Luminus SST-20-W 4000 K CRI95 color and output test

I think I have lights with 3000K 3500K and 4000K. I could try mixing them by just shining on a wall but problem is they are different output and reflector/tir.

If I recall the 3500K looks fairly good on its own (from Mouser / BlueSword) so if you haven’t purchased emitters yet I would suggest trying that over buying 2 emitters you know you dont like.

Shaving the dome off with a razor (not a very clean job), the tint got noticeably better.

Any beamshots? I’m tempted to do some slicing and dicing.

Also is there somewhere I can read up on how to interpret these charts? The only things I know are the numbers in the top right :blush:

Could you please append here a non-sacrificial output test of the razor-dedomed emitter, maukka? It would provide nice information regarding the potential output loss versus domed emitter.

I second this, if it’s not too much effort. My triple had a better tint after shaving but it took a big output loss. That’s the thing with this LED, it’s throwy without dedoming so shaving the LED just to fix the tint is not a great trade off.

I lose about 30% output slicing the dome at the bond wire height, with no increase in throw but a noticeably rosier tint. Increasing the intensity requires trimming the dome around the die and carefully slicing as close as possible to the phosphor while avoiding the bond wires. The stock 3500K SST-20s I’m using get about 600m throw in a D1S/GT mini, which I was able to increase to 685m with careful trimming. It’s not pretty, but it works, and it could probably be increased further. This is the best option for a high CRI thrower right now.

30%? Ouch. I guess it will be better using a minus green filter to correct tint as there will be less risk? Do they decrease throw/intensity at all?

What is it about slicing that makes the tint rosier?

Do different depths of slicing have different effects, i.e. do deep slices, as close to the phosphor as possible, result in the rosiest tint, with the best increase in throw but at the expense of the biggest sacrifice in output and vice versa?

How about Luxeon Z ES? It’s just CRI90 but maybe it will throw significantly better?

I just shaved 12 SST-20 domes in 3 D4’s. Been reading how they got rosier, had to find out. Started with the one I cared the least about just to experiment and ended up slicing them all. I thought my 4000K version was a wee bit green and my 3000K a bit yellow. Now I have white and rosy. The beam profile is maybe a little more wonky but only if you are looking for it on a white wall. I think these are pretty sweet little LEDs for what they are.

Any comparison beam shots or tips on how to slice?

Yeah when I get off work. Beam shots suck on my phone, but I had a 3000K 4000K mixed D4 and when I shaved them it made my straight 3000K D4 look pee yellow when in reality it’s not. Then I did the 3000K D4 and it’s a much nicer rosy color than yellow now. Took any hint of green at lower levels out of my 4000KD4. No I don’t have too many D4’s.

Hodor, my phone doesn’t really show the difference in tint well, but take more word that it improves. I didn’t get too excited with the preciseness of the slicing. Just got pretty close to the phosphor opposite the bond wires and when I got to the bond wires just angled up. So a side view looked like a little ramp (half) half pipe. Didn’t seem to be too crucial to get it as close to phosphor as possible.

I can attest to the tint improving but the output does go down and since the beam narrows even more it makes the beam pattern look different. You see a starburst type pattern (at least in a carclo triple) more clearly at the edge of the hot spot.

I honestly wouldn’t recommend it unless the green makes it unusable to you or you dont mind replacing the LEDs if you dont like it.

Good input contactcr. That is all true, but for me, preferable. I just went out in the snow to test against my 219C D4. I quite like the extra throw. A little bit more like a reflector light. I don’t have a lumen tube but I’d say the loss is not nearly as perceptable as the tighter beam profile. So I’m a vote for just do it. Depends what you like hodor.

I took some measurements of my dedomed SST-20 GT mini. I’ll add data for the stock LED when I have a chance to install one in a light. It looks pretty nice on turbo, though it’s really too warm for my tastes. But, it’s a rosy high CRI laser thrower, so I suppose I can suffer through it :smiley:

CCT = 3237K (Duv 0.0017)
Color Rendering Index (Ra) = 98.1 [ R9 = 93.0 ]

CCT = 3349K (Duv –0.0023)
Color Rendering Index (Ra) = 93.5 [ R9 = 71.8 ]

Nice, what meter do you use?

That’s measured using an X-Rite i1Studio and spotread.

TA sent me some more SST-20s. Tint bins FA3 and FD2. The FA3 turned out to be pretty good at the BBL already at 1A. Maybe he can guide us on where to buy them?

:+1: it looks like progress in getting what we like. Thanks TA and maukka for the effort.

Indeed. I so wish it will be generally available in the near future….It negates the biggest issue with this otherwise-great LED.