sonic boom calculations for the n+2 configuration · conventional cae solution with grid ... 0 50...

29
Sonic Boom Calculations for the N+2 Configuration Mark D. Sanetrik NASA Langley Research Center \ Presented at ASE Summit NASA Ames Research Center April 14, 2015 https://ntrs.nasa.gov/search.jsp?R=20160007032 2018-07-06T23:54:24+00:00Z

Upload: buidang

Post on 07-Jun-2018

215 views

Category:

Documents


0 download

TRANSCRIPT

Sonic Boom Calculations for the N+2 Configuration!

Mark D. Sanetrik NASA Langley Research Center \

Presented at ASE Summit NASA Ames Research Center

April 14, 2015

https://ntrs.nasa.gov/search.jsp?R=20160007032 2018-07-06T23:54:24+00:00Z

Outline

•  Introduction •  Requirements for Sonic Boom Calculations •  Challenges of Sonic Boom Calculations •  Sample Solutions •  Sonic Boom Results

–  Effect of Grid Resolution –  Effect of Domain Extent –  Effect of Near-Field Pressure Extraction Location –  Effect of Static Aeroelastic Deflections –  Effect of Engines

•  Concluding Remarks and Future Work

Introduction

•  Low sonic boom flight a major requirement for N+2 Program

•  Extensive work done to optimize sonic boom signature

•  Optimization done on a rigid structure

•  Structure will deflect aeroelastically under aerodynamic loads

•  How will static aeroelastic deflections change the sonic boom signature?

Requirements for Sonic Boom Calculations

•  Conventional Static Aeroelastic domain is rectangular

•  Body is defined at 0o angle of attack

•  Same grid can be used for multiple flow conditions

•  Sonic Boom domain must be shock aligned

•  Conical-shaped domain with angle of cone dictated by Mach angle

•  Body must be rotated to proper angle of attack

•  Separate grid for each flow condition

CAE Computational Domain

CAE Computational Domain Detail

Sonic Boom Computational Domain

Sonic Boom Computational Domain

Challenges of Sonic Boom Calculations

•  Separate grid for every flow condition

•  Extremely fine near-field grid required

•  Where to extract pressure signature?

–  Too close and the pressure signature not fully developed

–  Too far away and the pressure signature is dissipated

–  Current practice is 3 body lengths

•  Use of external program to propagate shock to ground level (sBOOM)

•  Interpretation of ground signatures

Conventional CAE Solution Inviscid, No Engines, Mach = 1.70, α = 2.25ο

11.8 MGP, CL = 0.14343, CD = 0.09220

Conventional CAE Solution with Grid Inviscid, No Engines, Mach = 1.70, α = 2.25ο

11.8 MGP, CL = 0.14343, CD = 0.09220

Sonic Boom Coarse Solution Inviscid, No Engines, Mach = 1.70, α = 2.25ο

15.9 MGP, CL = 0.14342, CD = 0.09252

Sonic Boom Coarse Solution with Grid Inviscid, No Engines, Mach = 1.70, α = 2.25ο

115.9 MGP, CL = 0.14342, CD = 0.09252

Sonic Boom Medium Solution Inviscid, No Engines, Mach = 1.70, α = 2.25ο

46.5 MGP, CL = 0.14343, CD = 0.09251

Sonic Boom Medium Solution with Grid Inviscid, No Engines, Mach = 1.70, α = 2.25ο

46.5 MGP, CL = 0.14343, CD = 0.09251

Sonic Boom Fine Solution Inviscid, No Engines, Mach = 1.70, α = 2.25ο

89.9 MGP, CL = 0.14342, CD = 0.09256

Sonic Boom Fine Solution with Grid Inviscid, No Engines, Mach = 1.70, α = 2.25ο

89.9 MGP, CL = 0.14342, CD = 0.09256

Sonic Boom Grid Resolution Study

5000 6000 7000 8000 9000x (in.)

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

Coarse GridMedium GridFine Grid

Near-field Pressure SignatureNo Engines, Undeformed, Mach = 1.7, Alpha = 2.25

Sonic Boom Grid Resolution Study (cont.)

0 50 100 150 200 250 300time, ms

-1

-0.5

0

0.5

1

1.5

Ove

rpre

ssur

e, p

sf

Coarse GridMedium GridFine Grid

Ground Level Boom SignatureNo Engines, Undeformed, Mach = 1.7, Alpha = 2.25

Sonic Boom Domain Size Study

5000 6000 7000 8000 9000 10000 11000x, inches

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

dp/p

Coarse Grid Short DomainCoarse Grid Long DomainMedium Grid Short DomainMedium Grid Long Domain

Near-field Pressure SignatureNo Engines, Undeformed, Mach = 1.7, Alpha = 2.25

Sonic Boom Domain Size Study (cont.)

0 50 100 150 200 250 300time, ms

-1

-0.5

0

0.5

1

1.5

Ove

rpre

ssur

e, p

sf

Coarse Grid Short DomainCoarse Grid Long DomainMedium Grid Short DomainMedium Grid Long Domain

Ground Level Boom SignatureNo Engines, Undeformed, Mach = 1.7, Alpha = 2.25

Sonic Boom Extraction Position Study

2000 4000 6000 8000 10000x (in)

-0.04

-0.03

-0.02

-0.01

0

0.01

0.02

dp/p

1.00 Body Length1.38 Body Lengths2.00 Body Lengths3.00 Body Lengths

Near-field Pressure SignatureNo Engines, Meduim Grid, Undeformed, Mach = 1.7, Alpha = 2.25

Sonic Boom Extraction Position Study (cont.)

0 50 100 150 200 250 300Time (ms)

-1.0

-0.5

0.0

0.5

1.0

1.5

Ove

rpre

ssur

e (p

sf)

1.00 Body Length1.38 Body Lengths2.00 Boby Lengths3.00 Body Lengths

Ground Level Boom SignatureNo Engines, Medium Grid, Undeformed, Mach = 1.7, Alpa 2.25

Sonic Boom Effect of Deflections Study

5000 6000 7000 8000 9000x (in.)

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

dp/p

UndeformedDeformed

Near-field Pressure SignatureNo Engines, Medium Grid, Mach = 1.7, Alpha = 2.25

Sonic Boom Effect of Deflections Study (cont.)

0 100 200 300Time (ms)

-1.0

-0.5

0.0

0.5

1.0

1.5

Ove

rpre

ssur

e (p

sf)

UndeformedDeformed

Ground Level Boom SignatureMedium Grid, No Engines, Mach = 1.7, Alpa 2.25

Sonic Boom Effect of Engines Study

5000 6000 7000 8000 9000x (in.)

-0.03

-0.02

-0.01

0

0.01

0.02

0.03

No EnginesEngines

Near-field Pressure SignatureMeduim Grid, Undeformed, Mach = 1.7, Alpha = 2.25

Sonic Boom Effect of Engines Study (cont.)

0 50 100 150 200 250 300Time (ms)

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

Ove

rpre

ssur

e (p

sf)

No EnginesEngines

Ground Level Boom SignatureMedium Grid, Undeformed, Mach = 1.7, Alpa 2.25

Concluding Remarks

•  Sonic boom signatures have been computed for the N+2 configuration •  Still a work in progress; much learning to be done •  Major questions to be answered

–  Exactly what is the configuration are we analyzing? –  What is the best procedure for static aeroelastic computations?

•  Future Work –  Additional computations with engines –  More realistic engines? –  Viscous Calculations

•  A good start has been made in the investigation of the aeroelastic effects on the sonic boom signature