Cree R2 LED consistency

[MrNaz]

I've just finished a mod on a dive light, and have noticed significant differences in the LEDs.

They are are an array of 4x XR-E R2 WG LEDs, each powered by its own 350mA AMC7135 board. When I turn the supply voltage down to about 2.2V, the boards go into DD mode and the LEDs are on, but dimly. At this level, I can see one LED significantly dimmer than the others, one significantly brighter than the others and a different tint, and the other two looking like I've come to expect R2-WG emitters to look.

When I turn the voltage up to 4.8V, the boards go into regulation, and the brightness of the array overwhelms any differences in the LEDs, making it impossible to actually see any difference.

Details of the build can be seen in my build photos here:
http://www.candlepowerforums.com/vb/showthread.php?t=278939

Could the differences be accounted for by the behavior of the driver boards? I expect that they should all be the same, especially when they are running with Vbatt < Vf and thus in DD mode. Or are the differences more likely to be due to variances in the emitters as supplied by DX. Or, could it be that I've damaged the dim one with insufficient heatsinking?

Is there a way around this, other than to have a reference set of LEDs and test each emitter against it before inserting? Is it even important?

 

[Th232]

First thought for me would be that one of them has a low Vf, given that Cree doesn't bin by Vf at all. Are you able to check the voltage and current going into each LED?

 

[AnAppleSnail]

Is there a way around this, other than to have a reference set of LEDs and test each emitter against it before inserting? Is it even important?

A little here, a little there. You might pair the weakest driver with the lowest Vf LED.

 

[MrNaz]

First thought for me would be that one of them has a low Vf, given that Cree doesn't bin by Vf at all. Are you able to check the voltage and current going into each LED?

I could have during the build, but as you can see from the photos, it's not really possible without re-dismantling the assembly, which I can't be bothered doing. Especially given that it is not particularly important.

 

[gcbryan]

That's just the nature of the leds. If you look at the spec sheet you will see a range. You are experiencing that range.

A manufacturer or customer builder would have a test each led and sort them by vf and color.

I'm surprised that 2 XR-E's don't cause problems eventually during a dive with heat build-up. I understand that you have a metal plate to transfer some of the heat away from the emitter but since plastic is an insulator eventually that heat should fill up the housing and it has no where to go.

 

[MrNaz]

That's just the nature of the leds. If you look at the spec sheet you will see a range. You are experiencing that range.

A manufacturer or customer builder would have a test each led and sort them by vf and color.

Ah. Ok from now on I will bin the LEDs as I buy them, by Vf. I won't bother binning by tint, as I won't be buying from DX, but from Cutter, who are usually better with getting their LEDs in the right tint zone. Variance within tint bins doesn't bother me, but this difference was well outside the WG tint range which is why I asked.

I'm surprised that 2 XR-E's don't cause problems eventually during a dive with heat build-up. I understand that you have a metal plate to transfer some of the heat away from the emitter but since plastic is an insulator eventually that heat should fill up the housing and it has no where to go.

Well, it had 3x P4 LEDs in there before, and I had the head running on for a good 20 mintues and it stayed quite cool. You're right though, that's different from having it on and in a housing. I'll follow up when I do a dive and let you all know if they burn out. Also remember I'm only powering them at 350mA, and not 1.5A like in some other crazy mods.

I do have a build planned for the same emitter array powered at 1A per LED, but that'll be an all-metal housing with the emitters mounted directly to a surface that is a single piece with the outer housing. You can see that here:
http://www.candlepowerforums.com/vb/showthread.php?t=240193
I'm just in the process of making the internals a bit more durable. It is very, very bright.

 

[gcbryan]

Ah. Ok from now on I will bin the LEDs as I buy them, by Vf. I won't bother binning by tint, as I won't be buying from DX, but from Cutter, who are usually better with getting their LEDs in the right tint zone. Variance within tint bins doesn't bother me, but this difference was well outside the WG tint range which is why I asked.

Well, it had 3x P4 LEDs in there before, and I had the head running on for a good 20 mintues and it stayed quite cool. You're right though, that's different from having it on and in a housing. I'll follow up when I do a dive and let you all know if they burn out. Also remember I'm only powering them at 350mA, and not 1.5A like in some other crazy mods.

I do have a build planned for the same emitter array powered at 1A per LED, but that'll be an all-metal housing with the emitters mounted directly to a surface that is a single piece with the outer housing. You can see that here:
http://www.candlepowerforums.com/vb/showthread.php?t=240193
I'm just in the process of making the internals a bit more durable. It is very, very bright.

I checked out your other thread. Nice build. Are you saying that you haven't got that completed yet? I take it that it hasn't been on a dive?

Nice work however!

 

[MrNaz]

I check out your other thread. Nice build. Are you saying that you haven't got that completed yet? I take it that it hasn't been on a dive?

Nice work however!

Yes, that light was completed and working, but I wasn't satisfied that the internals were robust enough to take a drop onto a deck. Given that the outside of the dive light seems like it could survive a nuclear holocaust, it didn't make sense to have the internals remain fragile.

I haven't dived it yet, but I have tested the body in a pressure pot, and it held up to 10 bar, or 100m of pressure. :twothumbs The o-rings are all viton, and I've taken the tolerances off an aerospace chart. They in theory should take up to 100 bar, as that is the material and seal gaps that I've used. The weakest point on the torch is the polycarbonate lens, and I don't know how to calculate the theoretical pressure tolerance, but at 4mm thick with a diameter of under 60mm, it'll be high.

I'm just in the process of getting the final bits and pieces made by a plastics company, then the whole thing should be nice and solid. It should be ready for my next dive trip in early July. I'll keep you posted how it goes.

 

[gcbryan]

Yes, that light was completed and working, but I wasn't satisfied that the internals were robust enough to take a drop onto a deck. Given that the outside of the dive light seems like it could survive a nuclear holocaust, it didn't make sense to have the internals remain fragile.

I haven't dived it yet, but I have tested the body in a pressure pot, and it held up to 10 bar, or 100m of pressure. :twothumbs The o-rings are all viton, and I've taken the tolerances off an aerospace chart. They in theory should take up to 100 bar, as that is the material and seal gaps that I've used. The weakest point on the torch is the polycarbonate lens, and I don't know how to calculate the theoretical pressure tolerance, but at 4mm thick with a diameter of under 60mm, it'll be high.

I'm just in the process of getting the final bits and pieces made by a plastics company, then the whole thing should be nice and solid. It should be ready for my next dive trip in early July. I'll keep you posted how it goes.

Dynamic forces are different than static but it sounds (and looks) pretty robust. I look forward to the report!