Yes, that is the problem if throw is needed.
It is seen here (XHP50.3 HI, XHP70.3 HI is more or less the same)
Only a few LEDs built in flip chip-design have clearly sharp edges around the LED chips, like CSLPM1/NM1.TG, Cree XP-P or Nichia emitters.
What is the LES of this emitter?
Don’t know what you mean.
Size? Type? Design? What a LES is?
Light emitting surface.
What aspect of the LES are you interested about?
The size. Lol.
An eyeball estimate puts the size of the phosphor at around 3mmx3mm. However, the actual light-emitting surface is harder to determine because of the sideways leakage–the entire emitter is lit up to varying intensities, so technically the LES is the entire 5mmx5mm, but that’s not a very useful figure.
What might make sense is to introduce a notion of “effective LES”, defined by taking the total flux (in lumens), and dividing it by the average intensity over the phosphor area (in lumens/mm^2). This would be more useful than phosphor size for the purpose of intensity calculations, and will be greater than the 3mmx3mm phosphor size, but I am unable to quantify exactly by how much. Koef3 would be in a better position to give an estimate.
Then write it clearly, can avoid misunderstanding.
@QReciprocity42 wrote almost everything to this topic. Earlier with classic lateral (lambertian) LES it was very easy to determine LES size, but nowadays with flip chip style chips it is almost impossible to get proper values with determining the LES in the classic way.
I have some ideas to get around this issue tho. Give me time to figure this out, I have a lot of other stuff in the pipeline.
I thought I was clear. When someone says what the LES of a led is, they mean size. This in my mind was universal. Apologies for the misunderstanding
Don’t want to be nitpicky, but LES means just ‘Light emitting surface’. From this it is not clear, if only the size is meant or the type of surface/LED chips itself.
Even outside of LED technology it could be used as designation for surfaces where light is coming from (displays, stars like the sun, animals like fireflies, filaments used in incandescents), but for these this term is virtually not in use at all.
Could be a language thing, for example in french surface often means area, if one is asking about the ”light emitting surface” then they’re asking about the area/size (unless specified otherwise). It’s easy to make mistakes in english because there are so many french words and a decent amount of them have a slightly or sometimes completely different meaning.
Recently peeled off the silicone layer of xhp50.3 HI r70 6v 6500k by accident while tightening the reflector. Power on… puff’
So the plague is real! That silicone layer is that fragile…
I’ve done that with two XHP50 HD’s when experimenting with TIR’s. They don’t go puff though- you end up with a bright almost UV blue LED.
450nm “Royal Blue” they call it. I think it looks cool but I’m always slightly concerned its damaging my eyes so I haven’t kept one about
I tested two XHP50.3 HI samples by Convoy shop for color and light quality.
Thanks to @ebastler for getting these! 
Both samples were bought in May 2024. Please keep in mind, that these results must not necessarily apply to other LEDs bought in same time and with same order code.
70 CRI 3000 K, order code: XHP50D-H-7A0-J20-D0-B-01
like given by color bin 7A, the duv is below 0, which is good. The tint is nice, although the low color rendition index is clearly visible, especially in direct comparison to 90 CRI.
90 CRI 3000 K, order code: XHP50D-H-30G-G20-D0-U-01
Unfortunately, this sample is more like urine yellow. Even the (on paper) still good duv of around 0.003 give an unpleasant yellow tint. I cannot recommend these LEDs if a good tint is necessary. These LEDs are binned in 3-step.
These LEDs can be distinguished by the color of the phosphor layer. The high CRI variant has more orange phosphor, and due to the thicker layer the gaps between the four LED chips are less visible as for 70 CRI variant.
Thank you very much for the test! Too bad the 90 CRI has such an ugly tint - it would be an awesome LED if it did not. At least the 70 CRI has a solid tint, although I would obviously prefer a high CRI LED with the same (apparent) tint…
I have then 3000K R70 in a Convoy S21E and like the tint and beam a lot. Glad Simon accidentally sent me a R70 instead of an R90 after looking at your test…
I have a bunch of XHP50.3 HI 2700k 90 CRI LEDs from Digi-Key (XHP50D-H0-0000-0D0UF227G) and the tint looks beautiful behind a color mixing optic. The Duv is less than .002 from two bare LED samples at various currents. These are F2 binned LEDs, which is a low bin. It has been posited by many that higher binned LEDs may exhibit less favorable tint characteristics. I cannot yet say if that’s true, but it makes sense. I’ll present some tests of a few samples soon.
I wonder if the failure-prone spots on the XHP correspond to the positions of nozzles that deposit the goop in manufacturing 