Ensuring That Products Will Work
by Tom Lecklider, Senior Technical Editor
If all products were operated in controlled,
benign conditions, there would be little need for environmental
testing laboratories. When used in the real world, equipment is
exposed to shock, vibration, temperature extremes, dust,
humidity, and many more factors that affect performance. As part
of a thorough design process, manufacturers subject products to
simulated stresses to ensure that specifications will continue
to be met in the actual use environments.
The nature of the product determines which
tests are most important. For example, an engine air filter
intended for rugged military applications would need to
withstand severe sand and dust tests in addition to mechanical
shock and vibration. If the filter were made of metal, it also
might require salt spray and high humidity tests to prove it
could endure these conditions for prolonged periods of time.
Because each kind of product is different and
is used differently, many types of environmental tests are
available. Several of these capabilities are demonstrated in
real applications submitted by three accredited testing labs.
Military-Grade Vibration Testing
Standex Electronics, a global manufacturer of
custom magnetics, reed switches, and sensors for the medical,
military, industrial, and automotive markets, must meet
stringent customer and regulatory requirements. Although it
operates its own product-line testing equipment, the company
also relies on expert, accredited testing services for
time-sensitive, large-capacity, and unique scenarios.
To provide rapid, complete military-grade
transformer vibration testing for an aerospace application,
Standex turned to ESSC Test Laboratory. The lab has 20 years of
experience and is ISO/IEC-compliant and accredited by The
American Association for Laboratory Accreditation (A2LA).
The transformer test required precise
equipment and customized fixturing. Successful testing and
useful reporting hinged on how the parts were attached to the
vibration table and how well the behavior was analyzed.
Each transformer weighed less than a pound,
and 12 were tested at the same time (Figure 1). An LDS
750 System Shaker with a 6,000-lbf rating was used. It was
coupled to a slip plate with a 19"x 21" interface plate to which
were bolted two hollow aluminum cubes. These fixtures provided
convenient mounting locations for six test boards, each with two
transformers and related components.
Figure 1. Standex Transformers Similar to Those Tested 
Courtesy of Standex Electronics and ESSC Test Laboratories
ESSC performed the test in accordance with
MIL-STD-202G, Method 204D Vibration, High Frequency. The test
entailed 20-minute logarithmic cycling through the 10-Hz to
2,000-Hz frequency range, repeated 12 times over a 12-hour
period, with the transformers oriented in each of three
directions. These requirements correspond to Method 204D Test
Condition E, which applies a peak-to-peak displacement of 0.06"
from 10 Hz to about 127 Hz. For frequencies above 127 Hz, the
displacement is reduced to maintain a peak 50g acceleration (Figure
2).
Figure 2. Vibration Test Graph
Courtesy of Standex Electronics and ESSC Test Laboratories
Method 204D supports testing with components
de-energized or operating under specified load conditions. In
this case, Standex elected to test the transformers together
with related electrical components in a de-energized state.
For temperature and humidity testing, Method
106G was followed. It requires cycling five times among 25°C @
95% humidity, 65°C @ 95% humidity, and -5°C, dwelling from three
to eight hours at each temperature.
The customer specified three temperatures for
thermal shock testing: 130°F, 155°F, and 170°F. The units were
soaked at each temperature for 24 hours with inspection every
eight hours.
According to Method 107G Thermal Shock, the
transformer temperature was reduced to 25°C in less than five
minutes after soaking at a high temperature. Similarly, the
transformer temperature was increased from room temperature to
the required maximum, also in less than five minutes. In
contrast to these tests run in air at different temperatures, an
almost instantaneous thermal shock can be created under Method
107G by using liquid immersion. The customer chose air in lieu
of immersion.
Because ESSC is a division of Cincinnati
Sub-Zero, the manufacturer of a long line of temperature-based
products and thermal testing chambers, a great deal of practical
environmental test experience is available. The lab delivers
tests specially suited for design verification, product
performance, regulatory compliance, failure analysis, life
cycle, and environmental stress. In addition, ESSC uses the
latest software to produce follow-up reports that document and
analyze testing for customer and regulatory purposes.
Vibration Test Fixturing for Small Parts
Designing and building a product mounting
fixture can be one of the most difficult aspects of a vibration
testing project. Fixturing for vibration testing is intended to
provide good transmissibility. This requirement may be
compromised by the needs to secure the unit under test without
damaging it and to access it easily for assembly and removal.
This is especially important when high-volume testing is
required. In addition, the fixturing has to be built quickly and
inexpensively.
Chris Finch, technical sales manager at Trace
Laboratories-Central, explained that traditional vibration
fixtures are made by drilling and tapping aluminum cubes or
plates and bolting down the product via its built-in flanges. If
the unit doesn't have built-in mounting holes, more creative
methods are necessary.
One method commonly used to secure test
samples clamps the unit to an aluminum plate. Typically, the
clamping fixture is built by drilling and taping holes in the
plate and inserting threaded rods. An aluminum bar then is used
to create the clamp, with protective foam between the samples
and the fixture. Another method is to double-stick tape the
product to the shaker table. Each method is application specific
and has cost and efficiency advantages.
There are instances when traditional
vibration fixtures are not an effective means of mounting a test
unit. Fixturing issues can occur when conducting vibration
testing on oscillators, transceivers, Ethernet adapters, flash
drives, and similar components.
Often, these units are small, and many are
tested at a time. They are round or have nonparallel sides that
can be crushed if not handled properly. These components
typically do not have mounting holes and cannot easily be
attached to an aluminum plate. Using the sandwiching method will
not work because the components can be damaged or slip out due
to their nonstandard shape.
Instead, Trace Laboratories uses paraffin wax
to rigidly mount these multiple, small, hermetically sealed
units to the vibration equipment. Paraffin wax is pliable,
inexpensive, and easy to use. It also is noncorrosive and an
extremely good electrical insulator with an electrical
resistivity of at least 1013
Ω-m.
Multiple test units can be placed into a
block of paraffin wax, and the wax is placed onto the vibration
adapter plate. The vibration inputs of the shaker system are
directly transferred to the test units. Paraffin wax allows the
vibration testing to be conducted without the significant time
and expense of complicated mounting fixtures or fear of damaging
fragile components.
Although some laboratories still are using
double-stick tape, paraffin wax is a better alternative.
Double-stick tape can provide quick, inexpensive mounting of
lightweight components; however, it cannot be used for all
product geometries and is difficult to remove. And, while
double-stick tape holds a unit by one surface, a unit can be
embedded into the wax, allowing for better support and vibration
transmission.
Upon completion of the vibration, the units
can be manually removed and easily cleaned. The paraffin wax is
removed by melting it with a heat gun.
Removal and cleaning can be conducted
quickly, and additional samples can be under test in a few
minutes. As long as it remains clean, the wax can be used
repeatedly.
There are some considerations when working
with paraffin wax. The testing should only be conducted at room
temperatures because paraffin wax can begin melting at
temperatures as low as 47°C. If the units are powered and
monitored during the testing, be aware of their heat generation.
In addition, the samples must be able to withstand 47°C during
removal of the wax.
Paraffin wax should only be used on
hermetically sealed units since the wax may be difficult to
remove from part openings or crevices. Care must be taken when
performing shock test fixturing because the wax may not properly
hold the units during high impact.
Always ensure proper transmissability by
mounting a response accelerometer to the test units. As a rule
of thumb, only consider paraffin wax mounting for components
that are too small or delicate to be mounted conventionally.
Much of the experience at Trace Labs has been with very
lightweight ICs, as shown mounted in paraffin wax in Figure 3.
Figure 3. ICs Mounted in Paraffin Wax for Vibration Testing
Courtesy of Trace Laboratories
Explosive Pressure Pulse Simulation
A customer recently asked Aero Nav
Laboratories whether simulation tests could be performed to
determine the survivability of a piece of equipment when
subjected to explosive pressure pulse blasts. The equipment was
designed to be robust and expected to survive moderate levels of
pulse blasts like what would be experienced in the survivability
zone of an explosion. This zone is defined as the area where the
explosive effects are less severe than would be seen at the
point of inception of a blast.
Sheldon Levine, Aero Nav Laboratories vice
president of marketing and business development, explained that
the lab is not equipped to perform testing using actual
explosives on-site and, furthermore, is limited by municipal
regulations. Another means of achieving the test objectives was
required.
A blast tube was constructed using a
heavy-walled steel pipe open at one end and equipped with a
pressurized air chamber at the other (Figure 4). A valve
was provided to rapidly release the entrapped pressure within
the chamber. The equipment under test was fixtured to simulate
its actual mounting and installed in the pipe. The open end of
the pipe was set up with a converging section to maintain the
pressure and velocity of the air as it traversed the length of
the pipe.
Figure 4. Blast Tube
Courtesy of Aero Nav Laboratories
When an air-blast event occurs, two effects
will exist: peak pressure in the wave front and drag loading.
The peak pressure is the maximum pressure in the wave front as
it passes the test item. Drag loading represents the effect of
the bulk of the air mass moving past the test item. Drag loading
is the same as that encountered by an object moving in a fluid
stream.1
In a ground-level explosive event in an open
field, the explosive effects radiate outward in a
semihemispheric manner. However, when constrained within a long,
relatively small-diameter pipe, the effects are directed axially
outward toward the open end. Consequently, the equipment under
test in the tube experiences the full effect of the sudden
release of pressure and the rapid movement of the air past the
unit.
An idea of the magnitude of the pressure
created by an explosion can be gained by reviewing several
documents on the U.S. Department of Energy website. For example,
Appendix B of a 1995 Defense Nuclear Facilities Safety Board
Memorandum presents an interesting comparison of the pressures
that might be developed by burning hydrogen
with oxygen vs. exploding a hydrogen-oxygen mixture.2
Burning a confined hydrogen-oxygen mixture
was calculated to produce a pressure about 10x higher than
atmospheric. A containment vessel must withstand twice that
value to account for reflected pressure. However, an actual
explosion with reflected shock waves would create a maximum
pressure greater than 122 atmospheres or greater than 1,800 psi.
The peak pressure at the site of an explosion
can be extremely high. In contrast, the testing carried out by
Aero Nav Laboratories simulated a lower peak pressure and drag
loading some distance from the blast site in the survivability
zone.
For the device being tested, a blast tube was
considered to provide a sufficient simulation of the actual
explosive event. The pressure buildup and subsequent decay in
the blast tube were measured by a pressure transducer placed on
the upstream side of the equipment under test. Figure 5
presents a typical trace showing the pressure vs. time.
Figure 5. Simulated Air Blast Pressure vs. Time
Courtesy of Aero Nav Laboratories
The pressures, usually shown as psig, are
presented as percentages of peak pressure since the actual
values are considered proprietary. Rapid buildup of pressure in
an almost linear manner is followed by a slower, more gradual
drop-off.
The entire event takes place in slightly more
than 400 milliseconds. The rise of pressure to its peak level
occurs within 100 milliseconds. The peak pressure would be much
higher very close to the actual site of a blast; however, the
pressure pulse shown in Figure 5 corresponds to expected
conditions in the survivability zone.
Multiple test runs were performed at various
peak pressures. The equipment under test was found to have
survived the effects of the air blasts with no physical damage
or functional degradation.
Summary
If it's done correctly, simulated
environmental testing can quickly prove a design's value or lack
thereof. A product designed to survive the environment in which
it will be used won't be damaged by testing. On the other hand,
marginal test results or outright failure should set off warning
bells.
Achieving a good correlation between
simulated tests and the actual use environment is critical. As
the Aero Nav Laboratories example showed, the actual test
apparatus doesn't need to resemble the real environment as long
as the values of important parameters are similar. In this case,
peak air pressure and drag loading had the greatest effect on
the product being tested. The tests were meaningful because the
simulated quantities sufficiently matched real values expected
in the survivability zone.
Sometimes, successful testing relies on a
lab's experience and ingenuity. Because they know how a test
should be run, seasoned test engineers can easily spot errors.
For example, resonance caused by a loose fixture could be
confused with resonance in the product itself. This problem and
others related to fixturing are compounded when the product
being tested is very small, delicate, or an odd shape. Trace
Labs' use of paraffin wax seems an obvious solution, but it's a
good bet not all test engineers are familiar with it.
Beyond technical test capabilities, sometimes
customers expect a test lab to help solve business-related
issues such as tight delivery schedules or unanticipated large
orders. These kinds of considerations were a factor in Standex's
decision to contract ESSC Test Labs for transformer testing.
What kinds of environmental testing do your
company's products require? They may not be as extreme as a
simulated explosion or as extensive as those called for under
MIL-STD-202G. Nevertheless, whatever is needed, it's likely that
standardized tests already exist and a number of labs have the
necessary equipment and experience to perform them.
References
1. Glasstone, S. and Dolan, P.J., The Effects
of Nuclear Weapons, Third Edition, 1977.
2. Cunningham, G. W., "Defense Nuclear
Facilities Safety Board Memorandum," 1995,
www.hss.energy.gov/deprep/1995-2/tr95m27a.pdf