Air Accident Investigation Branch: film lights caused window damage on A321neo
flightradar24.com
flightradar24.com
So we left it on. After one test we left it pointed at our test article, which was made of styrofoam. After a couple of minutes it started smoking and melting. If we hadn’t smelled the smoke I'm certain it would have caught on fire.
We made very sure we never left it pointed at a fire sprinkler after that.
The one time we had an inadvertent discharge of a fire sprinkler in the historic building we're in was when a film crew had a light positioned too close to one. It was not close enough that someone without specific experience looking at it would have thought it was too close. The light wasn't even a high-powered outdoor film spotlight; it was for an indoor shoot. Film lights can just be very bright, and very hot.
Screw Verilux with their overpriced "10,000 lux" 10W panel (only at less than 6 inches from your face), instead get a high-CRI studio light over my reading chair and I'll be happy all winter.
I previously used some adhesive LED strip lights on a 24x24 plate of aluminum. If DIY electronics are your aesthetic, go for it!
See also these posts:
https://meaningness.com/sad-light-led-lux
https://meaningness.com/sad-light-lumens
for a much more detailed writeup (no affiliation).
I wouldn't suggest going brighter though, 30klm is already bright enough I can't look at the light directly and need to bounce it off the walls for comfort. The cat doesn't seem to mind though: https://i.imgur.com/aTegmqR.mp4
I found that 200-300W is about the optimum from a dollars per high quality lumen perspective. The SmallRig COB lights are great if you can get them at a discount.
I got two of these, $350, but they’re fantastic and I use them in place of my crappy ceiling lights:
https://www.bhphotovideo.com/c/product/1753990-REG/amaran_ap...
One can also buy a couple of bright, high-CRI LED bulbs for one's existing lights and save themselves about $650, because the difference between those and the studio lights is typically red spectrum coverage, which is not relevant for circadian rhythm / SAD treatment. The film/TV LED lights are also designed to have a really smooth spectrum, minimizing gaps. If you want to do that...fine, but please don't advise people to waste lots of money on something completely unnecessary.
I assure you, a double-height high-CRI LED strip running along the edges of the room will throw a fuckton of light without the issues caused by having two very intense point sources, namely shadows and glare.
There's almost never a problem with too much blue light in the high CRI bulbs (in fact, the people doing the spectrograms are usually looking for bulbs without the high blue spectrum spikes, and many people are looking for bulbs without the blue spikes (and they're hard to find!)
Smart bulbs sacrifice on light quality for color and dumb bulbs can't adjust color temperature (even dumbly!). Studio lighting is expensive but has both.
I love the idea of a high-CRI LED strip, because, yeah, an intense point source is really not ideal to light a room. But the strip would need to be bi-color, but such a thing probably exists. I just moved and I have zero ceiling sockets, so A19 is dead to me :(
What does that mean?
You can clip one vertical on the vertical edge and one on the horizontal edge of the monitor. Just don't party on your desk or move it too much.
Ouch, hopefully the film production company had liability insurance! It’s easy to rack up six figures of damage by setting off fire sprinkler.
Usually it's just one in the fire area, they trigger with a heat sensitive capsule that breaks.
Anyway if I remember my highrise fire safety training right there is at least one building where on sprinkler going off sets the others off. IIRC the pressure drop from one unit activating triggered a pump and the subsequent increase in pressure popped the other.
Of course it would be quite annoying to have your apartment flooded because one of the neighbours let their pizza burn :P
A burning pizza just kicks off the fire alarm, not the sprinklers. Frankly I don't remember much past the the whole drop in pressure detection leading to kickoff of other sprinklers thing to suggest what level of partitioning occurred .
A jockey pump maintains pressure in the system and is sized at less than the flow rate of one sprinkler head. The main fire pump is activated when a pressure transmitter detects a drop in pressure. The jockey pump can’t maintain pressure when a sprinkler head opens, and the drop in pressure starts the main fire pump.
I don't know off the top of my head the amp rating, but I know we had to have the room electric supply upgraded (and this was already a well equipped robotics lab - it isn't like the electric supply was bad to begin with).
The light itself was about three feet in diameter, four feet long, and, like, 200 pounds. It was a big old hunk of photon-generating madness.
>I’m really surprised lights used to simulate sunrise placed 20-30ft away could completely melt the foam holding the exterior windows in place.
We loaned it to some NASA colleagues once, but their safety manager refused to allow it in the building. At all.
About 3am the fire alarm went off. One of the elderly nuns was sleeping in her room on the top level of the building, was carried out by a firefighter.
The school was a converted English Tutor Mansion, had a grand entrance with hand carved mahogany everywhere. The film company hired a crew of like 30 cleaners that were scrubbing the smoke out of the carvings with toothbrushes.
Although it's fairly common for schools to have tutors as well
I was quite impressed the with the evacuation procedures - almost 3000 people outside in under 3 minutes!
styrofoam melts super easily so not sure that says much
It was a brand new arc lamp.
Some other facts:
* While handling lamps, you are not allowed to be in the lamp room without hearing protection, just in case one implodes. * A single lamp emits enough light that it can cause permanent blindness. * Lamps are water-cooled.
Here's a picture of the back of the unit: https://www.esa.int/ESA_Multimedia/Images/2014/04/Sun_simula...
(it still lists the old power rating, but it was increased to 35 kW a few years ago, when BepiColombo was tested. The lamps were also refocused at that time, to cover a smaller diameter circle. The sun emits a ridiculous amount of light...)
Here's a picture of a spacecraft being lit up by the lamps (at a really low light level): https://www.esa.int/ESA_Multimedia/Images/2008/03/GOCE_ready...
I wrote the software that monitors lamp output during use, and also the software used to calibrate the alignment of the lamps (they need to be properly centered).
That article isn't as good (less technical/more 'pop' audience) but they both link the actual AAIB report which I found quite interesting and surprisingly readable, and I don't think either article has anything extra to offer. I'd suggest skipping straight to it: https://assets.publishing.service.gov.uk/media/6544b3089e05f...
Some directors prefer the spectrum completeness and profile from tungsten (or the ultimate, carbon-arc, which is virtually indistinguishable from the sun.)
I also think there are levels of light that aren't really feasible except with carbon arc because LEDs don't like heat and that limits power density.
I don't remember what movie it was, but there's a photo of an enormous balloon light - larger than an entire house - over a farmhouse somewhere in the midwest, at night. Pretty sure it wasn't LEDs as the source, but I could be wrong.
Not a balloon, but a big diffuser with a whole bunch of Cineo 410 LED sources.
The situation can also be flipped and apply to light sources, and those by definition are key in photography.
In case of LED light, its colour balance can be declared to match some reference Kelvin number, but because it is “fake”, a mix of spikes in the spectrum (roughly at R, G, and B for RGB LEDs, phosphorus-covered white LEDs have the spectrum more even but its own gaps and bumps), and materials of various colours can reflect inbetween those peaks, or right at those peaks, those materials can look 1) different from scene to scene and from light to light or 2) plain wrong in post production, compounding variance between camera sensors or films (which create colour from their own mix of R, G, and B), lenses, etc.
Added to other flaws of LEDs, such as longevity (of cheaper units), issues with brightness and colour reproduction consistency, PWM, etc., they make a poor choice for a variety of situations[0], but in photography particularly so, particularly where colour reproduction and continuity matter (TV and film).
By contrast, black body radiation—hot and more energy intensive—is a solid spectrum of even light, without spectral discontinuities or flicker at any brightness.
[0] In some situations those flaws are considered acceptable. You may have noticed how two identical OLED iPhones displayed at an Apple Store, even fully reset to defaults, can have obviously different white point when you look at them side by side—that’s colour reproduction/emission variance and/or degradation over time. Similarly, you can often spot PWM flicker if you reduce brightness and squint at an OLED phone with your peripheral vision. These things don’t matter much, since 99.99% of the time we look only at our own device and our colour perception and flicker tolerance is adjusted to it. Not so with photography; you can’t afford colour variance between two different lights even in cases where it’s not noticeable to the naked eye in the moment, whereas PWM restricts your FPS and shutter angle options.
Take a CD / DVD (if you can find one in 2023) and look at the spectrum from your phone flash, or any other LED lamp, and observe the broad spectrum. They all have some sort of blue peak, but it's very small in "warm white" lights.
Other issues, such as PWM flicker coinciding with your FPS and shutter angle or LED controller interference in audio recording, remain.
From my understanding, today mature LED-based solutions can only mitigate these issues with varying success, not eliminate all of them entirely, all this mitigation adding complexity and cost, while on the other hand one could just use a black body emitter so that those issues are not technically a thing, and try not to set things on fire.
That said, if you need a lot of light, halogen is without alternative. There's a reason the ArriMAX with its 18kW halogen lamp is still popular.
I believe they still use PWM to dim, and you still need to work around the flicker by choosing the right frequency depending on your FPS and shutter angle. They help by making PWM frequency configurable but that is mitigating the problem not eliminating it.
Second, these weren't ordinary stage lights. These were lights for use outdoors to simulate natural sunlight, which is much brighter (and hotter) than what you'd normally use indoors.
Third, they were probably positioned too close, and left on for too long. Just a guess on this one.
Maybe this kind of foam isn't entirely up to the spec either.
It’s an surmountable problem, but can be expensive if it doesn’t need to be overcome often.
(All speculation on my part)
I've become much less of a fan of the "switchilinear" approach. The control loop of a linear regulator isn't fast enough to respond to fast switching frequencies; the regulator is effectively just a resistor in a filter network at that point.
Even if your switching frequency is in the regulator's loop bandwidth, the loop's gain at that frequency is almost certainly very low.
Assuming you're using a global shutter camera: even if the ripple is completely unfiltered, the only remaining intensity effect would be that some frames get 8000 "blinks" in a frame and some get 8001.
Already, stage and film lighting is a relatively small market compared to the entire lighting market, but it's not so small that NRE can't be spread across a lot of units.
I believe the time constant for most small lamps is 30ms and longer for bigger lamps. 1 / (2 * pi * 30ms) is 5Hz, so 120 Hz will be about 30 dB down. This is probably just barely big enough to matter.
I wonder what the filament looks like.
E.g. I have both fluorescent and HID lamps with this characteristic; they take a couple minutes to reach full brightness, but still will dim when a big load kicks on and turn off immediately when I flip the switch.
Is the limiting factor the technology to make a 4500W LED bulb in the same form-factor/focus as the 18kW halogen, or does it just become prohibitively expensive to do so?
https://en.wikipedia.org/wiki/Color_rendering_index#Film_and...
But otherwise yes they would be less hot.
Light is energy, and filming requires a lot of light, so there's always going to be significant visible energy available to heat things up. (Unless you're filming a mirror, but in that case why are you illuminating it?)
Incandescent lights emit a lot of invisible energy in the form of infrared radiation, while LEDs are designed to emit most of their energy in the visible spectrum. This means LEDs need less energy to produce the same visual effect, and therefore impart less heat. Another option is to slap a "hot mirror" or "cold mirror" (like the one from the dentist's office) in front of your lights that separates the visible from IR.
Cold mirrors that large are expensive and it's just complexity/something to break or need servicing (cleaning) so I doubt they were stocked much by rental companies or saw much usage in tv/movie production.
Spotlights like these though, you might as well be holding a giant magnifying glass. The beam is culminated and will cook whatever it’s pointing at. Will it cook a human? Yes if you stand there, still, for long enough. You will definitely get sun burned. If you get within 10 feet it will feel like your standing in an oven. Within 5 feet and your skin could boil.
One thing I learned while webcasting Yves St. Laurent's fashion show in the mid 90's is that the flowers wilt within minutes of being placed due to all of the heat from the lights. So the way they dealt with that is that backstage there is an absolutely enormous amount of flowers ready to be deployed and they just cycle through flower arrangements continuously for the duration of the show so that the flowers are always fresh. It's tens of people walking to and fro without pause.
https://en.wikipedia.org/wiki/The_Thin_Man_(film)#Filming
(Great movie BTW!)
As to the number of takes and food: one particular set I recall was for a cookie brand, they did endless re-takes and the actress that had to eat the cookies started to get sick enough to vomit. We were all pretty happy when the director was finally satisfied with what must have been his masterpiece.
The reasoning was that:
- college students are busy/forgetful so will leave the light on
- there are times of day where no one would be in a dorm room
- BUT that dorm room is a bigger dorm with potentially lots of other students
- an unattended halogen lamp could set fire to an object on a desk etc
this post feels like a much bigger version of the above
Quite scary what kinds of causes trigger such incidents. As a frequent flier, I would have never imagined something like this.
How are there no regulations that should have prevented or caught this? I bet there will be now.
Commercial aircraft are required to be designed to maintain 15,000 foot equivalent cabin altitude after any probable failure, which includes multiple missing windows, per the requirements of 14 CFR 25.841 and it's EASA equivalent.
There is already regulation in place that should have caught this but did not. The captain is required to perform an external visual inspection of the aircraft before departure, which follows a checklist that includes looking at the windows. It appears that did not happen.
We've no idea if the damage was visible prior to takeoff. It may have easily passed a visual inspection on the ground before moving at speed with a pressure differential. As per the article at least one of them appears to have detached in flight and struck the horizontal stabilizer.
Also the engineer who signed off the aircraft will do a walk around.
The location of these windows might have made it hard to spot the damage.
I would think a pressure check would be trivial. I would also assume a visual inspection is part of takeoff (camera or human). I assume this sort of thing is rare enough to not warrant the extra time/cost of these trivialities.
* Damage to the left leading edge of the horizontal stabilizer was noted as well. Investigators found acrylic in the puncture, consistent with a window pane.
And:
* The aircraft remained properly pressurized throughout the flight.
Are not mutually incompatible? (i.e. a window pane hit the tail in flight and somehow the aircraft remained pressurized?)
pressurization of aircraft is not required till 12,500 ft (3810m) so the air pressure differential at 15000 is not likely to be that great.
Well, no. I don't think cabin altitude for passenger airliners has ever been that high, so you need pressurization much lower than that (typically starting around 6000 ft for modern aircraft).
Edit: Even the Boeing 307 Stratoliner was pressurized to 8000 ft.
And, I mean, think about it - there are _towns_ at 9k+ feet.
Without reading the report, the outside pane could have failed and hit the stabilizer while the middle pane continued to hold pressure as expected, therefore no contradiction in the two statements.
In this incident, the outer frame had melted and the outer pane had separated from the rest but the central layer was OK at that point in time.
https://thepointsguy.com/news/what-are-airplane-windows-made...
This is why the fear that a bullet fired in the airplane will pop it like a balloon is unfounded. Poking a hole in the fuselage won't do anything. A bullet would have to hit and disable something critical (and there's always a backup for critical parts).
Having so many of these huge lights, blasting out 12KWs each, for 4-5.5 hrs just to get a few photos of a plane sounds a little bit too much, but hey who am I to judge :-)
https://www.caa.co.uk/commercial-industry/aircraft/operation...
A321 Neo LR is approved for that.
That generalizes to 'any plane with more than one engine can fly with one engine out'.
> The over-water part is that it can fly on one engine for a long distance to get back to an airport - which tends to be further away if you're flying over the ocean.
That's where certification comes in: not all twin engine planes are ETOPS certified.
> Basically having a bit more margin in the design for extended endurance with an engine out.
Yes, and that translates into much more work than just 'a bit more margin' under the hood, to get that margin you also will need to take this into account during the design phase of the aircraft, crew training, maintenance schedule etc. Incidents under ETOPS conditions are rated more severely than those in other situations.
Something to think about next time you're over the pacific.
For the engine to melt a window, something has gone wrong enough that this tolerance isn't material. (You'd need a lot of heat. Plus enough turbulence to blow it laterally inward, but not so much that it's allowed to cool. That combination suggests a loss of power and a low-speed, i.e. low-altitude, stall.)
From the article: "They located the source of the noise as a dislodged window pane aft of the over wing exit."
The engines are under the wing; the affected seals were over it.
If you're melting the window panes in this scenario from the engines, you're having a really, really bad day. Plus, the noise increase from a missing window pant would likely be the smallest of the warning signs.
[1] The complete list of options are typically starvation, suppression, evacuation. Apply in that order and do so quickly.
"Starvation, suppression, and evacuation" needs neither "emergency landing" nor "don a parachute" to be added to it to be understandable. Both are just prerequisite steps to the "evacuation" mitigation.
That prerequisite step may never be completed, due to the aircraft being destroyed before it can be evacuated.
So no, if there is a fire, those 3 steps are your options regardless of the aircraft's location or altitude.
Interesting reversal: in my younger years I flew all over the planet, I can't even begin to estimate on how many flights I've been. And then, a couple of really bad flights in succession and I actually find it very difficult to contemplate flying again even if I know that statistically it's pretty safe and I'm exaggerating. Rationality only goes so far, apparently if you scare the lizard good a couple of times he remembers.
There is that one case that made the news worldwide for months because of how hard it is... and I never heard about any other one.
The part about smaller planes makes a lot of sense.
Also it gives me something to do other then sitting in an airport terminal because the plane is delayed due to maintenance issues.
Except for the toaster, the bloody toaster is always acting up.
What a silly thing to say. First of all not all water is ocean. Second the speeds and forces involved in a landing and in a crashing are very different.
Here is how an aftermath of a water landing looks like: https://youtu.be/x02cA6eamq0?si=-dQsJm353WySpv0Z
As you can see the airplane is designed to float.
I appreciate this, er, implicit understatement.
heat from an engine is directed straight out the back by nature of the turbines.
modern engines are also what are called "high bypass ratio" engines, where the outer ring of the engine (closest to the cladding) is really just air flowing by. the combustion area is smaller, in the center.
The engines are under the wing, the windows are above the wing, they are not very close to the windows at all.
The black-body radiation is the true heat coming from them
But good there is awareness about using equipment with old, faulty hardware and the need for measuring uv levels as part of technical checks.
To clarify, ARRI can only speak on behalf of what they legally aim to sell. Not what the factory reality lets out the door, or what any production does with their equipment. They aren't evil, just a company.
Depending on the amount of variation that trouble may arrive sooner rather than later.
All lamps that have a fixture which would be capable of emitting UV lights (mostly HMI lamps) must have UV filters in front of them. Typically they are part of the fresnel lense. For LED this is a non-issue.
If you got a sunburn in a studio, that means they either used shit lamps (non-industry standard), or the filters have been removed. Either way this is a work safety violation and any study worth their salt wouldn't stand for it.
The lamps are of course still something that you need to treat carefully. First because it is a ton of power and heat, but also because they are heavy. But if you use them in the way intended by the manufacturer they are not more dangerous than anything else.
A typical Arri HMI will even have a diagram with safe distances for heat radiation engraved somewhere on the body of the lamp. So you don't even need a manual to know you are doing it wrong.
I once had a HMI with a broken lense through which UV radiation would leak on a outdoor set. We noticed this because a fist-sized pile of insects started to accumulate right at the spot where the leak was. The smoke from that was quickly noticed.
You'd be surprised what shortcuts budgets end up creating...
Source: A decade+ on film/photography sets as DIT.
p.s. Work on macro phantom stuff. You'll see.
Is UV a standard thing to avoid at every case, or just in exceptional cases?
If you're too close you may get burned from the heat, but actors spend hours under them and don't get tanned.
But yeah, 72000 watts is alot of heat to pump at an object for 9 hours