Steady Flow Past Tube Bundles

The experiments were carried out in a purpose-designed tunnel with water at 20^oC used as the working fluid. The maximum flowrate was 290 lt/min, corresponding to an upstream bulk velocity, \(U_{bulk}\), of 0.93 m/s. Three different tube bundle configurations were tested, as shown in figure 1: a 3.6×1.6 staggered, a 3.6×2.1 staggered and a 3.6×2.1 in-line array. The two staggered arrays were employed to study the effect of tube spacing. The transverse pitch ratio in these was 3.6 and the longitudinal ones were 1.6 and 2.1 respectively; the bundle consisted of 6 transverse rows of rods. The in-line array consisted of 5 transverse rows with transverse and longitudinal pitch ratios of 3.6 and 2.1 respectively, and was investigated in order to study the effect of tube arrangement.

Fig. 1: Tube bundle configurations

All test sections were made of transparent cast acrylic with cast acrylic rods, 10 mm in diameter and had a square cross-section (\(72 \times 72\) mm). The length to diameter ratio of the rods was 7.2. The rods were rigidly mounted on the walls. Half rods were fixed on the side walls to simulate an infinite tube bundle and minimise boundary layer effects.

Ensemble-averaged and time resolved velocity measurements were obtained using Laser Doppler Anemometry. A single component, fringe-type laser Doppler velocimeter operating in forward-scatter mode was employed. The diameter and length of the control volume were 48.8 \(\mu\)m and 466 \(\mu\)m respectively. The Doppler signals were processed using a frequency counter.

The flow approaching the bundles was uniform and, away from the side walls, two-dimensional. Detailed ensemble-averaged measurements of the axial and transverse velocity components were taken in all three configurations at a Reynolds number, \(Re_{g,d}\), of 12,858. This Reynolds number is based on the gap velocity and the cylinder diameter. The third component could only be measured in a few locations due to restricted optical access. All measurements were taken on the plane of symmetry (\(z/d = 0\)) and in the region \(0.0 \le y/d \le 1.8\) since the flow was found to be symmetrical with respect to the \(x\)-axis. Detailed measurements were also made in the tube wakes for a range of \(Re_{g,d}\) in order to estimate recirculation parameters.

Time-resolved measurements were taken in selected locations in the 3.6×1.6 staggered configuration in order to identify flow instabilities and characterise fully the turbulence structure. Instantaneous values of the axial and transverse velocity components were recorded at regular time intervals over a period of time. Sampling rates in the range of 1 to 4 kHz were employed and blocks of 6000 velocity data points were acquired at each location. The accuracy of the mean velocity and turbulence level measurements was calculated as 1-5% and 5-10% respectively, with the higher errors occurring in regions of steep velocity gradients.

Profiles of the axial (\(U\)) and transverse (\(V\)) mean velocities, and corresponding rms velocities \(u'\) and \(v'\), are available at a number of \(x/d\) locations for the three geometries shown above. For the 3.6×2.1 staggered array skewness and kurtosis data are also available.

Sample plots of selected quantities are available.

Compressed archives of the files can be downloaded from the links below, or files can be retrieved individually from the tables.

• sftb-allfiles.zip
• sftb-allfiles.tar.gz

Balabani et al (1994) report predictions of the flow through staggered arrays using a standard \(k\)-\(\varepsilon\) model both with and without curvature corrections.

The 3.6×1.6 staggered and 3.6×2.1 in-line arrays have been modelled using \(k\)-\(\varepsilon\) models as part of a European Commission Joule project. A number of model modifications were tested, and the details and results can be found in Bouris (1996).

The experimental work was supported financially by the European Commission, and is described in the doctoral thesis and other publications below.

1. Balabani, S., Bergeles, G., Burry, D., Yianneskis, M. (1994). Velocity characteristics of the crossflow over tube bundles. Proc. 7th Int. Symposium Applications of Laser Techniques to Fluid Mechanics, Lisbon, Portugal.
2. Balabani, S. (1996). An experimental investigation of the crossflow over tube bundles. Ph.D. Thesis, King's College London, University of London.
3. Balabani, S., Yianneskis, M. (1996). An experimental study of the mean flow and turbulence structure of cross-flow over tube bundles. Proc. IMechE, Part C, Vol. 210, pp. 317-331.
4. Balabani, S., Yianneskis, M. (1997). Vortex shedding and turbulence scales in staggered tube bundle flows. Can. J. Chem. Engng., Vol. 75, pp. 823-831.
5. Bergeles, G., Bouris, D., Yianneskis, M., Balabani, S., Kravaritis, A., Itskos, S. (1997). Effects of fouling on the efficiency of heat exchangers in lignite utility boilers. Applied Thermal Engineering, Vol. 17, pp. 739-749.
6. Bouris, D. (1996). Numerical investigation of the flow, temperature field and fouling in heat exchangers. Ph.D. Thesis, Department of Mechanical Engineering, National Technical University of Athens (NTUA).

Indexed data: