Convoy S3 - highest continous current?

Nice!

i don’t care much about the host, not much to be damaged there, perhaps the o-rings or the reflector, but it’s pennies. I am somewhat concerned by permanently damaging the emitter thus reducing it’s output.

I am though more concerned about the battery doing the thermal runaway.

so perhaps I need to do a test to verify if something will get damaged.

Test it yourself if you like, but you aren't going to be running at higher currents than I do/did. These are direct drive lights with hot INR cells. The single LED lights run at a little over 5A, the triples are closer to 10A.

I’m running it max with 2.8A measured at the tail cap (8*7135), the reflector section of the light gets hot very fast. The star is aluminum, using very high quality thermal paste (gelid gc extreme).

What hosts did you use for the triple emitter lights when testing? I guess you know, but Convoy s3 is a pretty small light, just a tad bigger than the 18650. It’s got direct thermal path shelf for the emitter.

I have built several S3s. Single XML2 & XPG2, and a few triple XPG2s. You cannot make enough heat with only one cell, and the LED or LEDs on copper, to damage anything inside the light.

A quality cell should allow for non direct heating(say placing it in the oven, not attacking it with a blow torch) from ambience temperature to 130c for 60 minutes with no fire or explosions :slight_smile:

I wonder about hung lights vs tailstanding lights .

The heat hitting the battery harder since it's above the emitter .

that happens because of hot air streaming upwards, don’t think it’s significant in these cases.

keep in mind the li ion battery could get hot enough to make it dangerous, while the light is on and internal damage that does not become apparent till it explodes in the future

thanks for the info!

I feel safer now :slight_smile:


I can understand that. I'm wanting to find a way to log temperature in my high current battery tests. All I know right now is that some of them get hot enough that it hurts to touch and further shrink the wrapper, so I wait a few minutes before removing it from the battery holder.

I heat only rises where a convection current can exist. So in liquids and gases yes, but solids no.

Comfy when you are saying you can’t kill a light on a single cell, is that assuming a certain light model, copper base, and good thermal paste? Or just in general?

I think in reality in setting the light down to run on max on its own, the more probable concern is more what could a running S3 set on fire if you set it on the wrong material, or had someone try to pick up the hot light and reflexively drop it…

On any of the single cell small pocket lights. Threaded pill, integral shelf, fins or no fins, doesn't matter.

I only use cheap white MG Chemicals silicon-based paste in the big bulk 10oz tube... I keep going on like a broken record about how at these power levels the Miracle Homeopathic Nano-Quantum Astrology Paste doesn't perform any better, but nobody wants to hear it.

It’s impossible to have a sane conversation on this since neither side has empirical data. Or has that changed?

EDIT: Added my emphasis in the quote in case it wasn’t clear what I was referring to.

You can kill a LED on a non-direct-thermal aluminum MCPCB with a single cell even if said MCPCB is bolted to a 16-ton piece of copper.

I'm tired of trying to convince you guys that the blue sky is really blue, and not green. Sure, whatever, Tinkerbell is real. Happy?

Which do you think would make me happier, data or Never Never Land? The best data we have so far is from tests like this one by pflexpro, which does not directly address this topic.

Yes, good thread, unfortunately Fujik vs. solder is not the kind of thing I am advocating, not at all. Why do my arguments always get twisted around to "ha! he thinks we should all be using lights made of Fujik, asbestos, and styrofoam, lol!"

Look at that and then explain how you get more light with excellent thermal paste under the star vs. average thermal paste under the star. The farther you get from the source of the heat the larger the contact area, which makes each joint less and less critical.

The substrate is pretty good stuff, but it will always be compromised since it also has to still be an electrical insulator.

But still, if there's no heat sag, how is a copper pill or better paste going to make more light come out? If the Tinkerbell theories are correct, why does a light with an aluminum pill and crappy paste not show any significant dropoff over time? The initial output is going to be the same no matter what since none of the downstream parts are saturated yet, right? So where in all this is the extra light output supposed to come from?