How many of you set your battery max discharge under 20% ?

[OnGrid]

I saw on one of Will's videos that setting the max discharge of your Lithium batteries under 20% (max discharge) wouldn't be a problem. He said in his video that "calendar aging will kill the battery before cyling will ever".

He said he likes to set his to charge to 100% and discharges to zero.

Watch video at the 8:15 mark.

I want to set my EG4 48v batteries (two of them 10kw) to max discharge rate of 5%. I don't plan to discharge every night to 5%, but on occasion, if I need the extra power, I have it available.

 

[rpdom]

I've got my inverter set to let my Pylontech US2000Cs go down to 10% while in use and then 8% before they get charged a bit from grid. That's within the normal limits for those batteries so shouldn't affect the lifespan significantly.

 

[Lt.Dan]

I go down to 48v every night. Basically 3%.

@Will Prowse and others have stated repeatedly that calendar aging will kill batteries before cycling will.

 

[DIYPV]

I set my inverter to switch to grid at 10%, but it rarely gets that low.

 

[OPTIMALSOLAR]

The whole 80 20 thing sounded far more marketing than science. Sells more batteries. Eliminates 40% of your battery capacity if followed. Pass. Don't believe the extra 40% needed would ever be paid off with longer life, and more so 8-10 years from now batteries will be way cheaper. Think about the solar batteries/inverters introduced in just the last 12 months. Maybe by then even some will be made in states.

 

[sunshine_eggo]

To me calendar aging is deterioration associated with time where capacity is lost gradually as they age.

I don't think they're going to age out. I think the cells are just going to fail outright. I'm not real keen on chinese LFP cell quality - even when it's EV grade.

The whole 80 20 thing sounded far more marketing than science. Sells more batteries. Eliminates 40% of your battery capacity if followed. Pass. Don't believe the extra 40% needed would ever be paid off with longer life, and more so 8-10 years from now batteries will be way cheaper.

So, you're going with the conspiracy theory rather than spend 5 minutes Googling?

This hasn't been a part of LFP. LPF is typically 90 10 or 100 20.

Staying out of the knees of the voltage curve for improved life is 100% scientific.

80 20 is a 100% science thing for other Lithium chemistries that can impact cycle life by orders of magnitude. NCA can go from 500 cycles to 10,000 cycles with reduced DoD with a corresponding dramatic increased total energy delivery of 8-10X.

I'm using PHEV sourced Panasonic NMC cells - many of which are already 11 years old and abused in vehicles. They are still testing at 84% rated capacity with zero degradation over the last 11 months. I wouldn't say aging out is my biggest risk, but these cells are premium quality, and I'm using them at about 0.25C at the absolute maximum, so it's best case all around.

Reduced C rates, reduced depth of discharge and reduced charge voltage have been proven to extend cycle life notably. With LFP, this can be accomplished with negligible capacity loss due to its already robust nature.

Andy has demonstrated that his 280Ah cells have lost about 5% after just a year.

In 5-10 years, we'll find out if Will is right.

IMHO, a well designed system isn't going to utilize 100% of the capacity because you're not going to have 100% capacity for long.

Think about the solar batteries/inverters introduced in just the last 12 months. Maybe by then even some will be made in states.

LOL. Many already are and have been for decades. The difference is American workers demand a living wage, but American consumers don't want to pay it.

 

[2TrevorJ]

My setup is configured to switch to grid at an indicated 1% (0% & 2.94V/cell triggered some sort of forced grid charging - baked into the Sol-Ark code). I've yet to get that low during normal operation though...after a couple cloudy days with the summer heat and AC utilization I bottomed out at 14% yesterday morning (3.193V / cell). Finally recovered to 100% again early evening today.

 

[fafrd]

I saw on one of Will's videos that setting the max discharge of your Lithium batteries under 20% (max discharge) wouldn't be a problem. He said in his video that "calendar aging will kill the battery before cyling will ever".

He said he likes to set his to charge to 100% and discharges to zero.

Watch video at the 8:15 mark.

I want to set my EG4 48v batteries (two of them 10kw) to max discharge rate of 5%. I don't plan to discharge every night to 5%, but on occasion, if I need the extra power, I have it available.

I attempt to keep battery usage centered. At peak summer, that can mean swinging from 10% to 90% but I’ll change limits at least once per season to reduce that to 20% to 80% for wintertime.

May not make any difference but I just feel silly swinging between 10% and 50% for 3 months…

 

[Crowz]

I wouldn't go to low. Maybe 10% or maybe 5% and the reason has nothing to do with aging the battery it has to do with tripping the bms and the headaches that can ensue from that.

 

[toms]

Show me one instance of where calendar ageing has resulted in a LiFePO4 pack becoming unuseble.

High current rates near the voltage knees on the other hand have been the downfall of many cells.

 

[EastTexCowboy]

I'm not going to get into the discussion on whether it's okay to discharge to zero or not, but on the EG4 batteries when it goes below 20% they go into alarm. Just so you know.

 

[robby]

I don't know who is right but I do know that almost every Tier1 company states 20% is the lowest you should go to get the 6000+ cycles. This is of course assumes a low C rate and the batteries are kept in a fairly cool environment.

I have run across several graphs that paint this kind of picture of what they claim you can expect in dod vs cycles.


I suspect that they are being a bit conservative but with all of this money tied up in batteries I am not willing to gamble on it, especially since it has been my personal experience with older battery chemistries that going to the Knees really shortens the batteries life.

 

[fafrd]

I don't know who is right but I do know that almost every Tier1 company states 20% is the lowest you should go to get the 6000+ cycles. This is of course assumes a low C rate and the batteries are kept in a fairly cool environment.

I have run across several graphs that paint this kind of picture of what they claim you can expect in dod vs cycles.

View attachment 153076
I suspect that they are being a bit conservative but with all of this money tied up in batteries I am not willing to gamble on it, especially since it has been my personal experience with older battery chemistries that going to the Knees really shortens the batteries life.

10,000 cycles is 26.4 years of one cycle per day. I wouldn’t be surprised that calendar aging does not allow you to go that long (even if limiting yourself to 60% DOD cycles over those 26.4 years…

 

[OzSolar]

I'd like to figure out to approach to analyze this from a LCOE (lifetime cost of energy). In the bad old days of FLA math would show you that it was more cost effective to cycle your batteries deeper and replace them more often than shallow cycles and less frequent replacements.

 

[MurphyGuy]

25kWh LG N2.1 NMC battery bank.. I let my cells run between 3.3 and 4.15

 

[robby]

10,000 cycles is 26.4 years of one cycle per day. I wouldn’t be surprised that calendar aging does not allow you to go that long (even if limiting yourself to 60% DOD cycles over those 26.4 years…

Take into account that most of these graphs are based on .5C or less discharge rate and low charge rates.
Factor that along with the Temperature which is typically never ideal and your looking at maybe 4K-6K cycles at 80% DOD to get to a 20% loss in capacity.
I based my system on a 10 year replacement time so pushing the envelope and going with 90% DOD is not something I am willing to risk. As for calendar aging it's mostly a guess as to how long most of these batteries will last. The Separators are probably going to be the first thing that starts to deteriorate due to the daily expansion and contractions of the cells.

 

[sunshine_eggo]

25kWh LG N2.1 NMC battery bank.. I let my cells run between 3.3 and 4.15

Have you found you're using almost all the capacity? My limits with my Panasonic NMC/LMO are 3.5 and 3.92V. 3.5V is about 8% SoC and 3.92V is about 75% based on my own testing.

4.15V is 97%
3.30V is about 3%

 

[robot_zombie]

I use 10% dod limit, but only charge to ~3.46-3.47 VPC with a slightly longer absorption time at that voltage (45min). In winter 10% is hit often but in summer very rarely. my cycle rate is around 0.6/D after around 7 months. at 10 years thats around 2200 cycles, if they last that long with reasonable capacity remaining I will be happy.

 

[SeaGal]

It's complicated and unfortunately there is little historic data, especially for LFP cells.

We talk of getting 6000 cycles, but my heavily used cells are nearing a year old and have logged just 161 cycles. If I get (or rather the cells get) to 6000 cycles, that's like 37 years from now. I know I will have run out of cycles before then

Charging near capacity at a high C rate at low temperatures is definitely going to reduce the cells life, which is why I have implemented variable rate charging and heaters to keep my cells around the 20 to 25 degrees C mark.

But talk of charging to a specific voltage or discharging to a specific voltage (and keeping out of the knees) surely only has a true scientific meaning if that is the real knee due to SOC. Unless the cells have been resting for several minutes their instantaneous voltage will be highly influenced by charge or discharge rate. For example, my cells can be charging at (say) 20A and hit 3.4V, but as soon as the software kicks in and reduces that rate to (say) 5A, the max cell voltage will drop to something like 3.85V. Similarly a cell that would has a rest voltage of (say) 3.2V could drop to 3.1V when the dishwasher kicks in.

And in answer to the original question, I've set the charge / discharge settings to 100% / 15% on my inverter, but I know that my BMS thinks 100% is only about 3.4V, so my software will override that and trickle charge up to 3.45V. Similarly, my software will prevent discharge if the resting voltage of the lowest cell reaches 3.0V or the instantaneous voltage with load of 15A or more causes the lowest cell to reach 2.9V (though this hasn't yet triggered!). ?‍? ?

 

[SamDeleted]

To me calendar aging is deterioration associated with time where capacity is lost gradually as they age.

I don't think they're going to age out. I think the cells are just going to fail outright. I'm not real keen on chinese LFP cell quality - even when it's EV grade.



So, you're going with the conspiracy theory rather than spend 5 minutes Googling?

This hasn't been a part of LFP. LPF is typically 90 10 or 100 20.

Staying out of the knees of the voltage curve for improved life is 100% scientific.

80 20 is a 100% science thing for other Lithium chemistries that can impact cycle life by orders of magnitude. NCA can go from 500 cycles to 10,000 cycles with reduced DoD with a corresponding dramatic increased total energy delivery of 8-10X.

I'm using PHEV sourced Panasonic NMC cells - many of which are already 11 years old and abused in vehicles. They are still testing at 84% rated capacity with zero degradation over the last 11 months. I wouldn't say aging out is my biggest risk, but these cells are premium quality, and I'm using them at about 0.25C at the absolute maximum, so it's best case all around.

Reduced C rates, reduced depth of discharge and reduced charge voltage have been proven to extend cycle life notably. With LFP, this can be accomplished with negligible capacity loss due to its already robust nature.

Andy has demonstrated that his 280Ah cells have lost about 5% after just a year.

In 5-10 years, we'll find out if Will is right.

IMHO, a well designed system isn't going to utilize 100% of the capacity because you're not going to have 100% capacity for long.



LOL. Many already are and have been for decades. The difference is American workers demand a living wage, but American consumers don't want to pay it.

I don't know who is right but I do know that almost every Tier1 company states 20% is the lowest you should go to get the 6000+ cycles. This is of course assumes a low C rate and the batteries are kept in a fairly cool environment.

I have run across several graphs that paint this kind of picture of what they claim you can expect in dod vs cycles.

View attachment 153076
I suspect that they are being a bit conservative but with all of this money tied up in batteries I am not willing to gamble on it, especially since it has been my personal experience with older battery chemistries that going to the Knees really shortens the batteries life.

So if that's true and you should only use the middle 60% of your lithiums

Then to me it almost seems like it's not worth paying out to go LiFePo4? You may aswell save a boat load of money and stick with lead acid at 50% usable capacity