Optimal Charge Voltage - (LiCo)

Batteries, Chargers, and Battery Management Systems.
User avatar
DrkAngel   100 GW

100 GW
Posts: 5316
Joined: Dec 15 2010 11:14am
Location: Upstate-Western-Southern Tier NY. USA

Re: Optimal Charge Voltage - (LiCo)

Post by DrkAngel » Nov 30 2019 12:06am

Well, getting back to the tittle topic of LiCo ...
Rather than just full cycles, this chart illustrates optimal voltage ranges of partial cycles.
Optimal partial cycling.jpg
Optimal partial cycling.jpg (123.92 KiB) Viewed 1403 times
Notable is the difference in deterioration of 60% cycling from 100% to 40% compared to 85% to 25%. (red - green)
More telling is the difference in deterioration of 50% cycling from 100% to 50% compared to 75% to 25%! (dark blue - light blue)
Conclusion seems clear that higher voltages provoke faster deterioration!

.
Last edited by DrkAngel on Dec 01 2019 10:30am, edited 2 times in total.
A little learning is a dangerous thing;
Drink deep, or taste not the Pierian spring:
There, shallow draughts intoxicate the brain,
And drinking largely, sobers us again.

I enjoy enlightening ... and enlivening the spirit of the innovators.

New & Improved - Acronym Definitions

Index - Homemade Battery Packs - Updated - December 2019

EBike Toolbox - Bargains! $

Endless Sphere Wiki - Lost?

Pajda   1 kW

1 kW
Posts: 339
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Optimal Charge Voltage - (LiCo)

Post by Pajda » Nov 30 2019 5:12am

Graph without exact definition of testing method and particularly tested cell sample should not be even published, just because you can not add your own measure to compare. I am in this "bussines" for quite a long time and if it teaches me something, that almost all informations about ageing of cells you can easily google are just bullshits. I was not measure LCO cells for quite a while, because it just not make much sense if there are available NCA and particularly NMC cells. But I can tell, that this presented trend does not apply to NCA and NMC cells in 1865/21700 format. For both chemistries its cycle life (both DCIR rise and capacity fade) depends only on SoC window wide, there is no significant benefit in cycle life if SoC window is between 100%-50% or 75-25% (if we are talking about first few thousands of cycles)

And last one note, high quality cell like LG M36 can do 1000 cycles with 100% DoD (4.2-2.5V) at 0.5C-1C rate. If you use them in a modern BEV it gives you 500,000km range. Of course it is impossible to do each ride 500 km long, particularly with 1C continuous discharge rate. So your average will be 20-40% DoD with 0.2C dischare rate. So you can use "very shitty" HE cells in term of cycle life (and so very cheap) and still you do not have significant problem with battery life. Completely different story is ESS where you really need "high quality" cells in term of cycle life, but you actually do not need HE cells.

User avatar
DrkAngel   100 GW

100 GW
Posts: 5316
Joined: Dec 15 2010 11:14am
Location: Upstate-Western-Southern Tier NY. USA

Re: Optimal Charge Voltage - (LiCo)

Post by DrkAngel » Nov 30 2019 9:10am

Pajda wrote:
Nov 30 2019 5:12am
Graph without exact definition of testing method and particularly tested cell sample should not be even published, just because you can not add your own measure to compare.
Here you go ... https://www.researchgate.net/publicatio ... Assessment

Just acquired 12 - 36V 5.1Ah LiCo batteries for $300.
12 x 36V x 5.1Ah = 2.2kWh
$136 per kWh


Image


EVE ICR18650/26V looks to have greatest energy density between 15% and 60% (approximately from 3.5V to 3.7V)
I will likely parallel enough 36V batteries to limit controller draw to <1C and limit charge dependent on anticipated need.
1 advantage is the ability to "timer" charge at full Amps, dependent on battery voltage to obtain fastest charge, without the annoying trickle charge incumbent with CC\CV charges.
Set 36V .5C "charger" to 41V but limit with timer to add the required number of Ah.
As a "reserve", there is substantial energy density down to 33V.

Image
A little learning is a dangerous thing;
Drink deep, or taste not the Pierian spring:
There, shallow draughts intoxicate the brain,
And drinking largely, sobers us again.

I enjoy enlightening ... and enlivening the spirit of the innovators.

New & Improved - Acronym Definitions

Index - Homemade Battery Packs - Updated - December 2019

EBike Toolbox - Bargains! $

Endless Sphere Wiki - Lost?

User avatar
flippy   1 MW

1 MW
Posts: 1668
Joined: Aug 12 2015 3:07pm

Re: Optimal Charge Voltage - (LiCo)

Post by flippy » Dec 03 2019 5:30am

Pajda wrote:
Nov 29 2019 3:47pm
Flippy I told you many times that your information about Tesla batteries are not true. All Teslas from the first Model S could be charged to the cell 100% SoC which is 4.2V per cell or 403,6V per battery pack. There are available many videos showing that in range mode or 100% battery the display is showing the tappering voltage of 404V. Also they can be discharged almost to zero, lower than 3V per cell before bricking safety. Actually they lowered the charging voltage to ca 4.15V for the oldest 85 batteries just few months ago due to the problems with accidentl fire during charging sessions of this old packs.
Tesla 18650 and 21700 cells definitelly can not do 10k cycles, at least not at more than 50% DoD. I perform A lot tests of both TMS and TM3 cells and there is nothing interesting about their performance in comparison especially with modern NMC 18650/21700 production.
normally a tesla pack caps at 4.05V (charging at 4.1), only when you enable range mode will will peak the voltage to 4.2V during charging and when its "full" it will cut the carger and everything will sag to 4.1 -at the most- depending on balancing status. it will only go higher for wear compensation AND being in range mode, so a tesla battery that floats at 4.16 is reaching its limit on wear compensation and will show reduced range in the car. also probably cap the fast charging.

also: only the 85/90/100kWh packs are 403V, and the 60~75 packs are 352V as they are missing 2 modules in the "hump".
(the 90kW A generation pack is a whole different thing)

charge voltage is not float voltage.

dial down the agression a notch or 2. i am quite sure i am pretty well versed in how tesla batteries work, i was one of the first in my country to strip one.
Lithium beats liquid dinosaurs.

Pajda   1 kW

1 kW
Posts: 339
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Optimal Charge Voltage - (LiCo)

Post by Pajda » Dec 03 2019 7:48am

flippy wrote:
Dec 03 2019 5:30am
charge voltage is not float voltage.
Tesla have many user selectable charging options, where one of them allows to the user (he is warned by the car that it affect battery life) to set range mode (100% charge shown on display) and charge the cells in CC to the 4.2V (ca 404V pack voltage for "85,90 and 100" packs) and then continue in CV to the "almost zero" cut-off current, that means that Tesla battery can be charged with CC-CV algorithm to "99%". The reason, why the voltage in this particular option slightly drops after finished charging is the permanent vehicle current draw and the cut-off current per cell is not 50 mA but slightly higher.

There are other settings, like 90% which performs only CC phase to 404V and then immediately cut-off the charging process. So the voltage immediatelly drops to ca 4.1V or lower, because of missing CV phase.

User avatar
DrkAngel   100 GW

100 GW
Posts: 5316
Joined: Dec 15 2010 11:14am
Location: Upstate-Western-Southern Tier NY. USA

Re: Optimal Charge Voltage - (LiCo)

Post by DrkAngel » Dec 03 2019 9:46am

.
Soooo ... everyone ... DrkAngel, Pajda, flippy, Tesla is in agreement that charging to higher voltages produces faster battery deterioration!
And apparently everyone agrees that the severity of degradation is relatable to the specific cell formulation.
The only stickler seems to be the advisability of prolonging cycle life compared to initial and eventual usable capacity, which the individual user should decide as a function of their needs.
Recreational user might opt for slightly increased capacity with a more rapid degradation while a commuter would likely prefer prolonging usable lifespan at the expense a slight loss of range or cost of a slightly larger battery.

So could you move your mostly unrelated and petty bickering to some more relevant thread?
A little learning is a dangerous thing;
Drink deep, or taste not the Pierian spring:
There, shallow draughts intoxicate the brain,
And drinking largely, sobers us again.

I enjoy enlightening ... and enlivening the spirit of the innovators.

New & Improved - Acronym Definitions

Index - Homemade Battery Packs - Updated - December 2019

EBike Toolbox - Bargains! $

Endless Sphere Wiki - Lost?

Pajda   1 kW

1 kW
Posts: 339
Joined: Jan 24 2016 3:45am
Location: Czech Republic

Re: Optimal Charge Voltage - (LiCo)

Post by Pajda » Dec 03 2019 11:16am

DrkAngel wrote:
Dec 03 2019 9:46am
.
Soooo ... everyone ... DrkAngel, Pajda, flippy, Tesla is in agreement that charging to higher voltages produces faster battery deterioration!
Of course, but the main problem, which I am talking about is the level of deterioration at "higher" voltages. DrkAngel I have problem with the graphs in the first post of your thread. It is just higly missleading for for the casual reader. I repeat over and over that it is just bullshit that you kill the modern 18650 cells (it means almost all cells produced after 2012) in just 400-500 cycles when you use 100% DoD with 4.2V charging voltage. Many people, which I talked to, is OK with 1000 cycles or 5 years life but they simply thought, based on this "university" articles, that for achieving that it is absolutely necessary to use 70% or lower SoC window.

Scientific articles are aimed to find the absolute values of wear level and they did not comment what this implies for the average use case. If I use as example the cell from the article, which you are posted, or the data from EVE. Both are clearly showing that there are existing cells which can do easily 1000 equivalent full cycles with 4.2V charging voltage and high DoD. So I am only adding my comment to this results: If you choose to charge your battery to 4.1V or lower per cell, the benefit of doing that will be shown in average traction app no more than after 4-5 years or 700-1000 cycles (calendar ageing included) for many modern high quality 18650 cell. And so one should think that for this whole time you will suffer with the bigger, heavier and more expensive battery with lower range.

That is all from me to this topic. :thumb:

eMark   100 W

100 W
Posts: 210
Joined: Nov 02 2019 11:53am

Re: Optimal Charge Voltage - (LiCo)

Post by eMark » Dec 03 2019 11:46am

okashira wrote:
Dec 02 2014 2:06pm
NCA has two advantages over LiCo worth mentioning in regards to a topic many are ignoring.

1.) More energy is stored at the lower voltages. This is good because if you charge a NCA cell to 4.10, it's like 93% capacity compared to 88% for LiCo.
2.) NCA lasts longer when stored at high voltage. Put an NCA cell and a LiCo cell in storage at 4.17V, and the NCA cell will last three times longer.
Please post an authoritative link that claims NCA lasts longer when stored at a higher voltage (e.g. 4.0v-4.1v) than say any NCM 18650 cell stored at 3.7v-3.8v.

DrkAngel is apparently using (LiCo) in reference to NCM and NCA, and not just LCO (LiCo)? Voltaplex's (HK) list of available 18650s is as follows: 74 NMC(NCM), 31 LCO, 17 NCA, 12 LMO. Isn't it true that there's a few NCM(NMC) 18650 cells that are now more popular (performance and cycle life longevity) among ebikers than the Sanyo GA (NCA chemistry) ? That said of the two battery pack options for the Luna Apex one option is a Samsung 21700 28Ah 52v pack and the other option is a Panasonic GA 14Ah 52v pack ... https://lunacycle.com/luna-apex/

In your opinion is the Sanyo GA still the LiCo "Leader of the Pack" among 18650 cells or have a few choice NCM cells now become a better buy for the money (energy/longevity) than the NCA chemistry of the Sanyo GA? Voltaplex (HK) does list the Panasonic Sanyo GA as the Hot (most popular) NCA cell among the 17 NCA cells they list/stock ... https://voltaplex.com/lithium-ion-batte ... chemabv=21

What NCM(NMC) chemistry 18650 cell would you recommend as a better buy (cycle life and performance) for a DIY ebiker build offering more performance and cycle life than the Sanyo GA (NCA chemistry) ?

User avatar
DrkAngel   100 GW

100 GW
Posts: 5316
Joined: Dec 15 2010 11:14am
Location: Upstate-Western-Southern Tier NY. USA

Re: Optimal Charge Voltage - (LiCo)

Post by DrkAngel » Dec 05 2019 11:14pm

Pajda wrote:
Dec 03 2019 11:16am
DrkAngel wrote:
Dec 03 2019 9:46am
.
Soooo ... everyone ... DrkAngel, Pajda, flippy, Tesla is in agreement that charging to higher voltages produces faster battery deterioration!
Of course, but the main problem, which I am talking about is the level of deterioration at "higher" voltages. DrkAngel I have problem with the graphs in the first post of your thread. It is just higly missleading for for the casual reader. I repeat over and over that it is just bullshit that you kill the modern 18650 cells (it means almost all cells produced after 2012) in just 400-500 cycles when you use 100% DoD with 4.2V charging voltage
You do realize "the graphs in the first post of your thread" were posted more than 5 years ago?
Still hasn't changed - Higher charged voltages do cause more rapid deterioration!
But also, thanks for bringing it up, 100% depth (deep) discharges cause more rapid deterioration!
Add on that, higher rate of discharge causes more rapid deterioration!

I merely propose that you can get the most out of, most any, Lithium batteries by using the high energy density portion of it's discharge cycle, at moderate rates. Typically, this high energy density region is reasonably central towards, but not to, empty.
But remember, cell formulations can map very differently!
See- Capacity Mapping
A little learning is a dangerous thing;
Drink deep, or taste not the Pierian spring:
There, shallow draughts intoxicate the brain,
And drinking largely, sobers us again.

I enjoy enlightening ... and enlivening the spirit of the innovators.

New & Improved - Acronym Definitions

Index - Homemade Battery Packs - Updated - December 2019

EBike Toolbox - Bargains! $

Endless Sphere Wiki - Lost?

User avatar
TheBeastie   1 MW

1 MW
Posts: 1993
Joined: Jul 28 2012 12:31am
Location: Melbourne Australia

Re: Optimal Charge Voltage - (LiCo)

Post by TheBeastie » Dec 06 2019 12:23am

DrkAngel wrote:
Nov 30 2019 12:06am
Well, getting back to the tittle topic of LiCo ...
Rather than just full cycles, this chart illustrates optimal voltage ranges of partial cycles.
ImageOptimal partial cycling.jpg

Notable is the difference in deterioration of 60% cycling from 100% to 40% compared to 85% to 25%. (red - green)
More telling is the difference in deterioration of 50% cycling from 100% to 50% compared to 75% to 25%! (dark blue - light blue)
Conclusion seems clear that higher voltages provoke faster deterioration!

.
Fantastic post, especially the difference between the light blue and the dark blue capacity differences vs degradation when technically extracting the same amount of mAh...

One question about the chart is the "number of DST cycles", are you saying you really did discharge the cell 4000 to 5000 times or is this the mAh of the Lico cells? If the cells really can last that long in a good consistent environment than that is really quite interesting also.
I guess also the low charge and discharge rates are also important? C1 is quite mild, it's funny when we see folks post a photo of their 18650 pack complaining they are 'already' losing capacity when its clear the discharges have been so extreme that pack has blackened/barbequed spot-welded metal strip connectors between the cells.

Thanks
Speed Kills Range, 10mph = 46 miles range, 20mph = 20 miles, 30mph = 8 miles rangehttps://goo.gl/1JNL53
Over Charging Kills ur battery bit.ly/1hzWKl4
Consider PAS as your only throttle https://goo.gl/Kg1F8F
Fuel-Cell is the ultimate battery coupled with 4th-gen Nuclear
https://goo.gl/TcKtHs https://goo.gl/ZhFFot https://goo.gl/gfa215
10 Square Miles of solar panels = 0.12GW average power! https://goo.gl/Ub1S39

User avatar
flippy   1 MW

1 MW
Posts: 1668
Joined: Aug 12 2015 3:07pm

Re: Optimal Charge Voltage - (LiCo)

Post by flippy » Dec 06 2019 4:07am

TheBeastie wrote:
Dec 06 2019 12:23am
I guess also the low charge and discharge rates are also important? C1 is quite mild, it's funny when we see folks post a photo of their 18650 pack complaining they are 'already' losing capacity when its clear the discharges have been so extreme that pack has blackened/barbequed spot-welded metal strip connectors between the cells.
this is the actual issue for most people on this forum. they are doing a battery build based only on maximizing capacity and volume with low weight and then just take cell ratings as "more of a guide" instead of actual limits and then start complaining when shit starts exploding, leaking or have a lifespan of a carton of milk on a parking lot in mid summer in downtown Phoenix and genuinely act suprised when their pack has died after a single season.
the idea that lowering voltages will do anything meaningful when killing cells in this manner is not one that should be entertained as it gives a impression that people can abuse cells -but its fine because i stop charging at 4.1v-.
its basically filling up a gas tank halfway but constanty drive at the rpm limiter and thinking the engine will last longer with only half a tank.

in the end its retty simple no matter the chemistry: dont go to 4.2V and stay as far away from 3V as you can and keep as far away from the current limits as you possibly can.
Lithium beats liquid dinosaurs.

User avatar
madin88   1 GW

1 GW
Posts: 3069
Joined: May 27 2013 2:02am
Location: Austria

Re: Optimal Charge Voltage - (LiCo)

Post by madin88 » Dec 06 2019 8:06am

flippy wrote:
Dec 06 2019 4:07am
this is the actual issue for most people on this forum. they are doing a battery build based only on maximizing capacity and volume with low weight and then just take cell ratings as "more of a guide" instead of actual limits and then start complaining when shit starts exploding, leaking or have a lifespan of a carton of milk on a parking lot in mid summer in downtown Phoenix and genuinely act suprised when their pack has died after a single season.
the idea that lowering voltages will do anything meaningful when killing cells in this manner is not one that should be entertained as it gives a impression that people can abuse cells -but its fine because i stop charging at 4.1v-.
its basically filling up a gas tank halfway but constanty drive at the rpm limiter and thinking the engine will last longer with only half a tank.
how true! :mrgreen: :thumb:
Pajda wrote:
Dec 03 2019 11:16am
Of course, but the main problem, which I am talking about is the level of deterioration at "higher" voltages. DrkAngel I have problem with the graphs in the first post of your thread. It is just higly missleading for for the casual reader. I repeat over and over that it is just bullshit that you kill the modern 18650 cells (it means almost all cells produced after 2012) in just 400-500 cycles when you use 100% DoD with 4.2V charging voltage. Many people, which I talked to, is OK with 1000 cycles or 5 years life but they simply thought, based on this "university" articles, that for achieving that it is absolutely necessary to use 70% or lower SoC window.

Scientific articles are aimed to find the absolute values of wear level and they did not comment what this implies for the average use case. If I use as example the cell from the article, which you are posted, or the data from EVE. Both are clearly showing that there are existing cells which can do easily 1000 equivalent full cycles with 4.2V charging voltage and high DoD. So I am only adding my comment to this results: If you choose to charge your battery to 4.1V or lower per cell, the benefit of doing that will be shown in average traction app no more than after 4-5 years or 700-1000 cycles (calendar ageing included) for many modern high quality 18650 cell. And so one should think that for this whole time you will suffer with the bigger, heavier and more expensive battery with lower range.

That is all from me to this topic. :thumb:
Hi Pajda,
i have the meaning that we should differentiate between the cylce life of one single cell, and the lifetime of larger packs build of dozend or hundred of cells.
It is true that tests are showing that the benefit of charging to only 4,1V will be shown not before some years or hundreds of cycles (in terms of energy that could have been used compared to a full charge), but they normally are referring to one single cell as far as i could see, right? Please correct me if i am wrong.

The problem is that there is one layers in a LiIon cell (Graphite and Silicon on Anode) which does change it's size or volume during charge and discharge. This layer does expand during charge and it shrinks during discharge which could lead to internal damage the more often this happens. Internal damage which could show in small short circuits between the layers and higher self discharge.
Thats not a big problem if one cell from your flashlight or e-cig did suffer this, you probably will replace it without thinking too much about it, but it will be a problem once you have larger packs with one or more cells like this. Even if you have a good BMS and all cells always kept at same voltage, i believe that each cell has it's "own life" or "behaviour" in terms of that.

I have processed thousands of 18650 cells, mostly to repair ebike batteris from Bosch, BionX etc, but to speak about the ones which were custom made and which are used in my near field by friends, i can say the following:
Those which got charged to 100%, or where the charger was set to 4,2V, are practically dead after around two years or 5000km due to aweful high self discharge of a few cells (these numbers are from a 20s9p Samsung 30Q pack for instance), while the ones which are charged to only 4,1V are still fine after many years and show no drift.

When the first 18650 LiIon cells came out, the max charge voltage always was 4,1V. I remember that RC charger which had 4,2V for LiPo and 4,1V for LiIon as fixed CV cut off.
I think the big companies (ebikes, power tools etc) have nothing against that sort of "planned obsolescence" because they can sell a new pack to you every few years, while car makers like Tesla and DIY people don't do this for a good reason.

User avatar
DrkAngel   100 GW

100 GW
Posts: 5316
Joined: Dec 15 2010 11:14am
Location: Upstate-Western-Southern Tier NY. USA

Re: Optimal Charge Voltage - (LiCo)

Post by DrkAngel » Dec 07 2019 7:51am

DrkAngel wrote:
Nov 30 2019 9:10am
EVE ICR18650/26V looks to have greatest energy density between 15% and 60% (approximately from 3.5V to 3.7V)
I will likely parallel enough 36V batteries to limit controller draw to <1C and limit charge dependent on anticipated need.
1 advantage is the ability to "timer" charge at full Amps, dependent on battery voltage to obtain fastest charge, without the annoying trickle charge incumbent with CC\CV charges.
Set 36V .5C "charger" to 41V but limit with timer to add the required number of Ah.
As a "reserve", there is substantial energy density down to 33V.

Image
Older LiCo cells were near empty at 3.6-3.7V, this provided a reasonably narrow power band of 42V-37V full to empty for a 12% differential.
Newer LiCo like the Samsung ICR18650-22p "emptied" near 3.4V for a ~20% differential.
Even worse, the EVE ICR18650-26V cells I just ordered near "empty" at 3.3V for a ~22% differential.
This "differential" is the difference between performance from a full battery and a near empty battery.

As displayed a 42-33V differential has a drastic speed-torque-power variance.
42-33V.jpg
42-33V.jpg (245.25 KiB) Viewed 1152 times
Then I noticed a very dense region of energy in these EVE cells. A very large percentage resided between about 3.5V and 3.7V, more than from 3.7V to 4.2V.

Image

How best to take advantage ... ?
Well 37V to 35V is a nice tight differential ~`5.4%. But 37V is fairly pitiful on an eBike used to 42V!
So, decided to build a battery and add 1s to the 10s BMSed battery packs.
Current plan is for 4x 10s2p +1s10p for an effective 11s8p, with larger added 1s as safety, being without BMS support. (4.5 packs = 10lb - $112.50)
This will give me (11x 3.72V) 41V "full" and (11x 3.5V) 38.5V "empty". Now that is a performance differential that I can live with.
Compare to above - More than 4 MPH to barely 1 MPH
41-38.5V.jpg
41-38.5V.jpg (243.03 KiB) Viewed 1152 times
This would give me near 10Ah for daily use, with a nice safety buffer towards empty and a massive additional 11Ah if I decide to "fully charge" to 46.2V for some long trip?. ( I built and charge with "MeanWell Mega-mod" S-150-48 modded with dual meter and dual external pots for 29-54V, 0-5A charge customization)

The 15-60% cycle, within the best energy rich voltages, and the large pack ensuring low C - rate should ensure a very long life with stable performance throughout each cycle! (near the voltage stability of LiFe)
.
Last edited by DrkAngel on Dec 09 2019 5:36pm, edited 1 time in total.
A little learning is a dangerous thing;
Drink deep, or taste not the Pierian spring:
There, shallow draughts intoxicate the brain,
And drinking largely, sobers us again.

I enjoy enlightening ... and enlivening the spirit of the innovators.

New & Improved - Acronym Definitions

Index - Homemade Battery Packs - Updated - December 2019

EBike Toolbox - Bargains! $

Endless Sphere Wiki - Lost?

User avatar
flippy   1 MW

1 MW
Posts: 1668
Joined: Aug 12 2015 3:07pm

Re: Optimal Charge Voltage - (LiCo)

Post by flippy » Dec 09 2019 4:24am

yes, caring for your battery and using it within its specs extends its life.

i generally recommend cranking the voltage of the battery so you can simply cap the top speed so you dont lose speed when draining the battery or driving with partial charge often.
Lithium beats liquid dinosaurs.

Post Reply