« January 2008 | Main | March 2008 »
Episode 8:
Runaway Trailer
The Challenge: Unlike
a semi-truck, there is no way to stop runaway boats and motor homes attached to
trucks in an emergency. The Smash Lab
team has an idea.
The Material: Retro-firing rockets; enough said.
What to Watch For: Retro-firing rockets attached to a runaway trailer and, hopefully, bringing it to a stop.
When to Watch: Wednesday, Feb. 27, at 10 p.m. ET/PT
Episode 7: Fireproof House
The Challenge: Protect a home from radiant heat from a forest fire and the flames themselves.
The Material: Nanogel, a cutting-edge insulating material, also known as “frozen smoke.” Supplied in granular form, it is a lightweight solid that flows like water and is often used in green building construction. It is the only insulator better than air, but how will the team harness it for this challenge?
What to Watch For: The heat that the team has to create -- 500-600 degree F radiant heat that preheats a wooden home almost to its ignition temperature, followed by 2,000-degree F flames that instantly ignite it.
When to Watch: Wednesday, Feb. 20, at 10 p.m. ET/PT. Get a reminder.
What You Missed: Be sure to read Deanne’s behind-the-scenes calculations from High Rise Escape-- The Magnetic Pull of Experimentation.
Photos: AP
Deanne writes:
The magnet experiment is one of my favorites. Here are the basics of our exploration:
Rare earth magnets - commonly referred to as “NIB” magnets (neodymium-iron-boron):
NIB magnets are extremely powerful for their size. Everyone that came anywhere near the magnets had to keep fingers, cell phones and computers at a safe distance. The downside of NIB magnets is that they are very brittle and can shatter if dropped.
The scientific principle of Lenz’s law:
When a magnet is moved towards a coil of wire, it induces a current. The direction of current that is induced is such that it opposes the change in magnetic field that created it.
Translation to modern roller-coaster brakes: When you move a conductive, non-ferrous material, such as aluminum, by a really strong magnet, eddy currents set up and a force is produced that opposes the direction of movement.
The Halbach Configuration:
Klaus Halbach discovered a configuration of magnets that creates a concentrated magnetic field on one side by arranging the poles. Originally, he designed the configuration for particle accelerators.
Our tests:
Behind the scenes, we explored the possibility of using electromagnets and NIB magnets, but the elegance and simplicity of a passive rare earth magnet braking system (that requires no back-up generator) won us over.
The electro-magnet exploration involved magneto-man style electromagnet boots and gloves. It was so hilarious that the entire camera crew was brought to tears.
After a successful ramp test, it was time to get serious about the math and figure out a force versus velocity profile for the magnetic configuration we built.
Magnet braking systems have many fine-tuning adjustments that can be made. You can modify any of the following variables to alter your applied braking force:
Size of magnets (i.e., 1-inch cube, 2-inch cube)
Arrangement of magnets (standard stack, Halbach array, double Halbach array)
Air gap distance between magnets
Thickness of the aluminum fin
But, the braking force applied by the magnets is not a constant! The braking force due to Lenz’s law varies depending on the applied force, so I had to test multiple data points to get a trend curve.
Force is mass times acceleration, so I had to either change the mass or change the angle of the ramp. Changing the mass is much easier, so that’s what we did.
We did two series of ramp tests – the first was with a single Halbach array of 1.5-inch magnets.
It was a very basic physics model:
We had an inclined plane at 30 degrees (sin30g = .5g)
And a sled with “x” sandbags (mass in slugs = x lbs/32.2)
I plotted a graph of each sled’s velocity versus the applied force.
Fapplied = .5mg
v = distance/time
Using these points, I found a trend curve and then extrapolated what the final speed would be for a final applied force of 250 pounds (going straight down).
(My model took into account rolling friction as well.)
At low velocity, magnetic drag is linearly proportional to velocity. All of our tests were at low velocity, and my plot was linear, so I knew my data was looking good.
But our single Halbach configuration resulted in a final speed that was a bit too dangerous and/or fun (depending on how you look at it).
After referencing predetermined trend curves for changing magnet size, configuration and air gap, I decided to revise the system to use a 2-inch double Halbach configuration.
Then we repeated the test (which you don’t see) and calculated our approximate final test velocity – which was 3.1 feet per second.
But everything is in theory; we had to take it to the final test.
We dropped Chuck (275-pound load with the sled) at a descent rate of 3.4 feet per second. It was a nice controlled descent.
The test was a successful, calculable, proof of concept. The magnet design could be tweaked in a number of different ways to create whatever size escape pod that you want. We brainstormed many ideas of how to make larger-capacity pods with a variable design that adjusts fin engagement based on the applied load. These are just two of the many ideas we had to implement this on a manufactured system.
If I had to do it again, I would choose a faster descent rate. (Chuck was a little too controlled going down the tower.) And I would experiment with alternate pod designs for larger capacity loads.
Question:
What’s your take? How would you have used magnets to escape from a high-rise?
Episode 6: High Rise Escape
The Challenge: To engineer a quick escape from a burning high-rise building.
The Material: Magnetic roller coaster braking system.
What to Watch For: A stunt man sliding down a 100-foot tower.
When to Watch: Wednesday, Feb. 13, at 10 p.m. ET/PT. Get a reminder.
Deanne writes:
I focused my research on two main areas for this episode. The first was how to make a quake, and the second was how to defend against it.
I spoke with many earthquake specialists regarding earthquakes, the scales that are used to assess their magnitude and intensity, and how to best test a home using a shake table that cannot dial in a specific earthquake, rather can create simple harmonic oscillation.
Earthquakes are incredibly complex, as we already know. If you look at a seismogram, you will see the varied nature of an earthquake and how non-harmonic they are. An earthquake is comprised of different forms of waves that can constructively interfere and create enormous ground accelerations – referred to as PGAs or peak ground accelerations.
P-Waves are compression waves. They arrive first in an earthquake, but are the least responsible for structural failure. P-waves are small in amplitude and a house can easily defend against them.
S-Waves arrive later, but they are much more harmful to a structure because they are more likely to shear the home at its foundation and fatigue the structure. S-Waves are responsible for the very strong lateral movements that are extremely dangerous to a structure.
And, while the show didn’t go into depth about my math, the numbers I was crunching on camera were in regard to the home’s natural frequency. Natural frequency is another big player in structural failure. If the earthquake shakes at a frequency that matches the resonant frequency of a house, then the whole house will shake itself apart.
A double-story home will naturally vibrate back and forth once over the span of about .2 seconds (5 hz) when given a big whack. Even though you can’t see it, your house responds to that whack just like a tuning fork. Tall buildings will have a longer natural period – upwards to .7 seconds and beyond. Depending on the distance from the epicenter, earthquakes can have a range of frequencies from .01 Hz to 10 Hz, so this is a big issue in base isolation system design. Resonance is nothing but bad news for a structure.
You will notice that I never mention the word Richter on the show. After speaking with many experts, I decided to specify our test with intensity, another measurement used to assess an earthquake’s strength. Richter takes the log of amplitude, but assumes you have a typical seismogram obviously. And the amount of error in extrapolating a Richter value to our shaker was too great in my book. Intensity, on the other hand, is measured using the Modified Mercalli Intensity Scale. The Modified Mercalli Intensity (MM) Scale measures the ground vibrations that were felt during an earthquake on a qualitative basis.
Traditionally, simple surveys were sent out for residents to describe damage. Intensity ratings were given according to these surveys and plotted with relation to the epicenter. Intensity values range from I to XII based on damage, personal accounts and records.
http://earthquakes.usgs.gov/learning/topics/mercalli.php
While intensity is qualitative and rated according to damage, it is usually approximated to a range of peak ground accelerations. And to get peak ground accelerations, you need two quantities – frequency and displacement, both of which we have. So ... to conduct our test, we chose a quake, looked at intensities felt during that quake, and dialed in frequencies and amplitudes that would shake our house with corresponding peak ground accelerations.
The Quake: Northridge, Calif. Magnitude 6.7. 1994
Here is a MM chart of the Northridge quake:
http://www.cisn.org/shakemap/sc/shake/Northridge/intensity.html
The Tests:
We tested three intensities: a very small intensity felt many miles away from the epicenter; a sizable intensity felt near the epicenter of Northridge; and one huge, heaping helping of peak ground acceleration that is off the charts … because we can. =)
To give you an idea, a sizeable quake, according to experts, is on the order of .4Gs PGA. And I was told that the largest PGA ever recorded is on the order of 1G. I’ve definitely never heard of any peak ground accelerations above 2Gs, but our last test was above 2Gs of shaking!
Peak Ground Acceleration = displacement * (2*pi*frequency)^2
Our homegrown shaker dialed in a displacement and a frequency, based upon motor speed and crank arm length, to correlate with the PGAs estimated on the Northridge quake map.
Our Base Isolation Systems:
We split up into two teams:
1. Chuck and Kevin - Used ball bearings to create a near frictionless surface, and used shocks and dampers to contain the house.
2. Nick and Deanne - Used ball bearings to roll in a bowl, using the uphill slope (gravity) as the containment “spring.” We modified an existing system called the Friction Pendulum System that is used in skyscrapers today.
Nick and I went on a field trip to Pasadena City Hall to check out an installed friction pendulum system.
http://www.ci.pasadena.ca.us/cityhall/
Pasadena City Hall is a historic masonry building that has recently been restored and retrofitted to sit on a friction pendulum base isolation system. There is basically an underground moat circling the building. The entire building sits on a series of bowls in the middle island and has an overhang of metal along the perimeter that bridges across the underground moat (and stairs that are sacrificial in the event of a huge quake). The amazing feat of engineering is how it is installed. In between every load-bearing column, a bowl was placed, a spanner beam was added to transfer the load, and all of the load-bearing columns were moved 5 feet or so to the left … along the entire building. The idea is this – if an earthquake comes, the ground can rattle back and forth, but the building stays isolated in its bowls as the ground is moving violently underneath.
The friction pendulum system has two main design principles. The first is that the bowls are shaped to lengthen the building’s natural frequency response. Typically, a short, two-story structure has a resonant frequency of 5 Hz (.2-second period). The shape of the bowl extends the structure's natural frequency to whatever concavity you design it for. And the math is simple. If you treat the spherical radius of the bowl as the length of a pendulum arm, you can calculate how long the arm should be for a certain period using simple pendulum equations. We chose 3 seconds. So the idea is that if an earthquake shakes back and forth once in .2 seconds and the natural frequency response of the house is a period of 3 seconds, then resonance will never happen. Pretty cool, huh?
The second part of a friction pendulum system involves an articulated slider mounted to the structure that rests in the bowls. They describe them as hockey pucks at the city hall. This is where our big difference comes into play. We’re not using an articulated slider for our house, which relies on friction, but we’re using ball bearings, which are near frictionless and roll instead of slide. Basically, as the earthquake comes, the structure and the bowls shift together, then the static friction is overcome and the slider rests in the center of the bowl as the ground shakes back and forth.
What we implemented:
We basically designed a rolling pendulum system. We chose a natural period of 3 seconds for the house, but with ball bearings we have little to no friction, so the vibrational response will be undamped without any additional damper. (i.e., a saloon door versus a new modern door ), so we decided to line our bowls with an extremely tough durometer of rubber to dampen the response. And you see in our skateboard ramp demonstration that everything works as planned – until we put a heavy house on top. We knew that our failure point would be the rubber, but we were hoping that we could roll with the punches – literally. But the rubber was extremely problematic. So in our final test, we cut out the rubber, went for a pendulum system without anything to dampen the response, and it worked extremely well! The house just had to overcome the bit of friction in the point contact with the ball bearings, and settle into its groove. It was beautiful to watch. It was an isolation system in all its glory!
I guess I should talk about Chuck and Kevin’s stuff, too. The interesting thing about their test is that I measured the accelerations on the house during the test (using an accelerometer and my computer), and they were about .1 to .2 g's lower than the peak ground accelerations we input into the shaker for the mid-sized quake. So they were doing extremely well in my mind until our pendulum system was tested. Unfortunately, the accelerometer got demolished when their house went down, so I calculated the g’s (accelerations) of ours just by looking at the displacement and frequency of the house as it moved relative to the stationary ground. Our house was isolated and then some! The accelerations felt by the house were tiny during the large earthquake test. It was on the order of .01 g's felt by the house with over a 2g vibrational input into the shaker. That’s isolation, baby!
So, all in all it was a success for our ball-bearing proof of concept. Granted in the real world, you would use spherical bowls, not single dimension tracks, because the earthquake could come from any direction depending on your location with regard to the fault line. But this cheap, affordable and elegant solution is a big check in my book of science experiments with potential for much bigger implementation.
Episode 5: Earthquake Proof House
The Challenge: Prevent a house from shaking during an earthquake.
The Material: A ball bearing floor system used in the cargo hold of airplanes.
What to Watch For: An earthquake machine capable of going off the Richter scale.
What Happened Behind the Scenes: Deanne waxes poetic about frequency and intensity. Look out for her blog later in the week.
When to Watch: Wednesday, Feb. 6, at 10 p.m. ET/PT. Get a reminder.
Recent Comments