You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
\subsubsection{Specifying the Heat Flux at a Solid Surface}
2295
-
\label{info:net_and_convective_heat_flux}
2296
-
2297
-
Instead of altering the convective heat transfer coefficient, a fixed heat flux may be specified directly. Two methods are available to do this. The first is to specify a \ct{NET_HEAT_FLUX} in units of \unit{kW/m^2}. When this is specified FDS will compute the surface temperature required to ensure that the combined radiative and convective heat flux from the surface is equal to the \ct{NET_HEAT_FLUX}. The second method is to specify the \ct{CONVECTIVE_HEAT_FLUX}, in units of \unit{kW/m^2}. The radiative flux is then determined based on the \ct{EMISSIVITY} on the \ct{SURF} line and the wall temperature needed to get the desired \ct{CONVECTIVE_HEAT_FLUX}. Setting the \ct{EMISSIVITY} to zero will result in there being only a convective heat flux from a surface. If \ct{NET_HEAT_FLUX} or \ct{CONVECTIVE_HEAT_FLUX} is positive, the wall heats up the surrounding gases. If \ct{NET_HEAT_FLUX} or \ct{CONVECTIVE_HEAT_FLUX} is negative, the wall cools the surrounding gases. You cannot specify \ct{TMP_FRONT} with \ct{NET_HEAT_FLUX} since \ct{NET_HEAT_FLUX} combines radiative and convective flux, and FDS must compute the wall temperature to get the desired flux. Specifying \ct{TMP_FRONT} and \ct{CONVECTIVE_HEAT_FLUX} will result in FDS using the gas temperature adjacent to a wall cell to set the convection heat transfer coefficient to achieve the desired \ct{CONVECTIVE_HEAT_FLUX} for that wall cell. If a droplet lands surface with a specified heat flux, no heat transfer will be predicted between the droplet and the surface. This is because the high heat transfer rates associated with droplets can lead to nonsensical wall temperatures when trying to enforce the specific flux which are then likely to result in runtime errors or numerical instabilities.
2298
-
2299
2294
\subsubsection{Logarithmic Law of the Wall}
2300
2295
2301
2296
Near-wall treatments, such as wall models or wall functions, aim to mimic the sudden change from molecular to turbulent transport close to the walls using algebraic formulations without the need of resolving the otherwise computationally expensive region of the near-wall flow-field. The main theory follows dimensional analysis based on the idea that shear at the wall is constant. Accordingly, the non-dimensional velocity $u^+$ is calculated using a wall function \cite{FDS_Tech_Guide}.
where $h$ is the unadjusted heat transfer coefficient, $\dot{m}''$ is the mass flow rate per unit area (positive for blowing), and $c_p$ is the specific heat of the gas.,
2317
2312
2313
+
2314
+
\subsubsection{Specifying the Heat Flux at a Solid Surface}
2315
+
\label{info:net_and_convective_heat_flux}
2316
+
2317
+
Instead of altering the convective heat transfer coefficient, a fixed heat flux may be specified directly. Two methods are available to do this. The first is to specify a \ct{NET_HEAT_FLUX} in units of \unit{kW/m^2}. When this is specified FDS will compute the surface temperature required to ensure that the combined radiative and convective heat flux from the surface is equal to the \ct{NET_HEAT_FLUX}. The second method is to specify the \ct{CONVECTIVE_HEAT_FLUX}, in units of \unit{kW/m^2}. The radiative flux is then determined based on the \ct{EMISSIVITY} on the \ct{SURF} line and the wall temperature needed to get the desired \ct{CONVECTIVE_HEAT_FLUX}. Setting the \ct{EMISSIVITY} to zero will result in there being only a convective heat flux from a surface. If \ct{NET_HEAT_FLUX} or \ct{CONVECTIVE_HEAT_FLUX} is positive, the wall heats up the surrounding gases. If \ct{NET_HEAT_FLUX} or \ct{CONVECTIVE_HEAT_FLUX} is negative, the wall cools the surrounding gases. You cannot specify \ct{TMP_FRONT} with \ct{NET_HEAT_FLUX} since \ct{NET_HEAT_FLUX} combines radiative and convective flux, and FDS must compute the wall temperature to get the desired flux. Specifying \ct{TMP_FRONT} and \ct{CONVECTIVE_HEAT_FLUX} will result in FDS using the gas temperature adjacent to a wall cell to set the convection heat transfer coefficient to achieve the desired \ct{CONVECTIVE_HEAT_FLUX} for that wall cell. If a droplet lands surface with a specified heat flux, no heat transfer will be predicted between the droplet and the surface. This is because the high heat transfer rates associated with droplets can lead to nonsensical wall temperatures when trying to enforce the specific flux which are then likely to result in runtime errors or numerical instabilities.
2318
+
2319
+
The \ct{SURF} parameters \ct{NET_HEAT_FLUX} and \ct{CONVECTIVE_HEAT_FLUX} will not work with a \ct{MATL_ID}. When 1D heat conduction into the solid is considered, you can specify an \ct{EXTERNAL_FLUX} as discussed in Sec.~\ref{info:simulating_the_cone_calorimeter}, which \emph{supplements} the net surface heat flux. But in certain situations (usually for verification tests), you may want to directly control the incident radiative heat flux to the surface. To do this, you can use \ct{EXTERNAL_FLUX} and add \ct{SKIP_INRAD=T} on the \ct{SURF} line. The incoming radiation to the surface is then \emph{replaced} by the specified \ct{EXTERNAL_FLUX}.
2320
+
2318
2321
\subsection{Adiabatic Surfaces}
2319
2322
\label{info:adiabatic}
2320
2323
@@ -3302,9 +3305,8 @@ \subsubsection{Examples}
3302
3305
\paragraph{Case 11} (\ct{box_burn_away11}) The ``box'' in this case consists of 0.4~m by 0.4~m by 1~cm plates forming a collection of 10~cm cubical compartments. There are 15 plates in all, each with a density of 20~kg/m$^3$. The total mass is 0.48~kg.
\caption[Results of the \ct{box_burn_away11} test case]{Output of \ct{box_burn_away11} test case.}
3309
3311
\label{box_burn_away_11}
3310
3312
\end{figure}
@@ -3335,6 +3337,7 @@ \section{Testing a Pyrolysis Model}
3335
3337
The \ct{SURF} and \ct{MATL} lines describing the pyrolysis of real materials consist of a combination of empirical and fundamental properties, often originating from different sources. How do you know that the various property values and the associated thermo-physical model in FDS constitute an appropriate description of the solid? For a full-scale simulation, it is hard to untangle the uncertainties associated with the gas and solid phase routines. However, you can perform a simple check of any set of solid phase model by essentially turning off the gas phase. In the following sections, guidance is provided on how to perform a quick simulation of the cone calorimeter and bench-scale measurements like thermal gravimetric analysis (TGA), differential scanning calorimetry (DSC), and micro-combustion calorimetry (MCC).
3336
3338
3337
3339
\subsection{Simulating the Cone Calorimeter}
3340
+
\label{info:simulating_the_cone_calorimeter}
3338
3341
3339
3342
This section describes how to set up a simple model of the cone calorimeter or similar apparatus. This is not a full 3-D simulation of the apparatus, but rather a 1-D simulation of the solid phase degradation under an imposed external heat flux. While a full 3-D simulation of the cone heater and sample holder can be created in FDS, such a simulation would take some time to complete and would not disentangle issues with the solid phase model from uncertainties in the gas phase. It is worthwhile to perform a quick simulation like the one described here to test the solid phase model only.
0 commit comments