Years ago I built a Solarforce L2M that would be an 18350 size. But by creative shortening of the P60 pill and shorter springs I was able to fit an 18500. It’s an old light now with an XPG neutral white at 1.4 amps, but it still makes a nice pocket carry.
I also have a couple of Home Depot 3 AAA lights that I changed the LED to an XML and use an 18500 with a PVC pipe spacer.
I bought some 18500s for an RC transmitter once so I would buy a Convoy S2+ style flashlight that was made for 18500s - I don’t have anything else to do with them.
If you got the 18350 “shorty” tube, you can swap the reflector for a TIR lens, and if need be, shorten the springs on driver and/or tailcap. That usually leaves enough room to shoehorn an 18500 cell in there.
Also, as long as you’re not mounting it on a gun where the driver might be hammered by the cell by recoil, you can solder one of those small brass pillars/buttons, or just desolder the spring and fatten up the pad with a big blob of solder.
I accidentally bought a couple 18500s and returning them would have been a hassle.
For one use, I shortened the tube of a Wowtac A6 to take 18500s. Along with some other mods I like this light quite a bit.
I plan try the S2+ triple mod others have mentioned. Hopefully I’ll be able to use an 18500 with the stock 18350 tube, and an 18650 with the Jaxman 18500 tube.
Eventually I plan to bore out an old Nighteye aa light to take an 18500, although I don’t really like the light all that much so it’s pretty far down on my list of projects.
What method did you use to shorten the Wowtac A6 tube?
I’ve been thinking of shortening a Wowtac A6 tube myself, but without a way to cut new threads, the only method that presents is Old Lumens’ “human lathe” method. I’ve never tried it before and not convinced it would give a good result.
What did you do? If possible, please post pictures…
Firelight2, I’ll try to get a pic tomorrow. Basically, I didn’t worry about threads on the tailcap. Hack the tube down to size then use a rotary tool sanding drum to remove the threads from inside the tailcap. File / sand down the end of the tube so that the new smooth tailcap will fit over it. Use a conductive epoxy to glue the tailcap on.
Since there’s just a spring in the tailcap, and the head threads off for battery swaps, the removable tailcap is redundant. Just glue it on (but don’t electrically isolate it in the process).
Edit: Here’s the pic vs a stock A6 (above). The mod isn’t finished yet. I need to do some fine-tuning of the tailcap fit before gluing it together.
It’s a fun light to mod. I’ve been trying to find a suitable magnet for the tail but no luck so far. I also want to make a notch for a low profile friction fit clip. I already put in an SST20 4000k and an S2+ frosted TIR. Those TIRs fit great in terms of diameter but they are too deep for the A6, thus the base of the TIR needs to be sanded. I tried several options and only the frosted versions were free of ugly focusing artifacts.
I use 18500 similarly to mattlward. Way better performance than a 14500 and retains AA capability with the right driver. I have a pretty nice SST-20 P60 with an FM body set up like this.
I really like 18500, they have far more power than an 18350 which I hate. Yes 18650 has even more power, but 18500 is a nice compromise. Unfortunately 18500 lights are few and far between, and high capacity protected 18500 cells are scarce.
Nikon use 18500 cells in their camera battery packs.
Shorter TIR lens vs reflector + short springs lets you stick an 18500 (vs 18350) into a S2+ with shorty tube.
18500 fits my milk-frother vs the come-with 3×AAA holder.
Bunch of different reasons…
Doh. Already mentioned the S2+ not just once but twice above. Don’t recall this thread, so had no idea I already saw/replied before. Oh well…
About 18500 replacing 3×AAA, though, cheap lights rely on the internal resistance of the alkaleaks to limit current to the relatively puny LEDs, and sticking a Li cell without a ballast resistor may fry 'em in a jiffy.
Stuff like my motorised milk frother runs fine on a Li cell. Lower voltage vs AAAs but also way less IR drop so it maintains that voltage.
Yep, I made an S2+ triple like that. I used the T3 driver so I can run it off an AA, 14500, or 18500. I suppose I could make a battery spacer and run it off an 18350 (or 18650 with the regular tube, for that matter).
Power from a 'puter port so it can’t supply all that amperage if pulled.
Use it in small 5min bursts. Pull the cell and see how warm it is. Let it rest in between.
Check the cell in the light to see the voltage. If it goes from, say, 3.2V to 3.9V after a bunch of 5min blips, that might be enough. Once you hit 4.0V or so, charging naturally slows down (CC to CV) and you can just leave it in the rest of the way.
Lights don’t normally charge more than 1A or so, so while it might be approaching a bit more than 1C or so, it’s still a bit, hence the “resting” in between 5min blips to cool down and collect its wits.
Again, it’s not something I do on the regular, but if I can’t find my LK-101 or whatever, and can’t get arsed to unpack my Opus just for a quick hit, it’s doable.
OK, so it seems that you’re really cautious and don’t let it charge for long. I don’t know how these built-in chargers are designed… if they use a set of established “library” modules on governing the charge process or if they’re written from scratch every time. My guess is that a properly written routine would be compatible with a variety of cell sizes and amperages. I don’t think there’s any way for a test to be done in a matter of a few seconds to determine if a cell inserted is 10440, 18350, 18650, etc. It’s probably a matter of the increase in resistance as the cell reaches a maximum voltage. But the key is the amperage of the charging. And 1A is a safe bet.
What had me thinking that there’s different hardware or firmware involved was seeing some flashlights charging to different maximums. I had one that would go to 4.2v, several to 4.1v and one to 4.0v.