Demonstrate that the gravitational force acting on particle A inside a uniform spherical layer of the matter is equal to zero.
The force on A is zero. Every thin cone from A has an opposite cone, and the two patches pull with equal force in opposite directions: dF = γ M m dΩ / (4π a² cos φ) for both. For your sliders the pair gives 0 N and the sum over the whole layer gives N.
Problem statement
Problem 1.211. Demonstrate that the gravitational force acting on particle A inside a uniform spherical layer of the matter is equal to zero.
Interpretation. A hollow ball has a thin wall of equal density everywhere. Particle A sits anywhere inside the hollow. The task is to show that the wall pulls A with zero total force. This is a proof, so the page shows why the pulls cancel, and it also gives numbers for your sliders.
We assume the layer is thin and uniform, and the only force on A is the pull of the layer. The problem gives no numbers, so the radius a, the mass M of the layer and the mass m of A are sliders. The starting values are our own choice: a = 6.4 Mm, M = 6 × 1024 kg, m = 1 kg. We use γ = 6.67 × 10-11 N m²/kg².
Symbols: a is the radius of the layer, M its mass, m the mass of A, r the distance of A from the centre O, dΩ the solid angle of a thin cone, t the distance from A to a patch of the layer, φ the angle between a cone axis and the radius at the patch, γ the gravitational constant.
The big idea in plain words
A solid angle is how wide a cone is. Two thin cones from A that point in opposite directions have the same solid angle. The one that points to the nearer wall cuts out a small patch. The one that points to the far wall cuts out a bigger patch, because the cone has spread out more. The far patch has more mass, but it is also farther away. These two effects match exactly, so the pulls are equal and opposite.
What is a solid angle? Put a small ball around A. A cone from A cuts a small cap out of this ball. The solid angle dΩ is the area of that cap divided by the ball's radius squared. It is the 3D version of an angle in a flat drawing. Two opposite cones cut equal caps, so they have the same dΩ. The 3D view shows the small ball and both caps.
Why does A move in the simulation? In the problem A stays where it is. The slider r (and Play) only tries other places for A, to show that the pull is zero at every one of them. The grey dots mark the places tried.
Why match exactly? The area of a patch grows as t², so its mass grows as t². The pull is mass divided by t², so t drops out. Every direction through A has an opposite direction, so the whole layer splits into pairs that cancel.
The solution in each frame
Frame of the layer
The layer is at rest and A is inside it. The proof is done in this frame.
Other frames
A frame that moves or rotates adds forces that are not real. They are not needed for this proof, so no second frame is used.
The solution step by step
Step 1, two opposite cones. Draw from A two thin cones of the same small solid angle dΩ, along a line through A in opposite directions. The patch ahead is at distance t1 = Mm and the patch behind at t2 = Mm.
Step 2, the area of a patch. A cone of solid angle dΩ cuts an area dS = dΩ t² / cos φ out of the layer, where φ is the angle between the cone axis and the radius at the patch. The line through A and the two radii to its end points form an isosceles triangle (both radii equal a). So φ is the same at both ends: cos φ = . Then
Step 3, the mass. The layer is uniform, so dm = M dS / (4π a²). The masses are dm1 = kg and dm2 = kg.
Step 4, the pull. The pull of a patch on A is dF = γ m dm / t². Put in dm:
The distance t has cancelled, so dF1 = dF2 = N. They point in opposite directions, so the pair gives zero.
Step 5, the whole layer. Every direction through A has an opposite one, so the whole layer divides into such pairs. Each pair gives zero, so the force on A is zero. This holds for any position of A inside the layer, and for any mass m.
The page cuts the layer into 20000 thin rings about the line OA, adds the pull of every ring along that line, and gets N. That is percent of γMm/a² = N, which is zero within the sum's rounding. The same rings give the potential φ = J/kg, against −γM/a = J/kg from the formula (difference percent).
Second method: the potential
Cut the layer into rings about the line OA. A ring at polar angle β has mass dm = M sin β dβ / 2 and is at distance l from A, where l² = a² + r² − 2ar cos β. Differentiating gives l dl = a r sin β dβ. The potential of the ring at A is −γ dm / l, so
The result does not depend on r. The potential is the same at every point inside the layer, so the field, which is minus the slope of the potential, is zero. This needs a small integral, but no cones.
Which other methods apply?
Adding up the pull of each patch directly needs a hard integral, and the cone method avoids it. Energy and graphs do not find the force here. So two methods solve the problem: the cone pairs, which are the fastest and show why, and the potential. The ring sum in the numerical check is the most reliable.
What each slider changes
| Control | Changes the zero? | What it changes |
|---|---|---|
| Distance r of A (header slider too) | No, the net pull stays 0 | Where A is, both patch distances, both patch masses, the pull of each patch, all graphs and dots. |
| Direction θ of the cones | No | The cones, the patch distances, the angle φ, the single pulls and the dots. |
| Radius a | No | The distances in metres, the patch masses and the single pulls. |
| Mass M of the layer | No | The patch masses and the single pulls. |
| Mass m of A | No | The single pulls only. |
| Width ε of the cones | No | The size of the patches, their masses and the single pulls. |
Parameters
| Quantity | Symbol | Value used |
|---|---|---|
| Radius of the layer | a | Mm |
| Mass of the layer | M | × 1024 kg |
| Mass of A | m | kg |
| Distance of A from the centre | r | a |
| Pull from each patch | dF | N |
| Potential inside the layer | φ | J/kg |
Exam traps
- Saying the near patch pulls harder because it is closer. It has far less mass, and the two effects cancel exactly.
- Thinking the pull is zero outside too. Outside the layer the pull is γMm/r², toward the centre, as if all the mass were at the centre (third graph).
- Forgetting that the proof needs a layer of equal thickness. A layer thicker on one side pulls A toward the thicker side.
- Using the layer's thickness in the proof. The layer is thin, so the argument works for each thin layer, and a thick hollow ball is a pile of thin layers.
- Saying the field is zero because the potential is zero. The potential is not zero. It is constant, equal to −γM/a, and a constant potential has no slope.
- Dropping cos φ. It is the same at both ends, so it cancels in the comparison, but it makes each single pull larger than γMm dΩ/(4πa²).