What about dedomed 519A efficiency? Any changes? If dedomed more warm and rosy, so efficiency should drop?
This also isnāt quite apples to apples, as itās comparing different CCTs. Lower CCTs are always less efficient in lm/W.
thanks I calculated the difference in percent:
.
for dedomed 519a 4500k, I reduced the value by15%
Ideally dedomed 4500K would be compared to 3500K since thatās the resulting CCT.
yes, that makes sense
random data
SP10 Pro w AA Eneloop, momentary maximum:
LH351d 5000K 280 lumens
519a 4500k 260 lumens
519a 4500k dedomed 230 lumens
Tint progression:
.
How do I find out the bins for r9080 m500 from the link in the first post?
Trying to find the highest cri 5700k which produces the most lumens
It doesnāt exist, not yet at least, according to Clemence Nichia said the yield of sm573 M500 is too low.
Thoughts on L5 vs L6 bin please. Which one do I want to get and why?
Many thanks.
I feel like this has already been mentioned; has anyone taken any measurements of some dedomed 2700k 519As? Looking to get my hands on some 3535 high CRI emitters lower than 2700k for a tint ramping D4.
Update: 45 degree beads = donut :person_facepalming:
At least itās round this timeā¦ unfortunately lack of time to do any further experimenting with this such as sanding/padding. To be continued later in the year.
What optics are you using ? I just tested with H04/yajiamei 17mm 20, 45 and 70Ā° beaded optics and indeed the 45Ā° one creates a donut, not the 20 and 70.
Edit : the H04 comes with he 45Ā° so people wanting to swap the LED for a 519A should consider getting other angled TIRs.
Edit2 : testing 20mm from Convoy, 10, 20, 30, 45, 60Ā° : they all look good to me.
Strange we get different results. Iām using a 20mm Convoy beaded optic without gasket. The original 3535 gasket with TIR is too high and the bezel doesnāt close anymore (Convoy S3). Wanted to keep the glass (and rubber ring) in for stability (could take it out, but it keeps everything properly in place). Plus the TIR sat really poorly on the gasket.
This was bothering me, so I tried one more thing. I suspected the optic needed to be raised a bit. So I tried another gasket; Convoyās black B35AM one. Itās thinner (but too wide for the 519A). Figured the TIR and glass would keep it relatively centred and in place. Et voila! Now itās a very nice even beam (and the bezel stays on). Still not 100% happy with the whole stack of all parts, but good enough for now.
Sunwayman V10r T6 w 519a 5700K
Low 0.05, Max 400 lumens on Vapecell 16340
.
modding sequence
Interesting how the CRI drops around 900K (20) when the 4500K 519A is dedomed, but only around 400K (15) when the 2700K is dedomed.
Yes sorry, I meant to type CCT. 
I assume the CRI stays about the same after dedoming.
519a 3000K, 4500k and 5700K
.
519a 5700K momentary maximum lumens is 40% brighter than 519a 3000K:
.
. (photos for color comparison, not at max output)
Colour shift is measured in mireds, not Kelvin.
5700K - 1500K = 4200K CCT, 62.7 mired shift
4500K - 1000K = 3500K CCT, 63.5 mired shift
2700K - 400K = 2300K CCT, 64.4 mired shift
Very interesting scale, the mired. It accounts for the fact that as color temp approaches infinity, the black body spectrum, restricted to the visible range of wavelengths and normalized, converges to a fixed spectrum (a weird inverse fourth power distribution). Thus in any reasonable color space the points corresponding to the increasing sequence of temps would converge, and the BBL terminates in the interior of the color space, instead of becoming, say, infinitely blue. The behavior as temp goes to 0 is opposite; the spectrum does get infinitely red.
Out of curiosity, is the mired difference between two points on the BBL of the CIE 1931 chart proportional to either 1) their Euclidean distance or 2) arc length distance along the BBL?
Another curious question is why might one expect constant mired shift rather than constant CCT change as a result of dedoming.