SynopsisOne basic problem in the measurement of magnetostriction in laminations is that samples of silicon-iron sheet are rarely perfectly flat when tested. This can sometimes lead to spurious values for the longitudinal magnetostriction, with ceramic displacement transducers. A method has been developed whereby, using three transducers in turn, the true longitudinal component of magnetostrictive motion can be determined. The use of an oil film between the sample and the bed plate, to reduce or eliminate the vertical movement, has been shown to be unreliable.To measure the magnetostriction directly over a small region (say, from grain to grain), a special pickup has been developed from the basic parts of two transducers. Each part is calibrated independently, and the two outputs are connected in opposition. By varying the apparent sensitivity of one part of the double pickup in predetermined steps, the true longitudinal magnetostriction over the small region can be evaluated.