SST 50 mag turning blue

I finished building my SST-50 mag direct drive last night. I threw 3 nimh D cells in it and fired it up, after a few seconds it starts turning angry blue, scary blue...

I mounted it to the derwitchel heatsink using artic alumina epoxy... But the only thing I can come up with is that it didn't sink correctly...

Any input on this?

The first time I read the title I thought you meant the host was turning blue - that's one hot LED haha.

The LED light itself is turning blue? Seems a bit odd and not anything I've seen mentioned on these forums before.

Maybe you got an SST-50 that has quite a low Vf. Have you measured the voltage and/or current going across it in use?

if the cells were fully charged they could have more than 1.2 volts. maybe 1.3 or 1.4. that could driver your led way past 5 amps and be overheating it.
ctrl-f for: Sigbjoern's AccuPower D 11500 mAh cells.
http://www.candlepowerforums.com/vb/showthread.php?79302-NiMh-Battery-Shoot-Out
you can see that for around 5 to 15 min at the stat a cell will have a voltage over 1.25 which will overdriver your led. afteer they have lost about 1 amp hour they shoudl provide 5 amps at 1.25 or less volts.

but it is possible you got a lower Vf as others have said and your kinda screwed.

maybe try introducing some small resistance/resistor into the light to avoid that over driveing

Bad thermal path or to much current. Did the outside of the mag light heat up at all? If you dont feel some heat this is a good sign you have a bad thermal path. I would use a DMM and measure the current draw at the tail of the light. This should tell you if the currents to high.

Thanks for all the replies... I measured the current at the tail (Walmart meter with long thin wires for leads) first time was something like 2.1 amps.... Then I got the bright idea the cells may need topping off, so i threw em in the charger. When they were done I let them sit for about an hour and then three them back in the light... After about 7 - 8 seconds it started turning blue.. so no, the light isn't heating up, it doesn't have time...

I checked the amps again on the fresh charge and it was 3.5 (ish)... So maybe with the fresh charge and the resistance of the meter I was over 5amps... ( it didn't turn blue with the meter in the loop, ((during testing) but when not testing it did almost immediately which makes me think its not the thermal path)

And I guess i could tell ya that i disassembled and deoxit'd the switch, and used 22 gauge teflon coated wire about six inches long for the build... And that the cells I am using are old, and don't seem to hold up as long as they used to (in my p7 build 3d mag)..

did you put thermal paste below the socket?

How much pressure did you apply while the epoxy was setting?

MikeAusC said:

How much pressure did you apply while the epoxy was setting?

Click to expand...

Good point, was about to ask the same thing. You will need some pressure to ensure a thin layer of epoxy!

Der Wichtel said:

did you put thermal paste below the socket?

Click to expand...

Below the led, yes I used the artic alumina to secure it, I did not unscrew the plate and apply paste there, I do not believe it has the time to heat the pedestal...


I'm afraid I did not apply enough pressure while it was curing...but that doesn't make a lot sense. I did the same thing with several p7 builds, no trouble out of them.. I hate to trash the led taking it off...

And remember, this did not occur til after the cells had been charged...it was fine prior to that, but it wasn't as bright as my P7... Which is why I charged the batteries...

you also have to put thermal paste below the socket. I think that should solve your problem

Google of previous threads "LED turning Blue"

Norm

Well it seems heat is defintely the problem, I pulled the socket out and turned it on, and very quickly it got to hot to hold, within 5 seconds or so...., so I applied the thermal paste and reseated the socket and turned it on, it still turned blue in about 10 seconds, and then the led fell off the pedestal and the solder joints melted... me sighs, Im guessing it was the thermal path from the led to the pedestal, so I will be cleaning it and attempting to stick it back on top.. now I get to figure out how hard artic alumina really is... think I will use a clamp to hold it down this time..

I did a SST-50 build with DW's heatsink (see link in my sig). When I mounted the emitter to the pedestal I used tweezers to hold the emitter down for about 3-5 minutes (didn't touch the dome); enough time for the arctic alumina to dry. If you look at the pics in my build you can see a small amount of arctic alumina oozing out at the sides. This is a good sign as it indicates a very thin layer of epoxy is between the heatsink and emitter. When you do reapply the arctic alumina make sure to use the absolute thinnest layer as possible.

i had the same problem with my V10R Ti, and neutral white xp-g R5s. tried 3 different LEDs, and no luck. they all "tint-shifted" like crazy. cool white leds worked fine, so i put a High CRI xpg in it and it works fine.
my suggestion is to try a couple different leds.

Drmalenko, when measureing high current draw lights you need to replace your leads with some thick wire to get a more accurate reading like member moviles is using in this thread pic #3.

moderator007 said:

Drmalenko, when measureing high current draw lights you need to replace your leads with some thick wire to get a more accurate reading like member moviles is using in this thread pic #3.

Click to expand...

Yes, I know I should. The sad part is I have a clamp meter lying there I couldve used to get a reading at the led leads but forgot about it...

Good news is i got it working, but dedomed it in the process... LOL, just had shitty luck this week.. I will check the amp draw with my clamp meter tonight and see what this thing is pullin... Are there any metal reflectors that fit these things?

drmalenko said:

Are there any metal reflectors that fit these things?

Click to expand...

They are several places that carry these. You will have to look around to see which one you want. Kaidomain sku S004023 is one example. Or sku S004024. DX also has similar reflectors. You can even use a XR-E reflector but you may have to open up the hole for the sst-50. I prefer the mag rebel reflector for my builds. Comes in the led mag.

Ok.. so I put fresh batteries in the light, hooked my clamp meter up to the lead just before the led and at here are the results.. as soon as i hit the switch it read 5.8a, then immediately went to 7.4a, then again jumped to 10.??a, and then to 13.?? and was turning very angry blue so I shut it down, this took about 3-4 seconds.. any input?

With the same batteries, I used a different meter, and measuring at the tail cap, (using 8 guage wire for leads) measured 5.??a without it turning blue.. So Im guessing the resistance is what I needed (from the leads and the meter)..

I used the same batteries in my P7 (also a 3d mag build built by me), measuring at the tailcap and got 5.7a with no color shift...

So do I need resistance or is my thermal path not sufficent? And will I get the same results with a SST-90 as I know it can handle 9a?

(on a side note these were brand new batteries, ((new stock, not sure how old, but with a fresh charge)) not the same batteries I was using.. and they are Tenergy D 10,000mah)