HMI 5600-6500 Deg. Kelvin?

vladwolodarsky

vladwolodarsky

Active member
Hey Guys-

One afternoon this week I am shooting a scene with a couple stuck in traffic. We are going to have the car parked in a parking lot. There's going to be a bus parked behind the car and a car on the passenger's side and the driver's side of the talent's car. We are going to shoot the master shot through the windshield and then a cross close up for the passenger and the driver.

Our lighting scheme is to fly in a 12 x silk overhead to diffuse the sun. We are going to have a 1200 HMI Par (1200 Arri Sun) off to one side shooting through some sort of diffusion. We are also going to have a flag on stand by for any stray reflections, and there will also be a polarizer in front of the lens.

My big question is about a gel for the 1200 HMI. I read that HMI color temperature ranges from 5600 Kelvin - 6500 Kelvin. We're going to be shooting about two hours before sundown and the light is going to be a little bit warm (guestimating around 4400 degrees Kelvin). Since I'm not positive what the exact color temp. of the 1200 HMI, how am I supposed to know the specific gel I need to order? We probably only have enough cash for one gel instead of each level of C.T.O. . Thanks for taking the time out to read this and any input that you have will really be appreciated.

TAPPVCARD2ND300.jpg
 
My guess is that the diffusion would cut out so much light that this idea wouldn't work even with the overhead. Diffusion cuts out more than you think and a 1200w is perhaps a bit underrated for this kind of job already. The typical HMI bulb is 6000K color temperature (at least the strict Osram compatibles like Phillips and GE). A 1/8 CTO will correct from 5600K to 5000K A 1/4 CTO will correct from 5600K to 4600K. One of those or a combination of the two may do what you want to match the time of day look you need.
 
Thanks for the comments. These boards are such a great resource. Won't a par instrument will put out more light than a fresnel because of the mirror inside? So I guess using a fresnel will not solve my potential light level problem.

I'm going to have a dark blue bus behind the talent's car in the master shot to help the background from blowing out. There's also going to be ND9 gel (pretty heavy, that's all we have) on the driver side and passenger side window.

Will a 1200w HMI fresnel put out more light than 1200w HMI par?

TAPPVCARD2ND300.jpg
 
I was going through my vast collection of rolls of gel the other day. I found some stuff that someone else must have purchased - they were these nice little sheets of gel maybe 16" x 24", and they were not cut because they still had the Lee tag on the corner. I think they are sold this way so you do't have to buy a whole roll. Why not pick up a 1/4, 1/8, and a 1/2 CTO and try them out?

They can't cost very much. I'd be surprised if they were more than $20 per sheet.
 
Yes an HMI par generally will give more light output than an equivalent wattage HMI fresnel because of the mirror / bulb configuration which concentrates the beam quite a bit more. The par with a parabolic mirror (PAR stands for parabolic aluminized reflector) is about the most efficient lamp you can get.
 
Last edited:
Always important to use a color meter and take a reading of your HMI's when you get them. Depending on the bulb life, they can REALLY vary in color temperature. They are supposed to be around 5600, but I've had some bulbs that couldn't get past 4k or 4200. Pretty bad.

To figure out what gels you need to correct the color temperature you must find out (with a color meter) the temperature of the outside and the color temperature of your HMI. Using the mired system you should be able to tell exactly what gel will match up the best.

Use this past link to see an example and more in depth explanation of measuring color temperature and how to compensate using mired shift.

http://www.dvxuser.com/V6/showthread.php?t=102243

vladwolodarsky said:
Will a 1200w HMI fresnel put out more light than 1200w HMI par?
Click to expand...
Since they are both HMI's and rated at 1200w, they should output about the same amount of light, but since one is a fresnel and the other a par, they will most likely have different beam focus, falloff, and spread therefore being different from each other.
 
Last edited:
Why did you assume the temp would be 4400K? Are you just going by the color temp of the sky? A couple hours before sunset would be much cooler, somewhere around 8000K I believe.
 
You must log in or register to reply here.
Back
Top