Date of Award
Doctor of Philosophy (PhD)
The local and global phenomena in oscillatory steam flow reflux condensation were investigated both experimentally and theoretically.
A series of experiments were carried out in a 4.86m long vertical double-pipe condenser consisting of a 1.735cm diameter inner tube enclosed within a 5cm diameter outer tube. Four different inlet steam flow square-wave modulation frequencies: 0 (steady state), 0.05Hz, o.1Hz, and 0.2Hz, and a constant cooling side inlet temperature and flow rate of 30°C and 1kg/min respectively, were investigated. The experiments covered an average steam flow rate range of 10-300g/min, an amplitude range of 0-150g/min, and flow bias levels of 10-300g/min. These experiments were designed to investigate the effects of frequency, amplitude, and bias (steady component) of the inlet steam flow modulation on the observed process.
Five different operating modes were identified in the condenser as the inlet mass flow rate was increased. The operating modes, categorized according to the observed flow regimes in the condenser, were: (1) reflux condensation without water column mode, (2) reflux condensation with water column mode, (3) first carry-over mode, (4) second carry over mode, (5) climbing film flow mode. Complete reflux condensation was maintained only in the first two modes of operation.
Under steady inlet steam flow conditions, it was found that the flooding in the tube caused the system to undergo transition from the first mode of operation to the third mode of operation at very low inlet steam flow rates, thereby severely limiting the condensation capacity in complete reflux condensation. Under modulated steam flow conditions, and depending on the particular combination of frequency, amplitude, and steam flow bias, as much as 300% improvement on the condensation capacity was obtained. In general, improvement in the condensation capacity was favored by low frequency, high amplitude, and low to moderate steam flow bias levels.
Observations of the experimental phenomenon showed that once flooding occurred in the tube under steady inlet steam flow conditions, the resulting net upward condensate flow led to the formation and build-up of single-phase condensate (water column) on top of the condensing region. The back pressure exerted by this water column on the condensing region upstream led to a huge increase in the operating pressure and overall pressure drop across the condenser tube, which was directly reflected in the measured operating pressure. Experimental results showed that substantial reductions in the time-averaged system pressure and pressure drop across the condenser were achieved under modulated steam flow conditions, due to the tendency of the oscillations in the steam flow to destabilize the water column or prevent sustained water column formation and growth. Under modulated steam flow, the system pressure was influenced mostly by the frequency of modulation and the steam flow bias, and to a lesser extent the amplitude of steam flow. These factors were found to play dual roles in the time-averaged overall pressure drop, as this pressure drop was found to play dual roles in the time-averaged pressure drop increased with both the frequency of modulation and the flow bias. Additionally the experimental results showed that in addition to the remarkable reductions in the overall pressure drop achieved with steam flow modulation, the relative amplitude in the system pressure introduced by flow oscillation was unexpectedly very small (the maximum value obtained in the present experiments was less than 4%).
Experimental results of local wall heat fluxes showed that a significant part of the heat transfer augmentation achieved with steam modulation was due to condensate entrainment from the film, and more importantly from the water column where it existed. Estimates of water column entrainment rates made from experimental results revealed that extremely high entrainment ratios could be attained, depending on the amplitude of modulation. The high entrainment rates were manifested in the increased heat transfer rates around the water column region, due to the direct contact heat transfer made available by the interaction of the steam with the much colder water column. However, the local heat transfer under modulated steam flow conditions exhibited significant cyclic components, the magnitude of which depended on the minimum condensing region length (which was set by the steam flow bias and the amplitude in the inlet steam flow.
A fully implicit seven-equation two-fluid model was specifically formulated to model the reflux condensation process as observed in the present experiments. The model was solved by employing a simple time-dependant dynamic grid adaptation technique designed to track the unknown and time-varying condensing region length. A set of empirical correlations for interfacial mass, momentum, and energy transfer selected from the literature and developed from the present experimental results were employed in the solution of the two-fluid model. The model was used to investigate the effect of the tube diameter, tube length, and cooling side flow rate and inlet temperature on the process.
Numerical simulations of some of the present experiments, in the reflux condensation mode of operation, were carried out using the two-fluid model, but the model exhibited convergence problems in the presence of a water column, and so experiments in the second mode of operation could not be simulated with a reasonable degree of accuracy.
Theoretical results obtained from the model showed that as the tube diameter increased the pressure drop per unit length of condenser decreased, and the inlet steam flow rate at flooding increased, or equivalently the condensation capacity or operating range of the condenser increased, With respect to the inlet cooling water temperature, the model results showed that the heat removal rates in the condenser increased as this temperature was decreased.
Obinelo, Izundu F., "The Effects of Oscillatory Steam Flow on Reflux Condensation Phenomena" (1993). Open Access Dissertations and Theses. Paper 2980.