How to Develop Multiple-actuator Pneumatic Circuits Using the Cascade Method?

Most of the practical pneumatic systems involve the use of multiple actuators (cylinders, semi-rotary actuators, etc) which when operating in specified sequences carry out the desired control tasks. Double-pilot directional control (DC) valves are used as final control elements to control the forward and return strokes of the actuators. Sensors are used for confirming the actuator positions and providing signals to the final control elements. However, it is important to remember that the double-pilot DC valves are susceptible to the problem of ‘signal conflicts’. You may remember that the signal conflict is due to the appearance of pilot signals simultaneously on both sides of a double-pilot DC valve, due to which the valve will not be able to move its position in response to a control signal.

---

---

Control Task

A pneumatic circuit has to be developed for realizing the following control task using two cylinders A (1.0) and B (2.0) as shown in the schematic diagram. The cylinder A is to extend and bring a job under the stamping cylinder B. The cylinder B is then to extend and stamp the job. The cylinder A can return back only after the cylinder B has retracted fully. The sequential control task can also be expressed in a displacement-step diagram as shown.

Displacement-step Diagram

The displacement-step diagram for the control task is given below. The displacements of cylinders are plotted in accordance with the required sequence of the cylinder actions in equal steps. For example, the displacement of any cylinder from the retracted position (0) to the extended position (1) is shown in the diagram with a line from 0 to 1. Similarly, other displacements can also be shown.

Notational Form of the Control Task

In the notational form, the ‘+’ sign is appended to the cylinder designation (A, B, etc) to represent the forward stroke and the ‘-’ sign is appended to it to represent the return stroke. For the control task specified above, the sequence of the cylinder actions along with the details of the sensor outputs can be represented by:

As can be seen, the A+ action generates ‘a1’ signal and is used for the B+ action, the B+ action generates ‘b1’ signal and is used for the B- action, and so on. Usually, a “Start” signal is also required along with the a0 signal (the last signal of the control sequence) for obtaining the A+ action (the first cylinder action of the control sequence).

Adding the Control Circuit to the Power Circuit

Incorporate all the sensors and the “Start” pushbutton as per the required control sequence, as shown in the figure. Represent the sensors a0 and b0 initially in the actuated state.

Check for Signal Conflict

Next step is to analyze the circuit for the presence of signal conflicts. When the “Start” pushbutton is pressed, a signal appears at port 14 of the valve 1.1 through the sensor a0. It can be seen that a signal is also present at port 12 of the valve 1.1, resulting in a signal conflict. As result, the valve 1.1 is unable to switch over and the circuit will not work at all. You may check for other signal conflicts if any.

Elimination of signal conflicts

Various methods are devised to eliminate the problem of signal conflicts. Most popular methods are based on controlling the air supply to different sections of the control circuit. Any of the following methods may be used for the purpose of avoiding the signal conflicts: (1) Cascade method and (2) Shift register. This article explains only the cascade method of eliminating the signal conflict.

Cascade method

In this method, the sequence of operations of cylinders, that is, A+B+B-A- are divided into appropriate groups in such a way that there is no possibility of a signal conflict. That is, in case A+ and A- happen to be in the same group, there is a possibility of signals appearing simultaneously at both ends of the final control element (valve 1.1) controlling the cylinder A. Similarly, there is a possibility of signals appearing simultaneously at both ends of the final control element (valve 2.1) controlling the cylinder B, if the cylinder actions B+ and B- happen to be in the same group. Hence the sequence of operations is divided in such a way that the A+ and A- actions fall into different groups (G1 and G2), and the B+ and B- actions fall into different groups (G1 and G2), as demonstrated.

It should be remembered that the desired sequence should be maintained. In this method, every attempt should be made to keep the number of groups to a minimum so as to keep the number of valves to a minimum.

Next, divide the power supply for the control circuit into an equal number of groups in such a way that at any given point in time, only one group will have the supply with all other group(s) connected to the exhaust. By an appropriate interconnection of the 5/2-DC valves, the power supply can be divided into 2 groups, 3 groups, 4 groups etc. The different positions of the standard 2-group circuit are illustrated in the figures. It can be seen that initially, the supply is in the last group, G2 (see figure b). When the control signals are applied to inputs e1 and e2 in that sequence the supply changes to groups G1 and G2 respectively across the cascade. The group changing circuits for the three groups, four groups etc can also be drawn in a similar manner.

Circuit Design Using the Cascade Method

Add the group changing circuit for two groups just below the power circuit. The group changing circuit ensures that only one group will have the supply at any point in time with the other group connected to the exhaust. Add the control valves and sensors as specified in the notational form given above. Initially, the sensors a0 and b0 are shown in the actuated position. It may be observed that the port 14 (for A+ action) and the port 12 (for A- action) of the valve 1.1 are always connected to different supply groups. Similarly, the port 14 (for B+ action) and the port 12 (for B- action) of the valve 2.1 are always connected to different supply groups. This inhibits the signals from appearing simultaneously on both sides of each of the final control elements and hence avoids the possibility of signal conflicts. Finally, designate the valves and cylinders.

Conclusion

The cascade system provides a straightforward method of designing sequential circuits. It will always give a workable circuit and only rarely will it be possible to suggest any improvements.

---

Authored by JOJI Parambath, Founder/Director, Fluidsys Training Centre, Bangalore

---

---