"Optisolis"tm, Nichia starts with ultra-high CRI leds (mid-power currently)

Any update on these Clemence?

If nothing gets in the way, they should be available in May

- Clemence

output test is in the make

Packet to Maukka in the make too
:wink:

Some more informations:



- Clemence

Output test is up: Output and voltage test of 5000K and 3000K Nichia Optisolis leds

Saw it already. Looks like you could increase the current a bit further

- Clemence

I'm curious how the Optisolis performs in my setup. Overall the performance and efficiency looks very nice, even better than expected.

Maybe they can beat the 260 lumens at more than 1.0 - 1.4 Amps... - could be a great replacement for my JAXMAN M8 Triple...

Looks like a good choice for small key chain lights.

And good test!

So do these fit the quadtrix board? Else I might have to cut up a 5050 board like djozz did, but ofcourse that will give some insulation issues with DTP.

No quadtrix board doesn’t fit. Another quad board for 757 series is still work in progress.

- Clemence

I wonder if I could use a 3535 board with the central thermal pad electrically insulated by arctic alumina adhesive? Would an iptisolis fit then?

An XM-board works as quad board like that, see post #2 in my Optisolis test-thread.

You used the adhesive to insulate the board from the pill. What I had in mind was insulating the led from the board

The astrolux S1 has an anodized shelf right? Might buy one of those…

After seeing Djozz’s unexpected amazing result, we know that the Optisolis can be used for flashlight applications. At least for ultra high CRI small low powered EDC. I took some tint shots and the results were not as expected. Like Djozz said earlier, the extra “ultra” can’t be noticed as obvious as if we compare R70 to R9080. The difference compared to R9080 is too subtle for most people to notice in everyday scenes.

In natural scenes only those with extra CRI snobism will spot it. The biggest difference seen on blue (subtle), green (looks darker), and obviously any turquoise (cyan) colored stuff.
This perhaps looks like Lumileds’s “crisp white”. When shine onto white fabrics it looks whiter than the R9080 while still retain the strong blue, red, and green.

Don’t expect anything magical like the swap from R70 to R9080. IMHO, Red is more critical than cyan, even the move from R9050 to R9080 usually gives instant improvement. This is a great choice for ultra high CRI, low glare, evenly lit lighting system.
Retail price estimated USD 0,4-0,5/pc for the highest CRI rank. It’s cheap CRI/output wise, but very expensive for 757 class. The extra “hidden” performance tested by Djozz makes it an economical alternative for your own ultra high CRI home lighting.

See it for yourself. Not much difference from the already superior R9080:

I had to find suitable objects to show you the difference made by the Optisolis. The NF2W757GT-F1 (Optisolis) ran in 4S (4x series).

- Clemence

The more ‘niche’ a led becomes, the more I enjoy it! :party:

Thanks for the pics, I will try to make some comparisons too if I have time tonight.

It is interesting how color perception differs from one person to another. If my display renders the colors correctly, then the gain in green/blue area that I see is very considerable. And whites (the coconut) look very different, but again I need to see it in real life (1-2 days from now) to understand what is in fact more natural.

I tried a mouse-over but that did not work, so here’s a comparison next to each other, left Nichia 219C 5000K R9050 (in a D4), right Nichia Optisolis 5000K (as a quad in a diffused S2+). I do not have a 5000K R9080 source unfortunately. I must say that in reality there is some difference but I can not see much in pictures on a screen

It is different. But because the E21A sm503 R9080 has very slight pink tint (above 1A), this makes the 757 Rfc00 looks slightly green. I believe tint variation alone would ruin human perception. Using Maukka’s test result of E21A sm503 R9080, the 757 Rfc00 should be indeed greener.5231464
To my eyes and camera, 757 Rfc00 and Rfg00 have much less tint variation from as low as 50mA up to 1000mA, unlike E21A.
One way to correct this is to include a pure white paper in both pictures and use that as white balance datum (by software).
I just noticed that most modern phone like Sony Xperia Z5 and Samsung Note2/5/8 highly over saturates all colours by default

- Clemence