3D surface scanning is a technique brought forward for wound documentation and analysis in order to identify injury-causing tools in legal medicine and forensic science. Although many case reports have been published, little is known about the methodology employed by the authors. The study reported here is exploratory in nature, and its main purpose was to get a first evaluation of the ability of an operator, by means of 3D surface scanning and following a simple methodology, to correctly exclude or associate an incriminated tool as the source of a mock wound. Based on these results, an assessment of the possibility to define a structured methodology that could be suitable for this use was proposed. Blunt tools were used to produce 'wounds' on watermelons. Both wounds and tools were scanned with a non-contact optical surface 3D digitising system. Analysis of the obtained 3D models of wounds and tools was undertaken separately. This analytical phase was followed by a qualitative and a quantitative comparison. Results showed that in more than half of the cases, we obtained a correct association but the prevalence of wrong association was still high due to mark deformation and other limitations. Even if the findings of this exploratory study cannot be generalised, they suggest that the simple and direct comparison process is not reliable enough for a systematic routine application. The article highlights the importance of an analysis phase preceding the comparison step. Limitations of the technique, ensuring needs and possible paths for improvement are also expounded.