A SCADA (Supervisory Control and Data Acquisition) system collects real‑time data from sensors and actuators, transmits that data through field communication networks, and provides supervisory control through a human‑machine interface (HMI). Typical elements include:

• Remote terminal units (RTUs) or PLCs that gather signals from field devices, perform limited control, and relay data to the SCADA master.
• Communication networks such as fieldbus, Modbus, OPC‑UA or Ethernet that carry data between devices, controllers and the SCADA host.
• Human–Machine Interface (HMI) software that presents alarms, trends and dashboards to operators and allows manual control.
• Database and historian servers that store time‑stamped data for analysis and optimization.

Below are the types of signals handled by SCADA, common communication protocols, and the differences between programmable logic controllers (PLC) and distributed control systems (DCS). Example problems and exercises provide practice with transmitter scaling and network concepts.

Digital Inputs

Digital inputs are binary signals (ON or OFF) used to indicate the status of a device. A digital input module senses the presence or absence of voltage or current and converts it into a logical 1 or 0 for the controller. Typical digital input devices include:

• Push buttons and selector switches that start, stop or select operating modes.
• Limit switches and mechanical interlocks that sense position, doors or safety guards.
• Proximity and photoelectric sensors that detect the presence of objects without contact.
• Flow and level switches that indicate high/low limits.

Most industrial digital inputs operate at 24 VDC or 120 VAC. Two wiring conventions are common:

• Sinking input: the field device sources current and the input module connects to ground.
• Sourcing input: the field device sinks current to ground and the input module sources the supply voltage.

Diagnostic LEDs on input modules aid troubleshooting by showing which channels are active. Because digital signals are either on or off, they are relatively immune to noise and simple to process.

Digital Outputs

A digital output module is essentially a set of controllable switches used by the controller to energize actuators. When the controller writes a logic 1 to an output channel, the module completes a circuit to turn on a field device. Common digital outputs drive:

• Relays and contactors for motors, pumps, heaters and lights.
• Solenoid valves for pneumatic or hydraulic actuators.
• Alarms and indicator lamps to signal abnormal conditions.

Three main technologies are used:

• Relay outputs provide galvanic isolation and can switch high AC or DC loads; they have slower response and mechanical wear.
• Triac outputs switch AC loads without moving parts and allow faster operation, but only for AC circuits.
• Transistor outputs switch DC loads quickly and are well suited for small solenoids or interposing relays.

Like inputs, digital outputs may be sinking (negative potential to load) or sourcing (positive potential to load). Choosing the proper type ensures correct polarity and protects the devices.

Analog Inputs

Analog inputs measure continuously variable signals such as voltage, current, temperature, pressure or flow. An analog input module converts the analog value into a digital word using an analog‑to‑digital converter (ADC) so that the controller can process it. Typical signal ranges include:

• 0–10 V or 0–5 V for many transducers and microcontrollers.
• 4–20 mA for industrial sensors because current is less susceptible to voltage drop and noise.
• ±10 V or ±20 mA for bidirectional measurements.

Examples of analog input devices are thermocouples, RTDs, load cells, ultrasonic level sensors and differential pressure transmitters. Each channel of an input module often includes cold‑junction compensation, filtering and isolation to ensure accurate measurements.

Analog Outputs

Analog output modules perform the inverse function: they convert a digital word from the controller into a continuously variable electrical signal using a digital‑to‑analog converter (DAC). Analog outputs are used to drive:

• Control valves (through an I/P converter) to modulate flow or pressure.
• Variable speed drives that adjust motor speed.
• Heaters and pumps requiring proportional control.

Standard output ranges are 0–10 V, 4–20 mA or ±10 V. High‑resolution outputs (12‑bit or 16‑bit) provide fine control in demanding applications.

Field Communication Networks

Modern control systems use networked I/O instead of point‑to‑point wiring. Fieldbus and industrial Ethernet protocols allow multiple devices to share a single cable and provide diagnostic information. Common examples include:

• Foundation Fieldbus – A digital two‑wire, intrinsically safe bus used in process industries. Field devices contain microprocessors that execute control algorithms locally and report data to a host. It allows device‑level diagnostics and multiple topologies.
• Profibus/Profinet – Developed by Siemens, Profibus is a fieldbus for discrete and process automation, while Profinet adapts the protocol to Ethernet for high‑speed communication. Devices can be connected in line, star or ring topologies.
• DeviceNet and CANopen – Based on the Controller Area Network (CAN), these networks connect sensors, actuators and controllers over a two‑wire bus. They are widely used in packaging, automotive and small machinery.
• INTERBUS, AS‑Interface and HART – Specialized networks optimized for high‑speed machine control (INTERBUS), simple binary devices (AS‑Interface) or superimposed digital signals on analog loops (HART).

Using fieldbus reduces wiring, supports hot‑swap of devices and enables diagnostic data to be reported to maintenance systems.

Modbus

Modbus is one of the oldest and most widely adopted industrial communication protocols. Initially developed for programmable logic controllers, it remains popular because it is open, simple and lightweight. Key features include:

• Client/server (interrogator/responder) architecture – one primary client polls many device registers on networked servers for data, simplifying device implementation.
• Register‑based data exchange – devices store data in holding registers, input registers, coils and discrete inputs that the master reads or writes.
• Multiple physical layers – Modbus RTU is a compact binary protocol over RS‑485/RS‑232; Modbus ASCII encodes messages in ASCII characters; Modbus TCP encapsulates Modbus frames in TCP/IP packets over Ethernet.
• Error checking – RTU and ASCII include cyclic redundancy checks (CRC or LRC) to detect transmission errors.

Because of its simplicity and openness, Modbus is used in SCADA, building automation, energy monitoring and IoT gateways. Limitations include a lack of built‑in security and slower throughput compared to modern Ethernet‑based protocols.

OPC and OPC UA

OLE for Process Control (OPC) is a set of standards developed in the 1990s to allow Windows‑based applications to access process data from controllers, historians and HMI software. The original specification (OPC DA) used Microsoft COM/DCOM technology and had issues with firewall and authentication settings. To address these limitations, the OPC Unified Architecture (OPC UA) was introduced. OPC UA provides:

• Platform independence – it runs on Windows, Linux, embedded systems and mobile devices.
• Service‑oriented architecture – clients and servers can browse data, invoke methods and subscribe to events.
• Information modelling – structured data and metadata allow complex relationships between variables and objects to be represented.
• Built‑in security – encryption, authentication and authorization protect data integrity and confidentiality.

OPC UA supports both client/server (request‑response) and publish/subscribe models. Client/server is suitable for supervisory control, whereas PubSub allows high‑frequency or one‑to‑many data distribution in Industrial IoT applications.

SQL and Historian Databases

Structured Query Language (SQL) databases and specialized historian systems store time‑series data for analysis, reporting and optimization. While SQL is not a real‑time communication protocol, many SCADA systems write process data to SQL databases so that enterprise applications can query historical trends. Historians provide high‑speed storage, compression and retrieval of millions of data points and often include Web APIs for analytics and machine learning.

PLC versus DCS – What is the Difference?

Programmable Logic Controllers (PLC) were originally designed to replace relay panels in discrete manufacturing. A PLC is typically a central processor with local I/O modules and executes a simple cyclic scan (read inputs, run logic, write outputs). PLCs are fast, deterministic, and rugged. They excel at machine control, packaging, material handling and safety interlocking. Programming languages include ladder diagram, structured text and function block diagram.

Distributed Control Systems (DCS) are designed for large continuous or batch processes where thousands of measurements must be coordinated. A DCS consists of multiple controllers distributed throughout the plant, each with its own I/O. The controllers communicate over redundant networks to a central operator station. Key features include high availability, advanced regulatory control, alarm management, recipe management and integrated engineering tools. Programming often uses higher‑level languages such as continuous function charts and sequential function charts.

The main differences are:

Convergence: Modern PLCs have gained features once unique to DCS, such as redundant processors, networked I/O, integrated HMI and advanced function blocks. At the same time, DCS vendors have adopted high‑speed fieldbus, modular hardware and open protocols to reduce cost and increase flexibility. The distinction between PLC and DCS is blurring, and many systems are now marketed as “hybrid” or “programmable automation controllers (PAC)” combining the strengths of both.

Oil & Gas, Energy, Chemical Manufacturing

Pharmaceutical, Pulp & Paper, and Food

Laboratory and Prototype Systems

In research laboratories and educational settings, flexibility and rapid development are more important than ruggedness. The following platforms support data acquisition and control using Python or graphical programming:

Analog Scaling and Communication Examples

The following problems illustrate how to convert between physical quantities and transmitter signals, scale data for digital networks and mitigate signal noise. Work through the derivations and then check the solutions.

Problem 1 – Humidity Transmitter

A relative humidity sensor outputs 1 V at 0 %RH and 5 V at 100 %RH.

• Derive a linear expression for humidity as a function of the voltage output.
• Determine the gain, zero and span of the transmitter.
• What voltage is expected at 40 %RH?
• What voltage would the transmitter produce at 120 %RH if the signal saturates at 5 V?

Problem 2 – Flow Transmitter

A flow meter converts 0–300 L/min into a 4–20 mA signal.

• Derive the relationship between flow rate and output current.
• Find the gain, zero and span of the transmitter.
• What current is expected at 150 L/min?
• What current would the transmitter output at –50 L/min (reverse flow)?

Problem 3 – RTD Temperature Transducer

An RTD‑based transmitter produces 0.5 V at 0 °C and 2.5 V at 200 °C.

• Derive the linear voltage–temperature relationship.
• What is the gain, zero and span?
• What voltage would you expect at 50 °C?
• When the sensor cable runs past a variable‑frequency drive, the measurement exhibits random spikes. Suggest a possible cause and remedy.

Problem 4 – Ultrasonic Level Sensor

An ultrasonic level transmitter outputs 0 V at 5 m and 10 V at 0 m.

• Write the level–voltage relationship and find the gain and offset.
• What voltage corresponds to a 1.5 m liquid level?
• The signal is transmitted 20 m through an unshielded cable. Each time a large motor starts, the level reading fluctuates. What might be the cause and how can you mitigate it?

Problem 5 – Concentration Scaling Over Modbus

A gas analyzer reports concentration in the range 0.00–100.00 % as a 32‑bit floating‑point number. However, the existing Modbus register map allocates only one 16‑bit unsigned register per gas species.

• How can a 0.00–100.00 % value be transmitted using a 16‑bit unsigned integer without losing the two decimal places?
• What scaling factor is applied before transmission and after reception?
• Using this scaling, what integer value is transmitted for 56.78 % concentration?

Quiz: Communication and Control Fundamentals

Test your understanding by selecting the correct answer. Click "Select" to reveal the answer.

1. Which of the following is typically a digital input device?

---

2. Which protocol uses a simple client/server polling mechanism and stores data in coils and registers?

---

3. Compared with a PLC, a DCS is most accurately described as:

---

Exercise 1 (20 min): Voltage Divider for Signal Conditioning

A research group wants to interface a 0–10 V sensor with a microcontroller that accepts a maximum input of 3.3 V. Design a resistor divider to scale the voltage and determine appropriate resistor values.

Tasks:

1. Determine the required ratio of the divider to convert 10 V to 3.3 V.
2. Choose resistor values that meet this ratio while using standard resistor sizes (e.g., kΩ range).
3. Explain how input impedance and loading can affect measurement accuracy and how to mitigate it.

---

Exercise 2 (20 min): 12‑bit ADC Scaling

A pressure transducer outputs 0–10 V proportionally to 0–1000 kPa. The sensor is connected to a microcontroller with a 12‑bit ADC (0–4095 counts) referenced to 0–10 V.

• Derive an expression to convert the ADC reading `N`, in counts, to pressure `P` in kPa.
• What pressure corresponds to an ADC reading of 2048 counts?
• What is the quantization resolution of the measurement in kPa? (Hint: one count represents what change in voltage? Convert to kPa.)
• Suggest one method to improve the pressure resolution without changing the ADC hardware.

---

Exercise 3 (20 min): Modbus Polling Cycle

A Modbus client communicates with eight server devices on an RS‑485 network at 19 200 baud. Each poll reads four 16‑bit holding registers from a server. Assume the request frame is 8 bytes (address + function + start register + quantity + CRC) and the response frame is 15 bytes (address + function + byte count + 8 data bytes + CRC). Each byte includes 1 start bit, 8 data bits and 1 stop bit.

• Calculate the time in milliseconds to transmit one complete request and response frame to a single device.
• What is the total polling cycle time for all eight devices if the client polls them sequentially with no idle time between polls?
• If the system requires each device to be updated at least once per second, does this network meet the requirement? Explain.