Engineering topics

Anderson, David N.

Publications and source records attributed to Anderson, David N..

At least 19 records

NASA NTRS2008

Ice Shape Scaling for Aircraft in SLD Conditions

This paper has summarized recent NASA research into scaling of SLD conditions with data from both SLD and Appendix C tests. Scaling results obtained by applying existing scaling methods for size and test-condition scaling will be reviewed. Large feather growth issues, including scaling approaches, will be discussed briefly. The material included applies only to unprotected, unswept geometries. Within the limits of the conditions tested to date, the results show that the similarity parameters needed for Appendix C scaling also can be used for SLD scaling, and no additional parameters are required. These results were based on visual comparisons of reference and scale ice shapes. Nearly all of the experimental results presented have been obtained in sea-level tunnels. The currently recommended methods to scale model size, icing limit and test conditions are described.

Anderson, David N.↗

NASA NTRS2006

Further Assessment of MVD Effects in SLD Applications

The study reported here is part of an effort to develop scaling methods for super cooled large droplet (SLD) conditions. Previously reported results showed that SLD main ice shapes can be simulated quite successfully by appendix C conditions using scaling methods developed for appendix C. However, when the velocity was higher than 100 kt, the feather size and density for SLD tests at MVDs well above 100 m was not well represented by the scaled appendix C conditions. This paper reports additional results of a study of the feather region with the objective of identifying differences between SLD and appendix C feathers. Both the feather appearance and the angle at which feathers grow from the airfoil surface were recorded over a range of MVD from 20 to 190 m for airspeeds of 100 and 200 kt and stagnation freezing fractions of 0.3 to 1.0. Tests were performed in the NASA Glenn Icing Research Tunnel (IRT) using a 91-cm-chord NACA0012 airfoil model mounted at 0 degrees AOA. Photographs are presented to illustrate details of feather appearance. Appearance was noticeably affected by the stagnation freezing fraction of the test, but not by velocity or MVD. The angle of feather growth relative to the chord line decreased with increasing stagnation freezing fraction. For a velocity of 100 kt, no significant effect of MVD on feather angle was apparent, but at 200 kt, feather angle tended to increase with MVD for glaze conditions, but not rime. This finding is based on limited data, and its significance with respect to icing physics has not been determined.

Tsao, Jen-Ching↗

NASA NTRS2006

Latest Developments in SLD Scaling

Scaling methods have been shown previously to work well for super cooled large droplet (SLD) main ice shapes. However, feather sizes for some conditions have not been well represented by scale tests. To determine if there are fundamental differences between the development of feathers for appendix C and SLD conditions, this study used time-sequenced photographs, viewing along the span of the model during icing sprays. An airspeed of 100 kt, cloud water drop MVDs of 30 and 140 microns, and stagnation freezing fractions of 0.30 and 0.50 were tested in the NASA Glenn Icing Research Tunnel using an unswept 91-cm-chord NACA0012 airfoil model mounted at 0deg AOA. The photos indicated that the feathers that developed in a distinct region downstream of the leading-edge ice determined the horn location and angle. The angle at which feathers grew from the surface were also measured; results are shown for an airspeed of 150 kt, an MVD of 30 microns, and stagnation freezing fractions of 0.30 to 0.60. Feather angles were found to depend strongly on the stagnation freezing fraction, and were independent of either chordwise position on the model or time into the spray. Feather angles also correlated well with horn angles. For these tests, there did not appear to be fundamental differences between the physics of SLD and appendix C icing; therefore, for these conditions similarity parameters used for appendix C scaling appear to be valid for SLD scaling as well. Further investigation into the cause for the large feather structures observed for some SLD conditions will continue.

Tsao, Jen-Ching↗

NASA NTRS2005

Additional Results of Ice-Accretion Scaling at SLD Conditions

To determine scale velocity an additional similarity parameter is needed to supplement the Ruff scaling method. A Weber number based on water droplet MVD has been included in several studies because the effect of droplet splashing on ice accretion was believed to be important, particularly for SLD conditions. In the present study, ice shapes recorded at Appendix-C conditions and recent results at SLD conditions are reviewed to show that droplet diameter cannot be important to main ice shape, and for low airspeeds splashing does not appear to affect SLD ice shapes. Evidence is presented to show that while a supplementary similarity parameter probably has the form of a Weber number, it must be based on a length proportional to model size rather than MVD. Scaling comparisons were made between SLD reference conditions and Appendix-C scale conditions using this Weber number. Scale-to-reference model size ratios were 1:1.7 and 1:3.4. The reference tests used a 91-cm-chord NACA 0012 model with a velocity of approximately 50 m/s and an MVD of 160 m. Freezing fractions of 0.3, 0.4, and 0.5 were included in the study.

Bond, Thomas H.↗

NASA NTRS2005

Overview of Icing Physics Relevant to Scaling

An understanding of icing physics is required for the development of both scaling methods and ice-accretion prediction codes. This paper gives an overview of our present understanding of the important physical processes and the associated similarity parameters that determine the shape of Appendix C ice accretions. For many years it has been recognized that ice accretion processes depend on flow effects over the model, on droplet trajectories, on the rate of water collection and time of exposure, and, for glaze ice, on a heat balance. For scaling applications, equations describing these events have been based on analyses at the stagnation line of the model and have resulted in the identification of several non-dimensional similarity parameters. The parameters include the modified inertia parameter of the water drop, the accumulation parameter and the freezing fraction. Other parameters dealing with the leading edge heat balance have also been used for convenience. By equating scale expressions for these parameters to the values to be simulated a set of equations is produced which can be solved for the scale test conditions. Studies in the past few years have shown that at least one parameter in addition to those mentioned above is needed to describe surface-water effects, and some of the traditional parameters may not be as significant as once thought. Insight into the importance of each parameter, and the physical processes it represents, can be made by viewing whether ice shapes change, and the extent of the change, when each parameter is varied. Experimental evidence is presented to establish the importance of each of the traditionally used parameters and to identify the possible form of a new similarity parameter to be used for scaling.

Anderson, David N.↗

NASA NTRS2005

Evaluation and Validation of the Messinger Freezing Fraction

One of the most important non-dimensional parameters used in ice-accretion modeling and scaling studies is the freezing fraction defined by the heat-balance analysis of Messinger. For fifty years this parameter has been used to indicate how rapidly freezing takes place when super-cooled water strikes a solid body. The value ranges from 0 (no freezing) to 1 (water freezes immediately on impact), and the magnitude has been shown to play a major role in determining the physical appearance of the accreted ice. Because of its importance to ice shape, this parameter and the physics underlying the expressions used to calculate it have been questioned from time to time. Until now, there has been no strong evidence either validating or casting doubt on the current expressions. This paper presents experimental measurements of the leading-edge thickness of a number of ice shapes for a variety of test conditions with nominal freezing fractions from 0.3 to 1.0. From these thickness measurements, experimental freezing fractions were calculated and compared with values found from the Messinger analysis as applied by Ruff. Within the experimental uncertainty of measuring the leading-edge thickness, agreement of the experimental and analytical freezing fraction was very good. It is also shown that values of analytical freezing fraction were entirely consistent with observed ice shapes at and near rime conditions: At an analytical freezing fraction of unity, experimental ice shapes displayed the classic rime shape, while for conditions producing analytical freezing fractions slightly lower than unity, glaze features started to appear.

Anderson, David N.↗

NASA NTRS2005

Additional Study of Water Droplet Median Volume Diameter (MVD) Effects on Ice Shapes

This paper reports the result of an experimental study in the NASA Glenn Icing Research Tunnel (IRT) to evaluate how well the MVD-independent effect identified previously might apply to SLD conditions in rime icing situations. Models were NACA 0012 wing sections with chords of 53.3 and 91.4 cm. Tests were conducted with a nominal airspeed of 77 m/s (150 kt) and a number of MVD's ranging from 15 to 100 m with LWC of 0.5 to 1 g/cu m. In the present study, ice shapes recorded from past studies and recent results at SLD and Appendix-C conditions are reviewed to show that droplet diameter is not important to rime ice shape for MVD of 30 microns or larger, but for less than 30 m drop sizes a rime ice shape transition from convex to wedge to spearhead type ice shape is observed.

Tsao, Jen-Ching↗

NASA NTRS2004

Manual of Scaling Methods

This manual reviews the derivation of the similitude relationships believed to be important to ice accretion and examines ice-accretion data to evaluate their importance. Both size scaling and test-condition scaling methods employing the resulting similarity parameters are described, and experimental icing tests performed to evaluate scaling methods are reviewed with results. The material included applies primarily to unprotected, unswept geometries, but some discussion of how to approach other situations is included as well. The studies given here and scaling methods considered are applicable only to Appendix-C icing conditions. Nearly all of the experimental results presented have been obtained in sea-level tunnels. Recommendations are given regarding which scaling methods to use for both size scaling and test-condition scaling, and icing test results are described to support those recommendations. Facility limitations and size-scaling restrictions are discussed. Finally, appendices summarize the air, water and ice properties used in NASA scaling studies, give expressions for each of the similarity parameters used and provide sample calculations for the size-scaling and test-condition scaling methods advocated.

Bond, Thomas H.↗

NASA NTRS2003

Quantification of Ice Accretions for Icing Scaling Evaluations

The comparison of ice accretion characteristics is an integral part of aircraft icing research. It is often necessary to compare an ice accretion obtained from a flight test or numerical simulation to one produced in an icing wind tunnel or for validation of an icing scaling method. Traditionally, this has been accomplished by overlaying two-dimensional tracings of ice accretion shapes. This paper addresses the basic question of how to compare ice accretions using more quantitative methods. For simplicity, geometric characteristics of the ice accretions are used for the comparison. One method evaluated is a direct comparison of the percent differences of the geometric measurements. The second method inputs these measurements into a fuzzy inference system to obtain a single measure of the goodness of the comparison. The procedures are demonstrated by comparing ice shapes obtained in the Icing Research Tunnel at NASA Glenn Research Center during recent icing scaling tests. The results demonstrate that this type of analysis is useful in quantifying the similarity of ice accretion shapes and that the procedures should be further developed by expanding the analysis to additional icing data sets.

Ruff, Gary A.↗

NASA NTRS2003

A Preliminary Study of Ice-Accretion Scaling for SLD Conditions

Proposed changes to aircraft icing certification rules are being considered by European, Canadian, and American regulatory agencies to include operation in super-cooled large droplet conditions (SLD). This paper reports results of an experimental study in the NASA Glenn Icing Research Tunnel (IRT) to evaluate how well scaling methods developed for Appendix C conditions might apply to SLD conditions. Until now, scaling studies have been confined to the FAA FAR-25 Appendix C envelope of atmospheric cloud conditions. Tests were made in which it was attempted to scale to a droplet MVD of 50 microns from clouds having droplet MVDs of 175, 120, 100, and 70 microns. Scaling was based on the Ruff method with scale velocities found either by maintaining constant Weber number or by using the average of the velocities obtained by maintaining constant Weber number and constant Reynolds number. Models were unswept NACA 0012 wing sections. The reference model had a chord of 91.4 cm. Scale models had chords of 91.4, 80.0, and 53.3 cm. Tests were conducted with reference airspeeds of 100 and 150 kt (52 and 77 m/s) and with freezing fractions of 1.0, 0.6, and 0.3. It was demonstrated that the scaled 50-micron cloud simulated well the non-dimensional ice shapes accreted in clouds with MVD's of 120 microns or less.

Anderson, David N.↗

NASA NTRS2003

Measurement and Correlation of Ice Accretion Roughness

Measurements were taken of the roughness characteristics of ice accreted on NACA 0012 airfoils in the NASA Glenn Icing Research Tunnel (IRT). Tests were conducted with size scaled, using models with chords of 26.7, 53.3, and 80.0 cm, and with liquid-water content scaled, both according to previously-tested scaling methods. The width of the smooth zone which forms on either side of the leading edge of the airfoil and the diameter of the roughness elements are presented in non-dimensional form as functions of the accumulation parameter. The smooth-zone width was found to decrease with increasing accumulation parameter. The roughness-element diameter increased with accumulation parameter until a plateau was reached. This maximum diameter was about 0.06 times twice the model leading-edge radius. Neither smooth-zone width nor element diameter were affected by a change in freezing fraction from 0.2 to 0.4. Both roughness characteristics appeared to scale with model size and with liquid-water content.

Anderson, David N.↗

NASA NTRS2003

Ice-Accretion Scaling Using Water-Film Thickness Parameters

Studies were performed at INTA in Spain to determine water-film thickness on a stagnation-point probe inserted in a simulated cloud. The measurements were correlated with non-dimensional parameters describing the flow and the cloud conditions. Icing scaling tests in the NASA Glenn Icing Research Tunnel were then conducted using the Ruff scaling method with the scale velocity found by matching scale and reference values of either the INTA non-dimensional water-film thickness or a Weber number based on that film thickness. For comparison, tests were also performed using the constant drop-size Weber number and the average-velocity methods. The reference and scale models were both aluminum, 61-cm-span, NACA 0012 airfoil sections at 0 deg. AOA. The reference had a 53-cm-chord and the scale, 27 cm (1/2 size). Both models were mounted vertically in the center of the IRT test section. Tests covered a freezing fraction range of 0.28 to 1.0. Rime ice (n = 1.0) tests showed the consistency of the IRT calibration over a range of velocities. At a freezing fraction of 0.76, there was no significant difference in the scale ice shapes produced by the different methods. For freezing fractions of 0.40, 0.52 and 0.61, somewhat better agreement with the reference horn angles was typically achieved with the average-velocity and constant-film thickness methods than when either of the two Weber numbers was matched to the reference value. At a freezing fraction of 0.28, the four methods were judged equal in providing simulations of the reference shape.

Anderson, David N.↗

NASA NTRS2003

Effect of Velocity in Icing Scaling Tests

This paper presents additional results of a study first published in 1999 to determine the effect of scale velocity on scaled icing test results. Reference tests were made with a 53.3-cm-chord NACA 0012 airfoil model in the NASA Glenn Icing Research Tunnel at an airspeed of 67 m/s, an MVD of 40 microns, and an LWC of 0.6 g/cu m. Temperature was varied to provide nominal freezing fractions of 0.8, 0.6, and 0.5. Scale tests used both 35.6- and 27.7-cm-chord 0012 models for 2/3- and 1/2-size scaling. Scale test conditions were found using the modified Ruff (AEDC) scaling method with the scale velocity determined in five ways. Four of the scale velocities were found by matching the scale and reference values of water-film thickness, velocity, Weber number, and Reynolds number. The fifth scale velocity was simply the average of those found by matching the Weber and Reynolds numbers. The resulting scale velocities ranged from 85 to 220 percent of the reference velocity. For a freezing fraction of 0.8, the value of the scale velocity had no effect on how well the scale ice shape simulated the reference shape. For nominal freezing fractions of 0.5 and 0.6, the best simulation of the reference shape was achieved when the scale velocity was the average of the constant-Weber-number and the constant-Reynolds-number velocities.

Anderson, David N.↗

NASA NTRS2003

Evaluation of Methods to Select Scale Velocities in Icing Scaling Tests

A series of tests were made in the NASA Glenn Icing Research Tunnel to determine how icing scaling results were affected by the choice of scale velocity. Reference tests were performed with a 53.3-cm-chord NACA 0012 airfoil model, while scale tests used a 27.7-cm-chord 0012 model. Tests were made with rime, mixed, and glaze ice. Reference test conditions included airspeeds of 67 and 89 m/s, an MVD of 40 microns, and LWCs of 0.5 and 0.6 g/cu m. Scale test conditions were established by the modified Ruff (AEDC) scaling method with the scale velocity determined in five ways. The resulting scale velocities ranged from 85 to 220 percent of the reference velocity. This paper presents the ice shapes that resulted from those scale tests and compares them to the reference shapes. It was concluded that for freezing fractions greater than 0.8 as well as for a freezing fraction of 0.3, the value of the scale velocity had no effect on how well the scale ice shape simulated the reference shape. For freezing fractions of 0.5 and 0.7, the simulation of the reference shape appeared to improve as the scale velocity increased.

Anderson, David N.↗

NASA NTRS2003

Acceptable Tolerances for Matching Icing Similarity Parameters in Scaling Applications

This paper reviews past work and presents new data to evaluate how changes in similarity parameters affect ice shapes and how closely scale values of the parameters should match reference values. Experimental ice shapes presented are from tests by various researchers in the NASA Glenn Icing Research Tunnel. The parameters reviewed are the modified inertia parameter (which determines the stagnation collection efficiency), accumulation parameter, freezing fraction, Reynolds number, and Weber number. It was demonstrated that a good match of scale and reference ice shapes could sometimes be achieved even when values of the modified inertia parameter did not match precisely. Consequently, there can be some flexibility in setting scale droplet size, which is the test condition determined from the modified inertia parameter. A recommended guideline is that the modified inertia parameter be chosen so that the scale stagnation collection efficiency is within 10 percent of the reference value. The scale accumulation parameter and freezing fraction should also be within 10 percent of their reference values. The Weber number based on droplet size and water properties appears to be a more important scaling parameter than one based on model size and air properties. Scale values of both the Reynolds and Weber numbers need to be in the range of 60 to 160 percent of the corresponding reference values. The effects of variations in other similarity parameters have yet to be established.

Anderson, David N.↗

NASA NTRS2003

A Study of Scaling for Intercycle Ice Accretion Tests

The Ruff method with matched scale and reference velocity was used to determine appropriate 1/2-scale test conditions to simulate a full-size icing encounter for an NACA 23012 wing section protected with a pneumatic boot deicing system. Intercycle ice accretions were recorded on a 36-in-chord model used to represent 1/2-scale and compared with a hybrid reference model (full-size leading-edge and truncated aft section) representing a 72-in-chord full-size airfoil. The intercycle ice thickness and extent of icing for the scale tests generally compared well with those from the reference model. However, the scale tests did not reproduce the location and number of feather rows seen in the reference tests aft of the main ice shape. Many of the differences observed were believed to result from not scaling the pneumatic boot design along with the model size for these tests.

Anderson, David N.↗

NASA NTRS1997

Scaling Methods for Simulating Aircraft In-Flight Icing Encounters

This paper discusses scaling methods which permit the use of subscale models in icing wind tunnels to simulate natural flight in icing. Natural icing conditions exist when air temperatures are below freezing but cloud water droplets are super-cooled liquid. Aircraft flying through such clouds are susceptible to the accretion of ice on the leading edges of unprotected components such as wings, tailplane and engine inlets. To establish the aerodynamic penalties of such ice accretion and to determine what parts need to be protected from ice accretion (by heating, for example), extensive flight and wind-tunnel testing is necessary for new aircraft and components. Testing in icing tunnels is less expensive than flight testing, is safer, and permits better control of the test conditions. However, because of limitations on both model size and operating conditions in wind tunnels, it is often necessary to perform tests with either size or test conditions scaled. This paper describes the theoretical background to the development of icing scaling methods, discusses four methods, and presents results of tests to validate them.

Anderson, David N.↗

NASA NTRS1997

Tests of the Performance of Coatings for Low Ice Adhesion

This paper reports studies of the performance of low-ice-adhesion coatings by NASA Lewis and BFGoodrich. Studies used impact ice accreted both in the NASA Lewis Icing Research Tunnel (IRT) and in the BFGoodrich Icing Wind Tunnel (IWT) and static ice in a BFGoodrich bench-top parallel-plate shear rig. Early tests at NASA Lewis involved simple qualitative evaluations of the ease of removing impact ice from a surface. Coated surfaces were compared with uncoated ones. Some of the coatings were tested again with static ice at BFGoodrich to obtain quantitative measurements. Later, methods to establish the adhesion force on surfaces subjected to impact ice were explored at Lewis. This paper describes the various test programs and the results of testing some of the coatings looked at over the past 5 years. None of the coatings were found to be truly ice-phobic; however, the most effective coatings were found to reduce the adhesion of ice to about 1/2 that of an uncoated aluminum sample.

Anderson, David N.↗