To manufacture and test a set of phase plates for the calibration of ocular aberrometers and apply it to the calibration of an ocular laser ray tracing aberrometer. The set of phase plates is made by a greyscale single-mask photosculpture in photoresist method. Each plate induces a given amount of a particular aberration (Zernike) mode. The set contains two subsets: 1) pure Zernike modes to test the accuracy among different orders (from 3rd to 7th, approximately 0.3 to 0.4 microm); and 2) plates having different amounts of the same mode, 3rd order coma ranging from 0.11 to 0.47 microm. After manufacturing, the plates were tested twice, as a crosscheck, measuring the aberration pattern of each plate with a Mach-Zehnder interferometer and a single-pass Hartmann-Shack wavefront sensor. The set was then applied to the calibration of an ocular double-pass laser ray tracing aberrometer. Close agreement was found between the three types of measurement. The maximum difference between Hartmann-Shack and laser ray tracing measurements was 0.032 microm (ie, approximately lambda/20, half of the typical measuring error in human eyes). This permitted detection of a small bias in the ocular laser ray tracing aberrometer. The calibration set may be a powerful tool for the assessment of accuracy and reliability in ocular aberrometry. It discovered a small bias, which is almost impossible to detect when working with human eyes or trial lenses. This type of calibration tool is especially important in clinical environments.