damped?

what about this? http://www.youtube.com/watch?v=GuTMYgQDUzs

http://www.youtube.com/watch?v=4foY5r2TMOo

In the Milkovic example, the pendulum is not damped by the load, and so it can be fed back. 

What i meant to say is that the ALTERED Milkovic example is not damped.  Yes both Tagor and Exnihiloest are correct - the standard Milkovic example has dampening because the pivot point drops as the pendulum swings downward.  This takes energy from the system, Exnihiloest your explanation is perfect.

The Milkovic's pendulum is really a parametric pendulum with a pumping function. Nevertheless in such a system, the energy is always conserved, it is just shifted from the pumping system at the pumping frequency to the pumped system at the pumped frequency (for example, search for "parametric amplifier".

The key here is when properly designed gravity does the pumping.  Right now because the pivot point moves, gravity is only doing some fraction of the pumping, the operator of the pendulum does the rest.  If the pivot point doesn't move, gravity will do 100% of the pumping.  This is a very interesting result. 

And yes energy is conserved because the output is the initial energy input plus the energy required to keep the pendulum swinging.  But if Bob puts in the initial energy, and Mike keeps it swinging, it will appear to Mike that the device is over unity (assuming the device is designed so that the pivot point doesn't move).  Mike will think this because he is oblivious to the initial energy input that Bob placed.  So to Mike, the output (Bob energy + Mike energy) is much more than Mike's input. 

The damping of the pendulum is the most important aspect.  If you can never remove the pendulum damping, I don't think it will ever be possible to feed the energy back.

It is overunity and it isn't.  That's what I'm trying to say.  Its all in how you define your reference point.  If, for example, you include the initial energy to raise the pendulum up then include the energy it takes to keep the pendulum swinging, the system is not overunity at all (maybe 50% or worse - depending on your setup).  However, if you ignore the initial input and measure the amount of energy it takes to keep the pendulum swinging verses the output.  Suddenly the system appears overunity (or atleast very close - again depending on your setup). 

It is almost like the initial input is "non-radiating" energy and stays within the system and is re-used (if that is a good way to describe it).  The "radiating" part is what is consumed by friction and the dropping pivot point. 

@Charlie_V, I would be most interested to hear how you imagine that the enhanced 2-stage pendulum would be constructed. In my mind it seem totally impossible to make work at the output of this device if it is not moving its pivot (the pendulum pivot). And as we know, moving its pivot will cause it to loose some of its stored energy, this is the only real loss of this machine, friction is generally only a minor nuisance.

 

I'm still working on it.

What I meant by real loss is that it is the only one that counts, the others are both minimal and are not something which a certain design is forced to have, but loss of potential energy will always happen when we move the pivot, as you explained quite well earlier.

So not moving the pivot at all isn't really an option. What I had in my mind would be something dual, which both used the up time of the pendulum (or spring, as they work as well) and the down time of a second pendulum/spring. If they could somehow work together at the same time, and make up for each others losses, as it is only distance and not force*distance (real work) which is draining their energy, then they could work like some of the systems which A. Frolov described in the paper I linked. There you can see both a transformer and a magnet motor, the first which needs no power but only a signal, and the other which has no drag what-so-ever. The key to both is that we use two continuously interchanging actions to create a third action. The third one will not directly act on the source as a load, but will only be one of its parameters in constant change.

Hmm, this does seem pretty easy to unite with T. Beardens talk of engineering the causes instead of the effects.

To me the original principle of the 2-stage oscillator is still enough to power all that we need, but there are many ways to use it. We shouldn't limit ourselves to only the pendulum/spring on a hinge model. There numerous other designs which can be based on the same principle, and many of which are much more powerful and at the same time simpler.

Some time ago I tried to make a 3d model of a full-scale, completely self-powered Milkovic pendulum.
To make it completely self-sustained it would need many parts and gears in order to send its output back to its input, and in addition to this complexity its efficiency would probably be very low too.

Then, one day while watching the energetic forum a guy named Ted Evert made a very interesting sketch of a machine that centrifugally rotated water and then used the increased pressure to further aid in its total rotation. He hasn't made it yet, but it is a very intriguing idea, and has to me a very clear relevance to the 2-stage oscillator.

You input rotation (pushing the pendulum, or pushing the cylindrical water tank).
It outputs linear acceleration (centrifugal pull from the pendulum, or centrifugal jet from the tank).
Here's there difference; no complex machinery is needed in the water jet scenario, you only need to point it in the right direction and thats that. Then you will have yourself a full-scale completely self-driven 2-stage oscillator.

As of yet this seems to be one of the best ways to do it.
Other potentially even more powerful ways to use this effect can be acoustic resonators. Since the frequency of oscillation is higher, more power in comparison to its weight should be possible, but I haven't yet found way to handle the feedback.

Julian

Without a moving pivot, the lever could not take the energy of the pendulum to provide it to the load.

I think there are two distinct energy inputs that make the output in Milkovic's system.  The first being the non-radiating (for lack of a better term) effect of centripetal force due to gravity (this portion is recoverable when the lever moves down).  The second is the radiating portion which encompasses the frictional loss but also the fact the pivot point moves, damping the pendulum's oscillation.  Between the two, I'm not sure which is the more dominant - I suppose it varies for different setups.  I know the first will become more dominant the less the pivot point moves.  Of course, if it doesn't move at all no useful work is produced.  Centripetal force is what causes the lever to move in the first place.  But the moving pivot re-couples the input with the output so I wonder if feed back would ever be possible?

You know, everyone tries to load the lever arm.  I bet it would be easier to put a ratchet on the lever's fulcrum and attach a load to the ratchet itself.  Then you could gear the ratchet to get more speed and connect a small electric generator with capacitor storage bank.

Coils could be placed on either side of the pendulum arm with a Reed switch and a magnet on the end of the pendulum.  The generator could charge the capacitor bank, the capacitor bank could give power to the coils to keep the pendulum swinging.  A diode would need to be in the coil circuit though because you want current to flow only one direction (to attract the pendulum upward to its initial starting position but not to flow as the pendulum swings back downward).  Something like that should give auto timing to the pendulum.  You would find out really quick if the system can function in feedback or not.