The XblocPlus is a new type of single layer armour unit for breakwaters. The key difference between earlier single layer armour units is that the unit is not only placed in a regular grid, but also with a regular orientation. The main benefit of this is that it can directly be seen whether the unit is placed correctly. Additionally, the placement always takes place with the same repetitive movement, as is preferred by crane drivers. The benefits to the placement procedure would however be reduced when the requirements for the under layer become very tight. Since the time that is saved during placement of the armour units, then would be lost during the profiling of the under layer. This has led this study to focus on the influence of irregularities in the under layer on the stability of the XblocPlus and the allowed tolerances for the placement of the under layer. In the process of obtaining a stable armour unit, multiple researches have been conducted. The final shape has been further optimised by adding a hole in the middle of the unit and increasing the interlocking capacity. Better interlocking increased the connection between the units and thereby improved the resistance. The hole increased the porosity and thereby reduced the overpressure underneath the armour layer. Which is a result of a difference in water level inside and outside of the breakwater. The increase in stability due to the enlargement of porosity, validated earlier findings that the acting failure mechanism is extraction due to the overpressure. The stability was expected to be mainly effected by the amount of interlocking. Subsequently, the amount of interlocking is determined by the relative angle between the armour units. For this research physical model tests have been performed to check that hypothesis. Multiple tests have been conducted with different radii of convex and concave shapes in both long shore and cross shore direction. The goal was to find the critical relative angles of each combination of shape and direction. The convex shape in cross shore direction was the only irregularity that caused failure and proved to be critical. The relation between the relative angle and the stability number resulted to be linear, instead of the expected drop of stability at a certain critical value. The relation showed signicant spreading, for which tests with small deviations from the design profile seem to form the upper bound and large deviations the lower bound. Thereby indicating that the level of stability is not only determined by the relative angle but also by the deviation from the design profile. In succession of the tests with specific shapes and directions, more realistic tests have been performed with micro irregularities and S-proles. The convex shape in cross shore direction proved to be critical for these congurations as well. Furthermore it was found that for large S-proles, sudden failure could occur without previous indication of loss of interlocking. Further analysis of the divergent failure behaviour of large S-profiles indicated that the arched shape of the profile enabled the lifting of the armour layer. A mechanism similar to the buckling of beams. The pressure inside the breakwater, the drag of the down-rush and the weight of the upper part of the slope, cause the initial distortion of the profile to be enlarged. The differences between the measurements of the profile before and after the tests, showed that the same movements occur for smaller S-proles and the tested convex shapes in cross shore direction. The movement could even be seen in tests that have not failed. Which implies that the arching mechanism that causes sudden failure for large S-proles, also induces the loss of interlocking for other convex shapes. A simplied model of the forces on the convex sections in the profile indicates that the enforcing of the arching mechanism mainly depends on the length and steepness of the downward part of the convex section and the  atness of the upward part of the convex section. These findings correspond with the test results and explain why the stability performance is the highest for the short length scale micro irregularities and lowest for the large scale S-proles with large deviations from the design profile. Evaluation of the effect of irregularities is performed with the tolerances of the regular Xbloc, which are also intended to be used on the XblocPlus. To prevent differences in stability due to the size of the under layer, the tolerances are expressed in the unit size (Dn) instead of the under layer grain size (Dn50). The resulting tolerances are maximum 0.25 Dn for the deviation from the design profile and maximum 0.1 Dn deviation between subsequent measurements along the profile. The profile is measured every 10m of the breakwater and subsequent measurements along the profile have a distance of 0.3 Dn. The requirements are a little more liberal than the tolerances of the regular Xbloc since the tolerances of the largest under layer grain size are taken as a starting point. The test results indicate that the tolerances are sufficiently safe, with potential for more liberal requirements.