The tests reported here covers acoustic emission measurements on a typical model scale set-up with a model propeller in behind condition on a hull model in the cavitation tunnel. An AE transducer was mounted in rear end of the propeller shaft on an existing Aframax model. Two propellers were tested at several operating conditions. P2754 showed weak indication of erosion at a few bladed tips in the erosion tests. P5091 did not show any sign of erosion. The number of AE peaks at each amplitude are approximately proportional to the amount of cavitation. Even atmospheric condition, there are some, but very few, peaks up to 3.9 V amplitude. Whether these are due to sporadic cavitation incidents or some disturbance in the measurement is not clear at this moment. At the cavitating conditions, peaks up to 4.0 V amplitude are detected. All these conditions are judged to be barely erosive with just a few pit marks on a couple of blades. The question is if there would be even higher amplitudes in a highly erosive case? It is obvious that the tests must include conditions that are significantly more erosive. LR and DSME have suggested a threshold amplitude of 3.6 V for erosion based on systematic Sonotrode tests with erosion paint samples. It seems such amplitude threshold must be combined with a minimum number of incidents to be classified as erosive, otherwise non-cavitating conditions would also be classified as erosive.