Modern strategies for cost-effective manufacturing mandate that more information be collected from, and sent to, equipment throughout the plant. Often, a data acquisition and control system sends a myriad of signals from remotely located production equipment (frequently from different manufacturers) to a central control location. When these data collection points are widely dispersed throughout a plant, a dedicated wire is usually run to each point. This means miles of expensive multiconductor cabling and conduit, and substantial costs to install them.

Rather than installing many wires to collect sensor data and deliver control instructions, it may be more economical to digitize the signals in the field (on the plant floor) using a multiplexing system, Figure 1.

These wiring systems reduce the cost for sending multiple monitoring and control signals either long distances or through difficult environments. How? They replace dozens of wires with a twowire data link.

Generally used in a computer-based control system, a multiplexer is installed at, or in, the controller, where it converts process signals transmitted from various locations to a serial format that can be used by the controller.

Field multiplexers convert the process signals to a digital format before long-distance transmission to and from field locations. This allows the user to take advantage of economical and noise-immune digital transmission.

In a typical plant, analog and discrete (on/off) devices (sensors and transmitters located at motors, valves, and pumps) collect process data. The analog and discrete signals from these devices are sent to an analog-to-digital (A/D) converter, then a multiplexer chip, which converts data into a digital protocol (communication standard) such as RS- 485 or RS-232, Figure 2. This protocol is essentially a concentrated version of the dozens of analog and discrete process signals. At the opposite end of the data link, a multiplexer converts the digital data back to its original analog/discrete state, or delivers a digital format directly to a controller, like a PC or PLC. The control instructions are then sent back over the same set of wires. Hard wires don’t have to be run to each device.

Some types of multiplexers can be used with programmable logic controllers (PLCs), but this type of application is best suited for adding remote I/O to existing systems because cost and setup effort may be less than for additional PLC I/O. Also, a multiplexer with electrically isolated inputs interfacing with a PLC may be less costly than adding isolated PLC inputs. The multiplexer must be able to output a protocol the PLC can understand (such as Modbus) or convert serial data back to analog form for input to the PLC.

Inputs and outputs

To simplify system design and implementation, it is usually best to choose a field multiplexer that is compatible with an array of transmitter signals and field control devices.

A field multiplexer should be able to accommodate all common analog signals: 4-20 mA, 1-5 V, and 0-10 V. This enables it to accept inputs from any temperature, pressure, level, flow, or power transmitter. Analog signals are also useful for proportional control of valves, pumps, dampers, and louvers.

Discrete signals (contact closure, TTL) are used for tripping alarms to indicate unwanted process conditions. They provide on/off control of motors and valves.

Some multiplexers process only analog or discrete signals, whereas others can process both. When both analog and discrete devices are present in the system, it may be advantageous to use a field multiplexer that accepts both types of input.

A typical 16-channel multiplexer processes data at a scan rate of up to 0.5 sec per 16 channels. Thus, each monitoring point is scanned up to 120 times per minute. The scan rate generally depends on the number of channels per system, the baud rate, and the transmission distance.

Continue on Page 2