Nichia 519A

Flashlight Modding and DIY Parts
121

I dont think the Max lumens matter

the 519a has a larger hotspot, so it might work in place of the XM-L in a Sunwayman V11r,
.
instead of using LH351d that has lower R9. The 219b does not make a good beam w an XM-L reflector.

btw, slicing ruins the beam of an LH351d in an XM-L reflector…
.
so I see no benefit to slicing a 519a… it gives up all the size advantage, that could make it compatible in an XM-L reflector or Tir.

the 519a might also be viable to replace an Olight XM-L2…
.
not sure… I put a 219b in an Olight, but replaced the clear XM-L Tir w a Pebbled Tir, to fix the shadow artifact in the center of the beam, caused by the smaller 219b when paired w the stock clear XM-L Tir
.

the 519a is less pink than the sw45k, and is more similar in Tint DUV to the sw35 and sw40. So, choose 519a for the 2x larger hotspot, if the tight hotspot of 219b is not suited to the application or user preferences. Or for lights where 219b does not work to replace an XM-L due to beam issues, the larger dome of the 519a might produce a cleaner beam.

122

After seeing djozz's YooToo SD1 swap and recent reddit posts about the 4500K 519A, I'm getting the impression that the CCT is consistently and significantly lower than the rated 4500K. I just saw a reddit post that measures a 519A to be below 4200K. Does the warmer CCT seem to be the norm?

123

Those are my photos of the 519a posted by JS. The top is at Lightroom’s “daylight” white balance setting. The other is just iPhone 12 doing its thing. I don’t have a way of equalizing output between the 219b in an RRT01 and the 519A in a zebralight, so not sure if I could really show difference in hotspot intensity very well. Also the reflectors are a bit different. Edit: actually the reflector with the 219B is totally smooth so that would account for some of the difference.

I could try to get roughly the same output by eyeballing the ceiling bounce and then take a properly exposed hotspot photo. But I would be very surprised if the 519A wasn’t less intense, the upshot of less intensity/throw being arguably a more pleasant beam. For folks wanting more throw and rosiness dedoming seems like a good option. It will still be higher in output than the 219Bs. But I don’t understand how this would affect its performance in compatible reflectors. One question I have is whether the 519a is too big to fit in the Haiku’s 119 reflector.

I think my emitter example might be the one you saw measured. But I can confirm it is just barely cooler than my coolest and rosiest SST20 example, shown on the left in the reddit post here: https://www.reddit.com/r/flashlight/comments/sj8jli/nld_zebralight_sc64c_le_modded_with_nichia_519a/

124

some additional data points… all DUV is negative in his 519a sample posted by autumncrown on reddit:

data from this thread

fwiw, Im not sure ± 0.0010 is visually significant… basically the 519a seems to hover around the BBL in DUV… going slightly negative at max output…

LEDs are grouped in bins that vary by several percent. LED CCT also varies depending how high the output is.

A 4500k nominal CCT, is not an exact number to expect for every LED of that nominal CCT. Variation is normal.

to answer that we need to collect some info

1. can you see all 4 corners of the LED base, inside the bottom hole of the reflector. A photo would help answer that… but I did not find one.

2. does the 119 have the same size base as 519a, if it does it will fit the reflector if #1 is true. I have not researched the dimension specs.

3. we need to know the size of the mcpcb in the gizmo light engine… as it will need replacing to match the electrical pads on the 519a, which has a different configuration, than the 119…

125

I should have said that the McGizmo currently houses a 219C w/ Tana’s SuperLE comfortably. So I think that means #2 is true. Here is a photo of the 219C in the Haiku to help answer #1: https://imgur.com/a/MCJ9rd0

126

The bin that Bob_McBob would be the sm453. there is rosier bins though

127

The rosy crowd will likely prefer the “e” designated bins. In which case, I have a question for you pink-lovers: How do you tolerate sunlight, which is more green than my sm453?

128

that makes it easier, as the existing mcpcb could be reflowed w 519a.
afaik 219c and 519a have the same solder pad configuration, and the same size base.

and thanks for the photo… imo, yes the 519a will fit that reflector base hole… my guess is the 519a dome will be as large as the white parts of the 219c sides, which leaves good clearance for the reflector…

when my brain is adapted to sunlight, I tolerate LEDs w similarly green tint, nominally DUV 0.0032. I also choose cooler CCT lighting when my brain is set to Daylight White Balance (~5500K)

when my brain is adapted to incandescent (~3000k), I find LH351d and SST-20, greener than necessary. I also prefer warmer CCT, when my brain is set to Incandescent White Balance.

the pink tint of sw45k is not noticeable when I am Incandescent Adapted. It is only in relation to Sunlight, that the Pink Tint of the sw45k becomes evident.

People who use flashlights for work during the day, when they are in and out of sunlight, will tend to appreciate cooler CCT more, and wont care about green tint as much.

otoh, when relaxing at home in the evening, using incandescent at 3000k, the warmer CCT is more relaxing, less glaring, and green tint of an LED is more apparent, because it differs from the ambient Tint DUV of an incandescent lamp.

though it seems at first that there are two different Tint and CCT preferences, imo, it is the ambient color temperature and Tint of the environment in which the light is used, that determines the preference in CCT and Tint DUV in our choice of lighting.

For EDC, I like both, sw30, and sw45k, at different times of day. Im not much of a fan of 3500k or 4000k… When modding flashlight LEDs, I use a High CRI 5600k light on a headlamp. When cooking, I like sw45k. When camping, I prefer sw30… Courses for horses… :slight_smile:

My main interest in the 519a is to replace LEDs in lights that use LH351d and XM-L… better R9 CRI in the 519a

129

JS, your ambient light-adjustment hypothesis makes sense to me. I didn’t realize sunlight is that highly positive DUV. I am discovering my preference for neutral DUV lights, and I think part of the reason is that low DUV lights look more purply-blue at night than their neutral counterparts. I wonder if this is because of the larger purple/blue spike you get with low DUV blue-pumped LEDs:

https://i.imgur.com/dBCDDd5.png - e21a mix at 3700k
https://i.imgur.com/jZrjd8b.png - e21a non-mix at 3700k

https://i.imgur.com/CfehtBR.png - low DUV SW45k
https://i.imgur.com/2qXCLvT.png - more neutral e21a 4500k

It seems the blue spike makes up a larger percentage of total output when DUV is lowered.

130

Yeh. One of my pet phrases is that the 4C (B/C above the BBL, A/D below) of, say, the common S2+ but behind a TIR lens which mixes the light and gets rid of its fried-egginess, is like “warm sunlight”. It’s a great color for even indoors.

131

I just did my first 519a mod, to a V11r

not impressed by the Neutral Tint (personal preference)… but it does have 56% Higher Output than sw45k, also in a V11r.

.

more details: Sunwayman V11r LED output comparisons 219b, LH351d, 519a

132

Ouch!
That looks like a greenish 4000K to me

133

With that particular photos green/magenta balance, the –0.0117 which is very, very pink, appears quite neutral. So i don’t think the green of the 519 is represented accurately. I’d love to see the same comparison with a domed 4000k 351d for a reference for green.

134

Mind that tint pictures always fool you.

135

It would yield a better result to start with a higher CCT 519A and slice it to get 4500K. Rosy tint and no angular shift. I really hope Simon makes cooler temps available.

It seems that green tint tends to make a light look warmer compared to neutral, at the same CCT. My conjecture is that it’s due to curves of constant CCT not being normal to curves of constant duv (the BBL in particular) but slanted in a consistent direction. Thus, given a point P above the BBL if you try to find the closest point Q on the BBL by Euclidean distance (which is probably the CCT your eyes see), it will be warmer than the actual CCT of P.

136

I found the greenest LED I have… now the 519a does not look so green:

.

Tint perception changes based on the Tint of other lights it is compared to.

Ultimately, each of us has to try different LEDs for ourselves, to form our own impressions.

137

Looks like his camera white balanced to the middle light. Cameras white balance to both a blue/red slider (color temperature) and green/magenta slider (tint)

138

Sorry to hear that you’re not impressed with the 519A jon_slider. :frowning:

Mine will be going into a triple S3 using a 10511 Carclo lens. Hopefully that will help with the tint.

139

my opinion changes with the time of day

as long as the sun was up, the 519a looked nice, very neutral, not green, not pink

I think that is because the DUV of sunlight, 0.0026 to 0.0048 on my opple depending on time of day, is higher DUV than the –0.0003 of the 519a

iow, since the sun has DUV that is greener than the 519a… the Tint of 519a looks nice, during the day

I also really like how the 519a renders colors, including making red pop… The spectrum seems excellent to me.

otoh

after dark, when my brain adapts to the –0.0016 DUV of my house LED lights, the 519a, whose DUV is Higher, looks relatively green. And if I compare the 519a to sw45k at night, the 519a is just disgustingly green looking to me.

so…
I think the 519a works fine, during the day
in fact better than sw45k, which looks too pink, during the day

at night the sw45k looks considerably less pink than during the day

and, when I use the sw45k at night on red objects, they look great! Better than with 519a, to me.

courses for horses… my perception of Tint of a flashlight, changes based on the Tint my brain is adapted to. Ambient Daytime Tint DUV is greener than Ambient Nightime Tint DUV, in my world. :innocent:

140

Here’s why the 519A 4500k (sc64c host) is my favorite emitter right now:

-More neutral DUV than the SW45k, so colors look more natural (to me) compared with reference light sources like incandescent and sunlight
-less tint shift than SST-20, maybe partly because of larger die size (?)
-probably less tint variation between bins than the SST-20
-A little extra power and efficiency is never a bad thing
-It’s new and I like new shiny things
-Did I mention that it is neutral? It’s so neutral it’s hard for me to find anything to compare it to that makes it look green. I see 219b (SW45k/2700k) and e21a (6500k/2000k) tint mixes as pink, which objectively is true. Nice, but a bit jarring with neutrally tinted ambient light. That said, I have never been super sensitive to green. Even mediocre SST-20 bins (HDS) are tolerable to me. However, I can see the green in them. I do not see green in the 519A.

Looking forward to seeing this light in other hosts.