The nature and abundance of sulfur chemistry in protoplanetary disks (PPDs) may impact the sulfur inventory on young planets and therefore their habitability. PPDs also present an interesting test bed for sulfur chemistry models, since each disk present a diverse set of environments. In this context, we present new sulfur molecule observations in PPDs, and new S-disk chemistry models. With ALMA we observed the CS 5 − 4 rotational transition toward five PPDs (DM Tau, DO Tau, CI Tau, LkCa 15, MWC 480), and the CS 6 − 5 transition toward three PPDs (LkCa 15, MWC 480 and V4046 Sgr). Across this sample, CS displays a range of radial distributions, from centrally peaked, to gaps and rings. We also present the first detection in PPDs of 13 CS 6 − 5 (LkCa 15 and MWC 480), C 34 S 6 − 5 (LkCa 15), and H 2 CS 8 17 − 7 16 , 9 19 − 8 18 and 9 18 − 8 17 (MWC 480) transitions. Using LTE models to constrain column densities and excitation temperatures, we find that either 13 C and 34 S are enhanced in CS, or CS is optically thick despite its relatively low brightness temperature. Additional lines and higher spatial resolution observations are needed to distinguish between these scenarios. Assuming CS is optically thin, CS column density model predictions reproduce the observations within a factor of a few for both MWC 480 and LkCa 15. However, the model underpredicts H 2 CS by 1-2 orders of magnitude. Finally, comparing the H 2 CS/CS ratio observed toward the MWC 480 disk and toward different ISM sources, we find the closest match with prestellar cores. two of the Taurus disks: the Herbig Ae disk MWC 480, and the T Tauri disk LkCa 15. The spectral resolution for these lines was ∼ 975 kHz, i.e. corresponding to a velocity resolution of ∼ 1 km/s. The 12 CS 6 − 5 transition was observed in three execution blocks on January 17, 2016, April 23, 2016 and December 12, 2016. For the first execution block 31 antennas were included and covered baselines from 15 to 331 m. 36 antennas were included for the two other execution blocks, covering baselines from 15 to 463 m and 15 to 650 m, respectively. The first and third blocks used the source J0510+1800 as band-pass and flux calibrators, and the source J0438+3004 as phase calibrator. The second block of execution used the source J0238+1636 as band-pass calibrator, the source J0433+2905 as phase calibrator, and the source J0510+1800 as flux calibrator. The total on-source integration time were 17.6, 12.6 and 20.7 min., respectively.The 13 CS and C 34 S 6−5, and H 2 CS 8 17 −7 16 rotational transitions were observed in a single execution block, on January 17, 2016, with the same calibrator sources as those used for the first and second execution blocks used for 12 CS 6 − 5 (see above) but with 36 antennas and a total on-source integration time of ∼ 19.2 min.