Building a Clock with Periodic Motion

[IMDWalrus]

I need help with my homework; I figured that the engineering geniuses at CD would be good people to ask for help.

For one of my classes, I've been assigned to build a clock. To be honest, I have no idea how to do this...so I'm asking for any assistance that you can offer me.

The clock has to be able to keep track of time fairly accurately - within one second for a 30 second trial, and within two for a two minute trial. If I can get the clock to accurately measure three minutes as well, I get extra credit.

Problem is, there are a few catches...

1.) The clock can't make any use of electrical components
2.) The clock has to involve some form of periodic motion

Once the clock has been set into motion, I'm only allowed to touch it twice. The thirty second and two minute trials don't need to be measured during the same run; I can do thirty seconds first, reset, and do two minutes.

I'd like the clock to be unique; I've considered using one of those dipping birds but have no idea how to tie that into a timekeeping mechanism.

Any suggestions? The project is due on Friday, and I want to have some time to test it. Any help at all would be appreciated...

[edit]Slight clarification[/edit]

[KenWittlief]

what do you mean it cant have any mechanical components?!

any physical object is a mechanical component

can you be more specific about the rules and limitations?

[Tom Bottiglieri]

try a sundial that straps on to your wrist

yeah.. i wasnt being serious

[IMDWalrus]

Originally Posted by KenWittlief

what do you mean it cant have any mechanical components?!

any physical object is a mechanical component

can you be more specific about the rules and limitations?

Wow...it's too early for thinking. That didn't make much sense; I'll give it another go here.

"Machines are to be constructed from cardboard, wood, plastic, rubber, springs, etc. Electric motors are not allowed."

For some reason, I've been thinking that meant all mechanical components are not allowed...which is not only not what I've been told but also makes zero sense.

[KenWittlief]

wow you could have some fun with this!

water - sand - windup things - stuff that flys

how about this - get a bowlingball, some string, and a candle.....

this is one of the drawbacks of being on a FIRST team - your teacher is going to expect something incredible from you :^)

[Swan217]

Try water dripping into a graduated cylinder - water drips at a fairly constant rate, as you'd know if you've ever had a leaking faucet.

[Denman]

at first i was going to suggest using my incredably accurate stopwatch i did for some coursework
ok, why dont you try working out the time for a pendulum and just use it and count the swings, you could easily work out one second iirc...

[KenWittlief]

if im reading the problem correctly, he has to make a clock with a 30 second period - something you trigger and it runs for 30 seconds, then some end condition happens

making a pendulum that takes 30 seconds to swing once would be, ummmm.. big!

[Duke 13370]

The way a normal grandfather clock works would work for the task, but it takes time to create and test. Basically, there would be a large pendulum that swings back and forth (probably once every other second). when it swings in one direction, it trips a lever, advancing some sort of counting mechanism (like spinning an arm X degrees or pushing a beam X cm).

or, you could just build a pendulum and count the swings yourself if it's allowed.

[Eric Bareiss]

Does the clock have to show time or can you just count the number of times the pendulum swings.

[Biff]

Funny thing as I remember the rate that a 1 meter pendulum swings is once per second, you then need a escape mech to drive it and to count and your home.

[IMDWalrus]

Originally Posted by Eric Bareiss

Does the clock have to show time or can you just count the number of times the pendulum swings.

It doesn't have to show time; counting pendulum swings will work.

[Eric Bareiss]

Biff is right, a one meter pendulum should swing with a one second period no matter how many times it swings. There is a simple equation to find the period of a pendulum.

I don't know it off the top of my head, but I can imagine that you should because it is your class.

First step is going using the equation to be find a pendulum that swings exactly one second. If you build it right it should easily be accurate for 2 minutes.

[Rafi A]

i beleive it is:
period=2*pi*square root of (length/acceleration due to gravity)

[Elgin Clock]

Did someone say CLOCK??