Great! Yes, it is looking like the HDS lights might indeed be a good calibration light, they seem to be lining up very close.
Texas_Ace BLF Calibrated Lumen tube / Sphere No math skills needed - Several spheres still available
Nice description on that Lumen tube. I noticed you still list the tolerance as plus or minus 7%. Using the reference lights you got from Maukka aren’t you able to get the tolerances a little better?
I always under-rate things 
The highest deviance I actually saw from the stock spheres with disc installs was an outlier with just over 5% before I corrected it (with 2-3% being average).
If I ever make more of them I should indeed be able to get things closer.
For now though 7% gives a windows for any outliers in this batch that I was not able to recalibrate here and it lines up with the official Cree tolerance ratings.
As a side benefit it might help people realize that even these spheres are not accurate to a single digits.
All of that said, I could technically rate them at 5% and still be over-rating them on the ones I have here if I don’t include the spheres already in the wild. Just not sure if that is fair since they would all be compared equally. I can’t know for 100% sure there are no spheres I sold previously that fall outside 5% once the Correction discs are installed.
Cool. Thanks.
Has anyone with a TA Lumen tube measured a Rofis MR70?
I’ve seen a lot of varying tailcap amp draws so there’s not much consistency. I need to know what High actually is. It’s probably less than 1500, but I’m not sure how much less.
Thanks
Just FYI, maukka is now offering a calibrated light kit as well as light testing service.
Very nice, glad to see he is offering that. Worth every penny.
Ace are the larger tubes you make 4” or 4.5”? I see in the OP it says 4.5 but elsewhere I saw 4 mentioned. Either way can I buy a set of those diffuses and some of the white plastic rings?
TA, just out of curiosity… how much would the 3.5” tube cost without the lux meter?
I have the exact same meter already.
Tried my tube today, every light is about 20-40% off, they read way too low, wonder if I have a dodgy light meter.
They are 4” PVC but the opening is 4.5”, so kinda both.
I might have a few scrap diffusers left but pretty sure I used almost all the good ones, I would have to buy another sheet to cut out more. I suppose I could do this if you wanted enough of them.
I do have a few complete spheres left here.
It would be best to use the meter it comes with since that is what it was calibrated with.
Far as savings, I suppose I could knock off the $16 that the meter cost me from the price if you rally wanted to do it that way. Although the accuracy would be unknown at that point.
I tested all of these spheres personally as all of these latest batch were within ~+/-5%.
Remember that most of the readings you see posted online are too high and even my old readings were off by 32% before I got the ANSI calibration lights.
Do you have some examples of lights that are reading off? There are only a handful of lights that have shown to be reasonably close to the ratings in this thread.
Are you basing this off of the lights factory specs and you seem to be measuring 40% less? Some examples that you measured would help us know for sure.
Your lights are still as bright as before you measured them, it’s just that now you are seeing the real numbers.
Also, have you seen the right way to measure them? You turn the end tube up 90%. You find the disc that is closest to the flashlights head diameter. Put the shiny side of the disc facing down, then put the flashlight head through the hole just a little bit (maybe 5mm to 10mm) and move it up and down a little until you get the highest reading.
I had joked that it’s kind of like waking up from the Matrix. What you thought was real was all a lie. A light you thought was 1000 lumen may have been 600 or 700 all along. 
Welcome to the real world. Lol
My PT16 is measuring 950 lumens so is 10% off.
My Emisar D4 XP-L HI is measuring 3500 lumens so is about 20% off.
My DX80 is measuring 20000 lumens so is about 60% off making the calibrating tube redundant.
I’ll try another HS1010A light meter to see if that’s the problem.
What makes you think those numbers are off exactly?
They sound perfectly reasonable to me. As has been shown dozens and dozens of times in this thread, the ratings on lights means almost nothing. Fenix is the only brand that seems to get close to the ANSI ratings on a consistent basis that I have seen.
The D4 is an FET light and makes for a very bad comparison light, although 3500 lumens is actually pretty impressive, that must be with an 18650, not 18350.
IIRC the DX80 is well known to be overrated as well, although I do not have one to test msyelf. I am sure someone else can chime in though.
Overall, those readings sound very good to me.
Do you have any Fenix lights or other more consistent lights for comparison? I am pretty sure others with this tube can compare the numbers.
So far all the readings you have mentioned sound correct but they are not very good for comparing to others.
You have to ignore the rated lumens on the lights and realize that most light manufactures lie about the ratings. As we have seen throughout this thread, the official ratings are simply wrong most of the time.
After TA confirmed my readings seem right in a pm I can confirm the brighter the light you use on the TA calibrating tube the more the reading will be off, a 1000 lumen light will read around 800 lumens and a 100000 light will read around 50000 lumens the the thing is utter garbage people. £132 down the drain…
Have you tested it with using a neutral density or other dimming filter in front of the meter. If the difference (lux ratio) between a low and high output light is different with the filter, the meter isn’t linear in the high end.
Acebeam X45 - 8000 lumens
Acebeam EC50 Gen II - 2200 lumens
MecArmy PT16 - 920 lumens
Emisar D4 XP-L HI - 3600 lumens
Blf X5 Kronos - 920 Lumens
Imalent DX80 - 20000 lumens
Noctigon Meteor Nichia - 4550 lumens
Acebeam X80 - 15000 lumens
Astrolux MF01 cw - 7600 lumens
Olight S2R - 745 lumens
Manker Timeback II - 2000 lumens
DQG Tiny - 2200 lumens
Blf Q8 - 4800 lumens
TN40vn - 6500 lumens
All other testers reported more lumens especially on my lumen monsters, I have plenty of other lights to try too