kelvin change

[Miguel Bunster]

Hi,

Just checking something. Some times when using gels from diferent manufactures I get different kelvin readings even thoug the gels are for the same...small shift...a 100K here and so.

But in general I always had in mind a full CTO will shift 5600K to 3200K.....but one CTO gel goes grom 6000K to 3200K....and anyway I stick to the ones i use but for example a 1/2 CTB would take a 3200K lligh to 4300K right?

And a 3200K light with full CTO should shift to 2000K aprox or 1800K...?

I jsut wanted to check if I was on the right track here...this jsut came after testing some stuff with similar names and going through things...a doubt came in because I am doing somethign similar which I have done before but something came of.f I am shooting day interior correcting in camera and I want to get sun rays with 10k tungsten units. with the 85 filter on camera they should go down to aprox 2000K. And may want to bring them back up some ties a little where I would ad a 1/2 CTB to get them around 2800K (adding the 85 on camera)....This sounds right to me but wana double check with anyoen who has the time.

Thanks!

M

[David Mullen ASC]

You need to get ahold of the gel swatchbook for a manufacturer, which hopefully will list the MIRED shift of the gel, then with a little math, you can figure out the change in color temp from the MIRED.

 

To get the MIRED value, divide 1,000,000 by the color temp, so the MIRED value of 3200K is 312.5. 5500K is 181.8.

 

My Lee Filter swatchbook says that their 1/2 CTO has a MIRED shift of +109. So on a 5500K light, 1/2 CTO will be 181.8 + 109 = 290.8, which is 3439K. Of course, that's assuming your HMI is really 5500K.

 

Lee 1/2 CTB has a MIRED shift value of -78. So on a 3200K light (MIRED 312.5), a 1/2 CTB gives you a new MIRED value of 312.5 - 78 = 234.5, which is 4264K.

 

Now you'd think that 1/2 CTO on a 5500K HMI would get you the same color as 1/2 CTB on a 3200K tungsten lamp, but as you see, it's not exactly the same.

 

Gels fade over time with use, so that can cause variations.

 

But since you're talking about a sunlight effect, you can set the warmth by eye.

 

Looking at the Rosco Cinegel swatchbook, I see that their 1/2 Sun CTO has a MIRED shift of +81 (as opposed to Lee's +109) and their 1/2 CTB has a MIRED shift of -68 (as opposed to Lee's -78).

 

So in this case, it works out that Rosco 1/2 CTO on a 5500K light is 3805K and Rosco 1/2 CTB on a 3200K light is 4090K, a little closer than what the Lee gels gives you.

[Jonathan Bowerbank]

Here's a good question...

 

How do you pronounce MIRED?

 

I've heard it as "Me-Red" and "Mire'd"

[Michael Nash]

Since it's short for "MIcro-reciprocal degree," how would you pronounce it? ;)

[Jonathan Bowerbank]

Ahhhh, yes...it appears I've been saying it right all along :)

 

I've always assumed it pronounced like "mire", as in "ox in the mire"

 

danke

 

;)

[John Carreon]

I'm glad people take enough time to use "italics" in their answers...

 

I've learned a lot from those subtle nuances...

[Miguel Bunster]

David,

thanks for such a spot on answear!

I was looking into the Mired think now actually and I got a hold again of a Lee swatch book.

Thanks a lot!

Best

M

[David Mullen ASC]

If you do the math, an 85B camera filter (5500K to 3200K) has a MIRED shift value of +131.

 

But if you look at a swatchbook, a Rosco Sun Full CTO has a shift value of +167, their Full Straw has a shift value of +160.

 

Lee Full CTO has a shift value of +159.

 

This is mainly because HMI's burn bluer than 5500K, so the Full CTO/CTS gels are stronger than an 85B to convert an HMI to 3200K tungsten. But this explains why putting Full CTO on a window often warms of the view more than an 85B camera filter would.

 

So an 85B camera filter would shift a 3200K light (312.5 MIRED) over to 443.5 (312.5 + 131), which works out to be 2255K (1,000,000 divided by 443.5.

 

Adding Lee 1/2 CTB (-78 MIRED shift) to the tungsten lamp being shot with an 85B camera filter would change 443.5 to 365.5, which raises the color temp to 2736K.

 

I don't know why any of this is important though -- real sunlight changes color temp all day, so you're free to make it as warm as you want, by eye.