I just measured the die area, you are correct that it appears to be 1mm².
So maybe not much better than thXP-E2 in a single LED thrower.
Well, just like the XP-L HI, maybe the tint is good plus we don’t need to dedome it by ourself.
Cree, I would like XP-E2 red and green in HI version, thanks.
Someone once told me the CREE XQ-E Torch, was actually the XR-C die size. I had this initially in my head when I posted, which is like 0.78mm² die. Now I’m starting to wonder something about the XQ-E Torch. The thermal resistance is 12.5°C/W on the XQ-E Torch if you missed me say that (I re-edited the first post a few minutes after posting). Does what I just said make anyone wonder if the XQ-E Torch is possibly just using the wrong photo, besides me?
Because…
Regular XQ-E = 6°C/W
XQ-E HI Int. = 6°C/W
XQ-E Torch = 12.5°C/W
Smaller dies typically have higher thermal resistance, for obvious reasons, junction surface area, mainly.
Knowing that, and then looking at the above numbers, this throws a big RED flag. You don’t just double TRes by binning an output. Or adding phosphor types which are different (I.E. XP-E2 Torch).
That red flag keeps waiving, making me wonder if the Torch has that 0.78mm² die. I haven’t been able to find one, ANYWHERE posted, used, accept on Alibaba for sale. I’ve never used Alibaba. But there are some interesting things, primarily every LED we say we have trouble getting here in the states is listed on there. I start to wonder if we’re getting our LEDs from the wrong places when we have to use Mouser etc (not Mountain though, Richard’s great and deserves my business 
). There is the obvious “China scare” when you expect a bin. Like, “Will a U4 be a T6?”. But when you’re dealing with an LED the seller can’t physically swap because only one bin actually exists…well then I start to wonder if we should try one of those sellers out for those LED types. This LED, would even be a candidate, and I notice, on Alibaba they seem to have instant access to the dies we say we can’t get for weeks to a month after new release. “Shenzen Glow Electronic Technology Co” appears to be one seller that has everything all of the time that’s LED-related.
But back to the main issues. How about those goofy thermal resistances? I think it was someone here that told me the XQ-E Torch is the XR-C die. But I almost don’t think Cree would leave a picture up showing 2-bond wires, unless it’s a “new” XR-C die with 2-bond wires. We can pretty much bet the regular and HI XQ-E, is a better version of the XP-E2 die package, though. And I’ve always loved the E2. Knowing that, my statements were very conservative about lumens when overdriven. With 3° lower thermal resistance than XP-E2, the XQ-E should be outperforming the E2 all along (dedome to dedome), and it’s been right there the whole time.
Don’t let the small size make you hesitant to try one, they will reflow as good as any other board component will. PM me if you are having any trouble reflowing LEDs, and I’ll give you a method that’s easy and helpful that anyone can do with what you already (likely) own.
So no one makes a 1.6mm package, LED central-location copper board? Since there’s no center strip for reflow, we can’t physically get a DTP made for these, or can we? Because the DTP is only on the neutral path, right? If Noctigon can make the center neutral DTP work and remain neutral, the exact same method could be used to make it work on the +/- paths, there would just be two isolated copper columns.
The XQ-E HI is available in royal blue, blue, green, PC amber, red-orange, red.
Not clear on what your saying. If the LED has no neutral thermal pad then there is no way it make a true DTP that is neutral for the LED. If you electrically isolate a pad then it’s not DTP. The lack of the dielectric layer is what makes it DTP.
Holy freaking DUH. Haha, yeah, I didn’t think that one through, to the bottom of the PCB. 8) Or I mean, what I meant was, that this was a new special board design intended for “hot dedoming”. :bigsmile: :bigsmile: :bigsmile:
What you could have is a DTP on the cathode / negative, the star will not be neutral but you could isolate the star with non electrically conductive thermal compound. Dielectric at the led pad is more of a bottle neck than at the much wider surface of the MCPCB star.
I could be way off base here but isn’t the central thermal contact pad on Cree LEDs just a metal coating applied to the ceramic substrate (and tiny holes that go through the ceramic I think)? So when why wouldn’t a regular DTP system work but instead of reflowing it, you would need to use a thermal compound. I think a gallium based TIM might wet the ceramic well enough.
For the life of me I can’t remember who it was and forgot to bookmark it but awhile ago on the “other forum” a prominent member was experimenting with sanding the ceramic substrate to be really thin and in the process he removed the thermal pad but was then using a high performance TIM to glue it to the MPCB. The whole idea was to make the ceramic substrate super thin to improve thermals.
I mean pulling heat from the contacts may be slightly more effective due to the metal conduction and since that perforated metal filled thermal path is gone, at least I think that’s how the thermal path works, I could easily be thinking of something else.
I’m not completely following you. You mean, instead of reflow, a conductive thermal paste to get electrical conduction?
If so, I see resistance in the makings. Resistance=heat. See where I’m going?
Anyways, anyone could probably deduce this by now, but here’s what I think as of now. This LED is outcast I would say to begin with, even in commercial lighting, and it got the HI treatment. Would you even think for one second the XP-G2 wouldn’t? I think it’s probably within weeks from happening.
First I questioned myself as to what SC³ chip allows the 1.5mm² XP-G2 die to fit similarly in scale to the other “HI” LEDs? The XB-D.
The next “XP-G2” will be a HI Intensity on a 2.45x2.45mm SC³ chip. That’s my guess. Thermal resistance will probably drop ~25%.
Any of you guys have comments on the Alibaba thing I mentioned up a few posts? From past experience watching the web, this LED will probably show up there, before it shows up anywhere else.
I mean regular reflow for the electrical contacts and use a thermal paste for the center area connection to a DTP. Sure you would need to find said DTP board in the right size and I don’t think it would be quite as good as but much better than nothing.
As much as I’ve seen people say to stay away from alibaba and aliexpress I order things from aliexpress all the time and have worked with companies from alibaba numerous times without issue but that really means nothing as it depends on the seller. I would say you or someone just needs to make a test order. You can pretty much ignore all those MOQs they list and just tell them you want to make a test order of a small amount for inspection. I’ve never had anybody turn me down unless it was a custom product. And really even after the test you can get them to sell you smaller quantities with a little less discount more often than not.
Both points interesting. The first with the thermal paste could definitely be effective if the board was modified and setup the right way, so that the little tiny strip area that’s there had dielectric removed and top copper was showing. If you know the LED is permanent, I would put a fillet of thermal epoxy around the whole chip with a bit under it.
The MOQs were what was keeping me hesitant, always more than I want, even of items large and common they will have something like 3 minimum. Or LEDs, 1000-100,000…ehh no so much needed. Yeah, some test orders would be good, though—that makes sense, thanks.
Swap the word Ginger for flashlight and that is what a lot of this thread is like for me 
If you’re going for the maximum thermal path you could remove the dielectric, reflow on a thin copper shim that is as thick or just a hair thicker than the dielectric layer then pop the LED on there. I would try the gallium based TIM for the thermal pad. I’m not sure it will wet properly but when I’ve used it, it seems to work pretty well and it’s more thermally conductive than most other TIMs and epoxies.
I made a DTP (I think) for a /- Nichia LED before. I used a copper blank disk, I sprayed painted one surface to create a thin layer of electrical insulation. I then took two small pieces of copper and glued them to the copper blank with thermal paste. The copper pieces were separated just enough so they don’t touch each other. I then soldered the LED on the copper pieces, one serving as the and the other one as the -.
Something similar could be done for the XQ-E. I might try it. By the way does that qualify as a DTP?
Not really DTP. I don’t know what the official definition of Direct Thermal Path really is, but around here it means that the MCPCB is in direct, uninterrupted contact with the LED pad (generally the Thermal pad). Since yours is separated by paint and thermal paste, it would not be considered DTP.
I suppose that there is no way and no reason to make DTP board for leds that dont have additional (thermal) pad. LED is non-dtp itself.
Anyone tried them in terms of throw ?
I thought you put solder paste on the center pad, so the LED thermal pad is not directly in contact with the copper but there is a layer of solder paste between the LED and the MCPCB. Is that right? The thermal paste is my case is not between the copper pieces and the copper base, it on the side of the pieces. The only thing between the copper pieces and the copper base is a thin layer of paint.
+1