Worst case for low drain batteries in high drain flashlight?

vees · 2019-11-11T14:38:37.000Z

What are the most probable bad things that could happen putting a low-drain 18650 into a high drain direct-drive (D4, FW3, etc.) flashlight for regular use?

What is the likelihood of danger to the user or damage to the flashlight or battery?

Scenario 1: Use regulated mode or lower always.

Scenario 2: Regulated mode mostly, but accidentally click to turbo once or twice per charge cycle.

Scenario 3: Sustained turbo or direct drive use.

fixed_it · 2019-11-11T14:59:55.000Z

Someone correct me if I’m wrong… but I think it should go roughly like this:

1: Nothing
2: Probably not much besides lower output, assuming you don’t hold it too long.
3: Heat, lots of it. Good batteries should have a fuse that might melt and leave the battery dead. Crap batteries… anything goes. But that’s absolute worst case. Most likely you’ll just reduce battery life.

Keep in mind that even a “low drain” 18650 can put out a lot of current safely. In small lights like the ones you mention, you’re likely to burn your hands long before you damage the battery. Also, depending on the emitter involved, the voltage might sag so much that you won’t be able to pull that much current anyway, at least not for a significant amount of time. So I would put the odds of really bad things (eg. fire) happening as quite low. Although I’m sure someone will dig up a counterexample

RapidLux · 2019-11-11T16:00:28.000Z

Don’t flashlights have two thermal sensors? One for the Led / driver and one for the battery? How does Andruil manage this? Ie. for Astrolux EC01 / FW3A, could low drain batteries reach critical temperatures, while the Led / driver still being within safe range? Wouldn’t the temperature diffuse towards the driver sensor and then it will throttle down the output, long before the battery could be damaged? Ie. Samsung 30Q has been tested to reach over 100 degrees Celsius without any noticeable damages.

fixed_it · 2019-11-11T15:28:06.000Z

You have a good point. Unless you disable thermal regulation, the worst case for batteries is almost impossible to achieve on such a small light. They fall back to lower output very quickly. At least that’s my experience with the D4.

iamlucky13 · 2019-11-12T01:20:10.000Z

RapidLux:

Don’t flashlights have two thermal sensors? One for the Led / driver and one for the battery? How does Andruil manage this? Ie. for Astrolux EC01 / FW3A, could low drain batteries reach critical temperatures, while the Led / driver still being within safe range? Wouldn’t the temperature diffuse towards the driver sensor and then it will throttle down the output, long before the battery could be damaged? Ie. Samsung 30Q has been tested to reach over 100 degrees Celsius without any noticeable damages.

Many flashlights have no thermal sensors, although hopefully that is not common on very high power lights.

BLF style lights often use an uncalibrated sensor on the driver for temperature. This includes Anduril-based lights. The two main goals here are to keep the LED at an acceptable temperature (generally needs to be less than 150 degrees Celsius, but less for some LEDs), and to keep the user’s hand comfortable.

If the sensor does not get calibrated by the user, it may read low or high. Reading low increases the concern.

Normally, the majority of the heat is generated by the LED. The LED has a very good thermal path to the body, and the body usually has an ok thermal path to the driver through both air and the PCB.

If the heat is being generated mostly by the battery, the battery has a poor thermal path to the body through the foil and plastic layers inside the battery, through the steel case, and through the spring and air.

Based on that, if the battery self-heats enough, I could see it reaching concerning temperatures before the temperature sensor detects it.

Also, while a 30Q can withstand brief exposures to 100 degrees C, that’s no the kind of battery we’re concerned about here.

wle · 2019-11-12T02:08:34.000Z

well-

a. they are not called ‘low drain’- they are advertised as ‘high capacity’

a2. voltage WILL sag, that is the definition of ‘trying to take more current out than the battery is designed for’

b. the max output of the light may not be reached

c. it might take some life out of the battery to run it at high current levels, even though it will limit itself
(and you will never really know how much or if any, but you can read the specs for an idea)

d. it might also do this:
battery or the light might cut off due to low voltage under load (or just refuse to try them) but
still come back and work fine at lower loads

e. some lights may decide that, once the under-load voltage has occurred, the battery is dead and not try any output until you cycle the power

it should not hurt the light, but it may damage the battery some, and be inconvenient in a number of ways

wle

Sirstinky · 2019-11-12T05:01:08.000Z

There’s no way to autonomously monitor the battery temperature in a flashlight that I know of. The driver measures the temperature of the mcu. From experience, li-ion batteries of the ICR type (like in laptops) have so much internal resistance that they will basically act like big resistors and limit the amount of current the driver pulls from the battery. The voltage will also sag sooner triggering lvp and reducing output or shutting down before the cell can overheat. Cells from major manufacturers have thermal fuses built in, like PTC’s except they don’t reset. Once the cell reaches 60-80 C, the fuse will probably trip and the cell will shut down permanently. High drain batteries are more durable and can handle 60 C or even 75 C for short periods. Once you get up to those temps though you’re putting a 15-20A load on it, which you’ll only get close to in the highest-power single cell lights.

RapidLux · 2019-11-12T17:07:45.000Z

I’m using LG 21700 M50 5000 mAh in Astrolux EC01 with XHP 50B 3V Led. This battery seems to get fairly hot at 15A, HKJ Review

I don’t know how many amps EC01 draws in turbo, around 15A? I don’t think it has any point switching to a high drain 21700 cell for some less battery heat and a bit more lumens output.

flightless22 · 2019-11-12T14:31:47.000Z

Would name brand (22f Samsung) cells pulled from a 2008 battery pack vs new ebay ‘lightmyassonfire’ mystery brand require the same precautions?

My assumption

Scenario 1: <2A, Probably OK.
Scenario 2: Absolutely not.
Scenario 3: Absolutely not.

ggf31416 · 2019-11-12T15:04:08.000Z

Why would anyone in USA, where 18650 batteries are cheap and easy to get, use a laptop pull Samsung 22F with 2200mAh and 4.4A CDR in a high-drain flashlight? Extreme cheapstake?

Sirstinky · 2019-11-12T20:02:35.000Z

ggf31416:

Why would anyone in USA, where 18650 batteries are cheap and easy to get, use a laptop pull Samsung 22F with 2200mAh and 4.4A CDR in a high-drain flashlight? Extreme cheapstake?

Before I knew anything about this hobby, the first lights I built were based on laptop pulls. 6 batteries for $3 compared to $6 for $25 so-so ones. The first had 6 Samsung 22F running a 100W led from a boost converter. I built a second battery for it from a old, crusty 9 cell laptop battery I got for $2 from the thrift store. They were barely useable, so I built another 10 cell out of a 12 cell and it’s been working good! Even at a below cost discount, 10 Sanyo GA still cost me $40. That situation it’s okay to run laptop batteries since it’s series/parallel, but for single-cell just get a high quality high drain battery for $7-8.

cetary · 2019-11-12T22:53:58.000Z

I have several use cases for low drain batteries in FET driven lights. One of them is a Narrow Band Amber multi-emitter light that color shifts badly if pushed past around 8-9 amps. The other use case is another FET driven light with quad 2200K E21A. The foreword voltage on those emitters is so low, if I push it past 8 amps I’ll blow the emitters.

Scallywag · 2019-11-13T00:07:10.000Z

The absolute worse case on a truly crap battery is thermal runaway. So, a fire or explosion of the battery.

All respectable batteries have safeguards which should prevent that easily. The nature of the circuit as well should prevent it, as voltage sag brings the battery voltage to meet the LEDs forward voltage and limits current draw.

vees · 2020-07-15T14:09:34.000Z

wle:

a2. voltage WILL sag, that is the definition of ‘trying to take more current out than the battery is designed for’

This bullet point really helped me visualize what was happening, thanks!

Zappaman · 2020-07-16T03:53:04.000Z

As some have alluded… how you use the light can determine what battery to use. My IF25a is a perfect example. I have run it on three 21700 batteries: Sofirn 4K, Molicel p42a, and LG M50.

The Sofirn (unknown current rating but probably around 12-15A) and Molicel (very high drain 35A) ran fine of course, but they produced a LOT of heat in a very short period of time. Where as the M50 (7.3A) ALSO runs great, but heats up the light less quickly without giving up much from what I can see.

As the IF25a requires a glove to use on turbo for more than a minute or two, it’s really better suited for running with a limited ramp setting (at about 75% output ability). So ya, I might be giving up a touch of output on turbo with the M50, but a good high capacity battery like the M50 still has PLENTY of shorter-term current output to my eye- for THIS light. AND… as the IF25a runs cooler and longer with the M50… it’s a better battery for my use anyway.

With other lights that can handle the current (heat) better (and longer), I do love the Molicels though. They light up my modded C8F and C8G nicely and to a level I definitely CAN see when using the Molicel

Enderman · 2020-07-16T03:54:51.000Z

If the flashlight is direct drive the battery voltage will sag a lot, and lower Vf = lower current, so just a dimmer LED.
If the flashlight uses a buck driver when the voltage sags the driver will pull more current to compensate and if enough current is drawn the cell can overheat.

Rayoui · 2020-07-16T09:33:00.000Z

I believe you are thinking of a boost driver, or a buck-boost driver. A buck converter won’t increase its output voltage higher than the cell’s voltage.