Brake wear emissions are investigated during on-road driving with a midsize passenger car on a closed test track. A novel sampling system is designed that aims at monitoring the entire aspiration of brake wear particles. The wear particles are collected by a cone-shaped sampler, which is attached at the outer side of the wheel rim. Thus, the air flow direction penetrating the brake assembly from the vehicle underbody to the vehicle outside is preserved. For analysis, the wear aerosol is routed to the trunk of the car. In addition to the emission measurements, the setup flow is monitored, which enables quantification of the acquired emission data. A 3 h subsection of the Los Angeles City Traffic (LACT) cycle, representative for realistic driving behavior, is used as test cycle. For two different brake materials, PM10 emission factors are ranging from 1.4 mg km−1 brake−1 to 2.1 mg km−1 brake−1, while one material is found to be 18% less emissive. Due to high brake disc temperatures exceeding 170 °C, high particle number emissions occur through ultrafine particle generation. The unrealistic temperatures are caused by the limited brake cooling in the semi-closed measurement setup. In contrast, the reference brake temperature does not exceed 153 °C during the same test, thus ultrafine brake particle emissions are not expected during normal driving. Furthermore, well run-in brake material shows emission factors near the lower measurement limit at the unrealistic temperatures, suggesting that ultrafine particle emissions are characteristic for brand-new materials in combination with high brake temperatures observed due to the semi-closed housing of the measurement setup.