The Cree XT-E Royal Blue LED Explained

By Clive Bentley on Dec 21, 2011

You may have begun to notice Cree XT-E blue LEDs being included on some newer aquariums lights and unlike previous advancements, this generation of LEDs is designed for more than simple brightness and efficiency gains.

The Cree XT-E royal blue LED has been around for a little bit now, and is starting to get attention in the DIY community and some aquarium LED lights including the PacificSun Metis Hyperion. As with all things Cree, this LED is the next evolution in their product line, and is bringing some interesting advancements to both home builders and manufacturers. If you’ve been wanting to know more about the Cree XT-E blue LEDs here is some information about them and maybe some little nuggets of information that some of the more knowledgeable out there may have missed.

The obvious comparison for this LED will be against the current fan favorite, the XP-E. While the differences are fairly subtle, they are there, and are important in some respects. We will be talking strictly about the royal blue version of this series, as there is a recently introduced high voltage white XT-E, and didn’t want to confuse things.

The physical characteristics is where there are very few differences. The external footprint is the same as all the XP series devices, and has the same solder pad layout. This makes life a lot easier for those brave enough to try re-flow soldering at home for putting these LEDs on custom boards, or at least something different from the usual 20mm star. This also helps when it comes to making secondary optics fit. Many of the optics with mounting bases have tight tolerances on the opening for the LED. So much so, that even XP-G LEDs, which are only a hair larger, can sometimes be tough to make fit without modification.

There is a slight change in overall height though, going from the XP-E 2mm height, to the XT-E 2.36mm. It’s not a lot, and it won’t affect most people, but it should be known that the change in height will affect the performance of secondary optics. The average DIYer will never see a difference. This issue is more for manufacturers to deal with when trying to squeeze the most out of their product. For the aquarium world, that doesn’t seem to be that big of a concern right now, with most of the major manufacturers choosing to forgo TIR (total internal reflection) lenses in favor of either a lens free setup, or a simple reflector.

What most people instantly notice is the different die structure. This is really where the magic is with this LED. This is one of the first LEDs to use a three dimensional die structure. The reason behind it is to increase the viewing angle, and improve overall light distribution. This LED sadly wasn’t developed with us in mind. Cree is targeting this LED to the remote phosphor design set, where every little extra degree of viewing angle is of importance. The traditional flat die has limits on how wide it can project the light emitting from it before the use of primary optics. With that limitation, the primary optics can only do so much, and will typically be limited to 110-130 degrees, depending on the LED in question. With the three dimensional die structure, it’s now easier for the manufacturer to increase the viewing angle closer to 180 degrees. In this case, Cree has set the viewing angle to 140 degrees, but with a little work, could probably increase that angle a little more.

An extra 10-30 degrees more doesn’t sound like a lot, but there is more to it than that. Part of the difference comes into effect when you look at how the light is distributed through that range.

Typical spacial distribution for the XP-E series LED

Typical spacial distribution for the XT-E royal blue LED

It’s pretty easy to see from the pictures above that there are some considerable differences in the way light is emitted from the LED. The XP-E has a pretty typical smooth curve, where the output tapers off at a predictable rate as you start to go off center. The XT-E keeps a higher average intensity over a wider angle. This equates to a little more even light over a larger area than the hot-spot characteristics of a typical LED. It’s a fairly small difference for us, but a huge difference for remote phosphor applications. This more even light distribution is more important to us aquarists than the wider viewing angle, as after a certain point, the light no longer is directed into the aquarium below.

So far, this hasn’t been all that big of a deal. It’s an LED that’s the same size as existing XP-E LEDs, with a little increase in height, and a little different light distribution. Where the bigger differences start to show up is in the electrical characteristics and the light output, which is what most of you are more interested in.

When talking about light output, the first thing to note is that Cree is rating all their newer generation LEDs (XT-E and XM-L, among others) with a die temperature of 85C. Originally, everything was rated at 25C, and became an industry standard. Unfortunately, 25c die temperatures are not very practical outside of the lab, so Cree, and a few other manufacturers like Philips Lumileds started to use 85C as the benchmark point. This is a lot closer to real world conditions, and a much better predictor of performance. As a result, you need to do a little more work when comparing the light output of the XT-E to the XP-E. Rated output for the top bin XP-E at 350mA and 25C comes in at a minimum of 500mW. To bring it to a level playing field with the XT-E, the output is derated to approximately 460mW (estimated to be about 92% of the rated output based on the data sheet). XT-E top bins are currently rated to 550mW at 350mA and 85C. Now that things are apples to apples, that’s a 19.5% increase in output at the same current. Not bad.

Electrically, things improve again. Cree is gradually forcing the forward voltage down on their LEDs, which helps to increase efficiency, reducing driver requirements, or letting existing drivers run more LEDs in a single string. While most incremental LED updates bring the forward voltage down a tenth of a volt or two, Cree has managed to get the forward voltage of the XT-E down 0.35v over the XP-E. That’s a pretty considerable drop. The current XP-E has an average forward voltage of 3.2v at 350mA, while the XT-E has managed to get it down to an average of 2.85v. Keep in mind that these are average voltages and your experience may vary. As a result, this makes for a 12.3% increase in electrical efficiency at the same drive current. And you know what that means; less heat.

The last thing to note about this LED is something that a lot of people overlook; thermal resistance from junction to solder point. Most people don’t need to really look at this, as the heat sinks that are used in most DIY projects are vastly overkill, and can easily keep die temperatures in check. It’s still interesting to note that Cree has managed to increase the ability for the LED package to get heat away from the die by almost a factor of two. Thermal resistance for the XP-E is 9C/W, while the XT-E is down to 5C/W

So that’s the XT-E for you. It goes to show that Cree has a lot of tricks up it’s sleeves for improving the LEDs that we know and love, and has managed to improve a great LED quite considerably when you look into the details. To top it off, the average price of the LED is actually lower than the XP-E, making this a fantastic LED to use, and certainly one that will be catching the attention of DIYers and manufacturers alike.

For more “light” reading, check out the XP-E and XT-E data sheets linked below

Cree XP-E Data Sheet

Cree XT-E Royal Blue Data Sheet

Anonymous

great article. thanks
http://pulse.yahoo.com/_3PQR2S2EAZUDJKVYUCKLCWJ37Q Micheal

Too bad about the spread, but yeah I can see how these are more geared towards overall use where in a light fixture on the ceiling you absolutely want as wide a spread as possible (not sure why they’d make them in royal blue though…) Although I could definitely see these as the new actinic lighting strips, use a trough reflector similar to the ones ecoxotic uses for their strips (which FYI I dislike, but with a spread of over 140° I could see a use for that). Or for a FO type system where you want the color but not necessarily the intensity (or cost associated with that many LEDs) this could work nicely.
Clive Bentley

I’m not understanding why you dislike the spread? Is it the angle, or the emission pattern? While the extra angle isn’t all that useful for us as a raw emitter, the emission pattern that results from the design of this LED can help reduce the amount of intensity dropoff as you go off center that you would see with a typical LED like an XP-E.

Now, remember that this LED was initially developed for remote phosphor applications. This is where the phosphor is blended into the diffusor instead of the LED itself. The Philips AmbientLED A19 LED light bulb is a great example of this technology. Wide angles increase performance and light distribution in this instance. We as reefers (both DIY and manufacturer) are using this LED for it’s increases in output and efficiency for the most part. These LEDs are a direct swap for XP-E royal blues, have higher output than XP-Es, and are generally cheaper by a good margin in volume, so why wouldn’t you want to use this LED?
http://pulse.yahoo.com/_3PQR2S2EAZUDJKVYUCKLCWJ37Q Micheal

Well I’m not a fan of the spread specifically because I don’t want my light spread out, I really don’t care about the intensity drop off from center because quite frankly I use lenses to focus my LEDs into a tighter pattern, yeah there will be some natural drop off but that’s going to happen if you put lenses on these XT-Es as well.

Actually looking at the plots you provided, they really don’t look drastically that different as far as intensity as a function of angle, only real difference in increased intensity I see is that extra little hump after the +/- 60° mark, but if you superimpose those graphs it looks as if the energy drops off quicker with the XT-E, especially on the + side of that graph.
Anonymous

If I were buying new LEDs for a fixture I would go with these, but they don’t seem like a big enough gain to warrant swapping out working XR-Es or XP-Es if you already have them running (which I do). Still waiting for that “magic” blue LED to appear, a blue that covers 420-490nm. When an LED like that comes out I will swap most of my royal blues.
http://www.facebook.com/people/Gabriel-Manis/11901247 Gabriel Manis

Wouldn’t it follow here that with a really tight optic, say 10-20 degrees since that is what you like about LEDs, that the spread across the 10-20 degrees would be more even with the XT-Es vs. XP-Es. Also, these are more efficient and will likely produce the same, if not better, results.

I don’t want to put words in Clive’s mouth, but maybe he can clarify that much. I don’t know how there would be a downside to these if you used optics to achieve the desired spread angle.

Ultimately, these guys will probably end up being the next bench mark for manufacturer’s to distinguish their fixtures from “last years models”.
Clive Bentley

The effect will be less pronouced once you start adding secondary optics, as they have their own intensity drop off issues. Regardless of the emission pattern differences, the output and efficiency gains of this LED over the XP-E will make this the go to LED for almost anyone building an LED fixture. Add in the lower price, and it’s a no brainer.
Clive Bentley

Agreed. Most people that have built their own fixture won’t be in a big rush to replace their existing royals with these. Anyone building a new setup should seriously look at these though.

A broad spectrum blue LED would be great, but tricky to pull off. I have yet to see any LED compatible phosphors with an emission range that low. Even if they did exist, they would probably have to be excited by a UV LED of some description to get enough difference between the excitation and emission wavelengths.
http://pulse.yahoo.com/_3PQR2S2EAZUDJKVYUCKLCWJ37Q Micheal

Will the price be lower though? Efficiency gains are one thing, but if the cost per unit is not priced accordingly then it won’t be worth it. I notice that with the XM-L lamps, while yeah more efficient than the XP-Gs I’ve also seen them over 2 times as expensive (and they’re not that much more efficient)

Related posts: