Getting closer to making jtr1962 happy-High CRI at 5K

[saabluster]

I'd say these new LEDs while not being lumen sluts are quite a breakthrough. The news here is making such high CRI while still having a relatively high temperature and reasonable efficiency.

"The SCW1653 version comes in 5,000, 4,200, 3,500, 3,000, and 2,600K versions. With a colour rendition level of 95 (CRI), it stands out from the crowd making it an excellent choice for a variety of applications. Brightness levels are typically 60-70 lm; depending on the colour temperature."

 

[Linger]

Oh my. That's a very nice development.

 

[qwertyydude]

This will definitely be a good way to make good video lights, I'm always the first one to point out that cri isn't everything when your color temp is an incan 2800k I mean candles are 100 cri but try to light a video or photograph with them and color rendition is aweful even though technically it's 100 cri light. A dirty little secret is xenon flash lamps and hmi lighting for movies only average 70-85 cri at their 6000k but when I was at a tv show taping before and they lit up the mall I was at and you could have sworn they cut out the ceiling cause it looked like noon day sunlight. So if these guys can pump these out competitively unlike nichia which seems to have an output of like 10 leds/year for their good stuff, then this would be a great leap forward.

 

[phantom23]

Datasheet:

http://www.stanley-components.com/en/search/search_product.cfm?search_type=1#FFFFFFhttp://www.speedyshare.com/data/820580518/19377468/576236/w1643u e.pdf

There are two versions:
- "High efficiency type" (SCW1643) woth more yellow color, 111lm@300mA and 70 CRI
- "High color rendering index type" (SCW1653), more orange color, 60-70lm@300mA and 95 CRI.

 

[spencer]

But unfortunately they have a maximum input of 360mA. Something like that at 1A...

 

[EntropyQ3]

Still, output limitations noted, decent efficiency at high CRI are good news! I checked out the spectral response graphs, and they look pretty decent. These would do great in 1W LED spot lighting applications, in various circumstances. I would use those myself in a number of applications.

 

[Tekno_Cowboy]

These look nice, very similar to a Nichia 083.

If they allow distributors, these could be used to make some pretty nice fixture-type lights.

The problem with Nichia is that they don't allow distributors. That fact makes it very difficult for the smaller manufacturer to obtain their LED's, as they come on reels of 1400 parts.

I'd start a group buy for a reel of Nichia 183's (minimum 92 CRI, 800mA max drive current, 200 or so lumens at 800mA) but I just don't think I'd even get close to the 1400 part minimum.

With either of these parts, there's also the problem that they're surface-mount. Without a MCPCB, they're pretty tricky to work with.

 

[jtr1962]

Those are certainly a step in the right direction. The 5000K part has 70 typical lumens at 300 mA. With a typical Vf of 3.7V, that's 63 lm/W. The lower CRI cool-white parts are doing about 100 lm/W. So that's roughly a 37% penalty for high-CRI. Now if only Cree would do something like this with the XP-G die then I'd be even happier. I'm estimating you could approach 90 lm/W @ 350 mA with an XP-G high-CRI version.

But yes, nice development which fills a definite void! And I do hope they allow distributors with these. I'd certainly like to try a few, perhaps for a desk lamp or reading lamp.

 

[mds82]

That actually looks very similar to t he Cree MX-6 LED. high efficiency, very very tight color bins, and similar design as well.

 

[saabluster]

That actually looks very similar to t he Cree MX-6 LED. high efficiency, very very tight color bins, and similar design as well.

Although the story I linked to talks about two different versions, one of which is similar to Cree's new LED, this thread is focused on the high CRI version of which Cree has nothing similar.

 

[McGizmo]

Hi guys,
I added a couple graphs/ charts from a recent Nichia document to my High CRI thread which some of you likely didn't catch so I will offer them here as well.

At some point hopefully the manufacturers of high CRI LED's will provide us more detailed information regarding the CRI beyond the one CRIa number.

I believe there are different phosphors being used by the various manufacturers so the Nichia Data may be only germane to their products but it is also likely that the "hit" taken in CCT VS CRI is to some extent a case of physics for all of the manufacturers. :shrug:

 

[saabluster]

Hi guys,
I added a couple graphs/ charts from a recent Nichia document to my High CRI thread which some of you likely didn't catch so I will offer them here as well.

At some point hopefully the manufacturers of high CRI LED's will provide us more detailed information regarding the CRI beyond the one CRIa number.

I believe there are different phosphors being used by the various manufacturers so the Nichia Data may be only germane to their products but it is also likely that the "hit" taken in CCT VS CRI is to some extent a case of physics for all of the manufacturers. :shrug:

You are right I missed the update in your thread so thanks for posting that here. I don't think anyone doubts that the hit of high CRI is felt by all manufacturers but what has been noted in the past, notably by jtr1962, is that manufacturers haven't seemed capable or willing to come out with high temp high CRI LEDs. So this new LED is a step in the right direction as far as that is concerned. I think you are more familiar with Nichia's product line than me so do you know if they have a high CRI LED at 5K? The highest I can remember seeing off the top of my head was 4.5K.

 

[McGizmo]

..... I think you are more familiar with Nichia's product line than me so do you know if they have a high CRI LED at 5K? The highest I can remember seeing off the top of my head was 4.5K.

Here's a couple graphs from my integrating sphere based on the 083B LED's I am building with presently (I was stoked to get the higher CCT and still high CRI!)

 

[dengdam]

has anyone tried buying any of these leds yet to try and make some nice lights?

 

[saabluster]

Thanks dengdam for bringing this one back to the top as I had already forgotten about it.

Here's a couple graphs from my integrating sphere based on the 083B LED's I am building with presently (I was stoked to get the higher CCT and still high CRI!)

Wow 5800?! That is quite amazing. Have you bought a new batch or is it the same batch as what you originally had? I ask because I have some that I bought from you and was just curious what they were. I have been abusing a few of them.:naughty: I have been experimenting with phosphor transplantation. Frankenstein kind of stuff.

It would be very interesting to compare the LEDs I linked and the 083Bs side by side. The Stanley's look to be a bit brighter which is welcome. We need more truly high output high CRI LEDs than what we have now.

 

[blasterman]

I've been doing some contract work for a regional manufacturer, and in one of their large plants they have about 50,000 square feet of 5500K high CRI fluorescents (the expensive ones). I can't imagine what the annual bill is to replace the tubes, and I'm pretty sure they are eight footers.

The difference between the high CRI and standard fluorescents areas of the plant is like being kicked in the head.

So yeah, there's a market for this.