Improvements to a method of calibrating a high-density tape recording system are suggested. The small gap approximation is not made, and allowances for the effects of demagnetization are included. A better and smaller estimate of the effective head-media spacing is obtained. © 1992 IEEE