1
speed, or its payload capacity. This is because some of the loading that creates stresses in the moving parts is due to the inertial forces predicted by Newton’s second law, F = ma. Since the accelerations of the moving parts in the machine are dictated by its kinematic design and by its running speed, adding mass to moving parts will increase the inertial loads on those same parts unless their kinematic accelerations are reduced by slowing its operation. Even though the added mass may increase the strength of the part, that benefit may be reduced or cancelled by the resultant increases in inertial forces.
Iteration
Thus, we face a dilemma at the initial stages of machine design. Generally, before reaching the stage of sizing the parts, the kinematic motions of the machine will have already been defined. External forces provided by the “outside world” on the machine are also often known. Note that in some cases, the external loads on the machine will be very difficult to predict—for example, the loads on a moving automobile. The designer cannot predict with accuracy what environmental loads the user will subject the machine to (potholes, hard cornering, etc.) In such cases, statistical analysis of empirical data gathered from actual testing can provide some information for design purposes.
What remain to be defined are the inertial forces that will be generated by the known kinematic accelerations acting on the as yet undefined masses of the moving parts. The dilemma can be resolved only by iteration, which means to repeat, or to return to a previous state. We must assume some trial configuration for each part, use the mass properties (mass, CG location, and mass moment of inertia) of that trial configuration in a dynamic force analysis to determine the forces, moments, and torques acting on the part, and then use the cross-sectional geometry of the trial design to calculate the resulting stresses. In general, accurately determining all the loads on a machine is the most difficult task in the design process. If the loads are known, the stresses can be calculated.
Most likely, on the first trial, we will find that our design fails because the materials cannot stand the levels of stress presented. We must then redesign the parts (iterate) by changing shapes, sizes, materials, manufacturing processes, or other factors in order to reach an acceptable design. It is generally not possible to achieve a successful result without making several iterations through this design process. Note also that a change to the mass of one part will also affect the forces applied to parts connected to it and thus require their redesign also. It is truly the design of interrelated parts.
1.2
A DESIGN PROCESS*
The process of design is essentially an exercise in applied creativity. Various “design processes” have been defined to help organize the attack upon the “unstructured prob
-
lem,” i.e., one for which the problem definition is vague and for which many possible solutions exist. Some of these design process definitions contain only a few steps and others a detailed list of 25 steps. One version of a design process is shown in Table 1-1, which lists ten steps. The initial step, Identification of Need, usually consists of an ill-defined and vague problem statement. The development of Background Research
*
information (step 2) is necessary to fully define and understand the problem, after which Adapted from Norton, Design
of Machinery, 3ed. McGraw-Hill,
it is possible to restate the Goal (step 3) in a more reasonable and realistic way than in New York, 2004, with the
publisher’s permission.
the original problem statement.
6
MACHINE DESIGN -
An Integrated Approach
1
Table 1-1
A Design Process
1
Identification of need
2
Background research
3
Goal statement
4
Task specifications
5
Synthesis
6
Analysis
7
Selection
8
Detailed design
9
Prototyping and testing
10
Production
Step 4 calls for the creation of a detailed set of Task Specifications which bound the problem and limit its scope. The Synthesis step (5) is one in which as many alternative design approaches as possible are sought, usually without regard (at this stage) for their value or quality. This is also sometimes called the Ideation and Invention step in which the largest possible number of creative solutions are generated.
In step 6, the possible solutions from the previous step are Analyzed and either accepted, rejected, or modified. The most promising solution is Selected at step 7. Once an acceptable design is selected, the Detailed Design (step 8) can be done, in which all the loose ends are tied up, complete engineering drawings made, vendors identified, manufacturing specifications defined, etc. The actual construction of the working design is first done as a Prototype in step 9 and finally in quantity in Production at step 10. A more complete discussion of this design process can be found in reference 2, and a number of references on the topics of creativity and design are provided in the bibliography at the end of this chapter.
The above description may give an erroneous impression that this process can be accomplished in a linear fashion as listed. On the contrary, iteration is required within the entire process, moving from any step back to any previous step, in all possible combinations, and doing this repeatedly. The best ideas generated at step 5 will invariably be discovered to be flawed when later analyzed. Thus a return to at least the Ideation step will be necessary in order to generate more solutions. Perhaps a return to the Background Research phase may be necessary to gather more information. The Task Specifications may need to be revised if it turns out that they were unrealistic. In other words, anything is “fair game” in the design process, including a redefinition of the problem, if necessary. One cannot design in a linear fashion. It’s three steps forward and two (or more) back, until you finally emerge with a working solution.
Theoretically, we could continue this iteration on a given design problem forever, continually creating small improvements. Inevitably, the incremental gains in function or reductions in cost will tend toward zero with time. At some point, we must declare the design “good enough” and ship it. Often someone else (most likely, the boss) will snatch it from our grasp and ship it over our protests that it isn’t yet “perfect.” Machines that have been around a long time and that have been improved by many designers reach a level of “perfection” that makes them difficult to improve upon. One
Chapter 1