Jump to content

Difference between kinds of lights? Fluorescent? LED?


scott karos

Recommended Posts

One thing I want to know more about is the kinds of light filmmakers use. Im talking more specifically about the actual light source like LED or fluorescent light bulbs.

 

Whats the difference between them and how do they produce a different affect?

Link to comment
Share on other sites

  • Premium Member

That's an absolutely enormous subject that's really too broad to cover in a forum question. Still, to be brief:

 

main4171.jpg

- Traditional "tungsten" lighting uses electricity to heat a fine wire so that it glows white hot, kept inside a vacuum-filled glass bulb so that it doesn't burn out. The colour quality is good, but the colour temperature is low (the light is orangeish) and they create a lot of heat so efficiency is poor (not much light for the amount of electricity used). The electrical filament is small and the light can be applied hard, or diffused to be soft. Cheap to buy, expensive to run.

 

arrisun-with-cases.jpg

 

- HMI is a type of gas discharge light. Electricity is passed through a bulb filled with a very small amount of gas including mercury, which emits ultraviolet light when electrically excited. This ultraviolet light in turn excites gaseous iodine which emits white light. Colour temperature is high (the light is bluish), colour rendering is good. Efficiency is also good. Expensive. Large, heavy separate control unit, the ballast, sits on the floor near the light. Old types will not restart without cooling down first. Applications are as tungsten.

 

kino-4x4.jpg

- Fluorescent tubes are actually another type of mercury discharge light, but instead of exciting iodine, the UV light excites a phosphor on the inside of the tube. The electrical discharge is of lower intensity than an HMI and the lamp is therefore larger, so it can only really be used to create a soft light. Can be fairly cheap, but colour quality is not as good as HMI or tungsten, except on the very best units. Also uses a ballast, but the lower power means it's smaller and lighter.

 

1x1-bicolor.jpg

- The physics behind LEDs are too involved to describe here, but in white-emitting LEDs, a blue light is produced then used to excite a phosphor as in fluorescent tubes - the phosphor is the yellow colour visible in a switched-off LED. Efficiency is ranges from as good as HMI to somewhat better than HMI, although the more efficient types have poor colour rendering. Various configurations of LEDs have been built to provide hard light, soft light, or the sort of rather unusual projecting softlight created by things like Litepanels (and their many competitors). Can be very expensive and even the best types tend to have poor colour rendering.

 

Hope that helps.

 

Phil

Link to comment
Share on other sites

That's an absolutely enormous subject that's really too broad to cover in a forum question.

 

I wrote a white paper for our company newsletter on just the power considerations of each of the light sources Phil mentions above and it runs over 300'000 words. The white paper is available at http://www.screenlightandgrip.com/html/ emailnewsletter_generators.html.

 

Guy Holt, Gaffer,

ScreenLight & Grip,

www.screenlightandgrip.com

Edited by Guy Holt
Link to comment
Share on other sites

There has been a lot of discussion in the industry and on many websites about the use of the new LED fixtures in cinematography. Litepanel came out with the first really widely used units for filming with their on-camera bricks and their 1x1 panels. Many embraced them while others were very concerned with the lack of full color, harsh texture of light generation and multiple shadow rendition of LED sources.

 

Three years ago there was a special presentation made by the Academy of Motion Picture arts & Science and the ASC at the annual National Association of Broadcasters convention in Vegas that compared the three leading LED unit manufacturers against standard tungsten lighting. The results were an eye opener as all three recorded colors differently not only from the tungsten light units but also from each other. While there have been many Asian manufacturers that have been releasing very inconsistent lighting units at low price ranges since then, there have been great strides made in the last three years in standardizing the LED output in both color and consistency by more respected and recognized lighting manufacturers here in the USA. The most notable massive improvements have been made by Cineo and their remote phosphor units, by mole and their LED fresnels, spacelights and now softlight units and kinoflo with their new Celeb line of softlights.

 

My friend Joe DiGennaro,a local 600 cinematographer who works in the research dept of the Academy of Motion Pictures Arts & Science (The Oscars) pointed me towards Cineo during their development stage. He, personally, really likes their lights and he's definitely the expert when it comes to LED lights and how they record on digital and film - having tested every manufacturer against every major camera and film stock. I was fortunate enough to use them on a low budget feature and then again on a corporate video shoot. I love the softness and the color, which when i used my spectrometer on them was very high and rich.

 

The Academy has been working on an app that allows you to plug in the camera (with its sensor) and then the brand LED light which will then give you a color rendering idea. Its a rather complicated thing. Here's a link to one of their articles on the subject of LEDs and color.

http://www.oscars.org/science-technology/sci-tech-projects/solid-state-lighting-report

 

And here is a free app that the Academy has put out. Joe says there is a second app about to come out that is even more conclusive -

http://www.oscars.org/science-technology/sci-tech-projects/academy-color-predictor

 

The fact is that the industry has embraced LED lighting now. On Project Runway we use LEDs and Kinoflos almost exclusively for our locations and the work rooms lighting. The progress being made, especially by Mole and Cineo, has really brought the LED into the realm of dependability. There are still things to get over but much has been accomplished and more will be. England and the USA have made I clear that the governments will be terminating the manufacturing of tungsten element lighting in the very near future. Those of us in the lighting business have been fighting against this, but losing. There s a misconception that fluorescent and LED lighting is more green”. While they do consume much less power and generate much less heat, once they are disposed of they are actually more toxic to the environment than glass and tungsten wire. But they are looking only at the financial costs of power consumption and not the beauty of full spectrum light. And money always seems to win out over art.

Edited by David Landau
Link to comment
Share on other sites

England and the USA have made I clear that the governments will be terminating the manufacturing of tungsten element lighting in the very near future.

 

The last I heard from globe manufacturers was that certain industries, motion picture included, had been exempted. While they may get more expensive, Quartz Halogen globes will be available for the fore seeable future.

 

The fact is that the industry has embraced LED lighting now. On Project Runway we use LEDs and Kinoflos almost exclusively for our locations and the work rooms lighting.

 

A more accurate statement would be that those segments of the industry, like realty television, news, and sports for whom expediency out ways quality, has embraced LED lighting. We see very few LED fixtures come through on rental lists for commercial and feature productions except for DC operation on car or boat rigs. While Mole is doing great things with LEDs, it remains to be seen whether the industry will embrace it.

 

The most notable massive improvements have been made by Cineo and their remote phosphor units, ... when i used my spectrometer on them was very high and rich.

 

If you compare the Spectral Power Distribution Graph for the Cineo fixtures to that of a Phosphor White LED, they are similar and in no way resemble that of a true black body radiator like a Tungsten filament (use this link for graphs.)

 

The inherent limitation to the “Stokes shift” process by which a portion of a “pump” color is transformed from shorter wavelengths to longer wavelengths in Remote Phosphor LEDs, like the Cineo fixtures, is that it works in only one direction – that is why Remote Phosphor LEDs don’t emit color wavelengths shorter than their pump color. Another, inherent shortcoming to this approach to generating “tungsten” light from an LED is that there is a tradeoff between lumen output and warmer color temperatures (see my newsletter article for details.)

 

The Cineo fixtures clearly suffer from these limitations. Compared to a black body radiator source like a Tungsten filament, the output of the Cineo fixtures drop off steeply below its pump color (which is at 425nm verses 465nm of the Blue LED used in typical Phosphor White LEDs) so that it puts out no wavelengths below 400nm. By comparison a Tungsten filament continues to generate light with wavelengths well below 400nm which is why tungsten light will render violet colors better. While the Cineo fixtures appear to generate more light in the medium blue-cyan-turquoise range from about 465-510nm than the typical Phosphor White LED, its’ long-wavelength cutoff is still at about 625 nm where a tungsten filament continues to generate light all the way out. Because of this rapid drop off of wavelengths above 625nm, pinks, reds, oranges, and other long wave-length colors will look dull under the Cineo fixtures, compared with how they look under a Tungsten source which continues strong all the way out on the long-wavelength end.

 

Another problem with Remote Phosphor LED fixtures is that their output is inconsistent because it is dependent on maintaining a very specific wavelength output of their emitters (not easily accomplished with LEDs.) Trish Mass, with PRG (the fore-runner to Cineo), once wrote: “The specific wave-length of blue is paramount to fluorescing the rare earth elements (in the phosphor panels.) We have to hit it with a pretty specific nanometer. Without that blue, you just have one really dense yellow light blocker.”

 

However there are a number of factors that make the output of LED emitters inconsistent. For example they are affected not only by the imprecise binning practices and manufacturing tolerances of manufacturers, but also by the thermal management in the fixture, the ageing of the phosphors, and even the ambient temperature. For example, a one-degree shift in the junction temperature of the blue InGaN LED (pump color) in remote phosphor LEDs, will cause a +/- 2nm shift in the dominant wavelength. If compounded by the average wavelength variation of +/- 2nm of blue InGaN LEDs, a 5nm divergence from the prescribed 455nm wavelength of the pump color will create a color inconsistency of 5 MacAdams ellispses. While not readily apparent to the eye, image capture systems (both film & digital) will easily see this variation. For this reason the better multi-emitter designs, like the new Arri L7 LED Fresnels and the Gekko's Kleer Colour® technology, incorporate a color-feedback system of self-monitoring sensors to ensure stable color across a range of output levels, as well as correcting changes in performance caused by ambient temperature and component aging, which ensures consistent color temperature.

 

The Academy has been working on an app that allows you to plug in the camera (with its sensor) and then the brand LED light which will then give you a color rendering idea. Its a rather complicated thing.

When announcing the release of this app, Andy Malt with the AMPAS said:

 

“This app emerged from our conversations with cinematographers, production designers, costume designers and set decorators who were struggling to predict color reproduction when switching from traditional incandescent light sources to solid state lighting,”

 

If it is a struggle to switch, don’t do it. Quartz lights may run hot and consume a sh*tload of power, but they are comparatively inexpensive, offer a continuous color spectrum, and since all the imaging systems are designed for that spectrum, we know what to expect from them. The same can’t be said of LEDs.

 

For a little more than $200 I can buy a used Mole and put an $17 FCL globe in it. That works out to something like 7 cents per lumen. According to Cineo their LS fixture is 14 times as expensive per lumen. Except for rigs that require battery power, the difference in power consumption really isn’t that critical with these low-power units, whether tungsten or LED. But, the money saved by buying or renting a simple, conventional tungsten unit over that of a Cineo fixture could be invested in other production equipment that would actually add production value.

 

You have to be wary of all the claims made by LED head manufacturers because they all put a little spin on the scientific data, which has a tendency to cloud issues. For this reason, to pick the right LED luminary for a particular job it helps to have a thorough understanding of the technology. For our company newsletter I have put together an overview of the technology and what products are available for motion picture lighting (available at http://www.screenlightandgrip.com/html/emailnewsletter_generators.html#anchorHigh%20Output%20AC%20LEDs.)

 

Guy Holt, Gaffer, ScreenLight & Grip, Lighting Rental and Sales in Boston.

Link to comment
Share on other sites

I totally agree that nothing matches the color of tungsten. I greatly prefer it - as a gaffer and as a cinematographer. While Boston might be different, Shadowstone Lighting here in the NYC area rents out lots of LED and Kinoflo lights as do most of the other lighting rental houses. As you mentioned, LEDs are popular with TV news, sports, talk shows, corporate video and documentaries. In the past three years, I have never been on a feature that didn't have Kinos and litepanels in the truck. That's not to say they were always being used, but they are as common as china balls. They have their use and it is undeniable that they have become common place - thus my use of the term "embraced." Broadway now uses LED striplights for all their cycs and backdrops and most game shows use them for color accents and chasing. Will they replace tungsten? I hope not. On a recent sitcom I was the prelight best boy on we had all tungsten units in the air. The camera's had LED bricks on them and kinos were slipped into the set pieces.

 

LED units don't usually rent for much more (if any more) than tungsten units, so for a rental house who has to put out the big bucks to buy them, I can see a reason to avoid them. Productions rent rather than buy lights, so the cost factor isn't a factor. In fact, many producers want the LED fixtures to save on the electrical and air-conditioning costs, which is a reason why they've been installed in many TV news sets, talk shows and the White House press room. Their long lamp life saves money on replacing the lamps. The color mixing LED units save money on gels and the lumens of a red LED source four is higher than a tungsten source four with a red gel on it (which will burn out and have to be replaced).

 

There has been a lot of lobbying on behalf of the entertainment industry in the USA and England to keep tungsten lamps. Since home and architecture are the biggest markets for light bulbs, if they stop buying tungsten, sales will plummet, the price will go up and some manufacturers may decide that its not profitable enough to keep making them. I hope this will never become the case, but only time will tell.

 

I want to always light with tungsten. I also love shooting on film. Things change and we must adapt and make the best of it.

Link to comment
Share on other sites

  • Premium Member

There is a further practicality argument.

 

If I want soft tungsten or daylight balanced light, fluorescent is cheaper, has better colour, and has about the same efficiency as LED.

 

If I want hard daylight, HMI is still cheaper per lumen, has better colour, and about the same efficiency as LED.

 

About the only circumstance in which LED is overall a better bet is if hard tungsten-balanced light is needed and the cost of LED is smaller than the cost of power and aircon. This is all subject to caveats about the LED light being of any quality - there is terrible, cheap LED if you really want it. But to me the main problem with LED is not really colour output, or efficiency, it's price. It's just hopelessly too expensive for something that doesn't really do anything that can be done in other ways.

 

There are specialist exceptions - battery power (though fluorescents and small discharge lights compete) and that rather particular type of projecting softlight that things like Litepanels do, which isn't readily simulated with other things, at least not without a lot of messing about. I hesitate to say all this as it beckons accusations of luddism, but my concern really isn't the colour performance. It'll come, it exists already to some extent. It's price.

 

P

  • Upvote 1
Link to comment
Share on other sites

There is a further practicality argument.

 

If I want soft tungsten or daylight balanced light, fluorescent is cheaper, has better colour, and has about the same efficiency as LED.

 

 

Is this really true?

 

I think the problem with Florescent is the starter. My suspicion is that for small amounts of light output that LED's might be a lot more efficient by virtue of the lack of that starter. It seems on the face of it that you need something it the range of 5-7watt led to achieve a similar output to an 11w flo for example.

 

Also flo's use a lot of electricity when they first start up but then become more efficient over time but LED's don't have that start up cost. In contast LED's seem to really not like heat at all whereas flo's seem unconcerned about it mostly. Of course in most cinematography applications you just want to leave the lights on for a long time and so flo's seem well suited to that with the advantage of not pumping out a ton of heat. LED's also don't pump out heat much but they really don't like the heat they do pump out. I've toasted about 3 of those cheap corn on the cob LED's because there is nothing in the design for dumping the heat. Not even vents! You see LED's with heatsinks fins built into the design and this is because they need to dump the heat.

 

I note that larger LED panels tend to have fans in them too which can also not be such a good feature for cinematography. (...and probably doesn't help with efficiency either!) I suspect that LED's will be great in the bathroom where the light goes on and off a lot however.

 

Freya

Edited by Freya Black
Link to comment
Share on other sites

LED units don't usually rent for much more (if any more) than tungsten units, so for a rental house who has to put out the big bucks to buy them, I can see a reason to avoid them. Productions rent rather than buy lights, so the cost factor isn't a factor.

 

Let me know who rents LED lights for the same price as Tungsten lights, I would like to re-rent from them. The broadcast rental houses here rent the Arri L7 Fresnel, which has not quite the output of a 1k Fresnel for $125/day while a Tungsten 1k Fresnel rents for $40/day, i.e. you can rent three 1ks for the price of one L7. Cost is a factor and I have to agree with Phil when he says:

 

... to me the main problem with LED is not really colour output, or efficiency, it's price. It's just hopelessly too expensive for something that doesn't really do anything that can be done in other ways.

 

In David’s zeal for the Cineo LEDs, he is glossing over a major problem with all LEDs (remote phosphor included), which is that any gel (party, effects or color correction) on a discontinuous source like LEDs will not produce the same repeatable result that you’d expect from gelling a continuous source, i.e. tungsten. Interchangeable remote phosphor panels that limit you to only four Kelvin Temperatures, and at a cost of $340-$420 each, is a poor substitution for the hundreds of party, effects, and color correction gels that are available for tungsten lighting instruments at $6.50 a pop.

 

What you have to appreciate is that CTO, CTS, and CTB gels are a part of a finely calibrated imaging system that involves a highly specific light receptor (film emulsion or video sensor), light sources, and color correction or effects gels calibrated for both. Where that exists between film emulsions/video sensors and tungsten and/or daylight sources it is possible to mix dyes in a gelatin materials to create desired effects (it has taken decades to hone this system.) To use the available color correction gels (listed above) to correct LEDs is a misapplication of a finely tuned system of correction designed for continuous spectrum light sources only.

 

LED_3200K_Remote_Phosphor.jpg

 

While there is no spike in the green output of a remote phosphor LED, as you can see from the spectral distribution graph above for the Cineo Trucolor 3200 (black line), there exists a definite green/cyan bump, as well as a spike in the blue range that does not exist in the continuous spectrum of a tungsten light source (green line.) The greater proportion of blue and green/cyan in the Cineo Trucolor 3200 will result in an unexpected and undesirable result if a color correction (CTO, CTS, CTB), or color effect gel (Congo Blue, Bastard Amber, etc.) calibrated for the continuous spectrum of tungsten light is used on this discontinuous spectrum of a 3200K remote phosphor LED.

LED_5600K_Remote_Phosphor.jpg

 

The same is true of the Cineo Trucolor with the 5600K panel above. The quite prominent blue spike (black line) will likewise result in an unexpected and undesirable result if a color correction, or color effect, gel calibrated for the continuous spectrum of a continuous daylight source is used on the discontinuous spectrum of a 5500K remote phosphor LED.

 

Someday, Rosco, Lee, or Gam, will come up with gels calibrated for LEDs but I doubt it will be any time soon given that there is no standard spectral output to LEDs (Lee is beginning to provide gels, but they are only for the cool white LEDs commonly used in theatrical fixtures.) Perhaps, in the future, when LED technology has become standardized, and a system of calibrated gels exists, Konica Minolta will come out with a Color Meter suitable for measuring LEDs for photographic purposes; but for now no such color meter exists as it does for tungsten sources. Where there are no meters or gels calibrated to correct the discontinuous spectrum of LEDs (remote phosphor included), and the existing color correction gels have undesirable consequences when used on LEDs, the ability to color-correct LEDs is very limited (see http://www.screenlightandgrip.com/html/emailnewsletter_generators.html#anchorHigh%20Output%20AC%20LEDs for camera test results demonstrating this with a Lightpanel 1x1 Daylight Spot.)

 

Hal Smith summed it up very eloquently in a post on the CML when he said: “… If I light with tungsten, I know what the result is going to look like …. Yes, they're hot; yes, they're bulky; yes, they draw a poop pot full of electricity but dammit...I know what the result is going to look like...and don't have to give some post pro a bagful of money to straighten out what the LED's screwed up.”

 

Guy Holt, Gaffer, ScreenLight and Grip, Lighting Rental & Sales in Boston

Edited by Guy Holt
  • Upvote 1
Link to comment
Share on other sites

  • Premium Member

 

 

I think the problem with Florescent is the starter.

 

Very few modern fluorescent installs use starters and conventional ballasts in that way anymore. They're invariably electronic ballasts, especially in movie lighting. High frequency ballasts are a few percent more efficient than the older type.

 

 

 

Also flo's use a lot of electricity when they first start up but then become more efficient over time

 

 

That's not something I've seen substantiated in studies. The idea seems to be that older-style iron ballasts use a lot of power during the phase where the lamp is flickering during starting while the electrodes are warming up, but even if true, that's such a vanishingly small proportion of the total power that it ought to be entirely insignificant.

 

 

P

Link to comment
Share on other sites

 

Very few modern fluorescent installs use starters and conventional ballasts in that way anymore. They're invariably electronic ballasts, especially in movie lighting. High frequency ballasts are a few percent more efficient than the older type.

 

That's not something I've seen substantiated in studies. The idea seems to be that older-style iron ballasts use a lot of power during the phase where the lamp is flickering during starting while the electrodes are warming up, but even if true, that's such a vanishingly small proportion of the total power that it ought to be entirely insignificant.

 

 

P

 

Ah apparently this is a complete myth:

 

http://www.youtube.com/watch?v=qgM0N7GD5Ic

 

Good to know!

 

...but apologies for the tacky video.

 

Freya

Edited by Freya Black
Link to comment
Share on other sites

 

Ah apparently this is a complete myth

 

There is usually a kernel of truth to a myth and that is the case here. As can be seen in the Power Quality Meter reading of a single Image 85 fixture below, Image 85 fixtures exhibit a high inrush current (70A) when struck cold.

Image_85_Inrush.jpg

 

The reason for an inrush current 833% higher (70A) than the steady state load (8.4A) is that the electronic ballasts of Image 85 fixtures use large smoothing capacitors to convert rectified AC to DC before switching to a high frequency output (greater than 25kw Hz.) If we scale up this characteristic of Image 85 fixtures to the potential load placed on a company switch by a big effects show like R.I.P.D. (pictured below) that used 180 fixtures on their main stage, we see there is a potential inrush current on an individual 400A company switch of 2’660A on each phase leg (320A/8.4A per fixture = 38 fixtures per phase leg)(38 fixtures x 70A Inrush Current per fixture = 2660A Inrush Current per phase.) If the trip curve of electronic breakers on the company switches are not adjusted for this magnitude they will trip if all the lights are turned on at once. To avoid nuisance tripping, stage breakers should be programmed for an inverse time curve that accommodates a high inrush current that lasts for a duration of 24 cycles.

 

Image85_Product_Shot.jpg

 

Nuisance tripping is not the only potential problem related to the high inrush current of Image 85s. The triplen harmonics generated by Image 85 fixtures will accumulate on the system neutral, rather than cancelling each other out as the out of phase fundamentals normally do. In fact, the neutral will carry 176% of the average load on the individual fundamental phase legs. When you also take into account the high inrush current of Image 85s, there can be a devastating effect on the system neutral. If we take the RIPD rig for example, we see that there is the potential to severely overload the neutral bus of the switch.

Image85_RIPD.jpg

 

The 180 lamp heads pictured in this rig would require a service of 1’407A or 469A/leg on a 3-phase service. And since these ballasted lamp heads should not be operated on a dim channel, rigging crews will usually patch Image 85s into company switches in the perms, before bringing more power up from a company switch below. In this situation a rigging crew will not load the individual phase legs beyond 80% of their rated capacity – i.e. they will load each phase leg of a 400A/3-phase company switch in the perm to 320A. If each phase leg is loaded as such, we can expect a return on the neutral of 563A (320A x 1.76 = 563.20A) once the fixtures reach their steady state. When we factor in the high inrush current of Image 85s, we see there is the potential for a neutral conductor rated for 320A to carry 4’693A (563.20A x 8.33 = 4’693A.) Even though this magnitude of current will only travel on the neutral for 24 cycles it can cause damage to the conductor and upstream transformers because the neutral return of a distribution system has no over current protection. For these reasons hundreds of Images 85 should not be brought up all at once and “super neutrals” capable of carrying 200% of the load on an individual phase leg should always be used.

 

For more details about these issues see an article I wrote for our company newsletter that explains the electrical engineering principles behind these issues and how to resolve them. The newsletter article is available at http://www.screenlightandgrip.com/html/emailnewsletter_generators.html.

 

Guy Holt, Gaffer, ScreenLight & Grip, Lighting Rental & Sales in Boston

 

  • Upvote 1
Link to comment
Share on other sites

  • Premium Member

I don't think that's likely to be a particularly common circumstance for most people, to be fair.

 

Also, again, this isn't so much about lighting in particular as buck converters in general. They all have large input capacitances which will draw a lot of current when power is applied. I'm not sure there's much that could be done about that, other than engineering them with some sort of soft-start arrangements. For what it's worth, a modern electronic HMI ballast would do exactly the same thing if you happened to strike dozens of them all at once.

 

P

Link to comment
Share on other sites

To be fair, the tacky video I posted does acknowledge that the flo's use a huge amount more electricity on startup with break even occurring at a staggering 23secs compared to other fixtures but it probably won't be enough to affect usage in a lot of situations, the big issue being the power surge on startup as you suggest.

 

Thanks for your post on this by the way Guy! Really interesting, and actually this has been a great thread all round so thanks everyone!

 

Freya

Link to comment
Share on other sites

I don't think that's likely to be a particularly common circumstance for most people, to be fair.

 

 

No, it is only an issue when you have a number of lights controlled by DMX or controlled by a single breaker and only when striking them cold. I have sometimes had this problem with non-pfc HMIs or Kinos ganged together on a distro box circuit and controlled by the distro box breaker. When they are all turned on at once from the box breaker, the breaker immediately trips. When that happens, all you need to do re-set the breaker and try again. Since their capacitors received a charge the first time, they shouldn’t trip the breaker the second time.

 

A more common problem is the voltage waveform distortion that Kino fixtures that use the T-12 tubes (the older style Kino fixtures) and non-PFC HMIs will cause operating on conventional AVR generators. The ballasts of the older style Kino fixtures that use the T-12 tubes are not Power Factor Corrected (PFC) and draw harmonic currents. When used in quantity, they can constitute a source of considerable harmonic noise. For this reason, on their website Kino Flo cautions users of their older style fixtures that use T-12 tubes, that the ballasts “will draw double the current on the neutral from what is being drawn on the two hot legs. On large installations it may be necessary to double your neutral run so as not to exceed your cable capacity.”(FAQ “Why is the neutral drawing more than the hot leg”.) For a detailed explanation for why harmonic currents cause unusually high neutral returns see my article on the use of portable generators in motion picture production available on our website.

 

Put simply, when you plug a single 4’ - 4 tube T-12 Kino into a wall outlet you need not be concerned about harmonic currents. The impedance of the electrical path from the power plant is so low, the distortion of the original voltage waveform so small (1-3%), and the plant capacity so large in comparison to the load of the one light, that the inherently noisy load of the T-12 Kinos will not affect the voltage at the distribution bus.

 

 

wwaveform_kino.jpg

Left: Grid Power w/ Kino Flo Wall-o-Lite. Center: Conventional AVR Power w/ Kino Flo Wall-o-Lite. Right: Inverter Power w/ Kino Flo Wall-o-Lite.

 

 

It is, however, an all together different situation when plugging T-12 Kino fixtures into conventional portable generators. As a comparison of the oscilloscope shots above indicate, the harmonic currents drawn by conventional T-12 ballasts can generate voltage distortion in the power stream. Given the large sub-transient impedance of conventional portable generators, and the fact that the original supply voltage waveform of conventional generators is appreciably distorted (a THD of 17-19%) to begin with, you have a situation where the return of any harmonic currents by a non-PFC electronic ballast (HMI or Kino) will result in significant waveform distortion of the voltage in the distribution system.

 

Given the adverse effect of just a few conventional electronic fluorescent ballasts on a 5500W conventional generator, what is the accumulative effect of a typical lighting load made up of only non-PFC HMI & Fluorescent fixtures? To see, I ran a package consisting of two Arri 1200 HMI Par Pluses (with standard Arri non-PFC electronic ballasts) in addition to a 4’ – 10 tube Kino Flo Wall-o-Lite (with T-12 Ballasts) on a Honda EX5500 (a conventional generator). And, for the sake of comparison, I ran a comparable package but with power factor corrected electronic ballasts on our modified EU6500is (an inverter generator.) The difference between the resulting waveforms below is startling.

 

 

wwaveform_pkg_comp_AVR_In.jpg

Left: Conventional generator power w/ pkg. of non-PFC Elec. HMI Ballasts & Kino Flo Wall-o-Lite. Right: Inverter generator power w/ Pkg. of PFC Elec. Ballasts & Kino Flo Parabeam 400.

 

 

The adverse effects of the severe harmonic noise exhibited above left, can take the form of overheating and failing equipment, efficiency losses, circuit breaker trips, excessive current on the neutral return, and instability of the generator’s voltage and frequency. Harmonic noise of this magnitude can also damage HD digital cinema production equipment, create ground loops, and possibly create radio frequency (RF) interference. In other words, saving a few bucks using non-pfc lights can cost you a considerable amount of money in lost time or damaged equipment. For a detailed explanation for why this is, see my article on the use of portable generators in motion picture production available on our website.

 

- Guy Holt, Gaffer, ScreenLight & Grip, Lighting and Grip Equipment Rental and Sales in Boston

Link to comment
Share on other sites

  • Premium Member

Bit alarming that Kinos cause that problem, when common $50 fluorescent ballasts don't. I suppose the upshot of it all is that they're absolutely expecting people to use hundreds of those ballasts in a commercial building, whereas Kino-Flo apparently didn't anticipate the bit deployments people are using on these huge movies.

 

Still, it's incredibly sloppy design, when one is already paying a significant premium for a branded product. Paint me a vivid and lasting shade of unimpressed.

 

In another sense, I'd be interested to know what sort of stop they were getting out of that array of light. It's a lot of lights, but they're not that intense, and they're a fair way overhead.

 

P

Link to comment
Share on other sites

Bit alarming that Kinos cause that problem, when common $50 fluorescent ballasts don't. I suppose the upshot of it all is that they're absolutely expecting people to use hundreds of those ballasts in a commercial building, whereas Kino-Flo apparently didn't anticipate the bit deployments people are using on these huge movies.

 

Still, it's incredibly sloppy design, when one is already paying a significant premium for a branded product. Paint me a vivid and lasting shade of unimpressed.

 

In another sense, I'd be interested to know what sort of stop they were getting out of that array of light. It's a lot of lights, but they're not that intense, and they're a fair way overhead.

 

P

 

My guess would be the Kino engineers never expected Kinos (the old ones...) to be used on small generator supplies... because most productions, required big power even if at a location that required generators. And in studios or locations with power company power... the 'problem' was not an issue. (Unless shooting in an old house/building... and then the production would have brought in heave power anyway...).

 

Due to the running lighter in all senses of the modern digital age... more people are using smaller generators that exacerbate the problem.

 

This is not a problem limited to 'movie making'... my day job involves desiging and getting equipment working in the ugly places of the earth, and so I deal with this anytime someone mentions 'solar cells', 'hydrogen fuel cells', gas/diesel generators, to power equipment these problems arise with the computer based systems that are supposed to run 'flawlessly' in conditions ranging from subzero arctic moutain top conditions to the middle of the desert... (perhaps with a few changes in optional equipment...).

Edited by John E Clark
Link to comment
Share on other sites

Bit alarming that Kinos cause that problem, when common $50 fluorescent ballasts don't.

 

As can be seen in these pictures of a common electronic fluorescent ballast (top) and a Kino T-12 ballast (below), there is a big difference between the ballasts used in Kinos and those used in common Fluorescent fixtures.

 

Kino_Reg_Flo_Ballast_Board.jpg

Common Flo ballasts have only one small smoothing capacitor (the black cylinder on the upper right)

 

Given their intended use for cinematography, the Kino ballasts use much larger and more smoothing capacitors across the rectifier to remove any DC ripple that could cause flickering on camera. As such, Kino ballasts draw current in higher magnitude bursts (hence generate more harmonic currents) than do common ballasts because of the short interval in which larger smoothing capacitors must charge.

 

Kino_Ballast_Board_Sm.jpg

Kino ballasts have two larger smoothing capacitors (the black cylinders bottom right)

I suppose the upshot of it all is that they're absolutely expecting people to use hundreds of those ballasts in a commercial building, whereas Kino-Flo apparently didn't anticipate the bit deployments people are using on these huge movies.

I wouldn’t blame Kino Flo. Where hundreds of electronic Flo ballasts are to be used in commercial buildings, the Electrical Engineer designing the system will specify super neutrals to handle the higher neutral return current and if there will be a standby generator he will oversize the generator to minimize voltage waveform distortion. Rigging Gaffers have to know enough to do the same. After all there is a cost verses benefit analysis we go through whenever we design a lighting package for a scene – the extra cost of super neutrals and oversizing of generators is just another part of that analysis.

 

We went through a similar cost/benefit analysis when we were building our four 18’000 sq. ft. stages. In our case the cost of running double neutrals to each of our twenty-four 400A Disconnects (six per stage) was estimated at a quarter of a million dollars, and that did not even include upgrading our service transformers. Given the substantial cost, and the fact that we would not need double neutral capacity all the time, we explored other options. In the end, we went with conventional wet type transformers, and rather than de-rate them by 50% (the approach taken by many rigging gaffers), we picked up several Harmonic Mitigating Transformers (HMTs).

 

Kino_HMT_in_DimmerRoom_Sm.jpg

 

From front to back above are Strand Dimmer Packs, Cam Spider Boxes, a 400A 3-Phase Dual Output HMT, and finally the Dimmer Room Company Switches)

 

The zig-zag windings of HMTs create a low impedance path for triplen harmonics to return to the load, thereby off-loading the upstream neutral of triplens. So that we could use them on location with generators as well, we outfitted our HMTs with Camloks and built them into rolling cages. There is a substantial cost benefit to using HMTs with generators: not only do you save money and manpower in not having to run out double neutrals, but you also save money by renting a smaller generator that uses less fuel. For a detailed explanation for why this is, see my article on the use of portable generators in motion picture production available on our website.

 

Guy Holt, Gaffer, ScreenLight & Grip, Lighting and Grip Equipment Rental and Sales in Boston

Link to comment
Share on other sites

  • Premium Member
(the black cylinder on the upper right)

 

Are you sure that isn't an inductor? I would hazard a guess that it forms an LC filter with the two blue capacitors at the upper right, possibly to get the thing through interference checks. The black cylinders with metal tops on the top left are certainly capacitors and appear to be nearer the mains end of the board. I'd assume that the right hand end would be the output stages.

 

But that is a very small input reservoir cap, certainly.

 

Do my eyes deceive me, or is that a 555 timer IC in the middle there? Something that's probably a current regulating buck converter, based around the trusty 555? That's demented!

Link to comment
Share on other sites

My guess would be the Kino engineers never expected Kinos (the old ones...) to be used on small generator supplies... because most productions, required big power even if at a location that required generators. And in studios or locations with power company power... the 'problem' was not an issue. (Unless shooting in an old house/building... and then the production would have brought in heave power anyway...).

 

The harmonic currents generated by lights that use non-pfc Switch Mode Power Supplies (HMIs, Kinos, CLF lamp banks, LEDs) and lights that use power supplies consisting of SCR/capacitor circuits (SoftSuns) are also an issue for productions using larger generators and even grid power. The difference is that the means by which the industry has more or less successfully dealt with harmonics - namely the over-sizing of generators, the over-sizing of neutrals, the incorporation of power factor correction circuitry in large HMI ballasts, and finally the use of generators with 2/3 pitch windings – are generally not available to users of small portable generators as their primary source of power. It is generally not an option for small independent productions using portable gas generators by necessity to upscale to larger generators; and, given that there is not much that the end user can do to alter the power output panel of a portable gas generator, it is not an option to customize their distribution package for the requirements of higher neutral currents resulting from non-linear loads. All that users of small conventional AVR generators can do to remediate the adverse effects of harmonic currents is downsize their lighting package when it consists predominantly of non-linear light sources or use an inverter generator. For a detailed explanation for why this is, see my article on the use of portable generators in motion picture production available on our website.

 

Guy Holt, Gaffer, ScreenLight & Grip, Lighting and Grip Equipment Rental and Sales in Boston

Edited by Guy Holt
Link to comment
Share on other sites

...most productions [that] required big power... that required generators ... the 'problem' was not an issue. Due to the running lighter in all senses of the modern digital age... more people are using smaller generators that exacerbate the problem.

 

Here is an example from an earlier post (available at http://www.cinematography.com/index.php?showtopic=63910) that demonstrates that productions using bigger diesel tow plants are not immune from the effects of harmonic currents generated by non-power factor corrected Kino, HMI, and LED loads. The lighting set up consisted of the following: on each end of a football field the crew had rigged a 4k HMI Par and M18 into scissor lifts. Two additional M18s were rigged into the announcer’s booth at center field.

Underdog_stadium_Sm.jpg

 

From a 500 Amp diesel generator that was dropped behind the bleachers at the 50 yard line, the electricians ran out 100ft of 2/0 to a main distro box at the 50 yard line. From there they ran 150' of banded #2 AWG cable in each direction down the field towards the end zones where the lifts were. They terminated the banded #2 runs into a 100A/120V snake bite, and from there they ran 100ft of #4 AWG to a 100A lunch box, and from the lunch box they stepped down to #6 AWG cable to power the 4k Par and a stinger to power the M18. This set-up was mirrored on either side of the field. The two additional M18s rigged into the announcers booth at center field were powered by a 50' run of #4 AWG to a 100A Lunch box underneath the booth. Load balance was almost dead even between the legs of the generator operating single phase.

Underdogs_1_Sm.jpg

 

Shortly into shooting, one of the 4k Pars went down. In addition, one of the lunch boxes started buzzing loudly. The crew replaced the bad ballast (a non-pfc Power Gems) with another and it too went down. Since they had oversized their cable runs by using #2 AWG rated for a 160A for a 70A Load (52A (4kW) + 18A (1.8kW) = 70A), they did not suspect the problem had to do with voltage drop, besides the M18s ran flawlessly. When the best boy electric checked the generator, the generator meters read 108v. The best boy was a bit flabbergasted why the voltage had dropped on the generator when they had oversized the cable to banded #2 instead of #6 and no one had touched the generator. The answer to his questions has to do with the harmonic currents drawn by the non-pfc 4kW ballasts.

 

When powering lights that use non-pfc Switch Mode Power Supplies (HMIs, Kinos, CLF lamp banks, LEDs) you must take into account the harmonic currents these non-linear loads draw that can have a severe adverse effect on the power waveform of even large diesel generators.

2.5_gen450_CrestFactor_Sm.jpg

As is evident in the power quality meter reading of a non-pfc 2.5kW HMI above, the high peaked pulsed current (lower waveform) drawn by its’ smoothing capacitors is a distorted waveform that does not resemble the sinusoid of AC voltage or the current drawn by an incandescent light. As such, the current drawn includes a number of harmonic currents in addition to the 60hz fundamental (see Fast Fourier Transformation of a non-pfc 2.5kW HMI ballast below), which not only increases voltage drop but also causes voltage “flat topping.”

2.5_gen450_CurHar_sm.jpg

Voltage flat topping, a particular form of voltage drop, is caused by harmonic currents interacting with the high impedance and soft power of a diesel generator. Since smoothing capacitors consume power only at the peak of the voltage waveform, voltage drop due to system impedance occurs only at the peak of the voltage waveform – causing the “Flat Topping” we see in the top waveform of the first power quality meter above that is characteristic of capacitive loads on generators. If the voltage waveform distortion is severe, it can cause voltage regulator sensing problems and inaccurate instrument readings in a generator’s control systems as well as ballast failure, which would explain the problems the crew experienced above.

 

Let’s look at the regular voltage drop component of this problem first. As you can see in the Fast Fourier Transformation of the 2.5kW HMI ballast above, non-pfc HMI ballasts draw a distorted current waveform that is rich in harmonics. The higher harmonic frequencies create what is known as "skin effect" in the cable. Skin effect is a phenomenon where the higher frequencies cause the electrons to flow toward the outer sides of a conductor. Since the flow of the electrons is no longer evenly distributed across the cross sectional diameter of the conductor, more electrons are flowing through less copper and the resistance of the conductor increases. Even though the cable was oversized in this case, this increase in resistance reduces the ability of the conductor to carry current, resulting in greater voltage drop over shorter distances.

 

skin_effect.jpg

The area of the cross sectional diameter of a conductor used by DC current (left), Low Frequency AC Current (center), High Frequency AC Currents (right).

But, voltage drop alone was not the cause of the problems the crew experienced. The other contributing factor was the voltage “Flat Topping” caused by the high impedance of their distribution system. Always remember, there are two components to the impedance of a distro: cable and generator. The skin effect caused by the harmonic currents generated by the non-pfc ballast increased the impedance of the cable. The second contributing factor was that they were operating single phase on a small generator. A 500A generator operating three-phase provides about 166A/leg. The same generator operating single phase does not provide 250A/leg, but rather the same 166A/leg but just single phase. The set up as described had about 100A on each leg. So, the generator was pretty well loaded and half that load (52A) was generating harmonic currents.

 

Under these circumstances one could expect much more severe voltage flat topping than that in the example of a 2.5kW HMI above. What causes flat-topped voltage? According to Ohm’s Law current reacts with impedance to cause voltage drop. For example, when encountering the high impedance of a loaded generator, a 3rd harmonic current will produce a voltage drop at a 3rd harmonic voltage. Likewise a 5th harmonic current will produce a voltage drop at a 5th harmonic voltage, etc. Each harmonic current drawn by the non-power factor corrected 4k ballast flows through the system impedance, resulting in a voltage drop at that harmonic frequency. In other words, where a distorted current waveform is made up of the fundamental plus one or more harmonic currents, each of these currents flowing into an impedance will according to Ohm’s Law, result in a voltage drop resulting in voltage harmonics appearing at the load bus and distortion of the voltage waveform.

2.5kW_grid_CrestFacto_Sm.jpg

 

This pattern does not appear in the voltage waveform of the grid power above because of its’ much lower impedance. The impedance of a generator is not an easily known quantity. Depending on its’ size and design, the impedance of a generator will be 5 to 100 times that of a utility transformer and it will change as the load changes. But where they were using a fairly small generator for the harmonic load generated by the two non-pfc 4k ballasts, the internal reactance of the generator would have been sufficient to cause appreciable voltage flat-topping as well as a voltage drop.

 

Now for the reason the voltage output of the generator dropped even though no one touched its’ voltage regulator. The voltage output of the generator did not drop rather the flat-topped voltage caused an erroneous reading of the voltage by the meters. The flat-topped voltage manifests itself as low voltage on the generator’s meter because conventional electrical meters like those on most generators are designed to read only sinusoidal waveforms accurately – not distorted waveforms. Flat-topped voltages introduce errors into the measurement circuits of these meters, which result in low readings. Since the consequences of under measurement can be significant - overloaded cables may go undetected, bus-bars and cables may overheat, fuses and circuit breakers will trip unexpectedly - it is important to understand why meters based on "true rms" techniques should be used on power distribution systems supplying harmonic generating loads.

 

Most analogue meters and a large number of digital multi-meters are designed to read voltage and current quantities based on a technique known as “average reading, rms calibrated”. This technique entails taking a measurement of the average (or mean) value (0.636 × peak) and multiplying the result by the form factor (1.11 for a sine wave). The result is 0.7071 times the peak value, which is displayed as “rms.” This assumption is valid only for pure sinusoidal waveforms like the one pictured below.

Meters_Haromonics_2.jpg

 

To accurately measure waveforms distorted by harmonics, a meter that will measure the true rms value is required. For example, if you were to use a conventional “average reading, calibrated rms” meter to measure the waveform below distorted by a non-linear power supply (with a peak value of 2.6 A and an average of 0.55 A), its' display would give a “rms” current of 0.61 A.

Meters_Haromonics_3.jpg

 

A meter that measures true rms will give a more accurate measurement of 1.0 A for the distorted waveform above. By comparison, the reading of a conventional “average reading, calibrated rms” meter is almost 40% lower than the real value.

 

Now for the reason the older non-power factor corrected 4k ballasts were failing when the newer power factor corrected M18 ballasts were not. One adverse effect of flat-topped current is that it causes non-power factor corrected equipment to draw more current to maintain the power rating (watts) of the unit. This, in turn, can cause protective fuses on electrical boards of the equipment to blow. I experienced this first hand, when I first tried some years ago to operate a 4k HMI Par on a Honda ES6500 (a conventional AVR generator) with the first generation of electronic square wave ballasts - a Lightmaker. The ballast inexplicably failed when it had never given us problems on mains power. Upon closer inspection back in our shop, we found that a protective fuse on the main board had blown. We replaced the fuse and continued to operate the ballast off of grid power without incident. But as soon as we tried to run it again on the Honda the fuse blew. Since the Lightmaker ballasts are not Power Factor Corrected (PFC), the cause of the ballast's erratic behavior was the amount of harmonic distortion it was creating in the power stream generated by the ES6500. The harmonic currents were not a problem on grid power because, given the extremely low impedance of grid distribution, they did not induce voltage distortion. But, fed back into the power stream generated by our Honda ES6500, the same harmonic currents created voltage distortion and sufficient voltage drop from skin effect to cause sufficient increase in current to blow the protective fuses on the ballast boards.

 

Ballast performance has improved remarkably since that first generation of electronic Power Gems & Lightmaker ballasts. The latest generation of power factor corrected electronic ballasts include circuitry that virtually eliminates the harmonic currents the ballast draws. However, since power factor correction is still not commonly found in HMI ballasts smaller than 6kW, voltage waveform distortion can be a problem when operating HMI lights even on large diesel generators. For more detailed information on the source of harmonics and how to counteract their adverse effects use this link for a white paper I wrote on the use of portable generators in motion picture production.

BoxBookLinkGenSetFP.jpg

 

This article is cited in the 4th Edition of Harry Box's "Set Lighting Technician's Handbook" (http://www.screenlightandgrip.com/html/BoxBook.html) and featured on the companion website “Box Book Extras.” (http://www.screenlightandgrip.com/html/BoxBookExtras.html) Of the article Harry Box exclaims:

 

"Great work!... this is the kind of thing I think very few technician's ever get to see, and as a result many people have absolutely no idea why things stop working.”

"Following the prescriptions contained in this article enables the operation of bigger lights, or more smaller lights, on portable generators than has ever been possible before."

 

The article is available online at http://www.screenlightandgrip.com/html/emailnewsletter_generators.html.

 

Guy Holt, Gaffer, ScreenLight & Grip, Lighting & Rental and Sales in Boston.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...