Figure 1: Near-field acoustic holography and scanning measurement system with reference microphone
Challenge: Develop a portable near-field acoustic holography (NAH) system to measure high-amplitude jet noise from current and next-generation military aircraft, providing model correction and comparison, assessing the performance of noise control equipment, and predicting the conditions of ground maintenance personnel and the noise impact on communities.
Solution: Develop a cost-effective system based on NI PXI dynamic signal acquisition (DSA) equipment, which has advantages such as good portability, flexibility, scalability and high accuracy; by increasing the number of data acquisition channels and moving the microphone array, the measured area can be expanded and the measurement time can be shortened, while combining the technical requirements of NAH with the environmental conditions and safety constraints of jet noise measurement.
"Using the flexibility of LabVIEW software, we are able to customize monitoring and data verification functions."
Why does the Air Force Research Laboratory need to measure jet engine noise?
Military jet aircraft generate high levels of noise for ground maintenance personnel and communities. Therefore, the U.S. Department of Defense has invested in developing advanced modeling tools for noise suppression technologies and research into the noise impact on communities. For these tools to function fully, innovative measurement and analysis methods are needed to characterize the jet noise source region. Near-field acoustic holography (NAH) provides a best-in-class, universal method for measuring intensity, direction, and spectrum, as well as the spatial distribution of noise emanating from the nozzle.
The Air Force Research Laboratory selected Blue Ridge Research & Consulting (BRRC) to develop innovative measurement and analysis methods for feature extraction and mapping of noise emitted from jet engines. BRRC is an acoustic engineering consultancy specializing in addressing critical noise and vibration challenges, including sound and noise source measurement, general and specialized modeling, soundscape and transportation noise visualization, outdoor alarm system design, and rotating machinery monitoring. BRRC collaborated with the Acoustics Research Group at Brigham Young University (BYU) to develop this application.
Challenges of using NAH
The development of a feature processing and measurement matrix for large jet environments using a NAH presented several technical and logistical challenges. Accurate feature extraction of the near-field of military jet aircraft required the ability to record sound pressure levels up to 170 dB and frequencies from 5 Hz to 30 kHz. Furthermore, measurements needed to cover the entire jet length. The NAH system also had to be semi-portable because locations where military jet aircraft could operate at high speeds with static, high-powered engines were limited.
Figure 2: An F-22 starts its burners and uses NAH to operate the ground engines.
Measurement
A key requirement for this channel testing setup is to simplify system design and minimize startup time and cost. To ensure sufficient measurement space for feature acquisition of the entire jet noise source while minimizing the number of microphones used, we propose a scheme using a scan-based microphone array combined with a stationary reference microphone.
Building a multi-channel data acquisition system using the NI PXI platform
We recorded time waveforms from microphones on a multi-channel data acquisition system based on the NI PXI platform. The PXI chassis contains nine 16-channel NI PXI-4496 modules and two 4-channel NI PXI-4462 DSA boards, which feature synchronous sampling to ensure all 152 channels are in the correct phase. The DSA boards, with 24-bit analog inputs per channel and IEPE constant current signal conditioning, are ideal for precise microphone measurements.
The PXI-4496 module features a 113 dB dynamic range and can simultaneously sample all 16 channels at rates up to 204.8 kS/s. Furthermore, the module includes built-in anti-aliasing filters whose parameters automatically adjust according to the sampling rate, while also offering up to 20 dB of software-selectable input gain. The 113 dB dynamic range and 20 dB software-selectable gain adjustment allow for accurate measurements of both weak and strong signals. Additionally, the acquired data is AC-coupled via a 0.5 Hz high-pass filter.
Figure 3: NAH Multi-channel Data Acquisition and Monitoring Measurement System
We can perform high-amplitude pressure measurements using a quarter-inch GRAS 40BE free-field microphone and a 26CB preamplifier. It allows a frequency response from 4 Hz to 100 kHz ± 3 dB. Furthermore, two design modifications allow for customization of the sensor for this application. The microphone is designed with a nominal sensitivity of 1 mV/Pa, enabling the measurement of various sound pressure levels up to 170 dB. We also extended the preamplifier of the quarter-inch microphone to a half-inch BNC connector to minimize wiring, suppress sound reflections, and increase the robustness of the array mount. A constant-current preamplifier is provided by an IEPE conditioning board on the data acquisition system.
The NAH measurement system uses a two-dimensional microphone array containing 90 microphones, which can move on a plane parallel to the jet stream via four wheels and guide rails. The test area is decomposed into multiple two-dimensional microphone array blocks. The guide rails allow for precise positioning of the test equipment within the jet stream. Furthermore, the test equipment can be locked in a specific location.
Figure 4: NAH measurement team with F-22
Data from all channels is streamed via coaxial cable using an MXI-4 connection to a remote 1 U controller. The controller comprises an Intel Core 2 Quad processor and four 250 GB hard drives configured in RAID 0. This RAID 0 configuration allows for over 150 channels of data streaming while simultaneously running data monitoring and analysis software.
In addition, sufficient storage space is also crucial; 152 channels per measurement area at a sampling rate of 96,000 Hz generate over 1.75 GB of data in 30 seconds. We can perform jet feature extraction by measuring multiple areas along the jet at different engine energy states; therefore, the data is stored in a non-proprietary binary format. We can control and monitor data acquisition using a sunlight-readable remote desktop running Windows, either wirelessly or wired to the controller. The data storage system is housed in a MIL-SPEC shipping container.
The data acquisition system is located approximately 61 meters from the nozzle. A custom-designed InfiniBand cable connects the data acquisition system to the test equipment. This length minimizes vibration of the data acquisition hardware in the harsh jet environment and allows the test equipment to move along the entire length of the jet. Furthermore, this cable arrangement reduces the weight of the microphone test array. More importantly, with the flexible LabVIEW software, we were able to customize functions for monitoring and data verification.
We completed component testing to ensure the instruments were functioning correctly. Acoustic instrument testing was relatively straightforward; scalable data processing was the primary focus. We successfully tested the data acquisition system, which can synchronously sample 152 channels at a rate of 100 kHz. We will also introduce other visualization tools for future performance evaluations.
