Reynold's Number Effects on Deposition in Heated Fuel Flows, 03-9277

Printer Friendly Version

Principal Investigators
Clifford A. Moses

Inclusive Dates: 10/01/01 - 10/01/03

Background - Advanced, high-efficiency aircraft gas turbines will operate at higher pressure ratios and temperatures than current technology engines. As a result, the fuel is exposed to higher temperature environments. Jet fuel has a thermal stability limit beyond which the fuel begins to react with trace oxygen dissolved in the fuel to form varnish deposits which can foul heat transfer surfaces and/or block small fuel passages. Research has shown that deposition rates are affected by fluid mechanics and heat transfer, both of which are well-correlated with Reynold's number as a measure of turbulence in the flow; however, there are conflicts in the literature as to whether increasing Reynold's number results in higher or lower deposition rates. These conflicts make it difficult for engineers to utilize the results of laboratory deposition experiments in optimizing heated flow systems to minimize deposition. The objective of this research project is to determine if, and under what circumstances, Reynold's number is an effective correlating parameter for the effects of velocity and dimensional variables on the deposition rates in fuel flows through heated tubes.

Approach - This research incorporated both a heat transfer analysis and deposition experiments. The experiments were conducted in heated tubes, varying Reynold's number by varying velocity and diameter independently while holding residence time constant.

Accomplishments - The analytic approach used solutions from the literature for Reynold's number effects on the temperature and velocity fields. Transport of deposit precursors to the wall was modeled using a 1-d random walk approach for Brownian motion. This approach allowed some precursors to migrate to the wall and deposit while others diffused away from the wall according to the concentration gradient. Reynold's number, wall temperature, and fuel temperature were then treated as independent variables to evaluated their effect on deposition. Wall temperature was found to be the most important parameter on deposition because of the exponential effect on the formation rate of precursors near the wall. Increasing Reynold's number causes a thinning of the boundary layer and reduces the deposition rate because the temperature gradient, and hence the concentration gradient of precursors, is much steeper.

The experimental work yielded results that were consistent with the analytic predictions that deposition would decrease with increasing Reynold's number, but at the same increase with flow diameter. Not enough experiments were conducted to develop an empirical model.

Overall, the results of the two approaches are consistent with earlier SwRI data that led to the initial hypothesis that the precursors to deposit formation are formed entirely within the laminar sub-layer along the heated wall rather than the bulk flow as has generally been considered in the open literature. It follows from this that transport to the wall is by diffusion and that turbulent transport does not play a direct role. While neither approach was able to lead to a definitive deposition model, the analytical study, however, has shed valuable new light on the coupling of the chemical and physical transport mechanisms of deposition thereby providing new guidance for the development of a computer model.

2003 Program Home