With the introduction of BroadR-Reach and time-sensitive networking (TSN), Ethernet has become an option for in-vehicle networks (IVNs). Although it has been used in the IT field for decades, it is a new technology for automotive, and thus requires extensive testing. Current test solutions usually target specifications rather than the in-vehicle environment, which means that some properties are still uncertain for in-vehicle usage (e.g., fault tolerance for shorted or open wires). However, these characteristics must be cleared before applying Ethernet in IVNs, because of stringent vehicular safety requirements. Because CAN is usually used for these environments, automotive Ethernet is expected to have the same or better level of fault tolerance.Both CAN and BroadR-Reach use a single pair of twisted wires for physical media; thus, the traditional fault-tolerance test method can be applied for automotive Ethernet. However, because of the new transceiver circuit, some test methods need to be modified. This paper analyzes and tests the physical communication mechanism for BroadR-Reach to compare with the CAN bus. Test cases, such as a shorting a wire to BAT or GND, ground shift, and load capacity, are applied to TJA1100 from NXP. Although these test results are related to chip manufacturers and production batches, the analysis results still have guiding significance for the application of automotive Ethernet to avoid dangerous conditions.