There is a proven way to improve the reliability of your product, a way to precipitate and identify latent defects before the equipment is shipped. A carefully designed series of stress levels and durations—a stress profile—can uncover faults without damaging the product. This technique, environmental stress screening (ESS), subjects all products to accelerated environmental stimuli. By monitoring performance carefully, you discover potential problems, not your customers.
Is ESS needed on every product? “No, a mature product with a good track record generally is not a good candidate,” according to Lucy Baker, vice president of operations at Screening Systems.
But for product introduction, ESS is invaluable. Many companies find that the process in high-volume production applications can reduce or eliminate returns for repair and improve the mean time between failures (MTBF). It also can make the equipment less vulnerable to adverse environmental conditions in the customer’s hands.
There are many variations on the ESS theme. One of the earliest screens was the 48-h burn-in. Then came vibration in three axes, sometimes enhanced by using different levels of force. Temperature cycling had its day.
Variations and combinations of these techniques also have their benefits. Today, however, the most effective methods involve simultaneous temperature cycling and random vibration while keeping the screening within the product’s design limits.
Many ESS users begin with a highly accelerated life test (HALT) in which the product is stressed in steps, each increment designed to bring it nearer to destruction. This procedure provides valuable information from which to derive an ESS program.
“A new product should first go through the step-stressing sequence to determine its design weaknesses,” Ms. Baker of Screening Systems continued. “Those results should be fed back to the designers for corrective action. The HALT results provide the background for instituting an ESS program tailored to the product.”
Questions From Potential Users
Some companies that have never used ESS are reluctant to commit to a program of this nature. The engineers at Screening Systems shared some of their most frequent ESS-related inquiries of the past 21 years along with their answers:
- Will ESS induce faults?
No, the right levels of stress, properly applied, will not induce faults. Instead, they will bring existing ones to the surface. - Will such a program delay my shipments?
Not necessarily. First, production rates at many companies already have been reduced because of time lost in rework/repair, and a properly engineered ESS program will reduce such problems. Also, if shipments seem to be delayed after an ESS program is in place for a few months and its results are evaluated, it may be practical to go to highly accelerated stress auditing (HASA) rather than 100% screening. - But isn’t ESS expensive?
Some expense is inevitable as an ESS program is implemented. Compare this with the cost of operating without ESS. This may include factory rework, field-service calls, and the impact of defective products on the company’s reputation. Which type of expense is less painful?
And while on the topic of expenditures, many newcomers to ESS question why they should pay for vibration equipment if a product will be used in a vibration-free environment. The explanation is quite simple according to Neill Doertenbach, OVS technical sales manager at QualMark: “The vibration sequence is an accelerant for failure mechanisms that would show up later in quiet environments. For example, a cold solder joint can survive thermal testing and work well for a few months before failing, but a combined vibration and temperature screening will uncover that defect.”
Hardware Selection
Selection of an ESS station for simultaneous vibration and thermal ramping involves careful planning. The first step is to determine where in your company the screening will be conducted. For example, it may be a development laboratory, the quality-assurance area, or an assembly line. The location will likely determine the configuration—stationary, movable with a forklift, or on wheels for ultimate portability.
The size and form factor of the products to be tested will define the interior dimensions of the test chamber. The vibration table size must be specified. How heavy is the expected payload? Can the system handle this load?
Define the maximum acceleration available over the screening range, typically 5 Hz to 2,000 Hz. Also, specify how many grms can be applied at each frequency.
Stressing at frequencies higher than 2,000 Hz generally is considered counterproductive. It doesn’t produce parts displacements of sufficient magnitude to precipitate failures, and putting actuator energy into the higher frequencies just consumes power and lowers test-system efficiency.
The test system’s temperature range in °C and the maximum rate of change in °C/min must be known. It also is very important to know the air velocity across the UUT. Rapid changes in temperature will not produce the intended results unless the colder or warmer air is delivered to the product with minimum lag time.
Controller and Software
The ESS controller, typically a PC, and its peripheral hardware must be defined with their interface to the test chamber. Also, you will want to learn what software is available with the system and explore its operator interface and displays.
For meaningful results, the system should vibrate the product in six degrees of freedom—the X, Y, and Z axes plus three rotational axes—under software control. Temperature ramping must be integrated with the vibration sequences.
Vibration levels should be monitored continually by the software during operation, and screening should be shut down by the controller if an accelerometer output or the composite signal from all accelerometers exceeds specified limits.
Stress Profiling
Determining the appropriate screening levels is more an art than a science. For years, practitioners have been refining the elusive variables in the optimum stress profile. Today, they continue to fine-tune the process as they collect feedback from as many sources as possible.
An excellent example of fine-tuning was presented by Jeffrey Marshall, vice president of sales and marketing at M/RAD. “The appropriate stress levels are those that identify workmanship issues but don’t induce failures in good products,” he explained. “A good starting point is applying environmental stresses at low levels, raising them gradually, and observing the results.
“Vibration at a resonant frequency, for example, should uncover some latent defects. Some engineers even seed the product with a few known defects during the experimental stage to assure the sequence works,” Mr. Marshall concluded.
Ideally, stress levels and durations for ESS are determined empirically, using knowledge gained in HALT. Or, you can set up a slightly aggressive screen and tune it based on failure information collected over a few hundred units. Harmonizing temperature and vibration profiles is straightforward. You must vibrate during both hot and cold soaks and then adjust the level and duty cycle based on those results.
A screening program needs boundaries. First, it must be severe enough to precipitate failures that are inevitable in the product’s expected life span. Secondly, it must not be harsh enough to consume a significant percentage of the product’s design life.
A proof-of-screen sequence sometimes is used to test these boundaries. It shows that the screening you are using is effective in finding weaknesses and verifies that ESS is not taking more than 10% of the design life out of the product.
A common rule of thumb is to run a sample product through ESS 10 times. If no defects are found in that sequence or if failures seem excessive, the ESS limits probably are inappropriate.
No matter how you choose to implement an ESS program, you can never walk away assuming that the screening rules are optimum. Results must be monitored continually and the profile fine-tuned as often as necessary to weed out defects without significantly reducing product life.
Conclusion
Look for continuing refinements in the art of establishing stress profiles. Feedback from analysis of a variety of products will furnish the database for this trend, and communications between practitioners and consultants will provide the tools.
Published by EE-Evaluation Engineering
All contents © 2000 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.
December 2000