@i_ron: simply put your output hammer in the bucket of water and you will get water (fluid resistance) as the load

yes if he pump water = it is real work

It is real work non the less, the fluid will make friction which the machine must work against.
It is just not useful work but it will prove the point.

what about hammering? he is lifting up the weight + doing the work by hammering
http://www.youtube.com/watch?v=gC6Qlj1Mbo8

I like much of what you say, but I cannot believe you are buying into the inertial propulsion cart.  Can't you see that Milkovic is simply using friction to retard the gentler reaction force?  The forward thrust is quick, while the reverse is slow, hence the forward motion.

In space, with no friction, such a method would be impossible, so your proposed method for the astronaut to propel himself by moving weights back and forth has no chance of working.

there is still friction in space. just not much around the craft being moved. the moving parts of the device will still incur friction between the metals.
 the real problem with that type of propulsion in space is you have no inertial reference, it would be like triyng to move the space shuttle, by having the astronauts push on an inner wall of the craft.
the other is the lack of gravity which makes the pendulum operate....

According to pendulum physics, the force at angle=0 from verticle increases with mass and decreases with armerature length.  therefore, maximum force-gain would be achieved with a heavy weight on a short arm.
the other half of the force equation is the leverage-factor of the second fulcrum. ( big lever in top)
The key thing to note here is that this [max] force is short lived, with very little momentum. Great for pumping water or hammering something into oblivion - but not so great for things like steady, continious motion, or generating electricity.

For more momentum (which translates into torque), a heavier weight on a longer armerature might be desirable. This provides for a longer "stroke time", and thus more monentum from the weight is translated in the verticle direction. what im calling the stroke time here is the tiny pause at bottom-dead center, when the horizontal force switches direction, during which the verticle force maxes out and actuates the upper lever. This might be better for operating a generator or driving a  wind-up pendulum-train.


p.s. - if you guys want known physics documentation on this set-up, look at the specs for Milkovich's water pumps. it should have everything there you need to know. i.e. weight of pendulum, lift distance,  how much water it pumps from varying well depths. ect...

Could someone please post a correct calculation of energy from an experiment with one of these systems?  I tried reading the pdf someone posted but it is hard to follow since it doesn't detail the experiment very well, I'm not sure what "energy for third 32 amplitudes" means?  Is he taking an average of 32 runs at the third swing? 

This device is what is called a "parametric oscillator".  Using Milkovic's device, the lever mass oscillates at twice the frequency of the pendulum.  By definition, when this happens in a parametric system, energy is absorbed at a rate proportional to what the system already has.  So, although letting the pendulum swing down will give you the correct value of energy, it defeats the purpose of the device. 

@TinselKoala

I don't think a double pendulum describes this device very well.

The design is more closely related to this I do believe:
http://www.physics.uoguelph.ca/applets/Intro_physics/kisalev/java/pend2/index.html

Apparently in the link you gave, one page over is a spring pendulum.  This is the simplest parametric oscillator.  Milkovic's device functions similar to when you let the spring pendulum fall in such a way that it makes a U (not chaotically like when you first see the demo). 

I think gravity does the parametric pumping, all the operator must do is keep the pendulum swinging.  This device has two regimes, without a load on the lever it acts as a parametric pendulum, but if the lever is overloaded (so that it can't move), the system operates as a standard pendulum.  Someone has to apply an initial energy but then it takes every little to keep it going.  Yet because of gravity, the lever tends to continue to "reuse" the initial input. 

I'm still trying to learn more about parametric oscillators.  A child on a swing is one example, the other examples I've seen are using a rope to lengthen a pendulum's rod as it swings down then shorten the rod as it goes up.  Milkovic applies the technique backwards.  He uses centrifugal force of the pendulum's kinetic energy to lift the lever up - which i think is much more clever since now all the operator must do is keep replacing the leakage energy in the pendulum (otherwise it "radiates" away due to pivot friction and wind resistance). 

Once the pendulum is in motion, if you could apply a stronger force to the lever, so that as the pendulum is falling downward you cause the pivot point to raise up, you can damp the pendulum's movement.  But just think what that force would have to be?  Plus, loads on the lever wouldn't do that anyway.  They will try to stop its movement, not retard it in the opposite direction.  Thus, the loaded lever should not be able to damp the pendulum.  An overloaded lever would cause the pendulum to move out of the parametric regime and function as a simple pendulum. 

So energy in the system may be very different from the point of view of where you stand.  I think so far BOTH analysis are correct.  If you look at only the energy required to keep the pendulum swinging verses the energy output, you will have overunity.  However, if you take into account the INITIAL energy it took to place the pendulum at a certain amplitude PLUS the required leakage reducing energy, you should find the system operates below unity (because of losses).

But like I said earlier, letting the pendulum swing down defeats the purpose of this machine - but will give you the total energy of the system, which is conserved.  I think the idea is to input a large amount of energy into the pendulum, reduce the pivot resistances as much as possible, and then add a small amount of energy to keep it going for a long time.  I'm not sure what happens if you "close the loop".  Does it lock into a perpetual motion type device?  Do the losses of the system eventually get the better of it and cause it to stop?

Its an interesting idea to say the least.

Charlie

Ha i already changed my post, because maybe a double pendulum is a parametric oscillator as well?  I'm not sure haha.  Thanks for those websites, they are good!

just found on YouTube

Milkovic pendulum replication
http://www.youtube.com/watch?v=Ut6FVjKMrBo

Storing mechanical energy in either X, Y, Z transference directions even with leverage, it is still storing energy, the device just uses the stored energy more efficiently but it will not produce more energy than what is stored. that's why it stops.

it doesn't matter how elaborate you make a stored energy device, it by any other name is still a stored energy device.

Jerry