On Saturday, December 26, 2015 at 3:58:56 PM UTC-5, Dave M wrote:
It's the weight of the mercury that moves the bimetal coil spring such that
it takes a higher temperature to bend the bimetal coil spring so that the
mercury flows back to the other end of the bulb. When the mercury moves to
one end of the bulb, it makes the spring reposition itself so that it takes
a much higher (or lower, depending on heat or cool mode) for the spring to
return back to its original position.
Here's a link to a web page that describes the operation very well, saving
me a lot of typing.
http://inspectapedia.com/heat/Thermo...e_Response.php

Cheers,
Dave M


That would be how I would think it would work too. Once the mercury
rolls in one direction, it's going to take X amount of temp change
to make it shift enough to roll back the other way. That's the
hysteresis. As othere have said, the anticipator just heats the
thermostat so that it opens a bit early, to avoid overshooting. The
anticipator can be set. I haven't seen an old mercury type thermostat
where you could set the hysteresis.