Life cycle versus soc

[Johandup]

Has anyone figured this relationship out?
I have Pylontechs which are “claimed - not yet proven” that they will last for 6000 life cycles at 80% dod.

The usage for every 10% soc is 960wh which is (now and at current electricity prices) still affordable if drawn from the municipality.

A theoretical question for now is:

At what soc will the life cycle go up at lot (for Pylontech to extend the usage life time) and what will the electricity bill be for doing so?

Then one can make an informed decision on how to set your inverter. I know that technology and prices will change a lot in the future.

[Nexuss]

From what i have gathered over the past few years reading about LFP you should be cycling them as much as you possibly can. Calendar aging will kill the battery before it dies from too many cycles.  I try to do 90-95% DOD every day but with all the loadshedding it makes it hard to cycle as much.

[Scorp007]

1 hour ago, Nexuss said:

From what i have gathered over the past few years reading about LFP you should be cycling them as much as you possibly can. Calendar aging will kill the battery before it dies from too many cycles.  I try to do 90-95% DOD every day but with all the loadshedding it makes it hard to cycle as much.

There are also resources that indicate using say 85% of capacity daily would provide a lot less cycles and accumulated power than accumulated power from say between 20 to 80%. This 60% would provide a lot more accumulated power than say 35 to 95%.

[P1000]

6 hours ago, Scorp007 said:

There are also resources that indicate using say 85% of capacity daily would provide a lot less cycles and accumulated power than accumulated power from say between 20 to 80%.

Please link to these resources, presumably they are applicable to LiFePO₄?

[Scorp007]

3 hours ago, P1000 said:

Please link to these resources, presumably they are applicable to LiFePO₄?

Cannot confirm LiFePO4 but was on one of the many items on battery varsity. 

[Johandup]

Okay gents, I’ve gotten the final report on the Aus battery test project - which was concluded after TEN years in March 2023.

The report is VERY interesting to say the least. It is also of note to observe that we are so far behind Aus in solar (not just rugby ) that this guvament will neva eva catch up. They call it their Arena projects.
A quick summary on the Pylontech US2000B batterues of which I have four.

After 6 years of trial where they have been constantly (I think twice per day) discharged and charged they have completed close to 2800 cycles and retained 75% of its capacity. These batteries had zero problems during the trial.
Being on pension for 7 years and pushing 70 I won’t think about this issue anymore. 
An observation is that I need to add a big new lifepo4 battery at some stage when capacity starts being an issue - not before the time. Having a Sunsynk 8kw makes it easy.
The indicator will be when the soc reaches 20% after a night shift.

The link is: https://arena.gov.au/assets/2018/05/lithium-ion-battery-testing-public-report-12.pdf for people who have time to read.

Edited March 19, 2023 by Johandup

[rkistner]

I have a Revov R100 battery. Specs says it's "designed" for 7000 cycles, with a warranty of 10 years or 3500 cycles at 100% DOD. 50% discharge counts as 0.5 cycles. So with this, and considering I only have a single 5.1kWh battery, I'm going to cycle it as much as I can (down to 20% during the night, and 5% during load shedding). And if I do shorten the battery life a little because of this... by the time I'm over 3500 cycles, batteries may be a lot cheaper than they are now.

 

 

 

 

[Scorp007]

On 2023/03/18 at 3:47 PM, P1000 said:

Please link to these resources, presumably they are applicable to LiFePO₄?

https://batteryuniversity.com/article/bu-808-how-to-prolong-lithium-based-batteries

 

Edited March 29, 2023 by Scorp007

[P1000]

1 hour ago, Scorp007 said:

https://batteryuniversity.com/article/bu-808-how-to-prolong-lithium-based-batteries

 

Those numbers do not seem to match well with current LiFePO4 technology, It's not clear what they are referring to with "LiPO4" in the table, they have one reference in the article to Lithium Iron Phosphate, only to state that the tables are not applicable to them?

Quote

The following tables indicate stress related capacity losses on cobalt-based lithium-ion. The voltages of lithium iron phosphate and lithium titanate are lower and do not apply to the voltage references given.

 

Besides that, the article gives no substance to your claim that cycling between 20 and 80% has significant impact on cycle life for LiFePO4 batteries.

Edited March 29, 2023 by P1000

[Nexuss]

35 minutes ago, P1000 said:

Besides that, the article gives no substance to your claim that cycling between 20 and 80% has significant impact on cycle life for LiFePO4 batteries.

Exactly true ,in this research paper it shows clearly there is no significant difference between 0-100% and 20-80% for LFP.

https://iopscience.iop.org/article/10.1149/1945-7111/abae37

Edited March 29, 2023 by Nexuss

[Scorp007]

1 hour ago, P1000 said:

Those numbers do not seem to match well with current LiFePO4 technology, It's not clear what they are referring to with "LiPO4" in the table, they have one reference in the article to Lithium Iron Phosphate, only to state that the tables are not applicable to them?

 

Besides that, the article gives no substance to your claim that cycling between 20 and 80% has significant impact on cycle life for LiFePO4 batteries.

Yes they did not quote 20 to 80% as those are my figures. If you look in their tables/graphs you will see 25 to 85% which will not show much of a deviation from my figures which are still 60% use. 

Can you supply I link to make this void? 

 

Edited March 29, 2023 by Scorp007

[Scorp007]

1 hour ago, P1000 said:

Those numbers do not seem to match well with current LiFePO4 technology, It's not clear what they are referring to with "LiPO4" in the table, they have one reference in the article to Lithium Iron Phosphate, only to state that the tables are not applicable to them?

 

Besides that, the article gives no substance to your claim that cycling between 20 and 80% has significant impact on cycle life for LiFePO4 batteries.

Duplicate

Edited March 29, 2023 by Scorp007

[P1000]

1 hour ago, Scorp007 said:

Yes they did not quote 20 to 80% as those are my figures. If you look in their tables/graphs you will see 25 to 85% which will not show much of a deviation from my figures which are still 60% use. 

Can you supply I link to make this void? 

 

There is no claim in the linked article that that is applicable to LiFePO4. I am not disputing the fact that this is applicable to cobalt based lithium-ion batteries.

[Steve87]

I'm no scientist or claim to know a lot but I have observed a lot during my experience with building these batteries with specific reference to LFP. 

The stress happens at the Top of the cell voltage & at the very bottom of the individual cell regime. Stay away from there & those cells will last a very longtime. Out of particular interest is that already at 2.9V cell Voltage that cell has discharged between 80 - 85% of its SoC. 

I have verified this with at least 5 LPF packs I have built & making use of a Victron Smart shunt. To date I have had some 6MW of power transferred through 3 of these packs & I do not cycle below 3V per cell so that's 48V for the 16s pack. 

And they have done 300+ cycles without showing any signs of decay or loss in terms of capacity. 

Another question mark to me is that not all cells LFP flavour are the same. EVE will publish a specific Spec sheet for their 230ah cells Vs their 304 ah cells. 

EVE also have had some poor batches of cells that have tested below par & so you can see that it's quite subjective this discussion. 

[Nexuss]

29 minutes ago, Steve87 said:

The stress happens at the Top of the cell voltage & at the very bottom of the individual cell regime. Stay away from there & those cells will last a very longtime. Out of particular interest is that already at 2.9V cell Voltage that cell has discharged between 80 - 85% of its SoC. 

I agree somewhat but my exp with Pylontech differs slightly . If i cycle 95% the voltage range is 3.55Vmax  to 3.08V minimum per cell ,they mostly dont ever go below 3,1V even at 5% SOC. 

Edited March 29, 2023 by Nexuss

[Steve87]

A Shunt is also something that can vary results. I'm not saying the Pylontech is incorrect but a BMS is a very sophisticated piece of kit & it's Current senor or Shunt is a very important part of how it meters it's little Coulombs. 

Also your comment makes you really ask yourself the question? How much value is there in squeezing out the last 5%? Lots of stress that happens for maybe 1ah extra draw

Btw not saying you a culprit of such thinking just pointing out why drain to 0% or low voltages that so many do voluntary or not maybe with a Voltronics & have no idea what the Battery cut off voltage is. 

Edited March 29, 2023 by Steve87

[Nexuss]

11 minutes ago, Steve87 said:

why drain to 0% or low voltages that so many do voluntary or not maybe with a Voltronics & have no idea what the Battery cut off voltage is. 

I agree totally thats why i extensively monitor my cells to make sure i am not stressing them , i consider staying above 3 volts as very safe. If i were not monitoring individual cells i would not drain them as far . I also limit discharging current slightly at below 20% soc to avoid the voltage dipping below 3V. Gota take care of them a bit lol but i definitely use them fully . Over 4MWh out of my pack in 680days which consisted mostly of 7kwh capacity ,added a new battery 2 months ago.

[Scorp007]

24 minutes ago, Steve87 said:

 

Also your comment makes you really ask yourself the question? How much value is there in squeezing out the last 5%? Lots of stress that happens for maybe 1ah extra draw

What you mention is the reason why I don't use my lithiums below 3.1V per cell. Also no point in charging to 100% accept when I balance the cells. Also unlike most of us I only use my lithiums for my routers and only switch on the inverter for the house during evening LS or when power is required for any load during LS. 

Yes my lithiums will die of old age I hope. 

For the last 2 weeks I found I just need to charge them about 1-1.5kWh per day. Thanks to the battery balancer that was given a 2nd life with my 2nd life S-100's I can keep my routers directly on one of the batteries for the 12V supply. 

[Red Falcon]

So if I understand this correctly. 

Lifepo4 batteries should be cycled down to 20% often to get maximum life out of them? 

 

I'm worried because my current setup only use 5/10% of my battery capacity during a 2/4 hour led shed. 

How often should I allow my batteries to drop to 20%. If that is the case. 

Edited March 29, 2023 by Red Falcon

[Nexuss]

7 hours ago, Red Falcon said:

So if I understand this correctly. 

Lifepo4 batteries should be cycled down to 20% often to get maximum life out of them? 

 

I'm worried because my current setup only use 5/10% of my battery capacity during a 2/4 hour led shed. 

How often should I allow my batteries to drop to 20%. If that is the case. 

Not quite correct , you can safely cycle them low (5-20% SOC) to get the maximum returns/savings from them but you certainly dont have to. If you are using only 10% of your capacity that is totally fine. It just means you sized the battery slightly large for your needs. 

[Johandup]

I found that my pylontechs would go down to 19% soc before switching off.

On digging into the Sunsynk setup I found that I haven’t ticked the bms box.

So the switching off was done on the battery by the bms.

I also upped the minimum soc level to 21% in the Sunsynk setup to be safe.
It is now switching over to Escom at 20% soc.

In the light of this I do not know how one will be able to go down to say 10% as the bms is on auto inside the batteries.

Comments welcome.

[Nexuss]

25 minutes ago, Johandup said:

I found that my pylontechs would go down to 19% soc before switching off.

On digging into the Sunsynk setup I found that I haven’t ticked the bms box.

So the switching off was done on the battery by the bms.

I also upped the minimum soc level to 21% in the Sunsynk setup to be safe.
It is now switching over to Escom at 20% soc.

In the light of this I do not know how one will be able to go down to say 10% as the bms is on auto inside the batteries.

Comments welcome.

Interesting. Share some more details, what battery model ,how many units ,age? What do you mean 'BMS box'? The BMS _Err_Stop? Mine is not ticked. I have not ever had a battery shut down. Right at this moment they are on 5% SOC no problem.  My shut down is set to 3% ,Low batt to 4%. *Edit What i would suggest to anyone wanting to optimize their batteries to their settings is to use the battery monitor software to monitor the pack voltages and their individual cell voltages while you do a deep discharge to 5% .This will give you lots of information as you will see which pack hits 5% first and then you can compare that with the Inverters SOC at that time to know what a safe number would be to go down to.

Edited March 30, 2023 by Nexuss

[Johandup]

23 minutes ago, Nexuss said:

Interesting. Share some more details, what battery model ,how many units ,age? What do you mean 'BMS box'? The BMS _Err_Stop? Mine is not ticked. I have not ever had a battery shut down. Right at this moment they are on 5% SOC no problem.  My shut down is set to 3% ,Low batt to 4%.

4 off  US2000b pylontechs - now going for 4 years in use.

I don’t restrict usage at night as the setup changes over to Eskom (when available) when the battery shuts down. My batteries are set to charge only from solar.

I ticked the “activate” in the Battery setup - wasn’t done before.  
I have a 8kw Synsunk inverter. 
I don’t dig into the battery cell info like @Nexuss does.

Edited March 30, 2023 by Johandup

[Nexuss]

2 minutes ago, Johandup said:

4 off  US2000b pylontechs - now going for 4 years in use.

Those are most likely restricted to 80 or 90% DOD if i had to make a guess. I am only familiar with the 3000C so i can only speak for them..  You could potentially unlock the 95% ability with the latest firmware if i remember correctly but if you are not totally sure about how to do it i would not recommend doing it. My system specs are in my sig.