Static Friction
Friction is only as large as it needs to be, until it can't be any larger.
Before this Lesson
- Equilibrium of a Particle
- Reactions of Bent Bars, Brackets, and Booms
- Exponentials and Logarithms · Mathematics and Statistics
A 100 lb crate sits on a level floor, and the coefficient of static friction between them is 0.4. Nobody's touching it. How much friction does the floor put on the crate? You have a coefficient and a normal force, so 40 lb looks like the obvious answer. But picture what a 40 lb sideways force does with nothing to oppose it: the crate would set off across the floor on its own. So what's wrong with the obvious answer?
Friction takes only what it needs
The answer is zero. Nothing pushes the crate sideways, so leaves friction nothing to do. Push with 10 lb, and friction answers with 10 lb the other way. Static friction is a reaction, like the force at a pin: it takes whatever value equilibrium asks for.
Unlike a pin, it has a ceiling. The harder two surfaces are pressed together, the harder they can grip, so the ceiling is proportional to the normal force, and the area of contact doesn't enter. The Handbook's Statics chapter, under Friction, calls this ceiling the limiting friction:
Here is the friction force, is the normal force between the surfaces, and is the coefficient of static friction. The Handbook has no table of coefficients, so a problem always gives you . For the crate, the ceiling is .
The instant friction reaches its ceiling is called impending slip, and it's the only moment you may write . The Handbook's Dynamics chapter says so in those words, also under Friction. Problems signal it with phrases like "just starts to move." Past that point the crate is moving, and moving bodies belong to Newton's Second Law.
Which way does friction point? It opposes the slide that would happen without it. So erase friction in your head, see which way the body would go, and point friction the other way.
Does it slide?
That gives a test for any body at rest. Find the friction equilibrium needs, then the friction available, . If the need fits under the ceiling, the body stays, and friction equals the need. If not, it slides.
On a rough incline, the need comes from the weight. In Equilibrium of a Particle, a cable along the slope held a roller with while the surface carried . Swap the cable for friction, and nothing else changes:
The Handbook prints neither. Notice that isn't : the surface carries only the part of the weight aimed into it.
Tilt the board
Set a block on a board and lift one end slowly. The need, , grows. The ceiling, , shrinks. At some angle they meet, and the block goes:
The weight canceled, so a heavy block and a light one of the same material let go at the same angle. That angle belongs to the pair of surfaces, and it's called the angle of friction, :
So a block on an incline stays if and slides if , and you don't need the weight to say which.
The Handbook never names this angle, and appears only in its Screw Thread entry. You'll meet the idea again in Shear Strength, where soil resists shear with normal force times the tangent of a friction angle.
Friction can point either way
Now hold a block on a steep slope with a force parallel to it. There are two ways to lose. Pull too gently, and the block starts down, so friction points up the slope and helps you. Pull too hard, and it starts up, so friction flips and fights you. At each limit, friction is at its full :
Between those two, the block stays put and friction covers the difference. If , comes out negative, which just means the block holds itself and needs no help.
A horizontal push is more awkward. Part of it, , drives the block along the slope. The other part, , presses the block into the slope, which raises the ceiling:
Now hides inside the friction too. Point friction against the way the block is about to go, write the equation along the slope and the equation normal to it, and solve them together, by hand or with a calculator's system solver.
An angled force changes the normal force
Back on a level floor, tilt the force. Pull with a rope angled up, and its vertical part lifts some of the crate, so the ceiling drops. Push down at an angle, and the vertical part presses the crate into the floor, so the ceiling rises:
At impending slip, the horizontal part of equals . A 200 lb crate with and a rope 30° above the floor gives:
Pretending gives 92.4 lb, and pushing at the same angle takes 120 lb. That's the reason to pull a heavy sled and not push it.
Friction at every surface that can slide
Stacks and tied blocks are still isolate, draw, sum to zero, one block at a time. Every surface that could slide gets its own friction force, equal and opposite on the two bodies it touches. And each comes from what that surface actually carries, so the surface under a stack carries everything above it.
If one block is tied back, the other has to break free at both its surfaces at once, so those two ceilings add. If nothing is tied back, slip could start at either surface, so find the force each one allows and take the smaller, like the cable that governs a junction. A cord between blocks just passes force along, so a line of tied blocks moves only when the pull reaches the sum of their ceilings.
A ladder on a rough floor fits here too. In Reactions of Bent Bars, Brackets, and Booms, a stop at the foot of a ladder against a smooth wall pushed with 105 lb, while the floor carried 200 lb. Take the stop away, and friction has to supply that 105 lb, so the least coefficient that works is:
The Handbook has nothing on stacks or ladders. All of this is and equilibrium.
A rope on a post
Take a few turns of rope around a post, and one hand can hold a load that would otherwise drag you across the dock. The rope does it by squeezing the post.
Look at one short piece of rope lying on the post. Tension pulls at each end of the piece, and because the rope curves, those two pulls press it into the post with a force of about . So friction on that piece can reach , and that's how much the tension can change across it:
More tension squeezes harder, which grips harder, so each bit of angle multiplies the tension by the same factor. That's the change that multiplies from Exponentials and Logarithms, smeared out continuously, so its base is . The Handbook's Belt Friction entry gives the result:
Here is the force applied in the direction of impending motion, is the force resisting it, is the coefficient of static friction, and is the total angle of contact in radians, so convert turns and degrees first. is always the larger one. It's the tight side, and slip impends toward it.
Because the wrap multiplies, doubling the angle squares the ratio: .
| Wrap | (rad) | |
|---|---|---|
| Half turn | 3.14 | 2.57 |
| One turn | 6.28 | 6.59 |
| Two turns | 12.57 | 43.4 |
| Four turns | 25.13 | 1,881 |
To find , ask which way the rope would slide. When you're holding a load, the rope would run toward the load, so the load is and your hand is . When you're raising it, the rope would run toward your hand, so your pull is . Same post, same wrap, two very different numbers:
Now the numbers. Convert to radians yourself: is a plain exponent, so the calculator's degree setting never touches it.
Working backward
When the load is the unknown, multiply by the factor. When the angle or the coefficient is unknown, it's stuck up in an exponent, and brings it down:
To hold 1,000 lb with a 50 lb pull at , the ratio is 20, so . Dividing by gives 1.59 turns.
A belt on a pulley
A flat belt on a pulley obeys the same equation. Here is the contact angle, the arc where the belt touches the pulley, and the equation caps how much tighter one side can be than the other. That difference is what turns the pulley. Each side pulls tangent to the pulley, with the radius as its arm, and their moments oppose each other, so the torque the belt delivers is:
The Handbook gives the tension ratio but not this line, which is just two moments. Torque is largest when the ratio is at its limit. With a slack side of 100 lb, , and half a turn of contact, the tight side can reach 256.6 lb. On a pulley of 6 in radius, that's .