Preventing the destruction of a product during vibration test and ensuring that enough stress is applied are important concepts that must be understood by manufacturers to maintain quality, contain costs, and improve yield. Using relevant test profiles for qualifying the product is the basic requirement to prevent the overstressing and understressing of a product during design. Appropriate test-tailoring techniques to develop profiles based on measurements will help you eliminate unsuitable stress from your product testing.
Overstressing the product can lead to overdesigning, which may yield too costly a product, said Jan Debille, product manager at LMS International. On the other hand, understressing leads to poor quality. So you need to know how to correctly tailor tests for your products.
One example is the French military standard GAM EG13, which describes a systematic and efficient approach to test tailoring, said Mr. Debille. It has been used
by the military for the last two decades. A commercial implementation of this method also is available from some vibration test manufacturers, including LMS.
Measuring the real-world conditions that act on the product in its service environment will reduce the likelihood of overstressing or understressing, said Richard McCormick, president of MB Dynamics. This includes the measurement of real-world inputs and the product’s real-world response to those inputs plus the analysis of the power spectral density (PSD) shape, bandwidths, crest factors, transient-time histories, accelerations, and strains. If you understand and properly replicate real-world conditions during the validation of a product design, then there is less chance for overdesigning the product.
Operating life may be determined by testing a product in its natural environment at the normal levels of vibration or in an accelerated fatigue environment. The former needs an entire lifetime of a product to figure life expectancy; the latter allows the test to be compressed into a tiny fraction of its operating life.
Some software packages are available to address proper stressing of products. For example, the Field2Lab software and hardware package from MB Dynamics acquires and imports acceleration and strain-time histories of real-world responses of products operating in their field environment. You can edit and manipulate those responses to create vibration and shock profiles for export to your vibration controller. The controller manipulates electrodynamic and hydraulic shakers to perform lab tests that replicate the field environments.
Strategically placed accelerometers to monitor the vibration levels on the product- under-test will provide important stress information, said Neill Doertenbach, product manager of QualMark. For example, on a product that contains several circuit cards in different orientations, place the accelerometers on boards in each orientation as well as on any massive components, such as heat sinks or transformers.
Miniature accelerometers should be mounted at critical areas on the product, agreed Robert Mercado, US sales and marketing manager at Trig-Tek. They help measure the actual product response to the input vibration level.
Trig-Tek recommends that you connect the accelerometers to signal conditioners that have adjustable alarm settings. This allows you to program a predetermined vibration level. When this output level is reached, an indicator or relay is activated to alert the operator or signal the controller to shut down the test.
During the design phase, highly accelerated life testing techniques are used to stress the product beyond specification limits, in a controlled fashion, until the product fails. It is not as important that you avoid overstressing during the design phase as it is to understand the vibrational characteristics at the time of product failure, said Mr. Doertenbach.
During production, highly accelerated stress screening is used, continued QualMark’s Mr. Doertenbach. Typically, the product is stressed beyond its specifications to bring out latent, process-related defects. It is important to avoid overstressing so excess life is not removed during production test.
A proof of screen should be run as a product validation. In the proof of screen, a sample of the product is run through the production screen many times, then tested to verify that it still meets the specification.
Both overstressing and understressing a product can be costly to the manufacturer, said Susan Brooksbank, international sales manager at Data Physics. Overstressing can damage the product during test and, depending on the damage, may require complete replacement of the unit.
If any product is overstressed unwittingly, the manufacturer may conclude erroneously that the product requires redesign. Understressing a product may result in a design that fails during normal operation.
Overtest and undertest limits are entered into the vibration control software as the profile is generated, said Kevin Ewing, market manager at Thermotron Industries. The desired control response is surrounded by two sets of scalable limits: tolerance and abort.
Tolerance limits give you a visual or audible indication that levels are approaching an unacceptable condition. When the abort limit is reached, the vibration system is automatically shut down. These limits can be dictated by the product-under-test requirements or the test specification. Typical random vibration specs use tolerance limits of ±3 dB and abort limit values of ±6 dB.
During the limit phase of a test in swept sine testing, the drive signal reduces to accommodate the new control level as the sine frequency sweeps through the resonance of the critical area, said Data Physics’ Ms. Brooksbank. Once the drive frequency passes the resonance value, the original control level is reestablished, and the drive voltage increases to its previous level.
For random testing, a similar effect occurs. For example, limiting is provided in some control systems in random, sine, sine-on-random, and random-on-random control modes for protecting the UUT.
A simpler, yet highly effective, operation is the method of extremal averaging of multiple accelerometer channels in a test. A single acceleration reference spectrum or power spectral density is defined for the test.
All accelerometer channels identified as control are continuously monitored and compared, and the maximum response level is used for the control feedback. This process limits all accelerometer channels to a maximum of the reference level. Some accelerometers encounter a level lower than the reference; however, by extremal averaging, the most critical areas on the UUT experience the defined limit of vibration.
Undertesting occurs when the UUT is vibrated at levels lower than specified, said Ms. Brooksbank. It may happen when incorrect measurements are taken during field trials or from less-than-optimum test equipment. A dynamic signal analyzer that provides accurate vibration measurements will help avoid both of these problems.
Once the correct environment is established and defined as a reference, the equipment for reproducing the environment must transmit the necessary vibration levels to the product, continued Ms. Brooksbank. In some cases, the shake table or fixtures may have a resonance and antiresonance combination that obstructs adequate testing.
For example, if the shaker platform and fixture have an antiresonance at the point where the product is attached to the fixture or shaker, then it is possible that the drive signal from the controller exceeds what is considered a reasonable level. A thorough understanding of the structural characteristics of the shaker and fixture helps you to ensure that the product is placed in a location with good transmission properties.
A dynamic signal analyzer should be used to measure the modal properties of the shaker and fixture, allowing you to understand the possible obstacles to an accurate test, said Ms. Brooksbank. Transfer function measurements on the test equipment identify resonances and antiresonances on the structures.
Protection Features
Most vibration control systems have programmable abort levels that are activated if the product is subjected to vibration levels sufficient to damage it, said Christopher Williams, joint managing director at Ling Dynamic Systems. We recommend that you undertake a resonance search at low-level vibration by using accelerometers on all major components. This will identify any potential problems before you undertake high-level testing.
A redundant safety system should be part of every vibration test system, said Trig-Tek’s Mr. Mercado. The safety system has provisions for a smooth, transient-free shutdown. Some companies connect the output from a response accelerometer on the product to the redundant safety system. If preset levels are reached on the product, the shaker will shut down safely.
It is very important to protect the UUT with as many safety measures as possible, said Mr. Debille from LMS. Safety trips include rms limits in the random mode, peak limit in the sine mode, open loop checking, notching in the random and the sine modes, and abort checks on notch profiles. These safety measures should be active at all times.
Controllers not only should have rms, but also PSD aborts, said MB Dynamics’ Mr. McCormick. But the key for sufficient protection is to know where to set the abort conditions.
For example, if a delicate or vibration-sensitive component exists in a design, perform resonant surveys to determine the natural frequencies and modes of vibration. If these resonances are excited because of real-world inputs, the responses must be understood and characterized so the resonances can be monitored during tests.
Test engineers should set up the vibration controller and control accelerometers to replicate real-world inputs and responses for the UUT, continued Mr. McCormick. Monitor accelerometers measure vibration responses on sensitive items and can have different abort conditions from those used to control a test.
You should be able to set not-to-exceed process variable limits through the vibration control system software, said Thermotron’s Mr. Ewing. These limits can be set for a number of input variables including acceleration, velocity, or displacement. If any of these preset values are exceeded, the system shuts down.
Other safety features are built into some vibration systems. For example, Thermotron offers a centering system that uses optical sensors to safeguard against overtravel displacement limits that could damage the UUT or shaker. The system automatically centers the armature, ensuring uniform travel through the magnetic field without interference.
Operating Life
The operating life of a product depends on several factors that you need to know before you begin testing. Understand the product life expectancy and the life-cycle environment and identify the product service use, characteristics, and past failure modes, said Thermotron’s Mr. Ewing. Products and their end uses vary so dramatically that it is very difficult to offer a cookbook-style answer to determine the operating life of a product.
Understanding the operating boundary conditions for a product in several different usage environments is a daunting but important task to perform, said MB Dynamic’s Mr. McCormick. In-field data acquisition is one very important method for uncovering these conditions.
Measuring typical accelerations, strains, temperatures, pressures, forces, and speeds on a product provides enormous insight for product life. Additionally, time-history measurements are used to create random, periodic, or sine vibration as well as time-domain inputs that simulate the field operating environment.
Software is available to help display, edit, and manipulate this data and prepare test profiles for closed-loop control of lab tests on electrodynamic and hydraulic vibration systems.
Product life is shortened by increased stress and the number of stress reversals, said Ms. Brooksbank of Data Physics. By sustaining vibration at the largest natural mode of vibration at a high level of stress, a component may be brought to fatigue failure in hours, even though normal use would be thousands of hours. To perform such a test, the vibration controller has to accurately detect the resonance and track it as the frequency shifts because of fatigue.
Collecting information under different service scenarios helps further characterize the conditions under which products must operate and survive, continued Mr. McCormick of MB Dynamics. Fatigue-life estimation and prediction data is evolving, and more insight is available today than ever before to help determine the operating life of a product.
Vibration Test Products
Vibration Controller Uses
ActiveX to Automate Routines
The SignalCalc 550 Vibration Controller simulates various production and field environments with random, sine, shock, sine-on-random, and resonant search and dwell. ActiveX is supported in SignalCalc 550 and the company’s SignalCalc ACE dynamic signal analyzer to automate test sequences and share information with other test reports. Custom control panels can be created, and the test process can be controlled from an external user interface. SignalCalc 550 provides reproductions of vibration environments on electrodynamic and hydraulic shakers. Data Physics, (408) 371-7100.
Entry-Level System Offers
Sine, Random Capabilities
The Entry-Level Vibration System (ELViS™) provides broad-frequency capabilities with sine and random functions for test items measuring 48″ × 48″ and weighing up to 800 lb. ELViS has an actuator, a table, an inertia reaction base, a hydraulic pump, a fluid reservoir, and a servocontroller in one enclosure. It includes a Windows-based sine and random vibration control system and an air-cooled hydraulic power supply. Lansmont, (408) 655-6600.
System Tests Automotive Parts
For Squeaks and Rattles
A newly designed quiet vibration test system is targetted for the automotive industry for squeak and rattle testing. It consists of the company’s V870 Shaker with a large, guided-head expander to accommodate vehicle structures up to 1/4-buck size. The system is driven by the SPA25-40K Switching Amplifier with a dedicated vibration controller. In the quiet mode, it provides acceleration of 1 g peak between 5 Hz and 400 Hz with a 250-kg payload. Ling Dynamic Systems, (011) 441 763 242424.
New Module Displays
Four Plots Simultaneously
The Time Data Acquisition Module Version 3.85 is used with the company’s Roadrunner software. It offers X-Y orbit plots, real-time FFT, and a digital readout counter. The counter is used for level- or peak-hold tracking on any channel or channel combination. The module shows four display types simultaneously. The input from a channel can be processed with a virtual measurement channel. The virtual channels combine the FFT spectrum, an order cut, and an octave spectrum. Data can be resampled and reclassified for any reference channel and displayed in a different diagram. LMS International, (011) 32 16 384 200.
Vibration Exciters Used for
Squeak, Rattle Testing
The S&R Energizer family of vibration exciters is used for buzz, squeak, and rattle testing of vehicles and their components. The exciters use voice coil principles and does not require cooling during testing. Spring steel support elements are constructed to maximize radial stiffness to resist side loads, offset loads, and payload torqueing. Payloads up to 75 lb can be tested to squeak and rattle profiles. The Energizer comes with a 15″ dia mounting table for vibration tests to 750 Hz. MB Dynamics, (216) 292-5850.
System Offers Temperature
Change Rate to 60°C/min
The OVS-1.5 UltraRate™ Thermal Chamber and OmniAxial™ Vibration System feature a 19″ × 19″ × 16″ interior workspace, a +200°C to -100°C temperature range, and up to a 60°C/min change rate on the product. The vibration system consists of four exciters that generate six degrees of freedom. The vibration profile extends to 100 grms between 2 Hz and 5 kHz. The maximum air consumption of the chamber is 20 scfm at 90 psig. The system includes a PC controller in a separate cabinet, a temperature- limiting unit, and an accelerometer. QualMark, (303) 254-8800.
Test Controller Provides
Frequency Range to 20 kHz
The Accelerated Stress Test (AST) Control System has a library of preprogrammed step-stress and cycling profiles. It offers a selectable frequency range to 20 kHz and a resolution to 6,400 lines. The AST system manages 16 accelerometer channels and eight thermocouple inputs. The sampling rate is 51.2 kHz/channel with 12-bit A/D and D/A resolution. The temperature range extends from -200°C to +400°C using T-type thermocouples. The temperature change rate is 70°C/min. Thermotron Industries, (616) 392-1491.
Vibration Monitor Offers
Shutdown Protection
The Model 620B Vibration Protector-Monitor provides independent shutdown protection for test specimens and vibration systems. It accepts charge or voltage-mode accelerometer signals and processes the signals to provide digital display of acceleration, velocity, or displacement. Maximum displacement abort limits are independently set for the over- and under-test shutdown limits. An abort testing signal initiates a transient-free shutdown of the vibration system and activates two auxiliary shutdown relays. Trig-Tek, (714) 956-3593.
Hand-Held Measuring Unit
Offers Vibration Test Module
The MMS3000-VB1 Vibration Module for the MMS3000 Multi Measurement System has up to three channels for accelerometer inputs with 100 mV/g sensitivity and 500 kW input impedance. The unit features a 240- × 128-pixel backlit graphics LCD panel. It provides velocity indications with a maximum derived velocity of 4 in./s peak and a dynamic signal range of 95 dB with <-70 dB distortion. It supports FFT displays to 10 kHz and 1,600 lines of resolution. It provides 0.5 MB of nonvolatile memory and is expandable to 8 MB. Commtest Instruments, (888) 667-6006.
Portable Vibration System
Has 16 Inputs in Parallel
The Portable Acquisition/Analysis and Control System (PACS) operates from AC or DC power and features 16 parallel acquisition inputs. It acquires, measures, and analyzes data in the random signal, transient waveform, and long-duration waveform modes. PAC reproduces field vibration data for random power spectral density, shock, and time replication acceleration control of long-duration waveforms. Transient captured waveforms can be edited and compensation pulses added to be reproduced at the shaker. Unholtz-Dickie, (203) 265-3929.
Copyright 1998 Nelson Publishing Inc.
July 1998
|
|