Both the lifting force generated by the spinning cylinder and the drag force on the stationary cylinder push the wheel against the downwind side of the blade guide. For this reason, the cylinder will always spin with the correct polarity for propelling the cylinder along the blade guide, regardless of whether it traverses the track in the clockwise or counterclockwise direction. It would seem like this kind of turbine might be pretty good at starting itself. But how is it prevented from starting up in the wrong direction? Maybe the wheels could be ratcheted so that they can turn in one direction without the cylinder spinning. In this case, it would seem that the blade would eventually experience a perturbation that would tend to start it spinning with the power-producing polarity, and if so then this machine is self starting. Rotating tires may be used for reversing the spin polarity as the cylinder rounds one end of the track:
Writing this post has clarified one aspect of the Magnus Effect Rotor – if the rotor diameter is large then the time it takes to reverse the spin polarity of the cylinders may perhaps be insignificant.
Although the aerial view above shows an oblong blade guide (oblong track), a single blade guide could be used as well. A single track is possible if the turbine has only one blade (one spinning cylinder) and in this case the cylinder would stop when it gets to one end of the guide, and then travel in the opposite direction back to the other end of the blade guide. If it is not desirable to yaw the blade guide, then a giant circular guide may be used instead. Note that regardless of which blade guide configuration is selected, the cylinder will always spin with the correct polarity, regardless of whether it travels around the track in the clockwise or counterclockwise direction.
Many posts on the Salient White Elephant (such as this one or this one) describe ideas for generating electricity with turbine configurations that are similar to the Magnus Effect VAWT, so I won’t duplicate the descriptions of those ideas here.
A Magnus Effect Rotor that is similar to an H-Rotor may also be designed using the ideas presented here. In the case of an H-Rotor, the spin polarity for the spinning cylinders can be reversed using a mechanism similar to that described above. That is, the cylinders always push downwind, and this causes them to engage one spinning mechanism during the upwind arc of the rotor, and the other spinning mechanism during the downwind arc of the rotor. However, in the case of the H-Rotor, it isn’t necessary to synchronize or gear the cylinders to the rotation of the rotor itself, as their aerodynamic forces are sufficient for generating the rotor moment. In this case the spinning cylinders can be controlled by an electric motor. This makes it easy to reverse the spin polarity slowly and at the proper rotational angle of the rotor. In this application, the main lesson we learn from the Magnus VAWT described here is that the rotor diameter should be very large so that the time taken to reverse the spin polarity is insignificant compared to the time it takes the rotor to complete one revolution.
Since no utility scale wind turbines that use the Magnus Effect have been built (not to my knowledge anyway), I have to assume there’s some big disadvantage to this approach. But before closing, let me point out one big advantage of this type of rotor – cylinders are a lot stiffer than blades, especially given weight. From this perspective, I wonder why a Magnus Effect rotor similar to the turbine below wouldn’t be effective?