Cool, thanks for testing that. I have wondered how much of a difference it makes. I’ll go back to my old way. I used to simply put way too much on then when it melted I would pick it up with my hemo’s and tap it. The excess would bead up and roll off. Put it back down and then place the emitter on the pad.
This worked pretty much every single time. The hardest part with the XP’s is not getting too much in the center so that the outer legs don’t want to make good contact.
The issue with LEDs in my flashlights where that they just were not soldered properly.
They were holding by very small amount of solder. And the LED had any connection only to the some part of the contact pads.
Here is a bigger version of the picture:
You can clearly see that there is not enough solder there.
Contact pads should be tinned so the LED can make a good contact with them.
I’m currently waiting on high-amp batteries so I can do more proper lux testing… but I’m tempted to re-focus the reflector again first anyway.
I had been doing it by pointing the light at the ceiling on low while making tiny adjustments to the reflector position, watching the hotspot and trying to get the bezel tightened while the hotspot is in an ideal shape. I think I might try again with it aimed at a lux meter though, since even when it looks right by eye I still got 6kcd less than what it had initially. Then again, I might leave it that way because the spot actually looks a little nicer and more even now despite the 7% lux decrease.
Regardless though, the hotspot still has a sizeable and somewhat bright corona. It’d be nice if the reflector were able to aim the corona’s light into the spot instead of just near it.
How do the Efest 35a and Panasonic ncr18650pf compare to the 20r and 25r in terms of high current delivery? I need some high amp 18650’s and I’m trying to decide what ones to get.
Both of those are good cells Legolas, and you can hardly go wrong with either.
In my findings, the 20R reigns supreme in power delivery, followed by the Sony C4/Efest 35A, then Sony C5, Samsung 25R, Panasonic PF.
The issue is capacity vs delivery, and in that case the C4, 35A and C5 along with the 25R will all be nearly equal choices. The PF has the most capacity, but the least power delivery in these top cell choices while the 20R has the least capacity and most delivery. That’s how they generally work in my testing.
Thanks, DBCstm! I guess I will go with the Efest 35a. Seems to be the best of both worlds, at least for my uses.
+1
My ‘reflow’ is quite crude as I use a solder gun to heat the board, but always be sure to fully tin the contact surfaces under the LED, and that there is enough flux. (I don’t use paste either, just solder) …then just enough heat to ‘float’ the LED into position. Really need to check what happened here and fix the process.
Thanks for supporting me. I was actually quite surprised when I found why 18sixfifty’s light was not working.
At some point, I’d like to replace the lens with an anti-reflective one. When I was using my lux meter to help me focus the reflector, I noticed it gets like 10% less lux with the lens on.
I also seem to lose a few percent while tightening the bezel, though I did finally manage to get it somewhat close to where it was originally. It’s getting 86 kcd now, but with a better lens and slightly better focus I wouldn’t be surprised to get 95 kcd or even 100 kcd.
I took it out on a night-time bike ride just now, and it was a lot of fun. This seems like a really practical item to have for night adventures, since it lets me see about as far as the TK75 I used to have and (unlike the TK75) it’s actually small enough and light enough to carry.
I wonder though, if a different reflector might be able to focus more of the light into the hotspot instead of the corona. I’m not sure, but I think the stock reflector might be designed for XM-L and I hear the XR-E ones work best.
I have read those kind of stories too, but I don't understand them, a parabolic reflector is just a parabolic reflector, no magical shape possible that throws better than plain parabolic.
I think it is a myth.
I thought that too, at first, then it occurred to me that while a parabola is the ideal shape, placement of the emitter in the apex of this shape is critical and even a slight difference can cause a big change. The XR-E is quite a bit smaller than the XM-L and as such, is a much tighter point source of light and will need to be in a smaller area of the apex of the parabola. The larger XM-L footprint can be much deeper into the “bowl” and is why you see the large flat area…essentially the apex is cut off and the bigger source is located further into the parabola.
This is the only way I can see that would “optimize” a parabolic curve to suit the different emitters. Could be totally off track…
A parabola can be selected with the focus anywhere on the axis. That's why reflectors made for incandescents don't generally work well for LED's.
Hope the soldering problem got sorted out on future builds……
I just use an old Weller solder gun to heat the board and some kester solder. … no paste but be sure to fully tin the board before floating the LED into place.
Each and every buyer comes with a set of expectations.
Working on your own personal lights vs. multiple units becomes a job and no longer a hobby.
I believe E1320 may have experienced this in his UF2100 group buy.
For all your hard work all of us or practically all of us are grateful.
Thanks.
I did test each light and spent a good deal of time focusing each light as well. What I hoped to put out was a light with good run time that wouldn’t overheat and would maintain it’s throw without getting zapped by thermal sag. I was actually very happy with them. From my perspective they did what I hoped they would do. They seemed to be a light I could take out for a good long walk and maintain excellent throw for such a small size.
What I noticed from them was that they did their best job lighting up a couple hundred yards away. That seemed to be their sweet spot. The small hotspot in the center had a chance by then to open up and did a good job illuminating at that distance. Each thrower has it’s sweet spot. I suppose for each person it varies. I do know this though all of the lights worked just fine when I sent them out.
Also I am very sorry if anyone’s didn’t work as expected. Also I would have been more than happy to make it right with anyone who was disappointed with their’s. I don’t mind the posting of the mistake I made on the light, but as I said. I did test each light and spent a good deal of time focusing each of these lights. When modding a thrower getting the emitter centered and leveled to get the best beam is the hardest part.
I’m pretty happy with mine. I had it out on a bike ride again last night, and it’s a great way to see things at a distance without having to actually go look up close. I think I probably terrified a fox hiding out in the field; those reflective eyes are hard to miss.
At least it wasn’t a bunch of deer this time. One night I thought I saw a cat, and then suddenly there were several cats looking at me, and then the pairs of eyes lifted up far, far off the ground. This was up close, too, before I had a thrower to use. It really freaked me out until I realized they were just deer grazing on the ground.
That’s cool, I have tons of deer here and I never get tired of watching them. I had to take down some ceder trees this winter to get to my well to put in a new pump. The deer were all over those downed trees. I would have ten to twenty of them in each night just munching away. I also have a small orchard and they dig through the snow all winter long getting at the fallen apples and pears. Plus the grass is pretty long under the trees too so they like that.
Glad you like your light. I think they are great small throwers. I use mine fairly often and I have a bunch of throwers.
What that means is that you have to focus the led well, and I agree that is very important.
Quite few people agree that the XR-E reflector is the best one for the C8, the only reason that I can think of why that is so, is that that specific reflector is just a better made one, not that it is optically 'optimised' for a XR-E, whatever that would mean..
Really, just to point out a soldering problem and to help improve LED reflow technique. I have my own ‘hall of shame’; LED with solder burn on the dome, failed de-domes, burned or broke this, messed up that. (…don’t we all?) Most MCPCB are already tinned with solder, but looks like the copper comes bare. It’s important to make sure the pads are fully tinned with solder for good electrical & thermal path. I would fully tin the board first before reflow, but maybe there are other ways to do it. That’s all.
While the basic shape of a parabola remains the same old “x squared”, the other parameters can vary. Width of the curve, depth until the front edge, how much of the back is omitted, size of the rear opening, etc. All these proportions can vary independently, and that will change the beam pattern and how small its effective focal point must be and where exactly the focal point is. And that’s assuming it’s truly a parabola. You could also shift the entire thing outward from its center axis, turning the focal point into a focal ring… which might be a really good idea for use with LEDs, and would be highly sensitive to the size of the emitter.
Quality of the actual optics is a different matter entirely. And then there’s the matter of intentional flaws in the reflector, like the ridges used on a RRT01, or using texture to scatter light for a smoother beam.
Anyway, all I’m getting at is that there are at least half a dozen independent variables here and it’s very possible to optimize a reflector for different purposes and emitter sizes.