Reliable wireless communication for factory automation

Although wireless technology is now commonplace, reliable wireless is still something of a Holy Grail for industrial automation applications.

The benefits of using wireless for industrial automation applications are certainly alluring enough in terms of reduced cost, higher mobility and greater scalability. However, the downside of using wireless in industrial applications is that it is much harder to meet the reliability requirements demanded by industrial automation. Whereas a brief, several-second cessation of service for a routine office application might be viewed as a mere nuisance, for critical automation applications the same cessation could cause an unacceptable interruption in a factory’s operation.

Modern factories use many types of equipment, including robot arms on the production line, a variety of sensors and actuators for automation, and even unmanned AGVs to increase efficiency. Since much of the equipment is made from metal that protects complex electronics, it’s no wonder that this kind of environment creates a number of obstacles to implementing seamless wireless communications.

In particular, operators need to learn how to deal with multipath effects, which can generate interference that causes information distortion or packet loss. In addition, electrical disturbances such as ground loops need to be handled properly to avoid interruptions in wireless transmissions. And finally, when mobility is required for greater efficiency, providing a seamless roaming mechanism is essential to ensure that the moving objects can reliably receive and transmit data.

Multipath effects

Ideally, the radio waves emitted by wireless transmitters will travel unimpeded to the intended receiver, but in industrial automation sites various undesirable effects can occur, including reflections off large objects, scattering due to small objects, diffraction from sharp objects, shadowing from solid objects and Doppler effects from moving objects. When confronted with so many obstacles, the phenomenon of ‘multipath fading’ can occur.

Figure 1: Multipath fading occurs when the radio signal is split into multiple signals, each of which is affected by the environment in a different way. (For a large image, click here).

Multipath fading occurs when the radio signal is split into multiple signals, each of which is affected by the environment in a different way and each of which could arrive at the intended receiver at slightly different times. The resulting signal could be deteriorated to such a degree that whatever information was present at transmission is no longer decipherable at reception. Multipath fading occurs in factory environments that are subject to multipath effects, and when the radio transmitters and receivers are required to move around. Different versions of transmitted signals will arrive at the receiver out of phase with each other, resulting in a degradation of the radio signal.

Another type of interference caused by multipath effects is ‘inter-symbol interference’ (ISI), which occurs when different versions of the same signal travel along paths with different lengths, and consequently arrive at the receiver at different times. Since the frequencies of the various signals will be slightly changed, the combined signal will be distorted.

How to handle multipath effects

There are two common ways to eliminate interference caused by multipath effects: OFDM and MIMO.

OFDM

OFDM (orthogonal frequency division multiplexing) works by dividing one digital signal into different parts over lower data rate signal carriers. OFDM uses carriers orthogonal to each other to avoid interference. When the data is spread across different signal carriers the data rate handled by each carrier is reduced. The advantage of the lower data rate is that it makes the interference from reflections less serious.

Distributing data across a large number of carriers in an OFDM signal is paramount to playing the odds. That is, although interference from multipath effects will affect the signal transmitted by some of the carriers, it will only affect a small number of them. In addition, error correction code can be transmitted via a different carrier than the data, which allows data corrupted during transmission to be reconstructed at the receiver end.

OFDM plays an important role in today’s wireless data transmission technology. In fact, it is used with 802.11n Wi-Fi, LTE (Long-Term Evolution for 3G cellular communications), LTE Advanced (4G), WiMAX and others.

MIMO

MIMO, which stands for multiple-input multiple-output, uses multiple transmitters and receivers with multiple antennas. When multiple transmitters and receivers are used, simultaneous data streams can be sent, thus increasing the data rate. In addition, multiple transmitters and receivers allow greater coverage and longer distances between devices. The IEEE 802.11n standard uses MIMO to increase wireless data rates to 300 Mbps with two spatial streams and beyond; MIMO technology is also used in LTE and other wireless standards.

In a typical communication system using antenna diversity or MIMO, four methods are often used to deal with multipath effects in environments that exhibit multipath fading. The methods include: time diversity, frequency diversity, spatial diversity and path diversity.

IEEE 802.11n technology uses the spatial diversity method to counteract the multipath issue. If a specific antenna in the group suffers a severe transmission disruption, the spatial diversity method singles out the antenna signal that has suffered the least amount of transmission disruption, relatively speaking, and restores its function. The principle of this method lies in the belief that when multiple antennas are functioning independently while located in the same environment, the channels used by each antenna can be understood as uncorrelated with each other. Each channel may be subjected to multiple path or co-channel interferences, but the possibility of them suffering from the same type of variation at the same time is very unlikely.

Electrical disturbances

Three common types of electrical disturbances common in industrial environments are ground loops, interference from DC motors and ESD.

Ground loops

Ground loops, which are caused by unintended variations in the electric potential at different points in the application’s environment, can have a negative effect on communication signals and damage equipment. The effects can be particularly noticeable for integrated systems, such as AGV equipment, in which several different devices are attached to the vehicle. In this case, ground loops could interrupt the vehicle’s operation.

Figure 2: Ground loops can be particularly problematic in integrated systems. (For a large image, click here.)

Interference from motors

The electromagnets and electrical currents used to rotate motors can cause discontinuous currents and electromagnetic interference (EMI) at the start and transition stages, affecting the quality of the power supply, the surrounding electromagnetic environment and the operation of peripheral appliances.

ESD

ESD (electrostatic discharge), which results from the sudden transfer of static electricity between two objects with different electrical potentials, is also important. Factory workers wearing rubber boots and gloves can easily accumulate high levels of static electricity, and friction between objects rubbing against each other can also cause ESD. Physical contact with wireless devices can discharge several kilovolts of static electricity and permanently damage internal circuitry.

Figure 3: Electromagnetic interference and static electricity.

Dealing with electrical disturbances

To ensure that a wireless connection remains active and works properly, wireless devices must be tough enough to withstand all electrical disturbances in factory environments. The effects of these factors can be mitigated by using products designed with galvanic isolation.

Galvanic isolation

Galvanic isolation involves physically separating the electrical current in one part of a device from the electrical current in another part of the device to prevent unintentional interaction between the two parts. Although a variety of methods can be employed to implement galvanic isolation, the goal of all of the methods is to enforce the absence of DC paths between the parts of the system that need to be isolated from each other. In fact, most isolation methods eliminate all DC paths below 100 MΩ. Galvanic isolation provides three major benefits: circuit protection, noise reduction and rejection of common-mode voltage.

Figure 4: Wireless systems benefit from galvanic isolation between input, output and power supply.

Circuit protection

In order to ensure the reliability of your communications system, you should consider using a higher level of EMS protection in your device to guard against electromagnetic disturbances such as ESD, surge and EFT. Industrial wireless manufacturers usually use circuits and component methods to protect interfaces or dissipate energy, but using electrical methods is not sufficient, since the protection components or circuits could be destroyed by the effects of frequent abnormal electromagnetic disturbances. Galvanic isolation uses a physical gap to stop the flow of excess electrical energy. The gap ensures that the input path does not touch or connect to the output path and provides the most reliable method to protect against ESD, surge and EFT.

Noise reduction

The interference caused by a ground loop or DC motor can be resolved by simply cutting off the ground loop or by blocking the current and noise. Other methods include bundling the cable to reduce the loop or cutting the signal shielding wire or linking resistance, but neither solution resolves the issue completely. The first method cannot indefinitely prevent the occurrence of ground loops and noise, and the second method requires special attention be paid to whether the ground path has been removed, which could result in a floating ground and generate even more noise. We can conclude, therefore, that the best approach is to use galvanic isolation technology. Galvanic isolation technology can effectively remove the grounding line and the signal to prevent the occurrence of ground loops and significantly reduce or completely block off noise while transmitting data.

Rejection of common-mode voltage

For certain low-voltage signals, a specific level of voltage is required for the signal to be detected, and communication can only take place when the voltages are amplified to a certain level. For instance, the common-mode voltage range of an RS485 transceiver is -7 to +12 V, and it will only work when the voltage is in this range. When the common-mode voltage is outside this range, the stability and reliability of the communications will be adversely affected. In fact, the device could be damaged if the voltage reaches a level it cannot handle. Including galvanic isolation in the design of the device’s interface is an easy way of providing electrical isolation from the kilovolts of electricity that could occur between input and output, thereby avoiding problems caused by a high common-mode voltage level.

The demands of high mobility

AGVs, unmanned shuttles and other mobile equipment are used in today’s factories to increase efficiency. In addition, operators avoid wiring and space constraints by using wireless devices to control and monitor their mobile applications.

For mobile applications that use multiple access points (APs), roaming refers to when a client moves between two or more access points, with the speed of the mechanism used to implement roaming being crucial to maintaining a usable wireless network. As the client physically moves from one AP to another, the signal strength of the first AP will drop while the signal strength of the second AP will increase.

Factors that affect the smoothness of roaming include the topology of the access points, the gain and coverage of the antennas and the roaming threshold settings of the client. To ensure smooth roaming, you first need to take into consideration the route of the moving object, and then carefully plan the wireless AP deployment configuration.

Seamless roaming

The most crucial aspect of mission-critical wireless applications is ensuring uninterrupted communication between wireless clients and access points (APs), even when the wireless client is roaming at a relatively high speed between different APs.

A standard roaming mechanism only starts scanning for the second AP when the first AP disconnects, which can take 3-5 s or more to process, which is too long for critical industrial mobile applications. Such a long handover time can result in packet loss, which in turn can cause unmanned AGVs to temporarily lose contact with their control signal and then veer off course.

Ideally, mission-critical mobile applications require roaming times under 150-300 ms to ensure seamless wireless transmissions. Industrial wireless technology designed for mobile roaming applications should therefore implement a faster roaming mechanism. An example is the proprietary client-based roaming mechanism developed by Moxa: to avoid packet loss, the wireless client actively searches for APs emitting a stronger signal, without waiting for a complete disconnection. This pre-emptive type of roaming mechanism can react faster to the roaming event and hence provide a much shorter roaming time compared to standard roaming.

Conclusion

With so many critical factors associated with factory environments, system integrators are always on the lookout for the best comprehensive solutions to ensure wireless network reliability. Wireless devices must support several crucial features to ensure reliable wireless transmissions:

• 802.11n MIMO technology to improve the transmission and reception of multiple data streams. Not only can MIMO handle multipath effects, it can also increase the data rate up to 300 Mbps.
• Galvanic isolation for both the power source and antennas, since many different kinds of electrical disturbances can jeopardise devices and affect connectivity.
• Seamless roaming to ensure a smooth wireless handover from one AP to another in under 150 ms.