Does thermal paste and/or pads visibly outgas?

LED Flashlights – General Info
1

Hi,

I made a “shorty” STL-V6 recently:

and since the STL-V6 is now direct drive, I’ve been trying to improve the heatsinking a bit.

The way that the FandyFire STL-V6 that I have is designed, there’s a brass pill, which screws into the reflector from the tube end, and then the reflector+pill screws into the head from the front end.

So, the front end of the pill, where the emitter is, screws into the back end of the reflector, which screws into the front end of the head.

Thus, the thermal path from the XP-G2 emitter is through the star (just a regular star, from IS), which is glued using Arctic Alumina to another larger 25mm star (which I sanded the face - long story), to the pill, to the reflector, and finally to the head.

I initially just put some Arctic Silver 5 on the pill threads and on the reflector threads, but I didn’t think that it was good enough (the head wasn’t getting very warm), so I ran out of AS5, so I used some thermal grease I got from FastTech awhile ago:

https://www.fasttech.com/p/1148100

I also took a narrow strip-shaped Fujipoly thermal pad, and shoved it around the edge of the emitter PCBs on the front of the pill. Since the pill screws into back end of the reflector, my theory was that that would increase contact from the emitter PCB to the reflector.

This actually seemed to have worked, as the fins around the head now get pretty warm, almost hot, pretty quickly now.

So then, I was doing a timed test, to get an idea of how hot that would get, and I had the light tailstanding on my desk. When I did that, after awhile, I noticed that when I looked at light, the area inside the front looked a little “bluish”.

I quickly turned the light off, because I thought that I was seeing that bluish color because the emitter might be getting killed, then I waited awhile and then turned the light on, and it was fine, no more bluishness.

I did that a couple of times, and so far the same thing happens each time, and the last time, I noticed that in addition to the bluishness, it was kind of “foggy”.

However, when I turn the light off, the lens is still clear.

So, I’m wondering what is going on, and also wondering if maybe, as the light heats up, either that thermal paste or the thermal pad is “outgassing”, and since the reflector is in place, whatever it’s outgassing is remaining inside the head, and eventually becomes visible in the form of the “blue fogginess”.

I’ve googled and it seems like it might explain what might be happening, e.g.:

and:

Has anyone seen something like that with a flashlight where you added thermal paste or pads, and also, is it a problem long-term? I’m wondering if that is what’s happening, it might deposit a film on the inside of the lens?

Thanks,
Jim

2

Maybe if you leave the light on for a long time, with no lens, it will clear all the gas?

3

I don't even have to read the whole thing, only got about halfway through, so I'll just go ahead and tell you: XPG2 + aluminum MCPCB + direct drive = smoked LED (that's where your smoke came from - the dome of the LED overheated).

Copper star, or put an actual driver in it. Or both, preferably.

Run it with the LED out in the open, lens and/or reflector removed, and watch the little curly wisps of smoke rise off the dome. Then tear it apart and come up with a new plan.

4

comfy and chloe,

Since my OP, I’ve been doing that, running the light without the lens for up to 30 minutes or so at-a-time, and no smoke and no bluish fogginess.

With all of the “stuff” that I’ve “shoved” down the throat (at appropriate places, I think), the finned area (which is immediately around the pill), the head, and even the now short body get really hot, but the emitter, amazingly, seems to be surviving ok thus far.

I’m not recommending any of this, but just reporting information :).

I’ve been doing some research/googling, and apparently most thermal compounds and pads, including Arctic Silver 5, are silicone-based, and they all apparently outgas, so I’m thinking that whatever is being outgassed is being trapped inside the reflector-to-lens area, as Chloe suspected, so I’ve just order some non-silicone thermal compound and will try to replace at least the compound, when that gets in.

comfy,

I would’ve liked to have gone to an XP-G2 on Noctigon in the first place, but the way that this light evolved, time-wise, I had killed it earlier, and I didn’t have one available. That’s why I ended up with what I have now, and am just trying to make the best out of it with what I have.

If I eventually smoke the XP-G2, I’ll get an XP-G2+Noctigon and replace the one that’s in the light, but I’m holding on as long as it works; because of the “evolution”, the emitter PCB is now Arctic Aluminaed onto the original PCB+pill, so I really am not looking forward to having to get all of that off :)…

I appreciate the suggestion though :)…

Later,
Jim

5

We’re off to dinner, but I’ll try a more extended test later tonight.

Jim

6

the temperature of led in the reflector with glass lens is significantly higher than when lens removed. if you continue doing in closed compartment, I suppose the xml will die soon…and if there’s lube grease near the emitter thread, it might get smoked too…

7

Just for fun, since it's doomed anyway, try switching it on for about 5 minutes, then tap the head on the table and watch the LED slide off. Hot enough to make the dome smoke is above the melting point of the solder.

8

In my earlier post, I said that I hadn’t seen any smoke when running the light for up to 30 minutes?

Anyway, what you suggested was any interesting experiment :), so I just tried that, but the dome’s still intact after the “tap the head on the table” after the light had been on for 5 minutes.

9

With the reflector & lens in place? Same conditions as when it made smoke before..

10

soratantaz,

I suspect that you may be right that temperature might be higher with the lens in place.

Also, FYI, the emitter’s not an XM-L (or XM-L2). Per my earlier posts, it’s an XP-G2.

In most cases, with my “normal usage pattern”, I wouldn’t be having the light on for more than a minute or so. I was only doing the longer runs because I was curious about what I reported in the OP.

Thanks,
Jim

11

Hi,

I’m running another test, with the lens on this time. I probably won’t do it for 30 minutes this time, but more like 5 minutes, to see if the “fog” returns, i.e., to test what Chloe suggested earlier.

I’m hoping those longer runs earlier, with the lens off, caused all of whatever was going to get outgassed to outgas…

Jim

Edit: Finished the 5-minute test with the lens on, and I think that what I was seeing earlier was much better. It’s hard to tell for sure, since I’m doing this in a lighted room, but I didn’t notice what I saw earlier.

So, Chloe, maybe your “theory” was correct, that what I seeing earlier was visible because I had the lens on, and now it’s all “burnt off”/evaporated?

Also, I did notice something that I hadn’t noticed before. On the side of the reflector, near the bottom edge (where the emitter hole is), there are two small areas where there’s kind of “smoky shadow”, almost like a burst-type pattern in both cases. Like I said, I hadn’t noticed that earlier, because I was kind of avoiding looking directly at the reflector when the light was on, and even this time, I only saw it when looking kind of at an angle from the “other” side of the light from where those “shadow” areas are. It’s kind of hard to explain in words, and I’ll try to see if I can get a picture (if they show up in a pic).

I suspect (another theory again) that some of the thermal compound or pad got deposited onto those areas of the reflector, but I’m not sure whether that was from one of the earlier times, i.e., when I tested the light right after applying the thermal compound and pad. Those areas are not really visible when the light is off though, so I’ll have to see if they show up in a pic when the light is on…

Jim

Edit: This is about the best I can do. The 2 areas I mentioned are circled in yellow:

12

Direct drive XPG2 on aluminum won't survive 3x D cell NiMh but does fine with no change other than a copper MCPCB. The center thermal pad is so tiny, it's like 1/3 the area of an XML.

What's the tailcap current on this thing?

13

The last time I measured, tailcap current was ~2.2 amps:

It may not be clear from just that post, so FYI, the emitter+PCB from IS that I’m using is sitting on top of the original PCB which I lapped before Arctic Alumina’ing the IS emitter+PCB… because I didn’t have an ~25mm emitter+PCB (with XP-G2 or any other emitter).

BTW, this thread is getting way off-topic I think: So has anyone seen something like what I originally described? I started this thread mainly because I was wondering about the outgassing thing, and based on what I’ve seen thus far, that seems like a likely explanation… Also, I think that I mentioned that I ordered some non-silicone thermal compound today and will be replacing what I have with that.

Jim

14

Not off topic at all because yes, I have seen the exact thing you described, and it was always from an emitter being driven too hard and smoke coming off the dome and making the lens look foggy, and the fog was only visible when the light was on, it was nearly invisible when switched off. I have never seen any kind of silicone-based thermal compound do anything similar, and that's all I use when I can't use solder or epoxy.

15

It’s the solvents used to form a paste of solids in thermal compounds that are evaporating. AS5 claims it will thicken and stabilize as it cures which can take up to 200 hours of heat cycling. The thin amount you would normally use wouldn’t create the effect you’re seeing. In general, the higher viscosity compounds use less solvent but are harder to spread.

16

lightme,

Thanks. I think that I saw similar type information as I’ve been researching this. But, also, remember per the OP that I didn’t just use Arctic Silver 5. I ran out of that kind of in the middle of things, so I also used that compound I bought from FT, and I also used some Fujipoly thermal pad material (from FrozenCPU), so, at this point, I guess that any of those potentially could have potentially caused what I saw.

I’m in the middle of another run test, with the lens on this time.

Currently at ~43:26+, and the light is still going strong (using a KK 26650), and no sign of the “blue fog” that I saw earlier, and I’ve tapped the light onto a granite countertop several times (not really hard though) during the run.

I’m going to try to let this one run until the battery runs out or the light dies or I need to go to bed… probably the last one, because it seems to be ok so far :)…

Thanks again,
Jim

17

Ran > 50 minutes, and no “blue fog”. Light head, fins, and body very hot when I shut it down. Tapped against countertop, dome still ok.

Some boring pics (I think 2nd to the last was w/no flash - others with flash):

18

I have DD several XP-G2’s on aluminum. I scratch the backs of the star and the base of the pill and use a very thin even coat of thermal glue. The only time I have fried one was when I tried to use 20 gauge wire. That toasted it. When I run standard thin wire I have not had a problem. I have even done it with a P60 drop in and made sure the battery was topped off real good. The light itself was very hot to the touch after a few nice long walks and it never skipped a beat. Oh and that one is de-domed too.

19

Cool light ohaya.

20

I missed this thread! Thanks for the pics. It’s interesting to see this outgassing. I wonder how manufacturers get around this during assembly stage, to avoid potentially worried customers.