Graphical Thermal Analysis of LOD-CE on Li-ion Cell

[liveforphysics]

Many folks know how bright the LOD-CE gets when used with a Li-ion cell rather than a NiMh cell.

I've seen quite a bit of people talking about how it gets burning hot in seconds...

I saw the Fluke Ti20 sitting in my office, so I decided to show people exactly how warm it becomes over a 5 minute time interval.

Office ambient temperature is fixed at 75deg. Airflow is minimal. Cell used in the test is a 0.5Ah Ultrafire Li-ion purchased from www.DealExtreme.com .

This is the test setup.

Now for the test data. The first picture is after roughly 5 seconds of being on. This delay is present due to the time needed to switch the light on "High", then capture the thermal image.

All later pictures are taken on 30second intervals. The test runs for 5 minutes and 30 seconds. At the end of which time I picked the flashlight up (with my bare hands, it was hot, but not too hot to handle), unscrewed the top and slide the battery out. I measured peak temps on the Li-Ion cell to be 124.6deg.

The thermal images are as follows.

Begin time.

0:05 (begin time)

0:30

1:00

1:30

2:00

2:30

3:00

3:30

4:00

4:30

5:00

5:30



The LOD-CE on Li-ion is the brightest single LED light I own. It is also the smallest LED light I own. Its roughly the size of my pinky, yet makes more light than full body size overdriven K2 lights and all sorts of other LED lights I've wasted money on in the past. I find it perfectly useable for all my EDC needs when useing the Li-ion cell.

I am very pleased with the LOD-CE.

Best Wishes,
-Luke

 

[BentHeadTX]

Thanks Luke
The L0D CE is a flame-thrower on Li-ion cells and looks it with your thermal analysis. Although Eneloop NiMH AAA cells are more "my thing" on a keychain light, your results plant the seed of li-ion.
Did you notice any dimming of the light as it heated up? How hot was the cell after running for 5 minutes?

 

[liveforphysics]

At the end of the 5minute 30second test, the cell temperature was 124.6F (as listed above). This is a safe temperature for this battery.

I did not notice any visible dimming as the light heated up, however it likely is occuring, but on a level too fine for my eyes to observe it.

Best Wishes!

 

[Strauss]

I wonder how the heat is compared with the luxIII LOD version? I know mine(luxIII) gets really hot REAL fast with my 10440 cell! Even though the LOD-CE is brighter, I wouldn't be suprised if it didn't get much hotter since the CREE is so much more efficient.

 

[liveforphysics]

I would also guess the LuxIII version would get reach a higher temperature and faster, but I dont own one to check.

From the watching the thermal transfer occur, it would appear that a bit of thermal grease to lubricate the threads rather than the oil they come included with would make a pretty large improvement in the surface available for the light to dissapate its thermal energy.

This would likely cause battery changes to result in white smears on the users hands however, unless you were perticularly mindful to avoid contacting the threaded area.

If anyone wants actual battery current draw and voltage specs when useing a fully charged 10440 cell I can take some tonight at work. Its a slow night...

 

[HarveyRich]

Liveforphysics: Very interesting thermal pictures. However, can you conclude that the 124.6 degrees F is a safe temperature for the led even if it's OK for the battery? What if you had extended the test for 10, 15, or even 20 minutes? What would the temperatures be at those data points? Also, what is the short and long term effect on the led? I would be very hesitant to use the Li-Ion battery in my LOD-CE before getting substantially more answers to these questions. I even wonder if the long term effects on the led are adverse with only 5 minutes of periodic use of a Li-Ion battery in the LOD-CE?

 

[suvdave]

Great little thermography test. Few have access to a $6000 imager. The observed temps appear to represent the surface temperature of the flashlight. The internal temp could focally be much higher, with the heat dissipating over a larger surface area. What about a measurement looking end-on at the reflector with the glass lens removed? Perhaps also do a comparison with different battery types?

 

[txmatt]

Liveforphysics: Very interesting thermal pictures. However, can you conclude that the 124.6 degrees F is a safe temperature for the led even if it's OK for the battery? What if you had extended the test for 10, 15, or even 20 minutes? What would the temperatures be at those data points? Also, what is the short and long term effect on the led? I would be very hesitant to use the Li-Ion battery in my LOD-CE before getting substantially more answers to these questions. I even wonder if the long term effects on the led are adverse with only 5 minutes of periodic use of a Li-Ion battery in the LOD-CE?

Runtime with a 10440 is only 10 minutes to 50%. You'd have to swap multiple batteries to do a longer test.

The images after about the 2 minute mark look nearly identical, indicating the light had reached steady-state in terms of temp increase.

I don't think we know the long-term effects on the LED. It's likely shortening its life but by how much? Considering 2 or 3 years from now another generation (or two) of LED's will have made the L0D CE obsolete, most of us will never put even 100 hours on one before replacing it. We may never know.

The bigger issue with the 10440 is we're running the lithium cell outside of it's operating specs (max current) which can have more serious consequences than zapping the LED, though that more often occurs during charging.

I run a 10440 in mine for short periods. It will be interesting too see if we have any LED or 10440 failures in the L0D CE as time goes by.

 

[liveforphysics]

Hi guys, I've had a few more weeks go by of EDC'ing my LOD-CE on 10440's now, and I've seen no problems.

I've recently had a few times at work where I've run through an entire 10440 cell on "burst" until the cell drains, swapped batteries to a fresh 10440 and run through that entire cell on "burst" imeadiately after with no cooling time in between batteries.

If you are holding the light in your hand, it never get even clost to as warm as it did in these tests where it was just sitting out on a counter with minimal airflow.

So, after analysis, along with over a month of everyday 10440 useage, the LOD-CE has displayed no troubles or related issues, and continues to provide me with excellent light in a package the size of my finger.

Now, I dont know what it takes for this forum to think something is safe for a flashlight, but I think this myth that it's only safe on 10440's on burst for short little bits of time is busted!

 

[Streamer]

Great Work liveforphysics! :goodjob:

 

[mayo]

The images after about the 2 minute mark look nearly identical, indicating the light had reached steady-state in terms of temp increase.

Although the pictures looked the same the high temp was 130.8F at 2 minutes and 154.1F at 5 minutes.

Would love to see a longer test.

None the less, I still run mine on the edge too. If it pops, I'll just buy the next great thing anyway. :naughty:

 

[EV_007]

Awesome images. I too run/push the LOD CE with the 10440 with no ill effect, yet. I mainly use the lowest setting, but occassionlly pop it up to max for fun.

You still get all three levels, just a bit brighter of course. The low on the LOD CE running the 10440 is as bright as the high on the L0P SE funning NiMHs.

 

[LEDite]

Luke;

Great job on the IR photos. Thanks for taking the time to do the photos.

Have you done the photos on any other popular lights?

I still have to rely on my IR thermometer for my multi-Cree P4 assemblies.

It has shown me the need for better heatsinks!

Larry Cobb

 

[liveforphysics]

Hi guys, I was bored at work again babysitting some tech's who were re-assembleing a generator coupleing. Fortunately, I had 6 fully charged 10440's in my pocket, along with my favorite EDC buddy, the LOD-CE.

Long story short, I burned through SIX! (yes, 6) fully charged 10440's in a row, imeadiately swapping batteries as the light began to dim. No cool down time in between, just non-stop 'burst' mode to keep light on the equipment these guys were working on.

They were all equiped with maglights, and needless to say, we had to do some quick beam comparisons... They were simply blown away by the output, and imeadiately asked about where to buy them.

I still see lots of people acting worried about useing 10440's with this light, and I really dont understand the concern. Yes the light gets hot, as things become warmer than the things around them, the ability to transfer that thermal energy increases. When held in your hand, this light has proven itself to find a safe thermal equalibrium when on burst with 10440's.

Best Wishes,
-Luke

 

[2xTrinity]

I still see lots of people acting worried about useing 10440's with this light, and I really dont understand the concern. Yes the light gets hot, as things become warmer than the things around them, the ability to transfer that thermal energy increases. When held in your hand, this light has proven itself to find a safe thermal equalibrium when on burst with 10440's.

My concern has not been with the light getting too hot. As you mentioned, almost any time I'm using it, it's in my hand, getting waved around so there's both conduction through my hand, and convection/airflow, as opposed to mostly radiation to dissipate the heat (ie, light resting horizontally on a table). I was more concerned about the cells being ruined after relatively few cycles due to drawing more than the recommended amount of current. However, the thought of light output comparable to my 18650 Cree flashlgihts in my L0D is just too tempting. I have a few 10440s on the way, and I plan to just keep an eye on the cells when they're charging (make sure they aren't oveheating etc periodically) and if I ruin any cells, so be it -- they're a few bucks from Dealextreme, not much more than single-run Lithium primaries.

My L0D is already my most frequently used light. I plan to carry a 10440 in the light, a spare 10440, and standard NiMH AAA with me, to have my choice of output. If I even need to run the light in "candle" mode, or have minimum brightness, I can swap in the NiMH cell.

Earlier today, I needed to wire some speaker cables underneath a stage, a few people asked if I needed a flashlight, I said I was fine as I had my L0DCE with me... when I came out, they were shocked to see that so much light was coming out of such as small light the whole time, running a standard AAA cell. 10440 output should be out of this world.

 

[Dobbler]

Got my 10440 cells tonight.

:wow:

Gets warm fast, but damn that's a bright little sucker. :rock:

 

[jlomein]

That thermal imager is really interesting. Can you see through walls with it?

 

[liveforphysics]

Nope, I can't see through walls, but I can see through many types of thin clothing. However, my 1,300,000 ft^2 building only has 2 females working in it, and both of them are the sort of woman that you would rather not see better detail of the bodys under the clothing...

I could do images of any flashlights or LED's, but unless questions about other flashlights thermal stability arise, I won't bother.

 

[RonM]

The reports of people using 10440s in the L0D-CE is a testament to both the light and the Cree LED. However, I remain a skeptic. If the temperature of the flashlight head is showing 150 degrees, the temp within the LED itself must be substantially higher. The heat is generated at a small point and then spreads and cools, so what we're seeing is only the tip of the iceberg.

My guess is that for many this setup will work, but the statistical failure rate will probably be much higher. I'd also venture to guess that the heat will degrade the LEDs output fairly rapidly. At least that's always been the case with all previous high power LEDs.

 

[txmatt]

You only have to worry about those temps on high, though. One advantage of the regulation of the L0D CE is that all three levels still work with 10440 such that low and medium with 10440 are very similar to medium and high respectively with NiMH or Lithium primary. I keep a 10440 in mine all the time but mostly just use low or medium because that's all the light I usually need (and to preserve run time). At these levels the 10440 isn't pushing the Cree any harder than it would/could with normal batteries. And it is cool to have that very bright high. At least for me, the time spent on high is shorter by probably a factor of 100 than the time spent on low or medium so I'm not too concerned with LED degradation from a few seconds or minutes on high here and there.