ROAR of the Pelican (CR123 Explosion during use, firsthand account)

[shao.fu.tzer]

Another factor complicating the situation . . . Counterfeit CR123 cells do exist. Cheap versions branded to look like the good stuff. Always buy from a trusted dealer, store, or site.

I would consider L.L.Bean a trusted dealer... I'm pretty sure they buy direct from Surefire... I bought of a case during their last sale - all dated 2020 and the voltages vary on the cells up to .2V - whether or not that's dangerous to use in a multi-cell application, I'm not sure... Has anyone done any extensive testing on this? Someone with a few spare UF-501s should run some tests using cells reading slightly different voltages, starting from a .1V difference up and see at what differences in voltage will cause cell failure/venting/explosion.

 

[SilverFox]

Hello Shao.fu.tzer,

Open circuit voltage gives very little information on how evenly CR123 cells will discharge. These cells develop a passivation layer on the electrodes internally, and simply measuring voltage doesn't give you a good reading. The passivation layer acts like you put a resistor in series with the voltmeter and the voltage reads lower.

In an effort to figure out a way to look at this, the ZTS tester measures the voltage under a load, and compares the results to an internal look up table based upon testing a variety of brands of cells. This is a step in the right direction, but it also is challenged by the passivation layer. The benefit of this test is that very little of the cells capacity is used up during the test. The downside is that in order to get a better measurement, several tests need to be done back to back on each cell. When I use the ZTS I shoot for 5 readings that are the same. This may mean taking a total of 10 - 15 readings before things settle down to where you will get 5 similar readings in a row.

In order to test to see if a 0.2 volt difference in open circuit voltage makes a difference, you would need to run a series of tests. These tests would involve discharging the cells, so you wouldn't have information that is specific to all cells, but it would give a general idea. These tests would have to be repeated for various current draws and for each batch of each brand of cells.

Some of this testing has been done. In general a small voltage difference (less than 0.5 volts) does not seem to make much of a difference until the current is increased to the point where the cell is completely discharged in about 15 minutes. If your light uses up a set of cells in 15 minutes, your concern is closer to being valid.

I deal with this by buying quantities of cells. I get a box of 50 cells, then randomly pull 5 for testing. If the cells test close to each other I am good to go. If they don't, I get back to the supplier and discuss my concerns. As storage time goes by, I will check a cell or two to see how well it is aging.

This is where some interesting things have been observed. For example under a ZTS test you may end up with one cell at 100% and another at 80%. During a discharge test the 80% cell may outperform the 100% cell. The theory is that this is caused by a difference in how the chemistry degrades inside the cell.

A simple way to run some tests on this 0.2 volt difference is to put some of those cells in your light and turn it on. Run it continuous for 10 - 15 minutes, then pull the cells and measure the temperature of each cell. In order for a cell to vent with flame you need heat. If your "normal" cell is only warm and your cell that was 0.2 volt lower is so hot that you can't hold it in your hand, you have a problem. If both cells are about the same warmth, you don't have a problem.

Tom

 

[Steve-at-Springboard]

Just a small update from a dealer that has sold over 52,000 123s in the last 4 years (no known failures, 1 leaker).

There is only one manufacturer of primary 123 batteries in the U.S. That is the Panasonic factory in Colombus, GA. They private label every 123 cell marked "made in U.S.A" including (but not limited to): Panasonic, SureFire, Streamlight, Duracell, Universal, Energizer, Polaroid, Battery Station, Ray-O-Vac, and Radio Shack cells that are marked "made in U.S.A". Note that Battery Station switched to the Panasonic cell at some point after this thread started. There are certainly other brands I'm not aware of.

I tell my customers to only buy U.S. made cells in factory packaging from U.S. sources. I have seen Chinese counterfeit SureFires, Duracells and Panasonics. The Panasonic copies only registered 40% on my ZTS tester.

The U.S. made Panasonics and their private-label cousins are all rated at 1550 mAh. Every foreign brand I'm aware of has less rated capacity (1400 mAh, 1300, etc.) and, some of them are just crap.

Most of the private-label Panasonics are marked with an expiration date, which is ten years from the manufacture date. Don't buy old batteries! The ten-year date is when the battery has self-discharged to 50% capacity, a little less than 5% per year. They don't maintain full capacity for ten years and then die completely on the tenth anniversary date.

Also, all the U.S. made cells have the protection circuit. Some of the foreign and counterfeit cells may not be protected.

Hope this helps!

 

[MorePower]

Most of the private-label Panasonics are marked with an expiration date, which is ten years from the manufacture date. Don't buy old batteries! The ten-year date is when the battery has self-discharged to 50% capacity, a little less than 5% per year. They don't maintain full capacity for ten years and then die completely on the tenth anniversary date.

CR123A cells shouldn't lose 5% of capacty per year at room temperature. At 10 years, they should retain over 90% of initial capacity.

Also, all the U.S. made cells have the protection circuit. Some of the foreign and counterfeit cells may not be protected.

It's not really a circuit. US cells have a PTC, which increases in resistance as temperature increases. This protects the cell in case of a short circuit.