The last method is the most involved and is usually only available to the IC manufacturer. Here one of the parasitic diodes appearing between the pins of the regulator and ground as a temperature sensor is used. The control pins of the device will always have an ESD protection diode connected from the pin to GND, and sometimes to VIN. Biasing this pin so that the diode is turned on provides a voltage that is temperature dependent. The first step is to calibrate the diode voltage vs. temperature in an environmental chamber. Typically the diode voltage temperature coefficient is about -2mV/°C. Then, with the same current in the diode, the voltage is measured when the device is running in the actual application. The calibration curve is then used to estimate the junction temperature from the diode voltage. There are many pitfalls with this method, such as selecting the proper pin, noise from the converter corrupting the measurement, and the time consuming calibration procedure. This method can be employed when the use of a thermal camera would be very difficult, such as when the test hardware is inside an environmental chamber.