Originally posted by JeffersonRaley
Apparently the challenges come from magnetic interference from the massive solenoids required, and from the fact that the valves tend to destroy themselves because they "slam" open and closed. The ability to do this is one of the drivers behind the move toward higher voltage electical systems. I believe that BMW is working on it also.
Actually, Siemens has been researching electromechanical valves for some time, and while I am unfamiliar with the current state of their project, back in 2000 they had a software controlled solenoid system that allowed "soft landings" of the valve. (The system was capable of positional encoding, and could decelerate the valve as it reached the target position)
In a very different field, I am also familiar with extensive research into artificial heart design, in which power efficiency (heat dissipation and power supply capacity are both major constraints) was found to be optimal if the solenoid acted as quickly as possible. In other words, the shorter the duration of the positioning pulse, the less power required to facilitate the change of state. This is a little bit counter-intuitive, but when activated VERY quickly, it is actually possible to recapture some of the energy during the deceleration phase of the positioning; it acts like a generator, and the net power requirement is reduced.
I don't know how Siemens addressed the issue of holding the valve in a given position; there must be some kind of locking mechanism, because the efficiency of using the magnetic coil to hold the valve would be very, very low.
I have also heard of a system that BMW was experimenting with for Formula 1, in which the valves were pneumatically operated.