The thermal protection behavior can be split up into three categories of events that can happen. Figure 8-18 shows each of these categories.

1. Relative thermal shutdown: The device is enabled into an overcurrent event. The DIAG_EN pin is high so that diagnostics can be monitored on SNS and FLT (however, DIAG_EN being high is not necessary for all protection features to function). The output current rises up to the I_ILIM level and the FLT goes low while the SNS goes to V_SNSFH. With this large amount of current going through, the junction temperature of the FET increases rapidly with respect to the controller temperature. When the power FET temperature rises T_REL amount above the controller junction temperature ΔT = T_FET – T_CON > T_REL, the device shuts down. The faults are continually shown on SNS and FLT and the part waits for the t_RETRY timer to expire. When t_RETRY timer expires, because the LATCH pin is low and EN is still high, the device comes back on into this I_ILIM condition.
2. Absolute thermal shutdown: The device is still enabled in an overcurrent event with DIAG_EN high and LATCH still low. However, in this case the junction temperature rises up and hits an absolute reference temperature, T_ABS, and then shuts down. The device does not recover until both T_J < T_ABS – T_hys and the t_RETRY timer has expired.
3. Latch-off mode: The device is enabled into an overcurrent event. The DIAG_EN pin is high so that diagnostics can be monitored on SNS and FLT. The output current rises up to the I_ILIM level and the FLT goes low while the SNS goes to V_SNSFH. If the part shuts down due to a thermal fault, either relative thermal shutdown or absolute thermal shutdown, the device does not enable the channel until either the LATCH pin OR the EN pin is toggled.