Connect with us

Integrating Sphere coatings

Discussion in 'Lighting' started by JB, Feb 2, 2004.

Scroll to continue with content
  1. JB

    JB Guest

    I am looking for alternative coatings for a lamp measurement integrating
    sphere of approx 2M diameter. This will only be used in the visible
    (400-700nm) and so far I've only found Avian-D (Avian Technologies) and
    Spectraflect (Labsphere). The cost for a coating service is over twice the
    cost of the sphere itself!
    We can get our sphere coated here in the UK, but are there any other options
    available as to the coating itself? Most paint manufacturers in the UK don't
    seem to have heard of barium sulphate paint.
    This will primarily be used for measurement of total luminous flux of
    fluorescent and HID lamps from 14W-400W so I wouldn't have thought that a
    reflectance over the visible of >90% would be needed.
    Future extension to this system will also include lamp spectral
    Any advice, links, thoughts on this?
  2. What level of accuracy do you require in your measurements? Flat "Ceiling
    Super White" paint works very well if you don't need ultra-accurate

  3. JB

    JB Guest

    I will be using this system for detailed HID lamp spectral measurements in
    the future. Iinitially I'm only masuring luminous flux and could probably
    get away with a flat white paint. It would be a pin to remove and recoat
    later with a 'proper' coating though.
    What is a little amazing is that I have been told by some suppliers that
    15-40 (!) coats of a typical barium sulphate paint wold be required. (Mind
    you they are selling the stuff to me at $495.00+ per gallon).

    thanks again,
  4. What do you mean by "ultra-accurate". That term is not listed in any
    of the standard reference books I have :)

    Seriously, the problem is that spheres are usually calibrated with
    lamps of one SPD, such as incandescent lamps, and then used to measure
    lamps with a completely different SPD, such as metal halide lamps in
    this case. The sphere paint does not necessarily have to have very
    high reflectivity, but if the reflectivity varies with wavelength to
    any appreciable amount errors will be introduced when the sphere is
    calibrated with one SPD and used to measure another.

    I believe there is a paper on this subject by the guy at NIST who is
    the expert on sphere measurements. Unfortunately, I don't know how to
    find the paper right now.
  5. JB

    JB Guest

    Hi Vic,
    Please can you post a link should you find the paper. This is just the
    information I am after.
    Many thanks,
  6. It's the next entry down from "Full-Spectrum". ;^{)}

  7. As others have said, you can use an "Ultra White Flat Ceiling Paint" and
    get good results - 90% flat reflectance above 500nm, dropping to ~5% at
    400nm. However, since you're trying to measure fluorescents and HIDs
    (which are heavy in the blue), you'll get a quite a bit of inaccuracy.
    Most integrating spheres are calibrated using incandescents/Illuminant A
    which have very little blue in them. Also, since you'll be wanting to
    add spectral measurements in the future, you'll be better served having
    a coating that has a flat response - probably no more than a 2%
    reflectance variance throughout the visible spectrum.

    For a long time my lab was using Ceiling Paint in our sphere with very
    good results - but we were primarily measuring incandescents. When we
    decided to start measuring HIDs, we bought the Avian Technologies'
    Avian-D. We got much better results then we ever expected - 98% or
    better reflectance throughout the visible spectrum. This actually
    caused some problems we weren't expecting such as how much the shape and
    reflectance of the bulb sample holder would effect our results. We had
    to add an auxiliary or absorption lamp to our system to maintain any
    type of accuracy. We might have been able to get away without having to
    add the auxiliary lamp had we put on a coating that was only a 90% flat
    reflectance. Applying the coating was very easy and it's pretty durab;e
    too. I coated over two years ago and my spot check/calibrations show
    the same reflectance as when I first applied the coating.

    I recommend contacting Art Springsteen at Avian Technologies - he's a
    wealth of information on this subject. He may very well post a response
    to the newsgroup as well.
  8. JB

    JB Guest

    Thanks for the advice. I was seriously looking at the Avain-D coating. Easy
    to apply as you say.
    I'll contact Art directly if he doesn't post on here.
  9. JB

    JB Guest

    Filed under "snake oil"
    I still think that the nearest to a "full spectrum" lamp (God how I hate
    that term), is a compact source Xenon lamp. XE/D or similar. Funny how the
    junk peddlers selling these SAD lightboxes etc don't market those sources.
    (50-100kV striking volts and >60A DC operating current and cold fill
    pressures of >8atm on seems to sort out the men from the boys).
  10. The NIST Integrating Sphere guy is Yoshi Ohno. I did a quick look for
    integrating sphere coatings at the link below but didn't see anything
    obvious. But it may be a start for you JB.
  11. I thought so :)

  12. Thanks Doug. Yoshi Ohno is the guy I was thinking about. He and Rolf
    Bergman presented a very nice tutorial paper on sphere photometry at
    the 9th International Symposium on the Science and technology of Light
    Sources, Ithaca, 2001. Unfortunately the LS:9 Proceedings is not
    generally available.
  13. JB

    JB Guest

    Many, many thanks Doug. I think I've got a few days reading ahead of me now!
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day