Thermal Sensing Products
Edited by Miles Budimir
Thermostats are used in a wide variety of applications to monitor temperatures and prevent thermal damage. A prime example is monitoring the temperature of an engine manifold. Keeping the temperature at some defined point, or within some acceptable range, has always been a key design issue.
A typical car thermostat from the past used an inexpensive wax reservoir. This reservoir would expand when heated and push a plunger to activate a switch. However, it tends to drift and deteriorate over time, making it unreliable. Fortunately, there are better and more reliable alternatives.
Thermostats now come in several general categories such as differential expansion, liquid filled, mercury, and reed switch. Each has advantages. Differential expansion versions are the most common. In this type of thermostat, two metals with different coefficients of thermal expansion are bonded into one element. When temperature changes, the metal strips expand at different rates, bending the element in an arc to make or break a set of contacts.
One drawback to fused-bimetal strips is that the switchover tends to creep, so that there is not an immediate closing or opening of contacts. Another problem occurs when the sensing element is part of the circuit. Repeated opening and closing of the contacts produces arcing. This wears out the contact surfaces, raising the contact resistance and shortening the life of the thermostat.
Another differential-expansion thermostat uses a snap-acting bimetal disk. Like a fused-bimetal strip, a bimetal disk is made by fusing together two metals with different coefficients of thermal expansion. When the two metals are joined together and heated, the low-expansion metal retards the movement of the disk until such a point where the high-expansion metal just overcomes the energy that's built up retarding it. It then snaps over center, creating a positive snap make-and-break for both opening and closing. Isolating the disk from the circuit eliminates contact wear and chatter. It also means that it is not preheated and will not drift over time. Repeatability is also better than in creep thermostats.
An immersion thermostat is basically a temperature-actuated switch immersed in a fluid to monitor temperature. The switch can be configured to either open or close on rise. The industry standard has been a 0.5-in. bimetal disk, limiting the miniaturization to a 3 /8-in. National Pipe Thread (NPT), or M18 thread. However, with thermal protection being required more in the electronics industry, smaller disks are being used for thermostats on printed circuit boards to detect overheat problems in power supplies. They're packaged in industry-standard 8-pin DIPs and TO-220 packages. Smaller disks are also used in immersion theromstats, letting the thermostat package fit into smaller immersion housings, as low as a 1 /4-in. NPT or M14.
In the past, immersion thermostats used in engine protection were constructed of welded stainless steel. However, one-piece construction has become a more common way to manufacture theromstats. The one-piece construction uses machinable brass with a thicker cross-section to handle higher pressures. Standard pressure ratings of 1,500 psi can withstand pressure washing, meeting NEMA-4 and IP67 ratings. Also, one-piece construction is a little less expensive than the welded stainless steel.
Immersion snap-action thermostats are also used in so-called milestone applications. Here, it is important to alert an operator when a certain temperature has been reached. For instance, in one milestone application, the thermostat monitors the engine temperature. When the temperature drops below 80°F, it won't allow the operator to do anything but idle the engine until it warms up to 100°F. This avoids damaging a cold engine.