As compared to a XinTD C8 from IO which gets hot quick and seems to be well heat sinked, how much better would a C8 be if it was built using a copper pill. What driver, how long would the run time be, how quickly would it get too hot to hold?
The HD2010 seems to be a better candidate to be driven hard with it’s massive mass. If I purchased the copper pill what driver should I use, what would be the run time, how quickly would it get too hot to hold?
Just a heads up you have brass in title, copper in first sentence, brass in second sentence.
You mention copper and brass, what pill are you looking at? pure copper and brass are at each end of the heat transfer scale, but both have more thermal mass than alloy, but it’s hard to decipher what you want to use?
Chris
Copper pill, thanks for pointing that out.
The HD2010 definitely handles the heat better. In fact I wouldn’t worry about it ever getting “Too hot to hold”, even running at 5Amps from a 26650 it just gets warm really. The battery will fade in performance anyway so that will level out the temperature after a few mins. If long runs are your concern, the HD2010 wins hands down.
The main benefits of the HD2010 copper pill would be…
1. Higher thermal capacity and better conductivity, can pull more heat out of the emitter PCB and do it faster than Aluminium so it will keep the emitter cooler under the same drive current. As a result you’ll see less initial output sag as the emitter gets up to temperature. This is probably not going to be a very noticeable difference though, certainly not to the naked eye but lost lumens are lost lumens 
It also means you can push the drive current even higher without fear of burning up the emitter, although on a single cell and using an original XML you’re going to be struggling to get past 6A, with an XML2 apparently it’s hard just to get above 5A.
2. The copper pill allows you to solder the driver directly into it.(standard solder will adhere to copper and brass but not to aluminium). This makes installing a replacement driver like a Nanjg 105c or KD 8x71235 trivial since you don’t have to worry about making/finding a brass adapter ring or going crazy with stacking 7135s onto the back of a Nanjg 101 like I had to This is a pretty big deal if you ask me, as I really struggled to mod my hd2010 initially and actually kept the East 092 driver in there because of the hassle.
—
The c8 at the same drive current will get quite toasty and far quicker, it’s not really an ideal host for being driven hard.
But it’s still fun to try and get the most out of that little guy and the copper pill will arguably make more of a noticeable difference there than in the HD2010.
Edit: Forgot the most important reason of all, copper is just kinda cool! 
It's also more massive so will make your flashlight more effective when the day comes that you need to use the attack bezel. 
+1 for Cool factor, in bling bling and not as hot both.
I can add some other great reasons.
...your light gets heavy. :love: And heavy is good, because you just love to drag around excessive weight every time you take your light with you..
....because its expensive.. Makes your light roughly twice the price, and twice the price is good.. just ask anyone on CPF... 0:)
....because copper is awesome, and heat transfer should be better.. At least that is what you say to yourself.. despite that there have been basically zero measurable gain upgrading to copper pills in those lights so far... :X
pretty much sums it up.. :D
Because a nice copper will get you of a speeding fine.
A copper pill a day will keep the doctor away. :bigsmile:
Copper? I don't know nothin 'bout Copper.
Really, a Copper pill will transfer heat from the led much faster than Aluminum or Brass/Bronze. How much faster? I have no test data other than playing with this stuff for a long time now. By feel and my own experiences I would feel safe to say that Copper transfers heat about 1-1/2 times as fast as Aluminum. As an example, let's take a piece of Aluminum and say that by heating it with a torch, I can feel the Aluminum getting hot in 45 seconds. With the same sized piece of Copper I can feel it heating up in 20-30 seconds. I know, that's not the kind of test you want, but it proves to me, that Copper is much better than Aluminum, for heat transfer.
We've been down this road over and over, up one side and down the other. It is a proven fact that Copper transfers heat thru it much faster than Aluminum. Look at any chart on the subject and it will clearly show that. Look at led testing on Copper vs Aluminum stars and it clearly shows an improvement in getting heat out from the underside of the die.
Where the problem lies is, where is that heat eventually going to go? Unless you have made a light completely out of Copper, then it will contact Aluminum at some point. The smaller the Copper heat sink, the faster that will happen. I liken it to driving on a two lane highway and all of a sudden there's road construction and you are down to one lane, (for the next 14 miles). If that two laner was even halfway full, you are now bogged down in very slow moving traffic. Your mood worsens and you get hotter and hotter, because there's no relief in sight. This is what happens when a Copper pill runs into an Aluminum head. When the Aluminum acts as a road block, the Copper keeps getting hotter and hotter, with no relief. So, you will eventually get so hot that the led may fail.
Remember also, that since Copper has more mass, it will hold heat longer, once it's heated up, so if you run the light hard for so long that you can't hardly touch the Alunimum head and you quickly turn it down, the Copper pill is going to hold that heat much longer than the Aluminum head will.
Now, if you had just stuck to the Aluminum pill, the heat transfer, for all intents and purposes, would be slow, but smoother, as the pill to head transfer should stay at approximately the same rate. Will that be better? I don't think so, because the Aluminum is slower to take heat, the led will initially get hotter faster, than if it was on Copper.
General rule of thumb (for me), If you intend on hot rodding a light and pushing a led far above it's "normal output", you are asking for it to die sooner than later. We all do it and we all love it, but it's just stressing the led. Using Copper will give you higher initial output for as long as it takes for the heat to bottleneck, then it's just as bad, (or worse), as Aluminum after that point.
This is just my own thoughts and meanderings. I am not a scientific person. I learn from real life experiences and what little common sense I have left, tells me Copper does transfer heat much faster than Aluminum, but if it's going to bottleneck in a few minutes anyhow, is it really and truly any better?
I use it a lot! Why?
It's very easy to work with, when you are using hand tools. It can be heated to make it soft and worked to make it hard again. It is much heavier, therefore more mass in the same space. It looks Cool! I love the look of polished Copper! It takes solder like a duck takes to water. It makes the initial output of a high powered led more efficient, till the heat bottlenecks.
Each person has to weigh the pros and cons. The smaller the Copper pill, the less useful it will be. The larger/thicker the Aluminum head, the less effective the Copper pill will be. Ideally, I would want a large Copper pill with a large, but thin, Aluminum head, with lots and lots of deep fins on the outside, to take heat away from the Aluminum as best as possible from the surrounding air.
I'm done ranting and spell checking, time for bed.
Why should I get a copper pill for my HD2010?? - because I butchered the alloy one practicing on it with my new end-mill bits in the drill press…. S)
I have seen this misconception in quite a few threads. It is assumed that the amount of heat stored by a pill is directly related to the mass of the pill. What is ignored is that the specific amount of heat/kg/K differs in aluminium and copper, in fact aluminium is better at it (880J/kg/K) than copper (380J/kg/K). But because copper is so so much more heavy than aluminium, if you compare the amount of stored heat/volume (say: one flashlight pill), copper still wins (3.4 kJ/liter/K for copper versus 2.4kJ/liter/K for aluminium), but the difference is really not that great, at least much less than the difference in weight suggests.
Concluding: copper is better at heat sinking than aluminium but in real life you will not really notice the small difference.
(heat conductance is another reason to consider a copper pill, in that, copper is twice as good as aluminium, I'm still convinced that aluminium and even brass pills are way sufficient to form a adequate heat path for any overdriven led, but that is a different discussion)
I think we’re giving aluminium an unfairly bad rep and praising copper too highly in terms of thermal characteristic mostly because of the obvious superiority of the copper sinkpad/noctigon performance. Aluminium is badly handicapped in that fight due to it’s inability to accept solder.
It seems like something that could be relatively easily tested. The only aspect I see that would be pretty difficult to quantify would be the difference in led life longevity when using different metals for heat sinking, but someone with the proper equipment should be able to accomplish the simpler comparisons in an evening of experimentation.
Of course several of the various differences mentioned in above comments would only be valid on specific flashlights, but someone could do one of the popular heavy hitters like hd 2010.
The way I see it the only thing we’re ultimately interested in is led junction temperature. The increase in performance/output we see from an led on a copper sinkpad vs a standard aluminium pcb is directly as a result of a lower led junction temperature and thus better efficiency at high drive currents.
Lifespan of the LED at above the rated current is a bit of a mystery but again I would think is going to be mostly down to junction temperature. Ultimately that’s the only effect a mcpcb can have on the emitter anyway. Better thermal transfer away from the led = more output and longer life.
So who’s up for making a brass pcb with a short pedestal and flowing an led to it to test!
Plus aluminum is highly reactive, within minutes of exposing a fresh surface there will be a layer of transparent aluminum oxide that most people forget about as it's virtually invisible, unlike copper oxides.
I've linked to some great measurement data performed by a member of german TLF numerous times, it basically showed that with a nearly perfect interface contact it doesnt really matter if the heatsink is made of copper, brass or aluminium. Nearly perfect means: polished, lapped and with as little thermal paste as possible screwed down. Most of this is NOT possible in flashlights, and thats the problem.
Well, kinda. Its just the next "big" problem. The first big problem was the dielectric layer directly under the LED. Thermal conductivity of less than 1W/mK, solder is at about 60 (and I'll say it again, a heatsink made out of solder would be awesome, its just hard to work with or solder something to it because well.. it will melt. :D), brass, copper and aluminium are even better of course, but the effect is not nearly as big. Its not a factor of 60 times better but maybe 6 times better. The law of diminishing return. But I'm getting lost.. anyways, the problem with the dielectric was solved. First by the famous LED-tech.de copper boards, then Sinkpad, then Noctigon, then IS's own boards and some custom approaches.
So much for the first bottleneck. The next one is how you attach the star to the pill. Most people glued it down (Fujik, 2-3W/mK IIRC; Arctic Silver 8-9W/mK), some screwed it down, but in an uneven machined thing like most pills, you wont get perfect contact with tools you usually have at home. So some people grinded, polished and lapped pill and star, that worked fine, but again, without the proper tools, very hard. And then came the flashlights with brass pills and some with custom copper pills. If all the flashlight manufacturers would just use brass pills, all our problems would be gone. Ever soldered a copper PCB to a brass pill? The heat transfer is awesome! Sure, it will be a little bit better with copper, but not that much. Even hot rods stay under 30W, remember incan times? Its not that much. You cant really compare it with the heat a butane torch creates.
Two bottlenecks gone, where is the next one you ask? The threads. An interference fit would be superior, but thats VERY hard to achieve and by far not trivial. I'm actually waiting for brass pills with a built in PCB. Saves one layer of solder. Or even better: pre-soldered aluminium. Its hard to solder to aluminium with home tools when its bare, but if it comes pre-soldered from factory, its the same as anything else.
Some might ask "why copper if brass is almost as good?" now.. because people want copper. Brass is way better than its reputation. Biggest advantage imo: different form of oxidation. Copper can lose electric contact when it oxidizes, not so with brass.
Hope that helps. :P
Brass can oxidize and it will .
Great post, thanks
I’d also take brass pills over aluminium any day, if given the choice.
Hair-splitter. :P I'll go edit it so its clearer.
In flashlight use it doesnt affect electric conductivity. I've had copper oxidize (oxy? oxi?) and lose negative contact because of that. Wont happen with brass.