Bench Talk

(Source: Wang An Qi/Shutterstock.com)

We have come a long way from cave dwellings. For thousands of years, constructed homes and buildings had airflow, drainage, heating, cooling, and lighting. However, the way these systems are implemented has dramatically changed and is changing again.

Older passive and mechanical techniques for heating, ventilation, and air conditioning (HVAC) monitoring and control give way to digital solutions that incorporate newer state-of-the-art mixed-signal sensing solutions. Design engineers can take advantage of these to provide more accurate monitoring and control and allow for remote verification, diagnostics, and monitoring to take place through the Internet of Things (IoT).

Global events such as lockdowns and quarantines have recently refocused our attention to heating, cooling, and particularly airflow and air quality. Older buildings have shown that biological infestations can thrive in moist, dark places such as airflow ducts. Older mechanical control systems just can’t accurately monitor temperature, humidity, flow rate, pressure, and overall system integrity.

Although electromechanical and analog sensors have been available for quite some time, the ability to integrate them into a digitally controlled environment has been slow to become standardized. Design engineers have used a variety of techniques to monitor, control, and calibrate, with each engineer using their preferred method based on their constraints.

For example, a simple threshold detector for airflow can be a flapper switch that merely detects if the air is flowing (Figure 1 and Figure 2). It will trip even if the airflow is constricted by a filter that is slowly getting clogged. At some point, the filter will be so clogged that the flapper switch will not trip. By this time, the efficacy of the airflow system has been compromised for an extended period, and stress and strain have been placed on the airflow blowers.

Figure 1: Electromechanical airflow verification indicates when airflow is constricted, for example, a clogged filter. (Source: Mouser Electronics)

Figure 2: A digital sensor system will monitor the system’s performance even through varying environmental conditions and alert an operator before a system fault occurs. (Source: Mouser Electronics)

Calibrated temperature and humidity compensated digital airflow sensors, on the other hand, can monitor the curve of the degrading airflow. These digital airflow sensors can be programmed to alert a building supervisor before the airflow gets too constricted and will also protect blower motors from drawing too much current and getting hot, thereby extending the life of the motors and the system.

Take, for example, Amphenol ELV Analog and Digital Pressure Sensors with analog outputs, digital serial peripheral interfaces (SPI) or I²C serial outputs, in absolute and differential versions and feature a 0.25 percent accuracy with 14-bit resolution. These off-the-shelf calibrated devices span a range of 0.5psi to 150psi (3.45kPa-1034kPa) and can survive up to 245° C temperatures. Supply operating voltage options include 3V, 3.3V, and 5V, and a non-condensing humidity limit of 0 percent to 95 percent relative humidity (RH). A protective parylene-coated option can be ordered for harsh environment applications. These digital-pressure sensors can easily fit into your next design.

Running harnesses of wires to and from each sensor and actuator decreases reliability and increase complexity, the more connections, the lower the reliability. As a result, serial sensor systems have become the de-facto standard. The use of RS-485, RS-232, Controlled Area Network (CAN) Bus, I²C, and SPI means that sensor systems can be completely powered and controlled by just a few wires. Digital control systems can also detect whether a serial sensor is not responding. This permits control systems to incorporate safety mechanisms that protect the circuitry and the building.

Another benefit is that serial sensors can daisy chain. Design engineers do not have to add connectors and cables on their circuit boards for every sensor and do not need to run exceptionally long cables to the most distant sensors.

One such example of an initialized and ruggedized series is the Amphenol Advanced Sensors Telaire T9501 IP67 RH & T Sensor with Modbus (Figure 3). These fully calibrated and temperature-compensated sensor systems are water-resistant to IP67 specifications and can daisy-chain connect with Modbus and RS-485 signaling for simplified interconnect. These also feature 14-bit resolution and accuracies of +/- 2 percent RH and 0.5 percent for temperature.

Figure 3: This rugged and sealed relative humidity and temperature sensor easily integrates into a modern HVAC system using serial Modubus and RS485 signaling and is calibrated and accurate. (Source: Amphenol)

Instead, the sensor systems can act as a signal repeater that extends an individual communications transceiver (Figure 3). To eliminate IR drop along the common cable, AC can be passed along the serial sensor bus and rectified locally to provide power for the sensor subsystem.

A daisy-chained serial bus topology can interconnect long strings of sensor modules (Figure 4). Each acts as a repeater from the endpoint sensor—which loops back—to return a strong signal to the microcontroller. AC power is rectified locally by each sensor to overcome IR loss in the cabling.

Figure 4: A daisy-chained serial bus topology can interconnect long strings of sensor modules. Each acts as a repeater from the endpoint sensor—which loops back—to return a strong signal to the microcontroller. AC power is rectified locally by each sensor to overcome IR loss in the cabling. (Source: Mouser Electronics)

Note, with this approach, each sensor needs to be addressed individually. I²C sensors each have unique addresses, but RS-485, RS-232, and SPI sensors will need local intelligence. Fortunately, small, low-cost, and low-power microcontrollers can easily be incorporated into a specific sensor that can add a new level of self-test and diagnostic capabilities to the sensor systems.

Engineers understand that local access and control of a building’s temperature, humidity, lighting, and airflow need to be accommodated. Even multiple zones need to be individually controllable. Take, for instance, a hotel that might have dozens or even hundreds of rooms. Each room needs its own environmental controls because everyone has their own comfort requirements. Some elderly folks might want their hotel room to be set to a warm 27°C, but the average person would not want to stay in a hot hotel room.

Local access is a must, but what about global or remote access? A hotel room would most likely not need global access, but a local utility might want remote access. Peak-demand load constraints from a power company make it desirable to cap the total real-time electricity draw to a facility so as not to pay exorbitant prices.

In this case, certain non-critical systems can be controlled and sequenced to not exceed a threshold or real-time electricity draw. Although peak-demand load is not a universally implemented mandate today, it is coming, especially in a world becoming more sensitive to energy efficiency and greenhouse gas emissions.

Security, and the ability to guard against hacking and mischief, are imperative. Whether you’re working in a hotel room or a factory, you don’t want unwanted access.

Another factor with the digitization of technology is the ability to use smart artificial intelligence systems to learn characteristics and behaviors. In our hotel room example, smart sensors can detect when an occupant has left the room. Environmental controls can be shut down and save a lot of energy. When the occupant enters the hotel, radio-frequency identification (RFID) can detect and turn on the environmental controls.

Energy can also be saved by monitoring a compressor’s coolant pressure, for example. At some point, it will become saturated and continuous running will not make it colder. Turning it off early saves energy and has virtually no downside.

As it learns, AI can also determine what services can be momentarily turned off without affecting the overall performance. For example, a water heater can be turned off momentarily to allow a blower fan to turn on. While turning on, it can draw high amounts of surge current that will exceed peak-demand load limits, but once the blower is running smoothly, and the current draw is normalized, the water heater can turn on again. There is no noticeable interruption of service, but energy and money are saved. 

As society refocuses its attention on heating, cooling, airflow, and air quality, adding intelligence to our homes and buildings could make a difference in reducing our consumption and emissions, especially as the world population increases and resources decrease. We have seen analog sensors available for a while, but integrating them into digitally controlled environments comes with its challenges. Serial sensor systems provide benefits for engineers. It seems that this type of sensing makes sense, given the current digital environment and requirements that come with it.

After completing his studies in electrical engineering, Jon Gabay has worked with defense, commercial, industrial, consumer, energy, and medical companies as a design engineer, firmware coder, system designer, research scientist, and product developer. As an alternative energy researcher and inventor, he has been involved with automation technology since he founded and ran Dedicated Devices Corp. up until 2004. Since then, he has been doing research and development, writing articles, and developing technologies for next-generation engineers and students.