We have analysed a set of special large scale experiments run at SINTEF designed to characterize liquid accumulation in gas condensate pipelines at low liquid loads. The goal of the analysis was to clarify what the experiments actually measured: The critical point, where the two-phase Froude number equals unity, or the so-called accumulation point, where the liquid holdup starts to increase sharply with reducing gas flow rate. We developed an algorithm for computing the two points and performed a study to quantify the deviations between liquid flow rate and holdup at the critical point and at the accumulation point. We have found that the experiments give the accumulation point when the critical holdup is lower than the holdup at the accumulation point, which is normally the case in the current campaign. For steeper pipe inclinations, however, it is possible that the critical holdup is higher than the holdup at the accumulation point. In this case, experiments would be expected to measure the critical point.
An extensive experimental campaign was run in 2013 at the SINTEF Multiphase Flow Laboratory at Tiller, Kjølaas and Holm (1). The objectives of the campaign were to localize the onset of liquid accumulation for decreasing gas flow rate in two-phase low-liquid-loading flows for various pipe diameters, pipe angles and liquid rates (USL), to measure the extent of the regions where multiple holdup solutions exist and to measure the associated liquid holdups.
In the main type of experiments, from now on referred to as screening experiments, the test section was filled about half full of liquid. Then the liquid inflow was stopped and a gas flow rate inside the region of multiple holdup solutions was set, transporting the liquid out of the pipe at a low flow rate. A controlled injection of liquid at the inlet was used to keep the liquid level gradient in the middle of the pipe. The liquid holdup was then logged at several locations along the test section, and the liquid outflow was measured using a collecting tank with a level measurement system. An overview of the experimental setup is shown in figure 1.
The initial idea with the experiment was that when three holdup solutions exist at the actual conditions, the high holdup region in the lower half of the pipe represented the high holdup solution and the low holdup region in the upper half of the pipe represented the middle or the low holdup solution. The liquid outflow was interpreted as the "accumulation point" or liquid holdup take-off point at the set inclination angle and gas flow rate, i.e. the liquid flow rate at the jump from low to high holdup.
