Experimental and numerical investigation of fouling in heat exchangers

In this research an advanced monitoring system for fouling phenomenon was developed. Such a system can be used to monitor fouling in a wide range of tubular heat exchangers such as condensers and intercoolers. First, a mathematical model of fouling resistance in tubular heat exchangers was adapted. The model is based on the applied thermal power, the inside heat transfer coefficient, and geometrical characteristics of the heat exchanger under consideration. The resulting model is a function of measured quantities such as water and tube wall temperatures, fluid flow velocity, and some physical properties of the fluid flowing inside the tubes, such as viscosity, conductivity, and density. Second, an on-line fouling monitoring system was prepared, and the fouling thermal resistance for selected solutions was measured in real time by this system. The effect of concentration and chemical reactions on fouling was studied experimentally using contaminants such as sodium bicarbonate, sodium chloride, calcium chloride, and a mixture of sodium bicarbonate and calcium chloride.;The fouling effect of seawater samples collected from three New Hampshire beaches on a 90/10 Cu/Ni commercial heat exchanger tube was investigated. The filtered seawater samples were circulated through the closed loop experimental setup for durations of seven and fourteen days and the fouling thermal resistance was measured continuously. Analytical microscopy was performed on the tube surface before and after the experiments to see the effect of seawater fouling on the tube surface. The results showed different fouling behavior for the seawater samples. This different behavior was confirmed by the different composition of the seawater samples. Fouling monitoring experiments revealed a higher fouling thermal resistance for one of the seawater samples, Hampton seawater, contrary to the results of analytical microscopy which showed a lower crystallization rate for Hampton sample. Water decomposition analysis showed the lowest sodium content for Hampton seawater compared to the other samples. Accordingly, corrosion of the tube surface occurred with a higher rate for Hampton seawater due to the presence of chlorine ions and a lower concentration of sodium. The high fouling resistance of Hampton seawater can be explained as the result of several simultaneous fouling mechanisms, corrosion and crystallization indicating composite fouling behavior.;A modeling technique, typically used for aerosol deposition, was applied to simulate the deposition process of calcium carbonate nano- and micro-particles suspended in turbulent cooling water flows. The mean turbulent velocity field and the fluctuating velocities were determined by the k-epsilon and RSM turbulence models by simulating the water flow in a typical heat exchanger horizontal tube. Commercial software (ANSYS FLUENT(TM) 12.1.4) was used for turbulence mean flow modeling and the simulation of turbulence fluctuations was performed by stochastic models. Particle deposition velocities were obtained for the particles with diameters in the range 0.01-50 mum by the k-epsilon and RSM models and compared to those calculated by semi-empirical correlations available in the literature to investigate the effect of the turbulence model on the deposition velocity. Results show that the proposed numerical model can predict deposition velocity of micro-particles in water accurately and can be useful in determining the range of particle diameters in which the highest deposition velocity occurs. However, for nano-particles, the model's results do not agree with the correlations due to the higher lateral turbulence fluctuations calculated by ANSYS FLUENT(TM) code.;The effects of different parameters, surface roughness, flow velocity, and concentration on the calcium carbonate scale formation process were experimentally investigated by using the developed monitoring system. The real operating conditions of a tubular heat exchanger with 90/10 Cu/Ni tubes were simulated by performing prolonged experiments with durations of two to seven days. The solution used was a mixture of sodium bicarbonate and calcium chloride in deionized water. The experiments were repeated with the same procedure for 90/10 Cu/Ni tubes with different internal surface roughness, flow velocities, and concentrations. The surface was analyzed by analytical microscopy to investigate the morphology of the deposit layer. Comparison of the experimental results of smooth and rough surfaces showed a higher fouling thermal resistance for the rough surface. Increasing the velocity considerably reduced the fouling thermal resistance. A higher fouling thermal resistance and pressure drop were seen for the high concentration solution while a short initiation time was found. The experimental results agree with those in the literature as well as with theory. A good agreement is also seen among the results from different investigation techniques applied. Results show flow velocity has the greatest effect on calcium carbonate scale formation. Accordingly, the scale formation process is diffusion controlled rather than surface controlled.