Integrating sphere #4 (the fast and cheap one)

A few questions.

Will this smaller ball handle higher output lights? A 100 lumen light and a 3000 lumen light?
Would it perform better if the outside was covered in a black paint since the walls are thin? Maybe that’s what you meant by not using a coating?
Where is a hollow ball available?

As said in above post, using the Tondaj luxmeter with range from 1lux to 200klux, the lumen range is 1-17,000 lumen. If the sphere is bigger the range will shift to higher numbers, i.e. a 30cm sphere (26cm inner diameter) has a inner surface that is 5.6 times that of the 15cm sphere, and because the light is spread out over this larger surface, the range with this same luxmeter will be 6-95,000 lumen. This is of course a less desirable range, but still workable. My Mobilux luxmeter goes down to 0.01lux, so that would also help with bigger spheres. Perhaps really big spheres still need a hole for the luxmeter to get the range right, and then a baffle must be used to block direct light from the light source.

About the walls of the sphere: styrofoam directly reflects only part of the light, a lot is getting into the foam and eventually may get back into the cavity of the sphere, escape on the outside, or it gets absorbed. Most of the 2cm wall thickness of this sphere is actually needed for added 'reflection' of the light. A simple test: you can see a slight 'greying' effect of a black background through half a cm of styrofoam. Through 2cm I can not see any greying.

Painting the outside black should not have any effect, the sphere does not see a difference if light coming out of the sphere is absorbed by black paint or if it escapes into the world, it is lost either way.

In europe, there appears only one manufacturer of hollow styrofoam balls, all hobby shops and display material shops sell the same balls, they come in a range of sizes the wall thickness is 2cm for all sizes.

I do not know about the US.

I’m wondering where can I get those kind of styrofoam here in my country…

Just don’t step on it.

Djozz, do you need my address?

After a quick search I’ve found this one (for German buyers):

I think this size (30cm) would do even better?

It is the same series styrofoam balls as I can get here. Mine was from Opitec, I believe that is a german shop too.

With my 15cm ball I can not make the hole much wider than 30mm, with 30 cm you can go as wide as 60mm, making it suitable for more flashlights (but a smaller hole will give a bit better integration, but that was not bad anyway)

So, I’ve ordered the 30cm one at Opitec, thank you for mentioning this other source. The project seems easy enough to find the time for it and to make use of the Luxmeter I’ve bought some time ago from Banggood (Sunche HS1010A). The real problem will be (for me) to find reliable reference sources to somehow calibrate the setup for lumen readings…
What do you think: could I use LED light bulbs as a reference? Some of them have rather precise specs on their datasheets, e.g. this Osram: Deutsche Homepage | Licht ist OSRAM
According to the package it should have about 806 lm (10W/60W type). Perhaps E14 lamps would suit better (to make the opening not too large)…

Sorry for the late reply, missed this post. How are you getting on with the sphere?

An Osram led light bulb should be a very ok reference if you let it warm up for 30 minutes before measuring (it takes that long to get the light output constant, as I learned from texaspyro), but this bulb you linked is not suitable because you have stick it entirely into the sphere to let all the light in, and it will then also directly shine on the area on the inner surface where the luxmeter is behind. A led spotlight is a more suitable reference, like this one: http://www.osram.de/osram_de/produkte/led-technologie/lampen/consumer-led-reflektorlampen/led-superstar-par16/index.jsp, or any other led reflector lamp with reliable specs.

Very cool djozz, thanks.

Antenne, you could look up the measurements from some of our members of a flashlight that you own, and calibrate that way. That’s what I’d do if I built one of these.

Thank you for the suggestions. I have to admit, my sphere is still to be realized. I’ve got the styrofoam ball but had no time yet for the finish…

Anywhere to get the ball in the US?
Edit: I found smoothfoam balls at Jo-Ann Fabric and Craft. They have both 8” and 12” ones.

Hi, Djozz:
I just bought an 8” smoothfoam ball. Can you tell me what is the best way to make the holes on the ball?

I used a narrow blade scalpel but an exacto knife should work as well. I hope the sphere works for you, there has not been feedback yet of someone who build one of these.

Thanks for getting back to me… I found Exacto knife work pretty well. I have a couple more questions and home you can help.

  1. You used aluminum-tape to line the flashlight port. How this can prevent the direct path? Is it to prevent the light goes through the foam?
  2. Do I need to paint inside of the sphere? People suggested using matte white paint. I am wondering if flat paint will work well also.
    Will keep you posted about my progress.

If the lightsource is not sticked completely down the hole, its light does not illuminate the part of the inner surface near the entrance hole, so the direct path to the detector area is prevented. I found btw, because the aluminium makes the hole lining highly reflective, that the luxreading is fairly constant with the lightsource sticking at various depths in the entrance hole.

I did not try plain matt white paint, but I tested a latex/bariumsulphate mixture that I found an article on. It reduced the reflectivity of the surface and did not greatly change the wavelength 'response' of the sphere's inner surface. So my newest insight is to just leave the coating out, just carefully sand the styrofoam with fine sandpaper to a matt finish. If you do decide on a matt white paint finish, let me know how it turned out, and if you care to check it, take a luxreading on the same (stable) lightsource before and after the painting to see if the reflectivity has improved.

Thanks! Will get back to you.

I put it together and did some test. The reading was very high. With my 8”/20cm diameter sphere (7”/17.5cm inner) and WF501 with XM-L2-U2 bulb, the meter shows 40000 Lux. I got the same value if I point the meter directly towards the Lux meter with a distance of 11” (28 cm) between them. When I use the same meter on my 4” pipe 3 elbow tester, the reading was 1600 with the same meter. Does that make sense? If not, what can go wrong?

The actual luxreading is not a number that tells you much, so to compare it with different other ways of using the luxmeter does not make a lot of sense. Once it is calibrated with a lightsource of known output you're good to go.

But that the number is so high does get you into trouble with the range, if the maximum output that you can measure before the luxmeter gets out of range is under 1000 lumen, that would make it less useful. I think what is to blame here is the limited wall thickness of the ball, just 1.25cm, compared to 2cm in the ball that I used. That transmits more light and gives higher readings than my version. It also affects the integration somewhat because more light is leaving 'the system'.

A solution can be a longer (black coloured) tube for mounting the luxmeter sensor so it sits farther away and sees less light. A bigger sphere or one with a thicker wall will do the job as well.