correcting 100 watt 2900K bulbs

January 2, 2005

Odd question here,

I built a bunch of 2x2 foot enclosures housing four 100 watt bulbs each (2900K) with diffusion over the open face to be used for soft fill.

Any suggestions on how to correct this to blend in well with 3200K tungsten lighting?

The "right" answer is to drop 82B gels over the boxes, but I'm wondering if I might get away with simply painting the interior of the enclosures (which are now a somewhat shiny melamine-white colour). I actually see that my local film co-op has a "colour correcting box" in its inventory of rentals.. guess I should check that out.

Has anyone tried this before?

Should I just give up and shell out $50 for correcting gels?

(obviously with a colour temperature meter, I could pick the right blue paint for the trick, but I'm not sure how easily I can get my hands on one... although I do have the colorline 5000 in the basement for calibrating my colour photo darkroom... hmmm).

-Sean

Edited January 2, 2005 by Sean McVeigh

David Mullen ASC · January 2, 2005

You'd have to have a lot of softboxes to need $50 worth of gels. You can buy them in individual sheets.

Other options might be to use some blue tissue paper (like from an art store) in front of the diffusion to cool off the color. There are also some "cool" diffusion made (or there used to be) but that's not really going to save you money since you already have the diffusion.

You could also put one blue-dipped 250 watt photoflood (4800K) among the three 2800K bulbs and see if the end result is more like 3200K once it goes through the diffusion.

Or you can not worry about the fact that the softbox is 2800K. That would match surrounding practicals in the frame and any additional lights could be gelled with 1/4 CTO to match.

January 2, 2005

You'd have to have a lot of softboxes to need $50 worth of gels. You can buy them in individual sheets.

Other options might be to use some blue tissue paper (like from an art store) in front of the diffusion to cool off the color. There are also some "cool" diffusion made (or there used to be) but that's not really going to save you money since you already have the diffusion.

You could also put one blue-dipped 250 watt photoflood (4800K) among the three 2800K bulbs and see if the end result is more like 3200K once it goes through the diffusion.

Or you can not worry about the fact that the softbox is 2800K. That would match surrounding practicals in the frame and any additional lights could be gelled with 1/4 CTO to match.

<{POST_SNAPBACK}>

Actually, four 24x28" gels would be $30 in my neck of the woods.. so my original estimate was off :) That's actually about in line with buying paint and spending time getting it right.

Good idea about filtering the 3200's down to match instead.. that's definitely viable. 1/4 CTO you figure?

Thanks David.

-Sean

Laurent Andrieux · January 2, 2005

The thing is the gel will "die" after a while. Painting the boxes might live longer, if the paint is designed for high temperature (otherwise, it might turn odd with the heat).

mrbill762 · January 2, 2005

Sean,

Why not use two ECAs at 250 watt each or one ECT at 500 watts?

Both are rated at 3200K.

That said I doubt that you would see much of a difference between 2900K and 3200 K anyway unless the 2900K source was right up alongside the subject.

Mr. Bill

January 3, 2005

That said I doubt that you would see much of a difference between 2900K and 3200 K anyway unless the 2900K source was right up alongside the subject.
Mr. Bill

<{POST_SNAPBACK}>

Hmm.. well, check out these test shots (from a digital camera):

Kind of a weird setup.. I had CTBs on a 1K and 650 arrilite about 20 feet away, and I doubled the gels up on the 2900K boxes. (one of which you can see in the third image). They're still a tad warm for my liking... but of course, in hindsight, simply doubling up the gel probably wouldn't have given me the correcting factor I needed.. it would have just cut the intensity of the light further I guess. But you get the idea. That 300K difference is still fairly apparent.

I guess if I wasn't using them as keys, but just as a very soft fill placed far enough away, the difference may be more tolerable.

-Sean

David Mullen ASC · January 3, 2005

Where in these shots are 3200K and 2900K shown next to each other? I see very blue backlight (closer to 5500K) and tungsten fill. I don't see 3200K backlight and 2900K fill light, so where is the 300 degree difference visible?

Why don't you just use less blue on the backlight so that the difference between in the key and fill color is less strong, and then time it bluer if needed in post or with white-balancing?

January 3, 2005

Where in these shots are 3200K and 2900K shown next to each other? I see very blue backlight (closer to 5500K) and tungsten fill. I don't see 3200K backlight and 2900K fill light, so where is the 300 degree difference visible?

Why don't you just use less blue on the backlight so that the difference between in the key and fill color is less strong, and then time it bluer if needed in post or with white-balancing?

<{POST_SNAPBACK}>

Hi David,

There was a 3200K backlight and 2900 fill, but I had them both gelled with identical CTB filters. So relatively speaking, I figured they should still be off by around 300K, or am I way off there? (physics-wise, I suppose a CTB gel isn't actually shifting frequencies, but is attenuating the peak around 3200K.. on a 2900K light, some of the yellower stuff may still be slipping through).

Admittedly, this isn't the best example because of the filtration variable that's at play already. I'll have to try a setup this weekend without the CTB factor.

Thanks for your comments.

Cheers,

Sean

David Mullen ASC · January 3, 2005

Hi David,
There was a 3200K backlight and 2900 fill, but I had them both gelled with identical CTB filters. So relatively speaking, I figured they should still be off by around 300K, or am I way off there?

<{POST_SNAPBACK}>

Your photos show a LOT more of a difference than 300 degrees -- more like a 1000 degrees! -- so perhaps you used the wrong CTB filters or your two lights were not 3200K and 2900K. There is only a 44 Mired Shift difference between 3200K and 2900K, slightly more than a 1/4 CTB correction.

January 3, 2005

Your photos show a LOT more of a difference than 300 degrees -- more like a 1000 degrees! -- so perhaps you used the wrong CTB filters or your two lights were not 3200K and 2900K. There is only a 44 Mired Shift difference between 3200K and 2900K, slightly more than a 1/4 CTB correction.

<{POST_SNAPBACK}>

hmm, you raise a good point.. I'll have to meter these lights directly.

The 3200's were rented, so I can only assume they were correct, and what I'm calling 2900K is of course just a set of 4 100W bulbs from home depot. They may actually be much warmer than I am giving credit for.

Of course, the generator could have been undervolting something.

As to the gels, I picked them up the same day. Full CTB.

Oh, and just for clarity.. I'm not too concerned with this "night" setup, since the shoot has been postponed until May. I'm considering bringing the 2900K boxes along with me this weekend to play with on an indoor shoot which will be lit all 3200. (no pesky blues to muck things up).

Thanks again.

-Sean

January 10, 2005

Just as a followup to this thread.. I managed to borrow a 'color converter' box of some sort which did the trick beautifully. It was a rather heavy little brown metal box with switch on the front to select 29,32, and 34, and provided about 8 2-prong outlets. I am assuming that inside is just a rather large multitap transformer which was upping the juice from 120V to something slightly higher (maybe around 140 or 150?). In the process, presumably this also boosts the output of the 100W bulbs to something on the order of 150W or so. So instead of using gels and having less light output, I got the correcting factor PLUS more light :) -- at the expense of filament life no doubt.

Time to put one of these on my wish list :)

-Sean