Best 18650 battery (with criteria)

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
I want a battery capable of at least 8-10A cont. drain. higher capacity is better. unprotected. Am curious what battery will retain charge the longest in storage, as well as have the longest calendar life with minimal cycling (Maybe drain/charge once a month, and sometimes maybe not for 6mo). The two items in red are non-negotiable, as if they are not conditions that are met, the battery will not function in the role selected for it. Given this criteria, what is the best product on the market?
 

ChrisGarrett

Flashlight Enthusiast
Joined
Feb 2, 2012
Messages
5,725
Location
Miami, Florida
I want a battery capable of at least 8-10A cont. drain. higher capacity is better. unprotected. Am curious what battery will retain charge the longest in storage, as well as have the longest calendar life with minimal cycling (Maybe drain/charge once a month, and sometimes maybe not for 6mo). The two items in red are non-negotiable, as if they are not conditions that are met, the battery will not function in the role selected for it. Given this criteria, what is the best product on the market?

Sanyo-Panasonic NCR18650GA.

Chris
 

john61ct

Newly Enlightened
Joined
Jan 17, 2017
Messages
183
You mean just 1P?

Doing 3C continuous will mean nothing no matter how good will last many cycles

LG M36 aka MJ1
SONY VC7
Samsung 35E

The GA is good but not quite as consistent
 

dotCPF

Flashlight Enthusiast
Joined
Dec 20, 2018
Messages
1,702
Location
New Angleland
Sanyo NCRGA. Specifically the Japanese mfg but the Chinese Panasonic factory one isn't far off (and still tops out most charts). My cell of choice. Have about 50 of em at this point.

https://endless-sphere.com/forums/viewtopic.php?t=91821

http://52.25.253.50/forums/viewtopic.php?t=93719

The GA is good but not quite as consistent

Testing I have seen says the GA is a bit more consistent. My experience has been similar as well. Do you have any data or links? (not trying to be annoying, definitely wanna know!)



Though you'd be hard pressed to go wrong with the MJ1 either, that would be my second choice no question.
 

dotCPF

Flashlight Enthusiast
Joined
Dec 20, 2018
Messages
1,702
Location
New Angleland
WOW that NASA test is really interesting! Thanks for that.

My only thought on that is, they are probably testing for all kinds of things that have little, if any, impact on us terrestrial earthlings. That being said, they probably DID push the limits and find the extremes, great data to have.

Cheers!
 

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
I have been reading that if I cut off charge voltage at 4.1V, I will have 90% capacity, but will drastically lengthen service life of the battery? Does this mean a 3300mah battery would perform equal to a 3000mah battery with a full charge, if charged to 4.1V? If this is true, should I FULLY charge them for the first usage or two? Or never EVER above 4.1v? Is this like a car engine that you do not want to never rev to redline and THEN expect much from, or...?

Also, what battery will have the longest shelf/storage life if charged up and simply left in the light/storage for years, given all of my criteria EXCEPT capacity/mah? Is the answer the same, or is there a lower mah battery that meets the other criteria that has much better chemistry for "We need these on point in 20 years...".

I ask, as my humble 2005 battery pack harvest batteries endured years of abuse, and have held at 4.16v for over a year in storage, now, after sagging from 4.20v after 30 days, no further degradation, 11mo out.
 

WalkIntoTheLight

Flashlight Enthusiast
Joined
Jun 18, 2014
Messages
3,967
Location
Canada
I have been reading that if I cut off charge voltage at 4.1V, I will have 90% capacity, but will drastically lengthen service life of the battery? Does this mean a 3300mah battery would perform equal to a 3000mah battery with a full charge, if charged to 4.1V?

Correct.

If this is true, should I FULLY charge them for the first usage or two? Or never EVER above 4.1v?

The forming charge of the battery is done at the factory. So, there's no need to do anything special when you get the battery. Just charge it to 4.1v. If you ever need a bit more capacity, then top it up to 4.2v. Lithium-ion cells have perfectly happy getting partial charges and discharges.

I usually do a capacity test with a battery analyzer when I get new cells, and that automatically charges to 4.2v. Also, I like to charge to 4.2v and leave it for a few days, to see if there are any internal shorts that will slowly discharge the cell. But, I don't really need to do that. It's just for peace of mind.

Also, what battery will have the longest shelf/storage life if charged up and simply left in the light/storage for years, given all of my criteria EXCEPT capacity/mah? Is the answer the same, or is there a lower mah battery that meets the other criteria that has much better chemistry for "We need these on point in 20 years...".

We don't have any long-term results on any of the decent cells you can buy nowadays. Because, they've only been around for a few years at most.

I ask, as my humble 2005 battery pack harvest batteries endured years of abuse, and have held at 4.16v for over a year in storage, now, after sagging from 4.20v after 30 days, no further degradation, 11mo out.

Yup, lithium-ion cells hold storage remarkably well. You will notice that the capacity does drop over time, especially if stored at full charge. But, it's not as bad as what some people claim.
 

john61ct

Newly Enlightened
Joined
Jan 17, 2017
Messages
183
LFP is a much better chemistry for longevity, both calendar and cycle lifetime.

A123 is a top notch maker, was in on the initial development of LFP out of MIT, now part of LithiumWerks.

I am more familiar with their bigger cells, much more convenient for building high Ah capacity packs at high voltages

but I believe they still sell the tiny 18650 cells also.

Not cheap.

And LFP has way lower energy density, why it is rarely used in propulsion use cases, although Tesla's Chinese models are / will be using LFP packs from CATL
 

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
LFP is a much better chemistry for longevity, both calendar and cycle lifetime.

A123 is a top notch maker, was in on the initial development of LFP out of MIT, now part of LithiumWerks.

I am more familiar with their bigger cells, much more convenient for building high Ah capacity packs at high voltages

but I believe they still sell the tiny 18650 cells also.

Not cheap.

And LFP has way lower energy density, why it is rarely used in propulsion use cases, although Tesla's Chinese models are / will be using LFP packs from CATL

These?

https://www.18650batterystore.com/L...99iihbA31XMQMrOerDYkZCJnNiamDnIEaAlp0EALw_wcB
Do you have any data on your claims? I am not doubting, just looking for quantification, and figure you know best where to find it. Also how do they perform at very low temps vs the others in this thread?

I also note their nominal voltage as 3.3. My application will not function below 3.0v.
 
Last edited:

john61ct

Newly Enlightened
Joined
Jan 17, 2017
Messages
183

Yes those are the latest version, likely better than the below, as per older "A" model.

Stored in coolish conditions at 50% SoC isolated / no charge / no load will lose maybe 10% capacity at most over twenty years.

Cycled between resting 4.35V as 100% and

avg 80% DoD

below 1C current rate

maximum LVC 3.1V as 0% DoD only occasionally

charged at 0.4-5C in warmish temperatures

will go 7-8000 cycles, EoL defined as 70-75% SoH.

> Do you have any data on your claims?

Nope, there are a dozen variables that impact longevity, they all interact with each other, industry is focused on high-C rate EV type applications where they "only" last 3-5000 cycles.

> Also how do they perform at very low temps vs the others in this thread?

If by very low you mean in 0°C territory, the only LI chemistry that stands out is LTO, real odd man out.

With all the rest including LFP, charge rates need to **drastically** get reduced as 0° is approached

fast charging in freezing temps can instantly render the whole bank scrap.

Active warming is often implemented as needed, for very fast charging even to 35-40°

Discharging in cold temps does not cause damage, but rapidly becomes slower and much less efficient, as with old-school lead banks.
 

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
Yes those are the latest version, likely better than the below, as per older "A" model.

Stored in coolish conditions at 50% SoC isolated / no charge / no load will lose maybe 10% capacity at most over twenty years.

Cycled between resting 4.35V as 100% and

avg 80% DoD

below 1C current rate

maximum LVC 3.1V as 0% DoD only occasionally

charged at 0.4-5C in warmish temperatures

will go 7-8000 cycles, EoL defined as 70-75% SoH.

> Do you have any data on your claims?

Nope, there are a dozen variables that impact longevity, they all interact with each other, industry is focused on high-C rate EV type applications where they "only" last 3-5000 cycles.

> Also how do they perform at very low temps vs the others in this thread?

If by very low you mean in 0°C territory, the only LI chemistry that stands out is LTO, real odd man out.

With all the rest including LFP, charge rates need to **drastically** get reduced as 0° is approached

fast charging in freezing temps can instantly render the whole bank scrap.

Active warming is often implemented as needed, for very fast charging even to 35-40°

Discharging in cold temps does not cause damage, but rapidly becomes slower and much less efficient, as with old-school lead banks.
I have looked up as much data as I can find, and the discharge data for this chemistry doesn't show to be special at all.
https://www.batteryspace.com/prod-specs/9444.pdf
 

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
What do you mean by that?
Its ability to retain charge in storage/not being used does not exceed other much higher mah capacity batteries. 50% is only 5-600mah, while a 3500/2=more than the LFP starts with fresh and full.
 

john61ct

Newly Enlightened
Joined
Jan 17, 2017
Messages
183
Its ability to retain charge in storage/not being used does not exceed other much higher mah capacity batteries. 50% is only 5-600mah, while a 3500/2=more than the LFP starts with fresh and full.
I did state LFP is less energy dense.

It does retain its SoC much longer, in fact over just a few years loses hardly any at all.

But the two factors have nothing to do with each other anyway.

What 50% are you talking about?

If you want higher Ah capacity per cell, why are you messing around with these cylindricals?

How large are the packs you are building?
 

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
I did state LFP is less energy dense.

It does retain its SoC much longer, in fact over just a few years loses hardly any at all.

But the two factors have nothing to do with each other anyway.

What 50% are you talking about?

If you want higher Ah capacity per cell, why are you messing around with these cylindricals?

How large are the packs you are building?
This is for a weapon mounted light.
https://modlite.com/collections/wea...kw-18650-light-package?variant=20414940479547

I intend to feed it the most reliable food possible. I could sacrifice the runtime, but at 3% loss per month, it does not appear that the LFP battery is superior to the 15 year old LiIon cells I harvested from a laptop, which have only regressed from 4.20 to 4.15 volts over the course of 1 year.
 

john61ct

Newly Enlightened
Joined
Jan 17, 2017
Messages
183
3% loss is a complete fabrication.

Truth is well under half a percent per year.

Is 26650 size physically too big?
 

N/Apower

Enlightened
Joined
Jul 30, 2008
Messages
536
3% loss is a complete fabrication.

Truth is well under half a percent per year.

Is 26650 size physically too big?

Yes, 18650 or 18350 are the only two options.
Why would a company lie about their product much to their detriment in their data sheets?
 

john61ct

Newly Enlightened
Joined
Jan 17, 2017
Messages
183
I do not know of course, but my theory is if they told the truth people would not believe it, "everyone knows" batteries have some self-discharge rate, and they think their specs would lose credibility?
 
Top