Honestly until I am able to get some calibrated lights for more exact numbers (or if someone else has some like that), I would not really worry about getting it exact.
Just post the as-is numbers and maybe some numbers with whatever method you found got them in the right ballpark.
All I am looking for now is the average correction factor among the spheres at this point. So far it is looking well within my goal tolerance levels but more data is always good.
You are welcome to adjust your sphere however you want and to whatever you want for now. Once I figure out a fix then everyone can apply it and compare readings evenly (well, within ~10-15% or so, with everything factored in).
I will use fenix tk15 to calibrate temporary… i got 1430 lumen. Fenix got 1000 lumen.. so my correction factor will be .70
Fenix is legit in their lumen rating..
My 4” lumen tube correction factor is .70.
Very good, it looks like they are averaging out very good sphere to sphere. That is the hardest part by far so I am very pleased with those results thus far.
I got 2877 lumen @30 seconds for the lumintop odf30.. no plastic bag, and .70 factor has been counted.
Okay, so the raw number was 4111 I think. I remeasured mine at 30 secs with a Liitokala and got 4490 × .70 = 3143 so we are very close. Within 10%, which is well within Cree’s 14% tolerance.
Are you using a protected battery? I’m not sure if that will effect a boost driver, but it might. Our numbers might be even closer if we used the same battery.
—
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Got to remember that not only do you have the 14” tolerances of the LED to factor in but also the 1% (resistors are usually 1%) to 20% (capacitors, coils and other such components range from 10% to 20%) tolerances of the components on the drivers.
On top of that and many of those tolerances compound on top of each other. Particularly with buck and boost drivers as they have a lot of components that factor into the final output.
This is why using random lights off the assembly line is so inconsistent and hard to calibrate a sphere with.
Here are some fresh numbers from most of my lights. All CW.
Some interesting notes are that I lost 23% output on the X2R by switching the reflector to a 60° TIR lens. So to go from my headlamp measurements, which both use the same 60° lens, I need to add 30% which I did at the bottom.
The new Boruit 5K headlamp is kicking some butt. It’s rated at 5000 lumen, but that is Chinese lumen so completely unbelievable. I think the predecessor RJ02 was only about 300 or so lumen, but this one is an estimated 650 lumen. Not bad for a no name emitter.
Most of my lights match up to the JoshK Sphere by the same general percentage. The C8 was off quite a bit. I wonder if it’s due to the narrow beam.
The L5 and L6 are my custom lights which is why they are reading very high.
—
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Thanks you the info.. you should test fenix lights.. they have a higher lumen standard..
What does “lumen standard” mean?
Do you mean it’s more accurate to the rated output?
I never pay attention to rated output. I tend to go by emitter type, driver type and amp draw on battery. Then I compare it with emitter tests done by testers such as koef3, djozz, match and others. I have most of these charts in my head.
Then I calculate estimated driver, reflector and lens losses. Battery type can also play a factor. Anyway, I can pretty much tell you what it should be outputting and whether it’s more or less than the rated output. It’s not exact, but not bad for a back of the envelope calculation. Flashlight manufacturers can’t bend the rules of physics. Lol
The general rule of thumb is that most lights are below their rated output.
I never paid much attention to Fenix as they always cost a lot more than similar lights from other manufacturers.
—
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Hmm, that is possible. Although doesn’t look like he is very active, he has not logged in for almost 2 months.
Yea Randy at PflexPro haven’t been very responsive to emails in the last two months. He used to reply to every email right away. I hope it’s just because he’s taking a long vacation temporarily.
TA, I’ve yet to receive a response from either of the individuals I had contacted regarding using their spheres. Was hoping for better news for you. I know you are shipping your lights overseas, but I’ll keep you posted if anything changes.
OK, I redo my calculation with new Data from Maukka. I also take Maukka’s advice remove both Lumintop Tool AAA Copper & Tool AAA TIs’ DATA from the averge calibration#. Base on maukka’s review DATA; I simulation status best I can ( same model of lights with same CCT & use the same batteries). My new average calibration# for TA’s tube is 0.657939. I don’t trust any Mfg spec. I won’t recalibrate any integrating spheres with Mfg spec#.
Not sure I figured this properly (I’m not great with math) but it looks like my correction factor average is about .77 ?
We aren’t trying to find a specific number to make our readings match a manufacturers rated spec.
We want to know how far off the manufacturers specs actually are.
The bummer is that TA’s tube was calibrated to match a lot of other people’s lumen measuring devices on BLF and CPF. So people like DB Custom, Matt at Adventure Sport Flashlights and many others here were all measuring high.
We may have to scrap a bunch of old measurements from the last several years and start fresh based on guys like Maukka and certain TLF members, etc… who base everything off properly calibrated reference sources.
It sounds a bit scary.
—
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
OK, I redo my calculation with new Data from Maukka. I also take Maukka’s advice remove both Lumintop Tool AAA Copper & Tool AAA TIs’ DATA from the averge calibration#. Base on maukka’s review DATA; I simulation status best I can ( same model of lights with same CCT & use the same batteries). My new average calibration# for TA’s tube is 0.657939. I don’t trust any Mfg spec. I won’t recalibrate any integrating spheres with Mfg spec#.
!{width:120%}[img]https://i.imgur.com/FERCNMk.png[/img]!
Ok.. that look great.. now can you show us some lumen number for other stock lights? Skylumen lights? Thats way we know.
My lights are in Finland now. If we are lucky i’l have them by end of next Friday. And if i’m lucky i’l have my tube to test too
Did you measured them before sending? When Maukka has his lumen readings we can have a correction factor very quick.
Not quite. TA needs to recieve his measured sample lights and measure them across several lumen tubes. Any one particular tube could be in the plus 5% while another might be in the negative 5%. So we need multiple samples to find that middle ground. Then TA could tell us that correction number for all spheres.
I don’t know how good it is to have several sample lights measured from different people. I suppose if they are all measured against a certified standard calibration lamp, it should be fine.
If any one person wants to get their lumen tube even more accurate, these measured sample lights might be able to be shipped to customers with instructions.
I would be willing to give TA a deposit for the light and pay for shipping so I can find my particular lumen tubes exact correction number. Then the +-5% might shift down to +-1% or 2%. I could even ship it to the next person wanting to use it. I don’t know if TA or Micael would want to do this, though.
It may not even be needed. Of the 35-40 tubes made, I think statistically most are already in that +-1% or 2% range. The rest falling into the wider ranges. So passing a light around might be a waste of time. ???
—
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
I would be willing to give TA a deposit for the light and pay for shipping so I can find my particular lumen tubes exact correction number. Then the +-5% might shift down to +-1% or 2%. I could even ship it to the next person wanting to use it. I don’t know if TA or Micael would want to do this, though.
It may not even be needed. Of the 35-40 tubes made, I think statistically most are already in that +-1% or 2% range. The rest falling into the wider ranges. So passing a light around might be a waste of time. ???
For me right now I’d have no real need for the calibration light because TA is going to figure out what to for do re-calibration (something better than paper sheets / plastic bags) then he will send us the part to install or tell us what to do.
After that I could see maybe passing the light around to confirm accuracy of the fix, if TA recommends doing that, at that point I would be ok with paying for shipping the light to the next member.
I decided to test out some different mediums across the TA Tube sensor to see if they are consistent.
I used my S2+ on medium measuring from 210 to 225 depending on whether I turned it off and back on. So I would leave it on and keep getting reference levels to eliminate any discrepancies.
I tested one layer of the wax tissue paper that I talked about earlier. I measured the lumen drop relative to the reference brightness level. I tested 10 different sections of a single sheet. Here are all 10 results showing the lumen drop in lowest to highest order:
24 24 27 27 27 32 35 35 37 37
Almost a 50% difference. So not very consistent, especially when doing multiple layers.
Next was a Walmart white plastic bag. I just did 4 samples from all over the bag. No printed sections and no folds. Here is the drop in lumens:
37 41 45 50
About a 35% difference. So again, not very consistent. (On average this particular white plastic bag gives me a correction factor of .80)
Lastly I did the same test as above using a Kroger grocery brown plastic bag. Here is the drop in lumens:
72 78 79 82
About a 14% difference. (On average this particular brown plastic bag gives me a correction factor of .65)
I think the trend is that the more light a medium blocks, the more consistent it is.
As I find more materials laying around the house I’ll try and give some accurate samples from it.
I think the bottom line is that when TA announces his official correction factor we will need to find some type of paper, plastic, etc… medium that drops the readings that exact amount and then not mess with it. For example, if different sections of the same plastic bag can vary so much, then we will need to cut out a specific section to use and reuse. Then not get it mixed up with a different bag or different section of bag. See what I mean?
Then we will be back to direct lumen readings right off the meter.
Edit: Or TA might mail out a medium.
—
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
I assume TA is going to figure out something better than paper sheets / plastic bags etc. and send us the part that can then be permanently installed. (or tell us what/where to buy it)
Honestly until I am able to get some calibrated lights for more exact numbers (or if someone else has some like that), I would not really worry about getting it exact.
Just post the as-is numbers and maybe some numbers with whatever method you found got them in the right ballpark.
All I am looking for now is the average correction factor among the spheres at this point. So far it is looking well within my goal tolerance levels but more data is always good.
You are welcome to adjust your sphere however you want and to whatever you want for now. Once I figure out a fix then everyone can apply it and compare readings evenly (well, within ~10-15% or so, with everything factored in).
Texas Avenger Driver Series
My LED Test series - XP-L2 V5 - Nichia 219C 90+ CRI - Latticebright "XM-L" - XHP35 & PWM efficiency - XHP50 - XP-L V5 - XM-L2 U2 - XP-G3 S5 - XP-L HI V2 - Oslon Square & direct comparison to Djozz tests - Nichia 319A - Nichia 219B 9080 CRI - Nichia 219C D320 - Nichia 229AT - XHP70.2 P2 - XHP50.2 J4 - Samsung LH351D
Easy comparison tool for all my LED tests
I will use fenix tk15 to calibrate temporary… i got 1430 lumen. Fenix got 1000 lumen.. so my correction factor will be .70
Fenix is legit in their lumen rating..
My 4” lumen tube correction factor is .70.
You need more faith in Newlumen.
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
I got 2877 lumen @30 seconds for the lumintop odf30.. no plastic bag, and .70 factor has been counted.
So that is with or without the correct factor being applied? I assume that is after just not sure what the “counted” means in this context.
Texas Avenger Driver Series
My LED Test series - XP-L2 V5 - Nichia 219C 90+ CRI - Latticebright "XM-L" - XHP35 & PWM efficiency - XHP50 - XP-L V5 - XM-L2 U2 - XP-G3 S5 - XP-L HI V2 - Oslon Square & direct comparison to Djozz tests - Nichia 319A - Nichia 219B 9080 CRI - Nichia 219C D320 - Nichia 229AT - XHP70.2 P2 - XHP50.2 J4 - Samsung LH351D
Easy comparison tool for all my LED tests
Very good, it looks like they are averaging out very good sphere to sphere. That is the hardest part by far so I am very pleased with those results thus far.
Texas Avenger Driver Series
My LED Test series - XP-L2 V5 - Nichia 219C 90+ CRI - Latticebright "XM-L" - XHP35 & PWM efficiency - XHP50 - XP-L V5 - XM-L2 U2 - XP-G3 S5 - XP-L HI V2 - Oslon Square & direct comparison to Djozz tests - Nichia 319A - Nichia 219B 9080 CRI - Nichia 219C D320 - Nichia 229AT - XHP70.2 P2 - XHP50.2 J4 - Samsung LH351D
Easy comparison tool for all my LED tests
Okay, so the raw number was 4111 I think. I remeasured mine at 30 secs with a Liitokala and got 4490 × .70 = 3143 so we are very close. Within 10%, which is well within Cree’s 14% tolerance.
Are you using a protected battery? I’m not sure if that will effect a boost driver, but it might. Our numbers might be even closer if we used the same battery.
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
Got to remember that not only do you have the 14” tolerances of the LED to factor in but also the 1% (resistors are usually 1%) to 20% (capacitors, coils and other such components range from 10% to 20%) tolerances of the components on the drivers.
On top of that and many of those tolerances compound on top of each other. Particularly with buck and boost drivers as they have a lot of components that factor into the final output.
This is why using random lights off the assembly line is so inconsistent and hard to calibrate a sphere with.
Texas Avenger Driver Series
My LED Test series - XP-L2 V5 - Nichia 219C 90+ CRI - Latticebright "XM-L" - XHP35 & PWM efficiency - XHP50 - XP-L V5 - XM-L2 U2 - XP-G3 S5 - XP-L HI V2 - Oslon Square & direct comparison to Djozz tests - Nichia 319A - Nichia 219B 9080 CRI - Nichia 219C D320 - Nichia 229AT - XHP70.2 P2 - XHP50.2 J4 - Samsung LH351D
Easy comparison tool for all my LED tests
Here are some fresh numbers from most of my lights. All CW.
Some interesting notes are that I lost 23% output on the X2R by switching the reflector to a 60° TIR lens. So to go from my headlamp measurements, which both use the same 60° lens, I need to add 30% which I did at the bottom.
The new Boruit 5K headlamp is kicking some butt. It’s rated at 5000 lumen, but that is Chinese lumen so completely unbelievable. I think the predecessor RJ02 was only about 300 or so lumen, but this one is an estimated 650 lumen. Not bad for a no name emitter.
Most of my lights match up to the JoshK Sphere by the same general percentage. The C8 was off quite a bit. I wonder if it’s due to the narrow beam.
The L5 and L6 are my custom lights which is why they are reading very high.
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
Thanks you the info.. you should test fenix lights.. they have a higher lumen standard..
I Do —————-NewLumen is a Good Dude
What does “lumen standard” mean?
Do you mean it’s more accurate to the rated output?
I never pay attention to rated output. I tend to go by emitter type, driver type and amp draw on battery. Then I compare it with emitter tests done by testers such as koef3, djozz, match and others. I have most of these charts in my head.
Then I calculate estimated driver, reflector and lens losses. Battery type can also play a factor. Anyway, I can pretty much tell you what it should be outputting and whether it’s more or less than the rated output. It’s not exact, but not bad for a back of the envelope calculation. Flashlight manufacturers can’t bend the rules of physics. Lol
The general rule of thumb is that most lights are below their rated output.
I never paid much attention to Fenix as they always cost a lot more than similar lights from other manufacturers.
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
Yea Randy at PflexPro haven’t been very responsive to emails in the last two months. He used to reply to every email right away. I hope it’s just because he’s taking a long vacation temporarily.
Trust me. If thrunite, acebeam, lumintop, olight use the fenix integration sphere, their number will be much lower..
I tested my fenix pd35 at 30 seconds.. i got 1415 lumen x .70 = 990.5 lumen.. fenix rated 1000 lumen..
So its safe to say my correction factor is .70 or so.
Good info. .70 is looking like a good number so far but gonna be interesting to see a calibrated light to compare.
Texas Avenger Driver Series
My LED Test series - XP-L2 V5 - Nichia 219C 90+ CRI - Latticebright "XM-L" - XHP35 & PWM efficiency - XHP50 - XP-L V5 - XM-L2 U2 - XP-G3 S5 - XP-L HI V2 - Oslon Square & direct comparison to Djozz tests - Nichia 319A - Nichia 219B 9080 CRI - Nichia 219C D320 - Nichia 229AT - XHP70.2 P2 - XHP50.2 J4 - Samsung LH351D
Easy comparison tool for all my LED tests
My lights are in Finland now. If we are lucky i’l have them by end of next Friday. And if i’m lucky i’l have my tube to test too
Micael
Texas_Ace BLF Calibrated Lumen tube
TA, I’ve yet to receive a response from either of the individuals I had contacted regarding using their spheres. Was hoping for better news for you. I know you are shipping your lights overseas, but I’ll keep you posted if anything changes.
Not sure I figured this properly (I'm not great with math) but it looks like my correction factor average is about .77 ?
Using .70 correction factor the 2 sheets Deli paper is almost spot on for correction.
Model
TA tube @ 30 seconds
Mfg spec.
% of mfg spec
difference %
.70 correction figure
2 sheets Deli paper
Eagletac D25C2 (XP-G2)
625
530
84.8
17
437
435
Eagletac MX25L3 (MT-G2)
3350
2375
70.8
1.3
2345
2360
Jetbeam E20R
1290
990
76.7
8.8
903
895
Nitecore HC30
1453
1000
68.8
1.8
1017
995
Jaxman E2L 3*Nichia
895
720
80.4
1.7
626
629
Sofirn C8F v2
3340
2580
77.2
9.4
2338
2510
ThruNite TC20 (NW)
4240
3800
89.6
22
2968
2960
BLF Q8
6950
5000
71.9
2.7
4865
4910
Convoy C8 7135*8 HI 1A
1148
900 (est)
78.4
10.8
803
786
Average>
77.62222222
"Over 2000000 hours (about 200 years) standby time" (DQG Tiny 4th)
"27,157 results for zoomable flashlight" (ebay)
OK, I redo my calculation with new Data from Maukka. I also take Maukka’s advice remove both Lumintop Tool AAA Copper & Tool AAA TIs’ DATA from the averge calibration#. Base on maukka’s review DATA; I simulation status best I can ( same model of lights with same CCT & use the same batteries). My new average calibration# for TA’s tube is 0.657939. I don’t trust any Mfg spec. I won’t recalibrate any integrating spheres with Mfg spec#.
We aren’t trying to find a specific number to make our readings match a manufacturers rated spec.
We want to know how far off the manufacturers specs actually are.
The bummer is that TA’s tube was calibrated to match a lot of other people’s lumen measuring devices on BLF and CPF. So people like DB Custom, Matt at Adventure Sport Flashlights and many others here were all measuring high.
We may have to scrap a bunch of old measurements from the last several years and start fresh based on guys like Maukka and certain TLF members, etc… who base everything off properly calibrated reference sources.
It sounds a bit scary.
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
Did you measured them before sending? When Maukka has his lumen readings we can have a correction factor very quick.
Ok.. that look great.. now can you show us some lumen number for other stock lights? Skylumen lights? Thats way we know.
Micael hasn’t received his TA lumen tube yet. I just sent him the numbers, but you still have to wait a bit.
Not bad beam. i got 4662 lumen @30 seconds for the the stock blf q8. Samsung 30q fully charged..
TA’s calibration lights (S2+ 8×7135 XM-L2 3B and BLF348 219B) are measured and ready to ship as well.
Not quite. TA needs to recieve his measured sample lights and measure them across several lumen tubes. Any one particular tube could be in the plus 5% while another might be in the negative 5%. So we need multiple samples to find that middle ground. Then TA could tell us that correction number for all spheres.
I don’t know how good it is to have several sample lights measured from different people. I suppose if they are all measured against a certified standard calibration lamp, it should be fine.
If any one person wants to get their lumen tube even more accurate, these measured sample lights might be able to be shipped to customers with instructions.
I would be willing to give TA a deposit for the light and pay for shipping so I can find my particular lumen tubes exact correction number. Then the +-5% might shift down to +-1% or 2%. I could even ship it to the next person wanting to use it. I don’t know if TA or Micael would want to do this, though.
It may not even be needed. Of the 35-40 tubes made, I think statistically most are already in that +-1% or 2% range. The rest falling into the wider ranges. So passing a light around might be a waste of time. ???
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
For me right now I’d have no real need for the calibration light because TA is going to figure out what to for do re-calibration (something better than paper sheets / plastic bags) then he will send us the part to install or tell us what to do.
After that I could see maybe passing the light around to confirm accuracy of the fix, if TA recommends doing that, at that point I would be ok with paying for shipping the light to the next member.
"Over 2000000 hours (about 200 years) standby time" (DQG Tiny 4th)
"27,157 results for zoomable flashlight" (ebay)
I decided to test out some different mediums across the TA Tube sensor to see if they are consistent.
I used my S2+ on medium measuring from 210 to 225 depending on whether I turned it off and back on. So I would leave it on and keep getting reference levels to eliminate any discrepancies.
I tested one layer of the wax tissue paper that I talked about earlier. I measured the lumen drop relative to the reference brightness level. I tested 10 different sections of a single sheet. Here are all 10 results showing the lumen drop in lowest to highest order:
24 24 27 27 27 32 35 35 37 37
Almost a 50% difference. So not very consistent, especially when doing multiple layers.
Next was a Walmart white plastic bag. I just did 4 samples from all over the bag. No printed sections and no folds. Here is the drop in lumens:
37 41 45 50
About a 35% difference. So again, not very consistent. (On average this particular white plastic bag gives me a correction factor of .80)
Lastly I did the same test as above using a Kroger grocery brown plastic bag. Here is the drop in lumens:
72 78 79 82
About a 14% difference. (On average this particular brown plastic bag gives me a correction factor of .65)
I think the trend is that the more light a medium blocks, the more consistent it is.
As I find more materials laying around the house I’ll try and give some accurate samples from it.
I think the bottom line is that when TA announces his official correction factor we will need to find some type of paper, plastic, etc… medium that drops the readings that exact amount and then not mess with it. For example, if different sections of the same plastic bag can vary so much, then we will need to cut out a specific section to use and reuse. Then not get it mixed up with a different bag or different section of bag. See what I mean?
Then we will be back to direct lumen readings right off the meter.
Edit: Or TA might mail out a medium.
Texas Ace Lumen Tube and JoshK Sphere calibrated with Maukka lights
Click this to go to signature links.
I assume TA is going to figure out something better than paper sheets / plastic bags etc. and send us the part that can then be permanently installed. (or tell us what/where to buy it)
"Over 2000000 hours (about 200 years) standby time" (DQG Tiny 4th)
"27,157 results for zoomable flashlight" (ebay)
Pages