This site is not affiliated with the Discovery Channel or MythBusters. Please visit the Official MythBusters site for official content.

Episode 69: 22,000 foot fall, Lights On or Off

  • An explosion can save you from a 22,000 ft fall: busted
  • It's better to leave light bulbs on: busted

The 22,000 Foot Fall myth was in many was a revisit of the Cement Mix Up myth, i.e. setting off a really big explosion in a Calaveras County quarry for kicks, though they were actually testing a myth this time around. This explosion had better footage than the cement truck as you could see the shockwave more clearly, and there is even a fire-fighting copter swooping in afterwards to put out fire on the surrounding hillside. As this myth did have a true story at its core, there was some hint of belief from Adam that this might actually work... until they were reminded what a 1,000 lb bomb looks like.

I was interested in the lights on or off myth as I had been told that its generally best if you leave fluorescent lights on all the time. I believe I had been told something like, "unless you leave the fluorescent lights off for more than an hour, you shouldn't turn them off." As it turns out, the actual rule for fluorescent lights is about 23 seconds. For other lights, about a second or less.

22,000 Foot Fall

"Biggest Explosion in MythBusters History"

Myth: a World War II gunner feel 22,000 feet from a bomber gun turret onto a train station. At the moment he was falling onto the train station, a 1,000 bomb exploded in the station, cushioning his fall and allowing him to survive. (source: factoid listed on a calendar)

See also:


Adam and Jamie spoke to Joe Pruzzo of Castle Air Museum for background on WWII bombers. Pruzzo talked about one anecdote he had heard where a gunner fell 20,000 ft into a snowbank and survived with a broken leg. He also showed them a gun turret on a WWII bomber, which was so tiny that the gunners couldn't wear parachutes.

The MythBusters researchers were able to confirm that an airman did survive a 22,000 ft fall, but the question was whether or not the explosion was an element of the story.

15 lb. ball and air bag explosion test

Tory and Kari were tasked with creating a small-scale replica of this myth to test its feasibility. They dropped a 15 lb. ball onto a model glass-roofed house with a car airbag charge inside (no airbag, just the pyrotechnic charge). The shockwatches registered between 75g-100g for the ball falling by itself. After some trial and error, they got the charge to go off just as the ball was about to hit the roof. Side-by-side views of the ball falling with and without the explosion showed no real difference between the two; i.e. the explosion did not cushion the fall.

Experiment setup/design

  • Train station: Adam designed a 10'x20' mock train station out of welded steel and 2'x2' panes of glass. Jamie: "It's pretty, but it's going to go away pretty soon"
  • 1,000 lb bomb: a 1,000 lb GP bomb contains about 500 lbs of high explosives, with the rest of the weight being the bomb casing. The decided to give the steel tubing from the Sharammer one final hurrah (see XMB fuselage for Exploding Hair Cream, barrel for the Archimedes Steam Cannon, and Jaws Special). They used a bunch of Dyno Trojan Spartan explosives to generate the 500 lbs of explosive force.
  • Airman: they ruled out using Buster for the drop and instead used Ted (as in Busted), which was a dummy made out of ballistics gel.
  • 22,000 drop: The MythBusters knew human terminal velocity is 120 mph, so they decided to use balloons to lift their airman to an appropriate height for terminal velocity: 487 ft (5.5 seconds). They also planned on using a guide wire to direct airman's fall directly into the mock train station. (NOTE: In Escape Slide Parachute, they calculated with a 14 second fall and 2000 ft drop height)

The explosion

They setup at a quarry in Calaveras County. They were beset with difficulties at the blast site, from balloons popping to weather so hot that it was melting Ted and causing the fire marshals to cancel the explosion for the day.

The explosion itself was beautiful. The high-speed footage showed the shockwave radiating outward with a mushroom cloud forming upwards. The surrounding ground was set on fire and a fire-fighting copter swooped in to put it out. Steel shrapnel from the train station was also sent everywhere -- a piece of steel shrapnel even wrapped itself around a small tree, breaking the trunk. The shrapnel around the tree was evidence to Jamie that the myth was completely implausible: the airman would have been riddled by shrapnel as well.

As for Ted: that part of the experiment was a complete failure. Ted's guide wire failed and he fell far from the site, but still among the steel shrapnel.

busted: although they believe the story about an airman surviving a 22,000 foot fall to be true, the amount of explosive damage they witnessed led them to believe that there was no way that a explosive shockwave could cushion a fall.

Lights On or Off

Myth: You save on energy bills by leaving lights on. Some people believe that the energy to turn on lights exceeds savings of turning lights off.

They talked to Mark Reisfelt, manager of the Independent Electric Supply where they purchased their light bulbs. He felt that it was best to turn the lights off.

To test the myth, they needed to measure energy usage during startup, maintenance (steady state), and shutdown.

For steady state energy consumption, they turned on several different types of bulbs for 60 minutes and measured their consumption using a Kill A Watt: * Incandescent 90 Wh * Compact Fluorescent (CFL): 10 Wh * Halogen: 70 Wh * Metal halide 60 Wh * LED: 1 Wh * Fluorescent: 10 Wh

For startup energy consumption, Grant hooked up an inductive current loop to a computer and measured the amount of energy used when the turned on the bulbs. With an inductive current loop, you run a wire through the center, which induces a current in the loop. This current is then measured by a digital sampling oscilloscope.

Based on the amount of energy consumed turning on the bulb, they were able calculated how long the bulb would have to be turned off in order to make it worth the energy savings, i.e. "It's best to turn off the bulb if you are leaving the room for":

  • Incandescent: 0.36 seconds
  • CFL: 0.015 seconds
  • Halogen: .51 seconds
  • LED: 1.28 seconds
  • Fluorescent: 23.3 seconds

In other words, its almost always best to turn the bulb off. Even the 23 seconds for the fluorescent lights isn't very long, and the rest of the times are pretty much blinks of an eye.

Bulb Longevity

They tested one final element of this myth: frequently turning lights on and off decreases their life span, thus leading to greater costs. Grant setup a timer and relay to turn the bulbs on and off repeatedly every 2 minutes. After six weeks, only the LED bulb was still working. Based on this test, they extrapolated that it would take five years of ordinary usage to cause the bulbs to burn out.

* busted *

Side-note: 105-year bulb

Grant and Kari visited the Livermore/Pleasanton Fire Department to view their light bulb that has been burning for 105 years. It has a carbon filament that is much thicker than modern bulbs and also burns much cooler/darker. You can check on the light using the bulb's webcam.


For the cfl and flourescent, you really should compare equivalent bulbs. 90 watt incandescent compared to 10 watt cfls? How about comparing a 25 watt cfl to a 100 watt incandescent since cfls use about 1/4 the power to provide the same light.

There's an error in this article. W/hour is meaningless; the correct unit is Wh (which is the same as W*hour).

@Alex: thanks for the correction. I've updated the summary.

The probelem with the longevity experiments is that they didn't give us any numbers. All they said was that after 6 weeks all except the LED one were dead. So, does that mean one bulb only lasted 1 day of that treatment, or they all died after 6 weeks? Give us the data! How many cycles did each ones survive?

just FYI.. The current sensor is known as a hall-effect sensor. Also, I would think that the CCFL would be more economical to leave on for longer than the incandescent.

A more interesting test would be to measure the power factor of the different bulbs, or possible even measure reactive as well as real power.

I agree with David, "All they said was that after 6 weeks all except the LED one were dead. So, does that mean one bulb only lasted 1 day of that treatment, or they all died after 6 weeks? Give us the data! How many cycles did each ones survive?" Wasn't there a time-lapse date/time-stamped video recording made of the month-long experiment so they could review the recording to determine when each of the lamps failed. This experiment has potential to be extremely useful as a practical exercise.

I'd love to see the second experiment done with heating elements. Does it take more energy to start up "always-on" heating elements, like potpourri burners or heating pads, than it does to leave them on all the time?

What I want to know is WHERE I can get one of those LED bulbs. I have a thing for LED's and I have never been able to find that wicked cool bulb :-)

It would have been interesting to see the effects that a dimmer switch has on bulb life vs. a standard switch. My suspicion is that the dimmer will prolong the bulbs life even if it's taken to full on each time, because of the current is gradually applied to the bulb over a few seconds vs. the slam it gets from an on/off switch.

I think like the other posters that they dropped the ball on their bulb longevity experiment.

Nice episode but I too would have liked more detail -- you should be able to get duty cycles from a manufacturer to compare with the experiment.

Here's an article with some details.

Chris Taylor,

The style of bulb they used in the experiment, along with lots of more-conventional-looking LED bulbs, are available at . I'm sure there are other stores, but this one specifically tagged one of their bulbs "As Seen on Mythbusters." Whether they actually supplied said bulb, or just happen to carry the same model, I don't know. And their website says that the spiky LED bulb draws ten watts, not one. If it be true, that's a rather important mistake.

The dimmer would be interesting as someone else mentioned. My biggest question is when in the 60hz cycle where the bulbs turned on. If an Incandescent buld is turned on at the peak of the cycle (aprox 170v) - it's life will definitely be shortened, vs turning on at the zero-crossing. back in the 70's in a quest to make light bulbs last longer in University - we developed a zero-crossing swith that more than doubled the life of incandescent bulbs.
Also - if all the bulbs were 'burnt out' except the LED after 6 weeks - then turning them on and off certainly shortens their life-span - as even a regular Incandescent should last over 6weeks if on continuously.

I have multiple CFL's that are on virtually 24hrs / day and are over 5 years old.

Looks like the EPA needs to check out myth busters. I thought their recommendations for when to turn off lights seemed much too long.

The EPA ratings take into account something the Mythbusters seemed to overlook: heat output from the bulbs.

Ever notice how office towers leave many of their lights on all night? The main reason is shutting them off would require massive output of heat from the HVAC units to have the towers at a comfortable level in the morning.

There is a point where turning off incandescents and flourescents is less energy efficient than leaving them on all the time; this does not apply to residences, however, where the number of lights are so few and the space small enough to quickly heat with an HVAC and HRV.

On the 22,000 foot fall, two things. Ted's cable system didn't fail. When the balloons were cut loose the tension on the cable was gone thus Ted and the cable fell straight down to the ground. That, in my eyes is a big engineering gaff by the Mtythbusters. 2nd, I'm sure that falling into the blast wave of a large explosion is fatal. However, watching the video you'll see the mushroom cloud rising up and a column of heated dust pushing up underneath the mushroom. Could a body falling through such a column be slowed enough to survive?

Reguarding the lights on article, the EPA info seems to be pure garbage. Mythbusters actually measured the current being used. I can trust their data. The heat put off by the lights burning could only be cost effective if you were heating with electrical resistance heat. It would be particularly bad if you were cooling the house. Besides, if the MB data are right it's too small a factor to consider.

I've just watched this episode, and I'm really confused about the light bulb experiment - namely the numbers they came up with.

To start with, they measured "steady state power" over one second for the incandescent bulb at 59519Ws (Watt-seconds). Would that not imply that the rating of the bulb was about 60kW? Is that not 1000 times higher than you'd expect? I suspect the equipment was displaying mW rather than Watts.

Secondly, in reference to Jim's comment, it is MB themselves which use W/HR (which technically is not absolutely meaningless, but in this context is definitely incorrect). I'd say that the correct unit for steady state is simply Watts, as it is already a unit of rate. But can accept they were trying to measure total energy consumption over an hour. It looks like they may have realised their earlier mistake though, as the board ended up showing readings in Watts (eg 90W for incandescent) rather than kW.

And lastly, Jim's very first comment mentions comparing similarly rated bulbs. I think this is not necessarily as they are comparing a bulb being left on to the same bulb being switched off. It doesn't need to be similar to the other bulbs in the experiment.

Oh wait: I do have one more comment - relating to Tom's last post (before this one): When MB are making several basic errors in what the units even are, I would be not so willing to trust their figures as authoritative as you may be.

Regarding the 23sec for the fluorescent, they don't say whether this was an old-style magnetic ballast or a new-style electronic ballast. The CFL definitely reflects the payback time of an electronic ballast, so I can only guess the fluorescent tube they measured was using the older magnetic ballast. Electronic ballasts are much more efficient both in startup and steady-state mode.