Unimpressive Mod Results

G'evening folks,

Well, I've performed 2 different mods tonight and both ended up unexceptional. First up was the AA-S1. I swapped out the led to a Q5 and the driver with this one. Not having a flux meter yet, I settled for a ceiling bounce and also used my C3 as a reference. The aa-s1 was marginally brighter than stock after the update with a noticeable color shift (cool white now). Tried with both alkaline and 14500. The C3 still beat it. Not sure what else to do to save this light....

Second mod was on my X2000 (18650 3 mode). I swapped the P4 that came with it for an xr-e r2. All the spare drivers I had were too small for the pill so the stock one remains. For this one, I used a C78 for the control light. Before mod, the C78 with a 14500 was marginally brighter. Post mod they were about equal. Then I got frisky and bypassed the driver and set it up direct drive. Pulling @1.5amps at the tail I did another comparison. This time, the x2000 was marginally brighter. Btw, I did lap the star to the pill and thermal paste it to help with heat.

Maybe I just have unrealistic expectations of what an emitter / driver swap will do...

Btw, Oxy moron, I agree with you on the C78.... on a 14500 it's quickly becoming one of my favorite lights. I was just pondering on how I can mod it to improve it, but after the first two flops, I'm leaning to just leaving it as is... (already did the O-ring head mod).


If you want an easy and relatively cheap mod, the Nite Ize IQ tailswitch for AA Maglites works on the C78. You can get 'em off the 'Bay for about $6, but it needs the voltage from the 14500 to work.

The AA-S1 is my second preferred light.

The AA-S1 is perfect as it arrives.No need to change led or driver. It´s reflector is designed for OSRAM. A xr-e does not fit in the hole of the reflector. Maybe because of that, you dont see any improvement. HOWEVER I tested this AA-S1 with 14500 and another light with xr-e Q5 also with 14500 and no difference in lumens, only in beam characteristics and tint

The only thing that the AA-S1 need to be a perfect light, is a MULTIMODE driver. And a ramp mode driver will be even better.

However you can have a bad one..... so you decided to mod it

You nailed it... I did have a bad one (I think... I've never had another one to compare it to)

I'm not sure if you've tried it yourself, but the Q5 emitter fits the reflector Perfectly. Whether or not the shape of it is ideal for the q5, I'm not so sure...


Thanks for the tip on the tail cap. I'll have to pick one up.

Irregardless... I guess my question is, are my expectations too high for the aforementioned mods to these lights?

OK, your reflector fits nicely a xr-e. OK, perfect, so I have another batch different than yours.

I have the TF F20 and the AA-S1. The hole of the F20/akoray k106 is huge, it fits the xr-e.

In both AAS1 I have, the reflector hole is so small that the dome of the xr-e does not fit even using a hammer. Is very narrow.
ALSO the orange peel is different, is more aggressive in the AA-S1. Also I think that the reflector in the AAS1 is deeper than the F20.

It proves that depending the batch you get, you get different reflector, and different overall quality, among different tint of the LED (warm/cool)

Probably yours have the same reflector than the one in the F20 Akoray K106 (big hole)

About the expectations: As I told you, both of mine AAS1 only needs a multimode driver to be perfect!. No need to change the LED nor even the driver, because original driver gives enough juice to the LED


You seem a bit defensive on the aa-s1...I don't mean to put down anyones sacred cow. To me it was just a light I wasn't content with, so I decided to mod it. But I've got an idea...

Tonight I'll take some beam shots as it sits now (with the Q5) and then I'll return it to the stock driver/emitter and take a few more to post.

After that, if you'd like, I will happily mail you the aa-s1 as a gift. Then you can compare it to yours (hell, I might be crazy). If anything you'll have some spare parts :)

Moving on.....I guess I was really suprised that a direct drive R2 was only marginally brighter than the regulated Q3 in the X2000!

Ey dont worry, is only to inform you that with chinese cheapo lights, not always you will get the good units. Some units will be manufactured with high acceptable quality, but others will be crap. Also, among the good ones, there is a possibility that some will use some components (lets say drivers, reflectors, switches, etc...) and others will use other components.

Yes I am defensive on the AA-S1, because I own 2 and thinking into buy the 3rd one. The AA-S1, from lets say, 10 units manufactured, 7 are good, 2 not so good and 1 to throw to the garbage.

"Moving on.....I guess I was really suprised that a direct drive R2 was only marginally brighter than the regulated Q3 in the X2000!"

Not surprise at all, those lumen at full current given by the manufacturer are just a fairy tale for us in the real world.

Feeding a led with a power supply and an ampere meter you would be interesting to note that a XR-E give a increasing lot of light between said 600 mA to 900 m A , beyond that stop the ramp and begin to decrease.

With an improvised light box ( cheap 3,5 V solar panel ) and a stock C78's emitter XR-E attached to a 55 grs heat sink , noted the usual P60's pill weigh 22 grs, found some measures listed below.Of course with the crap pill of the usual zoomies you lost the battle before the first intent was made.

Note; up to 900 the output was stable even for longs periods, then when two measures were noted is when in a few seconds the light became to go down. at higher currents the output was less at the begining...in short, you can get more light at fewer current ...

Note 2; I don't wait more that 15" therefore beyond 1A as the heat increase the output decrease .

XR-E ez900 C78's stock emitter

100 mA 556 mV stable over time
350 mA 852 mV "
500 mA 955 mV "
610 mA 1.000 mV "
750 mA 1.061 mV "
900 mA 1.100 mV "
1.000 mA 1.119 mV to 1.112 mV at + - 15 ''
1.100 mA 1.129 mV " 1.120 mV " "
1.200 mA 1.144 mV " 1.125 mV " "
1.400 mA 1.140 mV " 1.130 mV " "
1.500 mA 1.130 mV
1.600 mA 1.100 mV

Sorry to hear the mods didn't work out as planned. I think I know what you mean though - actually, I think I know exactly what you mean.

I have a background in motorcycle racing. Not MotoGP-level stuff by any stretch of the imagination. Just a bunch of guys who enjoyed working on and racing bikes. Hell, Rossi, Stoner and Lorenzo could probably kick my butt riding lawn mowers. :D But one of the things you learn to pay attention to when you have a hobby like that is that details do matter. Not just because it's a dangerous sport but also because it's the small things that often make or break a race. Back then, there were no telemetrics and we didn't know the first thing about engine mappings because our bikes didn't even have digital fuel injection but we would still get excited about tinkering with this and that and finding half a second here and there. In other words, the grunt work was actually part of the fun.

The point of all that rambling is that when I got into flashlights, I still had that attitude. And I would get excited over small stuff. I didn't really do it for the results. I guess that might be one of the reasons why there's a lot of overlap between the outdoor/survival/gun, the flashlight and the motorcycle crowd. To some degree, we all do stuff that would make a regular person think we were crazy. Most people just want a flashlight to work. They don't care about the specifics as long as it does. Same goes for cars and motorcycles.

And then, at some point, my priorities changed and good enough started being, well, good enough. Life is short and working for hours on my bike just to suck a little less, relatively speaking, didn't seem like such a great idea all of a sudden. I still love motorcycles but obsessing about all the details was no longer fun. So I stopped. And that carried over to other parts of my life. That's why I hardly ever mod flashlights anymore unless I get a real dud. In terms of the effort/reward ratio, it never made sense anyway. Most budget lights are so cheap that there's no point economically in throwing good money after bad. But it was fun - for a while. But now I'd rather spend an hour reading a good book than try to push a Chinese light just a tad bit harder without breaking it. Maybe it's because I have enough lights that I am satisfied with. Maybe I'm just getting old. I obviously still like flashlights and I enjoy tinkering with them but I don't really like to turn a flashlight mod into a major project anymore.

Okay, that was way longer than it should have been. Sorry.

I'm glad you like it. Aside from improving the heat sinking, I usually just leave them alone. I'll definitely give the switch jekostas mentioned (thanks! :-)) a try though. I always thought the C78's clicky was one of its main weak points. Looks like there's one on eBay right now for $5.99 with free domestic shipping (190471992642).


Feeding a led with a power supply and an ampere meter you would be interesting to note that a XR-E give a increasing lot of light between said 600 mA to 900 m A , beyond that stop the ramp and begin to decrease.



Thanks for the very detailed info. I had susptition that I was entering diminishing returns over 1000ma, but didn't realize it would e so bad. The anemic heatsink on the X2000 doesn't help.


Thanks for the perspective. Good story too.... A few of my co-workers are into motorcycle racing, and it's interesting when I find they discover little things like angle of exhaust can make a difference. Closest I got was building up a mustang as a street bruiser. Thankfully no class restrictions to worry about so big gains were as easy as a pulley swap for more boost

I should have stated in my first post that even though the mods were disappointing, I'm still having fun figuring out what works and what doesn't. Btw, have you tried any of the glass lens replacements for your flood to zooms?

You're welcome on the mod idea - I can't really take credit for it. It's been posted a few times on the DX comments for the C78. I have done it, though, and it works. I gave one to my Dad as his emergency light over the summer.

just to throw my opinions in with you guys on the c78, although they do vary in quality, the good ones, with 14500's and the nite ize switches are very hard to beat as far as a small flood to throw is concerned. they're nice sized, run on AA's in a pinch, and their brightness and throw is as good or better than any other flood to throw i've had in my hands. i have two now. oddly, one has a pretty warmish emitter and the other fairly cool bluish, but they are both great. its hard to tell which is brighter from night to night.