Matthias Beyer

Dr. Dirk Lucas

Head Computational Fluid Dynamics
Phone: +49 351 260 2047

Experimental investigation of stratified two-phase flows


During the transport of multi-phase mixtures in horizontal or slightly inclined channels, mainly stratified flow regimes are generated, as in chemical industrial facilities, in oil production installations or in power plants. These multi-phase flows are difficult to control and can strongly influence the operation or efficiency of the system they flow through. A slug flow may for example affect the mechanical integrity of pipelines or valves because of the high pressure surges they may induce.

In order to simulate these complex flow conditions, the so-called computational fluid dynamics codes (CFD) are currently under development for two-phase flow applications. For this purpose, new models are implemented in CFD that must be checked against experiments. The aim of our experimental investigations of stratified two-phase flows is mainly to deliver high resolution data that is needed for the validation of CFD codes.

The Horizontal Air/Water Channel (HAWAC)

Horizontal air/water channel HAWAC

Fig. 1: Schematic view of the flow channel HAWAC

The HAWAC (Fig. 1) is devoted to co-current flow experiments. The 8 m long test-section has a rectangular cross-section of 100 x 30 mm² (height x width), leading to a length-to-height ratio L/h = 80. A special inlet device (Fig. 2) was designed to provide defined boundary conditions at the channel inlet. Therefore, air and water are injected separately into the test-section: the air flows through the upper part and the water through the lower part of the inlet device. In order to provide homogenous velocity profiles at the test-section inlet, 4 wire mesh filters are mounted in each part of the inlet device. Air and water come in contact at the edge of a 500 mm long blade that divides both phases downstream of the filter segment. The free inlet cross-section for each phase can be controlled by inclining this blade up and down. In this way, the water level at the test-section inlet can be controled. Both, filters and the inclinable blade, provide well-defined inlet boundary conditions for generic investigations of stratified two-phase flows and therefore offer very good code validation possibilities.

Flow field in a slug measured with PIV
Fig. 3: Example of the secondary flow field in a slug

Measurement techniques

The channel is designed for the application of optical measurement techniques, which deliver the high resolution required for CFD validation. Therefore, the acrylic glass test-section with rectangular cross-section provides good observation possibilities.


High-speed video observation was applied from the side of the channel. An algorithm was developed to recognise the interface in the images. This allows the extraction of quantitative values from the images (e.g. the water level in any cross-section). Further statistical data evaluation methods show the interface structure. The results were used for comparisons between experiments and CFD.

Further possibilities

It is possible to use a PIV (particle image velocimetry) system for velocity measurements in the water flow (Fig. 3). Furthermore, pressure sensors should be installed for the investigation of slug flow frequencies.


HAWAC flow map

The maximum superficial velocities achieved in the test-section are 2 m/s for water and 8 m/s for air. A flow map (Fig. 4) was established on the basis of visual observations of the flow structure at different combinations of the gas and liquid superficial velocities. The observed flow patterns are: stratified flow, wavy flow, elongated bubble flow and slug flow.

HAWAC flow map

Fig. 4: HAWAC flow map

Slug flow experiments

High-speed video observation was applied during slug flow. The camera images show the generation of slug flow (Fig. 5). Immediately after the inlet a stratified flow is observed with slight waves generated by the high air velocity. One of these waves grows rapidly and develops into a slug which travels along the channel. At the slug front an important droplet entrainment is visible which is driven by the air flow through the gap on top of the slug.

Picture sequence of slug generation

Fig. 5: Picture sequence of slug generation

The hydraulic jump in a closed channel

At high water flow rates, especially when the inlet blade is inclined down, a hydraulic jump can be occured in the test-section. The hydraulic jump is a discontinuous transition between super- and subcritical flow and is characterised by a steep rising of the water surface. From high-speed video observations and the interface capture algorithm the probability distribution of the water levels was calculated in each vertical cross-section and was ploted in a picture of the test section (Fig. 6). This shows both the structure and the dynamics of the interface.

Probability distribution in a hydraulic jump

Fig. 6: Visualization of the probability distribution of the water level measured in a hydraulic jump

Moreover, experiments were performed to investigate the influence of the air flow rate on the hydraulic jump in a closed channel. These experimental data obtained on a stationary phenomenon involving high turbulence by variation of the momentum exchange between the phases makes the hydraulic jump in a closed channel to a sensitive code validation case.

Code benchmark

The Horizontal Air/Water Channel is:

  • a OECD-NEA Benchmark test facility and
  • a reference test facility for the German CFD-network program.

In the frame of these code benchmark programs, the experimental results obtained at the HAWAC facility are made available for partners.

Selected references

Related links


This work is based on a scientific project funded by the German Federal Ministry of Economics and Energy, support code: 150 1329. The authors assume the responsibility for the content.



Matthias Beyer

Dr. Dirk Lucas

Head Computational Fluid Dynamics
Phone: +49 351 260 2047