Energy from where?
Sorry for the delay; here's the item I promised.
This is about a system to extract energy from a source not relating either directly or indirectly to chemical forces, fission, fusion, or tidal forces from the moon or sun.
I guess I didn't make it clear that this system could be built today (well, in a few months, if you had the money). It doesn't need any cutting-edge technology, let alone any black holes! There's even a patent. But I again want to stress that it is completely impractical.
The basic invention is this. Take an exceedingly large flywheel that can be spun up to display a very large angular momentum, the larger the better. From here on we'll call this flywheel a gyroscope because we want to utilize its gyroscopic qualities, not store energy in its intrinsic momentum. We will need some additional machinery to keep it spun up against losses, so mounting it in a sealed housing in a (near) vacuum would be a good idea. I think that sealed gyroscopes and/or flywheels are currently available for sale.
The simplest version is to have the gyroscope mounted with the bearings on the ends of its axis held fast in a circular ring which is itself mounted within a circular track allowing the interior ring including the gyroscope to rotate. While the gyroscope spins on its axis, the gyroscope's axis and the ring can only rotate within the plane of the track. The outer circular track is held fast, oriented parallel to the circle of latitude upon which the device sits. The track will stand at an apparent angle relative to the ground if the device is located anywhere but the Earth's equator.
After being spun up, the gyroscope will want to retain its current orientation relative to the "fixed stars" but the Earth will be turning beneath it. So from the point of view of a ground observer the gyroscope will make one compete turn within its track about once every 24 hours. This is true at any point on the surface of the Earth. The movement is slow but the torque generated will be relative to the angular momentum of the gyroscope, which we have set up to be very large. We oppose this movement with a gearing mechanism that eventually gears up to a speed acceptable to a generator.
Maybe one of you are aware of or can think up a better way to extract energy from this torque. A set of pulleys lifting heavy weights as in a grandfather clock, etc. Just like Steorn...
You'll remember from childhood experience with your toy gyroscopes, when a gyroscope feels a force attempting to twist it (as here), it will precess (move in a direction orthogonal to the twisting force, depending upon which way the gyroscope is spinning). IIRC, the existing U.S. patent on this device (the reference to which I have misplaced -- sorry!) describes a much more complicated mechanism than I do, so I hypothesize that it attempts to recover additional energy from these movements (as unlikely as that seems to me now). Either way, we do want to prevent the gyroscope from moving into an orientation where its spin axis is parallel to the Earth's spin axis -- no energy there.
In this version we instead simply prevent any precession, nutation, etc. that would move the gyroscope out of the plane of its track. We also ignore the other movements of the Earth, its revolution about the sun, the sun's revolution about the center of the galaxy, etc., that might minimally affect the gyroscope.
To sum up, this is a device which converts a tiny fraction of the Earth's angular momentum into a torque which we can then convert into useful energy such as electricity. Assuming the system has few losses from friction (mostly in bearing pressure) we should be able to obtain enough energy from it to keep the gyroscope itself spun-up indefinitely and still have enough left over to brew our afternoon tea.
Now you may be asking yourself, "Why isn't this instead converting the gyroscope's angular momentum into torque?"
The answer is that twisting a gyroscope does not affect its angular momentum except, again, from heat losses due to stress on its bearings (perhaps magnetic bearings are the way to go). But the torque from the gyroscope, through its bearings, track and energy gearing, eventually to the Earth's surface, are applied directly against the Earth's spin movement.
To reduce the required size of the gyroscope we might instead use multiple gyroscopes with the bearings of their axes held fast in a cylinder which is mounted in a cylindrical track. The gyroscopes don't have to be parallel to each other, just have their axes in parallel planes. (Mounting multiple gyroscopes in a disk invokes more complex forces.)
Finally, when the system needs maintenance you can use the spin-up motor instead as a generator to de-spin the gyroscope/flywheel, recovering a fair fraction of its energy.
Some of you engineers might enjoy working out the formulae to make this system more concrete, or conversely, show that it can't work. How much work can you get out of it per unit of angular momentum? What is the smallest model that could demonstrate the invention? Or perhaps you could sketch out methods to accept/direct the torque for energy conversion when the gyroscope is more like in the original patent where it is allowed to move in ways other than the one plane I describe. Or do a better patent search than I did.
That's it. Questions? Observations? Besides it being rather silly.
But I think it does show that you can get energy from the strangest places. Hmm... So maybe Steorn... No. No! That way lies madness.
-----
Later: I still haven't found the patent. Several patents by Frederick H. Mishler (5,353,655; 5,150,625; 3,726,146) from keywords might relate but it turns out they are for hand-operated gyroscopes with a mechanism that utilizes precession to increase spin -- which might be handy here. I've read about these toys: You start them with a pull string as usual, then you turn them back and forth in your hand until they reach about 10,000 rpm, making them almost impossible to handle. (Also see Wikipedia for DynaBee and NSDPowerball.)
You'll note Mishler's patents mention using the resulting energy for potential audio and visual output (which they do in the toy, seen in a video ad on the web). This shows the trend towards making every possible obvious claim in a patent. A recent Federal Circuit Court of Appeals ruling, just accepted for appeal by the U.S. Supreme Court, overturned the long-held (and written in law) standard of "obviousness" to instead require a previous publicly-available description of an invention to invalidate its patent due to obviousness. This includes combining existing things in patently

obvious ways. For info, google "patent obvious appeal".