BlueSwordM LED Group Buy 3: a 3V 5050 domeless Ra9070 4000-4500k LED, the perfect XM-L2/SST-40 (UPDATE 2: Something new!)

Over the years, we’ve always wanted to find a high CRI domeless replacement for the 3V XM-L2/SST-40 LEDs that are currently available.

Until now, it was practically impossible to find such an LED, for many reasons:

1. High CRI XM-L2s were only ever available in low color temperatures, and tints available were never really the best.
2. SST-40s were even worse: only available in 5000k and up, never available in high CRI.
3. SFT-40s, while domeless, are only available in low CRI 5700k and 6500k.

4. While the XHP50.3HI and XHP50.2 do exist in 3V, they basically don’t exist in high CRI or lower color temperatures because luminaire/light manufacturers probably don’t see a point in it.

So, let’s get on to the main subject: I actually found a manufacturer who’s willing to manufacture US a small batch of the LEDs with our own almost perfect specifications(more details below).

What manufacturer is it? Getian was my first try, but unless I commit a lot of money(100 000pcs), we’re not going to be getting a 3V 5050 High CRI LED from them any time soon :stuck_out_tongue:

It’s actually a smaller manufacturer called Rayten(Shenzen Rayten Lighting Co). They usually make stuff for automotive manufacturers actually: even though they’ve had the capability to make high CRI LEDs for a long time(movie and photographic sector for other LEDs), they’ve never actually done for the automative sector since most of them apparently went disgusting low CRI high CCT LEDs :confounded:

It is called the Rayten 5050W S55.

Anyway, before we go on, here’s the datasheet for the LED itself:

Manufacturer website:

Here is also a picture of the LED itself:

More information about the LED for those who don’t want/can’t download the datasheet down below.

LES and LED physical dimensions:

Electrical properties:

Luminous properties:

Let’s go onto what we really want:

Main specifications:

  • 3V 5050 size
  • 2.9x2.9mm LES
  • Domeless
  • 4000-4500k
  • Ra 9070 —- 90CRI minimum, R9 70 minimum. That’s why it’s not the almost perfect LED: no R9 80 minimum currently is possible, but that’s fine still.
  • Tint binning with a duv range of 0.010 to 0.000 — 219B SW45k grade to perfectly neutral grade. I personally wanted tighter binning, but tighter than that is currently not possible with their equipment and expertise, although it should become better with time.
  • Decent efficiency for the 70CRI model at a minimum of 110-130lm/W at rated current and 85C derated(1A).

Overall, this is a very exciting LED, especially since we’ll finally be able to replace so many XM-L2 and SST-40 lights(Olight and older lights come to mind).

High color rendition, decent output, good-great tint(especially since it’s domeless and we have good tint binning) and an actual way to make all 5050 mega-cheap LED lights with angry blue emitters actually usable.

Now let’s get down to business: because we’re doing a fully custom run of the LED, that means we won’t be getting any samples, and the samples is the actual order.

I’m willing to take a slight risk this time since this seems to be the almost perfect LED for the things we want to replace with it :smiley:

2100 pcs of this LED will be ordered for this run. We may order more if demand is very high, and future runs might be considered if demand is consistent and manufacturers aren’t picking up this LED.


  • 2.25$US/pcs for <20pcs
  • 2.05$US/pcs >=20pcs


  • 5$US internationally
  • Other shipping options are available in the US and Canada at an extra cost if you order a lot of LEDs. Europe is a bit different in this regard because of much higher shipping costs.

Link to interest sheet is below:

Voice your interest down below so that we can finally get the 3V 5050 High CRI LED we’ve wanted for years!

Current # of LEDs reserved:

Update 1:

1. It’s been confirmed by the manufacturer, but donut holes are not officially not an issue:

“because of the no dome lens design, classical focusing holes are not an issue with this LED design.”

2. QC will not be an issue:

” We work with a lot movie industry customers which have very strict specifications like yours, so it is not an issue.”

Interesting update 2

Interesting new information for you my friends.

I’ve actually found another manufacturer making a high CRI 5050 3V LED!!!

Even better from Rayten in some ways: single emitter die(so it should have higher intensity and no cross-die), Ra9580(not a “puny” R9070) and a nicely lower VF for better regulation :smiley:

The only downside? Higher thermal resistance(5C/W instead of 2C/W) and a bit lower efficiency.

Reserved for future use.



Put me down for 20 please.

Any idea about angular tint-shift?

If “bad” like most later Crees, then they might suck in a reflector (ie, fried-egg beam), but still be quite nice behind a TIR lens.

Even 519A has it… i’m enjoying XHP50.3HD and HI (HD sliced). You cannot have it perfect and we became real grumpy old grandpas

@Lightbringer, no idea about tint shift, but since it’s a domeless LED and we’ll ordering one with good luminous properties to our eyes, it should be pretty good overall.

Should I post about it on Reddit? That’d certainly boost demand for it :smiley:

Interested in 3 units

Put me in for 6. Sounds interesting.

Do they have a company site you can link us to? Google is returning zero hits for me although I see a “Rayton” optoelectronic group connected to auto light fixtures (complete).

It doesn’t seem like they have a website, but I’ll ask.

As always, there’s the unofficial manufacturer page on Ali :stuck_out_tongue:

I then find this monstrosity of an COB LED:

Maybe we should do a GB of this next :stuck_out_tongue: :smiling_imp: :stuck_out_tongue:

Yup, Reddit will bump this GB

I’m in for 20.

Isn’t their website?

This looks like 5050 led in question

I found something interesting

This looks just like the natural white version Yinding 5050 so maybe the are the actual manufacture of that?

Imagine a high cri version of that led :smiling_imp: :smiling_imp:

Damn, you actually found the manufacturer website only a few minutes after I found it.

You must be quick AF.

And yeah, you actually found that COB LED.

Sadly, not the same manufacturer.

Ah…how weird! Even when I search “rayten-lighting” or with the .com, google refuses to show it to me! What’s up with that.

It’ll be interesting to see how they test. Kinda looks like they won’t handle the heat of overdriving them much…marketing specs vs. graphs don’t quite jive (particularly concerning is that Vf/If graph…). I think for my plans that’d still be fine and hopefully the temp/tint is nice enough.

Not sure what to make of the nice furnishings and Latin gibberish on the single entry for Industry News: News Fifteen: weolcome to our website

Eh, I’m not really worried about overdriving potential.

It’s still got a good low thermal resistance(2C/W), so a bit lower than the LH351D and a decent bit lower than the SST-40, which itself can handle a nice bit of power.

If that’s accurate, with the 125°C Tj also. If that graph is accurate and applies, it’s hitting around 3.2V at 1A. Looking at the steepness of that line I don’t see how the stated 4A max is possible on a single lithium cell, looks like even 2A might be asking a little much from it, maybe? I dunno…proof is in the test once in hand. :slight_smile:

I’m in for 5!

@Correllux, yeah, it’s actually a bit higher than the LH351D and close to the XP-L HD in terms of VF (looks about 3.05V at 1A).

Certainly not bad, but nothing like the monster low VFs we’ve had these recent times.

Very important though: these numbers are at 25C on a actively cooled desktop heatsink.

I’m not surprised their numbers are higher than all of the other manufacturers which test their VFs at 85C…

That means real world VF will be quite a bit lower.

Good point…I did notice that but it’s a head scratcher because they are stating ambient temperature, not junction or solder point, and there was no temp indication at all on the Vf/If graph. The voltage drop isn’t much, though…comparing the SST40 the difference between 25C and 85C is like 0.07V, less than a tenth. Dunno…let’s see how they do. I hope these will be nice in a couple of host projects I have in mind.

Also….curious. They have that LeiTeng logo on the ali and web site…and they are mentioned in the notes of the datasheet. Also mentioned in the datasheet is Kasin. As if those two names are manufacturers. I know borrowing and copying is the accepted standard but three names juggled about. Would the real company name please stand up? lol

So, talking with my contact, I’m realizing something.

Even the smaller/unknown LED manufacturers can produce perfectly fine LEDs and whatnot, it’s just that generic luminaire/lighting manufacturers often ask for the absolute cheapest LED prices, making for stuff that’s rather unsustainable in business.

That obsiouvly requires corners to be cut, and not caring much about QC. scinating to see that basically everyone can make good stuff if they’re actually paid enough.

It’s really interesting talking someone who’s actually concerned about this and telling me this directly, and honestly kind of refreshing.

I’m now waiting to get beamshots to see if we get a donut instead of a waffle design.

Edit: Good news! Donuts shouldn’t be an issue because of the design, which is very nice.