Mechanical Actuators for Guidance of a Supersonic ... - CFD4Aircraft

Guided bullets to intercept mortarsAuthorized for public release. Distribution Unlimited.06/21/2007 4

Ball M33 Lab TestsOptical Verification of changesin flow due to mass injection atNoseCan Alter40mm Scaled Ball M33USAFA Tunnel / CFDMass Injection : Nose,Midbody, BoattailLong and Short BoattailMidbodyMeasurements of mass flowrequired to affect flowFlowJetLongBoattailBestAft MassInjectionBest40mm ARL ProjectileU. Texas Tunnel / CFD – forces and momentsAft Mass Injection : Tangential and NormalNormalMass InjectBetterProjectileUnstableAuthorized for public release. Distribution Unlimited.06/21/2007 5

Concept of implementing flow control near finActuators create shock patterns that impingeon fin and body surfaces to produce timedependent control forces.Roll controlAft View2 pins close to adjacent finsInduce Angle of Attack ChangePitch controlAft View2 diametrically opposed pinsInduce clockwise rotationM = 4.0AOA = 0°Authorized for public release. Distribution Unlimited.06/21/2007 6

Fin InteractionsGTRI TunnelMeasured forces generated by pins and mass injection in region near finy (in.)Total DifferentialForce: 1.46 lbsx (in.)Pitching Moment (N-m)151050-5-10Moment as Function of AOACg @ 90.2 mmCg @ 123.5 mm-15-15 -10 -5 0 5 10 15AOAPins Near FinsGenerateStrong TurningMoment½ Body Test RigToo MuchVolumeRequired forMass InjectionInput fromARL onSize, Shape,Mass Dist.Need System Study toDefine:Cg locationFin Shape/SizeRoll or Fin Stabilized?Actuation Concept andPreliminary Design3D EffectsSteering Force(Steady & Unsteady)CFDUnderstand FluidDynamicInteractionsPreliminaryActuatorsAvailableSteeringForceFire Roundat ARLTest Conceptand HardwareAuthorized for public release. Distribution Unlimited.06/21/2007 7

Parametric Study DetailsPin location test matrix 9 x 10 Matrix (90 locations) 0.55 in spanwise x 0.88 in streamwise Force measurements made for all pins atall points except trapezoid, whichexperienced structural failure 271 unique tests performed 1300 + data points (each locationperformed 3 times)Forces on fin directly measured asopposed to pressuremeasurementsFlowOriginal Test Matrix(March 2003)Extended Test Matrix(March 2004)M 1.7 C/D NozzleFlat PlateFin(-0.650, 1.065)(-0.1, 0.185)(0,0)Authorized for public release. Distribution Unlimited.06/21/2007 10

3-D D Contours of Force DataRectangle 0.2 Round 0.2 Round 0.1• Contour plots of side force vs pin location show same trend for all pins• Clear evidence of optimal regions for pin location Implies there is leeway in placement of pin Important as mechanical/space restrictions may not allow for location at optimal location Relative force forflat pin larger than round with same frontal area This likely due to stronger shock (no 3-D relieving effect)• Hypothesis that optimal location should scale with pin diameter, was proven wrong (compare 0.1 and 0.2 dia pins) The 3-D shock interactions are complex and do not lead to simple scalingAuthorized for public release. Distribution Unlimited.06/21/2007 11

Effect of Separation Distance (between(pin and fin)• Dividing the force by the frontal area of the pinprovides a 1 st order collapse of the magnitude• Several different parameters were explored todetermine the effect of separation distance The distance from the edge of the fin to the centroid of the pinprovided the best collapse Optimum separation distance appears to be about 0.41-0.42 in• Plots are at Y = 0.775inGap (pin-fin)Dist to CentroidPin DiameterAuthorized for public release. Distribution Unlimited.06/21/2007 12

Effect of Pin Geometry• For same frontal area, rectangularpin gives most force Has least 3-D relieving effect Seems to outweigh additional sideforcegenerated on trapezoidal pin• Optimal (X,Y) location independent ofpin geometry• Enough trapezoid data acquired(before structural failure) todemonstrate that flat pin is betterSide Force (lb)2018161412M 2.475 C/D Nozzle DataY = 0.565 in.TrapezoidRectangleRound1080.35 0.4 0.45 0.5 0.55 0.6X (in)Authorized for public release. Distribution Unlimited.06/21/2007 13

Mach 2.5 Experiments at GTRI½ Projectile (full scale)Ground PlaneMach 2.5C/D Nozzle50 mmLinear AirBearingShaft for modelrotationForce Sensor formoment measurementForce Sensors for side force measurementPressure Transducer andSignal ConditionersAuthorized for public release. Distribution Unlimited.06/21/2007 14

Effect of Pin Height8Force on ProjectileTrapezoid Pin 22.5 deg OrientationM = 2.47, Half Body10Moments Produced by Trapezoid PinM = 2.47, Half Body, 22.5 deg OrientationForce (lb)6420-2-4-6Increasing PinHeightFlush1 mm2 mm3 mm3.3 mmMoment (in-lb)50-5-10-15Increasing PinHeightFlush1 mm2 mm3 mm3.3 mm-8-6 -4 -2 0 2 4 6AOA-20-6 -4 -2 0 2 4 6AOA• Force dominated by AOA of projectile• Non linear effect of pin height on moment• Projectile should be rotate to about 5 degrees with pin deployedAuthorized for public release. Distribution Unlimited.06/21/2007 15

Second Generation Actuator Rotates into Flow• Rocker Pin Hardware Installed in WindTunnel Scale Model Rotation solves stiction problem• Consisted of Rocker Pin Assembly Pneumatic Cylinder Small Valve• Further work needed Not g-hardened Valve still too large• Using 90 psi Very large holding force Response time on order of 10 ms Rotates projectile over 4 degreesClick to PlayMovieAuthorized for public release. Distribution Unlimited.06/21/2007 16

Experimental Input to CFD• Experiments showedWhere to place guidance pinsEffects of pin geometry‣ Including material failure (not from CFD)Crude force measurementsMechanical design considerations (not from CFD)• Need CFD to complete pictureLittle flow understandingBetter drag and force measurementsUse full 3-D body• Combine EFD and CFD to predict Range TestsAuthorized for public release. Distribution Unlimited.06/21/2007 17

Using CFD to Predict Range Test Results• Drag and Roll Torque Predictedusing CFDAllowed for estimation ofperformance in rangeFewer shots required as we knew howmany rotations to expect downrangeM = 4.0AOA = 0°Authorized for public release. Distribution Unlimited.06/21/2007 18

ARL Range Tests to Measure Roll Torque• ½ Scale Projectiles Fired from 1 inch GunQuantify Rolling MomentsProvide Results for Validating CFDProvide More Accurate Aero Coefficients to 6 DOF• Total shots fired: 15 rounds 3 with no pins‣ 1 at Mach 3‣ 1 at Mach 2.5‣ 1 at Mach 2 3 with long pins (0.1 in height) at Mach 3 9 with short pins (0.07 in height)‣ 3 at Mach 3‣ 3 at Mach 2.5‣ 3 at Mach 2Picture of test facilityAuthorized for public release. Distribution Unlimited.06/21/2007 19

ARL Range Test Setup• 6 Orthogonal X-ray XStations Near Muzzle Showed that Sabot Separated Cleanly• 35 Shadowgraph Stations – to 100 m Downrange Generated Images that were used to determine;‣ Roll and Pitch Damping‣ Drag‣ Number of Revolutions – Spin Rate25-mm smoothbore gunAuthorized for public release. Distribution Unlimited.06/21/2007 20

Test Articles• Projectiles ½ Scale (25 mm)• Pins were round 1/16 th in diameter on opposing fins• Nylon SabotLong Pin Test ArticleTungstenAluminum25 mmNylon SabotShort Pin Test Article12.5 mmAuthorized for public release. Distribution Unlimited.06/21/2007 21

Shadowgraphs from Range – Count Rotations• The rotation of the round as it traverses the range can be tracked via a spin pin• The rotation rate leads to a measurement of roll torque developed by pinsPin Used to CountRevolutionsStations 22 and 276.7m to 8.2mLittle Spin ObservedAuthorized for public release. Distribution Unlimited.Stations 295 and 30090m to 91.4mOver 90° rotation06/21/2007 22

Range Test Comparison with CFD• Comparison with measured data not as good as expectedDrag under predicted at all Mach numbersRoll torque prediction worse as Mach number increasedAuthorized for public release. Distribution Unlimited.06/21/2007 23

What went wrong?• Compromises in machining small rounds led to significant differences between CFDgeometry and test rounds• New grid generated and new runs accounting forFin leading edge bluntnessFillet at base of finRound pin versus RectangularAuthorized for public release. Distribution Unlimited.06/21/2007 24

Comparison with Updated Geometry• Once a more accurate geometry was modeled, a much better correlation was foundbetween the computed and measured drag and roll torques• Allowed us to proceed with divert test on full scale rounds00.4-0.010.35Roll Torque (N-m)-0.02-0.03-0.04Long Pin (Range-Prodas)Long Pin (CFD)Short Pin (Range-Prodas)Short Pin (CFD)Drag Coefficient (C d)0.30.25Long Pin Cd (Range-Prodas)Long Pin Cd (CFD)Short Pin Cd (Range-Prodas)Short Pin Cd (CFD)-0.051.8 2 2.2 2.4 2.6 2.8 3Mach Number0.21.8 2 2.2 2.4 2.6 2.8 3Mach NumberAuthorized for public release. Distribution Unlimited.06/21/2007 25

700 ft Range Preliminary Tests• Outdoor Range• 75 mm smooth bore gun• Yaw cards set up• Problems encountered Stability Sabot SeparationTry doing this in CFDAuthorized for public release. Distribution Unlimited.06/21/2007 26

Sabot and Launch Package Resolution• New set of rounds made with increased static margin• Cup scored more deeply• Aluminum pusher plateAuthorized for public release. Distribution Unlimited.06/21/2007 27

700 ft Range Tests – Divert DemonstrationAim PointSeries of yaw cards show that projectile is clearly divingdown due to pins deployed after launch.Authorized for public release. Distribution Unlimited.06/21/2007 28

Transonic Spark Photograph Layout• 5 groups of 5 stations• Each Station provides Shadowgraphs for Vertical Plane – Wall Horizontal Plane - PitProjectileShadowgraphGroup 1 Group 2 Group 3 Group 4 Group 53 mNominal Pin Deployment39.5 m 45.8 m162 mAuthorized for public release. Distribution Unlimited.06/21/2007 29

Divert Demonstrated by ShadowgraphsAuthorized for public release. Distribution Unlimited.06/21/2007 30

Demonstrated High G Turn on Stable ProjectileStable projectile for testing (1.5 caliberstatic margin)~14 g divert maneuver10.4 g verticalacceleration~80 N force created by control pinsPreliminary data reductionMore data will be available in the nearfutureConcept promising for high g maneuvers9.4 g horizontalaccelerationAuthorized for public release. Distribution Unlimited.06/21/2007 31

Conclusions• A demonstration of steering a Mach 4 projectile using the guidance pins wassuccessful• The combined CFD and Experimental efforts led to a greater understanding ofthe effects of the pinsEFD and CFD each used to get different but required forces and moments‣ Results could have been easily done without IFDThis in turn allowed us to better predict the results of the range tests• Less range tests were required because once the predictions were validated, itwas proven we understood the aerodynamicsThis saved substantial amounts of money‣ $10,000 bullets and 5 range operators and 2 PhDs add up fast‣ (As does destruction of the ADT alarm box)Less Bullets Less $$ IFD=GOODAuthorized for public release. Distribution Unlimited.06/21/2007 32