Lumens from Lightbox readings - more accurate!

Hey all - just thought I would post a little update for those of you who know what this is all about.

At one point we thought that multiplying 1.43 times the Overall Output numbers in my charts (on FlashlightReviews.com) would give a close approximation to Lumens.

After some glaring discrepencies, I removed the reference to this calculation on my site.

Recently I had the opportunity to go over the numbers from the Lightmeter Benchmarking test that we did, and since I took part, AND did the Overall Output testing of the lights that were passed around, AND those same lights were fully regulated, AND they were tested by a reputable company for Lux and Lumens, I was able to get a much more accurate Lumen calculation factor. So, here it is:

LED lights, multiply the Overall Output chart number x 1.39
Incand. lights, multiply the Overall Output chart number x 1.62

I don't know why the're different, but after checking out Peter Gransee's list of Integrating Sphere tested lights, it looks like they're darn close, or right on the money. I didn't calculate the variation, but they were usually within a few lumens - I was really surprised.

Check it out and let me know what you think or if you see any glaring discrepencies!

Chart of Overall Output numbers

Doug P.

Doug-

Very interesting. Can we assume, (even though we all know about assuming), that those ratios are different from LED to incan because a lux meter responds differently to LED or incan light. Maybe this has to do with response of a lux meter to wavelength.

If the lux meter was accurate across the spectrum, than the quick conversion ratio to lumens would be the same regardless of source type.

The mean ratio is 1.505, which is easy to do in your head for a rough comparison.

Mark

I am glad that you did not use lumen output of Benchmark A2 as tested by LSI as a baseline. Tom and I agreed that the output of the A2 had probably deminished through the months of testing by Benchmark folks. I know that my A2 has lost some of its overall output as evidenced by my bounce tests compared to other lights. The Benchmark led lights are more stable in output, over time, compared to A2 incan.

Bill

:goodjob: Thanks, Doug!

Flash: That's my thinking exactly. The sensor responds differently, so we need a different conversion factor.

Bullzeye: Actually I DID use the Lumen test for the passed around A2. However, I was near the end of the passaround, and the Lumen test was done AFTER the passaround was done, so it's probably pretty close. When I checked the calculation against Gransee's numbers for incandescent lights, they were all pretty close except the Scorpion... until I checked their graph.

They used the Lumen reading at 2 minutes of burn-in, which was in the 70's. My calculation shows that my tested sample would have been in the 90 Lumen range. My readings are always taken at the start with new batteries, not after 2 minutes of burn-in. Their graph shows that at the start of their testing, their Scorpion was reading around 90 Lumens, and dropped to the 70's when they took the "official" reading 2 minutes later, so we actually do have a match - both methods find the Scorpion produces around 90 Lumens at the start with new cells. :thumbsup:

Sweet! This is awesome to know. I was wondering why the calculation dissappeared from the website. But now i know why. Are you going to put the calculations back onto the website?

One thing I don't exactly get....

On FLR the SL Strion is rated at 50, so that's 81 Lumens.
The Surefire E2D is rated at 40, so that's 65 Lumens.

I have both, and on fresh batteries when doing the ceiling bounce test, the E2D puts out noticeably more light. This was done in different rooms at different angles and everytime the E2D won. I cleaned the contacts and charged the Strion fully before doing the tests.

What gives?!?!?

Is it possible that the tighter beam of the Strion yields a higher value in the "Lightbox"?

The Strions I've seen would clean the E2D's contacts! Clock, this is. Strion LiIon problem, olde lamppe?

Thanks, Doug, for the detailed and latest listing. Your site is a treasure.

The Strions I've seen would clean the E2D's contacts! Clock, thati is. A Strion LiIon problem, olde lamppe, perhaps?

Thanks, Doug, for the detailed and latest listing. Your site is a treasure.

Flash, I agree with that as well and also assume that for coloured lights such as red LEDs, the conversion ration will be different from white LEDs.

Thanks alot Doug! It is very much appreciated that we can get a good indication of how much light a light emitting device puts in the real world. (as opposed to the not so real advertising stratagies world). Personally, I'd more likely trust your readings than readings from most manufacturers.

:thanks:

I have an adapter to put the strion bulb in my E2e, and it puts out more light than the surefire bulb. I'm using a protected 17670 for the strion, and 2 CR123a for the MN03. My observations would match up pretty closely to what FLR gets, so i'm not sure what gives on your setup.

schiesz

Shifty: :shrug: I don't know... I'd have to play around with it. Or as the posts above hint, the problem could be on your end...

Glad to help, everyone!

That is really handy to know, thanks Doug. i did follow your insructions on your website awhile ago and built myself a lightbox apparatus as well. Very handy to have and now even handier with the more accurate conversion factors. I must say though, your lux lottery chances are much better than mine! I only have one LED light where the OA numbers corresponds to your reported numbers. The rest are usually 1 or 2 OA units behind haha.
:goodjob:

Hello Doug,

I am glad you were able to use the test results to check the calibration of your light box. I agree that the differences you are seeing is a result of the light meters response to the LED vs incandescent light sources.

Tom

I did a fairly wide search of lux meters. Although I won't claim it is definitive, I found one meter that addressed this issue.

It is a no-name meter on eBay. Lux Meter

It contains the following note:Calibrated to standard incandescent lamp 2856k.

This would be a nice factor to know with any meter and a correction table based on different temps would be very useful.

Mark