So this is for measuring CCT and color rendition, I bought my i1 studio 340€, less expensive than the Aliexpress spectrometer linked by Correllux, it’s not handled though, but can sort of be with a smartphone and the paid Argyll Pro app. i1 pro and colormunki photo/i1 studio can sometimes be found second hand for less.
If color rendition is not needed then a colorimeter like eye-one display Pro (see Argyll CMS compatibility list ) can do and will be less expensive.
That's really interesting about the Gossen...never heard of those! Thanks for sharing that, ChibiM. Have no idea how those could work but I'm going to start googling to see what I can learn. Sometimes "old tech" amazes me in the ways that people came up with solutions and the theories behind them. We are so used to electronic gizmos that do everything imaginable that most of us can hardly even to the necessary math and thinking to start the walk to the solutions in a more basic way.
Thanks for all the info everyone! I’m starting to think I should just sell the KR4 with the not-5000k emitters though. What is the point in measuring it? The bottom line is that it doesn’t look like my other 5000k lights and measuring it won’t change that.
That sounds pretty good, I’d be satisfied too.
Mind that light that exits an integrating sphere has some spectrum changes compared to the light that is entered (unless you have a Spectralon coating inside your sphere): slightly more short wave light (the “blues”) are absorbed than longer wave light, so the CCT warms up a bit, it can easily become 500K lower.
Hey Djozz, no it's just styrofoam, without a coating. Good to know. I was also thinking about shining it onto/through normal copy paper for testing. Would that help? Or what do you think is the best way to measure CCT?
With Argyll CMS/x-rite spectrophotometers we can apply a correction for each wavelengths with the spotread command, so first we measure an uniform light source (e.g. an incandescent/halogen bulb), then measure the source in the integrating sphere, then apply the correction. Even with his professional sphere Maukka has to do this.
If you can get the spectrum data then you could apply this correction manually.
Yep, that is the advantage of using a spectrometer to measure light levels. Chibi and me are using a luxmeter in our spheres, so we can not correct for spectrum changes. I had a few lights measured by maukka in his sphere two years ago, with varying tint and CRI, the diffence between his and my calibration varied by as much as 4% for the different lights and I assume that the difference comes from the fact that maukka compensates for spectrum changes and I can not.
Just curious as someone who has never done this before… is measuring CCT & CRI supposed to be done in a sphere, or are there other/better ways of doing it? I suppose a sphere would help integrate angular differences of certain LEDs (think Cree flip chips with bad tint shift). But is there another way that wouldn’t introduce variation, like the sphere does by absorbing certain wavelengths more than others?
I use my sphere for measuring light output, not for CCT/CRI because I know that it changes the CCT. So if I post CCT/CRI numbers, I always add how/where it is measured (usually the middle of a hotspot from a reflector or a TIR because that is how we mostly use leds).
Maukka can measure integrated CCT/CRI of a led in his sphere because he does the spectrum correction, but he usually also measures in the hotspot of an optic because that has more relevance.
Depends what you want to measure, if you want to know the tint of the spot then you just shine the flashlight dead on the sensor. If you would want to measure the whole angle profile then you measures at several angle and that tells you how much it varies (see maukka reviews or my thread about LH351D slicing), which is quite important, you might have a nice integrated tint but an absolutely terrible angle profile, as you mentionned somme CREE LEDs are terrible in this aspect, or a 144A might have a terrible egg yolk effect and a nice overall tint.
If you’re measuring a LED bulb, the diffuser averages pretty well already, usually there is very little variation.
Honestly I’m not sure when integrated color measurement is an important metric, I guess for comparing bare LEDs mostly.
Yes as I and Djozz mentioned, the sphere must be color calibrated, even professional ones.
Dayum that’s a lot of money. I’d probably rather make a firearm purchase for that money. I think a lever-action rifle in .357 magnum would be great fun. Although the price of ammo is greatly inflated right now.