Surefire M6 / New Titanium 123's / Results / Not good *with pics and graphs*

[js]

OK. I'm at home on the dial up and I have to get back to SF M6-R orders work, so I'm not going to go back over all the posts in this thread. Not today anyway.

But, for what it's worth, I just did a runtime test of the New Titanium 123's in my M6 with MN21 HOLA installed.

I got 15 min 45 sec and then one or more of the batteries thermally shut down. I waited for a bit to let them cool, and then continued the test, getting another 8 minutes until obvious dimming. I could have gotten more, but it would have been increasingly dim and I believe I picked the proper cutoff point at obvious and increasing dimming.

The batteries were all equally hot--about 150 F or a bit less, going by the standard pain tests. (If you can't hold onto it for any length of time at all, it is >160F. If you can just hold it for a second it is around 160F. And if you can hold it for a few to several seconds it is 150-140 F. Etc.

Not very accurate, maybe, but you get the idea. I could hold these for a few seconds before chickening out.

I have done this sort of test with SF 123's before, and got similar results, except that thermal shutoff occured at 10 minutes, IIRC. And the cells were just as hot as these bad boys. Maybe even hotter.

OK. Hope that helped.

 

[InfidelCastro]

Unlikely a DMM will tell you anything useful. A battery with only 20% life left can have a higher voltage than a battery with 100% life left.

Yea I know, I'm just saying, at least I'll know it won't be totally dead. If it's totally dead it will be sub 3V I suspect.

 

[CostcoAAcells]

Thanks to this thread, I decided to order a ZTS mini tester also. I'm a battery hoarder, myself. Up to now, I've bought my batteries based soley on their reputation as they are represented here on CPF. That's still kind of blind faith, in my book. At least once I get this ZTS tester, I'll be able to check for duds on delivery day. As it is, if I currently have any duds I probably won't know for another 6-12 months--when I go to use them. And by then, if I want to return them it may be harder to convince the vendor to take them back. I doubt I'd still have the paperwork for the purchase after all that time either.

 

[AmondoTech]

Thank you to the CPFers for conducting such detail analyses.

I will report this to the technicians and look into this issue.

Regards,
Wayne

 

[bwaites]

Wayne,

I just wanted to point out that the M6 uses two (2) stacks of 3 cells, not a single stack of 6.

So theoretically, your testing to a stack of 4 should be fine.

But, the truth is that something about Lithium technology is still not quite perfect and just about ALL the brands have seen failures.

I myself had had M6 shutdowns on Surefires, BatteryStation, and Duracells, as well as a few other less well known brands.

Bill

 

[lebox97]

got my ZTS mini-MBT battery tester today (batteries not included! :ironic - very small (fit's in a shirt pocket)

wow, I was surprised... some of my used 123 batteries I thought were still "ok" showed 20% left.
on 25 new ones of 3 different brands (I won't pick on a vendor) I got mostly 100%, a couple of 80% and a couple of 60%'s

what was odd is I got a few that came up as 60-80% - and when re-tested it showed 100% - not sure what's going on with that! :thinking:

also, on a DMM - the 20% bats showed 3.02-3.03 v, and 100% ranged between 2.98-3.22v :thinking:

 

[SilverFox]

Hello Lebox97,

Can you give us your test procedure?

I have heard that the battery recovery effect can trick the ZTS tester a bit. I believe they recommend doing two tests 5 seconds apart and going with the second test results.

Tom

 

[InfidelCastro]

Hello Lebox97,

Can you give us your test procedure?

I have heard that the battery recovery effect can trick the ZTS tester a bit. I believe they recommend doing two tests 5 seconds apart and going with the second test results.

Tom

Thanks for the tip, I'll try and remember that. I definately want to get one of these testers.

Does this effect happen with alkalines and NiMh as well?

 

[lebox97]

Hi Tom
ummm, procedure?
I sorted about 200 batteries by brand and type (AAA, AA, CR123) and started testing them on the appropriate terminal with the probe? (and all came from same storage temp)

I will re-test and create a spreadsheet to show side by side the voltage and capacity % - as I found it odd that the capacity left %'s and voltage readings were not consistent. (ie. some full capacity batteries showed a lower voltage and vice versa?)

EDIT: re-sorted all my batteries by capacity % - was surprised to find some dead/low capacity AA/AAA bats mixed in with good ones in my devices. I think this was my screw up - not manufacturers though.

the instruction sheet does mention to repeat test after several seconds for best accuracy.
Tod

 

[Donovan]

The first batch that I purchased will be my last. They have less than 1/2 the runtime of battery station CR123's. I kept thinking something was wrong with my lights!!!

Granted I think I got the older style, but I wasn't able to return them for the so called "improved" ones so I was stuck with crappy batteries. I won't make that mistake again!

 

[atm]

I have found the same thing with the ZTS tester (tester's results not correlating with standing voltages) as noted here;

https://www.candlepowerforums.com/threads/112786

I did find though that results from flash amping the batteries ordered them the same as the tester (is flash-amping advisable for new batts, particularly ones you want to store for some time?).

I make a point when using the ZTS tester of doing more than one test per battery because of the variation in results with some batteries.

In the above post I ask whether people think the testers are accurate or not, these varying results do make me wonder.

Andrew

 

[Handlobraesing]

Re: Surefire M6 / New Titanium 123's / Results / Not good

The batterystation cell pulled THAT FAR ahead of the pack? That's very interesting. They're slightly cheaper than SF cells last I looked, and IIRC they're frequently considered to be comparable to SF in build quality. Did you just use one of each cell or are these repeated tests? How did you perform the test?

Also, the M6 uses two parallel stacks, so the draw will be 2.5A from each cell on the HOLA, not 5.0A (which would be way too much to expect from a CR123A except in very short, very sparse bursts.)

Remember that battery manufacturers usually don't recommend using lithium cells in any type of parallel config, because current sharing can't be guaranteed to be even. When one set collapses, entire loaded is shifted onto the other bank.

Someone should setup two banks of three CR123 and with a 0.01 ohm resistor in series with each bank and monitor the mV across each resistor with a two channel logger with the entire assembly connected to 5.0A CC load. Ideally both resistors would be giving a 25mV drop, but I bet in reality, one's going to be higher than the other one.

 

[OutdoorIdiot]

[Edit: Just to make it clear that this was in response to ATM's post!]

I think one of the more important things that this thread has addressed in earlier posts, is that the voltage of a battery cannot be used as a reliable indicator of it's state.

Indeed, the ZTS tester seems to have been created for this very reason. From my limited understanding of how the ZTS works, I think it is performing a very sophisticated kind of flash-amp test. It seems it measures not only the voltage, but also allows a lot of current to flow, so that it can measure the battery's ability to deliver current at that voltage.

And it does this in "pulses," presumably so that (1) It doesn't drain the battery too much (or damage it), and (2) It can measure the difference in readings over time (i.e. how the measurements change at each pulse), to get more insight into the battery's behaviour. And then I think it compares these results to some kind of internal look-up table that contains measurements of current and voltage from batterys that are "known" to be "good".

If all that is true, then I'll put my neck on the line and suggest this about the ZTS tester:

(1) It is fallible. Not all 123 battery's will have the same behaviour as the ones that were used to compile the look-up table, even though they may be "good" batteries, by some measure of the term "good". However:

(2) It's as good as it gets. I really can't think of what more you can do to test if a battery is AOK, or to check that all batteries in a group are in the same "state" before loading them into a flashlight.

 

[s.duff]

thanks for the info. i think ill be sticking with surefire batteries.

 

[atm]

Good points, thanks OutdoorIdiot. I think the tester is the best chance I have of getting some idea of a batterys health, and those which should and shouldn't be used together.

Andrew

 

[lebox97]

anybody want to try to decipher this? :huh2:

CR123 measured voltage, ZTS-mini tested, tested again after 5 seconds, re-measured voltage after 5 sec.

Known New Batteries (volts, % capacity, % capacity, volts)

3.27 100 100 3.14

3.28 100 100 3.12

3.28 100 100 3.14

3.28 100 100 3.11

3.27 40 60 3.11

3.28 100 100 3.11

3.26 100 100 3.12

3.03 0 20 2.98

3.27 100 100 3.11

3.25 100 100 3.06

3.25 100 100 3.11

3.25 100 100 3.09

3.24 100 100 3.09

3.26 100 100 3.08

3.26 100 100 3.05


New batteries (all initially tested at 100%) – after four 1 second blasts in SF M6 MN21 bulb – then rested 18 hours.

3.02 80 80 3.00

2.98 80 60 2.96

3.07 80 60 2.96

3.02 100 80 3.00

2.99 80 100 2.97

2.98 60 80 2.95


Known Used Batteries

3.00 100 100 2.97

3.03 100 80 2.99

3.23 100 100 3.09

3.23 100 100 3.09

3.23 100 100 3.09

3.19 100 100 3.04

3.20 80 100 3.04

3.20 100 60 3.06

3.14 100 80 3.00

3.19 100 100 3.05

3.03 80 80 2.96

3.01 80 60 2.96

3.03 80 80 2.98

3.05 20 40 3.04

3.05 40 40 3.03

3.02 20 20 3.00

3.02 20 20 3.00

I know I am completely baffled!!!

 

[OutdoorIdiot]

I'll have a go at deciphering those results, though I'm hardly qualified to do so. In fact I'm only doing this because I'm surprised anyone else hasn't had a go yet, and I'm getting impatient (I have to leave in a bit)!

Anyway, here goes some wild conjecture/observations from a novice:

(1) The one consistent thing (phew!) seems to be that the second voltage reading is always lower. Fair enough, because the voltage measurements and the ZTS testing will "use" the batteries to some extent. While voltage seems inadequate as a means of testing for battery life, I think it is still fair to say, that for any given battery, the voltage will drop with use (given same temperature and all else being equal, etc etc...).

(2) For the cases where the second ZTS measurement gives a higher "remaining life" reading that the first, I wonder if experimental error could be to blame? E.g, improper contact between the battery and the tester. Perhaps a few goes at measuring life, and averaging, would yield different results?

(3) The batteries that have provided a few blasts in an M6 MN21 are the most interesting part of the results, for me. The significant drops in measured life expectancy (in some of the cells) can possibly be explained by the fact that, even in a few short blasts of the M6, a huge amount of current will have been supplied, because, I believe, as the filament lights up a huge amount of current is drawn, until the filament gets hot. The fact that not all cells show the same huge drop could possibly be explained by (1) The statement that Handlobraesing made a few posts ago, "Remember that battery manufacturers usually don't recommend using lithium cells in any type of parallel config, because current sharing can't be guaranteed to be even. When one set collapses, entire loaded is shifted onto the other bank," and (2) Perhaps also experimental error again.

Just some thoughts. I confidently expect to be roundly slapped-down, but thought I'd have a go anyway...

 

[JimH]

anybody want to try to decipher this?

The only thing that stands out for me is that, if the initial voltage is below about 3.10, the odds are the battery will show a value less than 100 % on the ZTS tester.

 

[Size15's]

Has anyone taken an MB20 apart? I understood it had diodes to prevent reverse charging...

 

[Ringer]

Well most of the Titaniums I have are in the shrink wrapped 2 packs. Anyway to test these without breaking up the packs? I haven't experienced any issues with mine so far in various 2 cell Surefire and Streamlight lights.