Apr 14, 2014

Failures of Metals

Metal failure due to fatigue in a crankshaft.

Crankshaft that failed due to fatigue. Photo Credit: Peter Lewis: Fatigue of Material. English Wiki 27/2/2008.

The potential consequences of failed metal components can hardly be overstated.  Economic damage, potential injury, and loss of production capability are all likely outcomes of a metal component failure.  For this reason, metal failures are often investigated in-depth with the intent of avoiding a repeated failure in the future.  Determining the cause of failure requires a systematic, step-by-step investigation. Fortunately the science of forensic engineering is well established and allows the determination of the root cause of failures with a high degree of accuracy.

The primary types of metal failures are:
• Overload
• Fatigue
• Stress-corrosion cracking
• Hydrogen embrittlement

Overload failures are caused by loads that stress the metal beyond its ultimate strength.  These types of failures are generally found in components that were simply loaded well beyond their design specifications, whether purposefully, or accidentally.

There are two categories of overload failures. Ductile failures occur when the component stretches or bends to some degree before failure. Brittle failure occurs when the component breaks with little or no distortion. The manner of failure is determined by the type of material, its heat treatment, etc.  Overload failures occur very rapidly.

While dangerous and inconvenient, overload failures are not the most common type of metal failure, and are normally avoided by simply keeping the component loading levels well below design specifications.

Fatigue failures are caused by repeated or fluctuating loads and may occur even when the part is stressed below its yield strength.  Fatigue failures are the most common types of metal failures. Fatigue failures occur very slowly and yield recognizable markings on the fracture surface.  Repeated loading cycles allow a small material failure to grow unchecked into a larger one, potentially leading to total component failure.  The fatigue crack begins at the surface of the material, usually at a point such as a machine mark, corrosion pit, surface scratch, or other imperfection that allows the crack to gain a foothold and propagate.  The crack then progresses slowly across the fracture face leaving a distinctive pattern called “beach marks” which indicates the direction of progression. Final failure occurs when the remaining unaffected area is insufficient to carry the applied load.

Beach marks showing progression of fracture in transport rail.

Beach marks showing the progression and direction of a fracture in a failed railroad rail. Source: Transportation Safety Board of Canada. www.tsb.gc.ca

Stress-corrosion cracking can occur where stresses acting on a part, in combination with a chemical environment, can lead to failure.  This type of failure normally occurs under circumstances where loading alone would not cause failure without the addition of the chemical influence.  The failure can be sudden and unexpected in normally ductile materials.  This can be especially pronounced in metals experiencing exposure to chemical agents at high temperatures or pressures.

Branching pattern seen in stress corrosion cracking.

Stress corrosion cracking is often identified by its unique branching pattern. Image source: NASA Corrosion Technology Laboratory. corrosion.ksc.nasa.gov

Hydrogen embrittlement can lead to brittle fracture by the unintentional absorption of hydrogen during forming and finishing processes such as chromium plating. It occurs most frequently in high strength steel.

Each of the above failure types leaves unique identifiers, which can help identify the exact nature of the metal failure.  Each case has its unique properties, but in common, they share the characteristic that they can be difficult to anticipate, and account for during the design process.

Because each of the above failure types can occur unexpectedly, and sometimes occur while the material is loaded below design specifications, thorough testing under conditions that simulate the anticipated environment is often the best way to identify where potential failures may occur.

Mar 31, 2014

Design of Custom Testing Programs – Part 2

Changes in pneumatic pressure can be obtained in a variety of ways and with a number of devices, but determining a way to apply these pressures to one side of a train window, for example, may be challenging. A chamber or enclosure that the window will be mounted on must be fabricated so that pressures on one side may be varied according to the required conditions. In effect, the difference of the pressures acting on the two sides of the window is what causes a load on the glass pane. The test engineer must determine the means by which the pressure and vacuum cycles are to be applied to the side of the window within the enclosure. An air compressor, blower, or fan may be utilized, but other less conventional methods like a bellows system could also be considered.

Just like there are many possible means of applying pressure to one side of the window, there are also many different ways to complete the other aspects crucial to the test. Other important processes to consider are venting the pressurized air, applying a vacuum, measuring the pressure changes, timing the cycles, ensuring minimal deviation of pressure values between cycles, and accurately repeating these pressure cycles thousands, if not millions of times. Since it is the fatigue life of these train windows due to pressure fluctuations that is being tested, and it may take millions of cycles for fatigue cracks to develop, the most important thing to consider when designing the test setup is its reliability. The test engineer needs to diligently forecast the anticipated lifespan of each and every piece of equipment used during the test so that appropriate quantities could be ordered, and costly surprises don’t occur.

Reliability of equipment used in a test program must be closely considered, especially when testing is expected to last for an extended period of time, or if it simulates a large number of fatigue cycles. Source: en.wikipedia.org

Reliability of equipment used in a test program must be closely considered, especially when testing is expected to last for an extended period of time, or if it simulates a large number of fatigue cycles. Source: en.wikipedia.org

The test setup needs to be designed so that manufacturability of any custom parts is not unnecessarily difficult. Furthermore, when a test program is expected to last for weeks or months such as this one, the setup needs to be designed so that minimal labor needs to go into maintaining the desired test conditions. Automation is always a desirable option when it is cost-effective. Naturally, pricing is one of the most important aspects of designing a testing program. The setup must always be optimized so that material, equipment, labor, and power costs are minimized. A lot of the times when multiple units need to have the same tests performed on them, it is more cost-effective to test them simultaneously. Both the customer and the testing laboratory benefit from a streamlined, efficient, and cost-effective test program.

While attempts must always be made to minimize the cost of testing, concessions cannot be made which may affect the validity of the test program. The specifications to which an item needs to be tested must never be compromised; otherwise the test is inconclusive. The most important aspect of designing a custom test program is the ability to foresee any issues that may come up once the test is in progress. Issues relating to durability, reliability, strength, stiffness, stress concentrations, temperature, data measurement and recording, safety, and many others need to be considered. While testing may be the last step in the product design process, designing the test program may be as complex as designing the item itself.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Mar 20, 2014

Design of Custom Testing Programs – Part 1

Testing is crucial to the successful development of any type of new product, whether it’s meant for military, commercial, or consumer use. While testing is often done based on widely acknowledged specifications, it doesn’t mean that the process is similar for every item. As a matter of fact, testing programs are unique to the unit under test, even if the same type of test is performed on countless other items. While personalizing a test for a specific unit often only consists of designing a fixture which attaches it to a testing apparatus like a shaker table or shock machine, other times a complete test program needs to be developed around the specific test parameters for that unit.

Testing revolves around subjecting equipment to worst-case scenario conditions that could be expected in its service life. Clearly, depending on the type of item being tested, any of a multitude of tests may be performed on it. It is up to design engineers and acceptance personnel to determine the appropriate testing to be performed on an item. If equipment is designed for military shipboard use, it is expected that it will successfully pass such standard tests as shock, vibration, and salt spray. However, certain equipment will experience unique conditions that will dictate the type of testing it must undergo. For example, when a train enters a tunnel opening or passes alongside another train at high speed, there is a momentary positive and negative pressure pulse that acts upon the outer surface of the train. These aerodynamic loads and pressure waves occur due to the flow behavior of fluids (one must remember that although invisible, air is certainly a fluid). Not only do train windows have to be designed robustly enough to be able to handle this pressure pulse, they must be able to withstand the fatigue induced by the thousands of these waves encountered throughout their lifetime.

Pressure Pulse Measurement

Pressure pulse values are measured experimentally by setting up a false window with pressure transducers, and subsequently running the train through tunnels and beside other trains at specific velocities. This image, and experimental results were obtained from the paper "Aerodynamic Loading of Trains Passing Through Tunnels" by Tarada, Himbergen, and Stieltjes.

Determining the appropriate test setup for pressure cycling of train windows (as well as for any test) starts by taking measurements in the field to determine the magnitude of the conditions that may be encountered. A lot of times these values are specified in published test specs, but when dealing with such a specific condition as pressure waves on the outer surface of a train, it may be necessary to obtain data experimentally (see image above). Once the test condition is known, it is up to the test engineer to determine the best way to simulate these conditions in a laboratory setting.

Check back soon for the conclusion of this article!

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Catastrophic Structural Failures

AERO NAV LABORATORIES has prepared a seminar describing a number of structural failures. We thought it would be of interest to our readers to provide a descriptive summary of these failures along with their causes. The full seminar may be viewed by clicking on the following link:

Aero Nav Labs Seminar – Engineering Failures and Design Principles

A review of the causes of failures shown in the table provides some valuable engineering insights:

  • Advancing into new areas without a full understanding of the technology
  • Lack of conservative designs
  • Failure to fully understand the influence of external conditions

Addressing these factors should help to prevent failures in future designs.






Tacoma Narrows Bridge Collapse of bridge in Washington State Torsional vibration Low stiffness and low damping. Vibration caused by dynamic aeroelasticity or otherwise known as flutter. Theory of flutter was not well known at the time.
WW II Liberty Ships Hull failures Brittle cracking of hulls Transition of hull properties from ductile to brittle due to cold weather conditions. Also, use of all welded construction in lieu of riveted construction, which would have stopped crack formation. Theory of fracture mechanics was not well known at the time.
New Orleans levees Breaking of levees during storm Levees were breached Poor design of levees. Failure to anticipate worst case storm conditions.
Tower of Pisa Tilting of tower in Pisa, Italy Excessive tilting of tower Poor subsoil conditions
De Havilland Comet Jet Aircraft Structural failures of cabin. De Havilland Aircraft Company, England – First commercial jet aircraft Failures of windows leading to fuselage cracking Cabin windows had square corners leading to excessive stress concentrations. Theory of fracture mechanics was not well known.
Millennium Bridge Excessive vibration. Pedestrian bridge over River Thames, London. Lateral vibration due to the slight sideways motion of pedestrians while walking. Low lateral structural stiffness and low damping.

Deepwater Horizon Drilling Rig Collapse

Image from Aero Nav Labs' seminar on Catastrophic Engineering Failures. To view the entire seminar slideshow please click on the following link: Aero Nav Labs Seminar - Engineering Failures and Design Principles

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Feb 24, 2014

Torsional Vibration


Torsional vibration is a topic that is not always well understood. This type of vibration is typically the angular vibration of a component such as a shaft along its axis of rotation. However there are other types of angular vibration such as the torsional vibration failure of the  Tacoma-Narrows Bridge in Washington State. The bridge, a long narrow structure, was excited torsionally by the effects of flutter.  This is a phenomenon of concern in aircraft design whereby significant interactions between the ambient wind conditions and the structure occur. The torsional moments that were generated served to overstress the supporting cables and consequently ruptured the roadway.


In power transmission systems the generated torques and/or driven components may not react to these torques in a smooth manner. Components such as elastic drive belts, worn gears, and misaligned shafts can generate non-linear torques. Furthermore, the transmission components such as the shafting may itself twist and vibrate. Typical sources of non-smooth torques are those generated by internal combustion engines, reciprocating compressors, and universal joints if the shafts are not parallel. This is caused by the non-linear nature of the mechanisms.

Torsional Vibration

This cross section of an internal combustion engines cylinder, piston, and crankshaft shows how torsional vibration can be induced based on the geometry of a mechanism. Image source: http://www.musclemustangfastfords.com


The torsional failure of the Tacoma-Narrows Bridge was caused by insufficient torsional stiffness and low damping. Both factors contributed to the failure by allowing excessive motions with little ability to absorb and dissipate the energy.

In power transmission systems such as automobile engines, there is little inherent damping to reduce the vibration level. Therefore the vibration is controlled by the use of torsional dampers located at the front of the crankshaft. Another type of device is the tuned mass damper that limits vibration at specific torsional natural frequencies. This type of device is similar to that used in buildings to limit its motion during high wind conditions and earthquakes.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Feb 13, 2014
Comments Off

Fire Chamber


AERO NAV LABORATORIES has designed and constructed a fire chamber for flame testing items, such as hydraulic hoses, utilizing its own engineering resources.

The chamber is utilized to:

  • Test the fire flame resistance of items such as protective sleeving on hydraulic hoses while flowing heated, pressurized water. It tests the ignition resistance and burning characteristics of the material.
  • Subject materials to elevated temperatures
  • Subject materials to direct flame impingement. The chamber demonstrates the response of materials and products under controlled flame conditions.


The chamber consists of the components shown in the schematic below. It consists of a bank of jets supplied with propane gas at a pressure of 50 psi. The specimen under test is placed in the chamber above the gas jets. The circulating water is heated and pressurized by the heater and pump to a maximum temperature of 80 degrees C  (176 degrees F) and pressurized to 72.5 psi.  The maximum propane gas temperature is 800 degrees C (1472 degrees F).

Fire Chamber Schematic

Schematic showing the components of Aero Navs fire chamber and fluid circulating system. The photo on the right shows the complete system.


Hydraulic hoses are often provided with protective fire resistance sleeving.  The protective sleeving is typically rated, as “fire resistant”, not “fire proof”.  That is, it must remain tight (no leakage allowed), under the flame and circulating water conditions for a period of time, typically 30 minutes. This time frame allows for a fire extinguishing system to sense and extinguish the fire. Furthermore, it must pass a hydrostatic pressure test at 1.5 or 2 times the operating pressure for a ten minute period, after the fire test. The choice of pressure is dependent upon customer requirements. It is important to note that some degradation, such as charring of the protective sleeve, is allowed providing it meets the above criteria of no leakage, and passing of the hydrostatic pressure test.

The chamber can also be used to evaluate the effects on the material of the elevated temperatures and direct flame impingement.

Fire testing of a hose

Fire resistance testing of a hydraulic hose at Aero Nav Labs.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Feb 4, 2014
Comments Off

Vibration Theory and Applications


Vibration testing is performed to demonstrate the ability of equipment to withstand the expected dynamic vibrational stresses encountered in service.  Vibration is an oscillatory phenomenon, which may be periodic or random, or both occurring simultaneously.


One type of vibration, known as free vibration, occurs when a system is set off with an initial input and is then allowed to vibrate without restraint.  Vibration will occur at one or more of its natural frequencies and will eventually return to equilibrium after dissipating the input energy. The other type of vibration, known as forced vibration, occurs when a disturbance, which varies with time, is applied to the system.  The disturbance may be periodic, transient, random or combinations thereof.


Vibration testing is performed by mounting the unit under test on a shaker table with the use of a mounting fixture. The fixture is a key component, which if improperly designed can cause incorrect test results. The fixture design should consider the following:

  • Allow for ease of mounting of the test unit
  • Allow for testing in each of the three orthogonal directions, if required
  • Ensure the absence of fixture resonances within the test frequency range
  • Weight and force limitations of the vibration machine
  • Distribution of vibration energy uniformly throughout the test item


The basic vibration modes are as follows:

  • SINUSOIDAL  – vibration characterized by amplitudes which vary sinusoidally with time
  • RANDOM  -  vibration characterized by irregular non-repeating patterns and amplitude versus time . Random vibration more closely simulates the real world than sinusoidal vibration.  A typical example are road inputs to a moving vehicle.
Sinusoidal and Random Vibration

The figure above shows the distinct difference between sinusoidal and random vibration.

Mixed vibration modes are as follows:

  • SINE ON RANDOM  – Sine and random vibration are produced simultaneously to simulate sources where both sinusoidal and random vibration are generated in service
  • RANDOM ON RANDOM  – Two different random vibration patterns are produced simultaneously to simulate sources where two separate random inputs are generated in service

Mixed vibration modes more closely simulate the real world environment than either sine or random separately.


Test pulses are also available which simulate the vibration caused by overpressure pulses produced in the vicinity of armaments.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Jan 23, 2014
Comments Off

Effect of Harmonics on Power Distribution Systems

By: Artur Litovka, Project Engineer at Aero Nav Labs

If improperly designed or rated, electrical equipment may malfunction when voltage and current harmonics are present in the electrical system.

Harmonics have existed in power systems for many years. In the past, most electrical equipment used balanced loads referred to as linear loads (loads where the voltage and current follow one another without any distortion to their pure sine waves). The current drawn by the load is proportional to the voltage and impedance, and follows the envelope of the voltage waveform. Examples of linear loads are constant speed induction and synchronous motors, resistive heaters, and incandescent lamps.

The rapid increase in electronic device technology such as diodes, thyristors, variable frequency drives, electronic ballasts, battery chargers, and switching mode power supplies cause industrial loads to become non-linear. The non-linear load connected to the power system distribution will generate current and voltage harmonics. The current and voltage have waveforms that are non-sinusoidal, containing distortion, whereby the 60 Hz waveform has numerous additional waveforms superimposed upon it, creating multiple frequencies within the normal 60 Hz sine wave. The multiple frequencies are harmonics of the fundamental frequency. For example, if the fundamental frequency is 60 Hz, then the 2nd harmonic is 120 Hz, the 3rd is 180 Hz, and so on.

The graph above shows how a fundamental frequency and several harmonics combine to produce a resultant waveform

The graph above shows how a fundamental frequency and several harmonics combine to produce a resultant waveform. Source: www.hersheyenergy.com

To quantify the distortion, the term total harmonic distortion (THD) is used. The THD value is the effective value of all the current harmonics added together, compared with the value of the fundamental current. Normally, current distortions produce voltage distortions. However, when there is a stiff sinusoidal voltage source (when there is a low impedance path from the power source, which has sufficient capacity so that loads placed upon it will not affect the voltage), one need not be concerned about current distortions producing voltage distortions.

Power systems designed to function at the fundamental frequency, which is 60 Hz in the United States, are prone to unsatisfactory operation in the presence of harmonics.

  • There is an increasing use of variable frequency drives (VFD) that power electric motors. The voltages and currents emanating from a VFD that go to a motor are rich in harmonic frequency components.
  • The harmful effects of harmonic voltages and currents on transformer performance often go unnoticed until an actual failure occurs.
  • Many industrial and commercial electrical systems have capacitors installed to offset the effect of low power factor. Since capacitive reactance is inversely proportional to frequency, unfiltered harmonic currents in the power system find their way into capacitor banks. These banks act like a sink, attracting harmonic currents, thereby becoming overloaded.
  • The flow of a normal 60 Hz current in a cable produces resistance losses, and current distortion introduces additional losses in the conductor. Because of both the fundamental and the harmonic currents that can flow in a conductor, it is important to make sure a cable is rated for the proper current flow.

Often, the operation of electrical equipment may seem normal, but under a certain combination of conditions, the impact of harmonics is enhanced, with damaging results.

Once you have recognized that harmonics are in an electrical system, the next step is to carry out tests to determine the magnitude and type of harmonics. Harmonic analyzers are effective instruments for determining the wave shapes of voltage and current, and measuring the respective frequency spectrum.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Jan 15, 2014
Comments Off

Test Fixture Qualification

Upon the conclusion of the design of a vibration test fixture it is sometimes required that the fixture be qualified to determine its resonant characteristics. This requirement is imposed to demonstrate that the dynamic motion of the fixture will not adversely affect or skew the results of the vibration testing.

Test fixture qualifications can be performed numerically by hand analysis or finite element analysis (FEA), or in the laboratory by impact or bump testing. Another test technique is modal analysis. Except for simple fixtures, test methods are preferred, since hand analysis using basic static and dynamic methods usually results in imprecise data. FEA methods are complex and costly.

Finite Element Analysis

Finite element analysis is a computer-based analytic tool for solving field-flow problems. It allows for problem solving in the areas of structures, fluid flow, heat transfer, and electric fields due to the similarity in the partial differential equations which characterize these fields.

The FEA for structural design starts with modeling of the structure by dividing it into an equivalent system of simple elements, such as rectangles or triangles, with easily obtained stress and deflection characteristics. Upon specifying the material, material properties, boundary conditions, and loads, the analysis is completed by computer programs, utilizing arrays of matrix equations.

FEA allows for the determination of free-vibration natural frequencies and the associated mode shapes of a structure. It provides valuable information for use in the design stages of a program allowing optimization of the design by varying key parameters.

Impact or Bump Testing and Modal Analysis

Impact or bump testing is a powerful tool for testing the vibratory response of structures. It consists of a device for impacting energy, such as a hammer, into a structure. The frequency spectrum of the hammer force impact signal is nearly flat over a wide frequency range. Figure 1 shows the time history of the force impact for various hammer tips. It is a typical impulsive load response. Figure 2 shows the frequency spectrum of the impact force. Therefore, a single impact will excite all the natural frequencies within the frequency band. In other words, the energy content of an impulsive impact is broadband.

Time history of the force impact for various hammer tips

Figure 1: Time history of the force impact for various hammer tips

Frequency spectrum of the force impact for hammer tips

Figure 2: Frequency spectrum of the force impact for hammer tips

Utilizing a hammer with a force transducer and displacement transducers placed at numerous locations throughout the test item to measure the response motions, along with a Fast Fourier Transform (FFT) spectrum analyzer allows the determination of the structure’s dynamic response. The response is defined by the natural frequencies, mode shapes and damping factors. This method is called modal analysis. Electric impact hammers are also available to yield controlled repeatable impact forces.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com

Jan 6, 2014
Comments Off

Vibration Test Fixtures

Vibration test fixtures are required to allow mounting of the test specimen to the vibration table as well as to allow for testing in the three orthogonal directions.

The design of test fixtures is critical to avoid errors in equipment test response due to any resonances of the laboratory’s shaker, table (movable part of the shaker) and the fixture itself. Ideally, the laboratory mounting should replicate the physical conditions seen in service, such as stiffness, mass and the consequent resonant responses of the actual service installation.

Theory of Fixture Design

The key issues in the design of a vibration test fixture are as follows:

1. Allow for ease of mounting of the test item to the vibration machine
2. Allow for vibration testing in each of the three orthogonal directions with minimum cross talk, i.e., motion in the two orthogonal directions not being tested
3. Ensure the absence of fixture resonances within the specified test frequency range by tailoring the dynamic response of the fixture and table
4. Weight and force limitations of the vibration machine
5. Distribution of vibration energy uniformly throughout the test item


The fixture is the interface between the test unit and the table of the vibration machine. It must allow for proper mounting to the table. The number and size of attachment points must be sufficient to transmit the vibration energy equally to the fixture without losses.

Vibration Fixture for Ship Hatch Door

Directional Characteristics

Vibration testing is usually performed in each of the three orthogonal axes, one axis at a time. The real service environment is seldom uni-directional. However for testing purposes, to allow for control and standardization of test specifications and methods, uni-directional motion is desired.

To avoid or minimize orthogonal motion, commonly known as cross-talk, various steps are taken;
· Maintain the center of gravity of the test unit and fixture at or close to that of the vibration machine table
· Increase fixture stiffness and mass where required to minimize the effects of test item resonance

Dynamic Response

Ideally, the vibration response characteristics of a test fixture should replicate those of the actual service installation. Often, however, these characteristics are not specified by the customer or are not known. The designer of a test fixture must therefore use engineering judgment based on experience to achieve an optimum design. Generally the approach is to make the test fixture as rigid as possible, within the allowable weight limits of the shaker. The test fixture should therefore have no resonances within the frequency range as specified by the customer. That is, the first resonant frequency should be above the maximum test frequency.

Vibration Machine Limits

The weight (test item, fixture plus table) and force limits of the machine must be observed to avoid overloading and possible damage.

Energy Distribution

The test fixture shall be designed to transmit vibration energy from the machine table to the test item in an approximately uniform manner over its mass. This goal may be achieved by designing the fixture to be as rigid as possible within the weight limitations of the machine.

Contact Aero Nav Laboratories with any questions or comments.
Visit our websites for more articles and useful information:
www.AeroNavLabs.com & www.ContractWorkSolutions.com




Google Analytics