Design is the process of choosing the geometric shape, materials, manufacturing method, and other details needed to completely describe a machine, vehicle, structure, or other engineered item. This process involves a wide range of activities and objectives. It is first necessary to assure that the item is capable of performing its intended function.
For example, an automobile should be capable of the necessary speeds and maneuvers while carrying up to a certain number of passengers and additional weight, and the refueling and maintenance requirements should be reasonable as to frequency and cost.
However, any engineered item must meet additional requirements: The design must be such that it is physically possible and economical to manufacture the item. Certain standards must be met as to esthetics and convenience of use.
Material Selection in Mechanical Design
Environmental pollution needs to be minimized, and, hopefully, the materials and type of construction are chosen so that eventual recycling of the materials used is possible. Finally, the item must be safe and durable. Safety is affected not only by design features such as seat belts in automobiles, but also by avoiding structural failure. For example, excessive deformation or fracture of an automobile axle or steering component can cause a serious accident.
Durability is the capacity of an item to survive its intended use for a suitably long period of time, so that good durability minimizes the cost of maintaining and replacing the item.
For example, more durable automobiles cost less to drive than otherwise similar ones that experience more repairs and shorter life due to such gradually occurring processes as fatigue, creep, wear, and corrosion. In addition, durability is important to safety, as poor durability can lead to a structural failure or malfunction that can cause an accident.
Moreover, more durable items require less frequent replacement, thus reducing the environmental impact of manufacturing new items, including pollution, greenhouse gas emissions, energy use, depletion of natural resources, and disposal and recycling needs.
Iterative and Stepwise Nature of Design
A flow chart showing some of the steps necessary to complete a mechanical design is shown in Figure. The logic loops shown by arrows indicate that the design process is fundamentally iterative in nature. In other words, there is a strong element of trial and error where an initial design is done and then analyzed, tested, and subjected to trial production.
Changes may be made at any stage of the process to satisfy requirements not previously considered or problems just discovered. Changes may in turn require further analysis or testing. All of this must be done while observing constraints on time and cost. Each step involves a synthesis process in which all of the various concerns and requirements are considered together.
Compromises between conflicting requirements are usually necessary, and continual effort is needed to maintain simplicity, practicability, and economy. For example, the cargo weight limit of an aircraft cannot be made too large without causing unacceptable limits on the weight of fuel that can be carried, and therefore also on flight distance.
Prior individual or organizational experience may have important influences on the design. Also, certain design codes and standards may be used as an aid, and sometimes they are required by law. These are generally developed and published by either professional societies or governmental units, and one of their main purposes is to assure safety and durability. An example is the Bridge Design Specifications published by the American Association of State Highway and Transportation Officials.
One difficult and sometimes tricky step in design is estimation of the applied loads (forces or combinations of forces). Even rough estimates are often difficult to make, especially for vibratory loads resulting from such sources as road roughness or air turbulence.
It is sometimes possible to use measurements from a similar item that is already in service, but this is clearly impossible if the item being designed is unique. Once at least rough estimates (or assumptions) are made of the loads, then stresses in components can be calculated. The initial design is often made on the basis of avoiding stresses that exceed the yield strength of the material. Then the design is checked by more refined analysis, and changes are made as necessary to avoid more subtle modes of material failure, such as fatigue, brittle fracture, and creep.
The geometric shape or size may be changed to lower the magnitude or alter the distribution of stresses and strains to avoid one of these problems, or the material may be changed to one more suitable to resist a particular failure mode. 1.3.2
In making design decisions that involve safety and durability, the concept of a safety factor is often used. The safety factor in stress is the ratio of the stress that causes failure to the stress expected to occur in the actual service of the component. That is,
X1 = stress causing failure / stress in service ……(Equation 1)
For example, if X1 = 2.0, the stress necessary to cause failure is twice as high as the highest stress expected in service.
Safety factors provide a degree of assurance that unexpected events in service do not cause failure. They also allow some latitude for the usual lack of complete input information for the design process and for the approximations and assumptions that are often necessary.
Safety factors must be larger where there are greater uncertainties or where the consequences of failure are severe. Values for safety factors in the range X1 = 1.5 to 3.0 are common. If the magnitude of the loading is well known, and if there are few uncertainties from other sources, values near the lower end of this range may be appropriate.
For example, in the allowable stress design method of the American Institute of Steel Construction, used for buildings and similar applications, safety factors for design against yielding under static loading are generally in the range 1.5 to 2.0, with 1.5 applying for bending stress in the most favorable situations. Elsewhere, safety factors even as low as 1.2 are sometimes used, but this should be contemplated only for situations where there is quite thorough engineering analysis and few uncertainties, and also where failure has economic consequences only.
For the basic requirement of avoiding excessive deformation due to yielding, the failure stress is the yield strength of the material, σo, and the service stress is the largest stress in the component, calculated for the conditions expected in actual service. For ductile materials, the service stress employed is simply the net section nominal stress, S.
However, the localized effects of stress raisers do need to be included in the service stress for brittle materials, and also for fatigue of even ductile materials.
Where several causes of failure are possible, it is necessary to calculate a safety factor for each cause, and the lowest of these is the final safety factor. For example, safety factors might be calculated not only for yielding, but also for fatigue or creep. If cracks or sharp flaws are possible, a safety factor for brittle fracture is needed as well.
Safety factors in stress are sometimes supplemented or replaced by safety factors in life. This safety factor is the ratio of the expected life to failure to the desired service life. Life is measured by time or by events such as the number of flights of an aircraft:
X2 = failure life / desired service life
For example, if a helicopter part is expected to fail after 10 years of service, and if it is to be replaced after 2 years, there is a safety factor of 5 on life. Safety factors in life are used where deformation or cracking progresses gradually with time, as for creep or fatigue.
As the life is generally quite sensitive to small changes in stress, values of this factor must be relatively large, typically in the range X2 = 5 to 20. The use of safety factors as in Equation 1 is termed allowable stress design.
An alternative is load factor design. In this case, the loads (forces, moments, torques, etc.) expected in service are multiplied by a load factor, Y. The analysis done with these multiplied loads corresponds to the failure condition, not to the service condition.
(load in service) × Y = load causing failure
The two approaches give generally similar results, depending on the details of how they are applied. In some cases, they are equivalent, so that X1 = Y. The load factor approach has the advantage that it can be easily expanded to allow different load factors to be employed for different sources of loading, reflecting different uncertainties in how well each is known.
Prototype and Component Testing
Even though mechanical behavior of materials considerations may be involved in the design process from its early stages, testing is still often necessary to verify safety and durability. This arises because of the assumptions and imperfect knowledge reflected in many engineering estimates of strength or life.
A prototype, or trial model, is often made and subjected to simulated service testing to demonstrate whether or not a machine or vehicle functions properly. For example, a prototype automobile is generally run on a test course that includes rough roads, bumps, quick turns, etc.
Loads may be measured during simulated service testing, and these are used to improve the initial design, as the early estimate of loads may have been quite uncertain. A prototype may also be subjected to simulated service testing until either a mechanical failure occurs, perhaps by fatigue, creep, wear, or corrosion, or the design is proven to be reliable. This is called durability testing and is commonly done for new models of automobiles, tractors, and other vehicles.
For very large items, and especially for one-of-a-kind items, it may be impractical or uneconomical to test a prototype of the entire item. A part of the item, that is, a component, may then be tested. For example, wings, tail sections, and fuselages of large aircraft are separately tested to destruction under repeated loads that cause fatigue cracking in a manner similar to actual service. Individual joints and members of offshore oil well structures are similarly tested.
Component testing may also be done as a prelude to testing of a full prototype. An example of this is the testing of a new design of an automobile axle prior to manufacture and the subsequent testing of the first prototype of the entire automobile. Various sources of loading and vibration in machines, vehicles, and structures can be simulated by the use of digital computers, as can the resulting deformation and fracture of the material.
This technology is now being used to reduce the need for prototype and component testing, thus accelerating the design process. However, computer simulations are only as good as the simplifying assumptions used in analysis, and the limitations on input data, which are always present. Thus, some physical testing will continue to be frequently needed, at least as a final check on the design process.
Design changes may also be made as a result of experience with a limited production run of a new product. Purchasers of the product may use it in a way not anticipated by the designer, resulting in failures that necessitate design changes. For example, early models of surgical implants, such as hip joints and pin supports for broken bones, experienced failure problems that led to changes in both geometry and material.
The design process often continues even after a product is established and widely distributed. Long-term usage may uncover additional problems that need to be corrected in new items. If the problem is severe—perhaps safety related—changes may be needed in items already in service. Recalls of automobiles are an example of this, and a portion of these involve problems of deformation or fracture.
Thanks for reading about “material selection in mechanical design”. For more information visit Wikipedia.