9 CR123As in series at 3.6A, Danger??

[aerosimon]

HI CPF people, is my maglite a bomb?

I have a MR16 mag setup which utilises the above mentioned 9 in series cr123a at 3.6A giving 61W fresh. To date all use has been in burst mode of 15seconds max!

Preperation includes Flash-amp matching of cells and regular removal and retesting (monitoring) of all cells. Initially i even combined brands but still matched (flashamp) cells and had no real issues but did note resting voltage curve and recovery differences. My biggest change came about when on 2 seperate occasions, 1 SF battery in the string shutdown, and stayed dead, reading 2.7+ volts, but dribbling .1A.. Since i have stopped using SF in this torch.

I have now had 2 full sets of Titaniums over 3 months in use (all burst mode) without drama. Im confident with this arrangement, yet am being cautious. Its not something to give to a friend for christmas... Yet

I have searched and read up both great Lithium battery reviews by Silverfox (i think/thank/sorry if im wrong) and also scary venting/explosive stories.

Does anybody else have experience with similar power supply arrangements as i have setup. I am interested in long strings of Li cells, and similar or greater Current draws. Realistically, anything you think is useful, would likely be, so please jump in. Is it possible that the ends of a long stick drain or heat unevenly? (This is one thing iv never answered, but have thought about).

I think the surefire N62 lamp assembly uses 4 cr123 cells for 500Lumens, so think that has a similar current drain, so a commercial product with significanltly higher than 2A drain is available, but I have noted it is a burst use device. If there are any long run users of this N62, please jump in with comments and experience.

Is my cautious side just being overcautious. If so it is my left side, because the right hand always carries this maglite

but its definitely kept me safe.. If anyone can comfort me, with experience of similar safe abuse to cells, I would appreciate it, as ultimately i would like to know its safe for longer runs.

Otherwise, thanks and take care
aerosimon

 

[mdocod]

With 9 cells in series, I wouldn't want to run any higher than 2 amps... any time you have that many lithium primaries in a configuration, at any load, there is a lot of potential for problems to arise.. But I must say I applaud the creativity and your very careful approach with lots of testing and short burst usage.

I'm trying to think of a different configuration that could be safer and deliver similar power inside that amount of space...

is this a tri-bored 2D?

[edit in] which lamp are you driving with this? I'm taking a guess that this is a 12V 6000 hour 35W

[edit in again] I think ~2.5amp is considered the upper safe limit for CR123s. this is where Surefire runs their high power lamps. I'm not familiar with an N62 lamp... was that the optional 500 lumen lamp for the M4s way back in the day that was discontinued? I think I recall something about that- like 10 minutes estimated runtime and a warning about using it in short bursts only... but I could be pulling stuff out of a clogged subconscious dream/memory, so disregard this if I am way off base... maybe someone else who knows more about SF history can chime in on this...


[edit again] just noticed, this is my 1,500th post!!!

 

[aerosimon]

Gday mdocod

congrats on 1501 posts..

You are right about the N62, Im also just remembering what i think size15s has written in the past.

Im using a 6C mag, so all direct series with a paper/masking tape tube keeping it all central. 9 123s fit into it just about perfect but to prevent squashing the cells too much, I have reversed the spring and use the top off a C size energizer between spring and last cell so theres a bit of give. Its my favourite torch for house checking, its very lightweight, and nice length (personal preference).

My choice of bulb was as you guessed, but specifically the Osram Decostar IRC (heat reflective coating) thats rated at 12V 35W 4000Hrs.. I tried their 20W driven to over 35 but its a very different beam profile, more spread than 10degrees states and more blue tint, which i didnt like. Also it has two distinct regions, very odd in my opinion.

I am awaiting a 20W 8deg phillips I hope will put out a nice beam, and give a bit more overdrive room in the 9 cell setup. Initially i thought 7 cells for the 35 but it just kept going. In testing, I also tried the 50W osram, and for the record it took 105W input (5.2Amps if i remember) and was toasty.. But working... This was out of a host and using my 6V lead acid array plus steps with nimh... NO cr123s there

Im looking at this for my sisters bf's mag 6D on nimh.. Im sure the lumen rerating is not effective after that overdrive

So for me, its as is until the phillips arrives. I believe it will be right on the 2A threshold. Stepping back will be hard though..

take care
aerosimon





Also, just a note, iv actually pushed a 50W to 105W, it was about 5.2A running. BUt very hot.. I am awaiting a 20W phillips 8degree,

 

[mdocod]

50W to 105W, it was about 5.2A

I think that the instructions for starting a Lithium battery fire mention something like that, lol... YIKES!

interestingly, the voltage on fresh cells is sagging enough to prevent instaflash... keeping in mind, that 9 FRESH CR123s is almost 29V open circuit, they must be sagging to about 2.1V per cell under that load...

I was going to suggest... to reduce the risk of instaflashing your new philips, start with 7-8 cells and see how it looks. As you go to lower powered bulbs, you won't have as much voltage sag to save the bulb.

I think your best rechargable solution in that host is 5 of AWs new C size Li-Ions driving a 35W. 5-6 18500s might also work, but You'd be limited to 20W bulbs for safety reasons. (keep 18500s under 3 amps).

 

[aerosimon]

Yeah, the Titaniums say they have PTC kicking in at 5Amps, (unless my memory is off). That means its the fire that shuts the cell down from operation

Lucky leadacids dont have these 5A limits. I have done a fair bit of testing with CR123 primaries though, and some pelican bulbs, and MR16s. I note it all down too, only problem is my databook is in perth, and im in melbourne. I'll have it sent over as it had my years worth of sacrificing bulbs, and testing batteries. Ill share when i get it. Experimenting is great, but I could have had a serious turnkey for all the 3854s and 4303s iv blown. Ill definitely be starting slow on the phillips, thanks for the word.

As for Sag: Osram 35W IRC Checking my 6C now..
I cant rig up to measure Von right now, but V rest is 26.7V, 9cells=2.97/cell, and the starting spike current for my 35 (MR16 constant) is 4.63A inrush and settles at 3.45.. (fresh it settles at 3.6+) What would LiIons be like by comparison for instaflashing.. Better or worse.. Anyway, pulse current is looking like close on 100W,, for a 35W globe.

Ill see how the phillips 20W looks. If it looks nice, then it might be the solution. Also, this is mainly my by the bed/search the house and yard light, so primaries are probly financially reasonable after testing is finished. Also, i want to complete my ROP properly... Ahh,, more money..

aerosimon