Think I fired SST-90 w/ 3 D-cell CTA's on DD

I fired up my newly build 3D Mag running CTA D-cell batteries and everything was fine for the first ~5 seconds. Then the light started turning blue and the emitter went dark.

Before anyone asks, yes it was installed on a heatsink AND the heatsink was installed in the flashlight. And yes, I used thermal adhesive to install the emitter. This isn't my first build just my first DD build.

I've read about plenty of people running the SST-90 direct drive using everything from IMR cells, emoli, NiMh, etc. But I have yet to see anyone fry one of these emitters. Could it be that these CTA cells did not experience voltage sag when powering this emitter? The initial voltage was 4.24 V. Did I get a bad emitter? Is there anything that can be done to fix it, can't image what that would be?


POSTMORTUM:
I did not get good thermal contact between the emitter and the heatsink. When using the Britelumens heatsink, IF your wires are not exactly in the right location and oriented exactly the right direction it will make installing the emitter more difficult. In this case I THOUGHT the emitter was pressed fully into the pocket on the heatsink. As it turns out, I did not get the emitter pressed COMPLETELY down into the recess and make COMPLETE contact with the thermal epoxy. Cause of death: thermal overload.

-The emitter is not working any more?
-Did you try a multimeter to see amp draw for the led?
Making it all fine as you say, makes it impossible to have such a bad result. It might be a bad emitter though a bit hard to believe it. And that started 5sec after you turned it on. What thermal grease did you use? What is the host and the heatsink?
Are you sure it contacts well to the heatsink?

jwl said:

Then the light started turning blue and the emitter went dark.

Before anyone asks, yes it was installed on a heatsink AND the heatsink was installed in the flashlight. And yes, I used thermal adhesive to install the emitter. This isn't my first build just my first DD build.

Click to expand...

if the emitter gets blue then the thermal path was broken/not good enough, but i still think you know this...

markus

Sadly angry blue is the last you'll see from that emitter. (I can't say where on the death line blue comes in, and how reversable it is if immediate action is taken, but sure as weevils eat lumens, blue is overheating and followed by inoperability).
Sad about your lost emitter. A good analysis is essential before you tear it apart to correct this for the next one. Interestingly, the time-lines you give are very similar to an unmounted emitters on-to-fried. This wasn't the body being overwhelmed with heat, or the heatsink saturating and not transfering to the body: ~5s is pretty much all in the emitter. Now given that report installing it correctly, thermal should not be the cause of death. So voltage? But this is suspect as well as many have run the -50 and the -90 from more. IIRC a strong (+10A build) was from 4xD nihm.
*confirm the thermal path out of the emitter, if no possible cause is found there perhaps it a QC issue?

**pics would be awesome. In Cree's and Souels there is some indication of deforminity with-in the lens, the high heat leaves evidence with a deformed phosphor layer or irregularities in the columns. Rotate the emitter to adjust your viewing angle and you may see physical deformities.

Well, I pryed the emitter off the heatsink and what do you know, the thermal epoxy hadn't made contact across the entire back of the emitter. I am using a britelumens SST-90 heatsink and Artic Alumina adhesive. The problem occured when I pushed the emitter down onto/into the heatsink. I did not push it down far enough, the adhesive had only contacted the back of the emitter where the adhesive what thickest. Lesson learned.

Test fit to see how far the emitter should sit into the heatsink BEFORE soldering and applying adhesive, that way you know how deep it should fit into the heatsink. With the britelumens heatsink it is VERY diffcult to see if the emitter is all the way down onto the contact pad in the center of the heatsink.

Like you guys said, the thermal path was not good enough. I knew better and didn't push the off button fast enough.

Linger said:

Sadly angry blue is the last you'll see from that emitter. (I can't say where on the death line blue comes in, and how reversable it is if immediate action is taken, but sure as weevils eat lumens, blue is overheating and followed by inoperability).
Sad about your lost emitter. A good analysis is essential before you tear it apart to correct this for the next one. Interestingly, the time-lines you give are very similar to an unmounted emitters on-to-fried. This wasn't the body being overwhelmed with heat, or the heatsink saturating and not transfering to the body: ~5s is pretty much all in the emitter. Now given that report installing it correctly, thermal should not be the cause of death. So voltage? But this is suspect as well as many have run the -50 and the -90 from more. IIRC a strong (+10A build) was from 4xD nihm.
*confirm the thermal path out of the emitter, if no possible cause is found there perhaps it a QC issue?

Click to expand...

Yeah, I didn't think voltage was the problem either. As you guys have pointed out it was indeed poor thermal contact with the heatsink.

Sorry to hear that...I understand you, unfortunately this was a really expensive lesson for you...
I was almost to destroy mine sst90 too , but my experience kept it alive. I saw it turning to a blue hue and stopped immediately the power.

Anyway, i have destroyed many other emitters!

..same happend to me with my first SST-90 mod, built in into a stock light from DX, 5 seconds everything good, then emitter turned blue very fast and then darkness... In my case I grilled the stock switch, not able to handle the high amperage- with new modified switch emitter worked -but turned blue after about 10 seconds because of selfmade heatsinking- but turned slowly blue and got darker- and after about 20 seconds I knew it was time to turn the light off- and I did this for some times and the emitter never died... In an other light I used wrong wires- 1,5mm ordinary speaker wire- it began to smoke after 20 seconds and emitter turned slowly blue- then I used 20awg silver coated teflon wire and problem was solved... May this can help....

...Wow, I was too slow- wrote this after the third reply---

I thought i would let everyone know that I just received my 4500k sst-90 from avnet. I checked the site and it now says they have ~300 in stock. I test lit it without a heatsink and I think I'm going to like the color, though it was hard to tell in just a second or two that I had it on. I really like the cree 5a tint so that is what I was comparing it to. It's not the same but close. I'll post pics when I get it built.