I have to admit that my method is not as consistent as yours because I did not set the lumens of all 3 lights to the same output. I just place each light about 3 centimetre away from the sensor and wait for 5 seconds before I take a screenshot.
Use a variable distance on all your lights and see how it changes, then average. Maybe the W2 is actually off?
apparently there are 5400K and 5600K versions:
Osram KW CSLPM1.TG 5P-fcbB46-15 White 5400K LED Emitter
CCT bin: 5600K – 6100K ( 5850K typ. )
Maybe try 15cm from the Opple, and use an output that produces 20,000 lux? (in case the 3cm distance is not catching the hotspot well, or too dim an output to get a more accurate reading?… )
Thank you for providing me with some advice. I got better results by changing the distance to 15 cm (also did a test with 45cm).
from 15cm (CCT: 5650, Ra: 65, LUX: 22634)
from 45cm (CCT: 5715, Ra: 66.6, LUX: 22386)
I couldn’t ramp up the lux to exactly/close to 20,000 lux so I hope this is good enough. I will spend more time tinkering with this device. 
Yay! Thanks for sharing your success
I also cannot get exactly 20,000 lux. I think your approximation is great. And Im happy to see the results are closer to Nominal CCT of 6000K (If that is correct).
5650 CCT is 6% variance. This is normal. LEDs all vary, and manufacturers group them, within a range of tolerance of 7% (I read somewhere).
Enjoy your Opple… its a fun little tool… I use mine a lot. It gives me useful, approximate, information…
I also use the xy coordinates a LOT!
x and y are shown in this photo:
.
I type the x and y umbers into this website,
and it calculates the Tint DUV, which is very informative, for me, because I like to avoid LEDs with Tint DUV above the BBL. It also makes a very pretty colorful chart
in this photo, I compare two 3500k 219b LEDs… we can see their Tint DUV is not identical:
.
thanks for sharing the fun
Happy New Year!
Just pick a realistic distance and brightness that you might use in the real world. The reflector or optic needs room to do it’s thing, multiple emitters will blend differently, etc. Don’t place a light directly on the sensor and no need to target a specific lux that wont mean much across multiple lights.
thanks for nipping my Urban Myth, that Lux Matters… LOL
I did a test at 5 different LUX levels, and you are right… LUX does Not Matter
I got almost identical results in CCT, CRI and DUV
.
as long as the beam is hitting the sensor well, the intensity does not seem to make a big difference (Im not testing Turbo outputs, which might have larger CCT shifts due to higher temperatures…)
I dont see many specs on this but as long as you stay under the max lux (could be ~75K to ~200K) it should be accurate enough and yes turbo outputs usually shift CCT a lot. Really only hit max lux measuring throwers from too close or using these in an integrating sphere.
My opple measures my W2 5700K collimator light at close to 5800K.
Capt’n Obvious asks: You all measure in absence of other light sources, do you? duck
> You all measure in absence of other light sources, do you?
close the bathroom door… solved
> as long as you stay under the max lux (could be ~75K to ~200K)
maxes out at 50,000 lux… 90 lumens, 1” away, is too close… Im easily trained to avoid that…
my test distances range 4-8” (bathroom counter) and below 100 lumens… the Opple Lux are between 500 and 20,000 mostly
> The reflector or optic needs room to do it’s thing
good point,
no thrower experience
Im testing mostly 18350 and smaller lights
the Opple has helped me to confirm my feeling that the 219b 3500k is not very pink
I like the 219b 3000K better, and Opple confirms, that it has lower DUV, farther below the BBL, than the 3500K…
.
.
.
(my first Novatac Mod… Happy!)
Wondering if the Opple DUV values (xy values), are in the ballpark
with your spectrometer readings…
do you have any 219b tests handy?
thanks for your inputs
Happy New Year!
Like any other LED, it depends on the bin. The 219b sw35 that Bob_McBob put in my SC64c is around –0.003 duv. However, most of the examples I’ve seen have been at or slightly above BBL. Same with sw40.
Thank You Very Much
I have the same impressions. Your posts are extremely helpful.
Happy New Year!
ps, I remeasured the AAA Tool w sw35 because the DUV seemed much lower than yours. I discovered a significant difference in DUV, based on Output Level:
.
well, that difference in LUX bothered me, so I redid the 2 lumen test, with the light stading on the sensor. Now the DUV makes more sense:
.
conclusion:
Lux matters
20,000+ Lux gives a more realistic DUV than <1000 Lux
I found this good images of teardown of opple from Reddit post
Is there a similar device, even for 3x the price (lol, like $100) that does a similar job if not slightly better? Or is the next step up in the hundred$ ?
Looks like it got dropped and broke the plastic mounts.
It seems to have nearly everything marked with reference designators. Maybe a programming port with the 5 solder pads— wonder if the Rx,Tx could be used for a wired version.
Maybe if you can find an exellent deal on a second hand x-rite I1 pro, i1 studio or colormunki photo, but usually they are more in the 2 hundreds.
Happy New Year 2022!
Aliexpress Flash Sale, Opple Light Master Series 2 only $17.66!
I had the G3 Pro in my cart for $23 but it was gone by the time I went to place the order, so I had to get the Series 2, not sure what the difference is though.
Thanks to all for posting about this really cool tool!
Thanks for posting! It seems this chip is EOL. I wonder if there’s a device using the 7341.
I asked Opple on aliexpress the difference between series2 and G3, here’s the reply:
“The functions are basically the same, and the third generation links are faster. App is the same, the product size is different. The connection is nearly the same, three are no obviously differences.”
It sounds like the only major difference is the physical size of the device.
Not sure what was meant by “the third generation links are faster” can anyone elaborate?
I also thought the app was different…Blue background vs Black background in photos posted here.