Time

[Sato]

Hello, me and a friend are participating with our school's science club as a team and will be taking on an event that focuses on the topic of time. It will involve an exam covering topics pertaining to time and the presentation of an accurate homemade timekeeping device. Here is the event description and here is a sample exam.

 

Note that there are certain restrictions/regulations on the device which are outlined in the event description. Namely,

• The timekeeping device may not use electricity or chemical reactions.
• The device cannot exceed 80cm.

 

Any posts sharing pertinent knowledge/understanding would be much appreciated, that is, for the exam, and the homemade timekeeping device. We have a few months but are starting early.

 

Also, my local library has a 3D printer, so I was thinking maybe it could be made useful for building the device?

 

Thanks,

Sato

Edited September 18, 2014 by Sato

[fiveworlds]

five-minute-hourglass

Edited September 18, 2014 by fiveworlds

[Endy0816]

Gravitational potential energy(pendulum, hourglasses, etc), spring force, pressure, kinetic energy(flywheel) and temperature are all options.

 

If you have not already, you may want to try and build a virtual clock first to get a sense of what is going on. If one of you has Minecraft, it makes for a decent simulator(with schematics readily available online). There is also www.logic.ly as an alternative option.

 

Sounds like a fun project, good luck with it.

Edited September 19, 2014 by Endy0816

[imatfaal]

I would go heavy on the maths and try and keep the engineering to a minimum. A timepiece that measures to a tenth of a second is brutally difficult to make, to start/stop, and to count.

 

Off the top of my head I would set up an array of pendula - as many as you can all different lengths. you need to be able to start in unison and count each pass. If the lengths vary and the pendula are not in sync or harmonic then every tenth of a second will have a unique set of counts.

 

Setting off at same time is pretty easy to make, counting is not so easy but doable, use a single beam with as many stiff rod pendula as you can manage hanging off with good bearings (they need to swing for 350 seconds). In your three ring binder you have pages of counts from 10.0 to 300.0 sec. A quick excel mashup allowed me to get unique counts for all but two time intervals.

[swansont]

I would go heavy on the maths and try and keep the engineering to a minimum. A timepiece that measures to a tenth of a second is brutally difficult to make, to start/stop, and to count.

 

Off the top of my head I would set up an array of pendula - as many as you can all different lengths. you need to be able to start in unison and count each pass. If the lengths vary and the pendula are not in sync or harmonic then every tenth of a second will have a unique set of counts.

 

Setting off at same time is pretty easy to make, counting is not so easy but doable, use a single beam with as many stiff rod pendula as you can manage hanging off with good bearings (they need to swing for 350 seconds). In your three ring binder you have pages of counts from 10.0 to 300.0 sec. A quick excel mashup allowed me to get unique counts for all but two time intervals.

 

This may be difficult, though, because coupled pendula affect each other. I'm not sure what the effect on precision will be.

[imatfaal]

This may be difficult, though, because coupled pendula affect each other. I'm not sure what the effect on precision will be.

 

You are probably correct that the movement of the bar and other factors might couple them. Would it be repeatable? Probably not.

 

Hmm - completely separate fulcrums (fulcra?) would be better but much harder to engineer.

[swansont]

Here's what can happen with coupled oscillators

 

http://www.youtube.com/watch?v=kqFc4wriBvE

[studiot]

 

A timepiece that measures to a tenth of a second is brutally difficult to make, to start/stop, and to count.

 

 

The understatement of the year.

 

I cannot think of a way to guarantee this, within the constraints mentioned (notably size).

 

The simplest way I can think of is based on a model train set.

If you have a runner that runs around a large enough track (it would have to be mechanically powered since you are not allowed to use electricity) the track could be marked off to any desired accuracy and the measure time being the number of laps plus the graduated partial lap time.

However a track large enough to mark out in 0.1 seconds would be way larger than the 800 mm allowed. It may be worth doing some trials to see however.

Some figures:

If it takes the runner 10 seconds to complete a circle that's 360 degrees ie 36 degrees in 1 second ie 3.6 degrees in 0.1 seconds.

How big a protractor do you need to measure this?

Of course the next stage is to turn the runner and track into a fixed mark and have a spinning protractor.

 

 

I would advise against water clocks, they are notoriously difficult to regulate. The ancient Greeks had an hour of variable length because of this, there is an interesting story in my history of engineering in the ancient world about this.

Edited September 23, 2014 by studiot

[BJ LinOPo]

Thanks to everyone who responded,

sorry for responding so late; we've (me and Sato) just begun preparing for the event.

 

I would go heavy on the maths and try and keep the engineering to a minimum. A timepiece that measures to a tenth of a second is brutally difficult to make, to start/stop, and to count.

 

Off the top of my head I would set up an array of pendula - as many as you can all different lengths. you need to be able to start in unison and count each pass. If the lengths vary and the pendula are not in sync or harmonic then every tenth of a second will have a unique set of counts.

 

Setting off at same time is pretty easy to make, counting is not so easy but doable, use a single beam with as many stiff rod pendula as you can manage hanging off with good bearings (they need to swing for 350 seconds). In your three ring binder you have pages of counts from 10.0 to 300.0 sec. A quick excel mashup allowed me to get unique counts for all but two time intervals.

 

Thank you for the response, this sounds like a good / viable idea!

 

I have some questions:

1. How would we record each tick of the pendulum? Can you please explain your last remarks regarding the storing of counts in the three ring binder/excel?

2. How would we decide where to position each pendulum?

 

Also, do you know of any resources that go more in depth about building the pendulum clock?

 

 

 

 

 

The understatement of the year.

 

I cannot think of a way to guarantee this, within the constraints mentioned (notably size).

 

The simplest way I can think of is based on a model train set.

If you have a runner that runs around a large enough track (it would have to be mechanically powered since you are not allowed to use electricity) the track could be marked off to any desired accuracy and the measure time being the number of laps plus the graduated partial lap time.

However a track large enough to mark out in 0.1 seconds would be way larger than the 800 mm allowed. It may be worth doing some trials to see however.

Some figures:

If it takes the runner 10 seconds to complete a circle that's 360 degrees ie 36 degrees in 1 second ie 3.6 degrees in 0.1 seconds.

How big a protractor do you need to measure this?

Of course the next stage is to turn the runner and track into a fixed mark and have a spinning protractor.

 

 

I would advise against water clocks, they are notoriously difficult to regulate. The ancient Greeks had an hour of variable length because of this, there is an interesting story in my history of engineering in the ancient world about this.

 

This sounds interesting, but how would we power the runner/track?

 

Any ideas for how such a set up would be built? With metal pieces, plastic, etc? Do you think that it would be a good idea to use our library's 3d printer to make this track, modeling it in a CAD?

 

Could you elaborate on the set up with the track/runner as a fixed mark and the spinning protractor? Why would the runner be fixed?

 

And in any of these cases, how would we be able to record the number of runs/passes?

 

It would also be helpful if you could provide some material like tutorials/papers discussing the construction of such a clock.

 

Thank you for the help!

 

Here's what can happen with coupled oscillators

 

http://www.youtube.com/watch?v=kqFc4wriBvE

 

Why does this happen? How/why this happens and how we can prevent/predict it?

[swansont]

 

 

Why does this happen? How/why this happens and how we can prevent/predict it?

 

The metronomes couple because the board vibrates along with them. Each metronome is a forced oscillating system rather than a freely oscillating system. You need to make sure each oscillator is decoupled