Experimental analysis of local void fractions measurements for boiling hydrocarbons in complex geometry

Abstract

The present paper is part of a research program on two-phase flows and heat transfer studies in tube bundles. An experimental study was carried out to analyse the void fraction for vertical two-phase flows. Boiling across a horizontal tube bundle for three hydrocarbons (n-pentane, propane and iso-butane) under saturated conditions is investigated. The experiments were performed on a tube bundle with 45 plain copper tubes of 19.05 mm outside diameter in a staggered configuration with a pitch to diameter ratio of 1.33. An optical probe has been developed to determine the local void fraction at the minimum cross section between the tubes. The void fraction for the three hydrocarbons was found to be significantly less than the homogeneous prediction, and correlated well with the Zuber and Findlay (1965) model. A statistical analysis using Probability Density Functions (PDF) was performed to characterise the flow regimes. Two different flow regimes have been identified in the bundle, namely a "bubbly" and an "annular-dispersed" flow. A transition zone was evidenced by analysing the bubble size distribution. A comparison with the only preexisting map, established for adiabatic air-water flow, shows that the transitions between the flow regimes do not correspond to our results in boiling case.

Investigation of flow phenomena in a kettle reboiler

“…The proposed transitions are close to the bubbly-intermittent boundary of Ulbrich and Mewes [18] and significantly within the intermittent regime of Noghrehkar et al [19]. The data and flow regime boundaries of Asprin et al [20] and the flow regime boundaries of Noghrehkar et al [19] are shown in Fig. 14b.…”

“…This assumption would be poor in the two-phase region above the tube bundle, or in areas where the mass flux is low, because vapour-liquid separation would be expected there. However, within the tube bundle, where frictional effects are large and the mass flux is high because of the relatively [19] with boundaries of Ulbrich and Mewes [18] and (b) flow map and data of Noghrehkar et al [19] with boundaries of Asprin et al [20]. small gap between the tubes, it is more plausible; particularly when the available experimental evidence is taken into account.…”

“…The one-dimensional model described in Section 2 was used to determine the vertical mass flux in the column. The flow pattern transition from bubbly to intermittent flow is significantly controlled by the superficial vapour velocity [17][18][19][20]. The 1.5DT model assumes that bubbly flow exists until a critical superficial gas velocity is achieved and proceeds as an intermittent flow thereafter.…”

“…They reported that the bubblyintermittent boundary of Ulbrich and Mewes [18] could be in error at large water superficial velocities because visual observations at the front of the heat exchanger did not represent what was going on in the depth of it. Asprin et al [20] also used a voidage probe method. They obtained a flow map for boiling pentane, iso-butane and propane at various pressures.…”

“…Reasons for this could be that different tube bundles and fluids were [20] was obtained using a similar tube bundle with similar fluids to those used in this study. However, the pressures were different.…”