Heat transfer

From UMaine SECS Numerical Modeling Laboratory
Revision as of 19:16, 12 March 2019 by Mariama (talk | contribs)
Jump to: navigation, search

Heat Transfer in a Lava Lamp

When you flip on the light in a lava lamp, heat from the bulb works slowly to warm up the wax (or ‘lava’). This kind of heat transfer is called conduction, or movement of heat from one place to another. Heat energy flows from high heat (light bulb) to low heat (wax), and the gradient is the driver of conductive heat transfer.

Image: 220 pixels Image: 220 pixels Image: 190 pixels Image: 170 pixels


Until things start moving, conduction is the dominant heat transfer process in our lava lamp system.

  • Conduction, defined as κ∇²T depends on:
    • thermal diffusivity of the material through which heat is moving (κ)
    • thermal gradient (∇²T)

The density of the wax is inversely related to temperature, meaning that at higher temperatures, the wax is less dense. When the density becomes lower than the density of the liquid in the lava lamp, the wax will tend to float upward.

Image: 215 pixels Image: 200 pixels Image: 180 pixels

Wax reaches critical density (less than liquid) and tends to float upward. That movement of wax, and thus heat, introduces velocity into the system and is a process that is referred to as advection.

  • Advection, defined as v∇T depends on:
    • velocity of the fluid (v)
    • thermal gradient (∇T)


Heat Transfer Equation:

Heat transfer within the lava lamp can be represented by an equation. The heat transfer equation brings these aforementioned processes together to describe how heat changes through time (∂T/∂t). It is defined as:

 ∂T/∂t = κ∇²T + v∇T + A
 ∂T/∂t = conduction + advection + production

The production term (A) represents internally generated heat energy and we do not consider it in this lava lamp scenario. It becomes important in systems such as glaciers, when internal strain heating is an important source of heat.

Peclét Number:

The Peclét Number is a dimensionless number which indicates whether conduction or advection dominates the system.

 Pe = uL/κ

where u = flow velocity, L = characteristic length Height of lava lamp, κ = thermal diffusivity

  • Low peclet regime- dominated by heat transport by conduction
  • High peclet regime- dominated by heat transport by advection

With the movement of wax in the upward direction away from the heat source at the bottom of the container and into a cooler material the wax density increases again (more than liquid). This wax density increase causes the wax to sink back down to the bottom of the container towards the heat source. This process is called convection.

Image: 230 pixels Image: 180 pixels


Rayleigh Number:

The Rayleigh Number is a dimensionless number used to describe whether convection grows or decays.

 Ra ≡ (gρΔTαd³)/μκ
  • Convection depends on driving forces and resisting forces.
    • Driving forces (when these terms are dominant convection grows):
      • Acceleration due to gravity (g)
      • Density (ρ)
      • The temperature difference between the bottom and top of the convection cell (ΔT)
      • The volume coefficient of thermal expansion (α)
      • The height of the convection cell (d)
    • Resisting forces (when these terms are dominant convection decays):
      • Viscosity (μ)
      • Thermal diffusivity (κ)

At some critical Rayleigh (Racr) number (dependent on the system), the convection regime shifts.

Heat Transfer in Ice Stream Shut Down and Start (based on Joughin and Alley, 2011)

Stop and start of ice streams is dependent on:

  • Thinning (steepens basal temperature gradients)
  • Basal freezing (puts on the brakes for a system)
  • Thickening (build up traps geothermal heat)
  • Ice stream activation & the cycle is repeated

Watch the video below to walk through the steps of ice stream shut-down and speed-up, and see how this relates to the heat transfer equation introduced above.

Further reading:

Different theories exist for ice stream shut down, and here we present varying hypothesis. Please note that the behavior we model in our video is based upon the hypothesis of Joughin and Alley (2011).