What exactly is the problem with these devices?

[Koen1]

Ever heard of the Homopolar Generator? (aka homopolar dynamo, aka n-machine)
That works just fine as a motor on pure DC input. No AC at all, no switching of magnetic field polarity either.

[Honk]

http://en.wikipedia.org/wiki/Homopolar_generator

[Nostradamus2]

Simply put:
Magnets which is not changing magnetism on demand, cannot induce rotation or work. An electro motor is working because of the change of magnetism. All electromotors, even those running on DC-power supply are in fact AC-motors inside because they allways are able to change magnetic polarity as they rotate. That's why they work, and permanent magnet motors does not work.
If you even add loads of energy into DC electro magnets, as long as the magnetism does not change, they will just consume energy without doing any work. So how to make a permanent magnet motor to work, is to me a big puzzle.

But never give up. It's very educating to learn about magnets and magnetism, even if you should fail in your projects.

Br.

Vidar

You can switch iron or any other ferro-magnetic material On / Off with permanent magnet like current in electromagnet, so the right question is how to get more energy than put in ?

[Koen1]

No you cannot "switch ferromagnetic material on" using a permanent magnet!
There is nothing to "switch on".
Ferromagnetic material conducts magnetic flux better than other material, which is why the permanent magnet sticks to it.
When you remove the magnet, the magnetism in the frromagnetic material drops away.
To remove the magnet from the ferromagnetic material needs energy.

The only thing I can think of that might come close to what you are suggesting and that might work,
is a so-called adiabatic energy pump;
you could take a ferro bar, soft iron would be fine, and wrap a coil around it. Now have some current run through the coil and the bar will be magnetised.
Take the cirrent off the coil, and the bar will lose its magnetism. If properly done and tuned exactly right, you might be able to intercept the drop back to 'natural' unmagnetised state of the magnetic domains inside the ferro as tiny e/m fluctuations, then aligns them all using a weak field again, then have them drop back to normal unaligned domains at natural 'polarisation', etc etc, and keep intercepting the weak fluctuations...
In theory and properly done (I realise my description of it isn't very good ) you should be able to absorb a small temperature gradient from the environment this way, and with some luck that might even be more than the energy you lose in resistance etc.

[Nostradamus2]

No you cannot "switch ferromagnetic material on" using a permanent magnet!
There is nothing to "switch on".
Ferromagnetic material conducts magnetic flux better than other material, which is why the permanent magnet sticks to it.
When you remove the magnet, the magnetism in the frromagnetic material drops away.

Go back to school, learn physics ! Have you heard about BH curve ?