What's so hot about the SST-50?

[Fallingwater]

I'm seeing plenty of lights with this emitter, but I can't see why it's preferred over the MC-E or P7, since it seems to have a slightly lower efficiency, and SST-50 lights tend to cost more than MC-E or P7 ones.
Is it a matter of die size, i.e. the SST-50 is easier to configure as a thrower?

 

[Th232]

From what I can see:

* Since the SST-50 is a single die, there won't be any dark cross/donut hole like in some multi-die lights.
* An SST-50 can go up to 5A, at which point the output is way in excess of anything an MC-E or a P7 can put out. Note, however, that around the 2.8A level it'll be, as you said, beaten by an MC-E or a P7.
* I think that the tint bins were more consistent?

 

[aurum]

tint bins were more consistent?

What do you meant the tint bins are more consistent?

 

[Th232]

In the LED manufacturing process, while the variables like luminous flux and tint can be controlled to some extent, part of it is still random, and so manufacturers bin LEDs by those variables. Within each bin there's going to be some natural variation within the specific bin, this is true for any of the variables, flux bins are described as being between XXX and YYY lumens/350 mA (or some other current level), tint bins are described as being between four points on a CIE 1931 colour diagram. If there was no variation, flux would be bang on a single number, and tint would be described as being at a single point.

I'm going off memory here (which may very well be faulty), but I believe that what some people have noticed is that if you take 100, say, Cree XR-Es in a specific tint bin, and compare them to 100 SST-50s in a specific tint bin, there'll be a lot less variation in the tint of the SST-50s. I can't be bothered comparing the graphs, so I'm not sure whether it's just that Phlatlight have much tighter bins, or whether their manufacturing process is that much more consistent.

At least, that's what I've learnt.

 

[LEDninja]

SST-50 4500K Tint Comparison Beamshots

 

[Th232]

Yeeks!

If that is indeed tint variation then I definitely retract my previous comment. That said, it looks to me (as some noted in the thread) that it's a case of a bin mix-up rather than variation.

 

[easilyled]

I'm seeing plenty of lights with this emitter, but I can't see why it's preferred over the MC-E or P7, since it seems to have a slightly lower efficiency, and SST-50 lights tend to cost more than MC-E or P7 ones.
Is it a matter of die size, i.e. the SST-50 is easier to configure as a thrower?

What are you basing your evidence on for the assertion that SST50s are less efficient than MCE's or P7's?
Were the top bin SST-50's used for this comparison? ie. WJ output.

 

[alpg88]

What are you basing your evidence on for the assertion that SST50s are less efficient than MCE's or P7's?
.

From what I can see:

* An SST-50 can go up to 5A, at which point the output is way in excess of anything an MC-E or a P7 can put out. Note, however, that around the 2.8A level it'll be, as you said, beaten by an MC-E or a P7.

that is correct.

 

[ma_sha1]

I no longer have this light, but the spec below are simply not possible to achieve
with either MCE or P7:

Franken MagDEFT with SSR-50 (~5.5 Amp):

• Greater than 1300 emitter lumens
• Greater than 115,000 Lux at 1 meter
• Greater than 2x the throw of the stunning ARC Mania X6
• Out Performs the SR90 (SST-90 on 4" SMO) search light slightly in throw.
• Similar throw as DEFT FTP but with 5x the hot spot area, thus more useful as a search light.

http://www.candlepowerforums.com/vb/showthread.php?t=268894

What I don't quite understand is why most mfg. keep releasing SST-50 in 2.5/2.8Amp drive current,
in those cases, there isn't much advantage over MCE/P7,
except the no "cross" in the hot spot, which can be be overcome with proper reflectors on MCE/P7.
.
.

 

[Illum]

What are you basing your evidence on for the assertion that SST50s are less efficient than MCE's or P7's?

hmm...

[I][B]LED             Efficiency    [/B][/I]              [I][B]Current Limit [Absolute maximum][/B][/I]
SST-50          [URL="http://www.luminus.com/stuff/contentmgr/files/0/8bdf01102a2326a6cddaecf20c8e4d4c/miscdocs/pds_001345_rev_03_sst_50_w_product_datasheet_illumination.pdf"]>100 l/w at 1.75w[/URL]           [URL="http://www.luminus.com/stuff/contentmgr/files/0/8bdf01102a2326a6cddaecf20c8e4d4c/miscdocs/pds_001345_rev_03_sst_50_w_product_datasheet_illumination.pdf"]5A[/URL]   
SSC P7          [URL="http://www.acriche.com/en/product/prd/zpowerLEDp7.asp"]90 l/w[/URL]                      [URL="http://www.acriche.com/en/product/prd/zpowerLEDp7.asp"]2.8A[/URL]
MCEs            [URL="http://www.cree.com/press/press_detail.asp?i=1222776900692"]80 l/w at 9.8w[/URL]              [URL="http://www.cree.com/products/pdf/XLampMC-E.pdf"]700ma[/URL]

Unlike stringing up LEDs, where the total foward voltage is much higher than the supply voltage, SST leds are a bank of paralleled LEDs, totalling a foward voltage between 3 and 4V. In this case the inefficiencies of a boost converter can be avoided by using naturally higher efficiency buck circuits with high current capability and cells capable of sustaining the current needed. MCEs are certainly easier to address existing lighting applications because the waffle [die package] is nearly identical in size with all other lumiled/CREE reflector openings. But the opening alone didn't stop many lights from being force fed P7s

 

[easilyled]

Where are the figures at 2.8A comparing efficiencies of WJ flux SST-50 with D-bin SSC-P7s and M-bin MCEs?

 

[Illum]

Where are the figures at 2.8A comparing efficiencies of WJ flux SST-50 with D-bin SSC-P7s and M-bin MCEs?

Without first hand use of these LEDs I can only provide the general ballpark value described on datasheets, I have no idea where you can acquire information for such a detailed efficiency comparison.

I merely posted what I have on hand readily available to suggest Fallingwater that his claim may require some reevaluation. If you are looking for a more detailed scope of what different bins has to offer than I am sorry I cannot help you

it would be nice to see which a list consolidated though...information regarding different bins appear scattered as usual.

 

[was.lost.but.now.found]

hmm...

[I][B]LED             Efficiency    [/B][/I]              [I][B]Current Limit [Absolute maximum][/B][/I]
SST-50          [URL="http://www.luminus.com/stuff/contentmgr/files/0/8bdf01102a2326a6cddaecf20c8e4d4c/miscdocs/pds_001345_rev_03_sst_50_w_product_datasheet_illumination.pdf"]>100 l/w at 1.75w[/URL]           [URL="http://www.luminus.com/stuff/contentmgr/files/0/8bdf01102a2326a6cddaecf20c8e4d4c/miscdocs/pds_001345_rev_03_sst_50_w_product_datasheet_illumination.pdf"]5A[/URL]   
SSC P7          [URL="http://www.acriche.com/en/product/prd/zpowerLEDp7.asp"]90 l/w[/URL]                      [URL="http://www.acriche.com/en/product/prd/zpowerLEDp7.asp"]2.8A[/URL]
MCEs            [URL="http://www.cree.com/press/press_detail.asp?i=1222776900692"]80 l/w at 9.8w[/URL]              [URL="http://www.cree.com/products/pdf/XLampMC-E.pdf"]700ma[/URL]

Unlike stringing up LEDs, where the total foward voltage is much higher than the supply voltage, SST leds are a bank of paralleled LEDs, totalling a foward voltage between 3 and 4V. In this case the inefficiencies of a boost converter can be avoided by using naturally higher efficiency buck circuits with high current capability and cells capable of sustaining the current needed. MCEs are certainly easier to address existing lighting applications because the waffle [die package] is nearly identical in size with all other lumiled/CREE reflector openings. But the opening alone didn't stop many lights from being force fed P7s

I don't believe that's correct about the MCE. Cree's data sheet has 430 lumens (M bin) at 350mA/die @ 3.5V. That comes out to about 88l/w. Still not equal the others, but then the point again of the thread was, is it worth the difference in cost, specialty drivers, etc.?

 

[easilyled]

Without first hand use of these LEDs I can only provide the general ballpark value described on datasheets, I have no idea where you can acquire information for such a detailed efficiency comparison.

I merely posted what I have on hand readily available to suggest Fallingwater that his claim may require some reevaluation. If you are looking for a more detailed scope of what different bins has to offer than I am sorry I cannot help you

it would be nice to see which a list consolidated though...information regarding different bins appear scattered as usual.

Thanks Illum. I'm just trying to fathom out why Fallingwater makes this assertion which is backed by Th232 and alpg88.

I have a problem accepting statements that are written as facts but without some sort of data to show why.

 

[Fallingwater]

Thanks Illum. I'm just trying to fathom out why Fallingwater makes this assertion which is backed by Th232 and alpg88.

I have a problem accepting statements that are written as facts but without some sort of data to show why.

Going from memory, I remember someone stating as much in a thread, possibly with a graph - though I may be wrong.

So if I get it right, it's a bit less efficient at 2.8A, but will go with no problems up to 5A, at which point a MC-E or P7 would be less efficient (provided enough heatsinking existed to drive one so hard without burning it up). Is this correct?

If it is, I might just have found the LED for a high-power project I'm thinking of building... I was originally thinking of overdriving a P7 at 1050mA per die, but a SST-50 driven at 5A would be even better.

 

[mudman cj]

I question the validity of comparing those values of efficiency. In order to be comparable, they need to assume the same die temperatures and drive currents. Of course, in the case of the MC-E you would assume the dies are wired in parallel.

 

[Illum]

I don't believe that's correct about the MCE. Cree's data sheet has 430 lumens (M bin) at 350mA/die @ 3.5V. That comes out to about 88l/w. Still not equal the others, but then the point again of the thread was, is it worth the difference in cost, specialty drivers, etc.?

well, CREE says "At 9.8W, the XLamp MC-E LED provides up to 790 lumens at 6000K "

790/9.8 is about 80.6, that was all the reference I was able to come up for it :shrug:

Besides...efficiency is measured at Tj = 25C [that's junction temperature, not heatsink temperature] ...highly realistic for our purposes

 

[was.lost.but.now.found]

well, CREE says "At 9.8W, the XLamp MC-E LED provides up to 790 lumens at 6000K "

790/9.8 is about 80.6, that was all the reference I was able to come up for it :shrug:

Besides...efficiency is measured at Tj = 25C [that's junction temperature, not heatsink temperature] ...highly realistic for our purposes

Right on, we are talking about two different questions. The first question is, "what led is most effecient at any drive rate?" And the second question is "what led is most efficient at its maximum drive rate?". The point I was trying to make was it appears that your comparisons of the three leds answered question one with the SST-50 and P7, and answered question two for the MCE. Apples and oranges.

 

[easilyled]

Going from memory, I remember someone stating as much in a thread, possibly with a graph - though I may be wrong.

I think I remember a thread like that a very long time ago before the higher binned SST-50 leds were available for purchase.

So if I get it right, it's a bit less efficient at 2.8A .....

The top bin SST-50 leds are so much more efficient than the lower bins that I think a statement like this is pretty worthless without some updated evidence IMHO.

 

[alpg88]

The top bin SST-50 leds are so much more efficient than the lower bins that I think a statement like this is pretty worthless without some updated evidence IMHO.

imho, is worthless to compare efficiency of them, (mce, p7 sst50),
you use mce, or p7 when you wanna limit current at 3A and less, if you can drive your circuit to 5a, you can't use either of them, you need sst 50.

theoretically at 2.8-3 amps sst 50 wont be as efficient as p7, mce, but it is irrelevant, since you shouldn't normally use sst at low amps if efficiency is your goal.
the biggest advantage i see is monodie in sst50, but p7 is, IMO, more practical than sst50.