Explicit Heat Scheme: Stability Explained with Worked Examples

For , the explicit forward-time central-space scheme is stable only if .

Write 2x or 2*x to multiply, ^ for powers, and ; between simultaneous equations. You can also type factor, expand, derivative of, integrate … from a to b, limit … as x->0 or 15% of 80.

Try an example

Before you start

This page is easier if you are comfortable with:

What Explicit heat scheme: stability means

Each new value is : a weighted average, provided . Beyond the weights are negative and rounding errors grow at each step. With , , : , stable. The largest step is , and gives : unstable. The equation is in the heat equation.

Because is tied to , halving the grid spacing needs four times as many time steps. For , : . A Fourier-mode analysis (separation of variables) shows the amplification factor is , which has size at most 1 only if .

The restriction is much stricter than the CFL condition and is why implicit methods such as Crank–Nicolson are used for diffusion. The step logic is that of Euler stability, and the spatial part is the central difference.

What is it used for?

The explicit scheme for diffusion is stable only for small time steps tied to the square of the grid spacing.

How to use explicit heat scheme: stability

  1. Find , and .
  2. Compute .
  3. Check .
  4. If not, reduce to or less.

Graph

The ratio r = Δt / 0.01 with the stability limit 0.50.004: stable0.006: unstable00.0010.0020.0030.0040.0050.0060.0070.00800.10.20.30.40.50.60.70.80.9The ratio r = Δt / 0.01 with the stability limit 0.5Δtrr = kΔt/Δx²limit r = 1/2Graph with x from 0 to 0.008 and y from 0 to 0.9. Shows the curve r = kΔt/Δx²; the curve limit r = 1/2; point 0.004: stable at (0.004, 0.4); point 0.006: unstable at (0.006, 0.6).

The largest safe time step is 0.005.

Worked examples

These examples are worked by hand, step by step, so you can compare your own working.

Example 1: a stable step

Find for , , .

Worked example
Answer
0.4 (stable)0.4\ \text{(stable)}
  1. Formula

Example 2: an unstable step

Find for with the same data.

Worked example
Answer
0.6 (unstable)0.6\ \text{(unstable)}
  1. Formula

Example 3: largest time step

Find the largest for , .

Worked example
Answer
0.1250.125
  1. Formula

Special cases

A few situations come up again and again. Knowing them saves time and helps you spot answers that cannot be right.

Parameter

r = αΔt/Δx²

α = 1, Δx = 0.1, Δt = 0.004 gives 0.4.

Stability

r ≤ 1/2

FTCS scheme.

Largest Δt

Δt ≤ Δx²/(2α) = 0.005

For Δx = 0.1.

One step

u = (0, 1, 0), r = 0.4: centre becomes 1 − 2r = 0.2

Heat spreads.

Word problems

The explicit (FTCS) scheme uⱼⁿ⁺¹ = uⱼⁿ + r(uⱼ₊₁ⁿ − 2uⱼⁿ + uⱼ₋₁ⁿ) with r = αΔt/Δx² is stable iff r ≤ 1/2. Halving Δx requires quartering Δt.

The parameter

α = 1, Δx = 0.1, Δt = 0.004. Find r.

  1. 0.004 ÷ 0.01
  2. Check: 0.4 ≤ 0.5 ✓0.4.

Largest step

Find the largest stable Δt.

  1. Δx²/(2α)
  2. Check: ✓0.005.

One step Higher

From (0, 1, 0) with r = 0.4 find the centre value after one step.

  1. 1 + 0.4(0 − 2 + 0)
  2. Check: 0.2 ✓0.2.

Unstable r Higher

Find the centre after one step with r = 0.6.

  1. 1 − 1.2
  2. Check: −0.2 (negative, unphysical) ✓−0.2.

Common mistakes

Using Δt ≤ Δx

wave condition ✗

Δt ≤ Δx²/(2α) ✓

Diffusion is quadratic in Δx.

r ≤ 1

r = 0.8 stable ✗

r ≤ 1/2 ✓

Strict limit is 1/2.

Forgetting α

r = Δt/Δx² ✗

r = αΔt/Δx² ✓

Include diffusivity.

Halving Δx and Δt only halves

Δt ÷ 2 ✗

Δt ÷ 4 ✓

Δx² scaling.

Exam tips

  1. r = αΔt/Δx².
  2. r ≤ 1/2.
  3. Halve Δx → ÷4 Δt.
  4. Negative values signal instability.
  5. Crank–Nicolson avoids this.

Practice questions

Try each one on paper, then open the answer. Use the solver above to see the full working.

1. r for α = 1, Δx = 0.2, Δt = 0.01 Foundation

0.25

2. Largest Δt for Δx = 0.2, α = 1 Foundation

0.02

3. Centre after one step, r = 0.25, (0, 1, 0) Foundation

0.5

4. r for α = 2, Δx = 0.1, Δt = 0.002 Higher

0.4

5. Largest Δt for α = 2, Δx = 0.1 Higher

0.0025

6. Centre with r = 0.5 Higher

0

Practice generator

Unlimited fresh questions with instant marking. Choose a level, work it out on paper, then type your answer.

Solve: …

Key terms

FTCS scheme
Forward time, centred space.
Heat equation
u_t = αu_xx.
Diffusivity (α)
Rate of spreading.
Stability limit
r ≤ 1/2.
Time step (Δt)
Increment in time.
Grid spacing (Δx)
Increment in space.

Frequently asked questions

When is the explicit heat scheme stable?

When r = αΔt/Δx² ≤ 1/2.

What if Δx is halved?

Δt must be quartered.

What goes wrong if r > 1/2?

Oscillations grow.

What is the alternative?

Implicit schemes such as Crank–Nicolson.

Which page covers Crank–Nicolson?

Crank–Nicolson scheme.

Which page covers the CFL condition?

CFL condition.

More in Numerical methods and optimisation and the formula sheets.