I believe you need to size it for your inlet hole. Bigger headed lights need a bigger sphere to integrate. That’s all I’ve got. I went the easy route and built a light pipe with 4” PVC.
Soon I will have lots of time available (not by choice).
And I need to look for something to keep myself occupied, and this one is certainly in my to-do list: Integrating sphere #4 (the fast and cheap one)
In theory, maybe practice, yes, the sphere is/should be more accurate. Did you go thru all of djozz's threads on them? I have a PVC pipe lightbox I've used for a long time now. I'm look'n at building a couple of spheres though, hoping over the winter.
Can't answer all your Q's, but pretty sure definite you need white inside. djozz's threads though should be a big help.
Well, if I understand it, and not sure I do, the bigger the better. I'd post in djozz's threads directly or pm him. I know he feels the PVC pipe contraptions don't cut it, and I'm tending to lean his way
There are quite a few integrating sphere build threads (and I am the biggest offender ), and indeed I am for spheres instead of pipes because they are more accurate if different types of light sources are used.
I recommend building my 'fast and cheap one', and do not be fooled by the simplicity, it will do the job as well as some more difficult to make integrating spheres. A few recommendations:
-use a somewhat bigger styrofoam ball than I did, but not too big because that will give range problems, I'd estimate a 25-30cm inner diameter would be best. The entrance hole can then be made bigger too, 80 to 100mm diameter would still work (but I would go for 70mm)
-the styrofoam balls that I used all had 20mm wall thickness, and this thickness is needed for optimal reflectivity.
-carefully sand the inner surface with fine grit sandpaper to remove the shine from the styrofoam. No further coating is needed.
-do not make a hole for the luxmeter, this design needs measuring through the styrofoam wall, or else you will need baffles and you get range problems.
-do not bother with the most expensive luxmeter, it needs no calibration anyway in this application and for white leds with colour temperature from cool to neutral the measurement error is not so bad (warm white leds and colour leds give higher abberations). It just needs to be lineair and I did not see linearity problems in at least my two cheap luxmeters.
-an inbuild correction light source to correct for reflectivity alteration of different light sources in the entrance hole is recommended if you care to dive into how that works (why it is needed, how it works and how to build one). Else, just leave it out.
That ball from the UK looks very nice, the inner surface seems without any cavities. (the balls I have been using all have small cavities where the two halves meet, I used lightweight wall filler to fill them up).