123 Primary Lithium cell info/testing/links

[SilverFox]

Hello Steve,

If you look at the starting voltage of each cell you will see why the ZTS tester rated one at 100% and the other at 10%.

Tom

 

[soffiler]

Hello Steve,

If you look at the starting voltage of each cell you will see why the ZTS tester rated one at 100% and the other at 10%.

Tom

Yup, actually I did notice that. I still figured it would be worth posing the question. The "weak" cell per ZTS is lower in starting voltage but when given the opportunity to perform, it does perform. Therefore it's not really weak, and the 10% spit out by the ZTS is an anomaly. Call it a tester anomaly or a cell anomaly (most accurately a combination).

 

[SilverFox]

Hello Steve,

It could have something to do with the phases of the moon... More gravitational pull reduces the starting voltage... but does not seem to have an equal effect. OK, anomaly sounds good.

The ZTS tester seems to have a problem with unused cells that have something strange going on with the chemistry. It does seem to be able to match cells on starting voltage. I have found that the 10% and 20% cells seem to have almost the same capacity as the 100% cells. It's the 40%, and 60% cells that seem to be strange.

I have noticed that if you take a 10% cell and put it in a light for a few minutes, it will test higher. However, I haven't done anything formal with this.

NewBie mentioned that the heat from the lamp may contribute to what is going on. Perhaps we could load the lower capacity cell nearest the lamp in an effort to heat it up. That way the weaker cell may be able to keep up with the stronger cell because of the increase in chemical activity (due to the heat) of the electrolyte.

I will, for now, stick with matched cells and remember to shut the light off when the lamp dims. I think I will also keep an eye on temperature during run time tests.

Tom

 

[soffiler]

...I will, for now, stick with matched cells and remember to shut the light off when the lamp dims. I think I will also keep an eye on temperature during run time tests...

Hey Tom:

Let's think about that for a moment. Starting with the fact that you're matching cells to begin with, what is the concern when the light dims? Are you worried that no cell matching is perfect, and at the point of dimming, there is likely to be one weaker and one stronger cell? I submit that by that point, the stronger cell hasn't got enough energy left to push the weaker one all the way to the point of venting.

Or am I missing something?

 

[cy]

a common element of all reported light failures that I'm aware of, all involved clickies, multiple primary cells and one or more defective cell.

twisties (LOTC) like Surefire M6 have reported venting, but those have been traced to internal battery carrier failure. so far all twistie lights failing have been traced to possible internal short issues.

IMHO clickies play a major role in that you cann't easily tell clickie is on or off position. except by viewing light is coming on. if internal failure is occuring to point light is not coming on. a clicky will give no visual indication switch is off.

you put light away, thinking light is turned off, but current is still flowing. then when all condition are aligned, venting occurs!

Hey Tom:

Let's think about that for a moment. Starting with the fact that you're matching cells to begin with, what is the concern when the light dims? Are you worried that no cell matching is perfect, and at the point of dimming, there is likely to be one weaker and one stronger cell? I submit that by that point, the stronger cell hasn't got enough energy left to push the weaker one all the way to the point of venting.

Or am I missing something?

 

[Chronos]

cy,

You and I have come to the same theory. In my situation I believe the tailcap may have been faulty or possessed a slow drain. The cells were likely mismatched (though new and fresh from the same batch; Last night I pulled out another 2x123 stick from the same shipment- this stick was new and never used- and it couldn't light up my Gladius) and then a venting event occurred.

Once I receive my ZTS tester I'm going to parse through the threads here and learn how best to use it. I wonder if this is the path forward for us, and if it is truly acceptable: test every cell prior to use. Could I test every cell if I were in a combat situation?

 

[SilverFox]

Hello Steve,

Good point. If the cells start off matched and are dead, there is probably not enough energy left to reverse charge the weaker cell.

I still think shutting the light off is a good habit to get into...

Tom

 

[SilverFox]

Hello Cy,

The SureFire clicky switches that I have (E1E and U2) have a different feel when they are on. The plunger is down further from the rubber cap when it is on.

Kroll switches are the same way, but the QIII switch has the same feel whether it is off or on.

Tom

 

[cy]

Tom, thanks for pointing that out. never noticed all surefire clickie switches feel slightly depressed when on.

I've always just clicked switch looking if light is on or off.

another factor that could be important is clickies are easier to accidently turn on. it just take one dead-on push to turn on light. VS twisties or LOTC on for a split second. it'd have to wedge just right to keep light on with twisty. could happen, but not likely.

a chain of events all got to come together to produce venting. IMHO all venting factors have always been present except for huge numbers of defective CR123 cells from China.

 

[SilverFox]

Hello Cy,

I am inclined to agree with you. These factors have been present since the lights have been manufactured. The only thing that seems to have changed is the cell construction.

It will be interesting to see how NewBie's testing with ZTS mismatched cells turns out. I think I sent him 5 of each level (except for 10%) to check out.

Tom

 

[tvodrd]

Jumping the gun a little, NewBie's latest, and fully-instrumented test seems to indicate that the weeker cell's reverse charging appears to be the culprit. It also implies that the "system" has to me under a pretty good load for the "vent with flame" to occur. If this proves out, may I suggest that various brands be compared wrt to internal protection against reverse charging? Are they all potential (when reversed :green: ) "bombs?"

Larry

 

[NewBie]

NewBie: for the sake of clarity... these are both brand-new cells? (as opposed to you depleting one purposely down to 10%)

If they're both brand-new, I'd suggest this was more a ZTS tester anomaly than a real mismatch test. Which also helps explain why the "10%" cell pulled ahead of the "100%" cell, and why there was no venting.

Yes both are new, just tested on the ZTS tester by Silverfox, and put into the test rig.



Okay, here is a run that was done identically to the last one:

.

I took one of SilverFox's 10% cells, and put it in a glass of water that was at 59C for 10 minutes, and I am doing a run with that right now as I type this, to see how it affects the results.

.

Here are the results from a ZTS tested 10% cell that was preheated to 60C, inside a ziplock bag, for 10 minutes, paired up with a ZTS tested 100% cell. Notice how the plots changed...:

Now compare the venting plot (pay attention, the time scales are different):

It looks alot more like the venting plot now. I'll be doing a longer preheat next, about 30 minutes, so that the heat can get into the center of the cell a bit better. This heating has a very vague simularity to the heating that occurs in a performance flashlight, especially the PM6. Besides the head section of a PM6 heating up very nicely inside, it also has a very heavy guage spring, which helps to conduct the heat to the first battery.

I have a few more tests I'd like to try, then I'll get a PM6 rigged up for the test, and we can see how things go there.

 

[soffiler]

NewBie: continued great stuff! And thanks for the clarification.

The cells tested 10% on ZTS have an odd initial voltage characteristic; it is slightly depressed and recovers to a higher value during use. In my limited experience (using West Mt. Radio battery tester primarily with high-quality USA or Japanese cells... SF, Duracell, Sanyo) they always look like your 100% cells, that is, with a rather high initial voltage that drops over the first few minutes under load. Right now, it's looking to me like the ZTS is fooled by that low initial voltage which is apparently some manufacturing anomaly (??). So far, your venting has occurred exclusively with one purposely depleted cell, a true and known mismatch; highly ZTS-mismatched but otherwise fresh new cells have not vented yet (correct?).

I would not be surprised if your testing ultimately reveals that a certain level of reverse-charge voltage and current (thus power) for a certain time (thus energy) is necessary to initiate venting.

 

[vic303]

Here's a link to a story in FT.com concerning the Sony laptop Li-ion battery recall. It says in laymans terms what the problems is. That tidbit COULD apply to the cr123's too.
http://www.ft.com/cms/s/c2eab782-33...age=6700d4e4-6714-11da-a650-0000779e2340.html

 

[NewBie]

Okay, here is what happened after 30 minutes of preheat, in a ziplock bag in 60C water.

The video that goes with that logged data:
http://www.molalla.net/~leeper/zts104q.wmv

You will notice it vented at the end of the fuzzball at ~94 minutes. The test started at ~11:30 PM last night, and I ended it at about ~2:00 AM. (yes, been burning the midnight oil on this stuff for quite awhile now).

This is one of those ventings that happened a while after the bulb winked out.

As such, it appears the data is showing that the heat is also a big contributor to the whole ball of wax, so far.

Just a few tests to go, and I dig into a PM6, instrument it up, and find out what is going on in it. Still hoping for my ice point chip parts. I may end up taking another approach on this K-type thermocouple signal conditioning circuit, just so I can start getting the cell temp data.

 

[SilverFox]

Hello NewBie,

Any idea on what is causing the "fuzzball" voltage swings?

Tom

 

[Lost_In_Beijing]

Maybe this bit of info will help. Gleaned from twom news articles on Sone battery recalls (Dell and Apple): "Sony said the problems arise 'on rare occasions' when microscopic metal particles hit other parts of the battery cell and lead to a short circuit." From an earlier report Sony said the metal particles were from "overcrimping" of the batteries during manufacture.

So, correct me if I'm looking at this in the wrong way because batteries are not my speciality, but the metal flakes from the overcrimping may be the common thread. If, as a battery heats up from the imbalance betwwen the full and partialy discharged cells, this would cause thermal expansion within the battery. The expansion would cause movement of the metal flakes, leading to internal shorting. This in turn would lead to more heating and thermal runaway causing the catastrophic failures. The shorting might also be the source of the voltage swings. Of course, the internal heating could also be playing havic with the chemical reaction, which would also cause the voltage swings

It's just a theory and it was free, so take it for what it's worth.

Weldon

 

[NewBie]

Hello NewBie,

Any idea on what is causing the "fuzzball" voltage swings?

Tom

If I were to guess, it might be the PTC, but it is just a guess without merit. If it was the PTC, something should have shown up on the other cell, and in the current waveforms. So that isn't that likely.


.


On a side note, I repeated the same test that just caused the venting in the last run.

It didn't vent this time. It looks like it came close though. Take a look at the other venting graphs, you will notice that when there is a vent, it goes over a 1 volt negative. It looks like it just narrowly missed it. It might be due to the fact that it is cold outside, which affects the internal cell temperatures.

I've got a few more datapoint tests to run, but it will have to wait until tomorrow. You guys all have a great night, it's getting late here again.

 

[cy]

some real patterns are coming out...

yep sleep sounds good zzzzz...

 

[soffiler]

Hello NewBie,

Any idea on what is causing the "fuzzball" voltage swings?

Tom

My guess, and it's only a guess, is that the fuzzball has its root cause in chemistry. You're driving current in the wrong direction thru the dead (negative) cell's chemicals... however I am certainly no chemist, and cannot even begin to explain the details.