My "Light Box" or poor man's Integrating Sphere ...

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
My "Light Box" or poor man's Integrating Sphere ... with fan cooled HS!

The most amazing thing about this project is that I have been thinking about it since last summer. It was not until tonight that I got it running for it's intended, main purpose, so basically, on and off, I have been working on this project for over 6 months !!!

The main goal was to find a way to test LED's before I put them in lights. We all know that the LED's have certain bins of tint, voltage, and intensity, so many times we modders finish a light only to find too late that the LED was not all that awesome :(

Although I am not claiming that I met all of my design goals, this is what I was trying to do:
1) This Light Box (LB from now on) was conceived as my own way to test LED's against each other in a very repeatable way.
2) The LB was designed to be inmune to outside light so that I can run these tests anywhere in my house, day or night.
3) The LB had to use a power source that could be changed, so that if keeping the LED the same, I could test which driver and/or battery source would work best.
4) The LED had to be thermally stable, for at least the 30-60 seconds of a test run. The idea of course is that the testing would not damage/kill the LED under test.
5) Although soldering wires to a STAR would have been fairly easy, I wanted this setup to also work with bare emiters.
6) The LED "station" should be removable so that I could also test flashlights, even if not in the most perfect way (the LB is affected by the reflector/lens).
7) I could use almost any light meter. I have two different ones, so I made sure I would be able to test with both.
8) The LB must be fairly heavy/stable, and capable of working for years and years.
9) The LB must be movable/portable so that I could take to events and share with other flashaholics.

So here are some construction steps. Everything is glued with biscuits for long term stability/strenght:
IMG_1571.JPG


IMG_1574.JPG


IMG_1580.JPG


IMG_1589.JPG


All internal panels were sanded in between coats of white paint (4 layers), and then 2 layers of clear acrilic. The surface is not like a mirror, but it is very flat and even. Here is the outside, removable "house" prior to being glued permanently:
IMG_1591.JPG


Aftern being glued, I used some wood dowels to make sure that the house would line up exactly in the same place every time it was used. That inner piece helps reflect more of the light and prevents ANY light from hiting the light meter directly. This LB only measures reflected light:
IMG_2854.JPG


I made one "port hole" on the side (which when covered prevents any light from coming in) which allows me to see the LED or flashlight being tested, and allows me to see the tint of the LED which I can even photograph for comparing to other LED's later on:
IMG_2811.JPG


IMG_2847.JPG


IMG_2810.JPG



Here is not-completed-yet way in which I am bringing external power to the LED:
IMG_2851.JPG



Here is the completed setup. Yes, that is a modamag copper heat sink that has been "modified" to work in the LB. I took copper wires and trimmed them into the hole and used artic alumina to create a "fairly" even surface for the LED:
IMG_3752.JPG


IMG_3753.JPG



So when you are ready to test an LED, use a little thermal grease:
IMG_3754.JPG


IMG_3755.JPG



And then the two holders apply spring presure right onto the legs of the emiter to make electrical contact. For bare emiters, I first put a little grease on a generic star that I got some bare emiters I purchased earlier, and then put a little thermal grease between the star and the emiter. The spring presure keeps everything with the right thermal path, so I can test both stars and emiters on this setup:
IMG_3756.JPG



A little dark since I am underdriving this LED, but here is the working setup:
IMG_3758.JPG


IMG_3760.JPG


IMG_3761.JPG


IMG_3763.JPG



and with the house in place sealing outside light:
IMG_3765.JPG


IMG_3764.JPG



The mating surfaces between the base and the house have thin foam pieces which seal the light when compressed by the short elastic bands:
IMG_3766.JPG



Same setup above with the port open and with overhead lights:
IMG_3767.JPG


IMG_3768.JPG



and with the lights in the garage off:
IMG_3771.JPG



So does this qualify me as a flashaholic or what !!!

Will
 
Last edited:

CLHC

Flashaholic
Joined
Dec 25, 2004
Messages
6,001
Location
PNW|WA|USA
wquiles said:
So does this qualify me as a flashaholic or what! ! !
:huh: I would say so! Look at that detailed work you put into it! Way to go Will! :wave:
 

LEDcandle

Flashlight Enthusiast
Joined
Aug 15, 2005
Messages
1,943
Location
Mushroom S'pore
Looks like a great setup and thanks for the detailed pictorial article!!!
You aren't just a flashaholic, you're a flashaholic*! :laughing:
 

ICUDoc

Enlightened
Joined
Aug 20, 2004
Messages
907
Location
Sydney, Australia
wquiles

Great job!
Now we look forward to some data-how do you plan to calibrate it?
Do your light meters have a linear response so a single measurement of a known source will do the trick? I am still coming up with peak lux off my light meter as my only way of assessing differnet LEDs so your setup looks fantastic!
 

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
Thank you guys :eek:

The idea is to try to find an LED that has been tested already, preferably on a real IS, so that I can calibrate my setup. I already got a 750mA and a 1000mA constant current Downboy converters ready, so I can test one LED against another in a repeatable and controlled fashion. But without a "calibrated" LED with a known lumen output, all I can measure are relative numbers/units, like LBU - "Light Box Units" :D

I have many LED projects that I have not finished becuase I have been waiting for this LB to be finished. I will probably start going through my "collection" of LED's and start testing them for intensity, vf, and tint shortly. I will try to setup my camera so that I can take photos of the LED's through the port hole so that I can go back later and remember how they looked like.

I plan to keep a small database of my tested LED's so that I can pick/choose LED's for particular setups. For example, my current idea is to use the higher lumen LED's for single LED lights and the lower lumen LED's should be perfect for multi-LED lights (like on a M*g platform) where I can pick LED's of similar output and/or tint.

I can also use this setup to test for lumen vs. vf, which could be really exciting now that we have more "W" intensity, "S" vf 5W LED's - you could really hand pick a high lumen, very low vf 5W for that "special" light :)

Having this setup really opens many possibilities for LED projects ;)

Will
 

ICUDoc

Enlightened
Joined
Aug 20, 2004
Messages
907
Location
Sydney, Australia
wquiles

Yes I know what you mean- for that time when you have a great host and want the best "sub-bin" LED for a really special torch. Good luck!
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello Will,

I believe you are next in line to check out the CPF Light Meter Benchmarking Lights. You can look through that thread to see what Don came up with for them in his integrating sphere, and at the end of passing the lights around, we will be shipping them of to a certified laboratory to get the official word on those lights.

That should give you a good comparison of several different colors to start things off.

Tom
 

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
Yes, in fact, that is why I got my but in gear and finished the LB. With these test lights coming my way, this would be an excellent way to "calibrate" my LB to several known flashlights ;)

I will test all lights with my two meters and I will keep track of my results in my own "database" :D

Will
 

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
I just finished recording the lights used for the light meter experiement which will later be sent to be used in a real IS. This is at least some data that I will use to more-or-less know what to expect of the light box when testing flashlights, even if the primary design goal was to test LED's ;)

The first two sets of numbers were recorded using the reference lights in the usual manner, recording light directly aimed at the meters. The last two repeat the same tests, but using the meters to record reflected light inside of the Light Box.

Test #1 - Meterman LM631 Light Meter, Serial 0503014
Recorded on 3/21. Temperature 61F

Light=Measurement (Lux@1Meter) #1/#2/#3=Average
A2=2200/2260/2230=Ave=2230
A1 White LED=836/850/854=Ave=847
A1 Green LED=1279/1298/1286=Ave=1288
A1 Red LED=2370/2340/2420=Ave=2377
A1 Blue LED=823/817/820=Ave=820


Test #2 - Lutron LX-102, Serial Q030770
Recorded on 3/23. Temperature 53F

Light=Measurement (Lux@1Meter) #1/#2/#3=Average
A2=2520/2540/2490=Ave=2517
A1 White LED=1211/1147/1193=Ave=1184
A1 Green LED=2670/2650/2670=Ave=2663
A1 Red LED=1580/1600/1570=Ave=1583
A1 Blue LED=760/750/750=Ave=753


Test #3 - wquiles' "Light Box", Serial 001
Light meter used = Meterman LM631 Light Meter, Serial 0503014
Recorded on 3/23. Temperature 76F
Note that the Light Box measures reflected light only
Light=Measurement (in LBU's, or Light Box Units) #1/#2/#3=Average
A2=560/570/572=Ave=567
A1 White LED=298/289/293=Ave=293
A1 Green LED=230/233/234=Ave=232
A1 Red LED=630/630/627=Ave=629
A1 Blue LED=175/174/177=Ave=175


Test #4 - wquiles' "Light Box", Serial 001
Light meter used = Lutron LX-102, Serial Q030770
Recorded on 3/23. Temperature 76F
Note that the Light Box measures reflected light only
Light=Measurement (in LBU's, or Light Box Units) #1/#2/#3=Average
A2=570/567/578=Ave=572
A1 White LED=395/398/398=Ave=397
A1 Green LED=427/426/425=Ave=426
A1 Red LED=353/351/351=Ave=352
A1 Blue LED=39/39/38=Ave=39 => Note value is way too low. I did several times, but always got this range. Don't know why :(

I hope to test some LED's soon :D

Will
 

chevrofreak

Flashlight Enthusiast
Joined
May 10, 2004
Messages
2,543
Location
Billings, Montana, USA
look ma, it's an integrating birdhouse! :crackup:


I look forward to your readings on other lights as I am preparing to build an integrating "oddly shaped object" of my own before long :grin2:
 

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
chevrofreak said:
look ma, it's an integrating birdhouse! :crackup:

I look forward to your readings on other lights as I am preparing to build an integrating "oddly shaped object" of my own before long :grin2:
Yes, a birdhouse is a more appropriate way to describe this contraption ;)



Darell said:
Awesome! I've gotta say, I'm most impressed!
Thanks :eek:

Will
 

Macaw

Enlightened
Joined
Oct 8, 2005
Messages
565
Like your integrating bird house. Complete with lighting!! It would be a little small for me though.:grin2: Of course I can always chew a little larger doorway, and let my tail poke out of the hole...:lolsign:
 
Last edited:

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
wquiles said:
A1 Blue LED=39/39/38=Ave=39 => Note value is way too low. I did several times, but always got this range. Don't know why :(

Will


In your pictures, the surfaces look somewhat yellow, especially when compared to other white items in your photographs. This would easily account for rather low blue readings.
 

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
NewBie said:
In your pictures, the surfaces look somewhat yellow, especially when compared to other white items in your photographs. This would easily account for rather low blue readings.
Yes, I noted that as well :(. It is the Clear Acrilic Spray that I used to protect the painted surface. Either it is not really "clear", or the two coats that I gave it caused the color change. Either way, it is too late and I will have to live with my mistake - it is what it is ;)

Will
 

Pinter

Newly Enlightened
Joined
Oct 28, 2003
Messages
120
Location
Hungary, Europe
Hi,

here is a sample of the same Fenix L1P light, as a digicam sees it.
Red, green, Blue component and the overall in black.

First is a beamshot on a yellowish white wall (Click for the large view)

Note that blue component is diminished when reflected from a yellowish target.

Second is a beamshot on a sheet of white paper


Nice, even red/green/blue components on a white paper, also really white in real life.
 

chevrofreak

Flashlight Enthusiast
Joined
May 10, 2004
Messages
2,543
Location
Billings, Montana, USA
Some light meters are not very sensitive to blue, while others are extremely sensitive to it. Perhaps yours is one of the less sensitive ones, while mine seems to be the most sensitive one according to the light meter benchmark passaround.
 

brightnorm

Flashaholic
Joined
Oct 13, 2001
Messages
7,160
wquiles said:
... All internal panels were sanded in between coats of white paint (4 layers), and then 2 layers of clear acrilic. The surface is not like a mirror, but it is very flat and even. ...ll

Will,

Very impressive, beautiful job! Are the coats of white paint latex or oil-based? If oil-based they will gradually yellow, though I'm not sure how that will be affected by the acrilic coating.

Doug of Flashlight Reviews built a "milk carton IS" which he uses for his throw and total output tests. though his figures provide relative measure rather than actual lumens.

Brightnorm
 
Top