Interesting Lightbox Numbers...(Pics Added)

Hello All,

After seeing wquiles "poor man's integrating sphere", http://candlepowerforums.com/vb/showthread.php?t=111538, I decided to make something similar to measure overall output instead of just lux.Not knowing how my design would work I threw something together to see how it would go.It is by no means the quality of wquiles box,but I am very pleased with the results,so much so that I intend to re-make the box out of new material with the same dimensions.The material is white laminated MDF.Before anyone says so I am aware that this is by no means scientific,or even reflective of actual lumens,the results were very surprising though.All measurements were taken with the meter in the x10 Lux Range.

L2DCE P4

Advertised Lumens
Low 9
Med 40
High 80
Turbo 135

Lightbox Measurements
Low 16
Med 45
High 82
Turbo 135

I was so surprised to see the numbers so close I decided to charge the Eneloops and let them rest overnight,Then I repeated the tests.The results were surprisingly the same.Oh yeah,the meter is a Protek 6300.So what do you think.

Re: Interesting Lightbox Numbers...

It sounds like you've built a pretty good light box for the numbers to be so close. The low value was double the output, but the other levels were all close. Good job!

Re: Interesting Lightbox Numbers...

Question is, are Fenixs' numbers accurate?

Re: Interesting Lightbox Numbers...

I hopes that Fenix didnt use home made lightbox to measure advertised lumens

Re: Interesting Lightbox Numbers...

Good job, TexLite. Can you post a few pictures of your setup please.


:goodjob:

Re: Interesting Lightbox Numbers...

chevrofreak has approximate lumen numbers with his runtime graphs. Check out and see if he has done a runtime with same light and compare the numbers.

Bill

Re: Interesting Lightbox Numbers...

Hey Guys,
I'll try and get some pics tonight or tommorrow,I'll be busy most of the evening,I only had time to pop in for a minute.I dont remember exactly what Chevrofreak had measured but I know he did a review,I'll try to find it later when I get time.It wont really be that relevant though since its another light,the best thing to do is measure the same light.Irregardless of how Fenix measured the output(IS I'm sure),or how close the numbers are to real-world lumens,the most interesting aspect is how close the numbers are for ALL levels.I can understand them being bloated on high or turbo,but to be spot on in every level seems to point to some type of accuracy.I have also measured the output of some other lights,one of which was an original run L1P,rated at I think 40 real-world lumens(Fenix said 46 IIRC),I measured 33.Also a MKII,rated somewhere around 40 also I think,measured close to the Fenix,33 or 34.I'll get some pics as soon as possible,maybe even a video,I'll have to get the wife to help with that though.

Thanks for the possitive comments.

TexLite

Re: Interesting Lightbox Numbers...

I think that you will find that the diffferences between your lightbox numbers of approximate lumens and chevrofreaks will be revelant and maybe not just due to two different L2D CE P4's being used. Chevrofreaks runtime plots:

https://www.candlepowerforums.com/threads/155819

See NiMh's results.

Bill

Re: Interesting Lightbox Numbers...

I am by no means trying to claim my numbers are more accurate than Chevro's,in fact after looking I slightly more confident in saying the numbers are somewhat close to published specs.Thanks for the link.It goes without saying,(although I am saying)that Chevrofreak is a vital and highly respected part of CPF,(coming from someone who has learned an immense amount from his posts),so I think everyone would agree that his numbers could be said to be trusted as accurate.But they are still an estimation,I think coming from the formula everyone settled on when Doug was still around.Anyway,I trust Chevro,Peakbeam,and the others who have an immense amount of knowledge.Here's the thing,my meter always reads 6 even when you block all light,I dont know whether this should be taken as a defect or not,I'm at a loss,but perhaps I should start with 6 as the baseline.That would mean,

My Numbers,
Turbo 129
High 76
Med 39
Low 10

Chevrofreak's Numbers
Turbo 111-115
High 64-65
Med 30
Low 6.5

Taking into account that the P4's binning allowance is 80-87 Lumens,on top of the numbers being mostly within 10 of each other,and the fact that you have two different lights with two totally different setups,I would say it is surprising and cool that the numbers are so close.

Re: Interesting Lightbox Numbers...

I did not intend to depreciate your work. I wanted you to appreciate that it might be a possibility that your numbers might be very different relative to someone elses, and that differences might be revelant and get you to start looking at your procedure. That said, I did not know what chevrofreaks numbers were until I looked them up after my last post. In all honesty, your numbers did seem a little high, and is why I posted in the first place.

Congatulations on your work. You proabably will make adjustments in your baseline number, but that is part of the fun of it all. Maybe a diffferent lightmeter too. I will someday move toward making a lightbox as I have not moved away from simple bounce with a lightmeter.

Bill

Re: Interesting Lightbox Numbers...

TexLite said:

My Numbers,
Turbo 129
High 76
Med 39
Low 10

Click to expand...

Your adjusted numbers look great. Certainly close enough for most purposes, and better than published specs alone.

Re: Interesting Lightbox Numbers...

Thanks Guys,I appreciate all the comments.

Yeah,I think the box will work great for what I built it for,a way to accurately measure improvements from mods,with repeatable results.From now on I will post only adjusted numbers.

Kilovolt,I'm working on the pics,thanks for the interest.

Bill,no offence taken,thats the reason I posted,for feedback.

TexLite

Re: Interesting Lightbox Numbers...

Sorry for the wait,somehow I got it in my head that you had to have 20 posts in order to post pics??Well here are the pics of my setup,notice the 4th pic,the meter is reading 6 even though the flashlight isn't on.The rest of the readings therefore should be adjusted by 6.


Setup 1:


Setup 2:


Setup 3:


L2D-CE Off


L2D-CE Low


L2D-CE Medium




Continued...

Re: Interesting Lightbox Numbers...

L2D-CE High


L2D-CE Turbo



Thanks for looking

It seems to me that the LED manufacturers' specs for lumen output are a lot more likely to be realistic than flashlight manufacturers'. So I've taken a few LEDs from known bins and used their outputs when running at 350 mA to calibrate my Quickbeam type light box. I don't know how good it is, but I think this might be a fairly decent way to get into the ballpark.

c_c

It seems like this will be a great tool :thumbsup:

It seems like in comparing this to chevrofreak's graphs we should know how he tests lumens... I heard something about a formula, is that one applied to readings from a makeshift lightbox?

The doubled (not as much with the adjustment, but still very high) readings for the low setting can be easily explained: it is a know issue that with unintended battery types (or when switching from high mode?) many Fenix lights have a very high and brief low mode. So probably the reading weren't off at all

Looking forward to reviews with these readings :twothumbs

Congratulations, your setup is really first rate.

:twothumbs
:twothumbs
:twothumbs

Curious_character,
Agreed,the best way would be to take a known value and adjust accordingly,or even better,several known values.I will probably work on this later,but for now all I'm using it for are comparisons,so I think it will work for that.The most interesting aspect to me are the similarities with Fenix's published specs on all the levels.
Thanks for the input..


I came to the light...,
Thanks,I think it will be a great tool for measuring changes in output from emitter,reflector,and driver swaps,comparing lights and such.I wanted something reliable,repeatable and accurate,I would say it meets this goal.

I dont know what setup Chevrofreak has,but I think it's something like Quickbeams.I never felt comfortable in comparing my numbers with Chevrofreaks,wquiles,or anyone else's.Bill brought it up because my numbers were so much higher than Chevrofreak's,worthwhile because it caused me to go back and adjust the numbers.
Here are some links on the formula you asked about,there were some threads here on CPF,but I dont know where they are.
http://www.flashlightreviews.com/features/output_vs_throw.htm#lumens The Formula,
http://www.flashlightreviews.com/features/lightbox_output.htm Quickbeams Lightbox
Good suggestion on the flash,but the numbers were read after a five minute or greater warm-up to let everything settle in,the light does flash,but only for a split second.All of the readings are stabilized at the readings in the pics.


Kilovolt,
Thanks for the kind words,much appreciated...

My setup is actually spherical. I wanted to try to emulate a real integrating sphere as much as possible.



It's an old globe that I painted inside with about half a dozen coats of flat white paint. There's a baffle at the top to prevent light from directly shining onto the sensor.

chevrofreak said:

My setup is actually spherical. I wanted to try to emulate a real integrating sphere as much as possible.

It's an old globe that I painted inside with about half a dozen coats of flat white paint. There's a baffle at the top to prevent light from directly shining onto the sensor.

Click to expand...

Wow,nice setup.:thumbsup:
I had thought about fiberglassing over one of those Dollar Store bouncy balls,then cutting in in half and removing the ball and putting in a baffle.But I thought it would be to much trouble for as little as I'll use it.The old globe is a great idea.