Orion E-STA Acoustic Test: Evaluating Predictions Against Data

Transcription

1 Orion E-STA Acoustic Test: Evaluating Predictions Against Data Samantha Bittinger NASA Glenn Research Center Cleveland, OH LMD/Structural Dynamics Branch June 20, The Aerospace Corporation 2017 Spacecraft and Launch Vehicle Dynamic Environments Workshop El Segundo, CA June 20-22, 2017

2 Introduction & Background In May 2016, the Orion European Service Module Structural Test Article (ESM STA or E-STA) Acoustic Test was performed at the Reverberant Acoustic Test Facility (RATF) at NASA Glenn Research Center s (GRC) Plum Brook Station in Sandusky, Ohio Purpose: Dynamically test the ESM, which is the propulsion and power module of Orion All test objectives were met Verified mechanical resistance of the ESM to acoustic environment Verified the ESM modeling Validated the ESM components random vibration test levels Qualified the Solar Array Wing (SAW) Tested the empty (dry) propellant tank configuration Great collaboration between ESA, Airbus, Lockheed Martin, and NASA 2 This presentation will focus on the pretest prediction models and how they compared to test data

3 Test Article The E-STA article is comprised of: Flight-like ESM Includes one flight-like solar array wing (SAW) most acoustically receptive ESM component Heritage design hardware Fairing (SAJ) Crew Module Adapter (CMA) Spacecraft Adapter (SA) Crew Module (CM) mass simulator on top Blue steel derrick for mass Wooden heatshield as forward acoustic closeout MPCV Stage Adapter (MSA) Simulator as aft acoustic closeout Not pictured here 3

4 Vibroacoustic Prediction Models Two E-STA test-configuration vibroacoustic models created by Mike Yang/ATA Engineering with collaboration from NASA GRC Boundary Element Analysis Model (BEA or BEM) Low-mid frequency, Hz predictions Statistical Energy Analysis Model (SEA) High-frequency, 315-8,000 Hz predictions Acoustic models based on E-STA FEM (pictured) The FEM included updates based on prior testing E-STA ESM v1.5 Structural Stiffness Test Correlated SA & SAJ Correlated CMA Plum Brook Building Block Modal Test Partially correlated CM Simulator Plum Brook Building Block Modal Test 4

5 SEA Model Structure 5 Interior View Exterior View

6 SEA Acoustic Cavities CMA (Avionics boxes) SM Forward (Prop tanks) SM Outer (Solar arrays, radiators) MSA (Engine bell) 6

7 SEA Hybrid FE Subsystems Longerons, solar array supports, and T0 panel are modeled with FE, and incorporated into SEA model 7

8 SEA Model SEA Structural Subsystems (Flat Plates, Curved Shells, Beams) ATA Engineering determines panel properties and beam cross-sections through a rigorous averaging process Flexure Damping Loss Factor (DLF): 4% through 800 Hz, drops to 2% by 2,000 Hz Extension and Shear/Torsion DLF: 1% 30 SEA Acoustic Cavities 1% Absorption Hybrid Model: 9 FE Subsystems (longerons, SAW mounts, T0 panel) 4% DLF through 800 Hz, then drops off to 2% DLF by 2,000 Hz 2,398 modes imported (0-2,100 Hz range) > resonant and masscontrolled modes included 1000 resonant modes used around each solution frequency Diffuse Acoustic Field (DAF) excitation sources on exterior structure Solve Parameters 315-8,000 Hz 1/3 octave bands 8

9 BEA Model (BEM) 9

10 BEA Model 1 global Finite Element Structural Subsystem Structural DLF: 4% (constant over frequency) 598,441 structural nodes 6,176 modes imported (0-400 Hz range) 23 watertight bounded cavities (acoustically wetted surfaces) 1% Absorption DAF excitation sources on all exterior FE faces Sensors Pressure sensors placed at all E-STA mic locations Velocity sensors placed at all accelerometer locations Solve Parameters Hz 1/12 octave bands 10

11 Chamber Input Level: Equivalent DAF Methodology Motivation The vibroacoustic load cases from the most recent flight cycle ( v7 ) feature propagating waves, which cannot be directly replicated in the reverberant acoustic chamber Originated from Space Launch System (SLS) scale model acoustic testing at NASA Marshall Space Fight Center Need to create a diffuse acoustic field (DAF) that will produce the same internal cavity environment behind the fairings as v7 predictions (this cavity is called SM Outer Cavity) v7: envelope of 5 load cases (liftoff, transonic, max Q, liquid engine start, hold down) SM Outer Cavity chosen because it is the main driver of the ESM s structural responses Equivalent DAF Creation Run 100 db flat external DAF on E-STA model (BEM for low-freq, SEA for hi-freq) Scale the resulting E-STA SM Outer Cavity environment up to match the predicted v7 SM Outer Cavity environment envelope Scale up the 100dB input DAF by the exact same db amounts > Equivalent DAF 3dB added for Qualification Level Smooth the curve Applying this external Equivalent DAF to the E-STA model simulates a flightlike SM Outer Cavity environment 11

12 SPL (db, re: 20uPa) Input DAF vs. Test-Achieved DAF Black: Final input spec (derived from Equivalent DAF) Purple: Test-Achieved Control Average 8 control mics, full level test, dry prop tank configuration RATF was able to match the spec within tolerance over the full frequency range Particularly good match from Hz OTOB For post-test comparisons, predictions were scaled by the control mic average AC0243, s linearly averaged average control at 00:01:00 [db] Nominal 0dB SPL Spectrum (db, re: 20uPa) UPPER SPL TOLERANCE (db, re: 20uPa) LOWER SPL TOLERANCE (db, re: 20uPa) OTOB Center Frequency (Hz)

13 Sound Pressure Level, db (re 20 μpa) Overlays Test vs. Prediction 140 SM Outer Cavity SM Outer Test Average (No Vents) OASPL SM Outer Prediction OASPL /3 Octave Band Center Frequency (Hz)

14 Sound Pressure Level, db (re 20 μpa) Overlays Test vs. Prediction 140 MSA Cavity MSA Test Average OASPL MSA Prediction OASPL /3 Octave Band Center Frequency (Hz)

15 PSD (g 2 /Hz) Overlays Test vs. Prediction 1.E+01 SAJ Panel 3 Center 1.E+00 1.E-01 1.E-02 1.E-03 A760 Test Data 1/12 BEA Prediction 1/3 BEA Predictions SEA Prediction 1.E Frequency (Hz) 15

16 PSD (g 2 /Hz) Overlays Test vs. Prediction 1.E+01 Flightlike Solar Array Wing Panel (Normal Direction) 1.E+00 1.E-01 1.E-02 1.E-03 1.E E-05 1.E A525-A243Z (QL) - (Grms=15.43) Prediction 16.7 GRMS /3 Octave Band Center Frequency (Hz)

17 PSD (g 2 /Hz) Overlays Test vs. Prediction 1.E-01 OMS Engine Mounting Point (Normal Direction) 1.E-02 1.E-03 1.E-04 1.E-05 1.E A237-A201Z (QL) - (Grms=2.24) 17 1.E-07 Prediction 2.7 GRMS /3 Octave Band Center Frequency (Hz)

18 Conclusions In general, predictions matched test measurements well In both cavity and structural response, the models captured peak frequencies accurately The BEA model tended to overpredict amplitudes by a few db The SEA model both over and underpredicted, but by less amplitude than the BEA model Overlays build confidence in both of the models, however, also shows that further improvements in models can be made to match test data more closely 18

19 Current & Forward Work Quartus Engineering, Johnson Space Center, and Glenn Research Center are collaborating to correlate the acoustic models to test Quartus developed an FE Acoustic model in wave6 to compare to the VA One BEA model presented here Post-correlation, modeling lessons learned will be implemented in the next cycle of Orion vibroacoustic flight models 19