Objects in equilibrium have no net force acting on them, so they may be
"Static" equilibrium obviously refers to the case in which the objects are at rest.
Objects that take up some area or volume can be in static equilibrium as well, but this requires an additional condition: the sum of the torques acting on them also has to be zero.
Let's see how we can use the Newton's second law of motion to analyze these situations.
There are two conditions that must be met for an object to be in static equilibrium.
For an object to be in static equilibrium, it has to be not moving translationally (linearly), and not moving rotationally. It must not be moving relative to a given reference frame.
If this object is static—not moving—then we can also conclude that:
Only if both of these requirements are met do we have a case that can be considered static equilibrium.
For us to be able to understand static equilibrium and be able to use it to solve problems, forces are obviously a key factor.
Using the situation shown here, draw a free-body diagram that identifies six forces acting on the ladder.
The forces acting on the ladder are shown in the diagram to the right. They include, from bottom to top:
We'll be using these forces as well as the torques produced by these forces, calculated relative to some hypothetical axis of rotation about which the ladder might rotate. (It isn't rotating, however, because all of the torques are balanced.)
For an object that is in static equilibrium, where is the axis of rotation that we measure the torques relative to?


It turns out that if the forces all sum to zero for a a 2- or 3-dimensional object, and the torques sum to zero for a given point (axis of rotation) for that object, then the torques sum to zero for all points (axes of rotation) in that object.

We can demonstrate this mathematically.
Let's take a look at the torque τ1 relative to the original point O from F1, which is r1×F1. If we locate a new axis of orientation O′, located at a displacement r′ from the original O, we can determine the torque relative to the new axis O′ as τ1′ = (r1 - r′) ×F1.
See the diagram above to confirm this relationship.
Extending that analysis to the other three forces and their corresponding torques, we have:
Coming soon!
Solving static equilibrium problems involves a strategy similar to that we've used for other force-related problems:
Let's see how we can solve some problems.
A 5.0-kg lamp hangs from a b) post as shown, supported by a hinged horizontal beam (L = 2.0 m, m = 3.0 kg) and a supporting cable oriented at 30 ° to the horizontal.
The free-body diagram for the horizontal beam is about what you would expect, although we're introducing a new strategy for helping to solve this type of problem. The hinge on the left end of the beam supplies a force that helps to keep the beam static, but we don't know much about it. In this diagram I've drawn the horizontal force acting on that end of the beam as Rx pointing off to the right, because I'm pretty sure I need a horizontal force to balance the x-component of the tension in the negative direction. I didn't know if the y-component of that hinge force R is up or down, so I'm guessing up, and we'll see how that works.