Mythbusters bridge resonate attempt

[wingnut]

Hi

I would love some help and suggestions for a project I have been wanting to attempt for the last year.

Mythbusters had an episode where they tried to get a bridge to vibrate more and more by supplying a push at just the right time. It worked with a small linear motor and a huge steel arch bridge did noticeably vibrate. I think I detected a flaw in their method, which I would like to improve on.

Their linear motor vibrated at a fixed frequency. What I want to do is detect when the vibrating object reaches the end of its swing and at this point activate something to give a push.

The "push" troubles me too because according to Newton III, every action has an equal and opposite reaction. Which means that one should never be able to self-start a swing - but we do.

I bought a solenoid and an accelerometer (which looked big on the pic but was the size of a microdot when it arrived). I am not sure these are the best detectors and responders but any ideas as to how to best add to the resonance of an object would be appreciated. I am hoping to apply this from 1Hz to uHz frequencies of vibration. Like Tesla, I would like to try to vibrate objects to bits

 

[wingnut]

Maybe I could frame the question in another way...

Suppose one were to attach a sensor to a garden swing to detect when one reached the end of the swing, what sensor could one use?

My first reaction would be an accelerometer, but I cannot use these since my supplier supplies only ones too tiny for me to use. Any other suggestions would be welcomed.

 

[BobK]

Hey Wingnut, if you plan to knock down any bridges, could you let us know ahead of time so we can avoid them? Maybe even get a cool video, like this one:

http://archive.org/details/SF121

Bob

 

[Raven Luni]

I love mythbusters. But didnt they say never to try any of their stuff at home?

 

[duke37]

Why is the accelerometer too small? You could glue it to a brick.

 

[wingnut]

Hey Wingnut, if you plan to knock down any bridges, could you let us know ahead of time so we can avoid them? Maybe even get a cool video, like this one:

http://archive.org/details/SF121

Bob

Hi Bob, thanks for the reply.

I am surprised they did not try to save the Takoma bridge by anchoring a ship to its span to stop it vibrating. But maybe they also like to see stuff shaken apart

I will not post a video of my "Takoma bridge" consisting of a saw blade with a motor attached to the end of it, the motor having an offset flywheel. As the Arduino increases the pulse width, this ups the voltage, speeding up the motor. At a certain rpm the blade goes beserk as the motor reaches the natural resonance of the blade.

This is similar to what the Mythbusters did, but this is not good enough. I need the motor to intelligently syncronize with the blade's vibrations.

Maybe if I attached a laser to the blade, the wider the sweep of light, the closer I am to what I want.

 

[wingnut]

Why is the accelerometer too small? You could glue it to a brick.

Hi Duke

I bought two LIS33DE accelerometers, and when they arrived they were so small I could hardly find them in the packet. And the diagram of them looked so big on the datasheet

These 3x3x1mm chips have 16 leads and I have no clue how anybody attaches them to a circuit. It might be within my skill set however to glue it to a brick.

 

[CocaCola]

These 3x3x1mm chips have 16 leads and I have no clue how anybody attaches them to a circuit. It might be within my skill set however to glue it to a brick.

Get a stencil, solder paste and then reflow, is very doable even at home on a budget... You could also attempt it by tinning the pads, with a small amount of solder and letting it self level on the board with a heat gun, use a little extra flux, but this would not be my preferred choice for hidden pad chips, a stencil and solder paste is the way to go...

Shipping might make it cost ineffective, but the offer for me to set the chip on a breakout board for you is out there as long as you cover the shipping cost and break out board cost... I'll cover the stencil cost as I want to get a misc multiple small package stencil made up for my own use anyway...

 

[wingnut]

CocaCola - Thank you for your very generous offer of help.

If I understand you correctly, these chips need their very own circuit board. Then one gets a friend to hold it on the exact spot while I try to solder it.

I will first try a few other methods which I have in mind, and if these fail, I will take you up on your kind offer. One book suggests gluing a ballbearing to a piezo disk, and every time you shake it, it generates a voltage. I am hoping to crack this nut today, but I will keep you updated.

Good to have you back after what sounds like fairly torrid travel arrangements that you had to suffer.

 

[CocaCola]

If I understand you correctly, these chips need their very own circuit board.

No necessarily, but it sure make prototyping a whole lot easier...

Here is a breakout board, that will convert that LGA 16 chip to DIP 16

http://www.proto-advantage.com/store/product_info.php?products_id=2200123

And here is a whole kit if you want to DIY

http://www.proto-advantage.com/store/product_info.php?products_id=2210123

 

[Mongrel Shark]

The motor must want to oscillate as it moves. Mount it so it can, just enough to activate momentary switches. Placed as appropriate to your needs. Probably over and under. How you get them to "push" your motor I have no idea.

 

[wingnut]

Here are some of the things I tried, in order...

1. A computer speaker attached to the vibrating blade - no voltage generated.
2. A computer speaker with a nut superglued to the center of the speaker cone, to create resistance to shaking - no luck.
3. A piezo speaker attached to vibrating blade - a bit of vibration detection.
4. A Piezo speaker with metal glued to the disc - quite good vibration detection, but nada when it shakes the most. Only picks up high frequency vibrations.
5. A stiff conductor surrounded by a conducting spring. As it vibrates, acts like a switch. Quite good vibration detection, but touches randomly too, so no decent pattern.
6. A tilt switch, which when violently shaken produces a spike. Not sensitive enough, and random spikes.
7. Microswitch with a long actuator arm having a wingnut glued to the end of the arm to create momentum (resistance) - stuck with double-sided tape and cable-ties to the end of the vibrating blade - EUREKA

#7 worked great. As the Arduino increases the pulse-width the motor sped up - till it hit a frequency (38Hz). Then the blade shook the switch violently and regularly, creating a beautiful, clean switching action.
I have attached photos of the motor and switch, and oscilloscope graphs.

I am not where I want to be yet, but I am a lot further than where I started 5 hours ago. Thanks for the help so far.

Mongrel Shark - I think #7 is similar to what you suggested.

 

[wingnut]

Here is an update...

The best system so far to measure the swing of a saw-blade was to attach an IR LED to the blade and use a phototransistor to detect the resonance.

I seem to get a near perfect replication of what the blade is doing on my oscilloscope and frequency meter.

Here is where I need some advice from anyone who has experience with resonance...

What I seem to see is happening is that as I increase the pulse width to increase the speed of the motor with its offset weight, the rpm seem to "stick" on the natural resonant F of the blade which is 24 Hz. The blade will vibrate at other F but less so, and does not stay on these other F with the same persistance.

Is this observation true of most other natural phenomena involving F?

For instance, if there is a sound wave with a F CLOSE to the natural resonant F of a guitar string, will the guitar string vibrate in sync with the sound wave? The answer seems obviously yes?

Will a tuning circuit which is CLOSE to (but not exactly the same as) a broadcast F tune itself into that broadcast signal?

If the answer is yes, then it seems a fairly easy task to shake a bridge (or any object - I might try a glass first) to bits. You just have to attach a motor with an offset weight, and when you get close to the natural F of the object, the motor seems to "tune itself in" to the object it is attached to. That at least is my preliminary thesis, and I would love to hear if anyone has experience of the same phenomenon.

 

[Mongrel Shark]

I don't know much about resonance, Although I am trying to learn more so this interests me.

You seem to be talking about making a tuning fork.

Getting the natural frequency of an object would probably be the hardest frequency to shake it apart on. Unless the object is made out of crystal, or something verry brittle and resonant at the same time. Although Finding out the freq of the object, would make it easier to find the right freq for destruction.


I would think you would be looking for an opposite but resonant frequency. Ie the bridge can vibrate at its natural frequency forever. Thats what it does. Finding a frequency it won't vibrate at, then making it do it anyway would be the way to take it apart IMO. Look at bridge length and width. See if you can find a freq with a length half, quarter or 8th of the dimensions. I don't know that will work, but it's where I would start. Bridge thickness and flexibility will come into it too. Look at ossilations per division of bridge length/width.

It's my understanding, that you can make something (like a bridge or glass) vibrate faster than natural, and it wont necicerially break it. It's a specific freq for every item that must be reached. In the case of the famous old bridge footage. (which I saw in science class when we did resonance, many years ago) I think it was a much lower frequency. Getting close to one oscillation per suspension cable seems to have been what killed it (there where still a few theories when I was at school. Anyone know if it's been settled yet?). With the glass, I think it only works with crystal? Crystal has a natural freq it vibrates at, at all times. If you excite it at that freq too much the brittle material cant take the size of the oscillation and shatters. So material flexibility and density will play a role too.

I believe the Nazi's where working on a sound cannon that did something similar in WW2. Like some kind of big directional sub woffer. Maybe you can find some details from that?


Good luck, and keep us posted.

 

[wingnut]

Thanks for the reply Mongrel Shark

I spent much of yesterday Googling resonance, including the Tacoma Rapids bridge - which apparently is a bad example of resonance, but a good example of "flutter" - an aeronautical term - I don't know the difference between these yet.

I watched the mythbusters clip where they get a singer to break a glass with unamplified voice - apparently the first time it was ever filmed. Adam was able to break a glass with amplification of his voice.

When they tried to vibrate a bridge they used a linear motor. By sheer luck, I think I have found something better which works well with high frequencies, but will not work on a bridge. Its that old elephant in the room, viz. an unbalanced motor. In my case I used a toothbrush motor with an offset. If you mount this on a springy base of any sort, by experiment I found something wonderful (to me ... who still gets excited by science).

As you increase the F of the motor, it will hit the same F as the natural, resonant F of the spring. When it does so, the motor will go beserk and want to throw itself off its mounting. This has to be a problem which engineers face regularly.

What fascinated me is that the motor has no problem finding and staying in this resonant state. I thought it would take fancy electronics to get it to do so. On the contrary - using pwm just put it anywhere near that F and it takes itself into the resonant F of the spring, and stays there even while one changes the pulse width within quite a band.

So what I thought would be the biggest challenge, actually solved itself. Something rotating (with an offset) will hunt, and find and stay in sync with another object reciprocating/oscillating (like a bridge). If the object is not damped, the added impetus on each stroke may add up to shake the object apart.

As you mentioned, crystal is probably easier to destroy by this means, but even skyscrapers are subject to this shaking and so in earthquake zones they take this into account. Tesla is rumoured to have nearly shaken whole buildings apart - and I believe this possible. My wife is not keen on me trying this on our home. But I have noticed at times that a truck idling outside my door can cause all the windows to vibrate badly.

 

[CDRIVE]

Will a tuning circuit which is CLOSE to (but not exactly the same as) a broadcast F tune itself into that broadcast signal?

No, but the circuit "Q" will be the dominant factor of how well the resonant circuit rejects signals <> Fr. I'm not talking about an oscillator here because they can be pulled by a close proximity signal.

Chris

 

[wingnut]

No, but the circuit "Q" will be the dominant factor of how well the resonant circuit rejects signals <> Fr. I'm not talking about an oscillator here because they can be pulled by a close proximity signal.

Chris

Are you saying that the oscillator will synchronize with the signal Chris?

If I set up two 555's with close but not identical F's on the same breadboard, will they eventually synchronise do you think?

 

[CDRIVE]

Are you saying that the oscillator will synchronize with the signal Chris?

No, I'm not saying that. The active components of an RF LC oscillator can have their normal operating points pulled from a strong external signal.

Chris