@ventilator: Your perpetuum mobile looks nice, but it would not work.

Even if we assume that the valves are perfect (which is not possible without additional energy), it will not work.

Because it only works when the left tube is completely filled with water. But in a completely filled tube there is not enough space for the ball. When the ball rolls into the tube filled with water, the water cannot expand because of the closed valves. So the pressure of the water would be too high. The ball would bounce off.

PERFECT valves that ...
1> doesnt let the water escape into the air side.
1> doesnt let the air escape into the water side.
2> dont required any energy to open and close
3> dont restrict the speed of the 'ball' in passing


Side question::
Does anyoneone know if permanent magnets (eg boron) are
reduced in their strength by being used...
eg: two equal boron magnets HELD north to north for an
extended period, and then are released.
Is thier strength (gauss rating?) still as strong as they were
in the beginning?
Has their 'strength' been diminished by being distorted for so long...

Yes, permanent magnets (ferromagnetic materials) lose their strength. But they can be magnetized again by using a strong magnetic field. They even can be demagnetized by using a certain magnetic field.

on the upper side there wouldn't have to be a valve at all. you could use something like a basketball that jumps out of the water. hm... but the ball would always take some water with it too.

Thanks Harry, Ive always wondered about that.

So any perpetual-motion machine using permanent magnets
as a propulsion system would only ever be 'virtually' perpetual...


Now back to topic...
But what about if we re-design Ventilators doo-dad a bit...
First, throw away the valves altogether.
Then we fill the air side with another fluid.
One that CANT mix with water, for-instance alcohol.
Then find a new ball with a density that allows it to
still float in water, bit sinks in the alcohol. (or vise-versa)
Then we extend the vertical tubes until the ball gains
enough momentum during both the rise and fall times to be
able to "pierce" the surface tension between the two fluids...

That way, we no longer worry about the valve energy requirements,
and if any fluid gets 'dragged' into the wrong tube it will
then either rise or fall as a bubble (like a lava-lamp),
and re-merge with its own type.

[EDIT]After further thought, I think we still need the top valve, just to keep
the 'up' tube fluids 'mostly' separate from the d'down' tube.
But it no longer needs to be perfect, as long as we get the
ball moving fast enought to push it open, its allowed to drag
a few drops through with it...

@EvilSOB
One thing is wrong in your idea
alcohol ( if we talk about ethanol the one you are drinking ) mixes with water.
alcohols in general doesn't mix with water with a chain lenght longer than 4 ( if i remember right so pentanol wouldn't mix with water.
A fluid which won't mix with water could belong to the group of Alkane ( sorry doesn't know the english term must be similiar )for example octan.

In general i believe a perpeetuum mobile is possible if it doesn't power something else or looses energy.

muffel

two different fluids that don't mix wouldn't stay next to each other though. one would rise and one would sink to the bottom.

OK, how about if I drop the alcohol and replace it with an oil?

If I remember correctly, oil and water will try to stay separate,
and oil is lighter than water...
Only the higher surface tension may be an issue....

What about water and petrol....??

Anyway, water is not strictly necessary. The concept only calls for two
non-mixable fluids with significantly differing density.

Ventilator: check my final note on my above post. The valve
at the top should stop what you say...

Can also be demagged by being exposed to heat higher than the Curie point.
But I beleive you would have to force 2 like-poles closer to each other, past the point they naturally repel each other to make them lose strength - just a guess...

PS> *Anything* is possible, just some things aren't likely - not on this planet anyway.
I'm sure that the laws of thermodynamics absolutely don't apply *somewhere* in our galaxy/universe.
Just so the Universe gets the last laugh at mankind...

Since water is considered the universal solvent, I wouldn't use water at all.
Water even mixes with oil - if you create an emulsion/emulsifier.
If I remember, I think egg whites do that - as in mayonnaise.

Also, I think liquids have to high a coefficient of friction for PM machines.
Perhaps gasses/plasmas?