Some more tint comparisons featuring a couple 519As.
non-scientific:
519A from Convoy w/ Carclo frosted TIR
5700K, dome on, 300mA:
5700K, dome on, 1.0A:
Im using an Opple, not as accurate as your ColorMunki, but here are my DUV measurements:
3 CCT pairs, domed and dedomed:
the pair on left and right, are 3 step LEDs from azhu, the middle pair is 5 step from Simon
and compared to sw45k D180 w duv –0.0105
I consider the domed 519a as the replacement for any LH351d (same output, almost same hotspot size, less green, and better R9),
dedomed 519a 4500K as a replacement for sw35… (lower duv, 30% more output)
the dedomed 5700K replaces sw40 (more output)
the dedomed 5700K replaces SST-20 4000K (less green)
domed 5700K replaces 219c 5700K (less green, more output)
thefreeman’s test and my 5700K look pretty comparable. I assume his is from Clemence. I also have 4500K and 3500K from Convoy untested.
non-scientific:
519A from Convoy w/ Carclo frosted TIR
3500K, dome on, 300mA:
|
3500K, dome on, 1.0A:
|
non-scientific:
519A from Convoy w/ Carclo frosted TIR
4500K, dome on, 300mA:
|
4500K, dome on, 1.0A:
|
4500K must be the runt LEDs that don’t bin as 4000K and 5000K I mean who actually wants this CCT in reality?
you measured them at less than 4100K… and the duv is slighly negative…
plus you can dedome to get a 20% lower CCT and double the amount of DUV below the BBL
whats not to like?
LOL
what CCT and DUV would you like?
I don’t mean to say it’s not good just an observation that if this is really 4500K then it’s pretty far off in CCT.
2700,3000,4000 all have legacy bulbs to match but 4500K is neither daylight or artificial CCT so it’s probably not used in big numbers
> if this is really 4500K then it’s pretty far off in CCT
agree… iirc mine was 4300K, they have been running on the low side…
I could say the 4500K is so popular, and the 4000K is such a dog, that they are trying to unload the 4000K that dont move, by batching them into the 4500K that is in so much demand… lol
speculation runs rampant 
You’ve probably seen a lot more 4500K than you might realize. That’s the upper average of metal halide/HID lighting (when new, anyway). I think typical range is 43-45 but with hours or poor/weathered fixture lenses it looks warmer. For LEDs I’ve hardly seen it, just here and there, but I really like it. There were some auto 194 led bulbs that I found in that temp and wish I’d bought more before they disappeared entirely. To my eye it’s closer to “real” sunlight than the noon/5500K stat that has become the accepted norm (I recall discussion by the pros suggesting that 4800-5000 was more like sunshine, typically).
I consider my Opple more of a toy since it’s just not very accurate or repeatable but it does color temp fairly well. It reads about 400-600 low every time, but repeatability on that one feature is within its spec and is good enough. It’s really best suited for open air room testing for FM and low end lighting designers rather than concentrated beams like flashlights. Fun to play with though, notably how lenses and optics play with color temps.
Good to know about the HID. However, in Texas afternoon sun even higher than 5500k every time I have checked with my own meter.
Get an S2+ with the 4C chip, and stick it behind a TIR to mix the colors better. That’s about a nice 4500K, and what I’ve called “nice warm sunlight”.
A reflector is too fried-eggy, leaving the hotspot too warm and spill too cold.
That’s because you’re closer to Texas. lol
I’ve wondered about air conditions and how that must affect readings…around here we almost always have particulates (fine dust if not a host of other things) so crystal clear air and bluebird days are rare. Curious what the parameters were for the group(s) that measured and decided upon this years ago.
Note about my Opple statement/accuracy above. No idea where it came from or why we had it, but in the pile of hoarding junk at the office we found an old reference calibration bulb from Panasonic (with some deprecated but very expensive projector bulbs). Should have kept it but it went away to the recyclers, but I was able to light that up and compare the Opple readings…same low range. That bulb was marked as having over 20 burns on it, if that means anything. We don’t do anything remotely related to light measurement so how we came to possess it is a mystery. Now that I’m contemplating buying a Hopoocolor unit, shoulda kept that bulb.
Convoy has the S3 available with the 519A. I normally prefer a warmer light. Is there any compelling reason to choose 4500K or 4000K over the 3500K?
No, it is a mater of personal preference. With dome on, they are all not pink tinted, and have great color rendering.
I prefer the dedomed versions, because the tint is less neutral, slightly strawberry colored. I do like pink tint, Im not a fan of neutral DUV.
but fwiw, the dedomed 4500K is a personal favorite. It comes out to about 3400K and has, for me, a nicer tint than 219b 3500K, that I find too neutral. Bear in mind the dedome ruins the nice big round hotspot of the domed version. Dedoming makes the hotspot smaller, and not perfectly round.
here are my domed and dedomed pairs:
the duv is Opple data, for comparison purposes, though not as accurate as more expensive instruments.
Not really. Seems the 519A pretty much has a neutral tint regardless of CCT. Only when you slice or dedome them do they become more desirably rosy.
So just pick the temperature you like.
Lots of negative duv options with these emitters.
Here is a sample of what I’ve measured with an Opple and 519A emitters in Wurkkos WK01 hosts.
More details in my 519A slicing post.
FB
Nice work FBsLights. Thanks for sharing.
I made a post in reddit comparing the CRI of several of these emitters and mixes. I did not do a good job of controlling for power levels, though probably they were around 50-200 lumens for all. But take the Duv and CCTs with a grain of salt because power levels can change those:
https://www.reddit.com/r/flashlight/comments/vbmq8a/color_rendering_data_for_several_519a_emitters/
Direct link to color data: 519a measurements (CRI and TM-30) - Album on Imgur
Emitters/mixes:
519a 5700k domed
519a 5700k dedomed
519a 2700k domed
519a 4500k domed*
519a 5700k-2700k domed mix
519a 4500k-2700k domed mix
Bonus: Osram W1 amber-w1 white 6000 mix “pink w1”
Almost put an order for D4v2 dual channel with 219b 2700k+4500k and then found this thread so… 14 pages later, can you confirm whether the following is the right TL;DR; of the thread so far:
- 519a can emit more (twice as much?) light than 219b
- It’s a bit more floody but since it can be driven higher eventually can both overflood and overthrow 219b
- Tints are generally really good. Mb not as rosy as devoted sw45k lovers prefer but mostly below BBL (for high CRI versions)
- Greenish tint is observed on 519a 4500k by most
- Dedoming drops color temp about 1+ level down (i.e. 5700k becomes ~4250k)
- Dedoming impact on beam profile is FILL ME IN (so far I mainly observed random distortions on the screenshots posted here - donuts, coronas, etc.)
Is it safe to say that 519a is the more powerful and clear-TIR-capable version of E21A (in terms of colour closeness to BBL)? (My D4v2 E21A 4500k is amazing as the pic above confirms but bream looks horrible once I put clear lens on)