Looking for good vibrationsDesigning equipment to withstand intensive sound and vibration levels means first knowing details about the sound itself.
Rocket launches are noisy. So much so that the noise produced during a shuttle launch has considerable detrimental effect on the safety of spacecraft, ground facilities, and support equipment. The accumulation of stress induced by these high vibrations threatens the safe operation of any exposed system. MicroStrain, Williston, Vt., worked with NASA Kennedy Space Center researchers to install a wireless sensor system to monitor far-field shuttle launch acoustics, pressure, and vibration levels of the STS-134 Endeavour and STS-135 Atlantis launches and display the results on a Web browser.
The rocket-acoustics program lets NASA remotely monitor vibrations and sound levels deemed hazardous for equipment and humans. One system of particular interest is the Composite Overwrapped Pressure Vessel (COPV) used to house various types of pressurized fluids. COPVs are acutely subject to internal pressure and sit in close proximity to the shuttle launch.
Because of limited accessibility and the extreme operating conditions under which the COPVs operate, conventional hardware cannot monitor the strain these systems experience.
Strict mission protocols prevented researchers from accessing the test area for several days before and after a launch. Uncertainty over the timing of a launch further complicated the testing. As a result, the sensors had to remain powered and alert in anticipation of a launch over several days or weeks at a time. The wireless sensor network had to operate on limited power while still capturing the brief, but dynamic, nature of launch acoustics.
Information collected over this length of time generates a large amount of data that is difficult to navigate and transfer. Powerful data-management strategies are needed to identify and analyze key threshold data. MicroStrain’s approach to the problem combined advanced power-management techniques, high-performance wireless nodes, and a Web-based data-management platform called SensorCloud.
System hardware consisted of two wireless acceleration sensor nodes, one wireless strain gage, a wireless data-collection base station, and the data-management software. Nodes were instrumented on a cantilever plate with a low natural frequency. The plate was installed on Shuttle PAD 39B, the neighbor to Shuttle PAD 39A where the last several launches have taken place. On PAD 39B the plate is approximately 7,000 ft from the shuttle liftoff, the point where far-field acoustics are of interest.
After the spacecraft takes off and instrumentation captures its acoustic data, data reduction isolates the sensor data for the launch event. Initially, a local PC collected the data. During a five-day measurement time frame, the computer collected over 3 Gbytes of information. NASA handled the viewing, navigating, and sharing of such large amounts of data by uploading it to the SensorCloud software. On SensorCloud, MicroStrain support engineers collaborated with NASA researchers to seek and analyze key threshold data. Two sets of acceleration data from the strain gages clearly indicated peak vibration levels. These results let NASA compute the equivalent static load developed by the rocket and correlate the data to compare against their predictions.
The wireless sensor network lets NASA researchers scale the network across multiple locations and different equipment types.
NASA figures proactively measuring strains exerted on high-value and potentially hazardous assets helps protect valuable equipment. Smart data management lowered potential maintenance costs and reduced the chance of premature equipment failure and replacement. It enhances the safety of monitored components and supports any issues of availability and maintainability of equipment and structures.
Follow the sun
It takes an instrument that tracks the sun’s daily motion to focus in on solar magnetic fields.
The Advanced Technology Solar Telescope (ATST) could provide the sharpest views ever taken of the sun’s surface. Scheduled to be operational in 2017, the ATST should help scientists learn how cosmic magnetic fields are generated and destroyed, and learn what role the fields play in the organization of plasma structures and the impulsive releases of energy found everywhere in the universe. But more importantly, researchers are interested in the solar variability that eventually affects Earth.
The optic support structure that includes the mirror assemblies will weigh nearly 75 tons, the mount base nearly 90 tons, and the Coudé rotator 160 tons. (Coudé describes the construction of the telescope such that light reflects along the polar axis to focus at the mount point of the sensor array.)
The thermally controlled and highly ventilated 84-ft-diameter enclosure can rotate independently of the telescope or it can be linked to corotate as the telescope turns. The enclosure houses the controllers and much of the instrumentation.
Overall, the telescope uses over 100 motion actuators. That number is expected to grow as instruments are added to the system. Each individual actuator has little impact on the ATST Coudé environment. But all of the motion-control hardware taken together becomes a dominant factor in heat and EMI generation inside the telescope enclosure.
One critical area for the ATST motion control involved the expertise needed to maintain the multiple systems found in the telescope. If each control area used a different vendor for motion control, the hardware and software maintenance personnel would face a heavy burden in documentation, spare parts, periodic upgrades, maintenance of software code, and expertise in each of the systems. So the preference was for a single vendor.
A controller from Delta Tau in Chatsworth, Calif., handles the motion-control functions. The company’s PMAC Motion Controller is equipped with a sophisticated general-purpose computer that permits the use of widely used programming languages. In fact, the controller can be programmed in several languages simultaneously, including a built-in Script Language; general machine and I/O logic; industry-standard graphical programming in any of five formats specified in IEC-61131 including ladder logic and sequential function charts; and in C for advanced programmers who wish to write servo, phase, PLC and general-purpose applications. Other languages accepted include G-code, MatLab/Simulink, LabView, and EPICS.
Additional features of the PMAC include a built-in Web server to support direct browser access for development and maintenance; the ability to control up to 256 motors or 128 coordinate systems simultaneously; and the ability to accept RS-274 G-code programs.
Inside a single-phase digital point-of-load supply
This chipset smartly manages power systems while another conditions sensor signals.
Point-of-load supplies place the power management of electronic circuits near the devices that actually use the regulated power. POL supplies help eliminate noise pickup associated with long power leads and reduces voltage fluctuations as other circuits draw power.
The ZSPM1000 from Zentrum Mikroelektronik Dresden AG (ZMDI), Dresden, Germany, operates as a synchronous step-down converter in single-rail and single-phase configurations. It works with the ZSPM9000, a MOSFET and integrated power-stage driver. Together they make up a digital POL systems for constrained areas. Typical applications include high-performance areas such as servers, storage units, processor and FPGA boards, and other distributed-power loads.
Digital communication between the chips using an I2C bus provides energy savings and thermal-management benefits through system-level integration. ZMDI’s Pink Power Designer is a PC-based design program that lets designers balance trade-offs between cost, size, and efficiency to quickly identify optimal POL designs. The software also contains a user interface to the chipset via the I2C bus for configuration of additional features such as protection and sequencing. Changes can be made while the chips are running.
The ZSPM9000 DrMOS includes internal linear regulators that lets the ZSPM9000 operate from a single 5V supply. The regulators develop all other voltages needed by the chipset.
The controller IC comes in a standard 24-pin QFN package that measures 4 × 4 mm. The DrMOS power subsystem comes in a matching 40-pin PQFN package of 6 × 6 mm. Combined, the two devices save 72% of the space compared to conventional discrete components. An evaluation kit, the ZSPM8000-KIT, lets designers start developing POL circuits immediately. The Pink Power design tools are available separately from ZMDI.
Along with the POL controller, ZMDI also introduced the ZSSC3008 sensor signal-conditioning integrated circuit (IC) for sensors. The ZSSC3008 handles resistive bridge sensors that need second-order linearity correction for accuracy. Nonlinear compensation makes it possible to use less-expensive transducers to make cost-effective high-performance sensing systems.
Many sensors for those applications have an intrinsically low-temperature coefficient (TC). This includes ceramic pressure sensors, thin-film pressure sensors with good TC performance, and oil-filled stainless-steel or passively compensated sensors.
Chips for resistive bridge sensors today frequently include onboard diagnostics and protection features such as an EEPROM signature, bridge-connection checks, bridge input-short and open-bridge detection, power-loss detection, and output current limits for short-circuit protection. For sensor correction, devices may employ offset correction and gain programming to calibrate the bridge-output signal.
Programming and single-pass calibration of the sensor and the ZSSC3008 is done in a standard PC environment using the ZSSC3008KIT development tools. A development kit comes with the device in an SOP8 package, development board, USB cable, and calibration software.
The ZSSC3008 operates over a supply voltage of 2.7 to 5.5 V, or up to 30 V with an external JFET. Accuracy is 0.25% over the –25 to 85°C range, and 0.5% from −40 to 125°C. Depending upon configuration, current consumption typically ranges from 1 mA down to 250 μA. Response time is typically 1 msec.