Jump to content


Photo

Light Flickering


  • Please log in to reply
20 replies to this topic

#1 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 19 November 2009 - 02:53 PM

First off my camera is an Eclair ACLII with a crystal sync motor that runs at 8, 24, 25, 50, and 75 fps and has a 144 degree shutter.

Since this issue could related to electricity I should say that I live in the US.

This was my setup:
Interior w/halogen lights (500 watt)
Subject sitting on a chair facing the camera
2 keylights facing the subject at a 45 degree angle
1 backlight behind the subjects right.

I filmed at 50 fps. The issue is that it appeared that I got NO flickering from the front two lights, but I did on the backlight. All lights were 500watt tungsten and part of a Lowel kit.

I'm not sure if this is because of the frame rate, outlet, bulb, or what. If it was the 50fps that caught the flicker, wouldn't it be all over the shot, not just the backlight area? Can I shoot at 50fps without flicker? The only variable was that one light was plugged into a different outlet. It's a large apartment building that has 30 some units and was built in the late 20's. The electricity has been updated and has circuit breakers though. I was also using extension cords, some longer than others. I have shot many times at 24fps with no flicker.

Any help would be appreciated.
Thanks,
Tom
  • 0

#2 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 20 November 2009 - 03:16 AM

What does the flicker look like? Random or regular? If it's regular, single frame thru it and count out how many flicker cycles you have in a shot, and how long the shot is.




-- J.S.
  • 0

#3 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 20 November 2009 - 11:15 AM

Thanks for the response John. The flicker happens at regular intervals. I'm not in front of the footage, but I would say that playing back that (50fps) footage at 23.987, I would guess one on and off cycle would take approximately 6-8 frames. Although I'm not sure I completely understand your question. I'll have to go back and analyze it. Going by memory, always dangerous for a youngster such as myself:) I don't believe it was like one frame light and the next dark. I seem to remember it was more like a sine wave if that makes sense. I'll take a look tonight and post some footage.

Thanks again. I'm afraid to shoot more than 24 fps under lights :blink:

Tom
  • 0

#4 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 20 November 2009 - 04:18 PM

A sine wave does make sense if this is a beat frequency between the power and the frame rate. With 60 cycle power, we get a light pulse on both the positive and negative parts of the sine wave, with a dip at each zero crossing. So, that's 120 light pulses per second. You were shooting 50 frames per second. So, for every 5 frames, you had 12 light pulses. If the power frequency is the cause, you should expect a complete cycle every five frames. If it's a different cycle length, or the length varies, then it's not power frequency related.

With tungsten lights, the filament can cool down a little at the zero crossings, but it doesn't drop to zero like old magnetic ballast flourescents did. So, the difference between the high and low light levels of the flicker is usually much smaller with tungsten. Usually we don't see flicker from tungsten lights, it's so small. The bigger the lamp, the less the flicker, because the more massive filaments cool slower.

What's strange, and points away from power frequency, is that only the back light showed the flicker. That makes me suspect a bad plug, switch, lamp holder, etc., introducing some variable resistance or arcing. But those would give you a more random flicker.



-- J.S.
  • 0

#5 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 21 November 2009 - 12:17 AM

Hi John,

I'm hoping you can take a look at this clip. Or anyone else who can recognize the problem.

flicker @ 50 fps

Thanks in advance,
Tom
  • 0

#6 Adrian Sierkowski

Adrian Sierkowski
  • Sustaining Members
  • 7118 posts
  • Cinematographer
  • Los Angeles, Ca

Posted 21 November 2009 - 11:05 AM

Just gave it a look and have to say it's like nothing I've seen before. It doesn't much distract my eye, for what that's worth. it almost looks like a floro flicker... but I know that's not the case... it might've been the filament in the head if the head was somehow vibrating, I've seen that before but only on clear bulbs, where the filament "wiggles" a bit and can look strange... though normally missed @24 because it's subtle, @ 50 it might've slowed it down enough to be a distraction. Though this filament wiggle normally has happened to me on clear household bulbs at like 200/300 (these bulbs: http://www.acehardwa....aspx?SKU=35525 ). Best I got tom. Aside from that, really interesting shot!
  • 0

#7 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 23 November 2009 - 01:59 AM

I'm hoping you can take a look at this clip.


It doesn't work on the slow connection I have here at home. I'll be in the office again on Wednesday, and look at it if there's time....



-- J.S.
  • 0

#8 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 23 November 2009 - 09:29 AM

It doesn't work on the slow connection I have here at home. I'll be in the office again on Wednesday, and look at it if there's time....

-- J.S.


That would be great John. I didn't compress it too heavily as it's a subtle flicker, so pardon the file size.

Adrian,
Thanks for the feedback. This is the bulb I used:
Bulb
They are relatively new. Maybe 10 hours total.

It is just distracting enough. On this particular shot, the person, who is turning 360 will be composited onto a landscape (camera turning 360). Since both shots will be interacting with each other, one probably won't be able to see the flickering. However I don't want to shoot at 50 fps with my lights if this problem persists. In another month I'll have some tests back, but wanted to check with you guys as well.

Thanks to you both for your help thus far.
Tom
  • 0

#9 Adrian Sierkowski

Adrian Sierkowski
  • Sustaining Members
  • 7118 posts
  • Cinematographer
  • Los Angeles, Ca

Posted 23 November 2009 - 09:32 AM

Yeah doubtful that type of bulb would've had any form of of filament movement... perhaps, might be wise to shoot @ 48 in future as opposed to 50? Seems a bit "rounder" of a number here in NTSC land/24P land...
  • 0

#10 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 23 November 2009 - 09:58 AM

Yeah doubtful that type of bulb would've had any form of of filament movement... perhaps, might be wise to shoot @ 48 in future as opposed to 50? Seems a bit "rounder" of a number here in NTSC land/24P land...



you know Adrian, I would LOVE to shoot at 48, but alas the "Eclair" (being of French decent) only shoots 24/25, 50, 75. I would think at 48 fps, I would be golden. Oh well. At least there is the sun.

Tom
  • 0

#11 Adrian Sierkowski

Adrian Sierkowski
  • Sustaining Members
  • 7118 posts
  • Cinematographer
  • Los Angeles, Ca

Posted 23 November 2009 - 10:01 AM

Ahh, damn :/ Those crazy french! It's odd, though, that it's do 24/25 then skip right to 50.
  • 0

#12 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 25 November 2009 - 04:08 PM

I was able to download it here in the office. I see some density variation in the solid dark gray BG, but that looks like a stock or processing issue. The lighting on the actress doesn't show any flicker that I can detect.




-- J.S.
  • 0

#13 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 27 November 2009 - 11:10 AM

I was able to download it here in the office. I see some density variation in the solid dark gray BG, but that looks like a stock or processing issue. The lighting on the actress doesn't show any flicker that I can detect.

-- J.S.


Thanks John,

That's funny because I thought I saw an exterior shot in daylight, that had that same kind of strobing, but ever so subtle. I thought, well that cant' be a light.

So I need to show my thin, but growing, layer of experience and ask if you could clarify what you mean by "density variation in the solid dark gray BG, but that looks like a stock or processing issue." and how can I minimized the possibility of this in the future.

Thanks again,
Tom

Hope all that celebrate had a nice Thanksgiving yesterday.
  • 0

#14 David Rakoczy

David Rakoczy
  • Basic Members
  • PipPipPipPip
  • 1579 posts
  • Cinematographer
  • USA

Posted 27 November 2009 - 11:28 AM

I agree with John. Tom, under Tungsten lights you can shoot (any) speed.. except when you get into the 300fps realm. Under HMI you need to shoot at speeds divisible by 6. Fluorescents by 6 as well however some Fluorescents and Neon are hit and miss even at a speed divisible by 6. I have had Fluoro and neon flicker at 48fps.
  • 0

#15 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 27 November 2009 - 05:18 PM

This kind of thing is usually a raw stock problem, or a problem with handling of the exposed film prior to developing. Old film that was stored with the cans on edge will have the layers very slightly compressed at the bottom, while the sides can hang a little loose, letting more air get in there. Film that is shipped to or from distant location can sometimes get stowed in the airplane next to radioactive stuff like medical isotopes. We had that happen on an MOW years ago. If a stack of cans sits with one side facing the sun from a window or some other heat source, you can get this kind of once around the roll damage. Basically anything that makes one side of the rolled up film different from the other is what you're looking for.

Labs can also have problems that look like that, but that's less common than storage and handling problems. I'll leave it to Dominic Case and Robert Houllihan to describe those, as that's their area of expertise.




-- J.S.
  • 0

#16 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 27 November 2009 - 05:36 PM

Under HMI you need to shoot at speeds divisible by 6. Fluorescents by 6 as well however some Fluorescents and Neon are hit and miss even at a speed divisible by 6. I have had Fluoro and neon flicker at 48fps.


That's because speeds divisible by 6 will get you some right answers, but also some wrong ones, like 48. What you have here in the 60 Hz part of the world is 120 light pulses per second. What you need to do is get the same amount of light on each frame. There are two ways to do that, the easy way, and the hard way.

The easy way is to use a frame rate that corresponds to an integer (i.e., whole) number of light pulses: One light pulse would be 120 fps, two light pulses 60 fps, three light pulses 40 fps, etc. What you want is 120 divided by a whole number:

120, 60, 40, 30, 24, 20, 17.143, 15, 13.333, 12, etc.

Doing it the easy way, you can use any shutter angle you want.

The hard way is to use a frame rate other than those above, along with a shutter angle that works out to a whole number of light pulses reaching the film. For instance, the OP wanted to run 50 fps. With a 150 degree shutter, he would have an exposure of (1/50)x(150/360) = 1/120 sec.




-- J.S.
  • 0

#17 timHealy

timHealy
  • Basic Members
  • PipPipPipPip
  • 1252 posts
  • Other
  • New York

Posted 28 November 2009 - 02:09 PM

The only flickering I see is what was on the background. So what was lighting the background? I don't see flickering on the backlight. Was there and errant light hitting that background and not a film or tungsten light?

Am I missing something?

Best

Tim
  • 0

#18 Tom Hepburn

Tom Hepburn
  • Basic Members
  • PipPipPipPip
  • 341 posts
  • Other
  • Chicago-land

Posted 30 November 2009 - 10:09 AM

This kind of thing is usually a raw stock problem, or a problem with handling of the exposed film prior to developing. Old film that was stored with the cans on edge will have the layers very slightly compressed at the bottom, while the sides can hang a little loose, letting more air get in there. Film that is shipped to or from distant location can sometimes get stowed in the airplane next to radioactive stuff like medical isotopes. We had that happen on an MOW years ago. If a stack of cans sits with one side facing the sun from a window or some other heat source, you can get this kind of once around the roll damage. Basically anything that makes one side of the rolled up film different from the other is what you're looking for.

-- J.S.


I see. And you know John that might have been the issue as it was in the fridge for a week or so jockying for position. If that is the case, wouldn't the flicker happen every 10-15 frames based on the size of a 200 ft roll (even longer on a 400 ft) as opposed to the every other frame or so as it appears to be?

Tim, the lights are all 500 watt tungsten lights from the same Lowel kit.

Thanks again,
Tom
  • 0

#19 David Rakoczy

David Rakoczy
  • Basic Members
  • PipPipPipPip
  • 1579 posts
  • Cinematographer
  • USA

Posted 30 November 2009 - 10:38 AM

Film is a bit more rugged than that... It has been to to the top of Mt. Everest and all over the world under grueling conditions. Tom, I think you left the overhead Kitchen Fluoros on and it is they we pulsing on the background. ;)
  • 0

#20 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 30 November 2009 - 06:38 PM

If that is the case, wouldn't the flicker happen every 10-15 frames based on the size of a 200 ft roll (even longer on a 400 ft) as opposed to the every other frame or so as it appears to be?


I don't see anything happening as fast as every other frame. I see a longer period pulsation, and only on the dark gray part of the BG. It sounds like Tim Healy's seeing that same issue. Another thing to look for is if the pulsation rate speeds up or slows down between where it first appears and where it last appears. That's evidence that something happened to the film while it was rolled up.

As David said, it would take more than a week in the fridge to cause what we're seeing.



-- J.S.
  • 0


Ritter Battery

Paralinx LLC

Aerial Filmworks

Willys Widgets

Wooden Camera

Broadcast Solutions Inc

Opal

Metropolis Post

rebotnix Technologies

Gamma Ray Digital Inc

Tai Audio

The Slider

Rig Wheels Passport

FJS International, LLC

Glidecam

CineLab

Media Blackout - Custom Cables and AKS

Technodolly

Abel Cine

Visual Products

CineTape

Glidecam

FJS International, LLC

Abel Cine

rebotnix Technologies

The Slider

Technodolly

Tai Audio

Wooden Camera

Gamma Ray Digital Inc

Willys Widgets

Broadcast Solutions Inc

Aerial Filmworks

Paralinx LLC

Visual Products

Ritter Battery

CineLab

CineTape

Metropolis Post

Media Blackout - Custom Cables and AKS

Rig Wheels Passport

Opal