CR123A battery (circuit?) question - any explanation for this behavior?

[selfbuilt]

I have found a few Titanium cells (I suppose they have been included with a light, but because I am mostly using LiIon, I have not used the cells). I am going to do some test with the battery alone, i.e. no flashlight, only a resistor. This way I can log current, voltage and temperature at the same time

 

[SilverFox]

Hello Selfbuilt,

My "best guess" is that there is a difference in the chemical formulation.

I know that the major manufacturers of CR123 cells consider a 1C load large. In the early development of some high power incandescent lights there were a few incidents with cells blowing up under close to 4C loads. The formulation may have been changed to guard against that.

Another indication that the formulation may be different is that the Titanium cells start to form a pacification layer after a couple of years of storage and the Panasonic cells don't. Whatever keeps this passivation layer from forming may result in higher cell temperature under high loads.

I wonder what the internal cell temperatures are...?

Tom

 

[HKJ]

My first two runs are done. The temperature did not rise enough to trip the PTC, I will do another run with the cells in a plastic tube and hope this will isolate enough to trip the PTC.
The Titanium cells has a higher voltage and lower temperature during load, the final capacity is the same, but the energy is higher. This can be due to different chemical formulation as SilverFox guesses, but it can also be due to higher resistance in the PTC (it could even be differences in my connections to the cells, but because we have seen differences in the flashlight I doubt that).

The test is done with a 0.716 ohm (including wires and meter) load and a temperature is measured with a thermocoupler mount with adhesive tape to the CR123 cell. The connections to the cell is done with a clamp with the voltmeter connected directly to the screws.

 

[Justin Case]

Another indication that the formulation may be different is that the Titanium cells start to form a pacification layer after a couple of years of storage and the Panasonic cells don't. Whatever keeps this pacification layer from forming may result in higher cell temperature under high loads.

Passivation layer?

 

[45/70]

Passivation layer?

Nit picker.

And it's "large crystalline formations", not "granules".:nana:

Dave

 

[Helmut.G]

This is extremely interesting info, especially for anybody using primary CR123s in high power lights on the high power levels.
Thank you HJK and Selfbuilt for your measurements!

 

[recDNA]

Ok, I won't be trying this again. :sweat:

I ran the light without cooling on the Titanium Innovation cells until the PTC apparently engaged (occurred a little after ~20 mins into the run, as before on these cells). Keep in mind this is a not exactly fair comparison to the Duracells at 3.5 mins, as the longer run gave everything a lot more time to heat up. Here are the temps: (first pass is immediately upon removing from the light as the drop in output began, second pass is ~25 secs later).

Duracell (3.5min) first-pass, from front to back: 68.4C/56.4C, 51.9C/46.3C.
Duracell (3.5min) second-pass, from front to back: 58.3C/50.3C, 49.9C/43.9C

Titanium Innovations (20min) first-pass, from front to back: 80.6C/73.6C, 72.4C/70.3C.
Titanium Innovations (20min) second-pass, from front to back: 77.6C/71.3, 69.4C/66.6C

I don't believe you can draw any real conclusions about the temperature at which the PTC engaged this way. It's possible both types of cells responded at the same point - but because the Duracell/Panasonic ramped so much more quickly, they also dropped more quickly (i.e. by the time I got everything out of the light and measured it). As you can tell from the later TI run, the temp of the cells remained elevated longer (as you would expect from the more gradual increase in heat of the light and batteries). As such, it is really impossible to know what internal temp triggered the PTC throttle-down.

As an aside, I will say the light was a LOT hotter to the touch when getting these cells out (compared to the 3.5min Duracell runs, where the light was still relatively quite cool to the touch). Also, the wrapping of the cells had opened up near the positive terminals - slightly on rear cell, considerably on the front cell. Needless to say, I will not be doing any more of these runtimes without cooling!

The only way to know for sure what is going on would be to directly measure the temperature of the front cell while the light is operation, and see directly when the PTC engage. Given the wide time disparity here, my simple method doesn't really tell us much. As the more detailed analysis is beyond my capability to measure, hopefully HKJ (or someone else here) will be able to do it.

So, to summarize all of the above:

• the Duracell (and presumably Panasonic) cells heat up FAR more quickly than the Titanium Innovations cells in this setup.
• it is possible the Titanium Innovation cells are also tripping at higher heat levels, but the only real way to know is to directly measure temp under operation. This is beyond my ability at this point in time.

In the 20 min test the TI batteries got hotter than 3.5 min Panasonic test. Why not run BOTH for 20 min? My major concern is safety. If the Duracell are hotter after 20 min I want TI. If early ptc causes Duracell to be cooler after 20 min than TI I want Duracell.

Actually I have several boxes of Surefire and no Duracell's but I am drawing an inference that Surefire is more like Duracell than TI?

This thread is fascinating. I'm surprised it ever died. Maybe someone else could continue it with more tests? I would.also be very curious about results in single cell tests.

I noticed a Surefire heats up quickly in a sx32w in h1 2.2 amps. If I had a way to test I would try TI. Despite owning several boxes of SF I would buy TI if they run cooler in sc32w.

I have never heard of a sc32w blowing up on primary but maybe too few users with primaries.

 

[Lumencrazy]

The difficulty here is that we are trying to answer technical issues related to battery performance indirectly. Flashlights are not analytical test instruments used for battery evaluations. Especially when comparisons are being made with different lights under varying conditions. One really needs to use the correct equipment like a calibrated DC Load in order to be able to isolate the reasons for the varying differences in battery performance under controlled conditions.

 

[selfbuilt]

I am drawing an inference that Surefire is more like Duracell than TI?

Yes. You can see my 2013 CR123A round-up for some commentary on the source of all made-in-the-USA cells - and some direct testing results.

The difficulty here is that we are trying to answer technical issues related to battery performance indirectly. Flashlights are not analytical test instruments used for battery evaluations. Especially when comparisons are being made with different lights under varying conditions. One really needs to use the correct equipment like a calibrated DC Load in order to be able to isolate the reasons for the varying differences in battery performance under controlled conditions.

+1 :thumbsup:

I certainly didn't set out to do battery testing in this thread - it was just to explore what the likely source of the issue was. It really needs someone with the right equipment and expertise to investigate further (neither of which I have).