Match's Mods: Homemade Integrating Sphere

119 posts / 0 new
Last post
Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA

Piers said " ....but who wants enough light, when you have the option for far too much "

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

If you're planning on building a sphere like I did, I'd go with the Mastech simply because it'll make it easier to mount.  

agenthex
agenthex's picture
Offline
Last seen: 6 years 1 month ago
Joined: 07/14/2010 - 04:49
Posts: 3070
Location: Merica

They're actually both mastech. I have orange trim one and it works really well. The "head" on it is smaller and more precise than the other one.

Reading this makes you smarter: http://lesswrong.com/

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA

thanks guys, i went with the one with the smaller sensor. Measures higher, even though I dont need it really, has a smaller sensor so I could make a smaller baffle. most importantly it has the ring around the sensor for easier press fitting. wouldnt have thought of that

Piers said " ....but who wants enough light, when you have the option for far too much "

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Happy to have helped.  Let us know how your I.S. turns out.

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
The guy who invented paper mâché. What was his problem? The paper mâché ball is going well. I have no idea how many coats I have on it. The first few coats I used wallpaper paste because I had some. I didn't read match's OP real close the first time I guess and now I see that is what he used first. So I really am following his directions exactly. I think I put too many layers on at once in the beginning. I'm not sure the inner layers are completely dry. After the first layers with wallpaper paste I switched to the 50/50 mix Match used. It's going good. These layers are drying faster. I have no idea how many layers are on there but it's getting pretty solid. When I cut it open the those inner layers can dry if they are not already. It has been cold here, if I could let it bake in the sun I could go faster I think. Bottom line is it's going good. Nice project for the kids to help with also. The lux meter is coming from amazon and should be here real soon. Great project, can't wait to achieve my new level of geekdom. Gotta think about paint jobs also. That will be special , I promise.

Piers said " ....but who wants enough light, when you have the option for far too much "

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Nice work, Langcjl.  Keep us updated on how it turns out.  The good thing about using the wallpaper glue first is that it contains an anti-fungal agent in it, which will prevent the project from molding before it's complete.  As far as the number of layers, I lost count.  I just kept adding, then performing a "squeeze" test the next morning.  Once I was no longer able to squeeze and feel the ball underneath, I called it good (no need to squeeze too hard though).

I don't think I've asked before...what part of Wisconsin are you in?

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
Match, I'm in the southern part below Madison. Right in the middle of the huge cornfield.

Piers said " ....but who wants enough light, when you have the option for far too much "

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Very nice.  My folks are near the Twin Cities, and I still head over to River Falls/Ellsworth every now and then to visit some old friends.

Foy
Foy's picture
Offline
Last seen: 4 years 7 months ago
Joined: 01/02/2011 - 17:56
Posts: 3009
Location: Las Vegas

Okay Lang, I hope you're happy - you did it again.  I just ordered this meter for use in my upcoming Match inspired IS project.  Now, off to the drug store to find a bouncy ball.  This project will remove all doubt from those around me about my being a worthy candidate for some type of institutional psychological rehabilitation.

Foy

No referral links and nothing embedded . . . ever.

                      &nbsp

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
We knew you would come around soon.

Piers said " ....but who wants enough light, when you have the option for far too much "

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA

I finally cut the ball apart. It was pretty cool when I punctured the ball. I used kitchen plastic wrap and that worked nicely. No sticking or anything. I used a hole saw to cut the holes. I made the flashlight hole as big as my SF Masterpiece because it is the biggest one I have. For the Lux meter hole I only made a 2cm because I don't know how big the sensor will be. Hopefully it will come in the mail today. Match, how did you cut your holes? I used a hole saw but don't know how I am going to fine tune the sensor hole. I tried a sanding bit and a stone grinding bit on the dremel but they didn't do anything. Maybe the hole I already made will be perfect. It could happen.

Piers said " ....but who wants enough light, when you have the option for far too much "

Don
Don's picture
Offline
Last seen: 2 years 7 months ago
Joined: 01/12/2010 - 16:32
Posts: 6617
Location: Scotland

http://www.lightglassoptics.com/Newport-819-IS-2-Integrating-Sphere--2-i...

 

$1100 for a two inch lab grade one!

 

And the detectors (meters) are expensive!

 

The numbers from my light tests are always to be found here.

https://spreadsheets.google.com/ccc?key=0ApkFM37n_QnRdDU5MDNzOURjYllmZHI...

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Looking very nice, Lang!  I cut and trimmed my holes using a very sharp razor blade.

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
The only light I have that has a known lumen value is the UF 3 mode dropin. I know these values from Matches numbers. Using my lux numbers H 3600 lux M 1250 L 210 And Matches lumen numbers H 838 M 280 L 49 I came up with correction factors of H 4.30 M 4.46 L 4.28 So if I take the 3600 lux reading and divide by 4.35 I get 827 lumen. After I figured all of this out I remeasured the high and kept getting 3800 lux. I'll have to do more measurements and figure it out better. Using the numbers I have now, I measured the SF Masterpiece pro1 with one cell @ 202 lumens which seems spot on to me. ITP EOS A3 XP-G R5 w/10440 333 lumen XR-E Q5 14500 252 lumen These seem pretty close with the Cree charts, a bit high maybe.I think I'm going in the right direction. I know I was a bit off on my initial figures, I get hyper sometimes, so I'll go back now and try to refine things. If I'm doing it all wrong, let me know Just tested the Tmart 502B Q5 AK47A driver @ 223 lumen. The Cree charts show 233-251 @1A

Piers said " ....but who wants enough light, when you have the option for far too much "

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
It looks like my 202 reading for the Masterpiece is in line with what this review reports http://www.lumensreview.com/reviews/1-led-flashlights/104-solarforce-mas... It says aprox 200 with 1 cell and I got 202. By the way, have I reported how sweet the Masterpiece is? It is.

Piers said " ....but who wants enough light, when you have the option for far too much "

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Looking good, Lang.  I'd definitely say you're on the right track.  I used both of my zebra's (H51 and SC60) since they've been tested out the wazoo and seem quite repetitive.  My only concern with using the UF 3mode xml drop-in is that the high isn't current regulated, and dependent on the battery quality.  

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
Your right about that P60 Match. I think I got lucky with the one I started with. I tested a couple of others and got different readings. It is very battery dependent. Everything seems to match up now with the two lights I have good numbers for. This has been a great project. It's so much fun to play with. Thanks for sharing this one Match. All I have to do now is the final paint job.

Piers said " ....but who wants enough light, when you have the option for far too much "

Boaz
Boaz's picture
Offline
Last seen: 1 day 6 hours ago
Joined: 11/07/2010 - 09:31
Posts: 7173
Location: Birthplace of Aviation

I thought real otf numbers were something like 25 to 30% less than what the cree numbers were .

       καὶ τὸ φῶς ἐν τῇ σκοτίᾳ φαίνει καὶ ἡ σκοτία αὐτὸ οὐ κατέλαβεν

                            

       Dc-fix diffuser film  >…  http://budgetlightforum.com/node/42208

suri252001
Offline
Last seen: 8 years 3 months ago
Joined: 08/27/2011 - 10:41
Posts: 4
Location: India

It was Great!!!!!! But could you let me know making out Lumen from the Lux measured ?

Thanks!

suri252001
Offline
Last seen: 8 years 3 months ago
Joined: 08/27/2011 - 10:41
Posts: 4
Location: India

Match wrote:

Gentlemen,

  (Foy inspired quick summary:  I made my own I.S. using a $2 big rubber ball, paper mache, and a luxmeterWink)


  When I was into building disturbingly fast street cars, one tool that I always wanted was my own personal dynometer so I could accurately test/tune the fruits of my labor.  Seeing as how they cost over $15k and would take up the entire garage, I was stuck either finding a local shop with one or testing at the local drag track.  

  For reasons I can't fully explain (or completely understand), six months ago I found myself drawn into the flashlight world.  I would have never guessed messing with lights would be so much fun!  Yet, once again, I found myself in need of an accurate testing device...and for this hobby an Integrating Sphere (I.S.) replaces the dyno.  It serves the same function, but instead of reporting horsepower/torque it gives me lovely lovely lumen ratings.  Seeing as how there are no "hobbyist" IS's for sale, I was in a quandary as to what to do.  Thanks to some of the wonderfully inventive folks on this forum, I saw how one could be built fairly easily.  Being a traditionalist and an accuracy junky, I wanted to stick with the sphere instead of a box.  I've read where others have used large styrofoam balls, but the price for one (after shipping) was close to $100.  So, I decided to make my own from scratch -  which is more fun anywaysWink

  While perusing the aisles of my local Walmart, I saw the perfect "host" for this project.  A $2 big rubber ball in which I was going to entomb in paper mache to create my sphere.

  I actually ended up buying two of them, just so my young children could play with one while I destroyed the other.

First step was to wrap the ball in plastic so the forthcoming paper mache doesn't stick to it.  I used an old plastic picnic tablecloth cut into strips, but thinking back any type of food grade plastic wrap would work.  Once the ball is covered, it's simply a matter of layering thin strips of newspaper over it.  I tried a few differerent adhesives on this project.  The first was a leftover can of wallpaper glue, which seemed to work fairly well, but then finally settled on a 50/50 mix of white glue and water.

 Here's the ball in the beginning stage after the plastic is on and the first layer of newspaper is going on.  I honestly lost count on how many layers of paper mache I ended up using, but want to say it was around two dozen.  The test I would do to determine if it needed more was once it was dry I would give it a squeeze.  If I could still feel the rubber ball underneath, I put more layers on.  Eventually it felt solid (just don't push too hard), and ended up looking like this:

 The moment of truth comes when taking that first cut into it with a razor knife.  Carefully cut across the equator to create two equal halves.  Then, remove the remnants of the rubber ball and plastic wrap.  If all goes well, it should end up like this:

Next, paint the inside of both halves with a white matte paint.  Mine took a couple of coats to cover up the type on the paper. Once the paint dries, the two ports need to be cut into one of the halves.  One is for the luxmeter, and the other is the entrance for the flashlights.  These should be at a 90 degree angle from each other.  The only other addition required on the inside is a baffle.  From researching, this should be as small as possible but still cast a shadow over the sensor port when a light is shown into the input port.  This prevents any light from directly hitting the luxmeter.  Here's how mine turned out:

  Once this step has been completed, the sphere needs to be reassembled.  I used duct tape (or as I've been calling it, "Don's Tape") to hold the two halves in place while I re-applied paper mache to the outside.  It only took a few more layers around the cut to firm everything back up. Once back together and dry I decided to shoot it with a quick coat of spraypaint to make it look more complete. And here it is:

At this point, the only thing left to do is plug in a luxmeter in it's port and determine what the sphere's correction factor should be.  There's a lot of different ways to do this, but the most accurate one I tried was to simply take a light of known value, test it to get the lux reading, and then divide that reading by it's alleged lumen rating.  For example, I have two lights that have been tested a lot by others.  The zebralight SC60 (300L otf) and the H51 (200L otf).  I came up with a value of 12.65, which, when tested shows the H51 at 200L and the SC60 at 300.4L.  I also tested various other lights and the values returned seemed quite plausable.  One thing I found helps is using cardstock with a perfectly sized hole cut into it for the head of the torch.  I've got a bunch of these now of various sizes.  Here's me testing my L2P:

And a couple shots of the luxmeter and how it looks attached to the I.S.  It is a tight press fit, so no other retention is needed to hold it in place.

 

Summary:

 I'm very happy with the results of this and now having my very own 14.5" I.S.  Unlike the other flashlight projects I've done, I was able to get the kids involved with this one.  Gave them a tray of glue/water mix, some newspaper, and let them have at it!  Thankfully my 7yr old is as anal as I am, so it all turned out well.  The project was pretty simple and went fast.  The longest part was waiting  between paper mache layers for the glue to dry.  Total cost for creation was @ $5 in change, for the ball and a couple bottles of white glue (assuming one already has a luxmeter, if not add ~$20).  The best part is the repeatability.  I can take any torch I've previously measured and using a battery of the same charge level get results within a tenth of a percent.

Here's a few reasons why it's a handy device to have:

  • Accurately test recently purchased lights and ensure proper performance
  • Test outputs of a torch on various battery combinations
  • Compare before and after results after modification
  • Determine how various reflector/lens combos affect total light output (It's hard to subjectively compare the light output of a flooder vs a thrower by using the naked eye)
  • Troubleshoot a bad driver or emitter
  • Give lumen ratings on new flashlight creations!

  Some test results can be found here.  Thank you all for reading and humoring my flashaholic DIY sickness.

 

It was great, but could you let me know making out lumen from Lux measured?

 

Thanks!

Langcjl
Langcjl's picture
Offline
Last seen: 1 year 8 months ago
Joined: 03/05/2011 - 05:36
Posts: 2162
Location: Wisconsin USA
In the post you quoted above, the information you are looking for is right under the picture of the finished, painted sphere.

Piers said " ....but who wants enough light, when you have the option for far too much "

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

suri252001 wrote:

It was Great!!!!!! But could you let me know making out Lumen from the Lux measured ?

Thanks!

There's a quite complex formula that takes the area in sq feet inside the IS and extrapolates a lumen rating from the lux measured, while taking in the refraction and absorption of the sphere coating itself....  :Sp

Me?  I cheated and just took a light with a known lumen value to determine what my sphere correction value should be. 

For example.... A test light is known to give out 200lm.  When you shine it in your sphere you get 2,530 lux.  Dividing the lux by the lumens gives 12.65.  This is the correction value (for this example).

Then, all you need to do is divide any lux readings from your sphere by this set correction value to give you the lumen output.

Hope that makes sense. 

SPAMBOT
SPAMBOT's picture
Offline
Last seen: 1 year 11 months ago
Joined: 07/16/2010 - 18:40
Posts: 484
Location: Old World

Awesome build Match!

Now I need to find me some nice round inflated rubber balls Cool

Now with 100% all natural asbestos!

sixty545
Offline
Last seen: 6 months 1 hour ago
Joined: 10/25/2010 - 14:15
Posts: 702
Location: Denmark (GMT + 1)

Match wrote:

suri252001 wrote:

It was Great!!!!!! But could you let me know making out Lumen from the Lux measured ?

Thanks!

There's a quite complex formula that takes the area in sq feet inside the IS and extrapolates a lumen rating from the lux measured, while taking in the refraction and absorption of the sphere coating itself....  :Sp

Me?  I cheated and just took a light with a known lumen value to determine what my sphere correction value should be. 

For example.... A test light is known to give out 200lm.  When you shine it in your sphere you get 2,530 lux.  Dividing the lux by the lumens gives 12.65.  This is the correction value (for this example).

Then, all you need to do is divide any lux readings from your sphere by this set correction value to give you the lumen output.

Hope that makes sense. 

In fact the math is quite simple:

Hit the hole in the sphere with a uniform beam of parallel light (e.g. a throwers hot spot at several (5-10) meters distance).

Measure the intensity just in front of the hole with a luxmeter.

(This number is lumens pr. square meter that hits into the sphere.)

Multiply with the hole's area in sq.meters and the result is the lumens number that enter the sphere - simple and easy.

The only difficulty I can see here is the sagging intensity from the thrower, when you remove the outer luxmeter and read the one build into the sphere the light might have dropped a little. Just repeat the procedure a couple of times and you will get corresponding values.

suri252001
Offline
Last seen: 8 years 3 months ago
Joined: 08/27/2011 - 10:41
Posts: 4
Location: India

Thank You very much!!!!! It was really great..

suri252001
Offline
Last seen: 8 years 3 months ago
Joined: 08/27/2011 - 10:41
Posts: 4
Location: India

Thank you!

Match
Match's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/14/2010 - 11:57
Posts: 1488
Location: South Carolina, USA

Interesting discussion over on this thread about if a I.S. is even necessary.  Linked to there to save it for posterity Smile

Knobby
Offline
Last seen: 6 months 2 weeks ago
Joined: 03/24/2012 - 17:00
Posts: 57
Location: Vancouver, BC

Boaz wrote:

Need to make one of these from a nice globe  

Did you say ...6000 lumens ??  diggety ...

 

> You forgot one more reason .. to impress your BLF buddies ... Very cool

You are a good Dad ...Now go out there and give that lil girl a set of new streamers for her bike and 2000lumen bike light that will make the neighbor kids cry .

Wow, to use cut down and use a large globe is a brilliant idea.  Has anyone tried it ??????? ............ Knobby

Don
Don's picture
Offline
Last seen: 2 years 7 months ago
Joined: 01/12/2010 - 16:32
Posts: 6617
Location: Scotland

Will work.

Sixty545 built his from two lampshades. And he used to make measurement equipment for a living.

A beach ball is cheaper...

 

The numbers from my light tests are always to be found here.

https://spreadsheets.google.com/ccc?key=0ApkFM37n_QnRdDU5MDNzOURjYllmZHI...

Pages