3
\$\begingroup\$

I recently got an inexpensive (~$4 USD) Li+ 18650-powered flashlight from banggood.com, one of my favorite sources for inexpensive Chinese electronics to tear down and hack for fun :)

I wanted to validate its safety circuit before considering putting it into active service. I'm not liking what I found, but am wondering if I'm being too conservative. I'm having a hard time shaking visions of a Li+ battery charging on my counter bursting into flames suddenly :)

The flashlight takes a single 18650 non-protected Li+ cell. The flashlight won't turn on (draw any more than 10uA) when the voltage is below 2.37V. However, once on, it continues to draw current down to about 1.95V. Also, until it reaches that shut-down threshold, the turn-on "hysteresis" is defeated; that's basically how I've defined shut-down.

Here's my test setup. Only the LED holder and switch assembly are shown; everything else is mechanical (and ground conductor):

enter image description here

Here's a table of the various voltages (on the way down, lower brightness setting):

 V+  I (mA)
---- ------
4.2  65.5
3.8  61.4
3.4  45.2
3.0  22.2
2.7   6.3
2.6   2.5
2.5   165uA
2.4    83uA
2.3    60uA
2.2    46uA
2.1    36uA
2.0    28uA
1.95   0.2uA

And on the way up:

V+   I (mA)
---- ------
1.5  0.0uA
1.7  0.1uA
1.9  0.1uA
2.0  0.3uA
2.1  0.6uA
2.2  1.1uA
2.3  1.9uA
2.4  70.3uA
2.5  165uA
2.6  2.5mA

After this on the way up the V-I characteristics match those on the way down.

So I do note that the current consumption falls precipitously at about 2.5V on the way down (2500 -> 165 uA), and that nothing much happens at all until 2.4V on the way up. What I'm wondering is whether this is enough to constitute adequate protection.

The scenario I'm imagining is that someone places the flashlight vertically, lens down, on a desk, and then forgets it's turned on since the light isn't visible. If left for a couple weeks like this, I'm thinking the battery would discharge to levels that would make it prone to violent misbehavior on re-charge.

What I rather expected to see was the current drop a little sooner, say 2.7V, since the discharge rate wouldn't be great enough to allow much recovery. Then not take more than protection IC current up to about 3.0V on the way up.

Am I being over-conservative?

\$\endgroup\$
  • 1
    \$\begingroup\$ Seems like a pretty deep discharge for a protected 18650. \$\endgroup\$ – Scott Seidman May 18 '16 at 0:16
  • 3
    \$\begingroup\$ If you have a functional cell that is truly 2000mAh, you can expect its self discharge rate to equate to 0.2mA (0.1 to 0.35mA depending on build quality). That said, anything that doesn't show a discrete turn off is highly probably unstable and dependant on everything that surrounds it and should live on the inside of a trashcan. Which goes for the majority of stuff on banggood, eBay China and Alix. (Buy on the latter two myself, but with extreme prejudice and caution only) \$\endgroup\$ – Asmyldof May 18 '16 at 0:25
  • 1
    \$\begingroup\$ If it's a LiFePO4 (lithium iron phosphate) cell, then there is a remote chance that it's borderline safe. \$\endgroup\$ – Nick Alexeev May 18 '16 at 0:26
  • 1
    \$\begingroup\$ Try also posting in a flashlight forum, e.g. the CPF battery subforum. \$\endgroup\$ – Bill Dubuque May 18 '16 at 0:43
  • 2
    \$\begingroup\$ I merely want to validate the OP's concerns. The battery is being over-discharged in this light. \$\endgroup\$ – mkeith May 18 '16 at 4:13
1
\$\begingroup\$

Time Bomb

The true problem is the invisible damage that happens inside the cell, producing a few atoms short of a major short-circuit, until a little vibration, jarring, or a dropped flashlight, moves it the rest of the way, completes the circuit, and you find out you're holding a roman candle.

Active Cutoff Needed

In order for this flashlight to be safe with this battery, it would have to have a more active cutoff, because without it, you will eventually reach the 1.5 voltage point, and then this happens:

Copper Shunting

At battery university, at "BU-808a: How to Awaken a Sleeping Li-ion", the following section can be found:

Do not boost lithium-based batteries back to life that have dwelled below 1.5V/cell for a week or longer. Copper shunts may have formed inside the cells that can lead to a partial or total electrical short. When recharging, such a cell might become unstable, causing excessive heat or show other anomalies.

Venting With Flame!

Then, also at Battery University, on the page, "BU-304b: Making Lithium-ion Safe", it contains this little gem of what might be considered a good example of ultra-cheap-manufacturing double-speakery:

Manufacturers of lithium-ion batteries do not mention the word “explosion” but refer to “venting with flame” or “rapid disassembly.” Although seen as a slower and more controlled process than explosion, venting with flame or rapid disassembly can nevertheless be violent and inflict injury to those in close proximity.

The Longer I Think About This...

So, what happens if you charge the over-discharged battery, and it has shunts, but they smolder, or otherwise don't actually cause a problem until you've fully charged the battery, and screwed it up tight in the flashlight. Do you now have a grenade? I just wouldn't play with this. Convert it to use protected-cells, if you just can't let it go to waste. Let's say somebody else accidentally gets this flashlight? (a thief breaks in, and takes your flashlight for his little girl). Don't let it happen... I know... Put it through your own version of the inquisition: "convert, or die!" ;-)

Discharged == Harmless

In a talk by Professor Jeff Dahn of Dalhousie University entitled "Why do Li-ion Batteries die, and how to improve the situation", we find out the following things: Over-discharged cells are not dangerous until you recharge them. When Lithium ion cells are made, apparently, their ingredients are relatively safe, because Jeff says that you can assemble them in the open air. But when you charge them, both anode and cathode (and it's a jellyroll, so you have plenty) become very reactive.

So, then, based on Jeff's talk, if a Lithium ion cell is fully discharged, then it is no longer dangerous, but should at that point no longer be used, but recycled. I run mine down to zero volts, then I put them in a group to be recycled together. If somebody else should throw them away, they shouldn't cause a fireworks show in the trash truck while it's compacting because they have no stored energy, and there is no longer the highly reactive intercalated Lithium.

Charging Only to 4.1 Volts is Best

In the same video, which I consider to be some of the most authoritative information on the subject (I've watched the video many times for that reason), Jeff says a few things which stick in my mind and I want to pass on:

  1. "Time of exposure to high temperature while charging or discharging is the bad thing." He's talking about the electric cars and some real problems some of the manufacturers have gotten themselves into, most notably, one manufacturer who decided NOT to thermally manage their cells. (Think summer in Arizona, and discharging cells to drive a car, which produces a LOT more heat, and bad chemical reactions that quickly kill the cells under those conditions.)

  2. If you're storing them, do not store them fully charged. Discharge them so they only have a little left. And store them in a cool place, but not in the freezer. Refrigerator or basement would be good. For me, this is hard to do, so my compromize is to charge them just before I use them, with a little cool-down time in-between. And I try not to leave my rechargeables in the car!

  3. Charge to only 4.1 volts. (Found this in the Q&A at the end of the video). Then, in Jeff's words, "if they put the cell in your body, they'll last until you're dead... or until I'm dead..." Made his inquisitor speechless.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.