Mag 2C, RS KPR112 bulb, Li-Ion battery pack

[EvilLithiumMan]

I found a couple threads here telling of a quick and useful mod - put 3 CR123A cells into a Mag 2C and use them to overdrive a Radio Shack KPR112 bulb. I loved it. (I too favor the form and balance of the 2C). But being the evil guy I am, I'm not particularly fond of feeding CR123's every hour and a half to my flashlight. So I thought I would try to shove a pair of 18650 lithium ion cells into the 2C.

As it turns out, shove is definitely the operative term here. The two 18650 cells make for a tight lengthwise fit.

Two 18650 cells, plus a li-ion protection circuit from a discarded Canon BP511 pack make up the mod:

It is inconvenient to remove the pack for charging, so I made a 'charging adapter' from a PR bulb base:

How has it all turned out? - well, not too good actually. When I started, I wasn't using a fully charged battery so the KPR112's had no problem. But it turns out a full 8.4 volts on the pack will flash the bulb. I have flashed four of them now in an attempt to 'dial in' what voltage is the limit. The KPR118 (7.2v @ 600ma) works fine, but it isn't really being overdriven. (Although it does seem quite bright). At any rate, I guess until I achive decent reliability from the KPR112, the mod is a failure. /ubbthreads/images/graemlins/frown.gif
But I had a lot of fun, didn't I?

 

[Ginseng]

Hi,

Nice try, really! I'm not surprised the 2x18650 didn't fit all that well. They're 4x123 equivalent in length and 4x123 doesn't fit in a 2C without gyrations. It would go nicely in a 3C body though. You're not afraid to flash bulbs, that's a true sign of hot-wire worthiness /ubbthreads/images/graemlins/wink.gif I certainly wouldn't consider this a failure. It's a work in progress and a pretty good one at that.

We're thinking along the same lines. Since my ultra-high power torches are only used intermittently, I don't mind the use of SF123s. However, I'm dying to get more time out of my WA01185. I have purchased 3x18650 to feed this 30-watt superbulb. Since I do not have protection circuitry, I'll be using them bare but monitoring the charging (Pila charger) and discharing carefully.

I am able to get 5 3x18650 fully protected (OC/OD) battery packs made by a pack assembly house, but nobody seemed interested. 3.3A, 10.8V nominal. I guess $60 was too much.

Jolly good work!

Wilkey

 

[EvilLithiumMan]

Good luck with the naked li-ion cells. I recently bought a BN-V814 and ripped it apart for the two 18650's. I then wired the cells in parallel and, using the plastic from a milk carton and some Solder Wick, fashioned a replica of a battery pack for my SL Poly Stinger. It's standard pack consists of three 1.8AH sub-C nicads. By going to li-ion, I now had twice the capacity, 3.7AH, in a much lighter package.

Because the cells were new and I had no idea what level of charge existed, I wanted to discharge them first. Well, discharge them I did. /ubbthreads/images/graemlins/icon23.gif It was one or two in the morning and somewhere during the discharge cycle, I dozed off. When I woke, there was my new li-ion pack sitting at a staggering .57 volts! ("Those poor *******s never had a chance")

After complimenting myself on my gross incompetence, I removed the load and let the pack sit unattended for an hour. At that point it had recovered to 1.90 volts, so I decided to put it on a constant current of 100ma to see if I could get the voltage back to normal. After about an hour and a half it was at 3.5 volts. At that point I gave it a normal charge level, and it seems O.K. now. It's sitting in a Mag 3C, and I will probably test it more this weekend.

Anyway, my point here ( Oh good, 45 minutes later he's finally getting to the point. ) is watch those cells every minute.

Note: I couldn't use a protection circuit on the Stinger because the start up current to the bulb would cause the pack to shut down. (There is a solution for that: connect a n-channel FET with a time constant of 10-100ms. That would allow the bulb to do a 'soft start'. But for that to work, I need access to the 'output' side of the on/off switch and I saw no way to do that with the Stinger).

 

[Ginseng]

Wow,

0.57V. I would have guessed those cells would be dead for sure. At the very least, I would think their capacity would be severly reduced even if it does get back up into the high 3's.

I intend to be careful. Since I know, roughly where each cell needs to be (3.75V) I will charge to this level and then discharge to the target. I'll also be monitoring the time to discharge down to 3V and then be able to use that as a runtime guide. At least it sounds good in theory. I'd rather have a protected pack but I guess I'll make do.

We'll know in a few days.

Wilkey