Search

Techniques have been developed at the Engine Test Facility (ETF) of the Arnold Engineering Development Center (AEOC) to simulate flight through atmospheric icing conditions of supercooled liquid water droplets. Ice formed on aircraft and propulsion system surfaces during flight through icing conditions can, even in small amounts, be extremely hazardous. The effects of ice are dependent on many variables and are still unpredictable. Often, experiments are conducted to determine the characteristics of the aircraft and its propulsion system in an icing environment. Facilities at the ETF provide the capability to conduct icing testing in either the direct-connect (connected pipe) or the free-jet mode. The requirements of a spray system for turbine engine icing testing are described, as are the techniques used at the AEDC ETF to simulate flight in icing conditions. Some of the key issues facing the designer of a spray system for use in an altitude facility are identified and discussed, and validation testing of the design of a new spray system for the AEDC ETF is detailed. This spray system enables testing of the newest generation of high-thrust turbofan engines in simulated icing conditions.

The objective of this effort was to validate in flight, data that has been gathered in the NASA Lewis Research Center's Icing Research Tunnel (IRT) over the past several years. All data was acquired in flight on the NASA Lewis Research Center's Twin Otter Icing Research Aircraft. A faired 3.5 in. diameter metal-clad cylinder exposed to the natural icing environment was observed by a close-up video camera. The grazing angle video footage was recorded to S-VHS video tape and after the icing encounter, the resultant ice shape was documented by 35 mm photography and pencil tracings. The feather growth area was of primary interest; however, all regions of the ice accretion, from the stagnation line to the aft edge of run back were observed and recorded. After analysis of the recorded data several interesting points became evident: (1) the measured flight feather growth rate is consistent with IRT values, (2) the feather growth rate appears to be influenced by droplet size, (3) the feathers were straighter in the lower, spottier LWC of flight in comparison to those observed in the IRT, (4) feather shedding and ice sublimation may be significant to the final ice shape, and (5) the snow encountered on these flights appeared to have little influence on ice growth.

With the possible exception of tailplane-ice hazards, understanding of aircraft icing processes and their performance penalties and hazards has not changed fundamentally in 40-50 years. A survey is presently conducted of the aircraft design criteria, ice-detection and deicing hardware, and flight procedures, that have been developed to cope with icing phenomena. Attention is given to pilot icing conditions flight-training requirements, which are essential for the recognition of icing situations requiring corrective action.

Various types of ice formation have been studied by analysis of high magnification video observations. All testing was conducted in the NASA Lewis Icing Research Tunnel (IRT). A faired 8.9 cm (3.5 in.) diameter metal-clad cylinder and a 5.1 (2 in.) aluminum cylinder were observed by close-up and overview video cameras for several wind tunnel conditions. These included close-up grazing angle, close-up side view, as well as overhead and side overview cameras. Still photographs were taken at the end of each spray along with tracings of the subsequent ice shape. While in earlier tests only the stagnation region was observed, the entire area from the stagnation line to the horn region of glaze ice shapes was observed in this test. The modes or horn formation have been identified within the range of conditions observed. In the horn region, Horn Type A ice is formed by 'dry' feather growth into the flow direction and Horn Type B is formed by a 'wet' growth normal to the surface. The feather growth occurs when the freezing fraction is near unity and roughness elements exist to provide an initial growth site.

R Universal gas constant for air The altitude engine test cells at the Arnold Engineering Development Center are capable of simulating atmospheric icing conditions. The test cells are used to conduct tests in the direct-connect and free-jet modes. A key icing simulation parameter is the cloud liquid water content. The liquid water content setting at AEDC considers two factors, the bulk value and the spatial distribution of the liquid water across a specific region of interest. The techniques used to determine the bulk value from measurable quantities of injected water flow rate, test cell air mass flow rate, total air temperature, and static and total air pressure are discussed. The technique for evaluation of the spatial distribution of the liquid water content using ice thickness measurements is discussed. A recent development in automation of the ice thickness measurement process with ultrasonic transducers is described. The use of the ultrasonic technology removes several errors associated with the time-honored method of manually measuring ice thickness.

This paper deals with the measurement of the interfacial stress of a composite beam using an embedded piezoelectric PVDF (polyvinylidene-fluoride) strip. The composite beam consists of an aluminium beam on which an ice layer is deposited. The ice layer is formed from sprayed supercooled water droplets, simulating atmospheric icing of exposed structures. The beam is submitted to sinusoidal stress at the ice/aluminium interface by an electromagnetic shaker clamped to one end of the beam. The use of PVDF strips as a stress measurement apparatus was validated by numerical simulations. Experimental results demonstrated that a 25 µm thick PVDF strip used as an embedded interfacial stress sensor is sensitive only to bending stress and does not allow the measurement of interfacial shear stress. That way, the adhesion strength of an atmospheric ice layer cannot be measured. However, the use of embedded PVDF strips provides a new and original method for measuring the ice adhesion strength on different substrates, due to high repeatability and sensitivity of the measurement.

Atmospheric rime icing, resulting primarily from supercooled cloud droplet impaction on objects at the Earth's surface, was monitored and analyzed as a function of elevation on the west faces of Madonna Peak and Mount Mansfield in the Green Mountains, Vermont, and at the summit of Mount Washington, New Hampshire. Measurements were made of ice accretion rates on passive, manually operated collection baskets and automatic ice detectors. Icing rates increase exponentially with elevation above about 800 m, with secondary controls of rate suggested by microtopographic relief exposure. The illustrated dependence of icing rate upon elevation is largely a function of New England wind and cloud regimes and differs from other selected mountainous locations. The relationships presented may help assess the magnitude of frozen moisture inputs to high-elevation mountain ecosystems.

Four commercially available wind speed sensors have been tested in an icing wind tunnel to determine the relative susceptibility of each to atmospheric icing and to determine the influence of ire accumulations upon the operation and accuracy of each. A comparison of relative susceptibility to icing was carried out on the basis of an average collection efficiency. A clear advantage of the large sensors over the smaller ones was found. The operation of each sensor is affected by ice accretion in three specific ways. These are (i) interference between ice on the support structure and the hub of the rotating assembly, (ii) growth of ice from the support into the volume swept by the rotating assembly, and (iii) a change in cup shape. An estimate of an “effective” operating time under expected icing conditions was made. Results of this calculation indicate the useful operating life is only an the order of hours.

The atmospheric icing climatologies of two New England mountaintops with different elevations are compared: Mount Mansfield in northern Vermont and Mount Washington in New Hampshire. Atmospheric icing, as measured with Rosemount ice detectors, is twice as frequent on Mount Washington with about 12 to 20 times greater intensities and 25 to 50 times more accretion as on Mount Mansfield. Most of Mount Mansfield icing events are of low intensity, with periods between icing events averaging 35 to 45 hours on both peaks. Return intervals of ice events by length, intensity, and accretion amount are tabled. Approximately one-half of all severe icing on the two peaks occurs during and immediately after cold front passages. Icing is most intense when lows are about 450 km to the east and high pressure centers are more than about 450 km distant. Prolonged accretion periods occur when coastal and inland storms merge or follow closely.

In this paper the horizontal layer of clear ice sticking to the substrate is melted by comparatively short wave radiation similar to solar radiation for the purpose of removing ice from the surface of the material subject to atmospheric icing. The radiating source used for melting is 300 wattages halogen lamps whose color temperature is 3200‡K at 100 voltages. From the present investigation, a typical phenomenon of backmelting is observed clearly and it can be found that the predicted results including the melting rate of upper and lower layers which are melted by radiant energy impinged on or penetrated the ice layer are in good agreement with the experimental results.

This paper is concerned with melting of a vertical ice layer adhering to the substrate by using radiating heat source of halogen lamps having a large fraction of short wave beam or nichrome heater having a comparatively large fraction of long wave one. From the present experimental results, it can be seen that the heating of short wave radiation produces a peculiar melting behavior of strongly rough melting-surface due to the internal melting at the grain boundary of ice-surface. On the other hand, for the case of long wave radiation the melting-surface becomes very smooth. The melting rate of clear ice layer by short wave radiation obtained from halogen lamps is smaller than that of cloudy ice layer due to the good penetration of short wave fraction through the clear ice layer. Moreover, the raising of temperature of ice-substrate interface could offer a feasibility of removing ice layer from the structure subject to atmospheric icing. Concludingly, it is clarified that the melting rate of ice layer could be predicted numerically by using the band model of extinction coefficient or absorption coefficient presented in this study.

The literature reporting scientific and engineering investigations on ice has grown rapidly in the past 50 years, and increased by a factor of at least ten in the past 20 years. A review is presented of advances in engineering for ice problems during this period, including ice formation in rivers, lakes, and oceans; ice forces on structures; bearing capacity of ice covers; and atmospheric icing. The difficulties faced by engineers due to the unique physical and mechanical properties of ice and its normal thermal state are emphasized.

Ice accretion loads on cables can be measured through the end tension. For that purpose, the catenary equations are used as a subroutine in an iteration procedure to obtain the average ice loading on a cable from the end cable tension. To evaluate the accuracy of this approach, non-uniformly distributed ice loading is investigated. A catenary cable element computer code is used to calculate the end cable forces and tension as a function of a non-uniformly distributed ice load along a cable. The solution of the non-linear algebraic equations is obtained by a Newton-Raphson method. Numerical results show that for non-uniform icing the end tension can be used to give as good an evaluation of the ice accretion load as any other type of end support measurements. This technique is used to interpret, in terms of ice loadings, tension data measured on an instrumented cable at a site with frequent icing. A collected rime accretion sample is compared with the corresponding tension measurement and shows good agreement.

The heat transfer behavior of accreting ice surfaces in natural (flight test) and simulated (wind tunnel) cloud icing conditions are studied. Observations of wet and dry ice growth regimes as measured by ultrasonic pulseecho techniques are made. Observed wet and dry ice growth regimes at the stagnation point of a cylinder are compared with those predicted using a quasi steady-state heat balance model. A series of heat transfer coefficients are employed by the model to infer the local heat transfer behavior of the actual ice surfaces. The heat transfer in the stagnation region is generally inferred to be higher in wind tunnel icing tests than in natural, flight icing conditions.

The first experiments with the Swiss Hail Tunnel [1] have shown that the growth of hailstones under atmospheric icing conditions does not occur in the way usually assumed. It is quite normal to observe still liquid water in the resulting ice deposit. Figure 1 shows a model of this new observed mechanism while Figure 2 shows an ice framework without the liquid water normally contained in it.

In the third report, the author presented the results of his comparison between the growth rate and amount of ice accreted on patrol boats obtained from icing experiments conducted at sea and the rate and amount of ice estimated by the theories of forced convection of heat developed in the first and second reports.It is also advanced that the growth rate and amount of ice accreted on any vessels, can be estimated approximately under any ambient air temperature, wind velocity and duration of icing at sea through the results of these studies.This paper describes some materials for ambient air temperatures and relative wind velocities during icing on ships, which have been collected through questionnaires entrusted to various types of patrol boats in the northern waters around our country over several years.The weather condition of the growth of ice accreted on ship has been made considerably clearer by these investigations.Therefore, assuming conditional ambient air temperature, wind velocity and duration of icing on ships from the abovementioned materials, and picking up four vessels, whose navigations are scheduled in the northern waters, the amounts and heights of centres of gravity of ice accreted on them are estimated, and the change of ships' stability between before and after icing is also presented.It is also indicated that the centre of gravity of ice accreted on ship can be estimated satisfactorily by these theories.Finally, the author has made a comparison of the two kinds of amounts and centres of gravity of ice accreted on various types of patrol boats and two fishing vessels estimated by these theories and by the USSR icing norms respectively.It is pointed out through the results of the comparison that the USSR icing norms contain some defects basically and new icing norms are proposed by the author on the basis of the already reported investigations and studies.

Ice formation constitutes a hazard to aircraft operation both on the ground and in flight. This article deals with the protection of aircraft against ice formation in flight, but does not consider the counter measures which must be taken on the ground in winter conditions. The first part of the paper deals with the atmospheric conditions which give rise to ice accretion on forward facing surfaces and the types of ice which form at various ambient temperatures. A general survey is then made on methods of solving the problem and the weight and power penalties which they entail. Finally some recent developments in the electrical deicing systems arc reviewed.

A meeting of the Royal Aeronautical Society was held in the lecture hall of the Institution of Civil Engineers at Great George Street, Westminster, London, S.W.1, on Thursday 5th December 1946, at which a paper by Mr. J. K. Hardy on “ The Protection of Aircraft Against Ice,” was presented and discussed. In the Chair, the President, Sir Frederick Handley Page, C.B.E.

MIL-STD-810D, Method 521.0, "Icing/Freezing Rain," contains guidance for testing equipment exposed to freezing rain. This method was developed around a narrow set of meteorological conditions resulting in the buildup of clear glaze ice. The icing procedures described in Method 521.0 can be applied to a wide range of equipment and will help to predict equipment operations during natural freezing rain. However, the procedures do not give any indication of the operation of equipment during exposure to rime ice. The physical properties of rime and glaze ice differ significantly, as do their effects on equipment. Until MIL-STD-810D is modified to include rime icing, program managers and environmental test engineers will find no guidance or procedures in MIL-STD-810D for half of the icing (rime vs. glaze) that equipment can be exposed to in nature.

A time-dependent numerical model of ice accretion on wires, such as overhead conductors, is presented. Simulations of atmospheric icing are made with the model in order to examine the dependence of the accreted ice amount on atmospheric conditions. The results show that in wet growth (glaze formation) under constant atmospheric conditions, the growth rate increases with time until the process changes to dry growth. In dry growth (rime formation) the growth rate typically increases with time at the beginning of the icing process, but later decreases with time when the ice deposit has grown bigger. The effect of air temperature on the ice load turns out to be rather small for the first 24 hours of icing in typical dry growth conditions, but it is important for long-term icing. The ultimate ice load may either increase or decrease with decreasing air temperature, depending on the other atmospheric conditions and on the duration of icing. These results largely explain the difficulties encountered in estimating the formation of ice loads by simple methods using the routinely measured meteorological parameters.

The literature reporting scientific and engineering investigations on ice has grown rapidly in the past 50 years, and increased by a factor of at least ten in the past 20 years. A review is presented of advances in engineering for ice problems during this period, including ice formation in rivers, lakes, and oceans; ice forces on structures; bearing capacity of ice covers; and atmospheric icing. The difficulties faced by engineers due to the unique physical and mechanical properties of ice and its normal thermal state are emphasized.

Natural and artificial icing tests were conducted on a JUH-1H helicopter. An ice phobic coating was applied to both the main and tail rotor blades. A total of 4.0 productive hours were flown in the artificial environment, and 9.3 hours in natural icing conditions. Artificial tests were conducted utilizing the Helicopter Icing Spray System (HISS) with ambient temperatures ranging from -6 to -23 deg C, and relative humidities of 60 to 90 percent. Natural icing tests were conducted within the ambient temperature range of -2 to -12.5 deg C, with liquid water contents (LWC) ranging from 0.1 to 0.32 gm/ cu m. The artificial icing tests verified proper operation of the test helicopter ice protection system for use as a safety device in subsequent tests. Natural icing tests were conducted to determine the operational potential of G661 ice phobic compound and to gather data to aid in defining icing phenomenon and ice protection equipment design requirements. Due to insufficient quantitative baseline UH-1H rotor blade icing data, it was not possible to determine if the G661 ice phobic compound affected the operational icing capability of the UH-1H helicopter. Nine minor asymmetric ice sheds were encountered, however, no predetermined limit conditions were reached during testing. Excessive equipment and man-hour requirements degrade the practicality of G661 use. The LWC indicators were not adequate to determine ice accretion level and the Integrating Rate Unit (IRU) did not provide a repeatable cue of impending asymmetric ice shed.

Icing measurements were carried out in Spain during the Precipitation Enhancement Project experiment in 1979, with an instrumented DC-7 aircraft. The energy balance at the riming surface of a cylinder allows the prediction of the ice growth regime (wet or dry) for the present measurements with 85% success. The mixed conditions encountered show that the ice phase plays an important role in the ice accretion process. For a range of temperature from −21 to −8°C, the icing rate appears to be reduced ∼50% by the presence of large ice particles in concentrations above 5 L−1. The proposed explanation is that the ice particles erode the ice deposit formed by the supercooled water.

Aircraft exposed to an atmospheric icing environment can accumulate ice, resulting in a sharp increase in drag, a reduction in lift, control surface fouling, and engine damage all of which result in a hazardous flight situation. NASA Lewis Research Center (LeRC) has conducted a program to examine, with the aid of high-speed computer codes, how the trajectories of particles contribute to the ice accumulation on airfoils and engine inlets. For this effort, a computer code was developed to calculate icing particle trajectories and impingement limits for axisymmetric inlets. The original research-oriented NASA code was upgraded and modified to meet the requirements of the design engineer. The improved code is capable of performing trajectory calculations for any atmospheric conditions and droplet sizes. It can handle single droplets or a distribution of various droplet sizes. The four programs that comprise the code are described and the results of a test case using flight conditions for a Fokker F100 icing tunnel test are presented.

The literature on sea spray (superstructure) icing is almost entirely based on observations on moving ships. However, icing on stationary offshore platforms with their fixed vertical columns will differ significantly from ship icing, which is influenced by ship movement and wind and wave directions. An observation program on offshore drilling vessels is proposed, using 1-in.-diam x 8-in.-long cylinders in arrays as a standard measuring technique for spray icing. Atmospheric icing may be a source of ice accretion on derricks in some locations, and the best commercial device currently available for measuring it is the Rosemount detector. Improved devices for both spray and atmospheric ice accretion measurements should be developed. Icephobic coatings have the potential for reducing ice accretion, and testing of candidate materials should be undertaken. Well-documented icing reports by all types of ships or platforms should be made and collected at a central clearinghouse.

Flight tests were made in natural icing conditions with two 8-ft-chord heated airfoils of different sections. Measurements of meteorological variables conducive to ice formation were made simultaneously with the procurement of airfoil thermal data. The extent of knowledge on the meteorology of icing, the impingement of water drops on airfoil surfaces, and the processes of heat transfer and evaporation from a wetted airfoil surface have been increased to a point where the design of heated wings on a fundamental, wet-air basis now can be undertaken with reasonable certainty.

Surfaces exposed to atmospheric cold temperature and humid environments are prone to ice accretion. Airplanes, electrical power transmission cables, and wind turbines are typical examples for which icing has to be considered. The measurement of the resulting ice shapes is a challenging process. While macroscopic characteristics of the ice geometry can be observed using photography and optical scanning techniques, microscopic measurements are difficult to conduct because grooved surface partially occludes the geometry of chasms. To overcome this optical inaccessibility, we propose a method to carry out detailed high-resolution measurements of the accretion surface with microcomputed tomography. This approach provides a unique visualization of the empty spaces in the feather region. The information obtained by this technique can improve the understanding of ice accretion physics and its computational modeling.

The paper presents results of the numerical simulation of a hot-air anti-icing system model. A 20 Navier-Stokes CFD code is used to simulate jet impingement on (a) a flat plate and (b) th e inner surface a slat of a multi-element airfoil, a modified RAE 2822 airfoil. The j?at plate case is used to validate numerical predictions by comparison with known empirical conelations. Since these correlations are being used in most of the anti-icing simulation codes, the slat case is used to determine their applicability to concave surfaces. The results indicate that the empirical correlations are not n&able enough for use in anti-icing simulations. The CFD code is then coupled to an ice accretion and antiicing simulation code, CANICE. The overall computational prvcedum is presented with the help of an example. The merits of using the CFD tool in conjunction with the CANICE code are discussed. NOMENCLATURE c, = specific heat of air = airfoil chord length ii slot-tc+wall distance/height h = heat-transfer coefficient k = thermal conductivity L W C = liquid water content M = freestream Mach number MVD = median volumetric droplet diameter ti ?z mass flow rate NZ6 = Nusselt number, hs/kf Pr = Prandtl number, pk/Cp ke = heat flux = freestream Reynolds number, V,c/v Res = jet Reynolds number, isS/~ S = hydraulic diameter of slot, 2x width St = Stanton number, h/pre,VrerCP FL = surface arc length measured from origin = temperature Tf = film temperature, (Tw + Tin)/2 V = airspeed, velocity Q = angle of attack relative to chord line P = fluid viscosity v = kinematic viscosity, p/p P = fluid density i.2 = mean velocity at the inlet slot * Postdoctoral Fellow. Member AIAA. t Ph. D. Student Member AIAA. 4 Bombardier Aeronautical Chair Professor. Associate Fellow AIAA. Copyright @ 2000 by Farooq Saeed, FranGois Morency, and Ion Paraschivoiu. Published by the American Institute of Aeronautics and Astronautics, Inc. with permission. Subscripts: anti = anti-icing f = fluid, film in = inlet out = outlet ref = reference s = static t = total w = wall X = local 00 = freestream value INTRODUCTION Atmospheric icing presents a major hazard to aircraft operat ing under natural icing condit ions and is a cause of major concerns for the certification authorities as well as the aircraft manufacturers. The steady rise in the global aviation traffic means an increased likelihood of encounter ing natural icing conditions. This suggest an increased frequency of icing related accidents unless a considerable amount of effort is focused on the various safety issues concerning in-flight aircraft icing. To enhance flight safety under natural icing conditions, FAA has recently initiated a multi-year icing plan’>’ to address the various issues related to in-flight aircraft icing. One of the several key tasks outl ined in the plan is to ensure the validity and reliability of icing simulation/modeling methods currently being used/developed. In an effort to support the objectives of the FAA Icing Plan, and facilitate Bombardier Aerospace in the certification process, the main focus of research under the Bombardier Aeronautical Chair at l?cole Polytechnique, MontrCal, has been the development of a reliable ice accretion and anti-icing simulation code CANICE.3-5 The development of CANICE has

THERMAL protection from the effects of ice built up can be achieved by means of anti‐icing or de icing. Most aircraft structures are de‐iced since this consumes considerably less power, but in some cases it is essential that the surface be anti‐iced, (e.g. engine intakes, where ice ingestion to the engine is unacceptable and any formation of ice must therefore be prevented).

Icing on stationary structures is an increasingly serious problem as off-shore drilling operations in the sub-polar regions becomes more popular. Since this problem is less complicated than icing on a ship, an attempt was made to calculate accretion rate using existing data. The rate of ice accumulation R can be calculated from R = C f C c F where F is the mass flux of the water drops and C f and Cc are the proportions of spray frozen on the surface and coefficient of capture of drops, respectively. Cc can be close to unity for larger drops such as sea spray. Although many other factors may contribute, C f seems to be a strong function of the air temperature. Mass flux can be written as Vr3 dr where n(r) is the number of drops of radius r in unit volume, V is the wind velocity, ρ is the density of water; n(r) is a function of wind velocity and height of observation. For a stationary structure, the mass flux is primarily dependent upon the wind speed. The ice accretion rate R, calculated using published data on the size distribution of sea spray and using a capture efficiency Cc of I agrees surprisingly well with the diagrams given by previous authors for icing on ships.