Speaker
Description
Background of the study
Reflected shock tunnels are an excellent tool for studying high-temperature, hypersonic problems such as boundary layer transition, ablation, and shock-induced flow separation. These phenomena are known to be sensitive to the thermochemical state of the tunnel freestream. Although shock tunnels are a mature technology, much uncertainty in the tunnel freestream conditions remains. The temperature behind the reflected shock wave in the plenum results in dissociation of diatomic species, such as O2, and the production of additional species, such as nitric oxide (NO). The subsequent rapid expansion through the nozzle results in pronounced thermodynamic nonequilibrium where the vibrational temperature, Tv, of diatomic molecules can be much higher than the translational, Tt, and rotational temperature, Tr, at the nozzle exit. Moreover, the chemistry during the expansion can also remain ‘frozen’ where the freestream maintains a chemical composition having NO and high concentrations of atomic species akin to that in the nozzle throat. Thus, detailed multi-species, multi-temperature measurements are required to inform and validate predictive models for the nozzle expansion process and characterize the test section freestream.
The millisecond test times, high background emission, and low freestream densities in shock tunnels make measurements challenging. With recent advances in frequency-tunable, burst-mode laser diagnostics, time-resolved CARS measurements of homonuclear molecular temperatures are now possible. This presentation will present measurements of N2 and O2 Tr and Tv in the air freestream of Sandia’s Hypersonic Shock Tunnel (HST), while comparing to complimentary measurements of NO Tr and Tv temperatures acquired using laser absorption spectroscopy (LAS). Additionally, simultaneous freestream velocity measurements are performed using NO MTV along with concurrent measurement of stagnation and pitot pressures.
Methodology
The HST at Sandia National Laboratories is a research-scale, free-piston facility capable of producing high-enthalpy hypersonic flow with freestream velocities ranging from 3 to 5 km/s. A free-piston shock tube generated the high temperature and pressure stagnation reservoir gas for the shock tunnel. The shock tunnel nozzle has a throat diameter of 12.7 mm, with a circular throat cross-section blended to a conical expansion of 7.9 degrees. The area ratio was 784 corresponding to a 35.6-cm exit diameter. The core-flow diameter, as determined by a pitot rake, was approximately 25 cm. Pitot pressures are measured using a rake containing 17 PCB sensors in the test section. HST experiments were performed at four conditions to produce freestream velocities of nominally 3 km/s and 4 km/s with a test gas of either synthetic air (80% N¬2, 20% O2) or pure N2.
The CARS technique was used to measure rotational and vibrational temperatures of N2 and O2 in the freestream of the shock tunnel. The setup used a Spectral Energies “Quasimodo” burst-mode laser, which provided both the 532- and 355-nm, 10-ns pulses at 100 kHz for a burst duration of ~1.2 milliseconds (ms). To generate the broadband Stokes beam for the CARS process, a noncolinear optical parametric oscillator (NOPO) was pumped by the 355-nm output from the burst-mode laser. The output wavelength was centered at 580 nm for O2 CARS and 607 nm for N2 CARS. The CARS setup utilized the 532-nm from the burst-mode to form the CARS pump beams, and the broadband output from the NOPO was used for the Stokes beam to measure single-shot measurements of N2 and O2 temperature measurements.
The NO MTV setup was operated simultaneously with the CARS by splitting the main 355-nm output and pumping a secondary OPO to generate 622-nm output. The 622-nm output and the residual 355-nm pump beam are used for sum-frequency generation (SFG) to generate a laser at 226 nm with a bandwidth of ≈ 15 cm-1 to excite multiple rotational levels near the (0,0) bandhead of the NO A2Σ - X2Π system. The NO MTV beam is overlapped with the CARS set-up to generate NO LIF signal colinear with the CARS measurement and captured using a UV-sensitive image intensifier (LaVision HS-IRO S20) coupled to a high-speed Phantom TMX 7510 mono-chrome camera.
The LAS diagnostic used two quantum cascade lasers to measure rotational and vibrational temperatures and the partial pressure of NO at 25 or 100 kHz. The beams were both fiber coupled into a single-mode fiber. The collimated, collinear beams were pitched through the test section where 3D printed flow cutters were used to isolate the quasi-uniform core flow. The beams were positioned 2 cm from the nozzle exit and the absorbing path length was 23.2 cm.
Results
Ensemble-averaged temporal profiles of stagnation pressure, pitot pressure, freestream temperatures, and freestream velocity were taken for all four cases. Fig. 1 shows the temporal profiles for the 3 km/s – air test condition. SPARC CFD predictions initialized with NASA Chemical Equilibrium with Applications (CEA) of the stagnation reservoir are also shown on Fig. 1. For each case, there is pronounced thermal nonequilibrium observed between the rotational and vibrational temperature of all three species. The SPARC model accurately predicts the rotational temperature and the freestream velocity for each case. SPARC CFD predictions show that vibrational relaxation is fastest for NO and slowest for N2, an observation consistent with the experimental measurements. The current multi-vibrational temperature model captures this trend, but underpredicts NO Tv¬ while overpredicting N2 and O2 Tv¬.
To isolate oxygen chemistry effects on the N2 thermal relaxation during the flow expansion, pure nitrogen was also used as a test gas. With the lack of other collisional partners, the pure N2 condition showed elevated vibrational temperatures for both the 3 and 4 km/s conditions. Comparison of the experimental data with the CFD predictions showed good agreement with the rotational temperatures. The measured N2 vibrational temperatures came in lower than the CFD predictions at both conditions, consistent with the air cases.
An additional case was run using humid air to test the effect water has on N2 nonequilibrium at the 3 km/s condition. With 3600 ppm of water present, both the rotational and vibrational temperatures of N2 were nearly identical to the previous 3 km/s synthetic air cases showing little dependence on water at these levels.
Conclusion
A novel combination of high-speed laser spectroscopy measurements for freestream velocimetry and multi-species internal temperatures has been performed over an extensive number of repeat experiments in Sandia’s reflected shock tunnel. The resulting dataset allowed for meaningful trends between experimental conditions and molecular temperatures to be discovered. Additionally, the combined dataset served as a benchmark for comparison to nonequilibrium freestream modeling using the SPARC CFD code.
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.
Fig. 1. Time-synchronized plot of pressures, velocity, and temperatures for the 3 km/s condition. First row, stagnation pressure. Second row, pitot pressure measured on centerline probe. Third row, velocity measured from MTV. Fourth row, rotational temperature of diatomic nitrogen (from CARS) and nitric oxide (from LAS). Fifth row, vibrational temperature of diatomic nitrogen and oxygen (both from CARS) and nitric oxide (from LAS). Shading indicates variability between runs, plotted as +/- 1 standard uncertainty of the ensemble. Dashed horizontal lines indicate SPARC predictions.
Summary
Coherent anti-Stokes Raman scattering (CARS) and nitric oxide molecular tagging velocimetry (NO-MTV) are used to characterize the freestream in Sandia’s Hypersonic Shock Tunnel (HST) using a burst-mode laser operated at 100-kHz. Experiments are performed at nominal freestream velocities of 3 and 4 km/s using both air and N2 test gas. The CARS diagnostic provides nonequilibrium characterization of the flow by measuring vibrational and rotational temperatures of N2 and O2, which are compared to NO temperatures from separate laser absorption experiments. Simultaneous, colinear freestream velocities are measured using NO MTV along with pitot pressures. This extensive freestream dataset is compared to nonequilibrium CFD capable of modeling species-specific, vibrational temperatures throughout the nozzle expansion. Significant nonequilibrium between vibrational and rotational temperatures are measured at each flow condition. N2 exhibits the most nonequilibrium followed by O2 and NO. The CFD model captures this trend, although it consistently overpredicts N2 vibrational temperatures. The modeled temperatures agree with the O2 data. At 3 km/s, the modeled NO nonequilibrium is underpredicted, whereas it is overpredicted at 4 km/s. Good agreement is seen between CFD and the velocity and rotational temperature measurements. Experiments with water added to the test gas yielded no discernable difference in vibrational relaxation.
