Techniques for improving resolver-to-digital conversion accuracy

Several methods for reducing the position error caused by the existence of imperfect resolver signal characteristics are introduced. The most straightforward method is simply to calibrate each resolver and resolver-to-digital (R/D) converter and then correct the R/D converter output in real time. Although this method corrects all errors, including those with an origin in the R/D converter, it is also the most time-consuming and labor-intensive. By appropriate signal processing, it is shown that quadrature error can be eliminated by simple algebraic manipulation of the resolver signals. Similarly, it is shown that all even harmonics in the resolver signals can be canceled if the resolver is constructed with complementary phases. It is shown that amplitude imbalance can be used to reduce the position error due to inductive harmonics. >