Luminus Xnova Cube 1800K 90CRI tested

The Luminus Xnova Cube is not a new led, it was announced 5 years ago already, but it caught some interest on BLF as it is one of the few leds that come in the extreme warm 1800K colour temperature, and at 90CRI.

Mouser is usually too expensive for me to buy leds from because of the high shipping costs, but I had gathered a small wishlist over time so I did a larger order to get to the free shippings treshold (60 dollar before tax), and the Cube was one of the list.

The partnr. of the led tested is MP-1616-1103-18-90. I wish other manufacturers make their part nrs so easy to read: a 1616-size led of 1800K and 90CRI. The led is small, square and has a symmetrical two-pad footprint. Edit: I had not even noticed it but Agro pointed out to me that the Cube comes in two radiation angles: the 1103 of this test has a close to standard radiation angle of 130 degrees, while the 1100 version with higher silicon/phosfor layer has a very wide 170 degrees radiation angle.
The appearance and size of the 1103 version is very similar to the high-power Cree XD16, until you light it up, then you notice that it has 2 very small dies next to each other and thus the Cube has in fact more resemblance to other midpower leds like the Optosolis and the SunLike (I mention these two because I tested them but there are many more midpower leds with two dies embedded in a phosfor mixture) than the single high power die XD16.

(here is the 1800K Cube lighted at 1 mA to show the light source)

As you can see, the Cube with its two separated dies will behave bad in the common flashlight optics but I want to use this led for a simple mule build and a small lantern so that is no problem. But I do need the voltage/output charactistics so here is the test, with older data of the two other midpower leds mentioned added, for no other reason than to have some reference.

The datasheet mentions that the tint of the 1800K Cube is above the BBL at 25 degC, but lands on top of it at the working temperature of 85 degC. Perhaps my copper (non-DTP) board cools too well because even at 700 mA the tint is still well above the BBL. But I must admit that when looking at the light colour, that looks still very good and extremely cosy. :slight_smile:

Here is the tint data at 350 mA measured straight above the led. At 60 degrees the tint is slightly warmer (80K) but nothing else changes much. Also, the measurements at both 20 mA and 700 mA are not significantly different from 350 mA, so I leave them out.

I have no conclusion about this led because it has no typical flashlight use, it serves a very specific niche. But I hope that a few people find the data useful. :slight_smile:

1800K LED? Isn’t that a bit too orange? :confounded:

Think campfire. It is not for everyone :smiley:

Candle light is 1800K. Very warm and cozy. Too bad it is above the BBL.

Apparently at 1800K, tints above the BBL are fine, I can see nothing bad about the tint whatsoever, just: campfire :slight_smile:

So, a low fire, like a candle or a campfire, is around 1800K? Well, I guess that’s not so bad! I love fire! :smiley:

I really like your reviews! I think you are the only person who does LEDs reviews??

It's weird how the led in the picture above looks very very nice but in reality it will look very orange..

I like it though:)

I really wish to see your flashlights collection one day! I think you have amazing collection of high CRI lights???

God knows what kind of high CRI lights you have! :-)

There are a few other led reviewers, maukka does some, and for a short period koef3 has posted a few. They both are better equipped than me.

Yes, I have quite a collection of high CRI flashlights now, but the hosts are usually the typical BLF lights. My most used high CRI flashlight is a cheap no-name tiny 16340 tail-clicky host from Fasttech, with a BLF-A6 driver and 3000K SST-20 and a lighted tail. (the lighted tail is 95CRI btw :innocent: )

Btw, these Cube leds will make a fine lighted tail too.

Oh, I hadn’t thought of using these in lighted tail switch PCB’s! I wonder what it would look like to put these behind a pink button cover. With the extremely low CCT and high CRI, it should look really nice!

Even tailcap LEDs is high CRI??!!! You are crazy! Just like me! lol

Now I will keep thinking about the light you mentioned can we some pictures if possible please?

Someone should make a tiny beside ‘candle’ with this led!

The E21A 2000K with DUV of 0.000 is a great candle light too. It’s just slightly cooler temp than an actual candle but you get neutral DUV tint and 9090 color rendering. I love all of my E21A 2000K modded lights. Great night time lights.

For candlelight colour lovers there is nothing not to like about the 2000K E21A, it is a wonderful led. The one thing that this Cube does better is the price of just 12 dollarcents per led, it opens up applications that you would not normally use expensive E21A’s for, like use in lighted switches.

Good point. I hope LED4Power or Lexel would start offering these in their lighted switches

Very interesting, thanks djozz :+1:

Not much to see about it, the host was small (D25C-size), cheap and it looks cheap and used. I just rebuild it to the mentioned specs and to the point that it is very reliable, it always seems to work. The lighted tail are 3x Cree XD16 2700K 80CRI, and 3x 633nm red 0603 leds. The added red to the XD16 light brings the CCT to 2200K and the CRI to 95. Not that there is enough light in the tail to see that good colour rendering at all, but it is the idea that counts. :smiley:

Nice job! These are interesting emitters for sure.

Thanks djozz. :slight_smile:

I see that you ordered a 1616-1103 and not 1616-1100. Yours is specced to have 135 degree emission angle as opposed to 170 degree of the other. I see that there’s even some light going back towards the MCPCB but the photo doesn’t seem to allow me to judge whether there’s much of it…Could you comment on the emission pattern?