BMS balancer level, but cells come up to balance. I initially top balanced for 2 weeks using a 5A buck converter, (from solar powered 12v GEL batteries) which only gave about 1A, and never reached 3.4v
I’m new at this but trying to learn. By Andy’s charge/discharge graph he made with his battery tester I’m guessing your 70-75 percent soc at 3.397v. Battery will last many, many cycles there if that’s doing what you need. These batteries live long and well in the middle of their voltage range.
@@chuckziska988 so far it's keeping me pretty well covered. I have 65 kilowatt hours of storage and 8 KW solar panels. On average I harvest more than I use so I dumped the extra into a hybrid car. It goes 40 miles on a full charge so I rarely need to start its gas engine.
just use a coulomb meter :P
I tried to consistently charge to 90% but gave up. Too many variations using solar as a charging source.
Yes, I think it is impossible to do that with LFP. Once you're over a certain point it just charges to 99% anyway.
@@OffGridGarageAustralia Because I only need a small portion of my battery, I charge to 70%. Around 3.3V per cell.
@@SpeakerKevin Thats good for battery life
I believe ur cell drift are abnormal.
When you charge your batteries with as high current as you do alot of energy is not stored in a few of your cells but lost as heat in most likely a bad contact at the bus bar.
You could check the voltage between cells terminal and bus bar during high load.
With 40 amp load i bet the voltage difference between most of your cells terminals and their respective bus bar is less than 1mv except on the problematic cells you got.
If one cell has 10mv drop when other cells has less than 1mv drop at 50amp charge that is alot of cell imbalance after one cycle.
Note that only during charging cells can get imbalanced (i think?) because part of the current is not getting stored in cell but is lost as heat.
When discharging all of the current is however leaving the cell no matter the resistance
during charge period, voltage shown higher. we need to isolate the cells to get its true voltage when it is in rest mode.
during discharge period, voltage shown lower. we need to isolate the cells to get its true voltage when it is in rest mode.
non-isolated, non-rested cells can not show us true voltage
It's called charge voltage and discharge voltage and caused by the internal resistance.
Still cant see your VRM you know ..... have mentioned it before.
to stop getting to 100% ... then use the power before it fills the battery. if your battery is getting towards full and the sun isnt just about to go down your just not using enough power :)
Link is in the description. Not sure what the problem is. Do you get an error? tried different browsers, mobile phone and PC?
Your VRM is awesome! I’m Looking forward to next month, when getting a Raspberry Pi 3B+ to connect with Victron SCC, and wired Ethernet to WiFi router. Later will buy a Victron Smart Shunt, as I have a Epever 60Amp SCC (Tracer 6415AN, cost 1/8 of equivalent Victron) so will be able to see battery SOC. Victron gear has the best connectivity and software. Bluetooth has limited range, not even 10m through walls. Also ordered an Epever eLOG01 (stores 4 months data) so can monitor output on PC. Will connect that via RS458 to Ethernet too. Will probably store each 30 days of Victron SCC data on PC.
But in the end you didnt define the 90% charge point and its respective cell voltage (assume all balanced)? We dont want to go up the knee, stay on the level, then find the other 10% point and define that. SOC can then only be determined by Coulomb meter
That is very hard/impossible with LiFePO4. Up until 95%SOC, the voltages are all the same. When you leave the flat part of the charge curve, the cells are almost fully charged.
With LiFePO4 it is not worth searching for these 10-90% window. The cycle life is already so high, it makes no sense.
You also have to get out of this flat part to reset/calibrate your smart shunt from time to time and allow the balancer to work.
Well, your SmartShunt knows, he needs to tell you.
Maybe you can just file 10% of your cells, and can then fully charge to achieve 90%. Probably won't work...
The chase for 10%-90% soc…..
Where does this come from? Where is the available data that support this?
So far, the only reference I managed to find is a kind of weak comment in the eve reference sheet for these cells: “Recommended SOC range 10-90%”. There’s no explanation why, and what it will do for the longevity of the cells.
Battleborn have a recommendation too, that you may have smaller cycles, and that will give you longer cycle life in more or less a linear fashion. The say as example that discharging to 50% instead of 0% will double the number of cycles. That statement states within itself that it actually does nothing for the longevity of the battery in terms of available Ah over the lifespan.
If someone have a reference that explains the soc boundaries vs longevity for lifepo4 I would be grateful! I need hard data, not urban myths! 😉
Never fully charge ur battery. Internal resistance changes with SoC. At 100% to 90% and 10% to 20% SoC, resistance is higher.
Lithium CC/CV when the end amps hits zero it is full. the CV will hold it to your settings until the CC fills it up. There is no absorb or it is always in absorb.
Yep, that is correct. You can wither stop charging at a certain voltage altogether or keep the cells on CV for a while. The capacity gain is marginal though when you absorb at 3.4V or above.