I’ve been following the C8 threads sporadically, and I recently put down $12 to purchase one of these lights from an eBay user. Long story short, I got a flashlight with some fairly serious defects, and this happened:
I’m wondering what can be done to correct the fubar construction of this pill (if anything). I’m likely going to need a new emitter retainer ring, a new battery spring, and an LED drive circuit. The circuit on this one makes a high pitched noise and goes to ‘next mode’ every time the light’s turned on, both of which are exactly the things I hoped to avoid.
I saw a few really nice C8 builds here a while back, one of them apparently had a copper pill or something like that, so I was wondering if anyone might be able to advise me on what’s currently out on the market for options in a custom build.
I also assume there is a two prong wrench of some kind I’ll need to find to remove the existing retainer ring and pill. Any suggestions?
I think the retaining ring is made of low temp plastic. Material flaw.
As a quick workaround, you can try to cut out a bigger hole on the retaining ring center. This type of retaining ring is meant to secure the base on the sides, so IMO bigger center hole is ok.
You can get rid of the next mode memory with a pencil. It becomes “always-start-high” mode.
Copper pill, while very cool and purty, isn’t really necessary.
USA? I would head on over to mtnelectronics . Get a Qlite 3.04A driver (none of mine have ever had the high pitched whine). While you’re in there upgrade to an XML2 or XPG2 emitter on noctigon. The springs should be fine. Get some 22 AWG wire. And grab a pair of sturdy tweezers that will allow you to remove the pill from the head of the C8.
Or if you can live with the whine of the driver do as pulsar says and cut a larger hole in the plastic gasket that is over the star and burning.
That would be the logical assumption, but no. It was definitely the smell of ABS plastic burning.
I just took a second look.
The ring has managed to fuse to the bottom end of the reflector, so this time it came right off without issue. Melted ring gone, smoke gone.
Now I’m looking at the star itself, and after turning the light on for a minute it immediately cut out because the solder on the power contacts melted away and one wire became disconnected. This let me peek underneath, where I promptly discovered the source of my problems.
… THEY BUILT THIS THING WITHOUT SECURING THE STAR TO THE PILL OR USING THERMAL PASTE.
facepalm
The pill surface itself is as smooth as glass, no screws or mount points or anything.
The star itself is intact, and the LED works … it seems the solder failed before the emitter had sufficient chance to burn out.
Well it’s pretty much a bottom of the line cheap c8. You’ll find Ultrafire c8’s for as low as $4 sometimes (they make an okay cheap host at that price.) There are decent Ultrafire C8’s out there, but I see far more cheap quality complaints with Ultrafire c8 compared to say Convoy, tangsfire, or Xintd.
I did the same thing to myself. My first C-8 build I couldn’t fit the white plastic-shielding-centering doo-dad on there so I took the black ring from one of my old toasted XML’s and placed it on there. I needed to insulate the reflector from the emitter as I was shorting and not able to switch modes. I thought the smoke was from the excessive amount of flux that I used but nope, just like you discovered its that cheap black plastic. It melted the dome and I had to re-flow another emitter. All good now.
I got this problem too when I received my Tangsfire. Switch off immediately when smoke appears, this is to reduce further damage. Replace the burn centering ring/ gasket with another grade. You should be good to go if LED is intact. Main cause of the burn might be due to LED not sitting flash on star.
Did anyone else here experience “unsecured star, no thermal paste”?
What I was wondering at this point is whether the emitter will still be viable, or if I should junk the lot and start over. It’s an XM-L U2, I see no scratches or visible damage on the star plate, it seems to still be stuck to it OK, and the dome isn’t wrecked or discoloured, just covered in a fine coat of dust and detritus from whatever garage monkey assembled this thing.
Assembly was shoddy on all fronts, I didn’t even need tweezers to get the pill out, a light touch with a screwdriver on one hole was all it took to get turning.
i had it happen to a simmilar light, i discovered the pill was completely hollow ( no heatsink metal under the star at all) and the emiter got really hot to the point it melted the plastic centering ring.
The pill is threaded into the host correct? It looks like you’ve identified the main culprit(no thermal bond) but any high power light (and I start this catagory at >1A) will benefit from:
1). A copper star, the biggest that fits the pill, to draw the heat away from the led junction and maximize suface contact with the pill.
2). Thermal grease or adhesive between the star and pill, the smallest amount you can get to smear over the entire surface.
3). Thermal grease on the pill threads. Once again this is a bottleneck for heat.
4). At higher currents above 3A it would be worth looking into getting rid of any brass left between the led and the host. A press fit with a homemade copper pill will perform better than a piece of brass. Hold a piece of copper wire in one hand and same diameter brass wire in the other and put the ends in the flame of a stove. See how much longer you can hold the brass than the copper. Thats thermal resistance. That test will work with a steel coat hanger as well. Copper is better, period. Even alloyed plumbing parts are better than brass.
Brass is pretty. It polishes great. It is stronger and harder than copper or aluminum and with a bit of lead added is easy to machine. It is electrically conductive and can be soldered and brazed. It’s great stuff. But, it isn’t very good material for conducting heat from the star to the host.
I must have found a dozen variants of this light when I began searching around. Not even sure about the one I have. It looks similar. The pill is solid aluminum with two holes and a very roomy area underneath the star mounting surface where the driver would reside. The host body is solid aluminum, although knurling and threads and printing and several other details are all slightly off, and markedly different from other UFs. I’m assuming this must be one of the better quality copies of the C8. What I really need to find out is what ‘normal’ measurements would be in this host for threads and diameters, for buying parts.
One thing I liked in my WF-501b was a stock separator/spacer that sits between the battery and the actual compartment holding the P60 drop-in. It served to prevent undue stress on the battery spring, which was a godsend as it’s survived many drops on concrete, whereas my L2T cracked a dropin spring in one impact. Huge benefit and difference there!
I also heard some interesting tales about someone once shoving 3 XM-Ls inside one of these things with high current drivers. Can the C8 host even handle that much heat? I’d love to stuff as much power under the cap as I can, but in the end I’m only going to be happy with something that can do 100% duty cycle.
Brightness modes? Don’t need anything that flashes. Something that’s 1-mode tactical is best as it’s simpler and harder to break, but I wouldn’t mind going 3-mode with output of 100%/50%/5%. I’ve got an E03 with three mode, and that experience taught me if I go that way, I want persistent memory that stays put and remembers the current mode even if the battery is removed. Does this exist?
I have the convoy version from MtnElectronics Used to build my Olympic torch.
I think it could handle the heat longer than a lot of other hosts. But the light would be wasted unless you remove 20mm from the bottom of the reflector. Then you could add copper as a riser but I think the beam might look weird- might look decent. Interesting idea.
I found that if you really want to buy a “Ultrafire” brand C8 Lightmalls is a good place to go.
I own one with XM-L U3 in stock and its quality is good. Driver is not your favourable Nanjg 105C, but the high mode is direct drive. Two disco modes and visible PWM in lower modes though. For those who loves direct drive C8 this can be your choice. I know the Tangsfire C8 is hot now and I own one too.
The price of the LM C8 however is around the same with the Convoy C8.
I think smoke is hotter than an led should get, so if you use it anyway carry a spare light. Since it works, more likely poor heat sink or high resistance connection than a short circuit. One thread described smoke because the star was loose from the pill. Seems a choice between doing the standard emitter and driver swaps or ordering another C8.
Absence of thermal paste or even a hollow pill will not make a retaining ring melt, but a board not pressed down on the pill will, the c8 is sensitive to that because usually it needs the reflector to press down the board, so the tolerances in the head are critical.
The xml led itself can handle loads of overheating, I have never noticed any damage.
If the reflector fits too loosely, maybe he could sand or file down the front of the head and/or the back of the bezel until it is snug. That would lose the anodization of those two surfaces.
The leading inside edge of the head might do it, after all it’s aluminum and will passivate once exposed, so I wouldn’t be too concerned there.
Then again, I would just go with thermal epoxy for the star and avoid all of that. It’s clear the tolerances on this unit are pooched from the get-go, so no sense fiddling around with it any further.
I do however want to know one thing — what active sales sites exist for C8 OP reflectors and AR front lenses? All the ones I saw in previous posts either don’t have them anymore, or were flagged as discontinued.
