Radiation from Wavepackets

This paper introduces a neural network-based calibration for low-force sensors that operate in the sub-newton regime (0−1N) for wearable applications. The proposed calibration utilizes a fully-connected neural network to digitally reduce the sensor nonlinearity. The neural network is trained using data from a custom low-force measurement system with a novel compliant mechanism. Detailed study explores the tradeoffs between the neural network size and activation function with calibration accuracy. Measurement results demonstrate 4X improvement in the force sensor linearity, achieving errors less than 0.005 N. The proposed calibration is well-suited for wearable applications requiring precise low-force measurements.

The Small Body Dynamics Tool (SBDynT) is software written for the community of solar system small body researchers to perform dynamical classification, characterization, and investigation. SBDynT provides advanced simulation analysis capabilities that make it straightforward to determine mean-motion resonance occupation, proper orbital elements, and a variety of stability indicators. These calculations can be performed for small bodies that are known, newly discovered, or simulated; observational uncertainties can be incorporated through the use of dynamical clones. In this paper, we describe the methods for producing proper orbital elements and stability indicators, which serve as essential tools for characterizing dynamical stability and long-term evolution. Through extensive validation, we demonstrate that this code offers a robust open-source framework for investigating the dynamics of solar system small bodies with high accuracy. We also aim for computational efficiency allowing SBDynT to provide dynamical information for the several-fold increases in small bodies expected in the Legacy Survey of Space and Time era.

Machine learning is revolutionizing protein design by enabling the rapid generation of sequences with precise structural and functional properties. Controlling protein conformational states remains a major challenge, particularly for enzymes regulated by complex structural switches. Here, using high-resolution structural data and probabilistic sequence-structure models, a machine learning-driven framework for conformationally biased protein design is presented titled Conformation-Specific Design or CSDesign. This approach generates sequences predicted to favor a desired conformation while disfavoring alternative states. As a proof-of-concept, this approach is applied to extracellular signal-regulated kinase 2 (ERK2), generating variants predicted to favor the active or inactive state. Experimental validation of relative kinase activity in a controlled assay confirmed that an active-biased variant, CSD104, exhibits robust kinase activity without native upstream phosphorylation, while an inactive-biased variant, CSD101, remains inactivated. Structural analysis suggests that engineered interactions stabilize active-like features in place of phosphorylation. These results demonstrate machine learning control of protein conformational ensembles, with potential to design enzymes and other conformationally regulated proteins without relying on phosphomimetic mutations or extensive experimental screening.

This paper presents an aeroacoustic analysis of SpaceX’s Falcon 9 using data from two launches at Vandenberg Space Force Base. Acoustic measurements from 28 stations, ranging from 0.2 – 38.6 km, captured four key events: ignition overpressure, maximum launch noise, flyback sonic boom, and maximum landing noise. Sound exposure spectra show that the launch noise has a peak frequency of 30 Hz, an order of magnitude greater than the flyback boom peak frequency. Maximum 1-s overall sound pressure levels approach 150 dB at the closest stations and collapse well between launches across the full measurement range. Scientific Acoustic Tool for Understanding Rocket Noise (SATURN) predictions for launch noise in the time and frequency domains agree well with measured data, validating the recently developed model for Falcon 9. Using stations within 1 km, Falcon 9 was found to have a wide peak directivity region, spanning angles from 60 to 70°. The measured directivity angles are within the range of convective Mach number predictions and provide evidence that supersonic instability waves contribute to the main radiation lobe for rockets. Overall sound power levels were also calculated, producing an average sound power level of 195.4 ± 0.7 dB (1σ) and an acoustic efficiency of 0.30%. Sound power spectra peak at a Strouhal number of 0.010, lower than typical rocket assumptions and far below values for other supersonic jets. Overall sound pressure levels scaled to 100 nozzle diameters were similar to lower Mach number jets, suggesting that this metric may plateau at rocket-like conditions. These analyses extend understanding of the aeroacoustic source characteristics of rockets and better connect them to other supersonic, heated jets.

The role of nozzle configuration on rocket noise radiation is not well understood, particularly for multi-core vehicles where plume interactions may introduce azimuthal asymmetry. While tightly clustered engines are often assumed to radiate axisymmetrically, configurations with spaced nozzles may exhibit directionally dependent acoustic fields. This paper presents results from a measurement campaign conducted during the final Delta IV Heavy launch (NROL-70), supplemented by data from a previous launch (NROL-82), to investigate azimuthal variation in radiated noise. Measurements spanning a wide range of azimuthal angles show a consistent increase in sound pressure and sound power levels along the jet midplane relative to the jet plane. In sound pressure levels, differences of 5 dB are observed near the dominant spectral frequency (~30 Hz). In sound power, differences of ~2.5 dB are observed, particularly around the peak frequency, with smaller but persistent differences at higher frequencies. Strouhal number analysis indicates that the effective source length scale lies between the limits of fully independent and fully merged plumes, suggesting a partially merged interaction regime. These results provide field-scale evidence that multi-core rocket plumes do not behave as independent or fully merged sources but instead form partially coupled turbulent structures that produce directionally dependent acoustic radiation. The findings demonstrate that azimuthal asymmetry in large rocket noise is both angle and frequency dependent and should be considered in modeling of launch acoustics.

AB  - The role of nozzle configuration on rocket noise radiation is not well understood, particularly for multi-core vehicles where plume interactions may introduce azimuthal asymmetry. While tightly clustered engines are often assumed to radiate axisymmetrically, configurations with spaced nozzles may exhibit directionally dependent acoustic fields. This paper presents results from a measurement campaign conducted during the final Delta IV Heavy launch (NROL-70), supplemented by data from a previous launch (NROL-82), to investigate azimuthal variation in radiated noise. Measurements spanning a wide range of azimuthal angles show a consistent increase in sound pressure and sound power levels along the jet midplane relative to the jet plane. In sound pressure levels, differences of 5 dB are observed near the dominant spectral frequency (~30 Hz). In sound power, differences of ~2.5 dB are observed, particularly around the peak frequency, with smaller but persistent differences at higher frequencies. Strouhal number analysis indicates that the effective source length scale lies between the limits of fully independent and fully merged plumes, suggesting a partially merged interaction regime. These results provide field-scale evidence that multi-core rocket plumes do not behave as independent or fully merged sources but instead form partially coupled turbulent structures that produce directionally dependent acoustic radiation. The findings demonstrate that azimuthal asymmetry in large rocket noise is both angle and frequency dependent and should be considered in modeling of launch acoustics.

Vertically aligned carbon nanotube forest growth uses a thin-film iron catalyst on an alumina support. The iron catalyst thickness (typically, 1–10 nm) strongly affects forest morphology. We explored the use of spectroscopic ellipsometry (SE) as a rapid, sensitive, and nondestructive metrology method for these films. SE does have challenges, however, as it is difficult to break the correlation in the analysis between fitted optical constants and thickness of ultrathin films. Partial oxidation and optical absorption in the iron–iron oxide films add further complexity. We performed a multisample SE analysis of thermally evaporated iron films with target thicknesses of 1–14 nm. To improve sensitivity, we used interference enhancement by incorporating a 350 nm silica film on a silicon substrate beneath the iron film and alumina support. We used a consecutive-layer approach, collecting SE data and fitting the optical constants and thickness of each film before depositing the next. The iron–iron oxide film was modeled with an effective medium approximation layer. The model fit the data well with a mean squared error of 25. From the SE results, we estimated the thickness of the iron film before oxidation (“equivalent iron thickness”). We found that SE is highly sensitive to equivalent iron thickness and yields repeatable thickness measurements (ca. ±0.015 nm). We determined that the equivalent iron thickness variation we observed across different measurement locations on the same sample can be explained by error propagation from uncertainty in the underlying alumina thickness.