Wet central heating systems account for a very large portion of energy consumption in the UK and recent figures indicate that its usage in households will be increasing even further. Under such circumstances, it is desirable to use these systems in the most efficient way possible. However, dissolved gases that penetrate into central heating systems are later released as bubbles due to local supersaturated conditions occurring on the primary heat exchanger wall of the boiler. This leads to a two-phase flow throughout the pipework, causing microbubbles to escape to the upper parts of the system and creating cold spots in the radiators, thus, reducing its efficiency. There is an increasing trend in building services to install devices that remove these unwanted gases. Therefore, investigation of two-phase structures throughout different pipe installations will facilitate companies in enhancing their deaerator designs. In this regard, extensive experimental and computational investigations of two-phase flow structures were conducted within this study. Two-phase flow structures were measured by a photographic technique and investigated in means of void fractions, bubble sizes, and velocities. Fluid velocities in the range of 0.5 to 1.1 m/s at typical wet central heating temperature (60 to 80 °C) and pressures (2.2 to 27 bar) were utilized. Results show that that bubble production increases as temperature, boiler heating load, and saturation ratio escalate. On the other hand, it reduces when the pressure and flow rate of the system gets higher. A clear relationship between bubble sizes and system parameters was non-existent, except for the system flow rate (where bubble diameters decrease as the flow rate increases). Moreover, bubbles were evenly distributed during vertical flow when compared to horizontal flow, where bubbles tend to flow at the upper parts of the pipe. Furthermore, it was shown that bubble distributions were highly affected by obstacles like the 90 degree bend, thermocouple or pressure sensors. In addition, it was observed that axial flow development of bubbly flow was a continuous process and void fraction at the upper part of the pipe increased as the flow travelled through horizontal pipeline. Regarding the bubble velocity measurements, it was concluded that, bubble velocity profiles show development along both vertical and horizontal flows and approach to profiles which can be expressed with the power-law. Moreover, coalescence of two bubbles during horizontal flow was captured, emphasizing that the effect of coalescences should not be neglected at low void fractions. It was also found that bubbly flow in central heating systems was in a coalescences dominant regime and maximum bubble diameter observed at most positions were higher than theoretically defined values. Moreover, bubble dissolution effect was not observed at any of the test rig conditions. The reasons were thought to be the variation saturation ratio and axial flow development of two-phase flow, which supress the effect of dissolution and favour coalescence phenomenon. Finally, after evaluating conclusions from the experimental results and computational study regarding the effect of the 90 degree bend on void fraction distributions, it was concluded that the employed physical model and solver settings in ANSYS Fluent 14.5, can be utilized to predict bubble distribution developments throughout the central heating systems’ pipework. Keywords: Central heating systems, two-phase flow, bubbly flow, bubble distributions, bubble sizes, bubble velocities, coalescence, image processing, experimental fluid easurements.