# Radiation and natural convection in a sealed double-glazed cavity: how do you set it up?

**URL:** <https://www.simscale.com/forum/t/radiation-and-natural-convection-in-a-sealed-double-glazed-cavity-how-do-you-set-it-up/101398>\
**Category:** Fluid Flow / CFD\
**Created:** [October 21, 2026, 12:10pm UTC](https://www.simscale.com/forum/t/radiation-and-natural-convection-in-a-sealed-double-glazed-cavity-how-do-you-set-it-up/101398 "2026-10-21T12:10:50Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![marcomaesgs1](https://www.simscale.com/forum/letter_avatar/marcomaesgs1/32/5_5575768a8748004e209b776fc1b2916d.png) [@marcomaesgs1](https://www.simscale.com/forum/u/marcomaesgs1)\
**Post date:** [October 21, 2026, 12:10pm UTC](https://www.simscale.com/forum/t/radiation-and-natural-convection-in-a-sealed-double-glazed-cavity-how-do-you-set-it-up/101398/1 "2026-10-21T12:10:51Z")

</div>

Hi all,

I’m modeling heat transfer through a double-glazed unit, about [24 mm] overall with a [12 mm] gas-filled cavity, to estimate center-of-glass heat flux. I’d like to capture both natural convection in the cavity and radiation between the panes.

Here’s what I’ve done so far:

• Conjugate heat transfer with a [2D / 3D] model of the glass and gas layers

• Boundary conditions: [interior temp] / [exterior temp] with convective coefficients of [X] and [Y] W/m²K

• Radiation: [not enabled yet / surface-to-surface with emissivity 0.84 on glass and 0.04 on the low-e coating]

My results seem too high compared with a hand calculation, and I think I’m handling radiation wrong. Specifically:

• Is surface-to-surface the right choice for a thin cavity like this, or is there a simpler approach?

• Any advice on meshing the boundary layer in a narrow gap, so the convection cells are resolved without the mesh count exploding?

I’m happy to share screenshots of the setup. Thanks!