An experimental study has been carried out to investigate the effect of contaminants such as sand and leaves on the electrical isolation of a wheel/rail contact. Train detection, for signalling purposes, can be by means of track circuits. Signalling block occupancy is triggered by the wheelset of the train ‘shorting out’ the track circuit. Contaminants in the wheel/rail interface mean that contact between the wheelsets and the track may be compromised, inhibiting train identification. Tests were performed using sections cut from wheels and rail sections loaded together hydraulically with different combinations of contaminants in wet and dry conditions. The electrical circuit used was a simplified simulation of the TI21 track circuit. The sand results indicated that a transition exists in the amount of sand in the contact above which isolation occurred. This amount is higher than the specified sand application rate for train operation at 4.5 m/s (10 mile/h), much of which will not actually make it into the contact. At lower speeds, down to zero, however, if the applied sand rate is not reduced, there is a possibility that this amount could enter the contact leading to isolation. Dead leaves were found to present the greatest problem in terms of isolation. The addition of sand used to mitigate the problem of adhesion loss caused by leaf fall made the likelihood of isolation greater. With water, breakdown loads were lower and partial conductance was seen to occur, although it is not clear whether these levels of conductance could be detected by a track circuit. This needs further investigation. Leaves with a higher level of sclerophylly (leatheriness) were found to be harder to break down. Loads used represented an average value for the tread contact. It was clear that for tests with dead leaves, particularly, the breakdown loads were around this level. With train masses gradually reducing leading to lower wheel/rail contact forces, the problem could get worse. At the present stage of development, the test method is best used as a means to qualitatively assess relative effects on electrical isolation of different contaminants, but can provide information that may be able to help in assessing where greater attention to leaf clearance may be necessary.