-
Measuring light stops of gels
Since getting my Sekonic L758 light meter, I have decided to put a bit more science behind my lighting and try an understand more the parameters and effects of certain things, like light stops, and differences in fresnel focus brightness, and in particular the effect of the gels that I use on my lights. I predominantly have Lee filters of many types, including a #250 1/2 white diffusion. I have however, three other diffusion gels which I do not know the origin of and therefore, wanted to measure the light loss. The setup I have is a 300w fresnel tungsten light 1M away from my light meter. The light meter is set to 1/50s at ISO 320.
I have taken measurements with no filters for wide and spot, then put each filter on the lamp and taken readings. The bit I don’t understand is why the EV difference alters between wide and spot. Easier if I put the figures down.
Test1 No Filter: Wide: f/5, Spot f/8
Test2 1/2 diffusion: Wide:f/4 EV-0.5, Spot: f/4.5 EV-1.6The EV value above is in relation to the Wide/Spot readings respectively, of the no filter Test.
I would have expected the EV difference to be the same. It’s almost as if, the brighter the light the more effect the diffusion has on light stops.So I obviously have a piece missing from my understanding. If I carry out the same tests for say a CTB gel, I get what I expect, although 1/4 CTB produces -0.5EV, 1/2 CTB produces -1.0EV, and Full CTB produces, -1.5EV. Consistent across wide /spot though.
From the published values of the diffusion, I am trying to gauge which gel I have by the number of light stops it reduces the EV by. When a gel say 1/2 a stop, should that be constant across light levels ? If so, there must be a flaw in how I am conducting my measurements, so would appreciate some guidance.
Declan Smith
https://www.madpanic.tv
After Effects CS6/ FCS3 / Canon XLH1 / Canon 7D / Reason / Cubase“it’s either binary or it’s not”
-
2 Replies
-
When you set up your diffusion test, did you set up a target spot at which you can exactly place your meter each time you take a reading? And did you first focus the light in spot mode on that target before putting up the diffusion? (It is impossible to correctly focus a light with diffusion in place.)
Rick Wise
Cinematographer
MFA/BFA Lighting and Camera Instructor Academy of Art University
San Francisco Bay Area
https://www.RickWiseDP.com -
I set the lamp on a stand with the centre at 53 inches from the ground, and secured the 758 to another stand (via the 1/4″ socket) with the Lumisphere retracted and also at 53 Inches from the ground, then put the meter 1M from the lamp. The lamp and the meter therefore are at a consistent position between tests. I have done the tests fully spotted and fully wide. When checking the setup, I didn’t use a target, just the measurements and a rough view where the centre of the spot was landing.
I did a similar test with my open faced lamp of the same power and the relative difference again shows different light loss depending on whether the lamp is wide or spot.
Test 1: No Filter Wide: f/5, Spot f/6.3
Test 2: 1/2 White Diffusion Wide: f/4.5 EV-0.4, Spot f/5 EV-0.6
Test 3: Thick Diffusion (unsure of grade) Wide: f/4 EV-0.6, Spot: f/4.5 EV-1.0So I am guessing this must be a property of diffusion that is more pronounced using fresnels?
The reason I am trying to understand the light loss is that on reviewing the rushes following a technical rehearsal, where I measured the light setups with the 758, I want to soften the shadows a bit with some diffusion. If I know the expected light loss I can calculate the changes I need to make to the setups prior to the filming so as not to take up any more time on set (that’s my theory at least!). This approach appears to work with standard colour gels (like CTB), but I guess with diffusion, it acts differently and depends on the brightness of the source being shone through it?
Declan Smith
https://www.madpanic.tv
After Effects CS6/ FCS3 / Canon XLH1 / Canon 7D / Reason / Cubase“it’s either binary or it’s not”
Reply to this Discussion! Login or Sign Up