0
Research Papers: Evaporation, Boiling, and Condensation

# Flow Boiling in Minichannels Under Normal, Hyper-, and Microgravity: Local Heat Transfer Analysis Using Inverse Methods

[+] Author and Article Information
Sébastien Luciani

Ecole Polytechnique Universitaire de Marseille, Technopôle de Château-Gombert, 5, rue Enrico Fermi, 13453 Marseille, Francesebastien.luciani@polytech.univ-mrs.fr

David Brutin, Christophe Le Niliot, Ouamar Rahli, Lounès Tadrist

Ecole Polytechnique Universitaire de Marseille, Technopôle de Château-Gombert, 5, rue Enrico Fermi, 13453 Marseille, France

J. Heat Transfer 130(10), 101502 (Aug 08, 2008) (13 pages) doi:10.1115/1.2953306 History: Received August 20, 2007; Revised March 18, 2008; Published August 08, 2008

## Abstract

Boiling in microchannels is a very efficient mode of heat transfer since high heat and mass transfer coefficients are achieved. Here, the objective is to provide basic knowledge on the systems of biphasic cooling in mini- and microchannels during hyper- and microgravity. The experimental activities are performed in the frame of the MAP Boiling project founded by ESA. Analysis using inverse methods allows us to estimate local flow boiling heat transfers in the minichannels. To observe the influence of gravity level on the fluid flow and to take data measurements, an experimental setup is designed with two identical channels: one for the visualization and the other one for the data acquisition. These two devices enable us to study the influence of gravity on the temperature and pressure measurements. The two minichannels are modeled as a rectangular rod made up of three materials: a layer of polycarbonate $(λ=0.2Wm−1K−1)$ used as an insulator, a cement rod $(λ=0.83Wm−1K−1)$ instrumented with 21 $K$-type thermocouples, and in the middle a layer of Inconel® $(λ=10.8Wm−1K−1)$ in which the minichannel is engraved. Pressure and temperature measurements are carried out simultaneously at various levels of the minichannel. Above the channel, we have a set of temperature and pressure gauges and inside the cement rods, five heating wires provide a power of $11W$. The $K$-type thermocouple sensors enable us to acquire the temperature in various locations ($x$, $y$, and $z$) of the device. With these temperatures and the knowledge of the boundary conditions, we are able to solve the problem using inverse methods and obtain local heat fluxes and local surface temperatures on several locations. The experiments are conducted with HFE-7100 as this fluid has a low boiling temperature at the cabin pressure on Board A300. We applied for each experiment a constant heat flux $(Qw=33kWm−2)$ for the PF52 campaigns (Parabolic Flights). The mass flow rate varies in the range of $1 and the fluid saturation temperature $(Tsat)$ is $54°C$ at $Psat=820mbars$.

<>

## Figures

Figure 1

Coupling of the two rods used during parabolic flights (the left minichannel is used for measurements and the right minichannel for visualization). The thermocouples are located along the minichannel and the pressure gauges on top of it.

Figure 2

Top view of the minichannel cross-sectional components. The minichannels are engraved in the Inconel plate. The gravity direction is opposite to x.

Figure 3

Front view of the minichannel cross-sectional components. We can see the five heating wires and the several thermocouples weld represented in circles.

Figure 4

Doubling of the sensors. We copy the symmetrical weld thermocouple trough the x-axis and pass from 21 thermocouples to 42.

Figure 5

Influence of the thermocouples’ noise measurements on the predicted wall

Figure 6

Influence of the error estimation induced by the thermocouples’ locations on the surface temperature

Figure 7

2D meshing of the minichannel—the three domains are meshed but due to the adiabatic condition on the polycarbonate; we work with only two domains, the cement rod and the Inconel where the minichannel is engraved

Figure 8

3D Inconel meshing—only the faces are meshed with the BEM. All the elements are connected with each other. The minichannel is divided into seven parts in order to have good correspondence with three thermocouples located in the cement rod. We have then for each mesh, three thermocouples to have significant gradient to solve the inverse problem.

Figure 9

The problem of the unknown boundary conditions. We have N′ internal points (42 thermocouples) and we want to obtain the temperature and the heat flux density on the minichannel boundary where we have one equation and two unknowns.

Figure 10

The L-curve approach applied to the BEM. The optimal value is in the bend of the L where the best compromise is between stable results and low residuals (on the distinct corner separating the vertical and the horizontal part of the curve). It is around this corner at the maximum curvature that we find the best compromise.

Figure 11

Boundary conditions. According to our study, we consider only two domains: the cement and the Inconel. We made this choice simply because the polycarbonate had no influence on the studied zone (adiabatic condition). A variation in the different coefficients on the edge of the device does not change the output temperature profiles.

Figure 12

Grouping of the sensors by section. As we work on different sections of the bar, it is necessary that they contain the inversed information. However, the cutting cannot contain the real positions of the thermocouples and the temperatures. An additional step is required to bring the sensors to the edges of the cutout sections. Their grouping is done by translating the three closer (and their symmetrical) next to the section.

Figure 13

Profile of the heat transfer coefficient h and the wall temperature along the minichannel obtained with a 2D model (Qw=32kWm−2, Qm=0.45gs−1, χ=0.66). The sharp decrease at the inlet and the outlet of the minichannel is due to the edge effects, which are not taken into account in the 2D model.

Figure 14

3D temperature distribution inside the cement rod. The temperature decreases on top in the cement surface because the heat is pumped directly to the minichannel.

Figure 15

Local wall temperature and local heat flux along the minichannel (Qw=32kWm−2, Qm=0.26gs−1, χ=0.26) depending on the gravity level

Figure 16

Local heat transfer coefficient as a function of the main flow axis (Qw=32kWm−2, Qm=0.26gs−1, χ=0.26)

Figure 17

Local vapor quality as a function of the main flow axis (Qw=32kWm−2) depending on the mass flow rate. We plot the curves for seven points. The profiles increase from the inlet to the outlet minichannel. We observe that the final point corresponds to the outlet vapor quality, which is a way to validate our results. The error estimation is here around 2%.

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections