Electromagnetic Tracking (EMT) is a novel technique for error detection and quality assurance (QA) in interstitial high dose rate brachytherapy (HDR‐iBT). The purpose of this study is to provide a concept for data acquisition developed as part of a clinical evaluation study on the use of EMT during interstitial treatment of breast cancer patients. The stability, accuracy, and precision of EMT‐determined dwell positions were quantified. Dwell position reconstruction based on EMT was investigated on CT table, HDR table and PDR bed to examine the influence on precision and accuracy in a typical clinical workflow. All investigations were performed using a precise PMMA phantom. The track of catheters inserted in that phantom was measured by manually inserting a 5 degree of freedom (DoF) sensor while recording the position of three 6DoF fiducial sensors on the phantom surface to correct motion influences. From the corrected data, dwell positions were reconstructed along the catheter's track. The accuracy of the EMT‐determined dwell positions was quantified by the residual distances to reference dwell positions after using a rigid registration. Precision and accuracy were investigated for different phantom‐table and sensor‐field generator (FG) distances. The measured precision of the EMT‐determined dwell positions was ≤ 0.28 mm (95th percentile). Stability tests showed a drift of 0.03 mm in the first 20 min of use. Sudden shaking of the FG or (large) metallic objects close to the FG degrade the precision. The accuracy with respect to the reference dwell positions was on all clinical tables < 1 mm at 200 mm FG distance and 120 mm phantom‐table distance. Phantom measurements showed that EMT‐determined localization of dwell positions in HDR‐iBT is stable, precise, and sufficiently accurate for clinical assessment. The presented method may be viable for clinical applications in HDR‐iBT, like implant definition, error detection or quantification of uncertainties. Further clinical investigations are needed.