[] lights with no line banding

Z

zsoldier

Member
do you guys have any suggestions/recommendations for cheap studio lights(soft light & hard light) that don't produce line banding @ high shutter speeds?! I heard there cheap soft boxes $100-$150 but they produce banding @ 60 shutter speed!? List your lighting set up !!
 
I must admit, I had no clue what you meant by "line banding." However, a quick google fixed that. If what I found out about "line banding" is the same as what you mean, the lights have nothing to do with it. Line banding is a camera sensor issue, usually seen at higher ISO's.

Now.... Flicker...is probably what you meant. And, yes, certain lights flicker at high shutter speeds/rates. You asked for cheap, so stick with just about any tungsten source and shoot away at just about any shutter speed you desire.

Hope that helps.

Joe
 
Well, the lights do have something to do with it, in that certain types of lights will interact with MOS sensors to cause it, and certain other types won't. Sunlight and tungsten light, for example, are safe to use at any shutter speed and won't cause line banding on MOS cameras.

Magnetic-ballast HMI and fluorescent and mercury-vapor and sodium-vapor lights will always cause the banding if you're shooting at any shutter speed other than the power frequency (i.e., 1/60 in USA, 1/50 in Europe). The solution here is to always, always, always use a shutter speed that is an exact divisor of your power frequency (so, 1/60 or 1/40 or 1/30 or 1/24 in USA, 1/50 or 1/25 in Europe, but never 1/48 in either).

High-frequency electronic ballast HMI and fluorescent should be immune from the banding.

LED lights are a basket of maybe. Some work fine with no banding, some cause banding regardless of your shutter speed. Dimming LED's are especially prone to causing banding that cannot be fixed by changing the shutter speed.

Finally: banding is a product of using an MOS/CMOS sensor. CCD sensors are immune to the banding artifact, but can instead be affected by a bit of color shift, especially when using a high shutter speed under magnetic-ballast lights.
 
That link only goes to a general product listing, not any specific lights, so it's hard to tell which ones you were looking at exactly.

But they are all tungsten lamps on that page, which are continuous sources and should never present any problems with flicker, banding, etc.

- Greg
 
No Cool Lights fluorescents will cause a flicker problem. Whatever problem that is with a Fulham ballast seems confined to that brand as we never see this issue at all.
 
Hey Richard
Do you have any footage of your lights shot at different shutter speeds online? Not necessarily footage of something that's lit, but actual footage of the bulbs at different shutter speeds?
 
The main tests we've done are pointing at a white wall in the past. I ran one just now pointing at tubes. No problems through all the shutter speeds on my T2I and 24P. But we have thousands of these units out there since early 2007. I do think we would have heard about that by now if there was an inherent problem. Anyone can run their own test if they choose on any of those models past and present. Kino Flo lighting doesn't have this problem apparently but neither do we.
 
Last edited:
Sorry, I wasn't trying to imply that your fixtures wouldn't perform as you stated. I've read through many of your articles (especially DIY HMI ones) in the past and found them very helpful. I'm just trying to figure out why my high speed electrical ballasts do this! In your design, did you add anything specific to overcome this? I'm a little confused just because I've seen so many people talk about dividing by 60 or 50 (depending on where you are in the world) for finding acceptable frame rates / shutter speeds.
 
I didn't do anything other than a selection process. I just chose a ballast with the highest frequency I could find (40KHZ) then test, test and more tests back in the beginning. Then didn't stray away from those ballasts over the years because it is so hard to find a reasonably priced good one. In the beginning when I was doing my DIY work, I used some T5 ballasts from Advance and those are more expensive but far less in output frequency than what we use now in our studio and portable fixtures (Advance is 25KHZ vs. our 40KHZ). The problem didn't exist with those (at least at that time). So I think there must be some other issues with certain ballasts out there. I've heard mixed reactions on Fulham. Some love and some think they do have this odd flicker issue. I'd stick with an Advance or Philips (T8 or T12 if that's what you're using) and see what happens. A lot of T12 ballasts out there are terrible. We have looked at some in the past when I was thinking about doing T12 fixtures. Couldn't find a single one that I liked and didn't have some flicker issue. I think its mainly because T12 is a very mature flo tube and there just isn't a lot of new development going on for it. T8 is more likely to benefit from more competition and therefore more ballasts to choose from.

One thing I was reminded of just now when testing, always be sure there is no light contamination from any other place when doing these tests (you probably did, but just reminding). When I was trying it I was too lazy to turn out the overhead lights in our office. I know these aren't safe for video but was trying anyway. There was an odd frequency "beating" (for lack of a better term to put it in terms used in sound for similar phenomena) of our light at only one shutter speed with the overhead light until I turned off the overheads. Then all was fine at that same speed. So be sure there isn't some light coming from some other place where you're doing your tests. What you would be seeing in this case is the two lights out of phase with each other and that would only happen at certain shutter speeds. In particular when the shutter speed is higher and the tube is getting less bright so less competition for the overhead light to break through and cause issues with. Kind of the visual equivalent of frequency beating.
 
Last edited:
Ok, I redid all my tests just hoping that it had been some sort of light contamination that I had somehow missed. Also, it occurred to me to try one ballast at a time (I have 4 foot 4 bank, powered by 2 ballasts each powering 2 bulbs), hoping that it might have been some sort of phasing problem. Alas, I'm still getting that same banding. I did read of another guy having luck with the Advance ballasts, so maybe I'll have to do those next. Frequency of the Fulham WH5-120-L, as stated on their website, is "Frequency: 50/60 Hz." So I'm still at a loss. It's strange that lighting keeps me up at night -- I met an electrician the other day and I think I actually bored him. Anyway Richard -- thanks for the help. Also, since it still pertains to the thread topic, I feel like this problem would be solved by an induction light (hint hint)! You messing around with those at all? All the retrofit induction bulbs I see on the internet seem to originate in China. Those still don't seem like they're a good match for CFLs though, because of their lower luminous efficiency at low wattages. One day... ONE DAY!
 
You know, I'm pretty sure that "Frequency: 50/60 Hz" refers to the line input frequency... or is it saying that it will operate at whatever frequency the line inputs?
 
DaneCannon said:
Frequency of the Fulham WH5-120-L, as stated on their website, is "Frequency: 50/60 Hz."
Click to expand...

That's the input frequency and its the same on all AC ballasts. With electronic ballasts you have an input frequency and output frequency and they are most often different with the output one being quite a bit higher. With magnetic ballasts the input and output frequencies are the same and are always 50 or 60 hz.

You don't know the output frequency of your model because for some reason, its not a stated spec for most of these companies. To find out, you often have to call a tech support or application engineer within the ballast company. Not sure why they don't just tell people in the normal specs.

It sounds like you're using the T12 type bulbs and that will be a very hard road for DIY. So few people attempt it for this very reason. Sticking with T8 will bring a lot more choices in ballasts and bulbs as well.

On induction bulbs I have looked at using them but its an issue of standards. We pretty much stick with standards so people will be comfortable in being able to find replacement tubes if ours aren't available. Induction bulbs don't have any standards yet and there are so many different ones and they may all require their own proprietary ballast. I don't know if that's a technology which will ever find any traction since its coming out so late in the life of fluorescent. With LEDs on the horizon, it may never make it.
 
Beating is a good term, Richard. From my electrical engineering days, long, long ago, when you combine two frequency sources you get the original two plus the sum and the difference frequencies. The difference is called the beat frequency in any branch of physics from sound to electricity and in lots of practical situations it's low enough that you can hear or see it.
 
When you have two analog audio oscillators running at the same frequency, you can hear what we are talking about. They'll never be perfectly sync'd so you can hear the differences. All these concepts follow through to light as well of course.
 
You must log in or register to reply here.
Back
Top