Resolver with Rotor Output
An alternative form of resolver uses two ac voltages 90° out of phase, generated from a digital signal-generator board, to power the two coils of the stator.
The rotor is the secondary winding in this case. The phase shift of the induced voltage determines the angular position of the rotor.
An advantage of this arrangement is that it does not require slip rings and brushes to energize the windings, as needed in the previous arrangement where the rotor has the primary winding. But it will need some mechanism to pick off two stator coils are
u1=ua sin wt
u2=ua cos wt
When the rotor coil is oriented at angular position q with respect to the stator-coil pair 2, it will be at an angular position p/2-q from the stator-coil pair 1.
Hence the voltage induced by stator coil 1 in the rotor coil would be uasinwt sinq, and the voltage induced by the stator coil 2 in the rotor coil would be ua cos wt cos q. It follows that the total induced voltage in the rotor coil is given by
Ur=uA sin wt. sin q ua cos wt cos q
Urn=uA cos (wt.–q)
It is seen that the phase angle of the rotor output signal with respect to the stator excitation signals u1 and u2 will provide both magnitude and sign of the rotor position q.
The output signals of a resolver are nonlinear (trigonometric) functions of the angle of rotation. (Historically, resolvers were used to compute trigonometric functions or to “resolve” a vector into orthogonal components). In robot control applications, this is sometimes viewed as a blessing. For computed torque control of robotic manipulators, for example, trigonometric functions of the joint angles are needed in order to compute the required input signals (joint torques).
Consequently, when resolvers are used to measure joint angles in manipulators, there is an associated reduction in processing time because the trigonometric functions are available as direct measurements.
The primary advantages of the resolver include:
1. Fine resolution and high accuracy
2. Low output impedance (high signal levels)
3. Small size (e.g., 10 mm diameter)
Its main limitations are:
1. Nonlinear output signals (an advantage in some applications where trigonometric functions of the rotations are needed)
2. Bandwidth limited by supply frequency
3. Slip rings and brushes would be needed if complete and multiple rotations have to be measured (which adds mechanical loading and also creates component wear, oxidation, and thermal and noise problems)
