well, I used the max lumen at 966, and 918 at 30 seconds. Using that for my conversion I get a cal factor of 0.44. Mine is a WW and not a NW, but it should be close. If anything my calculation factor needs to be a little lower, say around .4 to .42
The A60’s have a XR-E that is supposed to be 310 lumen so my measurements are almost dead on.
The M24 has a believable lumen number.
If I throw out my one set with the T20 #1 where it was getting really high numbers with the panasonic batteries, they are all about the same and pretty close to what you had for your stock T20.
what do you think?
you mentioned Candelas is calculated based on:
(distance in feet to meter)² x .09290304 x (lux measurement)
There is no way to extrapolate this data correct? I actually have to use my meter to calculate candelas at say 10m?
You have to make a lux measurement at a distance, the lux and distance go into that equation and it’ll spit out your number. There is no accurate way to extrapolate a candela measurement from the reading you get in the light pipe. Distance to the meter does not matter too much as long as you’re a couple meters away or more (for really big fixed reflector lights it may need to be 5 meters or more to be the most accurate), just make sure you measure the distance.
If my numbers are any good, and I think they’re pretty close, yours are in the same ballpark. Close enough for you to use the numbers. Just remember to list them as approximate lumens if you use the numbers here, I’ve taken to calling them “measured lumens in pipe” so no one gets the idea that I’ve got a fancy calibrated IS.
My pipe looks almost just like that, it’s a dedicated PTrap though, not 3-90’s. Same meter. Manxbuggy1/rdrfronty built it for me, calibrated it with about 25 of their lights that were tested in their own pipe that was calibrated with over 40 lights. They bought quite a few high end lights like the Fenix and Nightcore and Eagle Tac that have ANSI rated outputs, those were heavily used to get their own pipe calibrated.
Mine has a conversion factor of .345. Tom E’s is also made by them and is (I think) .35 and even Richard uses one they made.
The absolute highest I’ve seen to date is my BTU Shocker with 3 XHP70’s and using (not recommended) 3 Panasonic PF cells. (I know, 12.6V into 3 6V emitters in parallel, scary for me, but I got a measurement) This was over 40,000 on the meter for some 16,000 out the front lumens!!!
I’m using 2 LG HE-2 cells in it, with the 3rd spot in the carrier having a 12Ga Turnigy wire soldered in, and getting 12,144 lumens out the front with a reading on the meter of 32,200 35,200. (Hope I’m remembering that right… I know it’s 12,144 but not 100% sure of the meter reading and am not at my own computer) This bad boy actually has about a 10 lumen moon mode as well. :bigsmile:
The lowest reading I’ve seen is using ToyKeeper’s firmware on a driver by Wight, the A17DD Hybrid-S with a single 7135 chip running the moon mode at 0.31 lumens.
So now y’all have me wondering…what if that sheet glass were replaced with a UCLp AR lens from flashlightlens.com? That regular glass simply cannot have the greatest light transmission and is almost assuredly costing us 10% or more right out of the gate. Maybe 10% isn’t much on a 600 lumen light, but what does that translate to on a 12,000 lumen light?
Gonna get me some UCLp……
Edit: (went and double checked notes….the meter read 35,200)
Hello, I have just purchased all of the parts to build one of these light pipes except the light meter.
It seems that most people on here have the same light meter. Is that for any particular reason or because of cost?
If you were going to get a light meter for this light tube which would you pick in different price points? $10-25, $25-50, $50-75… etc. What benefits would I get out of a more expensive light meter? Are those benefits worth it?
Out of curiosity, is there a reason why everyone went with 2 sections of pipe? I am currently building a similar design, but I found 1 section still allows the head of the light to not have a line of sight of the sensor on my lux meter.
Because PVC is a bit shiny, I believe that the purpose is to limit the chance of a direct reflection. Based on some testing I just did, my light pipe with a sanded interior is nearly the same correction for floody lights and throwers(.52xlux for flood, .48xlux for throwers). Without the sanding the correction factors may have been further apart.
Actually, that reminds me. Anyone who has a light pipe who wants to be involved in the calibration thread djozz and I have started, you should make a post in this thread.
I had initially wanted to scrub the insides with acetone to dull the finish but never got around to it. The shininess does in fact contribute to a slight advantage for throwers. Just a bit busy now but do plan on modding my LMD in the future.
Hi,
I am recently interested in flashlights. One problem is that I don’t have any equipment to measure their outputs. This idea using PVC pipe/90 degree elbow sounds really attractive to me. However, since I don’t have anything to calibrate, is there anyway I can estimate the lumen from lux by some analytical or empirical formulas? As long as I can have a ball park number, I don’t mind some errors.
Another questions is what size of pipe/elbow should I use. Would 4” or 3” better than 1.5” or 2”?
Really appreciate your help. Thanks in advance.
Here’s what you do. Find a light with a known and verified ansi rating. I use a Fenix, but there are others out there. So if it’s 117 lumens…. you divide 117 lumens by the lux on your meter… it’ll give you something like this .25, .03, .59. Then you have your multiplier. Then when you put in an unknown light, whatever the lux is, multiply the .25 (or whatever) by the lux and you have your rating. Formula example- 117lumens/890lux= .1314 There are some problems though. Different styles of reflectors or optics can skew the ratings. So if you’d like to test many types of lights, it’s best to have a variety of styles of lights with known lumen figures. Like a big reflector thrower, a floody EDC, a bare emitter light- etc. The better measuring device you build, the more consistent your figures are.
I use 4 inch for mine. I’m using two 90 bends right now, but might go with a third. Some people have suggested sanding or find a method to make the interior less reflective. Painting the outside of the pipes black is also helpful. You need a lux meter, of course too. This model seems to be popular, mainly because it’s cheap.
mhanlen, thanks!
I looked the Fenix lights, they are ansi certified but didn’t see they are verified. May be you can get me a link to look?
I don’t know if there’s any way to get an idea about the mapping from lux to lumen without a calibrated light. It will be really helpful if someone can publish their “exact” setup and the test results, so if I follow them as close as I can, I will have some better idea about my lux readings.
The illusive ANSI lumen scale - no one really knows what it is, and sadly all the different manufacturers seem to have their own interpretation of it, and the differences are huge (IMHO). Here’s a good example, this independent reviewer has calibrated on “Fenix lumens” and shows nearly identical output/runtime graphs between a D25A and SC52 (I did the same test with the same result). ET specs 75lms x 2.5hrs, ZL specs 108lms x 3hrs, Selfbuilt finds the SC52 to be 120lms x 2.5hrs and on Fenix scale, they’re in the 94-98 lm range.
There’s only one reviewer I’ve seen that does claim ANSI accuracy (ti-force), with laboratory tested calibration lights, and his scale would put these graphs in the 84-88 lm range. This scale also happens match the high-quality US manufacturers like Surefire, HDS and Malkoff, so I think it’s right, but it’s pretty far off from the likes of Selfbuilt, ZL, AT, TN, etc.
In the end, just pick a calibration light and start working with it. As you collect and test more lights, you’ll change calibration lights and develop your own scale that matches most of your collection. Although I’m not fond of Selfbuilt’s liberal scale, his relative measurements are good so you’ll develop an adjustment factor to translate, and be able to compare different lights you don’t own.
The one thing I would highly suggest for calibration, is to use lower output modes, or step down outputs, and NOT max output. Max is highly dependent on battery quality, age, and state of charge, not to mention more volatile sample to sample, and generally have steeply sloping curves (so highly dependent upon the point captured).
I wonder how close the BLF A6’s will be to each other when they are in the 7135 only modes, there will soon be hundreds of them going out to members here.
reppans its funny that you mention ti-force. I was just getting ready to mention him myself in another thread. He is actually the guy that taught me how to measure lumens in the first place.
He did the original tests on my shorty maglite, and I have a light that I use that he tested years ago I have always kept for calibration of my own sphere.
They do provide 'some' output standard, but the leds alone will vary by up to 7 or 8%, switches cause varying resistance etc., but worse is that as I understand, Toykeeper's user interface does not contain a mode with just the 7135 on 100%.
