HMI & Operator Interface

FEBRUARY 2015

STATE OF TECHNOLOGY REPORT

HMI & Operator Interface From BYOD mobile apps to powerful PACs that tackle control, too, machine designers have more options than ever when it comes to human-machine and operator interface functionality.

Still using text display ?

It's SAMBA time!

PLC HMI I/O

IN ONE UNIT

Starting at $301 PLC features • I/O options: Digital, Analog, including High-speed • Auto-tune PID, 2 independent loops • Recipe programs and data logging via Data Tables

Communication features • SMS and e-mail • TCP/IP via Ethernet, MODBUS, CANbus, DF1 Slave • Remote Access utilities

• Supplied with RS232 programming HMI features port in 3.5” model and USB • 3.5" Color-touch, 4.3" Color-touch, 320 x 240 pixels (QVGA) 480 x 272 pixels programming port in 4.3” model • HMI graphs • 2 ports can be added: 1 Serial • Troubleshoot via the HMI panel (RS232/RS485)/Ethernet & 1 CANbus

[email protected] w w w. u n i t r o n i c s . c om

Unitronics, Inc. 1 Batterymarch Park Quincy, MA 02169 USA Tel: 617-657-6596, Toll Free: 866-666-6033

Table of Contents Choices Abound in HMI Purchase Decision What OEMs Don’t Tell End Users about PACs Get in Touch With HMIs and Machines Remote Access Here or There Industrial PC Basics Beyond Network Security Passwords Combine Control and Operator Interface Take a Look at New Technology Sleepless in Software City PC-Based Control Goes Consumer Data Access, Mobility: ‘Open’ for Business Panel Meters Take on PLC Chores Interface Interference in the Machine Operating World Is BYOD Inevitable in the Manufacturing Space? Give Your HMI an Ergonomic Tune-up The Pros and Cons of Embedded HMIs For Machine Builders Machine Information in Your Hand Cooling Complications in Hazardous Locations It’s Free, You Say? HMI: Form vs. Function?

3

6 7 10 12 18 20 22 25 26 28 30 32 33 35 38 41 43 44 47 48

Advantech's Wide-screen, Multi-touch Industrial Monitors • 16:9 wide-screen display, viewing area increased by 40% • Projected capacitive touchscreen with reliable glass surface • Robust design with 24 VDC power input and IP66 front panel • Slim design for easy installation with panel, wall, VESA mounts

FPM-7151W / FPM-7155W

FPM-7181W

FPM-7211W

15.6" Industrial Monitor with Projected Capacitive Touchscreen, Direct-VGA/DVI or VGA/HDMI Ports

18.5" Industrial Monitor with Projected Capacitive Touchscreen, Direct-VGA and DVI Ports

21.5" Industrial Monitor with Projected Capacitive Touchscreen, Direct-VGA and DVI Ports

www.advantech.com

Advertiser Index Unitronics 2 www.unitronics.com

Advantech 4 www.advantech.com

IDEC 9 www.IDEC.comĥ0+1$

Red Lion

17

www.redlion.net

Pro-Face

21, 27

www.profaceamericaċ+)/remoteHMI

Triherdral Engineering

24

www.trihedral.com/cd

5

Choices Abound in Human-Machine Interface Purchase Decision

H

As with other aspects of industrial automation, HMI display and input technologies have followed the arc of commercial computing technology, adapting the latest consumer advances to the unique demands of the industrial environment. The CRTs once widely used in HMIs long ago gave way to flat-panel LCDs, and some of the latest HMIs even tout wide-screen 16:9 aspect ratios and can be mounted in landscape or portait orientations. Traditional 4:3 aspect ratio devices remain available, especially for those machine designers interesting in bringing legacy HMI applications forward onto new devices. Touchscreens, too, are now ubiquitous in the HMI marketplace; some three-quarters of all machine HMIs purchased by Control Design readers now include one. Configurable keys, trackballs and sealed keyboards have their place, but from an ergonomic standpoint the touchscreen is uniquely suited to plantfloor environments: it has no moving parts, takes up no incremental space and is readily accessible by a standing operator. More lately, HMI touchscreens have even added the multi-touch capabilities popularized on smart phones and tablets to the industrial mix. Industrial PCs used as HMIs can even double as controllers as well, communicating directly with I/O and effectively eliminating the need for a separate PLC. They also can integrate with plant-level networks and systems to accomplish a variety of other tasks, from facilitating remote troubleshooting to uploading production reports. This balance of this State of Technology Report explores in greater detail these and other technology developments in the arena of machine HMI. We hope that you find it useful.

uman-machine interface (HMI) solutions can range widely in sophistication and cost, depending largely on how much interaction and information exchange is required between the human operator and the machine in question. Further, because the HMI often doubles as a data collection and supervisory communications hub, these computation-intensive tasks also must be taken into account when selecting an HMI solution. A relatively simple, unconnected machine may need only a few hardwired lights and pushbuttons. But at the other end of the spectrum, a fully functional industrially hardened PC may be required. In between these two extremes are a range of microprocessor-based operator panels or operator interface terminals (OITs) that provide a configurable graphical operator display and interface in a more compact, less expensive package than a typical industrial PC. In short, the need for increased application flexibility and the need to handle more complex information management tasks will typically drive the machine designer to more capable solutions. Indeed, for all but the simplest of applications, increased functional and performance requirements are driving machine builders away from simple status indicator lights toward increasingly sophisticated HMIs equipped with graphical displays. On the human input side, with the exception of a few critical buttons (for emergency machine shutdown, for example) discrete switches, too, have given way to configurable input keys which, in turn, are yielding to integrated touchscreens. The increasing need to integrate machine production data with plant-wide information systems also is driving the movement toward more communications-capable industrial PCs on what were once standalone machines.

- The Editors 6

What OEMs Don’t Tell End Users About PACs Many end users don’t really know what a PAC is, or what the differences between PLCs and PACs are. The reality is, OEMs and machine designers need to educate users better. By Don Fitchett, Business Industrial Network

I

ogy that any equipment customer can easily visualize is: A PLC is to a PAC as a digital clock radio is to a computer. Sure, computers have clocks built into them, but computers are structured differently and, with all the added functionality, are much more complicated than a digital clock radio. The clock/computer and PLC/PAC comparison can be applied to the intended end-user consideration in PLC/PAC design, too.

f asked, “Is there really any difference between programmable logic controllers (PLCs) and programmable automation controllers (PACs),” the machine end user will likely answer, “No.” Or that person might ask, “What is a PAC?” You may even get a few that will argue that no difference exists between the PLC and the PAC. Additionally, users may claim the PAC is just a new name and acronym created by the PLC vendor sales department to generate new sales. The fact is a PAC is very different from a PLC, and it is important for original equipment manufacturers (OEMs) and designers to educate their end users regarding the differences between a PLC and a PAC prior to purchase and again during customer training on any new equipment. Machine designers should prioritize educating their customers on the difference between a PLC and a PAC. Doing so will create significant benefits for both parties by: • lowering warranty period cost • increasing customer satisfaction. Success in achieving these two benefits is multiplied when OEMs write programs that provide the appropriate level of detail for end users who might lack prior knowledge and experience with PACs. Clear and detailed program documentation is often more important when using a PAC to control a machine. There are some major differences and special considerations an OEM should take when machine control uses a PAC instead of a PLC. An equal anal-

See More: PLC vs. PAC Comparison Simplified, PLCs are designed with the electrician in mind, and a PAC is designed with the IT/computer programmer in mind. With PLC designs, simplicity and user-friendliness take priority over functionality. Therefore, the PLC design focuses on ladder logic, which electricians could easily understand from their knowledge of working with electrical diagrams, and the PLC has a very specific control purpose. In contrast, the PAC is designed to handle multiple control purposes, not just a PLC, but also a motion controller, a DCS, four additional high-level programming languages, and more. The primary purpose of automation control is to improve quality, efficiency and uptime. Over the years, PLCs have evolved to serve these purposes well. Nevertheless, the PAC is still greatly lacking in these three areas in circumstances where plant personnel working with PACs are involved in maintenance or troubleshooting the machine or process. 7

• recommending training for maintenance specific to PACs and mastery of PLCs before moving on to PACs • using PAC software’s documentation functionality, such as rung comments, to document in great detail AUTOMATION CONTROL • abandoning the use of Table 1: Machine designers would do well to consider their end users, who will have higher level programto maintain, operate or otherwise work with equipment controls, specifically related ming languages, such as to important differences in device architecture, ease of access and version control. structured text, blocks and user-defined instrucTherefore, it is the responsibility of the OEMs to comtions, unless it is required pensate for the PAC’s shortcomings in these areas. to obtain the desire conTable 1 shows a side-by-side comparison of the trol and then only where PLC and PAC. The major differences are easily it is required identifiable. Machine designers would do well to • carrying over best practices in PLC programconsider their end users, who will have to maintain, ming to the PAC programming (cross-reference operate or otherwise work with equipment controls, subroutine, startup subroutine, all HMI in its specifically related to important differences in deown subroutine) vice architecture, ease of access and version control. • creating a subroutine for key and commonly Regarding architecture, a PLC has a single scan used processor status data (end users no longer cycle, and a PAC multi-tasks as a computer does. End have processor status data files in PACs, as they users no longer have quick and easy access to the ordo in PLCs, for troubleshooting). ganized data tables that PLC software provided, as Manufacturers and designers of PACs should make the PAC uses only created tags. With a PAC, end us- their primary objective to be considering the mainteers now have to be aware of firmware versions, and nance and operating personnel who will make minor even access software revision numbers due to lack modifications to the PAC, and they simplify those proof backward compatibility and maintaining current cesses during every phase of the PAC program design. functionality. Additional consideration should be given to end usThere are other drawbacks including configuration ers facing stark differences between the PLC and the of a PAC, which is much more difficult than it is for a PAC and to assisting in the awareness of disparities in PLC. PLC configuring cards are plug-and-play with the two systems. Machine manufacturers should keep the click of a button. PACs comprise several options in mind during control programming and design that that often need to be set manually and firmware ver- the end users may not be strong in PAC/computer arsions to look for, creating additional complexity. chitecture, computer programming knowledge or exSome of the most helpful ways an OEM can help perience operating them. If they address the end-user their customers to reduce downtime and reduce de- needs for the additional level of detail required, they mand for additional support include: will see less warranty calls and happier customers. 8

This changes everything.

New HMI+PLC with Extensive Analog Capabilities! Introducing the new, better than ever, FT1A Touch 14 I/O. It takes a lot to be the next best thing, but with a host of new features, these combo 14 I/O PLC/3.8-inch HMI units make it look easy. Perfect for advanced analog monitoring and control, FT1A Touch 14 I/O combines an operator interface and control in a single compact package, all programmable with IDEC’s

• PID Control • 65K TFT Color • Modbus TCP and RTU

user-friendly software. The only question is, are you ready?

Visit us at www.IDEC.com/touch

• Ethernet connectivity • 2 built-in Analog inputs (0-10V DC, 4-20mA) • 2 built-in Analog outputs (0-10V DC, 4-20mA) www.IDEC.com/usa

800.262.IDEC

Get in Touch with HMIs and Machines HMIs today are the main point of decision-making for operators, and this will continue in the future By Hank Hogan, Contributing Editor

M

achine-to-machine communication and a modern, touch-enabled device means there’s no need to stand around an HMI to diagnose and solve a problem. Machine-to-machine communication might benefit from the right touch—specifically the right multitouch input displays. Along with the ability of modern HMIs to remotely collect data, the technology could give control engineers new tools and capabilities. An example of how this might unfold comes from Lenze. The automation supplier announced a panel controller a year ago that features an ARM processor, Ethernet connectivity, a USB port and a resistive touch display. Importantly, its operating system is Windows, and that brings a host of features, explains Lenze’s technology evangelist Tom Jensen. These include the ability to easily pass information around and the power to graphically display it, leading to some interesting usage scenarios. “If I have one HMI and two machines, when one

machine burps, the HMI will notice and automatically ask, ‘Hey, do you want videos to help troubleshoot this other machine? Yes or no?’” Jensen explains. The two machines operating under such unified control could be an application, such as a filler-capper combination, which are used in pharmaceutical or beverage processing. These devices might process 250 units per minute for pharmaceutical operations and as many as 1,200 per minute in the case of beverages, according to Jensen. In part, this new approach that could involve troubleshooting videos is now possible because the devices have the computing horsepower to oversee several multi-axis motion operations simultaneously. They also can handle the data load associated with a high volume of manufacturing throughput. Software running on the devices also can react to defined conditions, such as an alarm, a changeover request or a need for maintenance. In those cases, a video could pop up and guide personnel through the appropriate actions to take. 10

Some of Beckhoff Automation’s industrial display families have multi-point projective capacitive input. This means that swiping, flicking, zooming and other operations found on consumer devices are possible. That familiarity brings benefits. “It’s much easier for training in an international market to get operators to understand a machine and navigate the different HMI screens more efficiently,” says Nathan Eisel, Beckhoff’s North America support manager. He adds that the input technology can be used with thin gloves on, unlike some other multitouch technologies. Behind the scenes, the use of OPC UA means that the devices can exchange data with other machines. Beckhoff’s products can be either the client or the server without hardware add-ons. Thus, they can do machine-to-machine communication with other systems on the plant floor. Beyond that, they can also talk to management systems and move data from shop floor to top floor and vice versa, according to Eisel. Looking forward, he sees two trends. One is unification of HMI and controllers into a single unit that talks both upstream and downstream, interacting with machines and management systems. The other is a change in the input and display systems. For example, Beckhoff Automation has studied the use of Google Glass, which integrates a heads-up display with a camera, in an industrial environment. The technology could indicate

things to come—the birth of a wearable HMI. HMIs today are the main point of decision-making for operators, and this will continue in the future, notes John Dirks, global product manager for Rockwell Automation’s PanelView Plus. The product family has panel sizes as small as four inches, with a 19-in. display planned. In Spring 2013, the company announced a new version of HMI software. It allows its panels to connect to and display data from noncontrollers such as power monitors or smart overload relays. As time goes by, the computers behind the panels will produce a wider array of data and will interface with more systems on and off the plant floor. Some of this data will be accessed remotely. For instance, the manager of a beverage plant might need to access a screen showing a key performance indicator of a bottle-filling machine. That can be done by connecting to the filling room HMI and extracting the data. This sort of machine-to-machine communication and a modern, touch-enabled device mean there’s no need for personnel to stand at an HMI to diagnose and solve a problem. As Dirks says, “The support person, be they maintenance, operations or an engineer, can connect into the terminal, see exactly what’s going on and be able to walk an operator through some troubleshooting steps without having to come out on the floor.” 11

Remote Access Here or There Remote monitoring, diagnostics and control tools enable machine builders and integrators to skip the travel, but offer more services By Jim Montague, Executive Editor

D

neers,” Moore explains. “Ipsen’s aftermarket support team is prepared to offer technical advice and help diagnose problems, and remote access to both control systems helps our technical personnel see what the system is doing. In the past, we relied on an Ethernet modem, which required an analog phone connection at both the customer and Ipsen’s locations. Phone modems are notoriously slow, and in some cases, providing the analog phone connection at the customer site proved difficult or impossible. We clearly needed a better remote access solution.”

emanding applications such as heat-treating can be complex, so furnace control systems must do more than regulate temperature. For example, a 10-bar, quench-furnace system provided by Ipsen (www.ipsenusa.com), Rockford, Ill., also must control speed, pressure, flow direction and other variables throughout the quenching process because they directly affect load distortion in die-casting operations. These parameters change from product to product, so furnace controls need to allow users to develop and test batch recipes too. Users of Ipsen’s industrial vacuum and atmosphere furnaces use its CompuVac control system to look into their thermal-processing applications in the aerospace, commercial heat treating, medical, energy and automotive fields. However, users still need more help. “Local controls provide a window into the furnace’s process with standard features, including an integrated touchscreen for monitoring workloads, displays for programming, running, real-time and historical monitoring, almost unlimited recipe creation, modification and storage, and alarm displays, batch reports, quality control audits and record archiving,” says Larry Moore, electrical and software engineering manager at Ipsen. The company designs and builds industrial vacuum furnaces, atmosphere furnaces and supervisory control systems, while its aftermarket support team helps users around the world solve problems, plan furnace controls upgrades, replace hot zones and secure parts, maintenance and field services. “Though CompuVac makes it easy to create and run custom heat-treating profiles and batches, users often have questions or need support from our engi-

Saving Miles and Time Luckily, the expansion, diversification and growing sophistication of remote machine support makes it more practical for builders, integrators and other service professionals to access users’ equipment and production lines from a distance, and then monitor, maintain, troubleshoot, repair and upgrade them without being physically onsite. Instead of dealing with clunky, old-style, dial-in modems, or even jumping through hoops to get permission to access users’ internal virtual private networks (VPNs) or other networks, the latest remote-access components let outside experts work on safe versions of a machine’s operating software and data, which are served up to cloud-based services that don’t require users and their IT departments to allow access to their internal networks. “We encourage customers to install ports into their systems to allow remote access for monitoring and troubleshooting,” says Jon Ertle, vice president of sales at Criterion Manufacturing Solutions (www.criterionms.com) in Comstock Park, Mich. 12

IPSEN INC. AND PHOENIX CONTACT

Trends in Technology

The Right Router Though they’re relatively new in remote machine monitoring, VPN routers are being deployed to remotely monitor and control all kinds of machines and other equipment because they’re easier to set up, more secure and less intrusive than other monitoring methods. For instance, to achieve secure remote access to its furnace controls installed worldwide, Ipsen’s support team evaluated several remote access solutions and chose Phoenix Contact’s (www.phoenixcontact.com) mGuard VPN routers. These allow Ipsen to connect to a customer’s industrial network via the Internet with little intervention from its IT department, while secure communication is provided by the VPN and a stateful packet inspection (SPI) firewall. “The router’s wide-area network (WAN) port typically connects to the customer’s company network, which gives it access to the Internet through the corporate firewall/router. But because it tunnels outbound— that is, back to Ipsen—no ports need to be opened on the inbound side of the customer’s network. This satisfies the customer’s IT department security requirements because outsiders can’t detect a port,” Moore explains. “Conversely, the router can be connected directly to the Internet via its WAN port if a customer doesn’t want any connection to its corporate network.” Once its initial connection is made, the VPN router runs at 99 Mbps, which allows Ipsen’s engineers to view system data in real time and download program changes when needed. The router can be installed in the furnace’s control panel via a DIN-rail module, a PCI card or as a portable device that plugs into a USB port, depending on the customer’s requirements. Typically, there is a router at each end of the tunnel. Ipsen installs one per furnace, but only one receiving router is needed at Ipsen’s home base to accommodate up to 250 simultaneous VPN connections. “The network is configured in such a way that our service technicians can access each customer’s VPN from laptops,” Moore says. “A technician can see all the cus-

ASSISTANCE AT A DISTANCE Figure 1: Ipsen supports its vacuum furnaces with VPN routers over the Internet, which allow data in a user’s CompuVac furnace control system and other devices on the local control network to be accessed remotely.

The company manufactures CNC routers and CMMstyle gauging machines and delivers custom production, automation and gauging equipment. “In the beginning, the best way was to dial in,” Ertle continued. “Later, due to security concerns with the Internet and early VPNs, we usually phoned ahead to request access, but it could take days or a week for some IT departments to grant it. Most recently, we’ve been able to use VPN routers, such as eWon’s (www.ewon.us) Cosy 141, which plug onto our customer’s machine, establish a secure, SSL-based VPN tunnel, and can call our headquarters when they have a problem.” This gives Criterion a safe, remote link to the PLCs and HMIs on its users’ machines. “We can also monitor and manage serial connections to program barcode readers and other devices, or we can integrate cameras or other peripheral equipment,” Ertle adds. “Using these new VPN routers saves our customers and us a lot of time. Many times, users contact us with a problem that’s actually a symptom or the result of another problem, but now we can look at their HMIs and PLCs for the underlying situation and solution.” Dominique Blanc, eWon’s U.S. general manager, adds that, “eWon delivers pure VPN remote access to users’ control systems at their customers’ sites, but our technology doesn’t need to change firewalls or jeopardize users’ IT infrastructures. Our secure VPN connection only reaches what’s behind our devices, but has no access to the rest of a user’s plant. So we can tell a customer that our remote access only reaches what it’s supposed to, and that makes IT much more comfortable.” 13

Trends in Technology

tomer furnaces that are tunneled back to the mGuard at Ipsen in a hub-and-spoke topology. Once connected, the router lets our engineers access data from any Ethernet-connected device on the furnace’s local network, including PLC, HMI, DAQ instruments and video recorders. The router’s own configuration can also be accessed remotely through the VPN connection.” As a result, the VPN router can be used for start-up support, maintenance support or customer-requested enhancements. And, though these installations on equipment are relatively new, Ipsen already has performed many remote control modifications and diagnostics that previously would have required an on-site service technician. “Saving the cost of one service trip under warranty is enough to pay for the cost of a system,” Moore adds. “Remote access is a mature technology, but past iterations often lacked performance, cost-effectiveness and security. Our mGuard VPN remote access system overcomes these challenges and provides safe, secure, high-speed and low-cost access to users’ equipment worldwide from one router located at our headquarters.”

such as MTConnect, which is basically plug-and-play. The second way is to use a device that doesn’t speak a standard protocol, but does have an adapter that translates from the proprietary protocol to a common format, for example, using an MTConnect adapter to speak to a Fanuc controller via its standard Focas protocol. The third way is to use a machine that can’t provide information through a software interface, so the only way to get information is by intercepting electrical signals. One advantage of MTConnect is there are lots of options for using it with legacy equipment.” For example, Okuma (www.okuma.com) in Nagoya, Japan, and its U.S. subsidiary, Okuma America, in Charlotte, N.C., stopped counting when its users reached more than 200 machines with MTConnect for shop-floor monitoring of its legacy and current, open-architecture Thinc-OSP controls, according to Brian Sides, Okuma’s technology director. “One notable installation occurred recently in Europe, where our customer wanted to connect its new Okuma machines to its existing Freedom eLog shopfloor monitoring system,” Sides says. “Using our MTConnect agent, we were able to provide the customer with the necessary plug-and-play connectivity to allow them to monitor the productivity of these new installations from their U.S. headquarters.” Freedom eLog comes from 5ME (www.5me.com), which is a new business launched in July that includes the tooling and services, cryogenics and software business units of the former MAG IAS (www.mag-ias.com).

Standards Aid Oversight To help improve machine monitoring, some builders have pursued standards to help streamline communications with their devices—and between them. While many builders still use basic TCP/IP and other Ethernet varieties such as Profinet, EtherNet/IP and EtherCat to enable machine connections and ties to upper levels, some interoperability problems persist. As a result, several developers launched the six-year old MTConnect open, factory-floor communication protocol, which was initially used for machine monitoring, status reporting and other details, but is growing to include alerts and alarms, temperature, speed and other information. “There are basically three ways to get information from a machine,” says Dave Edstrom, president and board chair of the MTConnect Institute (www.mtconnect.org). “The first is native support for a standard,

Security and Documentation Of course, despite the ability of VPN routers and other networking components to segregate network traffic and conduct secure tunneling, many users remain concerned that remote monitoring will expose them to intrusions and possible attacks. To allay these fears, most suppliers give users physical keys and switches, so they can enable their VPN routers only when remote monitoring and support is needed, and disable them when the problem is resolved. 14

Trends in Technology

THERMOFORMING BUILDER CUTS SUPPORT COSTS IN HALF WITH REMOTE ACCESS

Meet in the Cloud Once a secure VPN router connection or other external link is established, another primary way that remote monitoring and control can become more approachable and workable for many users is by sending applicable operating information to a

Remote machine monitoring used to be possible, but it usually wasn’t simple or easy. Leslie Adams, technical services director at Maac Machinery (www.maacmachinery.com) in Carol Stream, Ill., remembers using phone-based modems to connect to its shuttle and rotary thermoforming machines, which are used worldwide to manufacture aerospace, medical, automotive and home products. “I remember the frustration with trying to monitor machines when it took a long time for information to make its way back via the modem,” Adams says. “In one instance, working with a machine in Australia, the delay ran up to 15 seconds.” Thankfully, remote monitoring and control is far quicker and more secure with today’s virtual private network (VPN) devices. Maac employs eWon’s VPN routers, which don’t impact its clients’ IT departments. It also uses eWon’s cloud-based Talk2M service for automated recordkeeping. “Using an Internet connection, we can connect to machines just about anywhere,” Adams adds. “We recently established secure VPN connections with our machines in Calgary and Montreal and in Minnesota and North Carolina. As long as the customer has an Internet connection, we’re good to go. It eliminates the need for any kind of special interface. Using VPN routers eliminates 50% to 70% of our support costs, in addition to significantly reducing hours of machine downtime normally associated with waiting for a service technician.” Using VPN routers lets Maac improve its customer service in important economies including China, India, and the Pacific Rim. “Those of us in the United States and Canada take solid phone infrastructure for granted, but MAAC MACHINERY AND EWON

Besides its software-based security, Blanc adds that eWon’s VPN routers also have a hardware key, so users can turn on their local VPN and allow remote access when assistance is needed, and then switch off the VPN after remote assistance has been provided. “We also have Talk2M, which is like a historian that reports who’s connected to the router, when and for how long,” Blanc explains. “Typically, users have to provide a name and password to access a router, but then lose control of it after that point. Talk2M lets administrators manage their routers better because they can see who’s trying to access it, kick out any unauthorized users or simply set up a whitelist ahead of time.” Similarly, to maintain its security, mGuard has a digital input that can be wired to the switch or relay to activate its VPN tunnel. This lets each of Ipsen’s customers activate their tunnel when needed, which increases peace of mind because they’re in control of their own remote access connection. “The remote access system is all hardware,” Moore says. “No software is required. This provides a high degree of security because changes to hardware require deliberate effort that can be easily monitored, as opposed to software changes that can be performed at the touch of a key.” Likewise, mGuard’s SPI firewall keeps track of the state of its network connections, such as TCP streams or UDP communications, as they travel through it. “For instance its algorithm distinguishes legitimate packets for different types of connections,” Moore adds. “Only packets matching a known connection state are allowed by the firewall, while others are dropped or rejected. We and our customer jointly set up the rules, so no other entity can intrude on the system.”

this is not true in other parts of the globe,” explains James Alongi, Maac’s president. “VPN routers get our customers away from the cost of running special phone lines into their manufacturing facilities, and they give us the flexibility to service our machinery using the best communication technology available locally, whether it’s Internet, cellular or anything else.” 15

SAVING ON SUPPORT Maac Machinery’s Mike and Paul Alongi show off their heavy-duty shuttle-forming machine, which is one the company’s many machines that can be remotely monitored and maintained using VPN routers.

GRONINGER AND PHOENIX CONTACT

Trends in Technology

they’ve installed more than 3,000 machines, including more than 500 in North America. To help reduce its considerable travel and phone time, groninger recently worked with Phoenix Contact to develop its Remote Video Service, which it offers as an option on new machines or as an upgrade to existing, Ethernet-enabled equipment. The service begins with a secure, key-switch-enabled, customer-initiated VPN connection between a user’s machine at its facility and groninger’s secure, internal service network in the U.S. and Germany. Both sides employ FL mGuard VPN NAT routers to maintain a secure, encrypted VPN connection and tunnel. Most onsite machine networks include the usual PLCs, HMIs, servo controllers and other Ethernet-enabled devices, which groninger’s service engineers can access to see live program statuses, make any needed changes, backup or restore programs, create new recipes and deliver machine or software updates or revisions. Once a problem is resolved or the machine’s PLC or program is updated, the users can switch off their VPN key to disconnect their machine network from groninger’s service network. However, groninger’s service also lets users connect a remote-controlled video camera to their same machine network (Figure 2). So besides viewing live PLC and I/O displays, groninger’s engineers also see the machine from an operator’s perspective by panning, tilting and zooming in the camera to examine particular areas. For easy camera setup, groninger uses a Power-over-Ethernet (PoE) module to supply its remote cameras with power and data over one cable. The company also developed remote monitoring and control over wireless networks, which is a setup option in its Remote Video Service. This method employs one router, one key switch and one wireless access point at each production floor. As a result, each groninger machine with the wireless option has an antenna installed that allows it to connect to the wireless access point. The firm reports that wireless is especially effective for many of its cosmetics customers, who must reconfigure their production lines regularly to accommodate changes in packaging size, shape and types.

VIEW TO A FILL Figure 2: An operator interface and other crucial points on groninger’s fill-and-finish processing lines at its customers’ plants can be viewed at the machine builder’s home office via remotely controlled video cameras.

third-party location, such as a cloud-based service. This strategy gives remote engineers and technicians the data they need to support the equipment, but doesn’t compromise the user’s internal network security. John Curie, business unit leader for Thiele Technologies’ Streamfeeder product line (www.streamfeeder. com), reports that Thiele has added eWon’s monitoring on bigger products, such as its large collating systems for printed materials, which can be examined remotely via through the cloud-based service. “A lot of users are concerned about not being able to get support for their machine when they need it, so we can add an optional eWon module, which ties in to the controller,” Curie states. “Then, the customer assigns it an IP address, which allows only predetermined users to communicate with it on eWon’s own cloud. This means we don’t have to touch our client’s internal operating system or corporate network, but we still get enough information via eWon’s cloud to monitor machine performance, check for glitches, capture new operating data, examine software, and even make changes at startup or on the fly if they can be done in a couple of minutes.”

Seeing is Believing Besides accessing operating data and conditions, remote monitoring and control increasingly means collecting and relaying real-time video and other specialized data streams. For example, Germany-based groninger GmbH and its subsidiary, groninger USA LLC (www.groningerusa. com) in Charlotte, N.C., design and build fill-and-finish processing lines for pharmaceutical and cosmetics manufacturers. Since the firm was formed in 1980, 16

More than just a pretty face. Expand your options with Graphite™ HMIs. Connect more devices with plug-in modules and support for 300+ protocols.

Red Lion’s Graphite™ advanced operator panels are the industry’s first rugged HMI solution to combine flexible plug-in modules with built-in protocol conversion, data logging and web-based monitoring. From factories to extreme remote locations, Graphite HMIs enable customers worldwide to connect, monitor and control processes across a broad range of industries to meet all of your industrial automation needs. Take a three minute tour and discover the Graphite features that are moving HMIs into the next generation.

+1 (717) 767-6511 I [email protected] I www.redlion.net © 2015 Red Lion Controls, Inc. All Rights Reserved.

Trends in Technology

Industrial PC Basics Failure and the future: A final set of selection criteria for the factory floor By Hank Hogan, Contributing Editor

A

PHOENIX CONTACT

s in real estate, it’s all about location, location, location when it comes to basic specifications for industrial PCs. Other important requirement drivers involve things that all automation solutions have to deal with: failure and the future. A final set of basic PC selection criteria arises because computers are general-purpose tools that interact with other machines and people. With regard to location, one of the most important factors to consider when choosing a PC for a factory floor is the factory floor itself. Is it a non-hazardous area? Or is it a Class I, Division 2 area where there’s a risk of a fire or explosion in abnormal conditions? The answers impact PC choices profoundly. “Once you start going into a hazardous area, the list of all the wonderful options that everybody makes gets narrowed down very quickly,” says Louis Szabo, business development manager at Pepperl+Fuchs North America. He adds that what’s important is the certification for both front and back of a PC. The front might be rated Class I, Div 2. This means it has been judged to be safe when hazardous gases could be present but normally are not. If the rear hasn’t been rated as meeting this standard, then some sort of enclosure purge will be needed to keep the concentration of flammable or explosive vapors at an acceptable level. Putting in a purge system could limit location and drive up costs of the total solution compared to a non-purged system, Szabo says. He adds that there are no industrial PCs rated for Class I, Div 1, which covers locations where flammable gases routinely exist in ignitable concentrations. There are touchscreens that are certified safe in such conditions, however.

The other aspect of location is the overall environment. A factory floormight be hot, cold, dusty, prone to vibration, occasionally drenched in water, or all of the above. An industrial PC might have to operate in these conditions, which means it should have the necessary NEMA or IP ratings to keep out moisture or dust. In general, the best system has the fewest moving parts because they increase system reliability and uptime. Thus, a fanless PC is preferred. A rule of thumb, Szabo says, is that the failure of its cooling fan leads to overall PC failure within 20 minutes or so. The second general category of specs arises because failure of an industrial PC can be expensive. It’s not the components or the labor needed for troubleshooting and installation that is a big ticket item, though. Instead, it’s the downtime, with lost revenue possibly running into millions of dollars an hour. For that reason, systems should have easily re18

Trends in Technology

appropriate processor and storage, or that these be modular and upgradable to the necessary level. If not, then the system might not have enough computing power to handle demanding tasks or new software. Consequently, another basic requirement is that enough computing resources be present or be installable in an industrial PC for the full range of possible applications. Along with a capable processor, it’s important the system have the right amount and type of memory. It should have a display of the appropriate resolution. Today, displays increasingly have a 16:9 aspect ratio and more offer multi-touch input. The ongoing shift from a 4:3 aspect ratio display and single touch is an example of why future proofing is so important in industrial PCs. A technology advance such as segmenting functions into different cores is relatively new. It has blurred distinctions, and made the performance of an industrial PC comparable to that of a PLC. But, says Sidney McLaurin, PC-based automation product marketing manager at Siemens Industry, a PLC generally can’t readily handle visualization or other advanced tasks. It’s also generally not as open as a PC. Finally, just as no man is an island, no industrial PC is totally isolated. There are network connections to the factory floor, which means that one of the basic specifications is that the system either have the necessary communication protocols built in or that it accept a plug-in card with those protocols. At a minimum, an industrial PC should offer Ethernet connectivity, but even here care must be taken. Ethernet undergoes periodic transitions to a higher data rate, so systems must provide this capability or permit a field upgrade to a faster link. These types of requirements could push end users away from a less-expensive option toward a more costly one. In the end, though, spending more up front could save a tremendous amount of money and headaches later.

placeable power supplies, hard drives, motherboards, screens, firmware and other parts. The resulting ease of maintenance minimizes downtime, and the modular approach helps to future-proof the system against technology advances — if the right approach is taken initially. “When you’re looking for that industrial computer, you want to think ahead,” says Connie Chick, product manager for control systems at GE Intelligent Platforms. “I need it right now for this purpose, but in looking around, what else can it be used for?” A hidden assumption in both maintenance and modernization strategies is that replacement and upgrade components can be obtained now and will be available in the future. For a standard office PC, that is a dicey proposition once a few years have passed. Clearly, one basic spec for an industrial PC should be a design that eliminates such issues and an accompanying vendor commitment to ensure a supply of parts. You have to consider what a system might be called on to do in four or five years, says Eric Reichert, product marketing specialist for industrial PCs and HMIs at Phoenix Contact USA. Systems might start out handling just the control function, but graduate to more complex tasks such as advanced visualization of data. This reality is reflected in a basic system requirement. “People want a lot of options when it comes to processors. They want myriad choices,” Reichert says. Multicore systems are good examples of that. Such setups can run a deterministic, real-time control system on one core, while another handles advanced visualization, interfaces with the network, interacts with operators, or does some combination of these tasks. The advantage is that the all-important control function can continue even if the operating system running on another core freezes up. However, setting up this type of arrangement after installation demands that the system have the 19

Beyond Network Security Passwords New HMI Innovations Promise Better Access Control By Hank Hogan, Contributing Editor

A

dividual access to a machine. An RFID-based approach doesn’t suffer from some of the drawbacks that are found with alternatives. “Biometrics is a challenge because a lot of operators have to wear gloves, or they work in a process where their hands get dirty,” Palus says. As for retinal scans, they’re frequently stymied because operators wear goggles or safety glasses. Rockwell Automation itself doesn’t offer an RFID solution, however, partners such as RF IDeas have readers that can be integrated into a machine. Often maligned, passwords actually can be secure — provided that user accounts are managed properly, Palus says. This typically means that user groups need to exist, with permissions given to the group at large and then individual users assigned to a group. It’s also necessary to be able to add, drop and modify user accounts. The next Machine Edition of FactoryTalk will have these capabilities, Palus claims. Now, making an HMI panel more secure can improve machine access control, but only if it’s used. That, in turn, often is determined by the difficulty level of administrative tasks. If users must jump through too many hoops, they’ll be tempted to bypass the access control system in some way. Siemens Industry kept that in mind for its WinCC software family. The software has tools to manage usernames, passwords and groups in an administration scheme. “Siemens has provided user group administration for years to provide customers with this power and flexibility,” says Wayne Patterson, U.S. Simatic HMI product manager. “It’s very easy to provide new individual users the rights that are already associated

t the machine level, access control involves managing who can do what using the HMI. For that human-machine interface, advances borrowed and adapted from other areas are now used to bolster usernames, passwords and other old security standbys. These innovations promise better access control. At B&R Industrial Automation, local access control for machines has been improved through the use of two technologies, says John Kowal, director of business development. First is the incorporation of the same RFID technology often used to control access to buildings. The second is locking down machine software through code compilation. RFID is found in fobs and cards, and company employees pass them near a reader to open doors and gain access to facilities. “It’s better than a password because people ‘pass’ passwords along,” Kowal says. “It’s certainly less complicated than a biometric.” The access that’s granted can be at different levels. Operators, maintenance techs and engineers could be granted permission to interact with an HMI to different degrees. As for code compilation, that offers some protection against inadvertent or intentional changes to a program. Kowal says B&R Industrial Automation uses IEC 61131-compliant languages, allowing function blocks to be locked. This is unlike the situation in a conventional PLC, where such safeguards don’t exist, he adds. Rockwell Automation also sees demand for RFID authentication, says Tad Palus, global product manager for visualization products. One of the reasons for this interest is a desire to track and control in20

with specific functional groups.” He adds that users want open, yet secure, systems. However, an HMI that’s locked down to prevent changes is not enough because the HMI itself can be replaced. A laptop, for example, could be plugged in to a network and a new HMI dropped into a machine. For that reason, device authentication is necessary, even for machines not connected to a network. Keeping with the theme that no machine is truly an island, Paul Forney, chief security strategist for the common architecture and technologies group at Invensys, notes that HMI access control must include strict limits and the monitoring of any access

from outside the machine. Threat-detection engines that pick up malware activity can be helpful for this, he says. For protection at the machine level, there are some promising developments, provided that system and operator panels are deployed securely. For instance, DeepSafe, a joint development between McAfee and Intel, assists security at a hardware level. That improves safeguards and access control efforts. “This new technology sits below the operating system and close to the silicon, allowing for an exceptional vantage point in the computing stack to better protect systems,” Forney says.

21

Trends in Technology

Combine Control and Operator Interface The upsides and downsides of contoller-display combo units By Dan Hebert

A

combo controller and operator interface can save big compared to separates. As the miniaturization of electronics continues its relentless march across the personal-consumer-device landscape, it’s only natural for it to proceed apace in machine and robot automation systems. One consequence is the combination of what were once separate components into a single housing, as with a machine controller and an operator interface device. Although relatively new on the scene, these combo units have seen significant adoption by machine builders because they are less expensive than separates, require no wiring or integration between the controller and the operator interface, and take up less panel space. For many applications, these benefits more than negate disadvantages, which include a single point of failure for both control and operator interface and a lack of the highest-end functionality, particularly for basic units.

These combo units come in two main f lavors. The first combines a PLC with an operator interface terminal (OIT) to create a unit designed for basic machines. The second marries a PC-based controller to a full-featured HMI, creating a unit capable of providing control and operator interface for the most complex machines. Combo PLC-OIT units were initially introduced with limited features and functions, but more recent products have upped the ante by adding more sophisticated capabilities. “The newest addition to our FT1A Touch micro programmable controller series of combo HMI+PLC units is the FT1A Touch 14 I/O, with new features making it suitable for advanced analog monitoring and control,” says Don Pham, a product manager at IDEC. “The FT1A Touch 14 I/O provides up to 158 discrete and analog inputs and outputs, using FT1A controllers as remote I/O slaves, PID control, Ethernet communications and a built-in 3.8-in touchscreen HMI in a compact package costing less than $500,” adds Pham. 22

This is obviously an attractive price point, one hard to match by purchasing a separate PLC and an OIT, particularly when the cost of wiring, integrating and installing two separate units is taken into account. A bit higher up the scale in price and performance, staring at about $1,000, is the Perspecto CP TV line of combo units from Wago. These units feature five sizes of TFT touchscreens: 3.5, 5.7, 10.4, 12.1 and 15 in. “Performance is optimized with scaled processing power up to 1.6 GHz processor, 256 MB of RAM and 128 MB of Flash memory,” notes Charlie Norz, the product manager for WAGO I/O Systems. “Our combo units are programmed using CoDeSys software, providing advanced programming tools, support for all the IEC 61131-3 programming languages and an easy-to-use graphic editor,” notes Norz. These units also have multiple interface ports, including CAN bus and Ethernet, and a built-in Web server that allows remote users to view and control the graphic screens using any browser. These combo units can be a good fit for machine builders not requiring large HMI-type screens, hundreds of I/O points or advanced control functionality. For applications requiring those features and functions,

the next step up the line are combo PC+HMI units. Readers over 40 years old may have not-so-fond memories of the sheer size, bulk and weight of older PC-based control systems. Not only was the CRT-based screen a monster, so was the industrial PC. Add, in some outboard I/O, the entire package was cost- and size-prohibitive for all but the most high-end applications. But times have changed, and new units simply tack a PC-based controller onto the back of a f lat-panel screen, creating a slim panel-mount package with reasonable weight and not much more depth than a monitor alone. A pioneer in this area is Beckhoff Automation with its panel PCs. “Rather than recommending a multi-component solution with separate PLC and HMI hardware, we offer customers an all-in-one approach combining an industrial PC and HMI, packaged as a streamlined panel PC,” explains Reid Beilke, the industrial PC product specialist at Beckhoff. “These units offer multicore processing performance, available multi-touch functionality and customizable housings. When running our TwinCAT software, one multi-tasking panel PC can handle the work of multiple PLCs, while also performing motion and robotic control,” adds Beilke.

23

Take a Look at New Technology Tools such as Google Glass could offer more than meets the eye By Mike Bacidore, editor in chief

S

data per day. And that’s just what Google processes. By latest count, the commercial Internet carries almost 2,000 petabytes of data each day. Just wait until we have an Internet of Things. Michael Ziesemer, COO of Endress+Hauser, once told me that Google understands the Internet, but Google doesn’t understand things. I’m not sure how much longer that assessment will remain accurate, if it even still does at this point. The online search engine company—does anyone even think of them this way any more?—makes continued forays into hardware with its visual interface and its self-driving car, not to mention its acquisition of Nest, the maker of smart thermostats and smoke alarms. For Google, the hardware is simply a means for interfacing with the data. And we all know that Google knows data. Rather, Google knows what to do with data. It knows how to access it, how to index it and how to make just the right data available at the exact moment you need it. Remember when you thought it was creepy for Google to be able to autofill your search request? Now you use it as a tool to find things you didn’t know you were looking for. So many of the brilliant things we have, from Teflon and vulcanized rubber to Coca-Cola and potato chips, were the result of a discovery that people decided to use differently. Google Glass will most likely take the same path. Put on a pair, and see what you think.

ometimes there’s more to a new innovation than meets the eye. Google Glass is the latest bit of see-it-to-believe-it commercialized “wow” to show up in exhibit halls. The technology itself has been around for quite some time, but now it’s being touted as an interface for accessing controls data, viewing an instructional demo video on resetting a valve, or asking the opinion of a colleague in another country about whether to use pneumatics, hydraulics or servos on a motion application. We assume we know what Google Glass will enable us to do, but we won’t really know for sure until we start using the tool and figure out its potential applications. I can almost guarantee that, in three years, it will be used to complete tasks no one has even considered at this point. The future’s so bright, I gotta’ wear Google Glass. The Internet of Things will enable access to an inordinate amount of data. How much? I recently attended a conference where one company rightfully boasted of the terabytes of data that it will be capable of managing in one application. And, on first blush, that sounds impressive. But think of how quickly we’ve moved from kilobytes to megabytes to gigabytes and now to terabytes. It won’t be long before we look back fondly on the quaint past when you might purchase an exabyte of storage to keep your hourly reports secure. Google already processes more than 20 petabytes of data on a daily basis. That’s a quadrillion bytes of 25

Trends in Technology

Sleepless in Software City Using and knowing more than one software platform and having more than one person involved suggests success By Jeremy Pollard, CET

I

had a nightmare. Indusoft has been scooped up by Invensys. Who will be next? Inductive Automation? Are those the last two independent software companies that were and are innovating? That’s the scary part. It’s not that big companies can’t produce innovative solutions. But I fear in the long run, we as users get the “it’s all because of you” speech, which can be rephrased as, “Thanks for the money, and now we can move forward the way that we think we should, but give you lemmings the impression that it’s all for you.” And then — as if the gods allowed it — I got an email from B-SCADA. So I spread my Google wings and wondered what the landscape looked like for HMI stuff, as I kind of did with programming software last month. I found an interesting garden. The biggest difference, in my opinion, is that HMI/SCADA is protocol-dependent and not hardware-dependent, as programming software seems to be. OPC has torn the protocol world apart by providing connective tissue for any device, anywhere. While this is good, it allows for a plethora of options to gather data in real time or historically with or without graphics. Writing a service in Windows or Java isn’t as hard as it might seem.

and of course, it’s everything mobile. Cloud-based stuff is a very intriguing technology. I fear, however, that having a third-party manage your systems might not be the right thing to do for many reasons. Inductive Automation, Indusoft and the big guys all have solutions for graphical interfacing for operators, management and for presentation of data. Is the market big enough to support all this activity? Based on past pricing models, I believe it is. However with web-based interfacing and the ability to use Linux or a Raspberry PI device as the local connective device to the web, where’s the business reason for spending a lot of money on SCADA solutions? Heck, Inductive Automation broke into the market by delivering a fully functional HMI solution complete with an OPC driver set for free. How that works is elementary, my dear Watson. The usability of any given system once was based on the development speed and rollout, and costs associated with multiple users. Management wanted information for business; engineering needed information for process, and operators for control. It’s no different now, but I have to wonder how hard it is to move from one platform to another. Ian Nimmo deals with abnormal situation management (ASM), which is essentially alarm systems. He has advocated for the longest time that SCADA screens are too busy, too colorful and overwhelming to most users. So if we have simple screens and a tag database in an open format, such as SQL, our movement should be easy. If we also choose a web-based server system, then there’s nothing to stop us from having four or five

So Where Do We Sit Right Now? Groov from Opto 22 provides a specific HMI platform that is hardware- or software-based. Because of OPC (in a future release) and a web interface, you can use it for any hardware. That really goes for all things SCADA and Internet. Cloud-based services are becoming available, 26

different servers using multiple OPC data servers. We can have best-of-breed alarm servers and data-logging software — and with web clients that shouldn’t cost an arm and a leg. There’s no excuse for any process or machine to have inadequate software. RSView, Wonderware InTouch, GE Proficy and WinCC are some of the big guns. The pricing models are so 1990s. When I did a cost comparison for a client that needed 36 client nodes, the costs were so prohibitive that I suggested custom software, which reduced the cost by over 80%. Big money.

To roll out what that client needed today would be even less, but as with all systems, having the ability to have multiple experts in the development stream is crucial to future success. Using and knowing more than one software platform and having more than one person involved suggests success. Programming software is so detail-oriented; SCADA and HMI not so much. A fully functional server can be had in minutes for a cost that will amaze you. Limits have been removed, which will take us back to the origins of SCADA. A spreadsheet and OPC — I wonder where DDE went.

27

PC-Based Control Goes Consumer The ability to create independently operating partitions could do more than ease real-time control. It could make cybersecurity more feasible, too By Hank Hogan

T

he trends in PC-based control are to divide and conquer, take a page from consumers, and get networked. For machine builders, these advances promise easier real-time control. The strategy of divide and conquer benefits from hardware changes. Chip vendors produce systemon-chip (SoC) solutions with a growing number of diverse and specialized cores. “Today’s multicore SoCs are comprised of multiple CPUs of both similar and different 32-bit cores with specific execution engines for graphics, networking, process control, plus analog and digital I/O,” says Dave Mender, vice president of business development for real-time operating system (RTOS) supplier Green Hills Software (www.ghs.com). The company’s software helps manage these various cores. He says this is part of a trend toward consolidating high-level PC-based functions with real-time process control into one platform. The advent of industrial PCs with multicore processors allows control and other functions to be handled in one system, says John Wilhite, product manager for PC-based automation at Siemens Industry. “In the past, you might have needed an HMI

with its own processor and a PLC with its own processor. There might have been a PC to capture that data, and send it to an enterprise system. Now all of that will be controlled by an industrial PC [IPC],” Wilhite says. He adds that the company’s real-time software controller and its failsafe version support several RTOSs. More processing power and resulting platform consolidation feeds another trend: consumerization. For example, consider the July announcement by Beckhoff Automation of its four-core, Intel-i7-powered panel PC. Reid Beilke, IPC and embedded PC product specialist at Beckhoff, points to its consumer-device-like, multi-touch technology and benefits. “Early adopters of the technology are implementing more intuitive and interactive HMIs that are much more in line with modern electronics users,” he says. Of course, not all processors are created equal. Some consumer experiences, such as gesture recognition or overly complex user interfaces, are computationally too taxing to recreate on a factory floor, particularly on embedded systems that are older and less powerful. 28

Products from RTOS vendor QNX Software Systems mitigate these resource constraints with adaptive time partitioning. This means the control loop or some other function might be guaranteed for at least 20% of the system, but that allocation is not fixed. “If you’ve got work to do and no one else is ready to run, we’ll actually give you the rest of the CPU,” says Grant Courville, QNX’s product management director. He adds that users of PC-based controls do want a consumer-like experience. If this is to be done in a cross-platform way, then one solution to use standards such as HTML5, Qt and OpenGL ES, Courville says. A final trend in PC-based control development is increasing connectivity, which is expected to explode with the arrival of the Internet of Things. The industry has incorporated sensors and automation in workflows for decades, yet only about one in 10 legacy systems currently is connected, notes Michel Chabroux, senior product manager at Wind River. Among the company’s products is the VxWorks RTOS. As things change and connectivity expands across the factory floor to devices that previously were isolated, embedded systems increasingly will

be concerned with cyber security. That will place additional demands on real-time operating and PC-based control systems, which in turn will require new must-have capabilities. “An RTOS must give customers the flexibility to design their embedded system to the necessary level of security by leveraging a comprehensive set of built-in features covering design, boot and execute, operation and power down,” Chabroux says. The ability to create independently operating partitions could do more than make real-time control easier. It could also make achieving needed cybersecurity more feasible. For instance, an HMI or a partition running Windows might be allowed to talk only via virtual Ethernet to a firewall. Such virtualization and abstraction of hardware can protect an embedded system against cyber assault. However, it must be done in a way that preserves determinism and PC-based control, says Kim Hartman, vice president of sales and marketing at RTOS supplier TenAsys. Moving everything to one platform might not be that painful thanks to multicore technology and power, he adds. “You can afford to bring your existing workloads and existing operating systems into this environment with little—or possibly no— changes whatsoever.”

29

Data Access, Mobility: ‘Open’ for Business HMI/SCADA remains the biggest area of vulnerability. By Jeremy Pollard, CET

I

believe we might have a new player in our space. Well, not really new, but the face is different. And it might be a bit angry. Microsoft’s Windows 8 platform has failed miserably. They have shaken up the division to be more agile and to become more competitive in the ever-changing marketplace of data access and mobility. Let’s take inventory. Most applications use Windows-based platforms and data structures like SQL. Most floor-based stuff is Wintel. Any server-based stuff is typically Microsoft. There’s wide acceptance of OPC-based communication drivers for interconnecting devices from all vendors. There’s OPC connectivity for IEC-61131 projects, and building and home automation. You can create your own driver for your fridge. The only wall of protection with most PLCs is the firmware that compiles or interprets the actual code in the PLC. That’s no accident. Industrial software always has been very expen-

sive. Hardware was, too. But hardware costs came down big time, and it became commodity-like. We still need software though. Those pesky support agreements, too. Revenue streams are becoming harder for all of us to find and maintain. That won’t change. Manufacturing is stagnant if not still declining in parts of North America. The global market is in the same boat. We don’t need all the widgets that are made, and we don’t have enough people with any money to buy them. So, expensive support agreements might be the first to fall. The biggest area of vulnerability is HMI/ SCADA. We don’t need the Wonderware’s and RSView’s anymore. We don’t need or want to pay for machine licenses, or by the number of screens. In fact, we have become the children of the net. We don’t want to pay for anything. We will have that chance, I believe. We’ve given the world the opportunity to take our software market. 30

We use commercially available standards and hardware. We use software platforms and interfaces that everyone uses. We promote the concept of putting it all online using web browsers to get at the data. I believe strongly that due to security and a lack of specific controls at the IT level that are intrinsic to industrial software, the bigger boys like Microsoft, maybe Google and others, could come calling. They don’t have to interface with anyone anymore because we’ve given them the key to the office. We said it wouldn’t matter because our industry is a comparatively small market for them to care. That is and will change due to the struggles that all companies are having with revenue streams. Heck, even I can create an HMI package or SCADA package using Java, VB.NET, Visual C# and distribute it on Cnet at no charge. Just because I could didn’t mean I would. Inductive Automation has done just that. A single machine interface is a no charge item. It includes the OPC server as well. It can run on Linux, so costs are reduced dramatically. pvBrowser is a GNU project which provides a free platform for SCADA. While it might not be as f lashy as some mainstream products, it isn’t an example of “you get what you pay for.”

The recurring revenue model is making a comeback. Some companies’ longevity depends on it. I talked to a few machine builders about recurring revenue models that could include gathering performance data so you can plan or schedule predictive maintenance, or provide a valuable service to their client by monitoring their machines tolerances so their production doesn’t experience unscheduled downtime. I was met with a blank stare. We sell machines. End of story. Sorry, I thought you ran a business. I must be mistaken. Opto 22, for one, gets the definite-purpose thing. Their new groov box requires the hardware to develop anything. It needs its own hardware to work, so one wonders a bit about the concept. I still believe that PC-based control was shunned by the “rent-collectors” of our industry. They couldn’t have their PLCs replaced by commercially available hardware. Proprietary is the only way to go, they tell us, but they’ve been pulled kicking and screaming into this land called “open.” This is Microsoft’s new focus: operating systems, applications, cloud computing and devices. Ring any bells? We’ve opened the door to any and all participants. After all, most of our stuff isn’t rocket science. Not to worry, there will be an app for that. 31

Trends in Technology Panel Meters Take on PLC Chores Panel meters incorporate capabilities to improve process visibility and keep up with transmitter enhancements By Hank Hogan, Contributing Editor

B

interface division of Red Lion Controls (www.redlion.net), notes that the automation industry is moving toward redundant control, driven by increased safety demands and regulatory requirements. Satisfying that need is something that can be handled by a panel meter that can do math and other functions previously provided by a PLC. “Should something happen to the PLC, the panel meter is still there, capable of performing the most critical tasks,” he says of this approach to redundancy. He points to the company’s series of panel meters. They offer the ability to display readings in red, orange or green, thereby supplying a visible indication of the status of a process or machine. This is particularly important when there are many different variables that have to be monitored, with operators having to ascertain the health of a system at a glance. As for the future, Red Lion is working to further expand functionality by enabling users to create a specialized panel meter that pulls in data from various sources. The result will be an output tailored to track the status of a specific process without the need to do so with an array of panel meters. The company expects to have these new products available shortly. Daniel Sparks, director of product marketing at Omega Engineering (www.omega.com), notes that one of his company’s latest panel meter offerings has a graphical display, full natural-language menus, USB and wireless communication, and onboard data-logging capabilities. Other panel meters offer a response time of 300 ms and provide a local indication of the state of a machine or process. Sparks notes there was a time a decade or so ago when discrete panel-meter-based solutions were losing market share to PLCs. He says that is no longer the case, in part because panel meters are more capable than before, while still being easier to implement than PLCs and still providing a local display of process variables. As a result, panel meters offer advantages both as a replacement for and for use in conjunction with PLCs. “They are a cost-effective alternative to adding PLC control and measurement loops, and most can be configured to communicate to PLCs for enterprise data handling and process management applications,” Sparks says.

ecause they’ve gotten smarter and have additional capabilities, panel meters now can take over some functions previously handled by PLCs. For instance, a panel meter with math capabilities can keep count and shut off a machine after a fixed number of cycles. For those who want to put these capabilities to work, a panel meter-based solution offers advantages. It’s less expensive to implement and easier to set up than a PLC-based system. Panel meters also can offer an alternative control path, important when implementing a redundant solution. Balancing these positives is a potential negative: Panel meters might be a bit slower in executing functions compared to a PLC. A final factor to consider is that panel meters are likely to be present in any case because they provide vital visibility, says Joe Ryan, marketing manager at Precision Digital (www.predig.com). “Operators always will need easy-to-understand information, thus, the presence of the panel meter.” While discussing panel meter technology trends, Ryan points to two of his company’s products in particular. The ProVu series dual-input meters offer math functions. The panel meters can add, subtract, average, divide and multiply. As a result, they can replace PLCs used to track diesel fuel consumption or ones that do mixing-ratio calculations. For these applications, panel meters are easier to set up and provide better process visibility, according to Ryan. As for the second category of advanced panel meters, he mentions the company’s Modbus scanners. These poll up to 16 Modbus variables from a variety of devices and can do math. Consequently, they offer a way to easily display information from multivariable transmitters, such as level and temperature, as well as provide analog output of the data. In the past, these tasks would have required a PLC. Panel meters incorporate such capabilities to improve process visibility and keep up with transmitter enhancements, Ryan indicates. Because this is an ongoing need, he looks for the trend toward more intelligent and higher-functioning panel meters to continue. Jeff Thornton, product manager for the panel meter and 32

Interface Interference in the Machine Operating World The Responsibility of Control Can be so Easily Given to Those Who Have Trouble Remembering What They Had for Breakfast. By Jeremy Pollard, CET

I

nterface, interface, interface. Oh, what Windows and the mouse did for the interface. But now, gestures and multi-touch on small, portable, surface-based devices have changed the game. They will create, if they haven’t already, a generation of non-verbal and frazzled participants. So what happens when the lines of communications get broken or simply get hidden on the really big, three-inch screen of a smart phone? Would a machine or process operator really use one? Do they stop trying? Or do they get weary or complacent and just forget to respond? If you believe that this interface type is the real deal, why would any company put its future in the hands of a commercial third-party like Google (Android) to provide the window into “my” world, not knowing if the window is going to be shut at a mo-

ment’s notice? We’ve done it, I know, but this time will be different. Sounds drastic? Maybe, but I fear that not enough of us are at least thinking that their worldwide access platform might be a compromised arena. I moved into the present day by grabbing a Blackberry Q5 smart phone. It runs BB10 and has many really cool features and apps, so the conundrum of “free” lives on in such esoteric and non-pervasive apps as Flashlight. However, as I noted last month, it “needs” to know your location and personal information so it can turn on. Interface? No. Intrusion? Yes. But the flashlight did come in handy while I was on an emergency start-up, and had to peer into the dust-laden panel and pour over drawings. So a Q5 phone employs multi-touch and scalability amid the illusion of modern. By that I mean that 33

everyone knows that a multi-touch gesture of two fingers ‘grows’ the screen. I write this thinking that our industrial operator stations are OK with these facts. “Oh, crap, is it a right finger or left finger hold or..?” This while Rome burns, and the system goes out of control. The responsibility of control can be so easily given to those who have trouble remembering what they had for breakfast. So I guess it’s clear that these new phones and tablets give us the interface we need for remote access and mobility with the interface of web-based commonality. Really? Remember the F1 key? That was for help in any application. But in this new touch-based world, does F5 mean refresh in every web-based, remote, mobile HMI app? I’m pretty sure the answer is no. So where have we gone wrong? When we got our new phones we were promised 10 hours a month of web-based real TV. The screen size is 3-in. diagonal, and I am well beyond 45. What were they thinking? Can’t wait for Surround Sound from these bubbas. One wonders how operators might respond to any alarm, issues, page, setpoint deviation alarm and setpoint change when they’ve used 24-in. screens for years, and I would suggest that they might not have dealt with or interfaced with them well.

Teamviewer is a common application for remote access internally, just as Remote Desktop Protocol (RDP) would be. This means that one would not have to have the SCADA client on the device. I installed it on the iPad to check the action out. The gestures are odd, since there is no keyboard/ mouse as such. Tap is left-click (easy). Tap and Hold is right click, and there are four others. Once you get used to it, then all is good. The Q5, however, is not the same. Tap is used often. Tap, hold and drag from various positions on the screen do different things, as well as introduce various components. My biggest concern is visual availability. You can’t see anything worthwhile because of the screen size. So an application such as Teamviewer accessing a normal PC with 100 tags on it would be silly. You can wonder how that works. It’s kind of like a mobile device vs. a fixed device accessing a normal website. You can get to the same data, but who knows where it is? Make no mistake. It’s not that we as a group can’t learn, but just because we can, doesn’t mean we should. As I said, the majority of us are not spring chickens. Long live the 17-in. laptop with mouse! 34

Trends in Technology

Is BYOD Inevitable in the Manufacturing Space? Will your next HMI RFQ include tablet or smartphone wireless connectivity? By Control Design Staff

I

t’s interesting to watch the bring your own device (BYOD) movement leak into the manufacturing space, but we have concerns. Many larger customers or potential customers are doing it to some degree in their IT groups. We’ve been successful with the plug-in or wireless pendant for multi-station HMIs; we know it well, and it does the monitoring and control our customers need. It seems inevitable that any day now an RFQ will include tablet or smartphone wireless HMI requirements. We’d like to have a few legitimate, technical performance reasons that argue for keeping what we have. Or is it time look for a seat on the bandwagon? —from February 2013 Control Design

access,” “kiosk” or similar secured modes to restrict application use. COTS mobile devices benefit from the sheer scale of deployed devices. They’re inexpensive, readily available, familiar and offer useful features. A COTS smartphone, for example, will cost much less than a wireless handheld terminal, use modern, non-proprietary networking standards like Wi-Fi and Bluetooth, and require little training to use. Ongoing software and hardware development occurs for an installed base of millions of units, not hundreds or thousands. Cases for industrial environments—even Class I Div. 2—are available to protect popular smartphones and tablets. Several mobile device management (MDM) software suites can secure off-the-shelf devices used at work. Finally, one of the key features of COTS mobile devices is their capable web browsers. Some automation companies, such as Opto 22, offer tools to develop web-based mobile operator interfaces for securely monitoring equipment and systems, without the need for additional software. In a nutshell, using COTS mobile devices lets you retain device control for security reasons, while using the advantages of off-the-shelf mobile platforms. Ben Orchard, application engineer, Opto 22, www.opto22.com

Answers HOP ON THE BANDWAGON Yes, an RFQ with tablet or smartphone HMI is inevitable, and it’s time to get on the mobile device bandwagon. The good news is that you can do this on your terms. In the manufacturing space, instead of supporting BYOD devices (which, by definition, are devices users already have and bring to work), consider supporting affordable commercial-off-the-shelf (COTS) smartphones and tablets specifically issued to users. This approach gives you better control over both application security and performance because you specify the device and accessories, configure security and application software, and maintain the device with OS and application updates. You can even lock down mobile devices in “guided

BEWARE BYOD The main problem of BYOD is found in the meaning of the acronym. You own the device, so you decide what is on the device and what you do with your device. This raises security concerns due to possible lack of control, and it demands extra measures. In 35

Trends in Technology

can bring valuable gains in accessible, convenient information and production control. As a web-enabled HMI application, it extends access to machine displays and dashboards to users anywhere for improved real-time decision-making. Operators also have the freedom to perform maintenance and troubleshoot remotely or walk along the conveyor line to check sensors in commissioning. With HMI mobility remote users can: • Gain access to download programs via FTP file transfer; • Increase diagnostic information gathering and remote troubleshooting capabilities; • Gain administrator login access to view terminal diagnostics via the VNC, with no disruption to the operator; • Check production rates and capacity or view key performance indicators from the road, home or an office terminal. HMI mobility can usually be accomplished without the need for costly new software or infrastructure changes. For example, an embedded, remote-connectivity feature on the Allen-Bradley PanelView Plus 6 HMI terminals can provide data access to real-time, plant-floor operations by extending the HMI content to a Windows, iOS- or Android-based device using VNC technology. Typical smart device VNC or remote desktop applications can be used. Such applications are likely embedded in a dedicated terminal already, so don’t wait to tap the power of mobile information. Remote visualization capabilities will put you ahead of the game in response time. John Dirks, global product manager, Rockwell Automation, www.rockwellautomation.com

exchange, companies obtain money savings and alleged increased employee productivity. For non-critical activities such as mailing or sharing a company’s calendar, security issues can be worked out, and the BYOD movement reaches its full potential. However, for critical operations such as controlling machines through HMIs, the risks could be higher than the benefits. Even if security concerns can be overcome, a device misconfiguration or a temporary glitch on a device might prevent a worker from properly dealing with a system requiring real-time operation. Attempting to support many disparate devices might not be the best idea in real-time scenarios. As developers of mobile apps for the manufacturing industry, we do not encourage BYOD. We require both user and device identification in order to run our native HMI apps. Unregistered combinations will not work. This approach still allows companies to adopt BYOD if desired. Otherwise they can ban unidentified devices, not just users, by simply not registering them. Ultimately, we encourage companies to supply employees with devices (such as iPads) only for the purpose of process control, using HMIs, as opposed to allowing users to bring their own devices. John Lluch-Zorrilla, SweetWilliam, S.L., www.sweetwilliamsl.com

MOBILITY RULES A control system continually produces valuable system and production data. But that information—and that system—is less valuable if you can access machine information only from a dedicated terminal. If operators can be more productive by accessing terminal data from a remote location, or if an operator needs to do a quick unit count from an HMI, but isn’t near a computer or the production line, HMI mobility options should be tapped. While the majority of industrial production companies likely will continue to use traditional HMIs as their main control interface for machine operations, remote access

[From LinkedIn’s Automation Engineers Group, where we posted the question, comes this thread about the problem:] SECURITY THE BIGGER ISSUE I would be far more concerned about security issues 36

Trends in Technology

selves why do employees want to bring their own devices. To my understanding, this is mostly because they have more up-to-date, sexier devices than those provided by their employer. With the prices of the devices dropping like a bad habit, why not provide the employee with the devices as a tool with the permission to use it privately? If they leave the company, the device can be remotely wiped and they can keep it. The company is in control about what device, how to use the ecosystem and what security measures will be used. Basically you want an ecosystem surrounding the device that supports the development of third-party modules. Next you need a way to be able to download this onto the device in a secure manner. The app store principle goes a long way in supporting this way of working. Performance-wise, the devices get more powerful with every release. There is, however, a mindset that needs to be changed. This seems to be the hardest part of dealing with change. The general idea is that most of the time the device must be able to do exact the same things as the HMI or desktop application. This is wrong. The apps for the devices need to be developed for the way they will be used. Do you really stand still in your factory with a tablet in your hand trying to check up on all the thousands of I/ Os? Or do you just want to quickly check some KPIs and copy the results into your presentation? Where did I get my wisdom? Well partly from listening to several discussions and reading some magazines. I listened to a good discussion on this topic from a podcast by RunAsRadio: www.runasradio. com/default.aspx?showNum=335. The debate is not an easy one and also heavily influenced by personal opinion and context. Personally, I have not taken a position yet. Just keeping my eyes and ears open. Robert Saunders, owner Eye-Concept Industrial Automation BV, www.eyeconcept.nl

than performance issues. Often the security on a control system network consists of a firewall between the business and manufacturing network, but not much security within the manufacturing network. Allowing a consumer device onto a relatively unprotected manufacturing network is not advisable without additional protection against the device. Steve Boyko, senior PI specialist, ADM Systems Engineering, www.admse.com BUT THERE ARE SECURITY ANSWERS Some mobile platforms address security really well. For example: • Separate NICs on the server or device so you can separate the control network from the IT network; • SSL-encrypted messaging; • Authentication certificates; • Ability to come in from the outside via VPN; • Ability to layer user access with permissions; • Ability to limit some screens to monitoring only. Given these six points, mobile devices can be just as secure or more so than traditional HMIs and OIs. I think BYOD is inevitable, but like a lot of things in the industrial automation space, it won’t replace traditional solutions, but it has its place for the right applications. Arun Sinha, director of business development, Opto 22, www.opto22.com [From LinkedIn’s Industrial Automation Group:] WHO OWNS THE DATA? There is one compelling factor. The devices are getting dirt cheap at a staggering rate. The biggest concern about BYOD is around security, and not just the infection part with viruses. Who is the owner of the data stored on the device? There are different solutions for this problem and, of course, different vendors have different implementations. This is not optimal if you need to decide what you want to do or use. We should step back from the BYOD and ask our37

Trends in Technology

Give Your HMI an Ergonomic Tune-Up Ergonomic tune-ups make sure operators are safely paying attention to the increased amount of operating data without being too comfortable to be attentive By Control Design Staff

O

ur multi-machine workstations haven’t changed much beyond the sophistication of the HMI software and better tactile input methods. The operators now spend more time at the workstation and less time patrolling the machines with clipboards since we have much more operational data feedback at the HMI. We need an ergonomic tune-up to make sure the operators are safely paying attention to the increased amount of operating data without being too comfortable to be attentive. Any experiences to share?

HMI complexity. And regulatory standards must be considered to meet industry criteria. All of these aspects influence the design of the interface in order to capture user attention and to ensure safe operation. This is why a mixed-technologies approach is best. Not one single technology has the ability to provide an all-encompassing solution. Once the application and user requirements are defined, a mixture of push buttons, cursor controls, keyboards, touch technologies and interactive displays can interface with industrial computers to inform, alert and efficiently update the user of machinery functions. Use of illumination techniques such as ring, halo or animation combined with audible alerts capture end user attention in both an aesthetically pleasing, modern appearance and forthrightly effective manner. The mixing and matching of components and technologies allow for a consolidated user interface along with a central and sometimes singular point of data feedback. Dan DiGioa, marketing manager, EAO, www.eao.com

Answers MIXED TECHNOLOGIES This can be best accomplished by employing a mixed-technologies approach along with ergonomic design principles when creating an optimal user interface. By utilizing and implementing a design which incorporates all forms of human-machine interface (HMI), a complete and consolidated user experience can be accomplished. HMI systems have to be designed with the user and application environment in mind. You first have to define the operational and functional requirements. This can encompass durability requirements and environmental stresses including exposure to moisture, vandalism, temperature extremes, cleaning agents and general rough use. Operator feedback is critical to capture end-user attention and to ensure overall effectiveness and efficiency. Understanding the application can dictate the degree of

ERGONOMIC OUTLOOK It’s important to take the time to do an ergonomic tune-up. Ergonomics plays an important role in health, safety and productivity. Technological advances have overloaded operators with information, and their scope of responsibility is ever expanding. Many operators work long hours in less-than-ideal conditions. Well-designed control rooms balance 38

Trends in Technology

MACHINERY VISIBILITY As your question points out, even with sophisticated output from software, visibility to machinery is still key. In fact, lean manufacturing techniques have led to a shift in the use of control enclosures to allow visual contact between operations and cells. Luckily, there are a variety of HMI enclosure systems available beyond traditional, static workstations to help you create a solution specific to your business’s needs. Great examples of this are vertical motion and suspension systems, which allow operators to reposition equipment as necessary throughout their shifts. Look for a system that allows for multiple combinations of components for innovative system solutions for any work environment. For even more flexibility, you may want to investigate industrial tablets, which blend all of the benefits of modern technology with the ability to patrol machinery as you would with a traditional clipboard. Ultimately, there are more HMI enclosure systems out there than ever which can be tailored to fit your application. Emily Delozier, global product manager for large and HMI enclosure systems, Pentair, www.pentairprotect.com

productivity with ergonomics, and a key component is the operator console. The console is the bridge that connects the operator to the technology, and therefore, has a significant impact on performance. Sit/stand consoles are a great choice for an ergonomic upgrade. Recent studies have shown that too much sitting can be detrimental to your health. Movement throughout the day is important to maintaining good health. Alternating between sitting and standing is a healthy activity that increases energy and reduces fatigue. Sit/stand consoles also provide adjustability to meet the needs of each individual operator. The ergonomic standards outlined in ISO 11064 are a good reference for any ergonomic initiative. ISO 11064 standards are designed to improve efficiency and reduce human error in the control room. A good ergonomics program translates into significant ROI in terms of reduced healthcare costs, increased productivity and fewer errors. Beyond the numbers, ergonomics can improve the quality of life for workers. Operators who are more comfortable and better able to do their jobs find more satisfaction in their jobs, which improves the morale of the organization. Brent Leimer, marketing manager, Winsted, www.winsted.com

OPERATOR ERGONOMICS HMIs are available as free-standing operator stations, console stations or pedestal or support arm systems. Depending on the application and available floor space, appropriate solutions can be selected. Systems are available in aluminum, mild steel and stainless steel to address differing application, environment and aesthetic requirements. Many users are transitioning to the support arm with enclosure solution because it can be safely moved in and out of the workplace when programming or data acquisition has been completed. Sup39

Trends in Technology

port arm systems provide options for vertical motion, swinging motion and swiveling to move the HMI enclosure into the most ergonomic position for the operator to access. Considerations for operator ergonomic use, safety, floor space, environmental performance, weight load, heat management, aesthetics and cost should all be considered when selecting the correct HMI solution. Greg Quick, product manager, wallmounts and HMI, Rittal, www.rittal.us

OPERATOR EVOLUTION It sounds like the right direction—machine data processing evolution—but operator evolution may be a concern. As more data is collected, the machine should process more data, so the operator has to process less data. The direction of less human dependency in the process equals less human error and risk, and greater safety, reliability and repeatability. But at the same time, operator evolution must take place to guide the operator to perform new tasks with the new time that was freed up by machine evolution. Excitingly and interestingly, machine data collection will eventually evolve to collecting data about the human operators, too, so as to further reduce human error and increase safety. The human motion/ gesture sensing on games such as Xbox will be incorporated; later, even health vitals and retina scans will be incorporated. So the machine may sound an alarm if my human operator has fallen asleep, is drunk, is sick or is in the wrong place at the wrong time, that is, for the few machines in that distant future that still require a human to operate them. Don Fitchett, president, Business Industrial Network, www.bin95.com

DATA FLOOD An interesting quandary you envision. Most processes I have worked with still involve regular operator interaction to load parts and/or renewable supplies and provide regular quality checks, some as often as every 10 or 15 minutes, so I think that we have freed up the operator to focus on the process and the product, rather than boring him so much that he becomes inattentive. As interface designers, we gather and present more and more data, and our challenge is to crunch the data and present it as simplified choices the operator can make quickly, rather than flooding his screen with distracting raw data and expecting him to analyze it. My goal is to always reduce the amount of data and choices and operations in any given sequence an operator will have to make. For every decision or step of manual workflow that you can remove for an operator, you exponentially reduce the possible outcomes and effectively reduce risk by the same factor. Steve Meredith, reliability electrical coordinator, Corod Division at Weatherford, www.weatherford.com

AUTONOMY I believe automation today is and should move toward autonomous automation, where whole facilities can be run from a central SCADA position with multiple monitors observing multiple machines and maybe even multiple facility locations, simultaneously. Monty Bass, engineering & maintenance manager, Exide Technologies, www.exide.com 40

Trends in Technology

The Pros and Cons of Embedded HMIs For Machine Builders Should you consider an embedded HMI instead of a PC-based HMI or graphics terminal? Here are some reasons and cautions By Dan Hebert

M

achine and robot builder OEMs have been using embedded control for decades, and many now use embedded HMIs. These embedded platforms typically are purchased from a vendor in a package that includes the display with the embedded operating system, usually Windows, pre-installed. Also included is the HMI programming software, often free, and the right to run the software in run-time mode on the display. This is in contrast to PC-based HMI platforms that require OEMs to buy the software and the hardware separately. The other HMI option is a low-end graphics terminal with very basic functionality and limited flexibility. Should you consider an embedded HMI instead of a PC-based HMI or a graphics terminal to provide an operator interface for your machine or robot? Here are some reasons and some cautions. A simple graphics terminal is cost-effective and is usually adequate for basic machines with few inputs and outputs, but on more advanced and automated equipment, it could be insufficient in terms of operator interface, connectivity and data handling.

At the other end of the scale are PC-based HMI platforms—top of the line in terms of price and performance. Within this option are two approaches, each of which requires the user to buy a PC. The first is to buy off-the-shelf HMI software and configure it for the application, and the second is to write software using standard programming languages such as Visual Basic. Either option will provide a full-featured HMI, but with a few caveats.

Compatibility and Reliability Issues The PC will have a short lifecycle of just a few years, so if it fails in part or whole, it might not be possible to make replacements using the same technology. Upgrading the PC or its operating system can cause compatibility issues with the HMI software and sometimes with connected components Rick Lamb, president of Midwest Technology Ventures, a distributor and system integration firm, concurs. “PC-based Windows systems get difficult to change/upgrade/support after about five years because of obsolescence of the operating system, drivers, utilities or hardware. One piece breaks, and 41

Trends in Technology

you can’t get a replacement, and nothing is compatible with newer hardware or operating systems.” Because the HMI is a PC, probably with a Windows operating system, there can be software reliability issues, as well as the temptation to load additional software onto the PC to perform other activities, both work-related and personal. In between a PC-based HMI and a simple graphics terminal with respect to price and performance is an embedded HMI, which can be the best solution in many machine control and robotic applications. Modern embedded HMIs provide a wide range of capabilities, flexibility and connectivity. “We provide an extremely easy-to-use software toolkit containing a very rich feature set that hardware manufacturers, OEMs and vertical industries use in their products,” says Richard Clark, an engineer at InduSoft, a supplier of HMI and SCADA software. “An embedded HMI configuration can be designed for many types of equipment, and can host a variety of external features such as thin-client server, web server, database access, a variety of I/O drivers and third-party reporting tools,” Clark adds.

“Such a configuration provides the flexibility required by machine and robot designers.” In many cases, a machine builder can get by with an embedded HMI instead of a PC-based HMI, resulting in substantial savings. For example, a vision-guided robotic system typically would have three controllers: a PLC, a robot controller and a camera controller. Most of the data exchange among the HMI and controllers would be discrete or floating point and typically limited to simple values such as setpoints, part numbers, lot numbers and measurement data. The HMI would have to display, collect and report machine status and alarms and configure functions. An embedded HMI fits well here, and if it fails, it’s simpler to replace than a PC-based HMI. When selecting an embedded HMI, you should consider the application, features, future expansion and customer requirements. You’ll find a variety of available embedded HMIs in cost-effective packages from a wide range of suppliers, and these expandable, customizable HMI solutions quite often are the best option to meet your operator interface requirements.

42

Trends in Technology

Machine Information in Your Hand One handheld HMI could be used to monitor and adjust many different machines By Dan Hebert, PE, Senior Technical Editor

A

at either the controller or the HMI level. This takes care of the hardware connection, but doesn’t address the software protocol issue. Most manufacturers use one or more Ethernet protocols in their plants, and it’s incumbent on the machine builder to ascertain which type of protocol is needed and to provide the machine with same. Although many companies use handheld HMIs for machine monitoring, few are using them to replace the primary machine-mounted HMI. The handheld HMIs are instead used to extend reach, usually via one-way communication of machine status. Any problems requiring adjustments to the machine control system typically still will be made at the machine. Future implementations might feature high-speed twoway access. This would allow your customers not only to monitor their machines remotely, but also to change control parameters to adjust machine operation. These types of adjustments probably will be made by operators in close proximity to the machine, paving the way to systems that completely eliminate on-machine HMIs. In these types of systems, your machine would be provided with no local HMI. Instead, a handheld wireless HMI would be used to provide full monitoring of your machine along with adjustment of machine operating parameters. The advantages of such a system to manufacturers are numerous. First, one handheld HMI could be used to monitor and adjust many different machines. In typical plant operations, an operator goes from one machine to the next to observe operating conditions and make adjustments. Second, an operator could access a machine from a safe distance, often outside a hazardous area. This not only could improve safety, but also save time as personal entrance into a hazardous area is often a time-consuming task requiring special protective personnel equipment (PPE) and lots of paperwork. Third, linking the handheld HMI to the machine and to the central control room could allow an operator to make machine adjustments with the entire process in mind.

s uour customers use handheld wireless human-machine interface (HMI) units more frequently, it’s important that you understand how best to make your machines fit their overall wireless infrastructures. Increasingly, compatibility with existing and planned wireless machine monitoring systems will be a required feature. For example, Mohawk Fine Papers (www.mohawkpaper.com) in Cohoes, N.Y., uses Transpara’s Visual KPI to monitor and control its paper machinery and other plant components. Key performance indicators (KPIs) are delivered to handheld HMIs, in this case Blackberry devices. Data isn’t accessed directly from machines, but instead primarily through OSIsoft’s PI data historian. Other data are delivered to the Blackberry devices through Microsoft SQL Server for data extraction, transformation and loading, and through Microsoft Sharepoint as an information portal. Everyone in the plant has access to the data via their Blackberry devices including technicians, managers, supervisors, engineers and even the chief operating officer. Handheld HMIs played an important role in Mohawk’s overall 2009 improvement in machine output, customer satisfaction and energy consumption. “Supervisors and senior managers now have real-time access to machine, production and order status,” explains Ben Whitaker, manager of enterprise process reengineering at Mohawk. “This allows more responsiveness to customer requirements and manufacturing issues. In the maintenance area, supervisors and senior engineers have access to energy consumption for better response to machine performance issues.” Machine OEMs supplying Mohawk and companies using similar systems need to make their machine control system accessible to the data repositories accessed by handheld HMIs. This means that links must be provided to software such as data historians and often to various Microsoft products as well. These links are most commonly Ethernet-based, meaning that your machine should have an Ethernet port 43

Trends in Technology

Cooling Complications In Hazardous Locations Are we better off with conventional gas-purge systems or another approach? By Control Design Staff

W

e build most of our custom, one-off control cabinets in-house. We’ve had success using vortex coolers to control the enclosure temperatures when we have heat issues. Now we’re getting into markets that place our systems in hazardous environments. What’s the experience with adding purging capabilities to vortex coolers? Are we better off with conventional gas-purge systems or another approach? We need an effective, but still cost-responsible solution. —from July 2014 Control Design

your application, we have systems that can monitor the temperature inside of the enclosure, and turn the vortex cooler on or off depending on the temperature settings entered into our purge unit. So if you have an enclosure in a hazardous location, and it has a vortex cooler, then using a purge system is normally more cost-effective, and you get all of the extra benefits of the temperature monitoring and pressure compensation features. Brent Dean, product engineer, Pepperl+Fuchs, www.pepperl-fuchs.us

Answers

[We received the following responses after posting the question on LinkedIn’s Industrial Automation and Process Controls Network.] PURGE WILL HAVE TO WORK HARDER Just use instrument air that is dry and filtered for the source of air for the vortex cooler. This would be the same source used for a purge system. The purpose of the purge is to actually pressurize the enclosure to prevent the ingress of ignitable vapors. The vortex cooler will cause the purge to work harder, since it will have to make up pressure (just a few inches of water column) to compensate for the air that is constantly flowing through the enclosure via the vortex cooler. Contact your purge system vendor to ensure the

A COMMON SOLUTION This is a very common solution for us. Most of the time customers prefer to use a purge system to protect their equipment over the use of an explosion-proof enclosure due to the large cost. The rating of the hazardous location that the enclosure is located in will determine which type of purge system you would need. One concern customers have when adding a vortex cooler or AC unit to a purge system is over-pressurizing the enclosure. You just have to make sure that the vent used on the system can handle the combined air flow of the purge system and vortex cooler. Depending on 44

Trends in Technology

are in their own hazardous-zone-rated enclosures, and you add a purge of sorts, also rated for hazardous zones, then you stand a chance of your entire installation being suitable for the hazardous-zone installation. If not, then I suggest that you just keep your enclosure out of the hazardous zones. Work with the site to determine your non-hazardous areas. They should have classification drawings that can guide you to where they are. Andrew McKeown, specialist process engineer, Carter Holt Harvey, www.chh.com

purge has the capacity needed to offset the demand of the vortex cooler. If possible, use all Div 2/Zone 2-rated components if you’re placing the enclosure in a Div 1/Zone 1 environment. This will permit you to use a less-expensive Y purge instead of an X purge. Robert Burgman, principal engineer, electrical, instrumentation and process controls, Ashland, www.ashland.com

GOOD COMBINATION Is this for purging systems such as standard atmosphere or concentrated atmosphere such as nitrogen? General positive pressure designs would strike me as a great vortex combo application as long as you’re not pushing moist air in. Any other gas would seem questionable. An exact application would help me answer with specifics. Theodore Sharpe, machining group supervisor, Ford Motor, www.linkedin.com/pub/theodore-sharpe/43/160/379

PURGE CONTROLLER HANDLES BOTH We recently provided modifications to an existing customer panel. From the beginning, the customer wanted both a vortex cooler and purge controller. Once the final installation was completed, our mutual conclusion was the purge controller could provide both functions—cooling and safety purge for C1/D1 area. Jim Robinson, control systems engineer, M.G. Newell, www.mgnewell.com

MAYBE MOVE THE ENCLOSURE If you’re a one-off manufacturer of an enclosure, you will find it difficult to satisfy the regulatory bodies that the enclosure meets the required rating for use in hazardous zones. Simply adding a purge is not sufficient. You have to contain or limit energy. Allowing the internals of your cabinet to vent, and thus forcing any energy out of the cabinet into the hazardous zone, is not acceptable. If all of your components and equipment

[We received these responses after posting the question on LinkedIn’s Automation Engineers Group.] PURGE + VORTEX Pepperl+Fuchs (Bebco) makes purge systems that can work in combination with vortex coolers. We use them at my shop, and it is a good product. Noel Jull, I&C designer, IDT Systems, www.idtsys.com 45

Trends in Technology

VORTEX GULPS AIR Vortex coolers place quite a demand on an air system. I would use a purged enclosure per NFPA 496, and then use something like a Noren air-to-water heat exchanger approved for use with purged enclosures. This is how I’ve done it before. Brian Rogers, senior electrical engineer, Transfuels, www.blulng.com

[We received this response after posting the question on LinkedIn’s Automation.com Group.] VORTEX + EX-N On previous projects, per customer’s requirement, we supplied IP65 stainless-steel panels with vortex coolers and Ex-n components. This is not a cheap solution, but allows operating the cabinets with the doors open in any situation, including the presence of gas, not needing purging or de-powering. George Eric Wootton, E&A solutions & engineering manager, Wärtsilä Brasil, www.wartsila.com

VORTEX ALONE WORKS I have no experience using a purge system in conjunction with a Vortec unit, but Vortec does make two units specifically for hazardous locations, the HazLoc Vortex A/C. Here are links to the operating and safety instruction sheets: bit.ly/1AVkdsb and bit.ly/1mTNknF. They have some useful information regarding sizing and using the two systems together. Fred Manning, pneumatic product manager, FCI Automation, www.fciautomation.com

[We received the following response after posting the question on LinkedIn’s Panel Building and System Integration Group.] GO AHEAD If you’re asking can you use both a vortex cooler and a purge system for your hazardous location requirement, the answer is yes. It’s not an uncommon practice. If your heat dissipation is too great, explosion-protected air conditioning units can be used effectively. If heat is a smaller concern, sometimes just using continuous purge (versus compensative) can assist. Hope that helps. Craig Yoss, director of marketing and business development, R. Stahl, www.rstahl.com

THEY’LL WORK INDEPENDENTLY You can use the vortex cooler as part of purging systems, but you need to be aware of the fact that all the components that make up the purging system will operate independently from the vortex. Also, you should use air where possible, since placing cooling or purge gases in contact with people could harm them. Alejandro Varga Meder, project/construction management, Devco USA, www.devcousa.com 46

Trends in Technology It’s Free, You Say, in the Industrial Community? Cloud-based programming using free tools doesn’t give you the same protection as if you had purchased the environment By Jeremy Pollard

S

are two additional operating systems that are free to the user, with free development tools, which leads us to the marketing platform of allowing the user apps to dictate the development environments, something Apple figured out a long time ago. Databases are usually MySQL or Microsoft’s SQLExpress —both are free. MongoDB is a leader in the NoSQL database race, which provides scalable and balanced platforms for document management. It’s automatic and takes the hair-pulling out of the database management. It uses a browser as its interface. Firefox, Chrome, Safari and Internet Explorer are all free. A free user interface running HTTP5 and cascaded style sheets and maybe some AJX and SOAP code with some PHP thrown in delivers a stunning application. All at zero cost, except for development, of course. Check out ObjectRocket.com. Facebook infected our minds as well as providing an unfettered landing spot for user comments and a way for companies to track individuals. I am not an old school guy, but I find the online stuff has only just reached the point where the reading is easy. It’s only a matter of time, I’m sure, that all content will be pushed based on our profile(s). I recently got introduced to Rockwell’s Connected Components Workbench for its Micro line of devices. It is developed using Visual Studio, uses VS runtimes, and sometimes just feels like a shell extension of the development environment. The application is free. The hardware platform won’t control a paper mill, but for the most part, it acts like a commodity. Teamviewer is the de facto standard for remote access. I use it because my customer base is local, and the security level isn’t all that important. But for state-wide access, one would be wise to be more vigilant, since all traffic goes through a server in Germany and other parts of the world. But it’s free. So what is it we should pay for? That’s my question. That 40-client application cost my customer around $100,000 over seven years. He can add 40 more clients for nothing. But what are we giving up for free? Is it really worth the aggravation or support or maintenance of such systems? Let me know what you think.

ocial media, open-source, back-door daemons, clouds, virtualization and COTS—are they a part of our beloved industrial community? You betcha, but where do they show up or, more importantly, where are they going to show up, and how many current companies will fall under the knife of “free?” Heartbleed brought fear into the open-source world, and so it should. So is it beneficial for a hardware company such as Rockwell or Siemens to employ and promote open source? One wonders. I recently priced out an existing application using a SCADA system provided by a major vendor—the application had two servers and 40 clients. I wrote all the software for this application(s) over the past five years using Visual Studio—the free version. To be clear, in 2014, this application would have cost more than $500,000 to license and implement. With free development tools and graphics, all I had to do was put the screen together and animate. Well, sort of, but you get the drift. There is now a version that really is cloud-based, so you can create and compile for the Windows platform on the fly. I haven’t used it, but the opportunity is there, and I wonder if there are companies that have developed applications using free software, and then turned around and sold the applications as “home-grown.” See, the issue is that cloud-based programming using free tools doesn’t give you the same protection as if you had purchased the environment. Most products once used a proprietary operating system. In the old days, the graphics were generated by the hardware and the OS. I remember Steve Rubin, founder of Intellution, which is now a part of GE Intelligent Platforms, telling me about how his wife burned UVProms with the graphics character sets for the software on the kitchen table after Sunday dinner. Now those images are almost free. And they’re much prettier. Embedded systems typically use a form of Linux, maybe with some FPGAs, but mainly a no-cost OS. Android and Chrome 47

Trends in Technology

HMI: Form vs. Function? Are industrial screens beginning to look more like smartphones? By Control Design Staff

A

long with the growing use of smartphones and other mobile devices used for remote HMI monitoring functions, we note that many operators and technicians are more comfortable with the touchscreens on a smartphone than on a conventional HMI. Are there starting to be deployments of industrial HMI screens that look more like phones on the shop floors? —From January 2013 Control Design

younger generation of workers. In my opinion, that alone is not enough to dictate an interface style, and it discounts one advantage that younger generations seem to have compared to us geezers: the ability to rapidly adapt to new (to them) electronic interfaces like those used on smartphones and industrial HMIs. Kim L. Ground, senior EE — controls, Surface Finishing Technologies IMAGINE THE POSSIBILITIES Ten years ago, this was the subject of obscure lab experiments and generally garnered laughter and ideas of science fiction. I certainly can envision deployment of smart devices in the industrial environment in the very near future. “Smartphone” is a misnomer as the “phone” part is now minuscule by comparison to all the other functions. Applications such as Aurasma, an augmented reality tool described by Matt Mills in a TED presentation, clearly demonstrate how easy this deployment can be. Imagine a production line tech pointing his cellphone camera at a piece of equipment on the line, and immediately the setup procedures, documentation, etc., appear on the screen. Why not use these smart devices on the line

Answers FUNCTIONALITY IS WHAT MATTERS HMIs used in the machines built by our organization have designs driven by functionality, not style. My experience with smartphones and the Windows 8 Metro interface indicates that they are perhaps adequate for entertainment, but are much less user-friendly for any application requiring serious computing or significant data entry. Nobody reads the manual unless they have a problem, and seldom even then, so if your interface requires an explanation in a manual before the user can operate it, the battle is already lost. The argument might be made that operator interfaces should take on the appearance of smartphones to make them more understandable to a 48

Trends in Technology

chine operator interfaces on the machine itself by making them much more intuitive to interact with. The main contributor to this revolution will be multi-touch technology, which already has had a great impact on how we interact with our mobile devices. Additionally, displays themselves will get larger as pricing continues to decrease, thanks to lower component costs. Supporting swipe and pinch and zoom gestures on larger displays allows designers of machine operator interfaces much more freedom to create a user experience that is richer in terms of graphics and content shown on the screen. In combination with the underlying control system gaining ability to show more production and diagnostic data, the need for special separate engineering and diagnostic tools will decrease, which will reduce training and maintenance costs for technicians and operators, lowering the total cost of ownership of machinery and boosting profitability. Robert Muehlfellner, director, automation technology, B&R Industrial Automation

rather than leaving them in the employee’s clothes locker? He is already totally familiar with the interface. Minimal training required. Gerald Beaudoin, automation project manager, Leahy Orchards

HIGH-PERFORMANCE HMI Industrial HMI screens are starting to take advantage of the usability research that has gone into making smartphones easy to operate. Apple and Google have invested in extensive research and testing to create simple, consistent and easy-to-use interfaces for iOS and Android operating systems, and both companies publish usability guidelines for developers creating iOS and Android apps. Beyond HMI screens, smartphone interfaces are arguably also influencing desktop PC user interfaces. Microsoft Windows 8 is a prime example. Smartphone usability guidelines and practices are beginning to be implemented in HMI screen design. This direction is particularly evident in a series of best practices for building effective HMI screens — called “high-performance HMI” — that emphasizes prioritizing on-screen data, using more informative graphics, and muting colors so key data stand out. High-performance HMI is described in The High-Performance HMI Handbook by Bill Hollifield and Eddie Habibi of PAS. Selam Shimelash, application engineer, Opto 22

USER EXPERIENCE KEY User experience is becoming a more important requirement in manufacturing automation. Users want interfaces similar to what they experience on smartphones and tablets. This is further complicated by the proliferation of available real-time data. We also need to provide a user interface that can leverage information from MES and IT systems in an intuitive manner. Scott A. Miller, business manager, Rockwell Automation

RICHER USER EXPERIENCE In addition to an increase in mobile connectivity (for the shop floor), we will also see a revolution in the ma49