I'm Looking to Make a Very Small "Nearly Lossless" Machine

[BillIsTheDill]

I know perpetual motion is by no means possible, but I want to make something self powered, in the same sense.
I don't have too many resources, but I am looking to make something whose own excitation would power the machine, for say, an hour.
I.e. pushing a button that would cause something to vibrate, and power generated from the vibration would power the vibrations to go for an hour. The button then could be pressed at the end of the hour to continue the vibrations.
Is such an idea/similar ideas feasible? I'd imagine something like this would be possible atleast on a small scale.

 

[cpsinkule]

there's no such thing. if you don't utilize a fuel source for chemical energy or use nuclear energy then the only energy the system would have would be the initial excitation.

 

[BillIsTheDill]

What about piezoelectronics? I've read quite a bit on how far it has come, and how some sensor systems are self powered.

 

[cpsinkule]

they aren't really self powered. the current they produce is powered by the pressure you exert on the crystal, so whoever is doing the pushing is actually doing the work.

 

[BillIsTheDill]

I understand that, but I was reading this the other day.
http://www.rh.gatech.edu/news/55147/improved-nanogenerators-power-sensors-based-nanowires
And from what I understand, when the nanogenerator is excited by mechanical energy, the sensor can operate.
Another way of saying what I'm asking is what if instead of being a sensor, it was something that could produce mechanical energy (i.e. vibrations).
Since perpetual motion isn't possible, what if it just lasted as long as possible, as I was saying, like an hour or so?

 

[cpsinkule]

again, even if there was no dissipation and the system was 100% efficient then the only energy you have available to you is what you started with. if you try to convert electrical energy into mechanical energy then you will have dissipation and the system will die out. your question depends on how efficient your device is. whatever device it may be, the motion will cease when all the energy you started with is lost to dissipation and friction.

 

[cpsinkule]

i should also point out that the term "generator" is a misnomer. all a generator does is convert one form of energy into another, you actually LOSE energy in this process.

 

[A.T.]

power generated from the vibration would power the vibrations

Converting energy back and forth just creates losses.

 

[Orodruin]

Grandfather (longcase) clocks work essentially like this. You draw up a weight and the clock works for quite some time (much longer than an hour or there would not be much sense in it) based on the dissipation of the potential energy of the weight. Once it reaches the bottom, you need to lift the weight again.

 

[sophiecentaur]

If you want to make a machine that will run for a long time then 'small' is not the direction to be going, with the technology that's available in your average home. The Grandfather Clock mechanism will run for a long time with relatively little energy input and it came along way before the equivalent small timepiece . If you want something that will run for a long time (days, possibly) then a free pendulum with a mass of several kg, on a wire, with a gimbal support with good knife edges may be your best bet. Such a system has a lot of energy and loses a very tiny fraction of that energy each cycle. It could do many thousands of cycles before finally stopping. A small system would need to be in a vacuum to avoid the effects of air viscosity (which is like treacle for small mechanisms).

 

[jbriggs444]

https://en.wikipedia.org/wiki/Automatic_watch

 

[russ_watters]

https://en.wikipedia.org/wiki/Newton's_cradle

 

[Paul Uszak]

There are some interesting rotary machines on YouTube. They are essentially cylinders held on neodymium magnets and spin for a long time. A large mass might spin for quite a while. I think that there's some law that says that no entirely passive magnetic system can be stable, so they have a hard single bearing /point of contact. If you try to reduce the bearing friction through a judicious choice of materials, all that'll stop the rotation is minimal bearing and air frictions. If the mass was several kilograms and shielded from extraneous air currents, it might just run for an hour. (Caveat: I can't swear to it, but I've always wondered if there might also be a drag induced from two magnetic fields sliding past each other )

 

[CWatters]

Battery -> LED -> solar cell -> another battery

Could easily run for a month or more depending on size of original batteries, their initial charge state and how much power you put through the LED.

Edit:..

For example a 1000mAH battery will deliver a 1mA LED current for about 41 days. Start with both batteries full and you could switch the batteries around after say one month and run it for another month :-)

Wouldn't fool anyone with a brain.