John Ecklin's SAG 6

The Stationary Armature Generator 6, designed by John Ecklin, is one among many of his inventions claimed to produce free energy. It is a fascinating, ingeneous, and simple concept which might or might not be over-unity, but is nonetheless worthwhile to investigate.
I first learned about this device when I was thirteen, have made several attempts at building and testing it, and learned a few things not mentioned in any of the source articles—which is why it is included here.
To understand the generator, some simple electromagnetism principles need to be reviewed.
Faraday’s Law states that the voltage between the outputs of a coil is proportional to the rate at which the coil moves through a certain amount of magnetic flux. Thus, the stronger the magnet, the higher the number of loops in the coil, or the faster the coil cuts through magnetic flux, the greater the voltage generated. This really states that a changing magnetic field is required for a coil to tap its magnetic energy in the form of electricity.
An example of this is a loop of wire rotating like a spinning coin between opposite poles of a horseshoe magnet. Electricity is siphoned off the outputs of the coil.
Lenz’s law states that any coil of wire will set up an opposite magnetic field to counteract any change in its externally applied field. So if you bring the north pole of a magnet toward a coil’s end, the electricity induced within the coil sets up its own north magnetic field which repels the magnet, causing you to put more energy into bringing them closer. Then, if you pull the magnet away, the coil’s end becomes south and pulls you back.
An example of Lenz’s law is the following:

Much energy is lost in pushing the coil toward and pulling it away from the magnet.
Ecklin’s SAG 6 attempts to circumvent Lenz’s law and produce electricity by combining two well known principles.
First, metal flying past a magnet loses no energy. The piece of metal velocity increases as it nears the magnet, and decreases as it leaves, but both in equal amounts. So from frame A to frame C, no energy is lost.

You can prove this to yourself by tying a piece of metal to a string, taping a button magnet to a table, and letting the metal swing back and forth above the magnet. It does so for a long time, and only slows down due to air friction and energy lost in the string’s vibration.
Second, the following setup changes the magnetic polarity of a coil without the need of a moving magnet OR moving coil:

Note that the only moving piece in this example is the shunt (piece of magnetic metal), and as illustrated above, shunts lose no energy when moving past a magnet. Still, from frame A to frame B, the magnetic field within the coil changes, so electricity is produced.
Thus, because the shunts lose no energy, but electrical energy is still produced, this set up is theoretically over-unity.
The SAG 6 is cleverly designed like this:

Here we have two opposite C-shaped magnets trying to influence the magnetic polarity of the I-shaped core. Without the shunts, no magnetic difference would exist between the top of the core and bottom. But since every quarter turn one of the gaps between core and magnet is bridged by a shunt, one magnet has greater influence over the core. By rotating the shaft, the magnetic field within the core flips back and forth as the strength of one magnet’s influence over the other switches back and forth.
Because only the shunts are rotating, no energy is lost, and yet the I-shaped core still carries an alternating magnetic field whose energy is tapped by the coil wrapped around the core. The magnets and core remain stationary, hence the name “stationary” armature generator. Zero energy in + lots of energy out = free energy.
My first several models of this device were pathetic, for I had no experience in using the tools in my basement, but my sixth model was sturdy enough to prove one thing: the SAG 6 does indeed generate electricity without motion of magnets or core.
It lit a couple LED’s, but since there was so much friction, with such large airgaps, testing for over-unity was out of the question. An oscilloscope reading measured ant-hill shaped sine waves of three volts peak to peak. Output current was in the milliampere range. Since input power was around 30 watts, the efficiency of my model was miniscule. But it did produce electricity, which was hopeful.
Materials employed were a race-car motor, blue radioshack C-magnets and shunts, a core cut from a small transformer, 300 turns of #22 wire, brass shaft and slip bearings, and a battery charger to run the motor. This was done using elementary household tools, so I was not expecting much.
I have talked to David Colishaw, and he had built a model of the SAG 6, but his was also characterized by shoddy efficiency.
Now, the largest factor contributing to loss of generator efficiency is wider than acceptable air gaps. There is much magnetic flux leakage, and since the intensity of the magnetic field falls off as the square of distance, any air gap whatsoever will contribute to weakening of the magnetic field.
In the SAG 6, the air gap factor is employed to alternatingly weaken one of the magnet’s influence over the core—but look carefully at how the shunts are designed. They are solid pieces of soft iron, which actually channel the magnetic flux from the north pole to the south, leaving little for the core.
Thus, a more accurate picture of what happens in the SAG 6 is as follows:

As you can see, the magnet with the stronger influence is the one with a larger airgap. The size of this air gap is around .7 cm, and if that is responsible for the strongest influence over the core, it is no wonder my or David’s models had terrible efficiency. A generator with air gaps larger than a quarter millimeter will undoubtedly fail at being practical.
Why did Ecklin’s design incorporate this fault? Was Ecklin naive? It is a common practice for inventors of suppressed devices to publish their findings with an artificially inserted flaw, such that hired scientists of the “thought police” build the device, see it as faulty, and abandon their persecution of the inventor. The wise reader, however, would see the flaw, make the correction, and enjoy his over-unity model.
The corrected Ecklin generator would appear like so:

Now, the shunts do not channel flux from north to south pole, but from the poles to the core. With careful construction, airgaps may be minimized, and friction reduced, leading to efficiencies hundreds to thousands times better than my model. I have not built this modified version yet, but will in due time.
From what has been said here, it would appear that the SAG6 is built upon sound theory. But a flaw in reasoning may still exist. Yes, a shunt will not lose energy when moving past a magnet…UNLESS the magnetic influence is diminished by an opposing magnetic field set up by the coil which cancels out the original field.

As the shunt moves toward and away from the core, the pull of the magnet may differ from frame A to frame C, such that extra energy is required to pull the shunt away from the magnet after the field is first weakened, then strengthened.
This may be the achilles’ heel of Ecklin’s reasoning. It can’t be said for certain until models are built. From various articles written about this device, actual over-unity models have been constructed which perform as theorized, so perhaps it truly is genuine.