Abstract: A method of determining the angular position of the rotor of a brushless direct current motor having a stator with multi-phase windings, control circuitry that provides excitations switched after one another to the motor phases such that the excitations produce stator flux vectors lying in different directions in the full 360 electrical degree range, and a measurement circuit that delivers responses for the phase excitations is provided. The position is estimated by selecting the main excitation stator flux vector which results in the minimal inductivity shown by the measured response, and selecting the two other vectors in the previous and subsequent directions with respect to the direction of the main vector, and by computing the ratio between differences of measured responses for the selected stator flux vectors.
Claim: What is claimed is:

1. A method of determining an angular position of a rotor of a brushless direct current motor having a stator with multi-phase windings, control circuitry providingexcitations switched one after another to motor phases such that the excitations produce stator flux vectors lying in different directions in a full 360 electrical degree range, and a measurement circuit that delivers responses for the phase excitations,the method comprising the acts of: selecting a main excitation stator flux vector resulting in a minimal inductivity shown by a measured response; selecting two other stator flux vectors, one of which is in a previous and another of which is in asubsequent direction with respect to a direction of the main excitation stator flux vector; and computing a ratio between differences of measured responses for the selected stator flux vectors in order to estimate an angular position of the rotor of thebrushless direct current motor, wherein the selection of the main excitation stator flux vector is done by selecting a vector that corresponds to a maximal value among {i.sub.0, . . . i.sub.2p-1}, where if more than one equivalent maximal value isfound, then any of their indices are selectable for the index of the main excitation stator flux vector; the ratio between differences of measured responses for the selected stator flux vectors being computed as a fraction, the numerator of which isi.sub.n minus i.sub..lamda., and the denominator of which is i.sub.m minus i.sub.n, if i.sub..lamda. is greater than i.sub.n, else if i.sub.m does not equal i.sub..lamda., then the denominator is i.sub.m minus i.sub.x, else the denominator is 1, where mdenotes the index of the direction of the main excitation, and .lamda., n denote indices of the directions before and, respectively, after the direction of the main excitation; and wherein the estimated position is obtained as m times .pi. over p plusthe calculated ratio between differences of measured responses for the selected stator flux vectors times .pi. over 2p.

2. The method according to claim 1, wherein the motor works in a non-linear (saturation) region of inductivity characteristics at least in a part of a position range.

3. The method according to claim 1, wherein the measured responses are proportional to an actual inductivity or counterinductivity of the multi-phase windings of the motor.

4. The method according to claim 3, wherein a measured response proportional to the inductivity is the time period of an excitation pulse to a stator phase until the response signal transient reaches a fixed maximum level.

5. The method according to claim 3, wherein a measured response proportional to the counterinductivity is the maximum of the current response transient of a stator phase for an excitation pulse of a fixed time period.

6. The method according to claim 3, wherein position dependent motor values are inverted before being applied in a further calculation, if they are proportional to the inductivity of the windings; the measured or calculated position dependentmotor values that are proportional to the counterinductivity of the windings being denoted by {i.sub.0,. . ., i.sub.2p-1}.

7. The method according to claim 1, wherein a measured response proportional to the inductivity is the time period of an excitation pulse to a stator phase until the response signal transient reaches a fixed maximum level.

8. The method according to claim 1, wherein a measured response proportional to the counterinductivity is the maximum of the current response transient of a stator phase for an excitation pulse of a fixed time period.

9. A method of determining an angular position of a rotor of a brushless direct current motor having a stator with multi-phase windings, control circuitry providing excitations switched one after another to motor phases such that theexcitations produce stator flux vectors lying in different directions in a full 360 electrical degree range, and a measurement circuit that delivers responses for the phase excitations, the method comprising the acts of: selecting a main excitationstator flux vector resulting in a minimal inductivity shown by a measured response; selecting two other stator flux vectors, one of which is in a previous and another of which is in a subsequent direction with respect to a direction of the mainexcitation stator flux vector; and computing a ratio between differences of measured responses for the selected stator flux vectors in order to estimate an angular position of the rotor of the brushless direct current motor, wherein the motor phases areexcited with at least 3 stator flux vectors selected from the 2p stator flux vectors lying at electrical angles of k.pi./p, k =0,. . ., 2p -1, where p denotes a number of phases of the motor, where selection is made such that at least the main vectorand the vectors lying in the previous and subsequent directions are chosen for excitation, wherein the selection of the main excitation stator flux vector is done by selecting a vector that corresponds to a maximal value among {i.sub.0,. . .,i.sub.2p-1}, where if more than one equivalent maximal value is found, then any of their indices are selectable for the index of the main excitation stator flux vector; the ratio between differences of measured responses for the selected stator fluxvectors being computed as a fraction, the numerator of which is i.sub.n minus i.sub..lamda., and the denominator of which is i.sub.m minus i.sub.n, if i.sub..lamda. is greater than i.sub.n, else if i.sub.m does not equal i.sub..lamda., then thedenominator is i.sub.m minus i.sub..lamda., else the denominator is 1, where m denotes the index of the direction of the main excitation, and .lamda., n denote indices of the directions before and, respectively, after the direction of the mainexcitation; and wherein the estimated position is obtained as m times .pi. over p plus the calculated ratio between differences of measured responses for the selected stator flux vectors times .pi. over 2p.

10. The method according to claim 9, wherein position dependent motor values are inverted before being applied in a further calculation, if they are proportional to the inductivity of the windings; the measured or calculated position dependentmotor values that are proportional to the counterinductivity of the windings being denoted by {i.sub.0,. . ., i.sub.2p-1}.

11. A method of determining an angular position of a rotor of a brushless direct current motor having a stator with multi-phase windings, control circuitry providing excitations switched one after another to motor phases such that theexcitations produce stator flux vectors lying in different directions in a full 360 electrical degree range, and a measurement circuit that delivers responses for the phase excitations, the method comprising the acts of: selecting a main excitationstator flux vector resulting in a minimal inductivity shown by a measured response; selecting two other stator flux vectors, one of which is in a previous and another of which is in a subseauent direction with respect to a direction of the mainexcitation stator flux vector; and computing a ratio between differences of measured responses for the selected stator flux vectors in order to estimate an angular position of the rotor of the brushless direct current motor, wherein position dependentmotor values are inverted before being applied in a further calculation, if they are proportional to the inductivity of the windings; the measured or calculated position dependent motor values that are proportional to the counterinductivity of thewindings being denoted by {i.sub.0,. . .,i.sub.2p-1}, wherein the selection of the main excitation stator flux vector is done by selecting a vector that corresponds to a maximal value among {i.sub.0,. . . i.sub.2p-1}, where if more than one equivalentmaximal value is found, then any of their indices are selectable for the index of the main excitation stator flux vector; the ratio between differences of measured responses for the selected stator flux vectors being computed as a fraction, thenumerator of which is i.sub.n minus i.sub..lamda., and the denominator of which is tm minus i.sub.n, if i.sub..lamda.is greater than i.sub.n, else if i.sub.m does not equal i.sub..lamda., then the denominator is i.sub.m minus i.sub..lamda., else thedenominator is 1, where m denotes the index of the direction of the main excitation, and .lamda., n denote indices of the directions before and, respectively, after the direction of the main excitation; and wherein the estimated position is obtained asm times .pi. over p plus the calculated ratio between differences of measured responses for the selected stator flux vectors times .pi. over 2p.