Congratulations on 10k subscribers Andy. Amazingly educational, I feel as a collective we've reached a more advanced understanding of lifepo4 and how best to look after them, keep them healthy and get the best efficiency out of battery banks. Truly thanks :)
This proves that LiFePO4 chemistry is the best available chemistry from a cost, capacity, and safety stand point. We will not see anything for at least 7 years to beat this chemistry! Stay in your 20/80 range (3.23V/3.32V) folks and they will out live most of us.
Uses the same tester as you, but got some varying results when testing 280Ah cells. Suspected included charging clamps, so I made new cables with ring terminals and GX20-5Pin plug, and got a more consistent result.
Note that because the voltage rises faster at the end of charge with a higher current, it is helpful to stay longer at that voltage for the current to drop to near zero.
Andy, It seems to me that if you charged & discharged your lifepo4 cells within a certain range, you will likely get the most life out of your cells. Below is what I would suggest: I call it the pre-avalanche cycling, by which I mean don't let the charge curve or discharge curves go into exponential (i.e. a steep rise or fall--like the graph of an exponential function) *I suggest the following:* *Charge Mode*: cc-cv Temp= min (4.44 °C (40 °F)), ideal (25 °C (77 °F)), max (29.44 °C (85 °F)) Vmax: 3.45 V/cell, then maintain Vmax until charge current = 1A Imax:
Yes, I will share all these info very soon. Filming it at the moment... I haven't changed any of settings due to the results, but I have lowered the absorption voltage to 3.45V now (from 3.5V before). I explain this in all details soon...
Thanks for the educating videos. Your higher voltage at higher charge current and lower voltage at higher load current are expected: The internal resistance of the battery is the same, U = I x R ;)
Andy, I see you charge/discharge batteries so many times and still find the SOC a little daunting. Could you please clarify once and for all( a graph would be wonderful) the optimal voltage ranges of this lifepo4 3.2 100ah battery? 3.65v volts would be over 29v on a 8s configuration, isn't it? Sounds too high. If we would want to do a 80/20 deep of discharge, what voltage are we looking at?
Normal use it makes little sense to charge above 3.5 or 3.55 volts. Look at his discharge curve... it drops really fast below 2.8v or so, indicating little capacity below that point. 2.8v to 3.0v is a good lower cutoff. Voltage ranges on an 8s are very similar to lead acid with 12s (nominally 24v). Think about it... you normally would charge a 12v lead acid to 14.0 to 14.5 volt. That's the same as 28v to 29v for a 24v nominal.
@@trevilights That should be fine. As long as you are within the absolute maximum limits it's all good, and more space inside the limits is even better. One thing I noticed on this video is the higher amps during discharge means even 3.0x8 would be a reasonable 0. I'd probably stay with the 2.8 because that is more certainly below the bend on the discharge.
@@Sylvan_dB Agree. Thanks. By the way, I've downloaded the software and the data from Andy's website and I was able to look closely at the values. He aslo have one he did for a 280ah battery. All data is there. I highly recommend you do too. Andy is providing us with awsome free data. Thanks Andy.
Great channel, congratulations! One quick question - these very popular EVE, CATL, REPT, Lishen lifepo4 cells, I understand the producers specify 3500-4000 cycles, how realistic are these numbers? How reliable are these cells in the long run? Are there users that have been using them a significant amount of time? So far I have been using Sinopoly or Winston, very good but crazy expensive. I am tempted to start building new inexpensive systems using these blue cells but I can't make up my mind. Thank you
Thank you. I'm not sure how long they will last. Nobody knows yet. They come with 5 years warranty. What that means, nobody knows either. I had two claims with the supplier in China and they have just replaced both cells. The LFP tech is around for quite a while, so nothing new here. I guess they will easily last 15-20 years if not kept on high or low voltages.
@@OffGridGarageAustralia Thank you Andy! Winston cells contain Ytrium, they claim 8000 cycles and usability at lower temperature, but they cost double the price of the blue cells. Keep up the good work! Best wishes, Adrian
@@OffGridGarageAustralia il possiedo un pacco di lifepo4 da 30 A da 12 anni Usate male ha oggi anno perso circa 5A. Considerando come le ho maltrattate in questi anni non mi posso lamentare.
I have heard that any battery will last longer (more charge/discharge cycles to 80% DOD) if they are charged and discharged at a lower C rate. While it is far too much to ask of you and this awesome channel, it would be nice to actually see this for ourselves, rather than just trusting the engineers from the battery manufacturers. I have one use that is a very low discharge rate and I can say that you get full capacity, plus some, every single time. I believe it is a matter of how much heat the battery is subjected to...less heat equals longer life.
That is correct, Jason. Smaller charge and also discharge currents will prolong battery life. It's very hard to test and proof as it may take years but overall we will see how my battery will perform over the next couple of years.
Hi andy. GR8 work and show. Would you mind making a video on the parameters you have used in your BMS and victron charger? I have the same system set-up only ,diff, 2x MultiPlus-II 48/5000 parallel and another for charging
Andy, I've one doubth about my battery bank, 58V 40A. My battery bank is charged till 54,4V it means 3,4V per cell and discharge till 48,9 V it means 3V per cell. And the question ⁉️ is. During charging 8h at 5A inverter is feeding peaks of my consumption. It means that is discharging too. If during charging It I discharge it too how many cicles will be during that process ?
You asked for it .... It would be nice to see a charge test series as such: 1 - Charge to 3.65v then discharge at .1c or 28A to 2.5v then recharge to 3.3v 2 - Charge to 3.65v then wait 30min, then discharge at .1c to 2.5v then recharge to 3.3v 3 - Charge to 3.55v then discharge at .1c to 2.5v then recharge to 3.3v 4 - Charge to 3.55v then wait 30min, then discharge at .1c to 2.5v then recharge to 3.3v 5 - Charge to 3.65v then discharge at .1c to 3.0v then recharge to 3.3v 6 - Charge to 3.65v then wait 30min, then discharge at .1c to 3.0v then recharge to 3.3v 7 - Charge to 3.55v then discharge at .1c to 3.0v then recharge to 3.3v 8 - Charge to 3.55v then wait 30min, then discharge at .1c to 3.0v then recharge to 3.3v I might buy a 100AH battery to do this. Also do the series above at a temperature of 10C and 25C As always your videos are not only fun to watch, but so educational. 👍 And with 76.9K subscribers now ... amazing.
So your saying if I charge my 400 ah 4s 2P setup at 20 amps cause that's the lowest my inverter charger will do it'll not make one cell actually 2 in parallel go to a higher voltage than the rest of them ?
You are in the ranges where measurement accuracy is a factor on results. The basic principles of C rate charge and discharging is how much power loss is happening for the I^2 * (R_s + R_kinetic) for cell. There is a secondary factor which is temp of cell. The higher the temp of cell the lower the kinetic energy necessary (lower kinetic loss). Temp can also change the Rs a bit. However you don't get much effect on cell temp and therefore change of kinetic energy required when cell current is less then 0.5 C rate of current. You do get effect due to ambient environment temp. R_s and effective R_kinetic losses are a little different for different cells, and both degrade the older and more used the cell. For 100 AH LiFePO4 cell in good condition, I will use R_s of 0.45 milliohms per cell and R_kinetic about 60% of R_s or 0.27 milliohm. If your capacity tester has four point wiring with separate cell voltage sense you don't have to worry about wiring loss to tester. If don't have remote voltage sensing then the wire loss adds to capacity loss. So running the numbers, I only get a capacity difference of 99.9%, 99.5%, and 99.3% for 5A, 20A, and 30A respectively. This is so small a difference it is within measurement error. By first measuring your R_s with four point battery impedance meter, you can get a better number for R_kinetic by the difference in curve voltage drop for different current rates and subtracting out I*Rs from this voltage drop leaving the effective R_kinetic drop. Remember R_kinetic has 1 to 3 minute time delay until it reaches equilibrum so initially upon first applying current there is only Rs caused voltage drop then within three minutes the cell voltage will drop a little more once R_kinetic settles to its equilibrium state for given current. Same applies in reverse for charging.
7:22, its almost the same because LiFePO4 has superb coulombic efficiency, both charge and discharge. Above 99%! So does li-ion to be fair, but i believe li-ion only achieves this at lower C-ratings whereas LiFePO4 can maintain high coulombic efficiency even at higher C-ratings
So basically you get an extra 4Ah out of the battery but I think that in a real world application, like you are about to parallel two bank's of Batteries, you would be discharging the batteries a half the current. So to be fair, you did one test at 30A discharge and the second test should have been done at 15A to simulate two battery bank's in parallel! Then divide the current by 3 to simulate 3 paralleled battery bank's. That way you can see how much more Ah you gain from sharing the discharge current across multiple battery bank's! As from what I can see, there was a 4Ah gain from 30A to 20A discharge which would probably increase to 5Ah or more per bank, across 3 bank's, it might yield 15Ah or possibly more and then there's the battery life expectancy that should increase from the lower current draw from each series bank! Every extra watt that is gained from the panels is as good as every extra Ah stored as is every extra year of service life! This all just probably means that you either go big or go home LoL... But you are the only person that I know who can find this out! I wish you nothing but success in your captain cook's endeavour in this upside down land that we call home!
Off grid garage I recently purchased eve cell 280 ah when I got the batterys they were 3.3 v I try to top balance cells but they don't go above 3.3 v are these cells good or did I get ripped off ?
I request a cycle of charging 33C and discharging 50C at a frequncy of 77 kHz and a duty cyle of starting with 66% dropping to 44% over 2 hours. Just kidding :-P I guess the difference of about 4 Ah was still left in the battery and rising the voltage back up to almost 3 V quickly. Thanks for sharing :-)
This isn’t the first time you’ve expressed concern about leaving the cell discharged all night. Why not just queue a charging cycle to run after the discharge cycle? 🤓🤔🤷♂️
With the fast bounce back your lost capacity is likely still in there. Will the tester allow one or two repeats until the bounce back is small or the same . I only mention this as I believe the cell is like a liter of water it's all you get whether you pour it through the top or a needle point all you get is a liter.
Congratulations on 10k subscribers Andy.
Amazingly educational, I feel as a collective we've reached a more advanced understanding of lifepo4 and how best to look after them, keep them healthy and get the best efficiency out of battery banks.
Truly thanks :)
Yabba dabba zabbuneeda! This was an excellent video. Good work! And good morning!!
This proves that LiFePO4 chemistry is the best available chemistry from a cost, capacity, and safety stand point. We will not see anything for at least 7 years to beat this chemistry! Stay in your 20/80 range (3.23V/3.32V) folks and they will out live most of us.
Yep, I agree with that.
Love it !
Especially those overlays of 2 graphs to match the horizontal scale.
Congrats Andy, reaching 10k subscribers and still creating nice videos!
Thanks a ton!
Uses the same tester as you, but got some varying results when testing 280Ah cells. Suspected included charging clamps, so I made new cables with ring terminals and GX20-5Pin plug, and got a more consistent result.
Thanks Andy
Any time!
So close to 10K! Congratulations
Note that because the voltage rises faster at the end of charge with a higher current, it is helpful to stay longer at that voltage for the current to drop to near zero.
Andy,
It seems to me that if you charged & discharged your lifepo4 cells within a certain range, you will likely get the most life out of your cells. Below is what I would suggest:
I call it the pre-avalanche cycling, by which I mean don't let the charge curve or discharge curves go into exponential (i.e. a steep rise or fall--like the graph of an exponential function)
*I suggest the following:*
*Charge Mode*: cc-cv
Temp= min (4.44 °C (40 °F)), ideal (25 °C (77 °F)), max (29.44 °C (85 °F))
Vmax: 3.45 V/cell, then maintain Vmax until charge current = 1A
Imax:
So with the 30A test done would you share with us your current charge controller and BMS settings?
Did the 30A test change anything?
Yes, I will share all these info very soon. Filming it at the moment...
I haven't changed any of settings due to the results, but I have lowered the absorption voltage to 3.45V now (from 3.5V before).
I explain this in all details soon...
Thanks for the educating videos.
Your higher voltage at higher charge current and lower voltage at higher load current are expected: The internal resistance of the battery is the same, U = I x R ;)
Danke Dir!
Yes, this is just because of the resistance in the cells, cables and connections.
Andy, I see you charge/discharge batteries so many times and still find the SOC a little daunting. Could you please clarify once and for all( a graph would be wonderful) the optimal voltage ranges of this lifepo4 3.2 100ah battery? 3.65v volts would be over 29v on a 8s configuration, isn't it? Sounds too high. If we would want to do a 80/20 deep of discharge, what voltage are we looking at?
Normal use it makes little sense to charge above 3.5 or 3.55 volts. Look at his discharge curve... it drops really fast below 2.8v or so, indicating little capacity below that point. 2.8v to 3.0v is a good lower cutoff.
Voltage ranges on an 8s are very similar to lead acid with 12s (nominally 24v). Think about it... you normally would charge a 12v lead acid to 14.0 to 14.5 volt. That's the same as 28v to 29v for a 24v nominal.
@@Sylvan_dB Thanks Sylvan. So then can we consider 28v(3.5x8) to be 100% and 22.4v(2.8x8) to be 0%? Rough numbers here.
@@trevilights That should be fine. As long as you are within the absolute maximum limits it's all good, and more space inside the limits is even better.
One thing I noticed on this video is the higher amps during discharge means even 3.0x8 would be a reasonable 0. I'd probably stay with the 2.8 because that is more certainly below the bend on the discharge.
@@Sylvan_dB Agree. Thanks. By the way, I've downloaded the software and the data from Andy's website and I was able to look closely at the values. He aslo have one he did for a 280ah battery. All data is there. I highly recommend you do too. Andy is providing us with awsome free data. Thanks Andy.
Great channel, congratulations! One quick question - these very popular EVE, CATL, REPT, Lishen lifepo4 cells, I understand the producers specify 3500-4000 cycles, how realistic are these numbers? How reliable are these cells in the long run? Are there users that have been using them a significant amount of time? So far I have been using Sinopoly or Winston, very good but crazy expensive. I am tempted to start building new inexpensive systems using these blue cells but I can't make up my mind. Thank you
Thank you. I'm not sure how long they will last. Nobody knows yet. They come with 5 years warranty. What that means, nobody knows either. I had two claims with the supplier in China and they have just replaced both cells.
The LFP tech is around for quite a while, so nothing new here. I guess they will easily last 15-20 years if not kept on high or low voltages.
@@OffGridGarageAustralia Thank you Andy!
Winston cells contain Ytrium, they claim 8000 cycles and usability at lower temperature, but they cost double the price of the blue cells.
Keep up the good work! Best wishes, Adrian
@@OffGridGarageAustralia il possiedo un pacco di lifepo4 da 30 A da 12 anni
Usate male ha oggi anno perso circa 5A.
Considerando come le ho maltrattate in questi anni non mi posso lamentare.
I have heard that any battery will last longer (more charge/discharge cycles to 80% DOD) if they are charged and discharged at a lower C rate. While it is far too much to ask of you and this awesome channel, it would be nice to actually see this for ourselves, rather than just trusting the engineers from the battery manufacturers. I have one use that is a very low discharge rate and I can say that you get full capacity, plus some, every single time. I believe it is a matter of how much heat the battery is subjected to...less heat equals longer life.
That is correct, Jason. Smaller charge and also discharge currents will prolong battery life. It's very hard to test and proof as it may take years but overall we will see how my battery will perform over the next couple of years.
@@OffGridGarageAustralia - I'm looking forward to that. :)
Please run multiple cycles on two batteries at the 5 and 30 amp rates then check the resistance of each battery
Hi andy. GR8 work and show.
Would you mind making a video on the parameters you have used in your BMS and victron charger? I have the same system set-up only ,diff, 2x MultiPlus-II 48/5000 parallel and another for charging
Yes I can. I'll see if I can fit it in this week. Thanks for your request and suggestion.
Hi Andy what's the name of the mobile app you are using to remote connect to your computer?
this now needs to be taken to the extreme at 1c CHG/dschg but I'm not sure how you'd get the graphs for it
The old school way, with meters, pencil, and paper. 👴
Andy, I've one doubth about my battery bank, 58V 40A.
My battery bank is charged till 54,4V it means 3,4V per cell and discharge till 48,9 V it means 3V per cell. And the question ⁉️ is.
During charging 8h at 5A inverter is feeding peaks of my consumption. It means that is discharging too.
If during charging It I discharge it too how many cicles will be during that process ?
So a very very small Peukert effect then. Great vid, pretty much what I thought was going on, but its great to have numbers/proof.
LiFePO4 do not suffer from the Peukert Effect.
@@dankarau2307 You're no scientist, so what would you call reduced output from a higher draw rate then?
@@dankarau2307 Probably because Peukert's Law is specific to Lead chemistry.
@@chrisfryer3118 heat
@@blackz06 I thought that was where the energy went in the P effect.
Andy DOES own a shirt with a collar! 😂
Two actually 😂
Is it ok to have cutoff current set to 0 in step3 (discharge to 2.5V)?
You asked for it ....
It would be nice to see a charge test series as such:
1 - Charge to 3.65v then discharge at .1c or 28A to 2.5v then recharge to 3.3v
2 - Charge to 3.65v then wait 30min, then discharge at .1c to 2.5v then recharge to 3.3v
3 - Charge to 3.55v then discharge at .1c to 2.5v then recharge to 3.3v
4 - Charge to 3.55v then wait 30min, then discharge at .1c to 2.5v then recharge to 3.3v
5 - Charge to 3.65v then discharge at .1c to 3.0v then recharge to 3.3v
6 - Charge to 3.65v then wait 30min, then discharge at .1c to 3.0v then recharge to 3.3v
7 - Charge to 3.55v then discharge at .1c to 3.0v then recharge to 3.3v
8 - Charge to 3.55v then wait 30min, then discharge at .1c to 3.0v then recharge to 3.3v
I might buy a 100AH battery to do this. Also do the series above at a temperature of 10C and 25C
As always your videos are not only fun to watch, but so educational. 👍
And with 76.9K subscribers now ... amazing.
So your saying if I charge my 400 ah 4s 2P setup at 20 amps cause that's the lowest my inverter charger will do it'll not make one cell actually 2 in parallel go to a higher voltage than the rest of them ?
You are in the ranges where measurement accuracy is a factor on results.
The basic principles of C rate charge and discharging is how much power loss is happening for the I^2 * (R_s + R_kinetic) for cell.
There is a secondary factor which is temp of cell. The higher the temp of cell the lower the kinetic energy necessary (lower kinetic loss). Temp can also change the Rs a bit. However you don't get much effect on cell temp and therefore change of kinetic energy required when cell current is less then 0.5 C rate of current. You do get effect due to ambient environment temp.
R_s and effective R_kinetic losses are a little different for different cells, and both degrade the older and more used the cell.
For 100 AH LiFePO4 cell in good condition, I will use R_s of 0.45 milliohms per cell and R_kinetic about 60% of R_s or 0.27 milliohm.
If your capacity tester has four point wiring with separate cell voltage sense you don't have to worry about wiring loss to tester. If don't have remote voltage sensing then the wire loss adds to capacity loss.
So running the numbers, I only get a capacity difference of 99.9%, 99.5%, and 99.3% for 5A, 20A, and 30A respectively. This is so small a difference it is within measurement error.
By first measuring your R_s with four point battery impedance meter, you can get a better number for R_kinetic by the difference in curve voltage drop for different current rates and subtracting out I*Rs from this voltage drop leaving the effective R_kinetic drop. Remember R_kinetic has 1 to 3 minute time delay until it reaches equilibrum so initially upon first applying current there is only Rs caused voltage drop then within three minutes the cell voltage will drop a little more once R_kinetic settles to its equilibrium state for given current.
Same applies in reverse for charging.
I have the daly 4s 200 amp BMS what should I set my high and low cell voltage to and the balance voltage ?
7:22, its almost the same because LiFePO4 has superb coulombic efficiency, both charge and discharge. Above 99%! So does li-ion to be fair, but i believe li-ion only achieves this at lower C-ratings whereas LiFePO4 can maintain high coulombic efficiency even at higher C-ratings
I've heard the LiFePO4's can suck up anything you feed them.. Well, thats for another video LOL. Very cool!!
can you charge and discharge at 1,000 amps please
Yes please
It will be quick too only what 6 minutes each way
we deserve to see what those cells really have to offer
Somebody get Photoninduction on the case... Booom
@@yodab.at1746 now he's back on the scene there's a good chance he'd take suggestions :)
When charging, don't you want to use a cut-off voltage of 5 Amps (C/20)? I've heard of C/20, and even C/30, but never C/10 like in this test.
I have heard 10% of the charge current but I mean with a discharge in the real world you just have whatever load you have it's not going to taper off
@@james10739 I was referring to charging only,
Could you do the same with a 12v lead-acid battery with ZKE discarger? cycle test
yes you can
Awesome!
So basically you get an extra 4Ah out of the battery but I think that in a real world application, like you are about to parallel two bank's of Batteries, you would be discharging the batteries a half the current. So to be fair, you did one test at 30A discharge and the second test should have been done at 15A to simulate two battery bank's in parallel! Then divide the current by 3 to simulate 3 paralleled battery bank's.
That way you can see how much more Ah you gain from sharing the discharge current across multiple battery bank's!
As from what I can see, there was a 4Ah gain from 30A to 20A discharge which would probably increase to 5Ah or more per bank, across 3 bank's, it might yield 15Ah or possibly more and then there's the battery life expectancy that should increase from the lower current draw from each series bank!
Every extra watt that is gained from the panels is as good as every extra Ah stored as is every extra year of service life!
This all just probably means that you either go big or go home LoL...
But you are the only person that I know who can find this out!
I wish you nothing but success in your captain cook's endeavour in this upside down land that we call home!
Off grid garage I recently purchased eve cell 280 ah when I got the batterys they were 3.3 v I try to top balance cells but they don't go above 3.3 v are these cells good or did I get ripped off ?
How do you charge them?
That is good. They will stay at 3.3 for a long time and then suddenly go higher....as per the graph!
I have the go power ic 3000 inverter charger inverter charger combo
@@zipsolo what settings are you putting in?
even if you have a tesla, it is illegal to use a phone while in the drivers seat in AU.
and it's $1000 in QLD ouchies.
Well, it wasn't a phone at all 😊
@@OffGridGarageAustralia I think a camera would encounter the same issue :P
can't be wasting calibration tool funding!
Who did drive the car?
Australia drive on lefthand side of road. Steering wheel on the right....
@@AdrSny you mean that Australian people drive on the correct side of the road LoL...
@@PeterMilanovski Yes. I am from South Africa.. Also driving on the correct side!!! wink wink
@@AdrSny good man! LoL...
I request a cycle of charging 33C and discharging 50C at a frequncy of 77 kHz and a duty cyle of starting with 66% dropping to 44% over 2 hours.
Just kidding :-P
I guess the difference of about 4 Ah was still left in the battery and rising the voltage back up to almost 3 V quickly.
Thanks for sharing :-)
Das ist doch Kaese!!! 🧀
Andy, on your compraison of discharge curves, your start voltages were not the same.
Yes that's true but for the discharge test they started at the same voltage!
Happy AMPS, 🍺🍺🍺
This isn’t the first time you’ve expressed concern about leaving the cell discharged all night. Why not just queue a charging cycle to run after the discharge cycle? 🤓🤔🤷♂️
Because that charge cycle will be shown in the curve and also in the data 😉
With the fast bounce back your lost capacity is likely still in there. Will the tester allow one or two repeats until the bounce back is small or the same . I only mention this as I believe the cell is like a liter of water it's all you get whether you pour it through the top or a needle point all you get is a liter.