Assessment of a 1:8 physical scale model for comparative analysis of modal behaviour and early reflections in small rooms

Physical acoustic scale modelling is widely used in architectural acoustics, but its applicability to small-room treatment studies has not been systematically validated. This paper evaluates a 1 : 8 physical scale model as a comparative tool for small-room adaptation planning by comparing measurements performed in room 17E in building D1 at AGH University of Krakow with measurements obtained in its scaled model. Scale-model impulse responses were interpreted on full-scale-equivalent frequency and time axes. The validation focused on modal-frequency correspondence and early-reflection timing, while treatment-induced changes were evaluated as comparative trends within the scale model. The scale model reproduced the main geometry-controlled modal-frequency regions with good accuracy. Across all five assigned 1 : 1 - 1 : 8 modal pairs, the mean absolute frequency difference was 1.10 Hz . This value was dominated by the vertical axial mode ( 0 , 0 , 1 ) , with a rigid-boundary theoretical frequency of 55.86 Hz , for which the scale-model peak occurred at 56.55 Hz , whereas the corresponding full-scale-room feature was observed at 52.24 Hz . The lower full-scale frequency was consistent with the compliant suspended-ceiling boundary and its coupling with the ceiling plenum, which were not reproduced by the rigid acrylic enclosure. For the remaining four assigned modal pairs, the mean absolute frequency difference was 0.30 Hz . Within the scale model, the added corner bass traps produced consistent changes in selected modal regions, especially around 35 - 50 Hz and 60 - 80 Hz ; these results are interpreted as treatment-induced trends rather than full-scale amplitude predictions. Peak energy time was used as a supplementary exploratory descriptor of the temporal location of dominant low-frequency energy and indicated earlier energy maxima after treatment in selected modal regions. Reflectogram analysis showed that the timing of the main early reflections can be assessed after time scaling, although reflection visibility and amplitude depended strongly on miniature-source directivity. Among the tested sources, the subwoofer-type source was most suitable for modal analysis, while the ribbon-speaker source provided the clearest isolated reflectogram peaks. The results indicate that 1 : 8 scale modelling can serve as an intermediate physical testing stage between numerical modelling and full-scale validation, particularly for comparative screening of small-room acoustic-treatment concepts with predominantly rigid boundary conditions.