Many people frequently confuse mercury flame sensors (MFS) with thermocouples because each has a similar sensor that extends into the pilot flame and a tube connecting the device to the gas valve. Moreover, they have the same function in the control system of a gas-fired appliance, but there are significant differences.
The principal components of a mercury flame sensor (MFS) device are (1) a pilot flame sensor (the portion of the MFS device extending into the pilot flame), (2) a diaphragm/SPDT switch assembly located at the main gas valve, and (3) a hollow capillary tube connecting the flame sensor to the diaphragm/switch assembly.
The operation of a mercury flame sensor device depends on the evaporation of mercury. The sensor end, capillary tube, and the SPDT switch are filled with mercury. When there is enough heat produced by the pilot flame at the sensor end of the capillary tube, it vaporizes and pushes the remaining nonvaporized (liquid) mercury down the capillary tube to the bellows-type diaphragm/switch assembly located at the main gas valve. Movement of the bellows diaphragm presses against a nonadjustable, calculated spring tension with enough force to snap the SPDT switch from one set of contacts to another. This action causes the switch contacts to move from one position to another. In an MFS device switch assembly, the normally closed contact opens and the normally open contact closes. This action deactivates the igniter (after the pilot flame is proven) and opens the main gas valve to allow raw gas to flow to the burners.
Mercury flame sensors are no longer used in gas-fired furnaces and boilers, especially those equipped with solid-state control modules. However, manufacturers still produce replacement MFS units along with their compatible main gas valves.