Jack, H., “Teaching PLCs With A Production Line”, ASEE North Central Section Conference Proceedings, Detroit, April, 1998.
Recently a new controls course (EGR 450: Manufacturing Controls) was developed at Grand Valley State University. This course includes a section on discrete logic control using Programmable Logic Controllers (PLCs). The basis for the labs is a four station line for manufacturing keytags. The stations include feeding, drilling, stamping and shearing. Each station has been designed so that it may be used stand-alone for initial programming and debugging. When each of the stations are functioning they are integrated into a complete line. Over four laboratory sessions, each team of students takes a turn at each station. By the completion of the labs the students have been exposed to a number of control problems, as well as the issue of integrating individual stations. This paper will be of value to instructors attempting to go beyond basic PLC programming to teach system integration principles.
We have recently begun to offer a course in manufacturing controls  to both the mechanical and manufacturing engineering students. This course examines the use of computers for discrete and continuous control. In particular this course uses Programmable Logic Controllers (PLCs) to teach discrete controls, while continuous controls is taught using data acquisition boards and Labview. The labs for the PLC portion of the course progressively introduce the students to PLCs. This progressive introduction is required because basic PLC programming is quite easy to teach, and can be explored with simple labs. But, the role of a PLC in a more complex system can be hard to teach without actual experience. We obtain this experience by using an operational manufacturing system.
The laboratory sequence begins by introducing the students to basic PLC programming using Allen-Bradley SLC-150 mini PLCs. These labs include wiring, logic, traffic lights, motor and encoder design. By the time the students have completed the introductory labs they are familiar with basic interfacing, timers, counters and internal memory. In the PLC laboratories that follow we attempt to teach them formal practices for good program design. This includes the use of flow charts, state diagrams, boolean equations and petri nets. To support the higher laboratories an integrated production line was designed and built by senior students as a capstone project (Terry Bridges, Mike Campeau, John Dejong, Eric Glass, Dana Silcox and Matt Werdon). The manufacturing line can be broken into individual stations, but then be reconnected allowing us to explore integration issues. The laboratory uses Allen Bradley PLC racks (PLC-2) with cards for TTL, DC and analog I/O. Teams of three students program four stations separately (on four PLC’s). Once the individual stations are working, the cards and programs are moved to a single PLC, then tested and debugged.
Each of the stations can be programmed and tested independently from the rest of the line. This is accomplished using a hardwired stand alone mode. Each station consists of a machine and a controller. When the stations are connected together they work as a single unit. This is a result of additional hardwired controls that combine emergency stops and active states for the hydraulic power unit. All of the processes, except the feeder, have hole detects to sense part positions. This allows asynchronous operation, but requires that after integration each station must be able to signal the feeder to stop.
The drill press has been made with a standard benchtop drill press that has been retrofitted with a geared down DC servomotor to feed the spindle. This motor is driven by an amplifier capable of running in multiple modes. In the simplest mode the input to the controller can be ordered to advance or retract and a hardwired PID controller will control the motion. In a more complex approach the user can supply their own analog voltages (possibly from a PID calculation) to advance or retract the drill at different rates. Limit switches are used to determine the end points of the motion.
The feeder station is basically two pinch rollers driven by a geared down stepper motor. In the simplest mode the program will advance the feeder by providing a pulse train until one of the other stations requests a stop. The stepper motor can also be used to accurately advance the material a fixed distance by providing a fixed number of pulses.
The embossing press uses hydraulic cylinders and valves to advance and retract. A scissor arrangement is used to get mechanical advantage, and limit switches allow the cylinder to be reversed or stopped as appropriate. The shearing station also uses a hydraulic cylinder to advance and retract the press, with limit switches for motion reversal and stopping. In the laboratory the students must wire the station controllers to the PLCs. This provides practice in practical electrical work, and reinforces understanding of the actual structure of the system.
Each individual station has programming issues. The hole detect mechanism must be treated as a one shot trigger to be properly used. When this is not done the station cycles repeatedly and does not allow the feeder to continue. The state of tool motion (advancing/retracting) must be tracked for all stations. When advancing the retract limit switch must be ignored. When retracting the advance limit switch must be ignored. After retracting this should trigger an end of the operation cycle state. The feeder station requires an oscillator to generate the pulse train for the stepper motor. And, the drill station may require programming of analog output values to control speed (advance is slower than retract).
The entire line requires effort to integrate. To do this the students must design individual programs so that they don’t overlap in internal memory locations or input/output locations. The programs must be able to communicate with the feeder program to indicate when they are idle (this usually means that the team doing the feeder station leads the integration process). While combining the code, the students are encouraged to add one piece of ladder logic at a time for modular testing and debugging.
In both modes the system has start-up procedures that are necessary. First the lexan has to have one hole drilled manually to start the asynchronous indexing (this hole can be drilled in manual mode). This is then fed to the drilling station where subsequent holes are drilled, based on the detection of the previous hole. Each station has a power-on stage first. In this stage the processes can be controlled manually. After this the stations must be placed in automatic mode before the PLC can assume control allowing the students to experience safety interlock systems. In some cases the stations have to be returned to an initial state before they can be put in automatic mode.
This paper outlined a simple production line for making keytags as part of a controls laboratory. The students program individual stations and attempt to integrate these into a more complete system. By doing this the students are exposed to a number of practical issues that will be encountered when doing an actual implementation. Over the four lab sessions required, the students get the line running fully at least once. Students have to interact with other teams to test and debug the system successfully. And, they get exposure to real processes and startup issues. In this environment mistakes could not hide, and all of the students work has to be correct.
This work was made possible by the foresight of Professor Paul Plotkowski who arranged for the necessary support. The majority of the design and construction work was done by now graduated students; Terry Bridges, Mike Campeau, John Dejong, Eric Glass, Dana Silcox and Matt Werdon.