1、 附录 A：英文资料 Selecting a PLC for the application Questions answered 1 how do we understand and estimate the requirement for a PLC? 2 how do we select the IO hardware? 3 how are IO circuits configured? 4 how do we size the processor and memory for our application? 5 how do we select a supplier? The cho
2、ice of a PLC for a particular application can be bewildering. The range of suppliers is vast, many offering a number alternative product ranges, with any number of modules to perform choice must meet the job and customers requirements, provide extra capacity to enable future modifications and provid
3、e an acceptable cost solution. We have to make choices balancing the cost of extra, or more expensive hardware against the time required to program algorithms that allow us to use cheaper hardware to meet the system requirements, each case has to be considered on its merits. Beware of the common tra
4、p of underestimating the time taken to write such code! 9.1 estimating requirements The starting point in determining any solution must be to understand what is to be achieved. In an ideal world our customer (even if we are building a system for ourselves) will have produced a detailed specification
5、 of the requirements. If this is not the case, we must start by preparing one. 9.1.1 System definition In chapter 7 we discussed program design, breaking down the task into a number of simple understandable elements, each of which can be easily described. The same technique of functional decompositi
6、on is equally applicable to defining the whole system, both hardware and software, as it is defining the program alone. The most common mistake is to attempt to handle the entire system as one unit. When such an approach is made we will immediately select solutions for the parts of the system we kno
7、w are going to be a problem, or the parts we immediately know how to solve. This approach diverts the design and equipment selection away from what is required to solve the real problems, and leaves us whit a solution that may be far from ideal. A worked example can be found in the appendix, which s
8、how a typical decomposition of fairly complex application and sample IO diagrams 9.2 choosing the correct IO hardware With an understanding of the entire system we can start to estimate the PLC requirements. For each module the inputs and outputs can be categorized for type and speed of operation. S
9、ection 6.4 described the various types of input and output modules but here we will consider the selection criteria. By knowing the number of any type of IO lines we need and the number of lines available on a given module, the final shopping list of modules and the size of the PLC system are determ
10、ined. In addition, burring in at least 20 percent extra capacity to allow for future modifications or to solve problem identified during commissioning. 9.2.1 Simple IO timing considerations For every element, we need to determine how fast the subsystem of input program and output must react to chang
11、ing input condition. The speed of operation will be the sum of the input hardware delays plus the PLC scan time plus any output hardware delays. In the vast majority of cases a time delay of 100ms or greater is not significant. Typical instances in which this may not be the case are pulse counters,
12、or where a movement has to be stopped in mid-stroke. To determine the required response speed, we need to consider each of our defined modules with its inputs and outputs. The effect of control decisions being taken at various rates can then be considered and the slowest rates determined. For exampl
13、e, figure 9.1 shows a simple tank level control application. if the flow rate is known to be 10 liter/s and we want to maintain the volume of liquid in the tank to 1 liter we need to read the level input and make a decision to set or clear the flow valve output , at the very slowest , every 0.1s . T
14、his can be determined by calculating the time it takes for our minimum control quantity to flow into the tank, i.e. 1 liter/s . Fig.9.1 Tank level control In the second example (figure 9.2) we need to stop the cylinder mid-stroke to an accuracy of 0.2mm .we know that is maximum speed is 100mm/s, so
15、to achieve this we would need to be able to make a decision to set or clear the control valve every 2ms, i.e.the time taken for it to move 0.2mm/100mm/s. Fig.9.2 Controlling pneumatic cylinder To achieve this we would require scan speed that would be difficult to guarantee except in the smallest pro
16、grams using a fast PLC. An interrupting input is indicated .the switching speed involved will probably also cause a problem because of delays in the electronic and pneumatic hardware .This will require us to stop the movement a known distance before the target position .Figure 9.3 shows how these de
17、lays are introduced. This, of course, assumes that the cylinder is moving at constant speed as it trips the switch and that all the electronic and pneumatic delays are constant. If d.c. inputs and outputs are used, this is a reasonable assumption. When a.c. I/O lines are in use there is always an additional 10ms uncertainty, as described below. The more normal action would be to use a second switch to slow
