For this to not be the case they would have to somehow change densities / tint with different output lights to read differently and this is not the case.
Like the photo-sensitive glasses that darken in the sun and I can assure you I am not using anything that advanced.
To put this in perspective Lux meters use exactly the same method with the plastic dome over the light sensor.
It has a fixed percentage that it reduces the readings by and they simply calibrate around that and the end result is fine because the reduction is more or less fixed.
So a certified 1000 lumen Fenix reads 800 and a certified 10000 lumen Fenix reads 8000.
We are getting there, then you have extra absorption from the floody spill hitting the tube walls at a different angle from hotspot heavy lights. Then there’s extra glare refection bouncing back from higher powered lights.
I’m gonna buy a globe and see if I can replicate a professional set up.
Thus far all of the fenix lights tested have read within margin of error.
The idea with the centering rings is that it integrates the light at the inlet thus removing the discrepancy of flood vs throw. I am sure there is a few percent variance but once again the percentages are fixed so it will be minor. This is what people have seen during testing as well as my own testing.
The reflective tape mimics a reflector and thus keeps all the light inside the tube.
Getting a real sphere is indeed a better option but a much more costly one.
The issue is still that you will need a way to calibrate it, I recommend getting one of maukka calibration lights for this task. Using ANY factory light is a recipe for disaster.
I edited the above post while you were posting with this information that you might find helpful:
Damn, that’s fucked up. From what I’ve seen in this thread, you’ve been nothing but professional with him. Might be he’s considering his interaction with other members here to be part of his negative experience. Hopefully you can get it sorted, man.
KG, we weren’t trying to be rude to you but wanted more info to diagnose the problem. Keep in mind we are all customers that bought the TA tube just like yourself and we have no reason to bash a fellow customer. All of us customers, like yourself, are hardcore enthusiasts and some with a great deal of knowledge in this matter. The only difference is except we are not experiencing what you are experiencing with this product (E.g., Your M43 measured 4,550 lumens, while mine measured 6,950 lumens.) It was easy for anyone to assume from your initial few posts that you had too much faith in manufacturer’s lumens, ceiling bounce, and other’s DIY lumen spheres calibrated using manufacturer’s ratings and since many of your readings were reasonable (e.g., 4,800 lumens for Q8), a very logical explanation would be some of your lights might be duds (I had three identical Zebralights that measure 44/68/74lux), experienced emitter degradation (like my D4 XP-L HI I explained above), or just manufacturer’s overrating their lights (I have about 100 lights or more but not all the ones you have so can’t confirm them all). With additional information you are giving us (e.g., you have multiples of the same lights though not sure what they measured), it seems more likely either you are using it incorrectly (have you tried unplugging and replugging in the lux sensor?) or there is a defect in your specific unit (which would make it the first).
The hundreds of comments in this long thread have been very constructive and we try to be objective and factual. Only until recently your posts initiated some emotions and sorry for your negative experience. But when you came out of no where and started attacking this product without providing more information to help us diagnose, to many hardcore flashlight enthusiasts/experts here, that may sound offensive (I’m not because I’m not that savvy). Customers here contributed a huge wealth of information through hundreds of posts in this threads that helped shape this product. Some of your comments made it sound like rest of the TA tube customers, all hardcore flashlight enthusiasts, are dumb for thinking this product works great and you’re the only one that realize it doesn’t (e.g., your quote in this post). Regardless, TA himself has been very professional and objective.
Just a FYI, Toykeeper, who contributed to the coding for the Emisar series and the upcoming FW3A summed up the “lumen” phenomena pretty well in here This is the reason why many of us here discount lumen measurements from other DIY solutions and manufacturer’s ratings.
This kind of crap is the reason I don’t sell anything—I have well over 300 lights that I would love to thin out—I’ll give them away before I deal with this crap—What I’d like to know— KG when you get a Sphere are you gonna adjust for flooders and throwers like you stated you do with other forms of testing—I have tested lots of lights ( big small high lumen Blf GT etc ) both with my Tube and Ceiling bounce test — I use one correction factor for each form of testing —I was amazed at how close the two forms of measuring were
. 
