Jump to content


Photo

best stocks to handle sodium vapor


  • Please log in to reply
20 replies to this topic

#1 F Bulgarelli

F Bulgarelli
  • Basic Members
  • PipPipPipPip
  • 409 posts
  • Cinematographer

Posted 17 July 2007 - 07:28 PM

Hello everyone,

I'll be shooting a low budget feature that involves a great deal of sodium vapor lights, it will be impossible for us to correct this fixtures, so I'm thinking about just matching the tungsten units to the sodium vapor lights and then remove it in post. It seems like there are different ways of going about this. What film stocks would be more appropiate for this kind of situation, I know that there is a Fuji stock specially designed for mixed light situations like this. I would appreciate any tips or advice.
Also, any lighting/gels recommendations for a test will be appreciated,

regards,

Francisco
  • 0

#2 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 17 July 2007 - 08:31 PM

You'll never be able to correct out sodium vapor lighting to look "normal." It has a discontinuous spectrum with significant spikes and gaps that make smooth, even color reproduction impossible. You can gel your lights to give a decent visual match to sodium vapor, but then you're just handicapping your tungsten spectrum in a similar manner (although it will still be more continuous than true sodium). I've seen people try to "time out" sodium before, and the end result is usually even more ugly than if they'd just left it alone.

The last time I put a color meter to sodium vapor it came up 1600 Kelvin and something like -35 on the magenta/green scale. Layering on that much minusgreen might get you technically close, but will look very odd by eye not to mention the stop loss. Also keep in mind that no two sodium units are going to burn exactly the same color, depending on age and so on. Look closely and you'll see some are very pink, some very yellow, often right next to each other.

You're probably better off going for a "visual" match, something that looks right on camera even if technically the numbers aren't right. My recipe on tungsten has always been two layers of Lee #162 Bastard Amber + 1/4 CTO. The nice thing about the pack I use is that you can subtly shift the color by substituting one layer, like swapping one layer of BA for another layer of 1/4 CTO to make the light a little more "friendly" on faces. But there are plenty of other gel packs people have recommended here before, so try a search.

As for film stocks, anything tungsten balanced is going have a slight advantage with all the red in sodium vapor. To my eye it's the magenta in sodium that makes it look harsh and ugly. The less sensitive a film is to that particular color, the less ugly it should look (in theory at least).
  • 0

#3 Eon Mora

Eon Mora

    New

  • Basic Members
  • Pip
  • 6 posts
  • Cinematographer
  • Chicago, Il

Posted 17 July 2007 - 09:37 PM

It's definitely best to go with a tungsten stock. The difference in color temp is quite a bit. But I know what direction your headed with trying to find the right stock. Fuji makes a line of stock called Reala. It's some really interesting stuff because it has a 4th strip of color that is magenta so you can time out green spikes (i.e. fluorescents). Its also a 500 speed stock so its a really interesting mixture. I've only seen it in action once and it was used to fluorescents because of their own broken spectrum (i believe that what it was designed for). It's a good stock but for your situation it has a 3rd peculiarity which is that its daylight balanced. You could correct it with a filter but that's throwing a whole nother factor into the bunch that could throw off the situation.

Honestly, in my opinion, I like the way sodium looks uncorrected despite what people say. I consider it an almost beautifully romatnic look, it just doesnt mix well with certain tones. I tried doing some tests where i shot color reversal with tungsten film and an 80A with sodium and that had a pretty sickly look with actors compexion.

I say shoot clean on some good kodak stock. Even pushing 500 looks good. You could try and dirty it up with high-con and some grain to work with the urban sodium look. I dont know what your project is but there's a lot of ways to go. the best for me is to shoot clean though.
  • 0

#4 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 18 July 2007 - 02:33 AM

Sodium vapour lamps don't have a colour temperature. It's a meaningless term.

Effectively if you try to correct anything lit exclusively by sodium vapour, it will come out neutral - ish, but mostly monochrome. It's one big spike at about 580nm, not much else. Even the "balanced" ones don't have a lot more to offer.

If you correct tungsten lights to make them look about the same yellow balance as sodium vapour, you will have removed virtually all of the blue component of the light. So that layer will be very underexposed, and will look lousy when you bring it up.

If the scene is lit with sodium, best bet is to accept that's what it is.

Yellow.
  • 0

#5 Richard R. Robbins

Richard R. Robbins
  • Basic Members
  • PipPip
  • 25 posts
  • Cinematographer
  • Omaha, Ne. USA

Posted 18 July 2007 - 09:45 AM

I find this topic very interesting because I occasionally m forced to shoot under these industrial type lights. I'm hoping that someone with accurate information can explain the differences between...
1.) Sodium Vapor
2.) Mercury Vapor
3.) Metal Halide

Of course, as they relate to production on video and film. What are the differences in color quality and light quantity?

Rich
  • 0

#6 F Bulgarelli

F Bulgarelli
  • Basic Members
  • PipPipPipPip
  • 409 posts
  • Cinematographer

Posted 18 July 2007 - 01:41 PM

Thanks guys, this is tremendously helpful.
I'm thinking that one option for me is to cover up the units closest to the actors and leave the ones in the background on, we'll always have the sodium vapor reference but never will interfere with the key, except for the spill, which I guess could be bad. we'll see if I can make that work.

Are these lights going to flicker? I'll be shooting with an SR2, mostly 24fps and a 180 degree shutter.

Francisco
  • 0

#7 Stuart Brereton

Stuart Brereton
  • Basic Members
  • PipPipPipPip
  • 3057 posts
  • Cinematographer
  • Los Angeles

Posted 18 July 2007 - 01:59 PM

They'll flicker at either 50 or 60hz depending on where you are in the world. If you're shooting at 24fps, you'll need to use either a 144 deg shutter (for 60hz) or 172.8 deg (for 50hz)
  • 0

#8 John Pytlak RIP

John Pytlak RIP

    (deceased)

  • Sustaining Members
  • 3499 posts
  • Industry Rep
  • Rochester, NY 14650-1922

Posted 18 July 2007 - 02:51 PM

"Low intensity" sodium vapor lamps have a nearly pure 589 nanometer spectral output. "High Intensity" lamps have a slightly broader spectrum. Mercury vapor lamps have mostly the mercury spectral lines with some broadening of output. HMI is a metal arc source that is relatively broad in output and nominally a daylight balance.

Sodium and mercury vapor lamps have lots of visible light output (very efficient for street lights and security lighting) , but have very poor color rendition.
  • 0

#9 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 18 July 2007 - 05:35 PM

I find this topic very interesting because I occasionally m forced to shoot under these industrial type lights. I'm hoping that someone with accurate information can explain the differences between...
1.) Sodium Vapor
2.) Mercury Vapor
3.) Metal Halide

Of course, as they relate to production on video and film. What are the differences in color quality and light quantity?

Rich


There's a good description here. Incidentally, the color they show for "high pressure sodium" is actually closer to what you see with low pressure sodium.

Sodium vapor are actually two types (high and low pressure), with two different color qualities. High Pressure are the pinkish-orange ones used in street lights in 90% of the US and many other parts of the world. Low Pressure are the extremely yellow ones Dominic described, that look like this:
Posted Image
Wikipedia


Mercury Vapor
are those bluish-green lights often used in street lighting and as outdoor security lights. They are a pale bluish-green to the eye, and significantly more green saturated on film and video (especially film).
Wikipedia

Metal Halide are often used in large parking lots and sometimes stadiums, and look kind of pale blue, almost daylight, to the eye. On film/video they appear a greenish-blue, but not as saturated as mercury vapor.
Wikipedia

Incidentally the color reproduction depends on the imaging device. Sodium vapor tends to look pretty "natural" on tungsten film, like the way it appears to the eye. But video will often see it as much redder, and digital still cameras usually capture it more yellow, without the red spike. Film and digital still cameras are also more sensitive to the green spike in all discharge lights, while video tends to be more forgiving to that color.
  • 0

#10 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 19 July 2007 - 01:33 PM

Hello everyone,

I'll be shooting a low budget feature that involves a great deal of sodium vapor lights, it will be impossible for us to correct this fixtures, so I'm thinking about just matching the tungsten units to the sodium vapor lights and then remove it in post. It seems like there are different ways of going about this. What film stocks would be more appropiate for this kind of situation, I know that there is a Fuji stock specially designed for mixed light situations like this. I would appreciate any tips or advice.
Also, any lighting/gels recommendations for a test will be appreciated,

regards,

Francisco

I saw a demo of the Fuji stock you mention. It was designed for mercury vapor and flourescent lights. In addition to RGB, it had one extra light sensitive layer that would specifically pick up on the blue-green mercury spectral spike, and create a sort of built-in hold out matte. I don't know if they made a second version for sodium, though.

The big problem if your light is all sodium is that it really doesn't have colors. It's just two strong spikes, at 589.0 and 589.6 nanometers. That comes from the way electrons change energy levels in the sodium atom. There really is nothing that any emulsion, chip, or retina could do about that. The Fuji hold-out approach works if you have other light sources, like the phosphor part of flourescent light. It discards the spikes, and if the only light you have is the spikes, that leaves pretty much nothing.

But remember that human eyes don't really see colors either under sodium light. Everything appears in shades of red-orange, like Baird system TV. Is it appropriate to your story to have nice pretty but unrealistic full color light here, or is it better told in sodium monochrome? Perhaps sodium keys with tungsten fill? Shoot tests and discuss them with your director.



-- J.S.
  • 0

#11 F Bulgarelli

F Bulgarelli
  • Basic Members
  • PipPipPipPip
  • 409 posts
  • Cinematographer

Posted 20 July 2007 - 12:19 AM

To all of you: THANKS! This is very helpful, I'll let you know how the test went.
  • 0

#12 Xavier Plaza

Xavier Plaza
  • Basic Members
  • PipPipPipPip
  • 288 posts
  • Cinematographer
  • Guayaquil - Ecuador

Posted 22 July 2007 - 01:26 AM

Hello everyone,

I'll be shooting a low budget feature that involves a great deal of sodium vapor lights, it will be impossible for us to correct this fixtures, so I'm thinking about just matching the tungsten units to the sodium vapor lights and then remove it in post. It seems like there are different ways of going about this. What film stocks would be more appropiate for this kind of situation, I know that there is a Fuji stock specially designed for mixed light situations like this. I would appreciate any tips or advice.
Also, any lighting/gels recommendations for a test will be appreciated,

regards,

Francisco



Hi Francisco i found the photography in 8 MILE very attractive in mixing fixtures sodium vapor and mercury vapor. Take a look on this post where Rodrigo Prieto talk about how he achieve this look on 8 MILE

http://www.cinematog...n...0&hl=8 mile

Hope help you

Xavier Plaza
  • 0

#13 F Bulgarelli

F Bulgarelli
  • Basic Members
  • PipPipPipPip
  • 409 posts
  • Cinematographer

Posted 23 July 2007 - 12:41 PM

Thanks for the link Xavier,

I'm working with an Old Sr2, it doesn't have an adjustable shutter, probably 180o only, i will double check, do you guys think that's going to create a lot of problems for me as far as the flicker is concerned? I'm going to test it, just trying to get ahead.


Thanks

Francisco
  • 0

#14 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 23 July 2007 - 01:38 PM

... do you guys think that's going to create a lot of problems for me as far as the flicker is concerned?

Unfortunately, yes. The sodium arc drops out around the zero crossing of the AC power, so you have 120 pulses of light per second, with dark periods between them. That means 5 light pulses in 1/24 second. If the pulses are narrow, you could have three on one frame, two on the next, then 3 and 2 over and over again. Not knowing how wide the pulses are, the only thing to do is test it.



-- J.S.
  • 0

#15 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 25 July 2007 - 09:50 PM

Unfortunately, yes. ... That means 5 light pulses in 1/24 second.


And 2.5 pulses per 1/48 second exposure, same as HMI's (except for the darker "off" period). Or am I missing something?

I've never had a problem with sodium vapor streetlights at 24fps crystal and 1/48 shutterspeed, nor video at 29.97. The only time I've experience flicker from sodium vapor was with a PAL video camera here in the US, and the flicker was very, very subtle. Simply switching the camera's shutterspeed to 1/60 took care of it.
  • 0

#16 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 27 July 2007 - 05:18 PM

I see now that I'm wrong about it varying that much from frame to frame.

Suppose that each pulse consists of equal durations of on and off, half and half. The shutter is open for 2 1/2 pulses, which is five half pulses. That five could be two ons and three offs at one extreme, or three ons and two offs at the other. If your power source for the lights is in good sync with the camera, that'll be constant throughout each take, but random from take to take. If the power frequency drifts significantly, the exposure will also drift up and down by about half a stop.

If the "on" part of the cycle is more than half, the effect will be smaller. That was the theory behind the square wave HMI ballasts. Changing the shutter angle to get the exposure equal to an integer number of light pulse cycles fixes the problem, too.



-- J.S.
  • 0

#17 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 27 July 2007 - 06:10 PM

I see what you're saying, and considered that before posting. But I figured as long as everything stays in synch each frame of a continuous take should be consistent. But maybe you're right that the chance of exposure difference between takes is greater compared to HMI's. Makes sense. I've just never noticed it in practice, so it's close enough to a "wash" for me ;) .
  • 0

#18 John Pytlak RIP

John Pytlak RIP

    (deceased)

  • Sustaining Members
  • 3499 posts
  • Industry Rep
  • Rochester, NY 14650-1922

Posted 30 July 2007 - 07:03 AM

The AC flicker issue with a direct discharge lamp (sodium vapor, mercury vapor, neon) will be more severe than with a flourescent tube, where the phosphor's light output has some "lag" or afterglow. Of course, in a large filament tungsten lamp, there is lots of lag or afterglow between cycles, and so very minimal flicker with typical studio tungsten lighting.
  • 0

#19 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 30 July 2007 - 01:48 PM

The AC flicker issue with a direct discharge lamp (sodium vapor, mercury vapor, neon) will be more severe than with a flourescent tube, where the phosphor's light output has some "lag" or afterglow.

The other interesting thing about flourescent afterglow is that it's not the same color as when the mercury arc is happening. The last ones I looked at -- a long time ago -- were more of a warm yellow, and not as bright, in the afterglow period. Of course, there are a huge variety of different phosphors, so I'd expect to see some variation in that color and brightness.

One way to get a feel for this sort of thing is to look at a distant light source at night through binoculars, and just jiggle the binoculars -- like doing a swish pan. Incandescents will appear as long streaks, and strobed sources will appear as a series of dots or dashes. This lets you look at light output spread out over small intervals of time, much like an oscilloscope lets you look at wave forms (but without the nice sync feature, of course). It'll let you make an estimate of what the duty cycle is -- what percent of the time the light is on.



-- J.S.
  • 0

#20 Michael Nash

Michael Nash
  • Sustaining Members
  • 3330 posts
  • Cinematographer
  • Pasadena, CA

Posted 30 July 2007 - 04:31 PM

The other interesting thing about flourescent afterglow is that it's not the same color as when the mercury arc is happening.


This might also explain the periodic blue-orange color shift you get from HMI's when using Cleascan (faster "nonsafe" shutterspeeds) with video.
  • 0


Rig Wheels Passport

Broadcast Solutions Inc

Media Blackout - Custom Cables and AKS

FJS International, LLC

rebotnix Technologies

Opal

Technodolly

Willys Widgets

CineTape

Paralinx LLC

Abel Cine

Gamma Ray Digital Inc

The Slider

Wooden Camera

Glidecam

CineLab

Visual Products

Ritter Battery

Metropolis Post

Tai Audio

Aerial Filmworks

The Slider

Visual Products

CineTape

Gamma Ray Digital Inc

Abel Cine

Glidecam

Tai Audio

Broadcast Solutions Inc

Technodolly

rebotnix Technologies

FJS International, LLC

Willys Widgets

Ritter Battery

Opal

Metropolis Post

Wooden Camera

Paralinx LLC

Aerial Filmworks

CineLab

Media Blackout - Custom Cables and AKS

Rig Wheels Passport