Comparison of Practical and Analytical Spray Performance in Defouling Process

Heat exchangers are widely used in heating, air conditioning, refrigeration, industrial plants and refineries, gas processing and water treatment. Fouling is a chronic operating and design problem for retaining highly effective heat exchangers. This leads to countless chemical and mechanical attempts for reducing fouling. Mitigation of fouling and removal of deposits can be addressed efficiently with proper design and gas conditioning modification. All previous mentioned factors and uncertainties can be examined, evaluated and optimized by using computational fluid dynamics (CFD) simulation. Typical defouling and cleaning process involves chemical solvent sprayed on the heat exchanger surfaces. Manufacturers and operators are facing increasing difficulty to preserve heat exchanger efficiency due to complex structures. Spray behavior and performance cannot be simply predicted due to the many variables involved such as nozzle characteristics, geometry, orientation, gas flow of heat exchanger and pressure resistance. There are tremendous challenges associated with fouling of heat exchangers, where CFD is known as a common tool to predict and resolve those problems. In this work, we will compare the performance of CFD simulation of spray with practical droplet test data in a scaled heat exchanger built and tested in Spraying Systems Co. spray laboratory for a pilot anti-fouling imitation. The spray characteristics based CFD will be used to validate the importance of reasonable and precise spray input simulation in a solvent based defouling process. * Corresponding author: [email protected] ILASS Americas 28th Annual Conference on Liquid Atomization and Spray Systems, Dearborn, MI, May 2016 Introduction Heat exchangers are process equipment that transfer heat. Through a heat transfer surface that separates the two fluids, heat is continuously or semicontinuously transferred from a hot to a cold fluid directly or indirectly. Heat exchangers are classified by flow arrangement and construction type, primarily bundles of pipes, tubes or plate coils. In processing industries like refineries, fouling is generally the deposition and accumulation of unwanted materials such as scale, algae, suspended solids and insoluble salts on the internal or external surfaces of heat exchangers. Fouling on process equipment surfaces is a chronic operating problem and can have a significantly negative impact on the operational efficiency of the unit. Fouling could cause a major economic loss on most industries. The total fouling related cost is estimated to exceed $4.4 million dollar yearly, which is about 0.25% to 30% of GDP of industrialized countries [1] . According to reports from Harwell Laboratories, maintenance costs of heat exchanger and boilers consist of about 15% of total costs in process plants. Heat exchange fouling associated cost includes costs due to over-design, additional fuel consumption and maintenance, loss of production by efficiency deterioration and shutdown. Also, heat exchanger fouling is responsible for emission of carbon dioxide, data from oil refineries suggested that crude oil fouling accounted for 10% of CO2 emissions of these plants [2] . Occasionally, heat exchanger surface is replaced rather than renewed. This circumstance arises when reuse is risky or more expensive than replacement, or shutdown is long. This strategy is seldom employed, which indicates the cleaning of heat exchanger is preferred to restore efficiency in process industries. Fouling inside heat exchanger can be reduced by appropriate heat-exchanger design, proper selection of operating conditions and exchanger geometry, mitigation/removal methods (mechanical and/or chemical) from the heat transfer surfaces and heat exchanger surface modification/coating. Numerous studies of the heat exchanger fouling and cleaning have been conducted. Today's oil refineries have to cope with heavier crude oils or densified residuals with higher risk of fouling from sources which until recently have not been economical to process. The production of hydrocarbon fuels from biomass and the co-firing of waste material and straw in power plants are associated with substantial fouling problems [3] . Further research on the problem of fouling in heat exchangers and practical methods for predicting the antifouling result, making use in particular of modern digital techniques, are still called for. Technical Approach Typical defouling and cleaning process involves chemical solvent sprayed on the heat exchanger surfaces, manually (see Fig.1) or mechanically by spray headers. Manufacturers and operators are facing increasing difficulty to preserve heat exchanger efficiency due to complex structures. Manual antifouling is time consuming, requires significant down time and hard to reach due to heat exchanger structure and safety issues. Injectors are commonly used. There are many variables that contribute to performance, including geometry, orientation, flow of heat exchanger, pressure resistance. Hence spray behavior and performance cannot be simply predicted. Figure 1. Heat Exchanger Fouling and Chemical and/or Mechanical Antifouling. Wilson and Crockford brought up discussion of cleaning and procedure effectiveness incorporating assurance and proof of cleanliness performance, which will dictate the choice of technologies and determine the consistent clean of a particular piece of equipment which cleaning can be verified [4,5] . There are tremendous challenges and requirements associated with defouling of heat exchangers, where CFD is known as a tool to predict and solve those problems. A scaled heat exchanger was built at the extension of a wind tunnel and tested in a spray laboratory as a pilot anti-fouling tube face. Various nozzles and arrangements were applied to this system. Spray performance was captured and recorded in the zone of interest. The data was then compared with CFD simulation, in which particle tracking was done with Discrete Phase Model. Practical droplet test data for specific nozzles was used as the inlet for model precision. The spray characteristics technology is combined with CFD analysis to ensure and validate the solvent spray behavior in defouling process, both qualitatively and quantitatively. Equipment and Methods The experimental setup consisted of spray nozzles, (with pump and flow meter), wind tunnel, and PDI with traverse. All tests were carried out with the co-current air flow. The injected fluid was water at ambient temperature ~293.15K. The nozzle was operated with a steady clean water supply for all tests as noted in Table 1. Wind tunnel and arrangement set-up schematics can be seen in Figures 2-3. Four different types of nozzles at relative layouts were used to complete the tests and compare the corresponding spray performance under the same liquid supply amount. Wind Tunnel The subsonic Wenham (blower-style) wind tunnel utilized in these experiments was capable of producing a co-current nominally uniform air flow at a velocity range from 2.5 m/s to greater than 50 m/s; the actual flow velocity generated during these tests was 5 and 10 m/s, respectively, for each arrangement. Before nozzles were placed in the test section, the wind speed and differential pressure was monitored and maintained using a Pitot tube method with calibrated TSI VELOCICALC meter reading at upstream and downstream of test section. This wind speed was chosen as it allowed for a reasonable representation of the scale down defouling systems commonly seen in the industry. The test section was set to ensure flow was fully developed before entering the test section, and pressure drop created by the heat exchanger was not influenced by the compressed air. During gas baseline stage, experimental results at two flow rates were collected for CFD comparison prior to nozzle installation. Figure 2 provides an image of the wind tunnel with heat exchanger structure (tubes) and the water line pipe for nozzle mounted ahead the test section. As the liquid drops bounce, splash and break up when hitting solid surface, it would be impossible to test exactly at tubeface. For conducting the nozzle tests with acceptable measurement resolution, a slide which was about 0.05m away from the tubeface, with 60% of the circumference length and approximately 0.025m width opening was cut as the test plane. PDI system was mounted at this plane on xand yaxis traverse and oriented vertically to allow data acquisition at various y-locations. Figure 2. Wind Tunnel System. Experimental layouts for specific nozzles are displayed in Figure 3 at the sequence of each line from left to right corresponding to each case. To reduce the repeats and save time in the tests, based on the layout, droplet behavior was assumed symmetrical from the water pipeline. Three horizontal locations, centerline and 0.075m increments, were set up as xaxis of traverse. Each measurement started at 0.025m away from the top of tunnel edge and 0.038m increments per test point moving toward the bottom of tunnel (yaxis of traverse) till no valid point or no laser detection was reached. per nozzle Units Case 1 Case 2 Case 3 Case 4 Injector Type Hollow Cone Hollow Cone Hollow Cone Hollow Cone Nozzle ID 1/8BD-3 1/4D3-45 1/4D2-45 1/4D1.5-45 Air Flow Conditions m/s 5 5 5 5