[en] This paper concerns the evaluation of the pull-off adhesion between a concrete added repair layer with variable thickness and a concrete substrate, based on parameters assessed using ultrasonic pulse velocity (UPV) method. In construction practice, the experimental determination of pull-off adhesion f _b, between added repair layer and a concrete substrate is necessary to assess the quality of repair. This is usually carried out with the use of pull-off method which results in local damage of the added concrete layer in all the testing areas. Bearing this in mind, it is important to describe the method without these disadvantages. The prediction of the pull-off adhesion of the two-layer concrete elements with variable thickness of each layer might be provided by means of UPV method with two-sided access to the investigated element. For this purpose, two-layered cylindrical specimens were obtained by drilling the borehole from a large size specially prepared concrete element. Those two-layer elements were made out of concrete substrate layer and Polymer Cement Concrete (PCC) mortar as an added repair layer. The values of pull-off adhesion f _b of the elements were determined before obtaining the samples by using the semi-destructive pull-off method. The ultrasonic wave velocity was determined in samples with variable thickness of each layer and was then compared to theoretical ultrasonic wave velocity predicted for those specimens. The regression curve for the dependence of velocity and pull-off adhesion, determined by the pulloff method, was made. It has been proved that together with an increase of ratio of investigated ultrasonic wave velocity divided by theoretical ultrasonic wave velocity, the pull-off adhesion value f _b between added repair layer with variable thickness and a substrate layer also increases. (paper)