CMG IU modification w/beamshots

Well, i had burnt out the circuitry in my CMG IU like 2 years ago? and it was just sitting around. I just bought some k2s and 2 new 14500s and thought 'what the hell'. So i stayed late afterwork tonight (to use their shop) and made my own aluminum heatsink (nothin fany, just sits in the head, no threads) with a UWOP K2 being direct driven by a protected AW 14500.

I sanded the anodization off the reflector with 400 grit then tried to polish it with polishing paper. Came out ok, not too great though. I drilled a small hole in the front of the reflector to let more light from the led out, but i might have to drill another one in the back and look at securing the whole internals better. I'll post up beamshots later tonight if i can get my 'issue' fixed.

I'm thinking i have a bad connection somewhere because when the light is on, the output level fluxuates. You can see the output jump up, then drop down for a bit, then up then down. It seems to hold the same output level for a while as in it isn't an infinitely variable connection.


My other question is, am i at risk of burning this k2 out being direct driven? It is a 'N' bin so it is 4.23 - 4.47 volts. What about current? Do i need a resistor? Also, if i want to add 2 stage, what resistor would be good for 10% level?


EDIT: What everyone loves My first mod btw.

Canon S50:
1/500s
f/2.8 iso400
6"


1/30s
f/2.8 iso 400
6"


1/30s
f/2.8 iso 400
3 feet.


Reflector Shot.
http://allyoucanupload.webshots.com/v/2001227950639815649

Solved the flickering issue. The wire i had coming around the heatsink - the insulation had cracked and was causing the slightest of shorts. Its brighter than in the beamshots. Not sure how much, but noticeably. No more flickering. Once i get the ok that i'm not going to destroy the k2 running it DD with no resistor, i'll do a runtime test.

Looks like the LED is sitting a bit low inside the 'reflector'. Maybe you'll have to make the aluminum below the hole a little thinner for the LED's dome to portrude through the hole a little more.

The K2 should be OK DD by the 14500 cell provided the heatsinking is good. Does the heatsink have a lot of its surface in contact with the head? Did you remove any anodizing from these contact points? If you have thermal grease, you can put a bit of it arround the heatsink to help transfer heat to the head/body of the light. Better yet, make another heatsink which fits snugly inside of the head and you should be good.

I'm not an expert though.

Great idea though. :twothumbs:

The heatsink touches the head all the way around and i put thermal paste all around the edges and on top in the head. The head gets warm in under a minute and heats up quickly with the whole body heating up rather quick. I like how you said 'reflector.' Made me laugh. I'm looking for a lens to mount in the front to throw the light a bit further. It looks to be about a 15mm opening? Anyone have any suggestions?

I think i'll have to monitor that. It gets hot. I don't think the IU is big enough to cope with the DD k2. Might add a resistor. What resistor should i add to get say 10% output?

Just did a runtest. 1 hr ~ 1:20 of considerably bright light (relative to max). Trailed off after that. Tested. 2.85 V, K2 is still firing UWON. Stopped test just b/c. Its enough light for an emergency at this point. More than enough to read by it in a dark room, but not enough to illuminate much

why are the shots at 400 iso btW?

No real reason other than i was awkwardly holding both and wanted to make sure i didn't get a crapload of camera shake *shrugs*