I’m no specialist in this stuff, but it seems to me that the heat is going to follow as direct a line as it’s allowed to. The tiny emitter base is where it’s coming from, if allowed to flow into the pill freely it will have an angle of travel much like the light that flows from the emitter, or a 125º emission angle. I’ve seen charts that indicated that to be so. Thermal imaging was used to show that heat flows into the heat sink at almost exactly the same conical pattern as the radiance of light out the front.
That is , unless it’s bottlenecked, then it has no choice but to mushroom out looking (as it were) for the way out. Given a nice copper pill, with the star soldered to it, the heat is going down before it goes out. So a larger star won’t see any real heat until the sink starts seeing saturation.
Maybe so, but some of us would still prefer to use a larger star. It's a lot easier to fit a 16mm star in a 20mm hole than the other way around. If I have the choice between a HD2010 pill that doesn't fit a 20mm star and a pill that does like the one vestureofblood made, I'd go with the one that does.
A 16mm copper star in a copper pill is exactly the same as a 20mm copper star in the same size copper pil. A flashligh doesn’t know where the star ends and the pill begins. It’s all copper.
I’d love to help but I never used a standard battery tube with a new pill. My battery tubes were made to suit longer pills. It would not be hard to measure up though.
leaftye! The issue of the reflector on the HD2010’s shorting out on the connections is much more easily handled with the larger star. I had forgotten that I enlarged the pill in mine for a 20mm star but then, I’ve done so many things to mine it’s hard to remember just exactly where it does stand. lol
So what’s the ETA for something like this? I’d order more if there was more lead time.
It doesn’t look like the Convoy pills isn’t practical like I was hoping for by looking at these pictures. That would have been real slick if their lights were modular.
Anyone know if the Fandyfire HD2010 pill is different to the Ultrafire HD2010?
From the picture ryansoh3 posted above, it looks like the UF takes a larger driver, or at least has a thicker wall for the driver cavity. The topside appears to be the same.
Would you love to help a fellow Aussie, & also keep Aust. Post working, by sending me up one of those copper HD2010 slugs
Tom E measured no appreciable difference between a copper c8 pill and a brass equivalent, there’s pretty much your best and worst case scenario, I can see replacing alu with a more easily solder able alternative but its not going to magically improve your light over the improvements a direct bonded copper star will make, and its just as easy to silver solder a copper star to a brass pill as it is to solder a star to a copper pill, you still have a junction there and lead/rosin solder is a pretty crap thermal conductor tbh.
Good luck with this either way, I just don’t see the benefit, I’d love to be proved wrong but there’s nothing yet.
Personal experience tells me that the copper pill will take much more heat away from the star than an aluminum one will…all other things being equal. I have soldered connecting wires to the copper star on my HD2010 pill too many times. With this light I have learned to mod. It’s not really difficult to solder to the star when it’s on the stock aluminum pill. But, after I put 1/2” of copper in the core of the pill, cutting out the middle of the pill so that the copper star is sitting directly on top of the copper….my Hakko 888 station is taxed to solder on it now. So for me it’s fairly easy to see that the addition of a mass of copper to the pill has greatly increased it’s thermal absorption rate. If it sucks the heat out of my soldering iron set to 1000º F, then it will also suck the heat from any emitter I put in there.
A full copper pill would give even more benefit, as it carries even more mass.
Edit: I have also made light engines for the L2P consisting of a 7/8” diameter copper rod 1” long. The star is re-flowed directly to the top of this chunk of copper. If I don’t put some solder paste on the connection pads when I’m doing the reflow, I will not be able to solder the wires onto the star later. My Hakko goes to 1100ºF, it won’t do it if the star is not pre-tinned.
That’s about 3 ounces of copper. Overkill? Beyond me to say. Many of us do things the way we do simply because it’s what we want/like/believe. Works for me.
@DBCstm: Its all correct what you say, but gords main point was that copper is not better than brass. There is enough data to back that claim, though I disagree with lead being a bad conductor. Its about 8-10times better than Arctic Silver.
The main benefit of copper is, that it can extend your runtime on high in lights that are too small to handle the heat, because it takes longer to saturate the heatsink. Downside of that is, that such lights will take longer to cool down.
I'm an ultralight backpacker and (sports)car guy. Paying for slight mods to gain an edge in performance that's barely perceptible at best is part of the game.
You may be right though, but that's part of the reason I'm asking for slight design changes that will have tangible benefits.
Testing of a copper star on Alu, Cu and Ms (brass) heatsinks.
What matters is the thermal properties of the first material after the LED - the PCB. After that, its pretty much irrelevant. Seeing how good brass performs, a heatsink made out of solder would still work great. See how aluminium gives a lower temperature reading than copper, although the thermal conductivity of copper is higher? Thats because there are other values to consider.
whilst the pill may absorb more heat, it still has to shed it through the threads, increasing the mass/using a different material merely gives you a bigger heat store, it doesn’t necessarily translate to an ability to allow the light to shed more heat.
I guess thats the point both myself and nightcrawl are trying to make.
yeah, all that copper will make it harder to solder to the pads, but it still might not get more heat out of a light, and that is really the aim here, myself, I’m just happy if a pill is brass rather than alu, and really, according to the facts above, I should be gutted if this were the case.
If that is so, where is all the heat from my soldering iron going when the copper is present under the copper star, as compared to when aluminum is present under same star? Why does my light run much cooler at more than 2A more energy consumption when the copper is present?
Perhaps the test was done in open air and the heat sink had an easy job of it. In a confined environment like our lights, it’s not as simple as that. The test results shown above do not make any sense, but then I am not an Electrical Engineer. I just build excessive lights in small packages and use copper to do it. It works. 1800 lumens in a 4” light from an 18350 cell. So I stick with what works.
Sorry, it just throws me as a relative beginner when someone first goes with one theory, then aludes to another. NightCrawl used copper for a heatsink in a fantastic build of a quad earlier this year, and is now saying it wasn’t necessary. I get lost easy I guess, comes with old age and meds.
the heat goes there… to the copper. once you apply enough calories to raise the temp to the melting point, you will be able to solder it. you must have in account the calories lost in the surface of the copper chunk, but most of the heat is being accumulated in the copper, because it doesnt have a very good cooling design. in fact i think it hast the worst possible design, IE smaller surface possible
That's important to me. I still want to use lots of current even in small lights, or big showerhead lights. I only use the high mode for short periods, so it's okay if it takes a while to cool down so long as it's able to soak up the heat in the first place.
It'd be nice to see these copper pills made to maximize thermal mass. That's what I liked about how MRsDNF designed that HD2010 pill. It could even cost the machinist less to make, if for example the driver walls are kept as thick as possible.
Where was the temperature being measured for that test?
Copper has a higher specific heat (as well as higher thermal conductivity) than aluminium, so while there might not be much temperature difference, the copper can absorb more heat before increasing in 1C in temperature.
I don’t think it would take too long. As soon as we finalize a design, I’ll probably be able to finish them in a week or so.
