Hello Kai

"that a DC-motor always needs a much higher voltage to reliably start, than for keeping running"

Of course you must use the motor at the voltage it was designed for, but if you do it will start just fine without the need for any "much higher" voltages.

It's a little bit different, and it has more to do with load resistance. The voltage is more the consequence than the cause in this case. The problem with putting a constant, ohmic resistor in series with a DC motor is that the motor will exhibit different resistances at different loads (as Steve said). The motor standing still is the same as it being loaded to the extreme. The rotor has no momentum yet, a certain amount of energy must logically be put into the motor to achieve that. At this point it will exhibit an extremely low ohmic resistance, so if you look at the series resistor and the motor as a voltage divider, after power-up the current through the resistor and the motor are the same, but because the resistance of the motor is so low (the cause), most of the supply voltage will drop across the resistor, leaving almost no voltage across the motor (the consequence). And P=U^2*I, so you can imagine what happens to the power available to the motor when it sees the voltage drop that much. It might easily not be enough to overcome friction, which is what probably happened with your fan (you will notice that if you give the fan a little push, it will spin happily). Also, the resistor will get to dissipate a huge amount of power, because it is almost immedaitely across the terminals of the power supply. The starting problem can be overcome by putting a big capacitor in parallel with the ohmic resistor. The capacitor will act as a short circuit for a short time after power-up, so the motor sees the full supply voltage and uses the charge current of the capacitor to start. But the solution with the series resistor is still an energy-inefficient one, and leaves the motor very sensitive to load variations.

Some devices I've seen in my profession (big audio tape recorders) did use a "kickstart" voltage, higher than the voltage the motors were rated for, to get them going, but that is not done because else the motors wouldn't start, but in order to get them up to the desired speed quicker.
OK. Back to the 8052.