UV-filter for compact flashlight?

Discussion in 'Lighting' started by AgX, Aug 9, 2017.

  1. AgX

    AgX Member

    Messages:
    14,236
    Joined:
    Apr 5, 2007
    Location:
    Germany
    Shooter:
    Multi Format
    I got such filter for a common GNm 40 flashlight. To my surprise...

    The filter is obviously from the same material, hue and thickness as the cover of the reflector: slightly yellow, good visible when looking through the edges.

    In general reflector covers vary in colour. Sometime, as with Metz, even at the same model over time. At some models the tube has a slight yellow hue. From all this I got the impression that manufacturer of flashes coped with UV-radiation, either at the stage of the tube or the cover, at least from the 70s on.

    When would a doubling of UV-filter density be necessary? I can't do UV-meterings, thus I would like to hear of your experiences.
     
  2. Mr Bill

    Mr Bill Member

    Messages:
    553
    Joined:
    Aug 22, 2006
    Shooter:
    Multi Format
    Yep, it's been a pretty big deal for portrait/wedding work ever since color film became popular. If you don't have "UV-coated" tubes (or cover plates) then white shirts, etc., generally fluoresce with a bluish tinge.

    Occasionally I see a question come up about flash tubes that someone says, "seem to have yellowed with age," in reality this just means they have a UV-coating.

    Pretty much, it's not necessary to double up. The add-on filter you got was probably intended for use on an uncoated unit. But if you have any flash where you DO get slight amounts of white shifting blue, the extra filter will probably kill it. The Wratten 2B, a very common one in printing systems, has a fairly steep cutoff curve. The 2B would probably get just about any electronic flash out of the range where optical brighters will fluoresce. So if you have something with a "low grade" coating that causes slight fluorescing, the supplemental filter would probably kill it.

    I've looked at probably 50 flash/tube or flash combinations via a general-purpose spectrophotometer, which barely sees under 400 nm (to about 370 or 380, I think). At the time my employer was having a fair number of uncoated tubes "coated" by a flashtube manufacturer, and the exact coating specification was up in the air. So they would make several test coatings for eval. I'd get spectral data, then we'd test shoot in a studio to evaluate the effect on white clothing with brighteners. Some of them leaked enough UV to have a noticeable effect. In these cases a single supplemental 2B filter would take care of things nicely.
     
    Last edited: Aug 10, 2017
  3. OP
    OP
    AgX

    AgX Member

    Messages:
    14,236
    Joined:
    Apr 5, 2007
    Location:
    Germany
    Shooter:
    Multi Format
    Thank you for your comprehensive reply.

    I only buy from rummage boxes etc, thus I must take stuff the way it comes... The flash is one of the best german compact flashes, though likely unknown in the US (GN 40, full swivel, 5 auto-apertures, 5 manual setting down to 1/16). I got it in many samples, the accessory filter holder too, but only super-WA diffuser inserts. This is my first filter insert, designated as "UV". Yes, it could be that there were samples that had no good UV filtration, though I only came across samples with the same light yellow cover. But would the manufacturer then offer a distinct filter instead of coloring the cover in first place? I rather assume that utmost cutoff with their means would create too much yelllow hue in the light, so that they added the doubling filter for extreme needs.
     
    Last edited: Aug 10, 2017
  4. Mr Bill

    Mr Bill Member

    Messages:
    553
    Joined:
    Aug 22, 2006
    Shooter:
    Multi Format
    Hi again, I have no idea why the "distinct filter" vs having the filtration built into the cover. But presumably there must have been SOME reason.

    I don't see an "extreme need" as explanation. In my view, there would only be two purposes for "filters," one would be for color balancing to match some other source, the other is to block UV to prevent bluish fluorescence of things with brighteners, such as wedding dresses or white shirts. And the UV is easily blocked by a single Wratten 2B. A second 2B would not improve the UV situation further, as it is completely done by the first filter, but it WOULD shift the average cutoff point upwards slightly, by (roughly) 5 or 10 nm, I think. If you calculated a CCT ("color temp") with a doubled 2B filter I'm sure it would shift slightly, but this is not a good way to control color temp - the better way is with filters that attenuate the entire spectrum, not just the UV-blue cutoff point.

    As a note, I did also look at spectrums of several hot shoe flashes. I was a little surprised that the higher-quality ones had almost identical spectral output, with almost nothing past 400 nm present. (I would not expect ANY UV-induced fluorescence from these.) I am guessing that your flash unit is in this class, and, although I didn't say it earlier, this is part of the reason why I don't think your extra filter would do anything useful for you.

    As a note, the practical sort of test we did was to include two samples of "white paper," one with and one without brighteners, in test photos. We photographed these with "professional" flash as a baseline, then compared against the "test" flash. Evaluation was to see if the paper with brighteners took on a bluish tinge relative to the other paper. This is something that can be done without instruments once you know what you have. Note: almost all high quality white paper contains brighteners, and the best source for non-brightener might be a piece of "museum grade" mount board from an art supply store. They can be verified under black light, such as might be found in many nightclubs - the brighteners will "glow" in the dark. (Ps, the "nearly clear" UV filter will make a dark shadow on the "glowing" paper, and presumably so will your "yellowish" cover.) (Pps, obviously a "pure white" image of paper won't be able to show any color, so must be darkened a bit to see the bluish tinge.)
     
    Last edited: Aug 10, 2017