sst from Luminous- not up to specs? Bad quality control?

[ergotelis]

Hi, has anyone in here test a good bunch of this leds? I have bough two sst50, from different reputable sellers but both don't seem to be up to specs. Basically they seem to have an output of 10-20% less than a good cree MC-E. I guess they were supposed to have the same output or better. What do you think?
My tests were done under the appropriate heasinking, the same driver with measured Vin/Vout and Iin and Iout.I used a lux meter to test the output and a diffuser on the leds.Anyway the result is visible via my own eyes.

 

[Th232]

What flux bins did you buy?

Also, it looks like the best MC-E is more efficient than the best SST-50. Going off the datasheets a top bin SST-50 at 2.8 A should have around 560-680 lm, while a top binned MC-E with all four dice in parallel and run at 2.8A (700 mA per die) should be 730-830 lm. That said, I don't know what the highest available flux bins are.

The difference is that the SST-50 can be driven at much higher currents.

 

[Optical Inferno]

I used a lux meter to test the output and a diffuser on the leds.Anyway the result is visible via my own eyes.

Did you have a setup where the diffuser and luxmeter were at the same distance from each LED?? Also, what kind of diffuser?

The only other thing I can think of is that the LEDs are different (even slightly) in colour temperature. This may result in the lux meter's photosensor to be slightly more sensative to one LED resulting in a higher output. Is your lux meter sensor corrected or does it follow the photopic curve?? I have the same problem with slightly different binned rebel LEDs.

 

[ergotelis]

Did you have a setup where the diffuser and luxmeter were at the same distance from each LED?? Also, what kind of diffuser?

The only other thing I can think of is that the LEDs are different (even slightly) in colour temperature. This may result in the lux meter's photosensor to be slightly more sensative to one LED resulting in a higher output. Is your lux meter sensor corrected or does it follow the photopic curve?? I have the same problem with slightly different binned rebel LEDs.

-Leds are top bin SST50.
-Distance is the same for all leds.
-Diffuser is one from jetbeam.
-Tint is almost the same between the leds.
The difference is visible, luxmeter comes to support what i am thinking of. I think that th232 is right. So i guess we have to wait for XM in order to see a really nice improvement in performance.

 

[LilKevin715]

I took a look at the SST-50 data sheet and it looks like the flux vs foward current (page 9, top graph) is a linear relationship. Doing some simple math I was able to generate a linear equation:

Flux Percentage (relative to 1.75A rating) = 53.85 X Amps + 6

The flux rating is relative to the rating at 1.75A, as Luminus rates their bins on page 3 of the spec sheet. A WJ bin is rated at 500-600 emitter lumens at 1.75A per their spec sheet. So if the drive current was 2.8A then the LED lumens would be:

53.85 X 2.8 + 6 = 156.78%
500 X 1.5678 = 783.9 emitter lumens (min)
600 X 1.5678 = 940.68 emitter lumens (max)

Take the min/max numbers times whatever percentage due to reflector/lens losses. This is debateable but generally speaking somewhere around 66-80% (.66 to .8) and you have OTF lumens.

 

[qwertyydude]

We're all still waiting for the xm-l. I think that will ultimately kill the sst-50 like the xr-e killed the old school luxeons.

 

[saabluster]

I took a look at the SST-50 data sheet and it looks like the flux vs foward current (page 9, top graph) is a linear relationship.

The only way that graph is correct is if Luminus got rid of droop which they most certainly have not. Therefore any attempts to use that faulty information as a base will lead in the wrong direction. Just say'n.

 

[UberLumens]

What Vf are you getting?

Maybe it needs a shock treatment of some heavy amps like some members have done with the 90.

 

[LilKevin715]

The Luminus data sheet accuracy is probably questionable, but its all that we have at the moment. JTR's white LED lumen testing thread hasn't included a SST-50 yet so I guess we will have to wait for a while for some independent results.

 

[flashfiend]

sst-50 are best direct driven with either a single li-ion or 3x nimh. Using a driver will prevent it from reaching its full potential unless you are using a driver that will give it 4-5 amps with an appropriate power source.

 

[ergotelis]

Direct draw is useless for me, you get full power in the beginning and after about 10-15 min nothing special.
The problem is the performance against other reputable leds such as mc-e and P7.