Hey krash, here is the photo… its the rough side the picture you see…
Texas_Ace BLF Calibrated Lumen tube / Sphere No math skills needed - Several spheres still available
Wow. I tested my pelican 7060, 368 lumen light… it was right on the spot… 369/368/366. By the time I took a pic, it was 366 lumen…
I also tested the fenix tk15 and it was around 1045 lumen around 20 seconds range…
So far so good…
Is that a 4” sphere? That could be why we are getting different results.
Is there a difference in readings if you align the two discs with each other with regard to the ridges on them or if they are at some random oblique angle?
I used an S2+ short tube with aspire 18350’s - medium mode, fresh battery each run, two runs each configuration, measured beyond 30seconds and got:
No disks - 503
2 disks smooth side in - 313 (.62)
2 disks smooth side out - 370 (.73)
2 disks rough side facing each other - 342 (.68)
I’m going to use the last configuration - seems good to me.
It seems clear to me that we should use a stable test light and then experiment with the discs until we get as close to a 0.68 correction factor as possible.
Let’s say it reads 1000 lumen before then we want it to read 680 lumen with the discs.
500 before
340 after, etc…
Please don’t try and make a light match it’s rated specs. That is not the point of all this. 
Yeap… . I dont even try to match… TA said two rough disc should be facing the tube… thats what i did, and results are great…
Here are the results in my 3.5” tube with a 150 lumen light:
Both smooth sides facing sensor 0.725
Both rough sides facing sensor 0.63
Both rough (or smooth) side facing each other .68
There was a 2 lumen of difference whether it was the 2 smooth or the rough facing each other. Too small to matter in my opinion (see next post).
Next I will do a comparison at a higher lumen level just to confirm my results.
I did some comparisons using my ODF30 which has a boost driver and seems to have very consistent output.
High level (about 1500 lumen)
Both smooth sides together .695
Both rough sides together .688
Turbo level (about 3100 lumen)
Both smooth sides together .689
Both rough sides together .684
So I’m getting consistency at higher levels. I’m going to be putting the two rough sides together on my tube to get as close to that .680 correction factor as possible.
A-plus job on using these diffuser discs TA. :+1:
Now I just need to talk you into loaning me those “Maukka measured” lights. :laughing:
TA, what about me sending you a light that I know has very stable output in the 200 lumen range?
Could you compare it to your reference lights and tell me what it’s real output is?
I can include shipping money as well to send it back.
Hmm, those are the same basic numbers I got, just with the wrong orientation. Very strange.
Well luckily as long as you get the right number it doesn’t matter. Just odd that people are getting different results compared to the ones I tested here.
That is what is supposed to happen, it sounds like it is working like it should. Your 4” sphere does read slightly different then the 3” but only very slightly.
Nope, they look exactly the same to the human eye. They just got slightly different readings when I tested them. Although another possibility is that the numbers changed over the length of the sheet and that is why some are getting different numbers.
The ones I installed in these spheres were from the first section I cut out, I had to cut another set a few days later to have enough. It is possible they read slightly different, although I tested a few random samples in my sphere and they seemed to be within margin of error.
Either way, luckily it is pretty easy for people to do the math and get the correction factor dialed in on their particular sphere. It annoys me that is needed though.
Indeed, the 4” does get slightly different results from the discs but only slight.
All of these numbers are numbers I saw with various combos of orientations and discs from the old and new batch. I am thinking that there must of been some inconsistency in the sheet that caused the second set I cut to read slightly different for some reason.
With the old batch I was going to recommend installing them with the smooth sides facing each other,.
So as annoying as it is for me, I would say everyone should do some basic math and double check the orientation of the discs. As you can see, you should be able to get it well within margin of error by adjusting the directions of the discs.
Sure, send one my way and I will measure it on my sphere.
Send me a PM to work out the details.
OK… thank you TA!! And thanks for the tube…… I’m loving it. :+1:
TA, just read through all the updates. Sent $5 via PayPal as well. Fiddled around with the disks, and I finally got a .68 correction factor with the rough sides both facing out (smooth sides facing each other in the middle). Appreciate all your hard work with this. I’m charging batteries now so that I can spend all night testing my lights 
Received my discs. After installing & re-testing , my MT09R modded by TA & shipped at 20k lumens now test at 13K :cry: . when I first received the sphere it was tested at 19.2K lumen same now when I take out the disc to retest. Maybe mine is pretty accurate without the correction?
Remember, it is only numbers. The light is just as bright as it was yesterday.
That said, yes, it does appear that the lights were sadly overrated even though that is the last thing I wanted.
If I could afford it I would offer refunds for all of them.
If someone is truly unhappy with any of my work please contact me and I will make it right one way or the other.
These spheres have thrown a wrench in the what we consider a lumen here on BLF. I know my numbers for a given light were actually on the low side when compared with others readings on here in many cases.
That said, I trust these new numbers with an actual standard to base them on much more. They are simply much much lower then expected.
I have been hesitant to retest my own lights, it really hurts to get take readings on lights I used to think were doing much better then they really are.