Using compact flourescent lamps for lighting

Discussion in '35mm Cameras' started by Peter Chant, Nov 18, 2004.

  1. Peter Chant

    Alan Browne Guest

    In a studio environment the test images can be quickly loaded onto a computer.
    This is what I meant (not checking the camera monitor ... not even the fantastic
    large monitor on the Maxxum 7D (sorry, I had to get that in!)).
    Flourescent isn't as bad in that sense, partially (I assume) due to the
    UV->'white' conversion in the phosphor. The phosphor is stimulated to emit over
    a wider (or sets of wider) spectra.
    See, told ya.
    And that is sufficient unto the day for probably 80% or more of studio photos.
    Again, this whole sub thread started with me emphsizing "_digital_"

    Cheers,
    Alan.


    --
    -- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
    -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
    -- [SI gallery]: http://www.pbase.com/shootin
    -- [SI rulz]: http://www.aliasimages.com/si/rulz.html
    -- e-meil: there's no such thing as a FreeLunch.
     
    Alan Browne, Nov 20, 2004
    #21
    1. Advertisements

  2. Peter Chant

    Alan Browne Guest

    Crawl up the thread a bit.
    The pulse train can be pretty fast. I don't think there would be a problem at
    10,000 Hz or even higher.

    --
    -- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
    -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
    -- [SI gallery]: http://www.pbase.com/shootin
    -- [SI rulz]: http://www.aliasimages.com/si/rulz.html
    -- e-meil: there's no such thing as a FreeLunch.
     
    Alan Browne, Nov 20, 2004
    #22
    1. Advertisements

  3. Peter Chant

    Bruce Murphy Guest

    blue, UV, same thing. A narrow blue-area wavelength with the
    'whiteness' produced by some levels of fluorescence.

    B>
     
    Bruce Murphy, Nov 20, 2004
    #23
  4. Peter Chant

    Bruce Murphy Guest

    Not true. All your colour correction filters are designed assuming
    that you're going to have some continuous blackbody light
    distribution. LEDs and in particular white LEDs don't have a
    distribution which looks anything like this.

    The spectra from white LEDs in fact, looks like a standard 40-60nm
    wide LED peak in whatever their native driving wavelength is, and then
    a tail of fairly non-continuous fluorescence tootling out into the
    red. I speak from having actually measured several of the damn things
    last week as possible light source candidates for a project we are
    working on here (not photographic)

    Now, when you start taking non-continuous light sources, colour
    correction (As per channel balance) doesn't work anymore for exactly
    the same reasons that you can't reproduce arbitrary colour filters in
    photoshop. (Or correct for arbitrary lighting spectra in photoshop,
    which comes out to the same thing)

    B>
     
    Bruce Murphy, Nov 20, 2004
    #24
  5. Peter Chant

    Alan Browne Guest

    Excellent points...I'll convert to the one true faith. Then those marketing
    them as ring lights and other sources for digital shooting are not being totally
    honest about the suiability for digital?

    Cheers,
    Alan


    --
    -- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
    -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
    -- [SI gallery]: http://www.pbase.com/shootin
    -- [SI rulz]: http://www.aliasimages.com/si/rulz.html
    -- e-meil: there's no such thing as a FreeLunch.
     
    Alan Browne, Nov 20, 2004
    #25
  6. Fluorescent bulbs have a much higher frequency than 100/120 hertz if you
    use a electronic ballast, which the newer Compact Fluorescent fixtures
    do. They're in the 10-20 kHz range and you won't notice any flicker.

    The 55w mothers are pretty freaking bright and I've used my winter
    "daylight" as a lighting source with good results.
     
    Brian C. Baird, Nov 21, 2004
    #26
  7. Peter Chant

    Bruce Murphy Guest

    I imagine they'll work fairly well for all subjects whose colour
    responses are themselves nicely behaved. However, if you combine
    serious discontinuties in light source with odd subjects, Very Bad
    Things will happen, uncorrectably. Of course, white LED ringlights are
    also being sold in things like dermatoscopes where their horrifically
    uncalibrated (And changing!) nature is starting to worry people w.r.t
    accuracy of skin lesion diagnosis.

    There was a particular synthetic green PVC colouring that wasn't
    continuous in its colour response, and one of its colour bands fell
    neatly into a hole in colour film's response, lime green turned into
    something sickly.

    One other source of seriously non-continuous colours would be certain
    flowers (which is where people shooting the damn things first learned
    about metamerism and colour films). Other fantastic examples are
    objects which use photonic crystal-like structures to generate
    colours, butterflies and feathers spring to mind here.

    So, while you /can/ get away with it for quite a few things, knowing
    where you can expect to get really nasty surprises is pretty
    important. Given that exciting synthetic dyes turn up reasonably
    frequently in fashion or product shoots, even restricting the light
    sources to studio work isn't going to save you entirely.

    B>
     
    Bruce Murphy, Nov 21, 2004
    #27
  8. Peter Chant

    Bruce Graham Guest

    Another possibility is to use an array of RGB LEDs. I *think* I have
    seen reference to that.

    A further complication with the white LEDs is their colour varies with
    angle. The few I have played with have a blue spot in the centre of the
    beam, with various yellow/green casts as you move off axis. Pretty ugly.
    So you would need a diffuser panel.

    Bruce Graham
     
    Bruce Graham, Nov 21, 2004
    #28
  9. Peter Chant

    Bandicoot Guest

    As I wrote that I did think of a tethered back with a calibrated laptop
    screen - but that is the exception rather than the rule, of course.
    That's why I said it was an "extreme example" - I was using the simple case
    to illustrate the point...
    You mean, I told you... ;-)


    Peter
     
    Bandicoot, Nov 22, 2004
    #29
  10. Peter Chant

    Peter Chant Guest

    I'm now feeling guity for starting a rather vigourous thread to which I am
    only just following up.

    I realise that LED's especially white have come on leaps and bounds, but
    unless doing macro or wanting extreamly long exposures it would seem I
    would need an exorbitant number (and hence expense) to produce a
    reasonable output.

    Pete
     
    Peter Chant, Nov 22, 2004
    #30
  11. Peter Chant

    Peter Chant Guest

    Hmm,

    that sounds very cheap. If true I feel an experiment in high speed
    photography coming on,

    I've plused a standard LED (15 or 30mA typical forward current) at up to
    about 1A ish for a microsecond with no harm. I wonder if I could try a
    panel.
     
    Peter Chant, Nov 22, 2004
    #31
  12. Peter Chant

    Bruce Murphy Guest

    I'm very curious how this works out, but I expect that the fluorescent
    nature of the white LEDs is fairly likely to cause Bad Things to
    happen[1]. (and will itself degrade fairly fast)

    B>

    [1] colour shifts and the like.
     
    Bruce Murphy, Nov 22, 2004
    #32
  13. Peter Chant

    Peter Chant Guest

    Very good point. Phosphors would not have time to react. That kills that
    for white light. However, I could just make do with an approximation.
    Do blue LED's use a phosphor, I think not? I wonder if amber plus blue
    LED's make a whiteish colour?

    I did not think this was acheivable at reasonable cost until I saw LEDs
    are rather cheap on ebay, as pointed out by other posters.


    Now, the $64000 question. How does the luminance of a LED vary with
    current? Is it linear? If so at small scales I might be able to produce
    the sun on earth with much more control than with a flash gun.

    Just thinking out loud.

    Pete
     
    Peter Chant, Nov 22, 2004
    #33
  14. Peter Chant

    Peter Chant Guest

    That's just a leg pull.

    I personally am not a great fan of the yellow glow of GLS bulbs and warm
    white seems to be no improvement. Some CF bulbs seem whiter ad more
    plesent. However, I think having a very, almost blue/white CF in my
    bedroom with light blue walls does go too far the other way.

    Pete
     
    Peter Chant, Nov 22, 2004
    #34
  15. Peter Chant

    Bruce Murphy Guest

    I doubt it. LEDs tend to have fairly narrow emission spectra and most
    colour filter curves I've seen have a bigger green/red gap than the
    eye.

    Even if it /did/ make your sensor see 'white', your colour
    reproduction would suck rocks for all the reasons discussed elsewhere
    in this thread to do with non-continuous light sources.
    Linearly to a point then starts to fall. You should be able to find LED data
    sheets for some of the high intensity things about the place and they
    tend to have these curves.
    I'm curious as to what it would look like. Since I've got my
    spectroscopy lab downstairs, it might even be possible to measure some
    of these effects.

    B>
     
    Bruce Murphy, Nov 22, 2004
    #35
  16. Peter Chant

    Alan Browne Guest

    Alan Browne, Nov 22, 2004
    #36
  17. Peter Chant

    Alan Browne Guest

    Vigor, even in the somewhat rambunctious NG fashion, is part of the education.
    First off, these bright LED's are VERY bright. They are cheap and can be ganged
    up cheaply as well. LOTS of light.

    Cheers,
    Alan




    --
    -- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
    -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
    -- [SI gallery]: http://www.pbase.com/shootin
    -- [SI rulz]: http://www.aliasimages.com/si/rulz.html
    -- e-meil: there's no such thing as a FreeLunch.
     
    Alan Browne, Nov 22, 2004
    #37
  18. Peter Chant

    Alan Browne Guest

    Peter Chant wrote:

    I would bet at that much over current repeated often enough you would burn them
    up a lot sooner than the 30mA X hours would normally do.

    Heat is heat, if you pulsed a 3V, 30mA LED at 1A and more than 33.33 kHz you
    will be over driving the LED's... shorter life... and that is really all that
    happens... I've seen products where LED's were purposely over driven by x as
    that would still give an MTBF that was wholly acceptable for the product.

    Cheers,
    Alan



    --
    -- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
    -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
    -- [SI gallery]: http://www.pbase.com/shootin
    -- [SI rulz]: http://www.aliasimages.com/si/rulz.html
    -- e-meil: there's no such thing as a FreeLunch.
     
    Alan Browne, Nov 22, 2004
    #38
  19. Peter Chant

    Alan Browne Guest

    LED brightness is usually controlled by pulsing the power to it and varying the
    duty cycle for brightness.


    --
    -- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm
    -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm
    -- [SI gallery]: http://www.pbase.com/shootin
    -- [SI rulz]: http://www.aliasimages.com/si/rulz.html
    -- e-meil: there's no such thing as a FreeLunch.
     
    Alan Browne, Nov 22, 2004
    #39
  20. Peter Chant

    Peter Chant Guest

    Well, I may just buy a load cheap off ebay and try.

    I still don't think they are that bright. The best I have seen produce 8
    candela.

    How does that compare to a 60W GLS bulb?
     
    Peter Chant, Nov 22, 2004
    #40
    1. Advertisements

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.