Alf-Erik Almstedt

Influence of pressure and fluidization velocity on the bubble behaviour and gas flow distribution in a fluidized bed

Abstract Measurements of the visible bubble flow rate and the through-flow velocity of gas inside bubbles were made in a pressurized fluidized bed, using capacitance and Pitot-static pressure probes, respectively. The influence of pressure and fluidization velocity on the bubble behaviour and gas flow distribution was studied. The bed, of 0.2 × 0.3 m cross-section, was operated at pressures between 0.1 and 1.6 MPa and at excess gas velocities between 0.1 and 0.6 m/s. The bed material was silica sand with a mean particle diameter of 0.7 mm. The mean bubble frequency, the mean bubble rise velocity, the mean bubble volume fraction and the visible bubble flow rate were found to increase with both increasing pressure and excess gas velocity. The mean pierced length of bubbles increases with increasing excess gas velocity but decreases, after an initial increase, wtih increasing pressure. With both increasing pressure and excess gas velocity, the bubbles tend to redistribute themselves towards the center of the bed cross-section. The pressure probe measurements show that the through-flow velocity of gas inside the bubbles decreases with increasing pressure. Furthermore, gas flow balance calculations show that the ratio of dense phase superficial gas velocity to minimum fluidization velocity increases with both increasing pressure and excess gas velocity. Pressure drop measurements show a decrease in minimum fluidization velocity with increasing pressure, in accordance with predictons from the Ergun equation.

Fluid dynamics of a pressurized fluidized bed : comparison between numerical solutions from two-fluid models and experimental results

A validation of four different two-fluid model closures was carried out to investigate the effect of gas-phase turbulence, drift velocity and three dimensionality on the fluid dynamics of a bubbling fluidized bed. At atmospheric conditions, it is verified that gas-phase turbulence has a negligible effect for the bed material and operating conditions used in the investigation, whereas the validation shows some evidence that the gas-phase turbulence has a significant contribution for higher pressures. The drift velocity shows no noticeable effect on the results at any pressure. A comparison between two- and three-dimensional calculations at atmospheric pressure shows that the three-dimensional effects appear to be considerable.

Hydrodynamics, erosion and heat transfer in a pressurized fluidized bed: influence of pressure, fluidization velocity, particle size and tube bank geometry

Measurements of hydrodynamics, local tube erosion and local instantaneous bed-to-tube heat transfer were carried out in a cold pressurized fluidized bed, with two horizontal tube banks having different tube packings. The influence of pressure, fluidization velocity, particle size and tube bank geometry was studied. Two size distributions of silica sand were used, one with a mean particle diameter of d(p) = 0.7 mm and one with d(p) = 0.45 mm. The bed has a cross-section of 0.2 m x 0.3 m, and was operated at pressures between 0.1 and 1.6 MPa and at excess gas velocities of 0.2 and 0.6 m/s. The results show that, if plotted vs the excess gas velocity, the hydrodynamic behaviour is similar for the two different particle sizes. However, the smaller particles generally give rise to less erosion than the larger particles, as an effect of their momentum being lower at a given particle velocity. The small particles also give a higher heat transfer than the large particles, as a result of a higher particle convection. The hydrodynamic behaviour, erosion levels and local heat transfer differ significantly between the two tube banks. The denser tube bank causes an earlier transition to a turbulent bed behaviour with increasing pressure or fluidization velocity. The dense tube bank gives rise to considerably less erosion but also gives a somewhat lower heat transfer than the more sparse tube bank, at corresponding operating conditions. The tube erosion is strongly related to the bubble rise velocity. The heat transfer coefficient is generally coupled to the bubble frequency, except for the high excess gas velocity with the dense tube bank where, at high pressures, the bed assumes a strongly turbulent behaviour and no distinct bubble pattern exists. The results indicate that the most severe erosion will occur in sparsely packed parts of a tube bank. For the sparse tube bank investigated, at high pressures, the erosion decreases with increasing pressure. The bed-to-tube heat transfer coefficient generally increases with increasing pressure. Thus, it should be favourable to operate a bed at high pressure levels.

Influence of pressure on the minimum fluidization velocity

The influence of pressure on minimum fluidization velocity has been measured in a pressurized fluidized bed. Three different bed materials, corresponding to Geldarts group B or group D particles, were investigated at pressures between 0.1 and 1.6 MPa. The onset of fluidization was determined both visually and by pressure drop measurements. The results show a clear decrease in the minimum fluidization velocity with increasing pressure. The experimental results are in good accord with results obtained using the Ergun equation. A number of simplified correlations derived from the Ergun equation were also investigated. While many of these give a good description of the pressure effect, their accuracy varies significantly for different bed materials.

Simulation of the fluid dynamics of a bubbling fluidized bed. Experimental validation of the two-fluid model and evaluation of a parallel multiblock solver

A mesh refinement study and validation of two-fluid model closures were carried out for a bubbling fluidized bed application. The mesh refinement study indicates that a higher degree of mesh refinement is required for atmospheric than for pressurized fluidization. Statistical bubble parameters (bubble frequency, mean bubble rise velocity, mean pierced bubble length and mean bubble volume fraction) were evaluated. The simulated statistical bubble quantities are computed from voidage signals derived from the transient multidimensional solution of two-fluid models. The algorithm for computing these quantities is taken directly from the evaluation program treating the measurement signals. To remedy the long simulation times required to obtain acceptable statistical values, a parallel version of the two-fluid model solver was developed, based on a domain decomposition method for distributed memory computers. A number of problems related to the parallelization are investigated. These are optimal treatment of velocity components on multi-block boundaries, frequency of data exchange at multi-block boundaries, local errors at multi-block boundaries and simulation time requirements.

Hydrodynamics of a pressurized fluidized bed with horizontal tubes: Influence of pressure, fluidization velocity and tube-bank geometry

Measurements of the bubble hydrodynamics were carried out in a cold pressurized bed with horizontal tubes. The mean bubble rise velocity, the bubble frequency, the mean pierced length, the bubble volume fraction, and the visible bubble flow rate were measured using capacitance probes. The absolute gas velocity through the bubbles was measured using Pitot-static pressure probes. The bed expansion ratio was determined by measuring the pressure difference between the freeboard and the bed at different heights and extrapolating the pressure difference down to zero. The fluctuations in the pressure drop over the entire bed height were also measured, and the power spectral density distribution of these fluctuations was calculated. The influence of pressure, fluidization velocity, and tube-bank geometry on the bubble behaviour and gas-flow distribution were studied. The bed has a cross-section of 0.2 m x 0.3 m. It was operated at pressures between 0.1 and 1.6 MPa, at excess gas velocities of 0.2 and 0.6 m/s. Three different tube-bank geometries were used, one with a fairly dense pitch and two with more sparse configurations, and comparisons are also made with previous results obtained without tubes in the bed. The bed material was silica sand with a mean particle diameter of 0.7 mm. For the three tube banks investigated here, all the measured parameters except the mean pierced length consistently increased with increasing excess gas velocity. The mean pierced length increased with increasing excess gas velocity at low pressures, while the velocity effect at high pressures was less obvious. When the pressure was increased, the mean pierced length first increased to a maximum value, at p = 0.4 MPa for the low excess gas velocity and at p = 0.2 MPa for the high excess gas velocity, then decreased again as the pressure was increased further. The bubble frequency and the bed expansion increased as the pressure was increased. The absolute gas velocity through the bubbles as well as the gas velocity relative to the bubbles decrease as the pressure was increased. For the sparse tube banks at the lower excess gas velocity, the mean bubble rise velocity, the bubble volume fraction and the visible bubble flow rate increased as the pressure was increased. For the dense tube bank, however, these parameters showed a maximum at p = 0.4 MPa. At the higher excess gas velocity, these parameters showed a maximum at about p = 0.2-0.4 MPa for all three tube banks. It appears that, at this velocity, the presence of tubes prevents a further increase in these parameters. This behaviour differs significantly from the behaviour without tubes. From the power spectral density distributions of the pressure fluctuations over the entire bed height and from visual observation, it appears that the bed is slugging, or close to slugging, at atmospheric pressure for all the tube configurations. When the pressure is increased, the power spectral density distribution becomes wider as the large bubbles/slugs break down and the bed moves toward a more dispersed bubbling behaviour.

An investigation of fluidized-bed scaling: heat transfer measurements in a pressurized fluidized-bed combustor and a cold model bed

A method is proposed for predicting heat transfer coefficients in hot fluidized-bed combustors by translating results measured in a scaled-down, cold model bed. Provided the beds are scaled to hydrodynamic similarity, local heat transfer coefficients measured in the cold model bed can be translated into local hot-bed convective coefficients with the aid of existing correlations for the gas and particle convective components. To obtain the total hot-bed-to-surface coefficients, a radiative component is then added. The chief advantages of the proposed method are that existing convective heat transfer correlations can be applied locally in a bed, and that no a priori knowledge of the voidage close to the transfer surface is required. In a previous paper by Almstedt and Zakkay, measurements of the bubble activity in a pressurized fluidized-bed burning coal and in a scaled-down pressurized model bed operating at room temperature showed that a good hydrodynamic similarity can be obtained between the beds. The heat transfer translation method proposed here has been validated by comparing heat transfer coefficients measured in the same two beds, operating under scaled conditions. Average heat transfer coefficients for four different horizontal tube bundles, as well as local coefficients measured with probes in four different positions were compared. The results indicate good agreement between the hot-bed results measured and the results translated from the model bed measurements employing the proposed method. Furthermore, the present paper presents heat transfer measurements from the cold model bed for three different bed materials, at pressures of 0.1, 0.24 and 0.5 MPa and fluidization velocities ranging from 0.15 to 1.3 m/s. The results are in good accordance with existing theory, but indicate that the gas convective component (as well as the particle convective component) is significantly dependent on the fluidization velocity.

Turbulence modification by particles in a horizontal pipe flow

Measurements were made of turbulence intensities and turbulent energy spectra in a fully developed, turbulent air-particle pipe flow. The influence of the particles on the turbulence was studied. Measurements were made with spherical particles and particles with a large aspect ratio (pulp fibres). There is a significant change in turbulence intensity at higher particle concentrations with loading ratios of m = 0.1 and 0.03. The measurements show that the turbulence intensity increases close to the centre of the pipe while the turbulence intensity decreases close to the pipe wall for the spherical particles. These results are in agreement with earlier measurements found in the literature. For the fibres, the turbulence intensity decreases over the whole pipe cross-section. Fibre flocs, however, give variations in the mean velocity that result in the production of turbulence in the lower part of the channel.

Numerical simulation of fluid dynamics in fluidized beds with horizontal heat exchanger tubes

A numerical code, Gemini, based on the implicit multifield method (IMF) of Harlow and Amsden for Eulerian two-fluid modelling, is used to simulate the fluid dynamics of bubbling fluidized beds, assuming no turbulence in the gas or solid phase. The paper gives a formulation of the equations of motion and empirical closure laws in general curvilinear coordinates for calculation of the fluid dynamics in beds with complex internal geometries. A special discretization method for general curvilinear structured grids with multiblock connectivity is implemented, and two-dimensional non-stationary calculations are performed for a bed with a cross-sectional width of 0.3 m, containing two horizontal heat exchanger tubes. The local visible bubble flow and the gas and particle motion around the tubes are briefly discussed and compared with experimental fluid dynamic results at different pressure levels.

Influence of pressure, fluidization velocity and particle size on the hydrodynamics of a freely bubbling fluidized bed

The hydrodynamics have been studied in a cold, freely bubbling, pressurized fluidized bed. The bed has a cross-section of 0.2 m x 0.3 m and was operated at pressures between 0.1 and 1.6 MPa and at excess gas velocities of 0.2 and 0.6 m/s. The bed material was silica sand with a mean particle diameter of d(p) = 0.45 mm. Comparisons were made with previous results obtained with particles of d(p) = 0.7 mm. The hydrodynamic results are similar for the two different particle sizes when plotted vs the excess gas velocity. The results also show that the bed expansion, bubble rise velocity, bubble volume fraction and visible bubble flow rate fall on single curves if plotted vs a dimensionless potentially available drag force, while the bubble frequency, the mean pierced length and the through-flow velocity of gas through the bubble do not. The dimensionless drag force is a suitable scaling parameter as long as the particles do not respond to the gas-phase velocity fluctuations and as long as the dense phase does not expand. At high pressures, an increased gas-particle interaction, in combination with turbulent fluctuations in the gas phase, can be used to explain the increased bubble instability, with a corresponding increased bubble splitting and dense phase expansion. The gas-particle interaction also increases with decreasing particle size, which may help explain the maximum stable bubble size for group A particles observed by many workers.