November 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Sun, Nov.13.2005:12:28 PM

MythBusters: Steel Toe Amputation, Bottle Rocket Blast Off

This site is not sponsored or affiliated with the Discovery Channel/MythBusters. Please visit the Discovery Channel's MythBusters fansite for official content.

Myths tested:

  • You can launch someone 30-40ft with a bottle-rocket-powered backpack: mythbusted (the engineering on this one was a bit questionable)
  • Steel toe boots have a higher amputation risk than regular boots: mythbusted

bottle rocket shearing

Update: screenshots removed at the order of the Discovery Channel

Read on for more episode details.

Steel Cap Amputation

Myth: Steel-toed boots are more dangerous than regular boots -- if something falls on the boots, the steel can curl in and cut off your toes

They were able to find one occurrence of amputation while wearing steel-toed boots occurring in real life. In 2002, an Australian worker lost his 3rd toe when some steel pipes feel from a forklift.

Adam and Jamie constructed various tests for this myth using both a guillotine toe-smasher and an arbor-press. Initially they used frangible feet that Adam made, but it turned out that they made a mistake in assuming that their frangible feet would model real human feet being crushed. For better comparisons they ended up using clay.

Frangible Feet Construction

Adam constructed frangible feet to test with based on landmine frangible feet. After testing chicken legs, bamboo, and fiberglass as substitutes for human bones, he decided to use fiberglass bones. The bones were set in a ballistics gel cast of Adam's leg.

For those wondering, the full frangible leg construction process was:

  • Pour dental alginate over leg and surround with plaster bandages to get leg mold. (link to website on making plaster casting with dental alginate)
  • Fill alginate mold with silicon-based rubber to make a rubber leg
  • Make plaster mold of rubber leg
  • Make a plaster mold of a skeletal foot
  • Fill mold with hard resin fiberglass cocktail to make bones
  • Use hot glue gun to make tendons to connect bones
  • Place resin bones inside plaster leg mold and fill with ballistics gel mix (used a different ballistics gel mix than usual)

Guillotine drop test 1

NOTE: in turned out that the results from this test were somewhat invalid. After testing with the steel-toed boots they tested with the regular boots and discovered that the ballistics gel was too springy and was invalidating their results. While ballistics gel is good for simulating bullet impacts on flesh, it's not so good for testing crushing.


  • Guillotine-style toe crusher that drops a flag metal bar onto the toe of a boot beneath.
  • Used the highest-rated (ANSI-75) steel toe boots.

boot guillotine


  • 75lbs from 3 feet (official ANSI test height and weight): mashed the leather down a bit, but nothing injurious.
  • 400lbs from 3 ft: more deformation in the steel plate, but only damage to frangible foot was a broken metatarsal (big toe). Adam: "I want to see some toes cut off or crushed beyond all recognition"
  • 400lbs from 6 ft: a lot of pancaking of steel cap and lots of broken bones beneath, but no toe amputation.

400lb/6ft boot:

pancaked boot

They didn't detail the results from the regular boot because of their discovery about the ballistics gel being too springy.

Guillotine drop on boots filled with clay

Because of the ballistics gel problem they decided to use clay instead of the frangible bone legs they had constructed. Clay is the method ANSI uses to test boots.

At the official test height of 3ft with 75lbs there was 0.5" of clay compression with the steel-toe boot, which is exactly to spec. The regular boot failed horribly, with the clay being completely splattered.

setup splattered

Arbor press test to find total failure point

They used an arbor press to squish boots to their total failure point. The steel-toe boot was able to take 6000lbs of pressure before total failure; the regular boot was only able to take about 1200lbs, which was hard to measure as it failed so quickly.


complete failure from steel-toe boot:


complete failure from regular boot:


Shearing attachment tests

In order to test a worst case scenario, they made a shearing attachment, which was a thin metal plate that would hit the boot on edge.

They mounted the shearing attachment to the arbor press: at 750 lbs it broke every bone in the frangible foot; at 1400 lbs it severed all the bones in the feet.

They then mounted the shearing attachment on the guillotine and raised it to it's max height of 6ft and max weight of 400lbs. The blade glanced off the steel plate, shearing the entire shoe in half. They tested again and got the same result. In this particular scenario, were a heavy blade to drop on your foot you could actually lose more of your foot as the steel cap could direct the glade further up the foot as it did in the test. This isn't the failure mode described in the tests, though, and regardless of what type of boot you used there would be amputation.


Mythbusted: They had to mount a blade in order to get amputation with the steel toe boot and all their other tests showed much more damage to the foot when regular boots are used.

Bottle Rocket Blast Off

Myth: You can launch someone 30-40ft with a bottle-rocket-powered backpack.

This myth is based on a Japanese game show clip. In the clip, a person with soda bottles strapped to their back was able to fly 30-40 feet using only water and a bicycle pump for power. The world record for a bottle rocket is 517m, which was set by US Water Rockets' X-10 in Galway, New York. They don't know how that rocket was built, but according to Adam "... it seems to be not a soda bottle but actually an 8ft tube of fluorescent light protective covering polycarbonate wrapped in high-strength composite powered solely by tap water and air pressure"

Update: this appears to be the Japanese clip of the bottle rocket launch. It looks pretty real to me.

bottle rocket bottle rocket bottle rocket bottle rocket

The plan for the mythbusting was to launch Kari using a rig constructed according to the game show clip. They first had to test bottle rocket models in order to figure out how much water and what pressure to use. They also needed to figure out how many bottle rockets they would need. They focused on a bottle rocket made out of a soda bottle, but though they also did some testing with a water-cooler-bottle-based model.

Parameter tests

Grant determined that 1:3 was the best water:air ratio for their soda-bottle rocket by repeatedly setting off a bottle rocket and measuring the height.

bottle rocket

They tested a 2kg payload at 60 psi to figure out how many bottles they would need:

  • Soda bottle: 20 feet of altitude
  • Water cooler bottle: "Holy Crap run!" off the charts

At 60 psi they would Need 28 bottles, which is more than the 15 used in the game show clip, so they decided to see how much more pressure they could put into the bottles. Using a bike pump they were able to get a the soda bottles up to 95 psi before their arms gave out. In order to find the failure point of the bottles they hooked up the pneumatic pump from Grant's "Deadblow" battle robot. The soda bottles exploded at 150 psi and the water cooler bottles exploded at 95 psi, so they decided that using soda bottles was better for the test even though they got much more lift out of the water cooler bottle.

Retesting with a 3.5kg load at 90psi they were able to get the soda bottle rocket to go 16 ft. The load was 1/15th Kari's weight, so it's possible that 15 bottles could lift Kari 16 ft.

Kari: "Wow that went exactly to the top of the scale, that means we have exactly 16ft... and uh that's not metric so let me do some conversions really fast and it's really high... we'll come back to that"

Launch Setup

They built two rigs for the launch. The first rig followed the constraints of the myth, so it was bicycle-pump powered and only used 15 bottles. Other build details for the first rig:

  • Bike inner tube valves glued to rubber stoppers
  • 15 cable-tie quick release systems secured in plywood housing (aka "quick release platform')
  • 15 PVC collars fixed to a trigger plate and placed over the quick release platform

setup quick release platform quick ties setup

The second rig was meant to replicate the results in case the first rig failed, so it's didn't follow the myth constraints. It used 50 bottles and was fed with compressed air.

Kari tes launched the first rig. It launched but the rockets came apart in mid-air because they were not secured well enough together.

The Launch

They launched from a pier on the San Francisco Bay. For safety reasons they launch with Jane the [Simulaid dummy][] first. For similar safety reasons they never ended up launching Kari.

They tested with both the 15-bottle and 50-bottle rigs. Tory had finished the 50-bottle rig at 2am the night before and never tested it, so it was questionable how sound it was.

Launch 1 (Jane, 15 bottle rig):

The harness failed and the rockets launched in an arc over the simulaid's head, then yanked the simulaid down into the water below. Not very safe.

bottle rocket bottle rocket bottle rocket bottle rocket bottle rocket

Launch 2 (Jane, 15 bottle rig):

They positioned Jane higher on the harness but Kari didn't like the way they wrapped a tether around the simulaid's neck: "This is supposed to me." Jane flipped into the water but didn't get much height or distance. Tory: "Well I'm a little bummed though that we're not going to actually launch a person on it. I mean, looking at the high speed that was fun.. it's didn't look that dangerous but you know I guess we're all worried about breaking necks and people getting paralyzed"

bottle rocket bottle rocket bottle rocket bottle rocket bottle rocket bottle rocket bottle rocket

Launch 3 (Jane, 50 bottle rig):

The quick release failed. As Kari wailed on it with a hammer it finally launched, looking more like a water explosion than a rocket launch. The rockets on the left released cleanly but the rest got caught on the o-rings. The simulaid flew about 1 foot.

bottle rocket bottle rocket bottle rocket

Launch 4 (Jane, 50 bottle rig):

They lubricated the o-rings to get a cleaner release and lowered the angle of launch so that the water in the bottles would be more level. Once again the quick release failed. Tory stomped on the platform to get it to trigger. The simulaid flew far enough to make it into the water with maybe a foot of lift.

bottle rocket bottle rocket

mythbusted (couldn't achieve the height or the distance in the myth, nor was the rig safe)

Although the bottle-rocket myth was judged to be busted, the build team's engineering on it was a bit questionable. There were enough problems with the rig that their results were far from optimal. If you compare the screenshots of the first and second launches, the rockets continued to shoot water for much longer in the first launch; in the second launch it looks like most of the water was already expended before the rockets even left the launch rig.

More MythBusters episode summaries

Subscribe to comments:
rss comments netnewswire bloglines Subscribe to comments with Shrook Subscribe to comments with My Yahoo!

What the fuck are you smoking man, the asian video looks more fake than anything i have ever seen. He would of flown just the same way as it did in the mythbusters.

Posted by: ErosNemesis at December 11, 2005 12:48 PM

I beg to differ. The man in the video didn't look as though he weighed more than a buck ten.

The real trick is to figure out how many foot pounds of pressure that a single bottle could give off. I bet that a single bottle could give off about 3lbs for about 2 seconds. At that pressure you would need 37 bottles to move him.

But if you could pressurize the bottle to 140lbs of pressure (I have done it without explosion) then you could easily move a person 40 feet.

Posted by: Arnold at December 16, 2005 02:00 PM

I also disagree. The guy is a small Japanese man. Those are not regular water bottles that we have here. The pack they use is much bigger than yours by volume. They could achieve better liftoff with a high pressure nozzle than what was tried on your model. Japanese traditionally will not make something and be happy with it. They will continue to improve it never believing it is finished. That's why I believe they put much more R&D into their pack than what you have done. Believe it or not.

Posted by: kelcey at December 17, 2005 06:36 PM

the japanese video was definatally a fake. with the crooked takeoff and spining as he did, it would be impossible to achive the perfect arc shown in the video.

Posted by: Figglet at January 2, 2006 05:07 AM

I'd love to get a high-rez of the video to really analyze it. It doesn't look composited - no mismatches in lighting or color. it does look like he was lifted off the podium and carried the distance by a crane or helicopter - the "thrust" of the water bottles doesn't match his trajectory at all. It's pretty easy to remove wires or cables in post-production. That's what I'm betting happened.

Posted by: Raschied at January 13, 2006 03:46 PM

can u send me a list of all the things i need to make a water powered bottle rocker plz and thanx

Posted by: bobby at January 14, 2006 09:46 AM

It could be real, i had to do an Extended experimental investigation assignemnt for physics about the properties of water rocket acceleration. Because the bottles have large openings and not being launched at a vertical angle, all the water that is possible would have been released in less than a second therefore his intital trajectory would reamin relatively unchanged. so the perfect arch could very well be achieved. Water rockers don't exaclty have a long "burn time" they just have an immense impulse at the start. During the investigation we had a person break his scapula after the launch pad tilted and shot straight into his back after he turned to protect himself, and the water bottle was only at about 70 psi.

Posted by: Ben at January 18, 2006 10:29 PM

I remember making these 25 years ago and being amazed at the impulse produceable by even a small bottle iv even done glass ones ( not for the back vard physicist!) I think I remember the champaign bottle being mental.

Experiment with exhaust nozzles and dont be scared to go too small as reducing the size increases the exit velosity and lengthens the burn.

find the safe pressure that you are confident will not go off in your face and you really can drive a truck over a coke bottle without the lid sheering off or splitting. (but seriously, wear goggles) eyes are really soft!

once you have those two factors then decide on the weight the air to water ratio i have done one with helium and petrol both substantially lighter. Not on your mums lawn!

I remember using the thermoset plastic bell from the end of a bathroom light cord for the smaller ones and just drilling larger and larger holes in it and epoxying it into place the later ones i made from other plastic bottles cut curved and glued. but really a hole in the lid is quite good enough

The trigger was usually restraining the bottle onto a grolsh type lid basicly a grommet on a cone setup drilled to allow preasurisation. Just need ever larger grommets and a file for the cone to accomadate the ever increasing diamiter of exhaust.
then have a breakable stick just resting accross the other end restrained by a u shaped arm comming in from the side this arm is moved down bending the stick and you get a sudden release when it goes if u have the right material it will snap clean and not effect the launch.
Yes ispent alot of time on this one i then progressed to rapid burn/explosives resulting in a friend ending up with a very special hand.(not advised)
This ones for free my little invention ever told anyone this one....2" dia x 2" leingth metal tube....feed a baloon over the end centrally so nipple is in centre ...drop 1 ball bearing in ...pinch between fingers while encircling the tube/baloon pleats with other hand....pull back and release straight as to not catch your fingers (please not at anything with blood!) cos it will spill!

Posted by: Corholio at January 20, 2006 07:01 PM
Post a comment
NOTICE: Comments are encouraged, but I reserve the right to edit or remove any comment that is offtopic or spam.

Remember Me?