18650's Care and Maintenance

[ktate749]

Thanks terjee clears it up.

 

[degarb]

One cell phone app, on playstore, claims 55% charge does zero damage to cell. But I doubt that. Moreover, isn't there a 3 year shelf lifespan on Lithium Ion Chemicals in the cells?

Rather than .025% damage at 70%, I would feel better, if the chemist that makes the cells, had a voltage charge to kilo amp hour duty over entire cell life at 1 amp and .350 amp discharge.

While I am a skeptic as to wisdom of undercharging, I do hope it will make people think about demanding higher cell capacity formats.

 

[Gauss163]

One cell phone app, on playstore, claims 55% charge does zero damage to cell [...]

Generally using shallower cycles will prolong life, but it certainly won't completely eliminate cycle degradation. Further, to maximize longevity, besides using shallow cycles you should center them around 50% SOC, e.g. if you're going to use cycles of 50% depth then instead of using 0-50% you should center them, i.e. use the 25-75% region. See this post for further details, including graphs of results from studies on such matters.

Moreover, isn't there a 3 year shelf lifespan on Lithium Ion Chemicals in the cells?

No, quality cells have much longer life if stored properly. But that may apply to certain installed batteries that are subject to external parasitic loads that could drain them much quicker than self-discharging would. In particular if they drain too low it may no longer be safe to charge them so the BMS will prevent such.

 

[etc]

What the voltage of two 18650 ..........is it more than 9 volts ?
LOL.....I know nothin
I have a Malkoff M-61 drop in that I want to use in a two 18650 batt body ...........can I ??

Malkoff M61 in a 2-cell body, with two 18650 is a very good idea. You will get extra lumens and great runtime.

The current will be lower, meaning the cells will be hit less. Generally the more batteries you include, the less each one is drained individually.

M61 series of course can only take up to 2 Li-Ion cells. Other Malkoff drop-ins can take up to three Li-ion cells. E.g. the Hound Dog.

 

[etc]

Generally using shallower cycles will prolong life, but it certainly won't completely eliminate cycle degradation. .

Interesting point you make. I've always thought the same thing also. I agree.
On the one hand, you don't want many recharge cycles but it appears that shallow cycles are better than discharing the thing down to 2.5V - beyond empty.

 

[etc]

I would prefer to have protected cells in my collection because they protect from overcharge and overdischarge by disconnecting the battery

Yeah. *But*, one huge problem with protected cells is that they suddenly go out. In a multi-cell configuration but even 1x18650. The circuit kicks in and you are down.

Now imagine you are in a cave. Or hanging on a mountain. Or in any number of other dangerous situations where suddenly you lose all light. There is no warning period, it goes "Poof".
Yeah, I realize you have a spare that's 30 seconds away but that can be critical.

This is one *huge * advantage of running 1x18650 lights off *un* protected cells since you will get a long, gradual decline into the sunset.

M61 case in point, or 18650 Hound Dog.

That is one huge advantage of running your lights off 123s, as expensive as that might be. They don't go out in a nanosecond.

 

[WalkIntoTheLight]

This is one *huge * advantage of running 1x18650 lights off *un* protected cells since you will get a long, gradual decline into the sunset.

It depends on the light. Some lights have built-in low-voltage protection. However, those lights usually have a period of step-downs, which give you plenty of warning. For example, Zebralights have a low-voltage protection at 2.7v, but they also step-down well before you're left in the dark. In the case of Zebralight, the built-in LVP would kick in sooner than the 2.5v LVP of a protected cell.

With most 1x18650 lights that don't use boost drivers, it would be tough to actually be left in the dark with a protected cell. The voltage is just too low for the forward voltage of the LED.

 

[etc]

It depends on the light. Some lights have built-in low-voltage protection. However, those lights usually have a period of step-downs, which give you plenty of warning. For example, Zebralights have a low-voltage protection at 2.7v, but they also step-down well before you're left in the dark. In the case of Zebralight, the built-in LVP would kick in sooner than the 2.5v LVP of a protected cell.

None of my Surefires or Surefire clones or Malkoff MDx-series have built-in low-voltage protection. I don't have any Zebralights. Neither do most people that build LEGO assemblies.

With most 1x18650 lights that don't use boost drivers, it would be tough to actually be left in the dark with a protected cell. The voltage is just too low for the forward voltage of the LED.

The cell declines to the voltage where the PCB is triggered and you go from xxx lumens to zero lumens in a nanosecond.

That is how every protected cell has worked with every Lego light with every P60-drop-in module I've had in the last 10 years. This is kind of good as it protected the cell from overdischarge, that's the point of it. It's not so good if you critically need light that very second. Case in point, you are an LEO holding a suspect or suspects in a dark basement. Your lights go out as the cell hits the threshold for cutting out, whatever it is, 2.5V I assume, and that jeopardizes things. I can think of 100 and 1 situations like that.

Not only does an unprotected cell give you some warning by gradually declining but in an emergency, when you have no spare to reload, you can run it down as low as you have to maintain light. Quality 18650 only cost $7 these days, we are talking 18650 3400 mAh. I remember when they first came out, which seems like yesterday and were close to $20/each.

Really risking $7 is not that much more expensive than primaries that are 2 bucks a piece retail. Use it several times and you break even.

 

[WalkIntoTheLight]

The cell declines to the voltage where the PCB is triggered and you go from xxx lumens to zero lumens in a nanosecond.

That is how every protected cell has worked with every Lego light with every P60-drop-in module I've had in the last 10 years. This is kind of good as it protected the cell from overdischarge, that's the point of it. It's not so good if you critically need light that very second. Case in point, you are an LEO holding a suspect or suspects in a dark basement. Your lights go out as the cell hits the threshold for cutting out, whatever it is, 2.5V I assume, and that jeopardizes things. I can think of 100 and 1 situations like that.

My point is that with (non boost-driver) lights, they will already be very dim by the time the protected cell shuts off at 2.5v. For example, an XP-L emitter at 2.5v will be barely lit, if at all. IOW, there's not going to be much practical difference between running the light with a protected or unprotected cell.

This is completely different if your light has a boost-driver, such as a Zebralight. But, AFAIK, those kinds of lights all have built-in LVP anyway, so it will still be the same regardless of your cell protection.

 

[Gauss163]

Interesting point you make. I've always thought the same thing also. I agree.
On the one hand, you don't want many recharge cycles but it appears that shallow cycles are better than discharing the thing down to 2.5V - beyond empty.

This works because the shallower cycles (centered around 50% SOC) serve to keep the cell away from extreme voltages (which accelerate degradation). Generally to prolong life you want to minimize the amount of lifetime the cell spends at extreme voltages (and temperatures) - both in cycling and storage. Even small changes can yield large improvements because the degradation processes are nonlinear. Follow the link I gave above for further details.

 

[AlpineBatteries]

This is such an fascinating and illuminating thread. I just bought a new Lenovo laptop, and among the pre-installed Lenovo firmware is a function called "Conservation Mode". When activated, it prevents the battery from charging beyond 60%. Since I rarely unplug and carry my laptop around for any extended period of time, I keep the setting on. If I know I'm going to unplug for a while, then I'll deactivate the setting and charge it all the way up first.

Another interesting tidbit is that my father just got a Tesla Model 3, and told me that Tesla recommends not topping off the battery. So it's nice that the company is spreading the word about best practices for Li-Ion care.

 

[Monocrom]

I've never experienced any issues with topping off my 18650 and similar cells. One of the biggest things is not to use a fast-charger. Might be annoying waiting 5 hours to charge up one 18650, but patience is worth it. Last thing you want is an 18650 becoming the world's greatest hand-warmer the instant you take it off of a charger.

 

[WalkIntoTheLight]

I've never experienced any issues with topping off my 18650 and similar cells. One of the biggest things is not to use a fast-charger. Might be annoying waiting 5 hours to charge up one 18650, but patience is worth it. Last thing you want is an 18650 becoming the world's greatest hand-warmer the instant you take it off of a charger.

Yeah, I think some people get too anal about managing their cells to perfection, including myself sometimes. Though, I'm trying to be less strict about it, and just charge and use most of the time. I usually try to charge at 1 amp, because I'm not in a big hurry. A lot of chargers now seem to default to 2 amps, which is okay for high-drain cells, but it's still a little more stressful than 1 amp.

If I notice the voltage on the charger is up to about 4.1v, I'll often pull the cells off early. That's about a 90% charge, which is fine for most of my use. Lately, though, I'm usually letting it go to the full 4.2v. Again, trying to be less anal. Most of my cells will probably tire out from age, before they wear out from cycle counts.

Note that charger specs are usually 4.20v +/- 0.05v. That means that some chargers might charge your cells up to 4.25v, which IMO is putting more stress on them than you probably want. You'll get about 5% more capacity out of them, but probably lose 25% of the cycles.

As a rough rule, you can charge up to 4.3v, and lose half your cycles for a 10% gain in capacity. Or, charge to 4.1v, and double the number of cycles, for a loss of 10% capacity.

 

[Scimmy]

I wanted to share this video. It explains protected vs. unprotected 18650 cells. He does an excellent job explaining the dangers of over charge and over discharge whole showing how the protection circuit works.

https://www.youtube.com/watch?v=1rg3ZWxBNUE&t=31s

Bud

Thanks for the video, very interesting.. They sure can be dangerous if not treated properly

 

[doctordun]

Ok, I encountered a question trying to determine how long a protected 18650 cell rated at 3100mAh would power a 5 volt led light running at a drain of .07 amps. In actually testing I got 24 hours before the light stopped. A second test with the light function at .23amp draw and it lasted around 10 hours.
I have read that 18650 batteries have from 2 amp hours to 3.5 amp hours of available power.
In a protected battery, I would assume that numbers would be lower because of the protection circuit.
I would like like to get access to more data and perhaps a spreadsheet with formulas to play with.
How much does the amp hours vary with the mAH ratings? Does the protection circuit limit a percentage of available power output and if so, how much.
Any and all information would be most welcome.
Thank you. all.
Albert

 

[SilverFox]

Hello Albert,

Each cell manufacturer and sometimes each batch of cells can vary in capacity. The labeled capacity on the cell is often "optimistic." This means that if you want accurate information you end up having to do some cell testing yourself.

HKJ has done a lot of testing and posts results frequently. In addition he has a website that lists the results from the various tests. If you can find your brand you may be able to get an idea of how long your cell will last under various loads.

Right now you have one run with a drain of 70 mA that lasted for 24 hours. This would give the cell capacity starting at 70 x 24 = 1680 mAh. Your second run had a drain of 230 mA for 10 hours giving a cell capacity starting at 230 x 10 = 2300 mAh. At first glance it would seem your first run was done with a cell that wasn't fully charged...

The capacity of a cell is related to how fast the current is used from it. There is a point where the cell heats up and energy that could be used is consumed producing heat. I don't think your loads are in that range so something else is going on with the differences in run time.

It looks like you need to do a few more tests involving runtime to get a better idea of how your cell is performing. Once you have that down you can repeat the testing using other cells to see how they match up to what you have now.

Tom

 

[wayben]

Great thread, lots of useful info!!!

 

[PaladinNO]

I know this is an old thread, but how does the batteries handle cold?
Specifically, having a 16650 / 18650 in a flashlight in a car all year, including during -35°C? The runtime is obviously degraded, and I try to stay at low outputs if it is too cold and I have to use the light at all, but can the cold temp damage the cell just sitting there, with a charge?

And are those "cold resistant" cells something special (such as Nitecore NL1835LTHP), or just BS?

I do take the light indoors, letting it get to room temp first for safety, and then charge the cell occasionally.

 

[fulee9999]

in short, not well... electric cars have batterypack heaters for a reason, and the reason is that li-ion cells don't really like the freezing weather.
Anywhere below freezing you should use frost resistant cells like the molicel P28A, but for just leaving it in the car I think you really should use lithium CR123 primaries, they handle it a lot better.
Also leaving li-ions "just sitting there" is a very bad idea, they usually self-discharge over time a lot faster than primaries.

 

[PaladinNO]

Thank you. Sounds like I either need to revert to CR123's, or get a new (yellow) flashlight to keep in the car then.
The Wurkos Golden WK03 seems interesting, though I would have preferred an inexpensive plastic light, because this light frequenctly gets rattled around in the door pocket, next to a small can of emergency oil.

Currently got a Streamlight PolyTac. Yellow colour for visibility, and it got ON or OFF. Just what I need. The PolyTac is 2x CR123 by default, but I found it works with a single 16650, so I "upgraded" it to that, because the light never seemed to work when I really needed it while I used CR123's. The KeepPower 16650 has been flawless for many years now, despite the very infrequent charging and use.

EDIT:
Forgive me if I'm mistaken, but...is your avatar a picture of Bud Spencer?
I love those spaghetti western movies of his, either alone or with Terence Hill!