Experimental Investigation of the Vortex Flow around an AxisymmetricBody with the Five-Hole Probe and Hot Wire Probe

Anaxisymmetric body experiences the vertical flow around itself at incidence angle. If the adverse pressure gradient is significant, the boundary layers separated and a vortex is formed. The flow over a submarine at AOA (angle of attack) has specified separation of boundary layer and large vortex structures around the body. This flow influences body drag, acoustic and maneuverability. A propermethod to decrease and control the impacts of this separated flowis to use vortex generator. The mainobjective of the present study is to investigate the flow field around a Suboff model with applying the vortex generator by using the hot wire and five-hole probe in 0° ≤ α ≤ 20° angles of attack. The novelty of present study is application of two experimental method, (hot wire probe and five-hole probe) which can help to precisely study the structure of three-dimensional vortical flow field, the boundary layer velocity profiles and probability of the separation on the model with and without existence of vortex generator. The results indicate that vortex generators significantly decrease cross-flow separation, the size of vortices and the vortical flow.


Introduction
When separation occurs, the flow in the boundary layer is retarded to a point where it can no longer counteract the pressure gradient and separates from the surface. This results show adverse flow and thickening of the boundary layer downstream of the separation point. If the boundary layer flow remains attached, only a relatively small drag due to the skin friction will remain. Commonly employed methods for flow control today provide favorabletool to augment the performance of aerodynamic systems when the boundary layer is separated, or at least close to separated state (Velte 2009).
However, implementing an effective control requires to better understand the vortex and wake behavior in the close to the body and thus to better understand the relationship between the strength of vortices and the drag forces.
It depends on the strength, the size and the path of the vortices in the vortical region as well as on the speed they are moving away from the body. Some theoretical approaches (Collins and Keswani 2004) have been developed taking into account different simplified kinematic motions in order to better understand the flow behavior.
The aft wake of the blunt body like (submarine) is composed of 3Dvortex structures with different flow characteristicthat is given depending mainly on the Reynolds number as the geometry (Xueying et al., 2003, Pantelatos andMathioulakis, 2004). They have a reciprocating interaction with the near wake flow supplying the vortex street generation and being conditioned by the initial generated vortices. Turning maneuvers of blunt body result in flow separation that generates large hydrodynamic forces and moments. The vortex generators as a flow control device can be used to control this cross-flow separation on submarines.The role of the vortex formed by a vortex generator is to provide a way whereby fluid outside the boundary layer flow with higher energy can swirl with the body surface, low-velocity fluid in order to re-energize the boundary layer and delay or prevent the flow separation (Manshadi et al, 2015.
The flow visualization techniques have an important role in analyzing of the flow structures over the body.
Method of the oil flow visualization is a useful technique for identifying of the skin-friction line on the body surfaces (Saidinezhad et al, 2014, Ashok and Smits, 2013, Kumar et al., 2012,Liu et al. (2011), Etebari et al., 2008, Gregory et al. 2007, Mackay 2003, Hosder and Simpson 2001.
Saiedinezhad and colleagues (2012)  Also, the drag force was reduced and the secondary separation was eliminated with the vortex generators.
Review of the previous studies indicates that the five-hole probe was used to study the vortex fields. Maseland et al. (1992) numerically and experimentally studied the flow around a double delta wing.In order to validate their numerical code developed to analyze the flow field to double delta wing, a five-hole probe was used at the apex angle of 45 degrees. By this method, vortical flows induced on the wingswere measuredand the results were compared with each other. Blomhoff (2012) put a model wind turbine into the wind tunnel in order to observe the changes in the sequence of the wind turbine, and by a five-hole probe measured wake of the wind turbine at different section and also measured the various parameters.
The main aim of the present work is the study experimentally the effect of the vortex generators on the flow field characteristic of a Suboff model. The present work is devoted to determine the effect of vortex generators on the flow structures over and around a submarine model in an angle of attack by the hot wire anemometer and five-hole probe at Reynolds number of 1 × 10 6 . In this study, two experimental methods are used and the effect of the vortex generators on the standard submarine model was studied and discussed.

Wind Tunnel for Hot Wire Probe Measurements
The experiments in the present work are conducted in a closed loop subsonic wind tunnel with a test section size of 370×280×1200 mm. The axial fan of the tunnel provides air flow with maximum speed of about 30 m/s and turbulence intensity level of 0.25 % in the test section. The test section has 45 degree corner fillets which enlarge the cross-sectional area gradually in downstream direction to maintain a constant static pressure along the test section, therefore decreasing the buoyancy effects.
The boundary layer velocity profiles in the plane of symmetry on the leeward side (φ=180°) of the submarine were measured by using a single probe of hot wire anemometer (Fig. 1). The single probe with 0.5 mm tungsten wire is mounted on a three-

Wind Tunnel for Five-Hole Probe Measurements
All experimentsin this study have beenperformedinanopen circuit wind tunnel at Yazd University (Fig.2).The wind tunnel has a maximum fan powerof32 Kw and maximumvelocityof30m/s inside the testroom.The wind tunnel test chamber has a cross-sectionof457×457mm 2 and length of 1200mm 2 and has four transparent Plexiglas window glass at dimensionsof 267×572 mm.Inordertoavoid the interference of boundary layers bysidewalls,the corners of 45degreeangle were mounted between twowalls.The size of the cornersalong the test room gets small and it causes the enlargement of the cross-section and destroys partly their floating longitudinal effect along the test room.Coefficient of obstructions or blockage ratio in models plus its kickstand in the wind tunnel test chamber is equal to 4.9%percent.That's less than the standard value (5%) and therefore is reliable.Allof this researchtests is at velocityof 16 m/s.Turbulenceintensityin the test chamber is about 0.25percent.
In order to use the five-hole probe,itisneededtomeasurethefive pressure channels.For this reason, a 15channelpressure converterdeviceis used. This pressure sensor device is capable ofmeasuringthe maximum differential pressure±1270Pa at accuracy of±3Pa.Measured Data are entered directly through a 12-bit analog to digital voltage converter card into the computer and the necessary software and computing has been done on them.
According to the five-hole probe relations, measuring the relative uncertainty for pressure and average velocity are obtained of 0.16 and 8.7, respectively.

Introducing the five-hole probe and its calibration
Multi-hole probe is one of the sonic laboratory equipment to determine the factors of flow. The use of multi-hole probes began from the second decade of the 20th century and continues today. The probes are used not only in the laboratory but also are used in the aircrafts and helicopters. Multi-hole probes are used to determine the components of flow velocity and total pressure. The probe can be used to study three dimensional vortical fields. According to the five-hole probe advantages compared to other measurement methods, the study of a five-hole probe is applied to evaluate the three dimensional vortical fields.
The probe isbasedonthe operating principle ofthe pressure distribution on the surface in front of a circular object that is placed in fluid flow.Pressure distribution on the surface has a maximum value at the point of stagnation flow and pressure on the surface of the model at thedownstream reduced. Maximum pressure for round objects is the total pressureof 0 p that is sum of static pressure ( p ∞ ) and dynamic pressure ( q ) away from the body and the lowest pressure is in the region where the slope body is parallel to the direction of flow.According to this point it canbemeasuredbydifferent pressure points on the surface of a circular object, the angle of the flow in respect to the object can be determined.The probe is intended for experimental study of five metal pipes with outer diameter1mm.A schematic of the probe is shown in the Fig.3. These tubesare cut at an angle of 45 degrees and placed side by side.
A five-hole probe must be calibrated before using.Calibration means that the relationship between measured pressure and flow angles by the probeisobtainedin the measurement location.Theserelationsarederived for dimensionless coefficientsthatcanbe usedaccording to the flowangle.After the calibration process, measuringfive pressures by a five-hole probe should be conducted to determine the flow angle. Calibration of the airflow is done by an open circuit wind tunnel. To gain a uniform flow a honeycomb mesh is used in the wind tunnel.In this study to calibrate five-hole probe is the following: 1) the mechanism of the probe angle 2) direct flow generated in the wind tunnel test section, 3)pressuremeasuring system,4) Software for obtainingdata. Fig. (4) shows the five-hole probe calibration equipment. Fig. (4a) indicates five-hole probe, (4b) 15-channel pressure transducer) and (4c) mechanisms of the probe angle.
The probe pressure values are takenand by using the static pressure and total pressure getting from pitot probe the dimensionless numbersare obtained.This processis repeatedfor different yaw and pitch angles.The calibration curve of the probe is obtained based on four parameters of yaw and pitch and dimensionless coefficients.Since pitch and yaw angles can be attributed to the read pressures by different holes, it is necessary to determine the condition of pressure dimensionless coefficients.There are several ways to define the pressure dimensionless coefficients.In this study, the relationship proposed by Treaster and Yocum (1979) is used.Dimensionless pressure coefficients are defined for equation (1). (1)

+ + + =
In these relations the parameters ofCPαandCPβ are dimensionless pressure coefficient for flow angle vertically (pitch) and horizontally (yaw) respectively that through those probe angles in the calibration phase and flow angles in step recovery flow parameters are associated to five-hole pressure probe.Also CPtotalparametersandCPstaticare thetotal and staticdimensionless pressure coefficients created by these coefficients of total and static pressure and can be achieved at any point mentioned using two pressures mentioned and by using equation (2) the velocity is gained at each point. ( Generally, five-hole probe calibration procedure is so that first five-hole probe within an open circuit wind tunnel is installed on the angle device andprobepressure holes are connected to electronic transducer.Then putting the probe indifferent angularpositionsofthe pressure in each measurement and using relationships described in (1) the pressuresare converted at any pointinto dimensionless pressure coefficients.As noted above, after calibrating the process and obtainingdimensionlesscoefficientsare plotted on a curve as well.The resulting curve is known as a calibration curve.In Fig.5 calibration diagram obtained from five-hole probe calibration curve is shown at the inlet flow velocity16 m/s.

Testing process
The general test trend is so that after verifying five-hole probe, the probe shallbe placedin a wind tunneland the flow passing through the model by shifting the five-hole probe through the points in a perpendicular plate to the flow, the pressure values for each hole are read andrecorded by a pressure transducer.
Tocheckthe accuracyof measurementsperformancebya present experimental studythe reproducibility of results and values of uncertainty was carried out. In following to investigate each of thesepartswillbe discussed.

A) Reviewof Results Repeatability
In experimental studiesthe measurementofvalues usually are different at the each iteration.But each timemeasurementresults will be as reliableas possible, difference is less than to each other.It has been installed inside the tunnel while thedataforaline in the flow field around the modelweremeasuredtwice, at two differenttimes.As it canbe seenin both testsmeasuredpressuredifferencesare notsignificantand are much closer to each other.Sowecanclaim that they are repeatable experiments.To repeat calibration curve, three tests were conducted and the results were compared for all three tests.The resultsfor 6 . 21 = α in all three tests are shown in the Fig.6.As canbeseen,there is very little difference between the results and theconclusion that they are repeatable experiments conducted by researchers with good accuracy.

B) Determinationof Uncertaintyin Measurement Parameters
The measuring process ofa physical quantity is always contaminated to some error.Lack of knowledge about the size and sign of the error in the measurement is called by uncertainty.Estimation of uncertainty, description and evaluation of themeasurementerroris a statisticaland results of measurementonly to complete when a quantity of uncertainty is described.

B.1) uncertainty of pressure coefficients ,
According to the definition ofdimensionlesscoefficientsof pressure to five-hole probe in equation (1), it canbe calculated using equation (3) level of uncertaintyachieved for each of thequantities.For example, the uncertainty associated the quantityzisas a function of two independent quantitiesx and y definedasfollows:

B-2) Uncertainties of the anglesα and β
The amount of velocity isobtained at every point of equation (2). The uncertainty velocity by using equation (3) isobtainedfrom the followingequation: Kg per cubic meter for the uncertainty density and maximum differential pressure of 256 Pa for total and static pressure, the uncertainty of velocity is1.47meters per second and it'swill be relatively equal to0.087.

Experimental Results (five-hole probe)
The streamlines patterns and vorticity contours for the model without the vortex generators for the plane YZ for α=20 o is shown in Fig.8. At X/L=0.5, the streamline patterns show the separation of flow field from the suction side and the two counter vortices be larger at the downstream of the body. Also, core of the vortex moves away from the body surface result inform the circumferential pressure gradient to interact with the vortex core and it's diffuses and becomes larger. This is seen at X/L=0.7.

Experimental Results (hot wire probe)
In order to investigate the effect of the vortex generators around the submarine model, in addition to the five-hole probe, the hot wire anemometry is performed. Using the hot wire probe, the distribution of the velocity profile of boundary layer, turbulence intensity and instantaneous velocity can be obtained at different angles of attack and also, the effects of using the vortex generators on the flow structure can be studied.
In this section, the velocity profiles of boundary layer on the symmetry plane of the submarine nose (φ=180°) are studied for Re = 10 6 at two angels of attack α=0 and 10°. Moreover, the boundary layer profile in the location X/L = 0.9 is measured to compare with Huang and Liu result. Fig. 13 shows the boundary layer velocity profile in location X/L = 0.9 on the leeward symmetry plane (φ=180°) for suboff model at zero pitch angle and Re = 10 6 . The result of Huang and Liu is also shown in this figure. The velocity profile of the boundary layer (U/Ue) depicts in nondimensional distance from the surface (Y-R0/RMax). Comparing the present results with the experiments shows that the trend of the velocity profile of the boundary layer are very well predicted by present work, but the differences are seen in range of distance 0.5≤(Y-R0/RMax). In the present work, Ue is the velocity at the edge of the boundary layer in position X/L=0.9 but Huang and Liu used the velocity of the flow outside of the boundary layer in position X/L=0.75 that its value is unknown. This may be the reason of the discrepancies between the results. Also, this figure shows the model with the vortex generators has small vortical region and thus a small separation zone. It means the model with the vortex generators is more effective for controlling the boundary layer separation and prevents the flow separation from the hull compared to the model without the vortex generators. Fig. 15 shows the variation curves of the axial wake against the circumferential angle with and without the vortex generators at X/L=1.1. Fig. 15 shows that the vortex generators reduce the variation extent of the axial wake velocity. It suggests that the vortex generators can weaken the horseshoe vortex and improve the uniformity of the wake at the propeller disc, which is beneficial to the working propeller.  models. According to the diagrams it has been shown that the instantaneous velocity of the model without the vortex generators increases. This increase is due to growth of flow vortices and flow separation because of horseshoe vortex that created by presence of sail on the model. It also is seen that the use of the vortex generator helps to control the vortical flow and prevent increasing the moment speed. It means that using of the vortex generators can decreases the horseshoe vortex and control the flow.
In Fig. 18, the diagram has shown the turbulence intensity for the vertical distance of surface model. As it is clear, distancing from the surface first the turbulence intensity increased then decreased to get the free stream turbulence intensity. In addition, according to the figure it has observed that the vortex generators reduce the turbulence intensity around the model. In fact, as mentioned earlier, the vortex generators, controlling the boundary layer flow and prevent the vortex flow separation (which in this area it has shown high turbulence intensity). In fact, at the stern of the model is the location of the propeller, and the vortex generators by controlling the flow separation reduce noise and voricity which in turn leads to increased efficiency and reduced noise of the propeller and drag force.

Conclusion
In the present study, to investigate the physics of the flow field structures around the submarine model is investigated experimentally with two technique, five-hole probe and hot wire probe. The results of the five-hole probe show patterns of the vortical flowaround the model. The results of the hot wire probe show the velocity profile of boundary layer, turbulence intensity and instantaneous velocity.The major conclusions are summarized as follows: • At the angles of attack, the vortex generators reduces the vortical flow region and the separation zone. Note that the vortex generators can energize the boundary layer prevent the growth and strength of the vortices around the body. The drag force is also decreased by using the vortex generators.