Just bought this Emisar D4v2 SST-20 5k mule flashlight and I noticed that even though the center tint looks very neutral white the edge looks ridiculously yellow... This is my first mule light so I wonder if without a reflector they do be like that....
SST-20 5000 K is a pretty bad choice because they tend to be green and ugly. But the tint shift you are experiencing might come from reflections on the bezel / inside of the head or from the domes (or both).
Thanks for the info. I just did an experiment by using a black tape to mask the interior of the bazel tube to avoid any possible reflection and the tint is still ungodly 
sounds like I lost the tint lottery or something.
In fact I think even the center is no where near 5k and quite more yellow compared to my nichia 219c 5k
btw, sorry for a dumb question: reflection on the dome can also be yellowish if looking from side?
It is normal for sst20 5000k as it is famous for its green side light wich normally is focused with the reflector in hot spot.
For beam without tint shift you need any domeless led or some dome leds like hicri sst20 4000k and below or lh351d or 219b will have almost none tint shift. 519a has some tint shift but not as much.
Damn, I should have done more research before pulling the plug 
Thank you so much for the information!
who in the right mind would choose the SST-20 5000K (mule or not)
But this type of distortion is present in mules, but usually not as bad (it could be just the SST-20 thing)
here my D4v2 4 x Nichia 219BT-V1, R9080 4500K, 4 x Nichia 219BT-V1, R9080 3500K (ISO 50, WB 5000K, 1/250s)
and Fireflies PL09MU 4 x Nichia E21A R9050 sm2050 2000K 5 x Nichia E21A R9080 4500K (ISO 50, WB 5000K, 1/250s)
Wow, that fireflies look terrific, I need one for me too!
Yeah even the E21As in my Nov-MUs (both 4500k and 2000k) exhibit green shift around the edges. I think most if not all LEDs will exhibit this without focusing.
E21A is domeless but it has too much luminophore around the small led chip. That is why it has more tint shift than other domeless leds. As a rule domeless leds has almost no visible or significant tint shift, E21a is exception here.
That’s what causes a fried-egg beam in general.
Yellower light coming out the sides hits the reflector and gets bounced forward, creating a yellower hotspot, ie, the yolk.
Bluer light coming straight out creates the bluer spill, ie, the egg-white.
With 3535 LEDs, the ones which would has the lowest tint shift and so exhibit the least of what you’re seeing are dedomed 519A, if you can live with the rosyness, or sliced LH351D, both have noticeable tint shift domed though.
Also as a general rule the higher the CCT the higher the tint shift, this is very noticeable with E21A, which as explained above has significant tint shift due to the phosphor area being much larger that the LED die, 2000K E21A has much less tint shift than 5700K for example, even if it still has some amount of it.
From my notes, a couple of (bare) LEDs with very bad tint shift (0° is above the LED) :
E21A 5700K (50mA) :
0° 2784.7 Lux, CCT = 6398K (Duv –0.0026)
12.5° 2663.9 Lux, CCT = 6300K (Duv –0.0021)
25° 2016.4 Lux, CCT = 6105K (Duv –0.0010)
37.5° 1819.2 Lux, CCT = 5850K (Duv 0.0006)
45° 1593.0 Lux, CCT = 5710K (Duv 0.0016)
57.5° 1188.4 Lux, CCT = 5424K (Duv 0.0038)
70° 863.2 Lux, CCT = 5142K (Duv 0.0060)
80° 587.3 Lux, CCT = 5032K (Duv 0.0070)
XHP50.2 for what is probably among the worst out there :
0° CCT = 8061K (Duv –0.0065)
25° CCT = 7191K (Duv –0.0031)
45° CCT = 5690K (Duv 0.0076)
60° CCT = 4802K (Duv 0.0199)
75° CCT = 4713K (Duv 0.0208)
That does not look good dingdong22!
If your disappointed enough, you may want to consider swapping out the SST-20 with 519A (or other) LEDs.
your photo from the side, makes the beam look worse than it really is, in actual use
if you aim the light straight at the wall, like this:
.
you may feel better, seeing that the middle of the beam, the part you actually use, is not the same color as the outer perimeter of the beam in your side photo
Jon already gave a great explanation as to why your angle against the wall is making you see it as obviously as you are. But there is a very detailed and specific reason why this happens in mules.
So that “color” that you see around the edges, is the color that your led is supposed to spread evenly through the whole beam. And when you put the LED behind an optic, it does just that. That color gets somewhat evenly split within the beam. Some LED’s are known to be better than others at having an even tint spread, like the 219B for instance. But you still have to adjust the optic and LED to get the best and most optimal beam and tint spread throughout it that you can.
It is well known that you can take the exact same LED and put it into different lights behind different styles of optics, and that color you see around the edges that’s supposed to be spread evenly, will behave differently in each light. This is because each optic distributes and spreads that color into the light’s beam at different angles. D4Sv2 has been a great light to highlight this tint behavior. Some LED’s are not as good tint-wise in that host as they are behind other hosts optics.
Anyway, the reason why I explain what that ring of color is supposed to do and how it’s supposed to be injected into your beam. Is so that you understand that when there are no optics in front of an LED like in a mule, that color ring you’re seeing has no way of being anywhere but there. The color and warmth of any LED will always just shoot straight to the edges in a mule. This is the exact reason why any LED will always seem cooler in a mule too.
So in your case, your SST’s are notorious for having a yellow greenish beam. So naturally your outer ring on a mule with that LED will be exactly what you see there. If you were to get a mule with 21B SW45K’s, the center would be more cool too, but the outer ring will be more rosy pink, because that’s what that LED’s tint is known to look like behind optics.
So like Jon said, that ring is very normal. I just wanted to explain the very normal and expected reason why it happens. Knowing this tidbit will help buyers understand that they have to shoot downward to achieve their desired CCT in a mule. So if you wanted a mule that “felt” like 4500k, you would need to get a 4000k LED or even lower to get it to “feel” like a 4500k.
Hope this was helpful, even if long-winded.
Very interesting discussion thanks. Could you explain why the underlined part above is that way? Am I wrong in thinking lower CCT LED is further from blue therefore requiring thicker phosphor, therefore more tint shift?
Also… where were you when I ordered my E21A light? My note on the LED a few months ago:
Emisar D4V2 w/ E21a 4500k
I’ve come across post where someone waxed poetic about this E21a LED like it’s the second coming, and I’ve come across post where it was the worst LED the person had ever seen. Based on my pictures and of course live view observation, IMHO this LED is extremely close in beam color of the 219b 4500k, but very slightly more yellow. Of note Emisar needs to use frosted optic Carlo 10623 for the E21e to smooth out the beam, but this frosted optic results in a dim hotspot, too dim for me. I’ve tried the clear optic Carlo 10622 with this light, but then the beam has a yellow greenish ring at corona (see pictures later in this thread HERE). 219b has no such problem using the same optic. Unless you see something that you really like, personally I would pass on this LED and get the 219b instead.
*