PURPOSE The aim of this study was to improve the scanning peripheral anterior chamber depth analyzer (SPAC) for measuring anterior chamber depth (ACD), differentiating eyes with shallow anterior chamber, and automatically categorizing eyes into subgroups based on the ACD values. We also investigated its accuracy for measuring anterior ocular biometry. MATERIALS AND METHODS The improved SPAC system was equipped with an auto-focusing system for measuring ACD from the optical axis to the limbus without contacting the ocular surface, in addition to an auto-measuring system for determining central corneal thickness (CCT) and corneal radius of curvature (CRC). This system was also equipped with an auto-diagnosing program for differentiating eyes with narrow angle and an auto-classifying program for categorizing eyes into 12 subgroups according to the ACD values. A dummy eye was used for investigating the measurement accuracy and reproducibility of CCT, CRC, and ACD. Measurement duration and clinical availability were investigated using healthy subjects and glaucoma patients. RESULTS The SPAC-measured CCT, CRC, and ACD values were very similar to the theoretical values obtained from the dummy eye and their coefficients of variation were less than 1%. This system completed ACD measurement in 15.6+/-1.0 seconds and the mean coefficient of variation of ACD was 8.8+/-2.3%. The improved SPAC system successfully differentiated eyes with narrow angle from those with wide open angle. CONCLUSIONS The improved SPAC system can measure ACD easily with good accuracy and reproducibility, and is considered to be suitable for evaluating ACD in routine examinations and screening eyes with narrow angle.