The bimetallic coil thermometer is not a satisfactory device for measuring food temperatures for the following reasons as shown in the photographs.

1. The coil extends in the stem from about 1/4 inch to about 3 inches up the stem. This coil measures the average temperature of the food surrounding the coil. In a pan of hot food, this can be +/-50F, and the coil only gives the average temperature. For example, the food might be 100F at the top of the pan and 200F at the bottom of the pan, and the coil reads 150F.2. The specification that the thermometer is accurate to +/-2F is only true under perfect laboratory conditions. The thermometer must be perfectly calibrated in slush ice, not cube ice, and the pointer must be precisely on 32F. The companies that make these thermometers provide no guarantee of accuracy at 150F or higher, because the coil, after perhaps as little as a week of use, can corrode in the stem, causing it to malfunction.
3. For many mechanical reasons, the stem can get filled with liquid from food or wash water that seeps in around the gasket at the nut. This device is not designed to be immersed. The Oregon Department of Agriculture has reported that when inspectors check thermometers in operation, they are often off by more than 10F.
In summary, because of the problems with construction and calibration, this device cannot be trusted to give an accurate reading. The thermistor or thermocouple should be used.