Rising maintenance costs of ageing infrastructure necessitate innovative monitoring techniques. This paper presents a new approach for detecting axles, enabling real-time application of Bridge Weigh-In-Motion (BWIM) systems without dedicated axle detectors. The proposed Virtual Axle Detector with Enhanced Receptive Field (VADER) is independent of bridge type and sensor placement while only using raw acceleration data as input. By using raw data instead of spectograms as input, the receptive field can be enhanced without increasing the number of parameters. We also introduce a novel receptive field (RF) rule for an object-size driven design of Convolutional Neural Network (CNN) architectures. We were able to show, that the RF rule has the potential to bridge the gap between physical boundary conditions and deep learning model development. Based on the RF rule, our results suggest that models using raw data could achieve better performance than those using spectrograms, offering a compelling reason to consider raw data as input. The proposed VADER achieves to detect 99.9 % of axles with a spatial error of 4.13 cm using only acceleration measurements, while cutting computational and memory costs by 99 % compared to the state-of-the-art using spectograms.