A simple python implementation of thermodynamic models for dropwise condensation.
This Jupyter notebook provides some examples for the plotting functions. Please refer to the corresponding GitHub repository for additional information.
import DWC_plotting as DWC # import plotting functions
Defining the input parameters for the calculation. The plots below can easily be changed by adjusting the input parameters and rerunning the script. Please refer to the CoolProp documentation for a list of supported fluids.
inputParameters = {"medium":"Water", # choose the condensing fluid
"p_steam":337.8, # pressure in mbar
"deltaT_sub":5, # surface subcooling temperature in K
"Theta":90, # contact angle in deg
"CAH":10, # contact angle hysteresis in deg
"k_coat":0.2, # thermal conductivity of the coating in W/(mK)
"delta_coat":0.000001, # thickness of the coating in m
"h_i":15.7, # interfacial heat transfer coefficient in MW/m²K, if no value is given, h_i is calculated
"c":1, # numerical constant, "depends on the shape of the drop and on the steepness of the substrate surface"
"N_s":250} # number of Nucleation sites in 10^9 1/m²
You can plot the heat flux vs. the surface subcooling temperature:
fig = DWC.plot_q_deltaTsub(inputParameters)
For a simple parameter variation, this function also takes some optional arguments:
fig = DWC.plot_q_deltaTsub(inputParameters, theta=[90, 120, 150]) # plotting q vs. delta T for different contact angles
instead of theta = [Value 1, Value 2, ...] for a variation of the static contact angle, the following optional arguments are also possible:
Additionally, the heat flux for filmwise condensation on a vertical wall can be plotted for comparison by setting filmwise=True
. The height of the wall can be specified with the variable h_fw (in meter, default value: h_fw=0.1
. For example:
fig = DWC.plot_q_deltaTsub(inputParameters, medium=["Water", "Ammonia"], filmwise=True, h_fw=0.2)
You can also plot the rate of heat flow and the heat flux through a single droplet...
fig = DWC.plot_qdrop_theta_r(inputParameters, radii = [0.000001, 0.000005, 0.000010])
... and the various thermal resistances of a single droplet vs. the droplet radius:
fig = DWC.plot_Rdrop(inputParameters)
The drop size distribution can be plotted with the following funtion. Currently, three optional parameters are possible:
fig = DWC.plot_Nr_r(inputParameters, theta = [90, 120, 150])
the most important calculated values are given by the following function:
DWC.print_results(inputParameters)
fluid properties: fluid: Water T_sat: 71.84774592128895 °C sigma: 64.20573774100335 mN/m h_fg: 2328.439485181256 kJ/kg rho_l: 976.6705040387072 kg/m³ rho_g: 0.2137643354220767 kg/m³ R_s: 461.52295930318655 J/(kg*K) interfacial heat transfer coefficient: h_i: 15700000.0 W/m²K h_i_calc: 6717136.6 W/m²K results: q: 163729.00061053934 W/m² q_n: 22174.572734600122 W/m² q_N: 141554.4278759392 W/m² q_N/q: 86.5 % r_min: 3.8961625003973076e-09 m r_e: 1e-06 m r_max: 0.0010561535753410094 m misc.: {'Bo': 0.166455207325625, 'h_i': 15700000.0, 'k_c': 0.661171503942419} model used: KimKim2011