Long term effects of de-doming SST

[Techjunkie]

I'm not sure if the blackening that I'm seeing at the edges of the de-domed SST-50 is the LED die burning up or if it's the gel that was left behind being baked and dried like jerky. Maybe it's black mold!

Sorry for the crappy pic. I had to use my cell phone for this one. You get the idea.

This is the emitter from this torch, by the way.

 

[Techjunkie]

Also, here's the before and after dome/no-dome in the right column of this pic:

I don't know if my experience is typical, but I'd be interested in hearing other people's long term observations.

 

[ma_sha1]

I've only done two de-doom SST-50 on purpose.

The first one I did here at begining of July : http://www.candlepowerforums.com/vb/showthread.php?t=275711
Not comfortable with exposing the die to air, I scrap off the gels & then added a drop of Norland 61 over the die.
Then put under the Sun for about 1hr cure it.

It still looks like new But it's not that old. the die gets pretty hot, I hope the Norland can handle the heat long term.

I just did another one the same way, this time I added a bit more Norland & it looks like a very shallow doom, die image is still very small, but I haven't put it into a flashlight yet.

 

[saabluster]

I'm not sure if the blackening that I'm seeing at the edges of the de-domed SST-50 is the LED die burning up or if it's the gel that was left behind being baked and dried like jerky. Maybe it's black mold!
Sorry for the crappy pic. I had to use my cell phone for this one. You get the idea.

Hard to tell with the picture as bad as it is. I have had the phosphor burn like that if it is directly exposed to air. If you remove the dome just right though you will end up with a very smooth layer of gel left that protects the phosphor just fine.

I've only done two de-doom SST-50 on purpose.

The first one I did here at begining of July : http://www.candlepowerforums.com/vb/showthread.php?t=275711
Not comfortable with exposing the die to air, I scrap off the gels & then added a drop of Norland 61 over the die.
Then put under the Sun for about 1hr cure it.

It still looks like new But it's not that old. the die gets pretty hot, I hope the Norland can handle the heat long term.

I just did another one the same way, this time I added a bit more Norland & it looks like a very shallow doom, die image is still very small, but I haven't put it into a flashlight yet.

Norland is a very bad thing to be coating that die with. You now have essentially the same thing as the 5mm LEDs only much larger. Better to either leave the silicone gel on there or replace with another layer of silicone. There is a reason you don't see high power LEDs encapsulated with epoxy.

 

[ma_sha1]

Norland is a very bad thing to be coating that die with. You now have essentially the same thing as the 5mm LEDs only much larger. Better to either leave the silicone gel on there or replace with another layer of silicone. There is a reason you don't see high power LEDs encapsulated with epoxy.

One possible "bad thing" I am mostly worried was heat cracking it up or turn it yellow, pretty much the only thing I was afraid off. The spec sheet said after curing, it can tolerate heat up to 125C (257F). I am not sure what the top surface Temp would be? (Running at ~ 4Amp).

I know it's pretty hot but maybe not 257F? Plus, mfg. spec usually is conservative, as they don't want people sue them on some injuries due to Norland 61 cracked up within spec.

So far it hasn't cracked or yellowed, and my SST-50 throw in the Deep Rebel SMO went from 36K lux to the 50K lux range, what other "bad thing" could there be?
.
.

 

[bshanahan14rulz]

it will turn yellow with time. I have an XR-E that was inadvertently de-domed. I proceeded to cover it with a layer of NOA61, and it has since yellowed quite a bit. I'm more sad about my XP-E (only P3 bin, but 7C tint). Dome sheared off, taking half the 7C phosphor.

 

[saabluster]

One possible "bad thing" I am mostly worried was heat cracking it up or turn it yellow, pretty much the only thing I was afraid off. The spec sheet said after curing, it can tolerate heat up to 125C (257F). I am not sure what the top surface Temp would be? (Running at ~ 4Amp).

I know it's pretty hot but maybe not 257F? Plus, mfg. spec usually is conservative, as they don't want people sue them on some injuries due to Norland 61 cracked up within spec.

So far it hasn't cracked or yellowed, and my SST-50 throw in the Deep Rebel SMO went from 36K lux to the 50K lux range, what other "bad thing" could there be?
.
.

The yellowing is what I am talking about. It is just a matter of time. The maximum junction temps for those I think are about 150C and that shouldn't be too hard to get to if run hard and especially in a portable application. Even if you are under 125C a little bit it is a matter of time before it yellows. Once that happens it will create a sort of domino effect because the yellowed epoxy will not be as transmissive and some of the light will die as heat. This causes the epoxy to heat up even more which accelerates the degradation.

 

[ma_sha1]

The yellowing is what I am talking about. It is just a matter of time. The maximum junction temps for those I think are about 150C and that shouldn't be too hard to get to if run hard and especially in a portable application. Even if you are under 125C a little bit it is a matter of time before it yellows. Once that happens it will create a sort of domino effect because the yellowed epoxy will not be as transmissive and some of the light will die as heat. This causes the epoxy to heat up even more which accelerates the degradation.

Thanks, that make sense. Looks like I'll need to try something with higher Temp. resistance next time.

I remember seeing your thread that you were experimenting with alternative dooms for R2, did you get to a solution that won't yellow? Any link you cold recommend?

In one of the SST-90 de-doom exp. I did, I glued a tiny glass window over it, but ed the led as I was scraping gels away, never get to try & see it in a light. That might be a way to go especially there are tiny glass windows available for watches.

Alternatively, something like this might also work?
http://webcache.googleusercontent.com/search?q=cache:qHCKqZWZXhwJ:www.topsecretcoatings.com/auto-ts-100_1-part_epoxy-clear.htm+heat+resistant+%22clear+epoxy%22&cd=2&hl=en&ct=clnk&gl=us
.
.

 

[saabluster]

Thanks, that make sense. Looks like I'll need to try something with higher Temp. resistance next time.

I remember seeing your thread that you were experimenting with alternative dooms for R2, did you get to a solution that won't yellow? Any link you cold recommend?

In one of the SST-90 de-doom exp. I did, I glued a tiny glass window over it, but ed the led as I was scraping gels away, never get to try & see it in a light. That might be a way to go especially there are tiny glass windows available for watches.

Alternatively, something like this might also work?
http://webcache.googleusercontent.c...istant+"clear+epoxy"&cd=2&hl=en&ct=clnk&gl=us
.
.

Although that sounds like some cool stuff you do not want anything epoxy at all. I used the same stuff that an LED manufacturer would use. Just google for a silicone used for encapsulating LEDs. That is the only thing you want. Nothing else will do.

 

[Techjunkie]

I forgot that I had started this thread until I revisited my photobucket account tonight and saw the cell phone pic. Here are some (slightly) better images taken with out flash and only a little camera shake:

What's the verdict, cooked phospor?

 

[jgraham15]

To me it looks like cooked phosphor but I am no expert! Though I have cooked a few led's The dome fell off of my SST-90 a while back and I noticed it seems brighter and has a more neutral tint. I use it all the time in my bored 6P with an IMR 18650 and haven't had a problem. Maybe I got lucky and it still has a thin layer of gel on it???

 

[saabluster]

What's the verdict, cooked phospor?

Looks that way. Did you scrape the silicone from off the top of the LED? I have never seen it do that unless the phosphor was exposed to air.

 

[ma_sha1]

I put in a sample request for Dow Corning's Silicone used for LED encap.
They only sell by the bucket loads. I can't find anywhere where I could buy small qty of any silicon used for led encap.

Their sales rep. called me to "qualify" the application.
I couldn't lie to him so I coughed up the application is for SST/flashlight & for myself rather than for my company.

I am pretty sure I failed the qualification & won't get any samples

 

[saabluster]

I put in a sample request for Dow Corning's Silicone used for LED encap.
They only sell by the bucket loads. I can't find anywhere where I could buy small qty of any silicon used for led encap.

Their sales rep. called me to "qualify" the application.
I couldn't lie to him so I coughed up the application is for SST/flashlight & for myself rather than for my company.

I am pretty sure I failed the qualification & won't ant the samples

Yeah there certainly are some benefits to having a flashlight company. To really make it work well anyway you'd need to vacuum degas it.

 

[Techjunkie]

Looks that way. Did you scrape the silicone from off the top of the LED? I have never seen it do that unless the phosphor was exposed to air.

Nope. The dome popped off nice and clean. It only left a small chunk of gel in the upper left corner. I tried to trim it there a little, but left some from fear of tearing a bond wire.

The dome "incident" did involve some isoproply alcohol that I was using to remove epoxy from the dome. Maybe some of that worked its way into the phosphor through the gel.

The good news is that it still works and still looks ~80% as bright as day 1.

 

[Techjunkie]

Re: Long term effects of de-doming SST (higher Vf?)

Tonight I used a hot soldering iron to soften and scrape away the glow power infused epoxy from the star of the de-domed emitter. (Did you know that glow powder lights up when a hot soldering iron hits it?) In the process, the metal dome base also popped off. After that, I was able to remove the star from the torch and clean it off. I didn't think to grab a good photo of the star while I had it out, because despite all this abuse, I was actually able to salvage the SST and performed a star swap between the 5A regulated torch it was in and a small direct drive torch that had an in-tact SST-50 in it.

Here's a close-up of the de-domed emitter installed in the smaller torch and a few observations.

Close-up of de-domed SST-50 and what appears to be cooked phosphor

Now that the metal dome base is removed, it can be installed and focused perfectly in an UltraFire C8-A1:

Here's two quick and dirty beamshots taken with an automatic digi-cam in an unfinished basement. The pics don't do the perfectly round hotspot justice.
Against cardboard:

Against black sealed cement:

Now here's an interesting observation - the de-domed emitter in the direct drive torch (1*IMR18650) draws significantly less current than the in-tact emitter did. On the same battery that the in-tact emitter would draw ~5A from when the tailswitch was bypassed to measure current, this damaged emitter only draws ~3A.

I assume the reduced current draw is because the damaged parts of the emitter surface effectively reduced its working size or surface area. In effect, this has become an SST-30.

I'd be curious to know if de-doming alone raises the Vf of the emitter. I'm inclined to think not. It's more likely that the Vf of my de-domed emitter is higher becuase less of mine is in operation due to the damaged areas.

I like the beam of it in this new host so much that I'm considering leaving it that way. (I had intended to drop a CREE XM-L in that one as soon as they come out.)

 

[CKOD]

Re: Long term effects of de-doming SST (higher Vf?)

If you use a multimeter on diode check setting, can you light up the die and see what the burned portions do? or are the leads inaccessible?

 

[Techjunkie]

Re: Long term effects of de-doming SST (higher Vf?)

If you use a multimeter on diode check setting, can you light up the die and see what the burned portions do? or are the leads inaccessible?

My DMM doesn't have a diode check setting (that I'm aware of), but I usually use the tone/circuit-test setting to check LEDs. Unfortunately, it doesn't tone-test with enough juice to light up a Phatlight LED, it just returnes some impedance number. (When I really need to test a Phatlight, I use a DC wall wart supply that only puts out a few hundred mA, but that makes it still too bright to stare into directly without seeing spots afterward.)

In lieu of a really low level (few mA), stare directly into it, I took a picture of it light up at full blast, point blank into my camera phone. The result below appears as if at 3A, the entire square is lighting up uniformly:

It's not the same as using a welders mask, but I know this method works. I've used it on damaged MC-Es in the past and when one or two of the LED dies was damaged and only a small part of one of the four MCE dies was lighting up it was obvious in the photo. Here's kind of a crappy example:

 

[Techjunkie]

Re: Long term effects of de-doming SST (higher Vf?)

Re: Long term effects of de-doming SST (higher Vf?)
Written by Techjunkie on 01-17-2011 01:02 PM GMT

I thought I may have finally answered the question of whether or not those damaged spots do anything. I recycled this damaged, de-domed SST-50 in a flood to throw lighthere, and it appeared to me in the "throw" beamshots that the damaged parts are not lighting up at all. Here's a pic:

I've just now realized looking back at some old pics above, that there were more carbon marks at the perimiter of the emitter then there are now. In more recent photos, only two corners contain those "burnt" spots, whereas above, they were all around the edge. It must be that what is left is just something sitting on top of the phosphor or the dome adhesive and not the phosphor itself. Dare I try to remove what's left?...

...I did dare. A little work with a toothpick dipped in isopropyl and viola! No more carbon marks on the LED and the whole die shown lit in the "throw" focus mode:

-----------------------------------------------------------------------------------------------------------

Re: Long term effects of de-doming SST
Written by Mettee on 01-17-2011 01:17 PM GMT

I have gently rubbed the silicon off that is left over with a q-tip, try that.

 

[Mike Painter]

I'm guessing that the cost of putting the dome on in the first place is more than justified for some reason. While not as dramatic or fast as removing the glass part of a light bulb I would guess that the short term results will be the same.

I wonder how long before "light bulb" joins the ranks of "records" and "dialing" a phone?