Ethernet networks and the Web have revolutionized how we communicate information and distribute tasks throughout a company. Incorporating test and measurement data into the network improves access to information and allows the organization to make decisions effectively.
With data sharing, the automotive and telecom industries are improving product functionality and quality while simultaneously decreasing product design and development time. Both of these industries share information among departments to avoid repeating tests, data collection, and data analysis.
In the automotive area, for example, the design time for a new model car has been reduced from 48 to 24 months, and the number of defects reported has fallen from 200 per car to 48 in just seven years. Automakers achieved this by sharing data between computer-aided design (CAD) and computer-aided manufacturing (CAM) software.
The Web
The Web has changed the way businesses operate by providing a new way to deliver data, connect processes and measurement systems, and conduct business-to-business transactions. The Web often refers to a broad spectrum of tools that facilitates the exchange of information. These tools include TCP/IP, Web browsers, and HTML. When correctly deployed, these tools can give companies the power to change the way they design, manufacture, and test products.
Networked Measurement Systems
Today, most organizations have invested in a large network infrastructure to access corporate databases or share e-mail or Web pages. Just as the network has become indispensable in the office, networked measurement systems are finding their place in design and production. As the network improves productivity by giving office workers ready access to information, the same infrastructure helps make designers and manufacturing engineers more productive by sharing reports and live data.
Most measurement systems include a stand-alone computer connected to measurement devices through a bus such as PCI, GPIB, USB, or RS-232/485. These measurement systems feature traditional GPIB rack-and-stack instruments and PC-based data acquisition boards. In all cases, these systems can be added quickly to a network via a network card or Ethernet-to-local bus converter.
Software packages such as Enterprise Resource Planning (ERP) and Materials Resource Planning (MRP) require information from the manufacturing floor to manage resources. Operations include datalogging, generating reports through Microsoft Word and Excel, and performing off-line analysis in MathWorks MATLAB or similar packages. Once this information is shared via the network, it can be connected easily to the Internet and Web browsers.
Companies implementing networked measurement systems most often develop them to reduce equipment downtime. Many are used for monitoring and emergency shutdown.
For example, ECD Solutions, with the help of Data Science Automation, provides remote access and monitoring to desolate oil-drilling sites throughout the world. The measurement system runs continually for weeks and months. By monitoring the performance of drilling operations remotely, ECD can observe trends and predict failures. This saves hundreds of thousands of dollars in downtime and travel costs since experts do not have to be located at the remote locations.
ECD now has an unforeseen benefit. By monitoring from one location, the company can collect and record data from many drilling sites. Drilling operations are optimized by spotting trends in equipment wear and planning maintenance schedules.
Improving Efficiency and Quality
There are three ways to share information within a networked measurement system:
- The first and easiest way is through online reports. This is the electronic version of the traditional printed report.
- While a report is a static view of the information being presented, some applications may require streaming the actual data to another application for processing, storing, or monitoring.
- The third way uses what is perhaps one of the most significant advantages of a network: the capability to harness the computational power of many computers connected together to share the execution of an application, known as distributed execution. For example, while one computer may sit in a hazardous or remote location acquiring data, another computer on the network can perform the data analysis and reduction. The results of that application then can be logged to a database on a third computer—a server.
Online and Web Reports
Many instruments available today can be configured or their results viewed through a Web page. These instruments are designed with Ethernet ports and Web servers and can be added to the network infrastructure. Displaying the raw data an instrument provides may be useful, but the power of Web reporting becomes apparent when that data is transformed into analysis, graphing, and trend comparisons before being published.
It is very easy to integrate Web report generation into instrumentation and analysis software. Products such as National Instruments’ LabVIEW and MATLAB can quickly publish reports to the Web. Excel is one of the most popular and easy-to-use tools for generating Web reports.
Streaming Data
In many applications, real-time access to acquired data is needed to control or monitor a process or test. The results of one measurement or automation process can be passed directly into the next process or any monitoring task that requires live streaming of data.
Software that allows you to stream information between applications includes Real Networks’ Real Audio and National Instruments’ DataSocket for measurement and automation applications. For example, Oak Ridge National Laboratories uses data streaming to remotely monitor experiments over the Internet. The lab monitors emissions from industrial stacks. The measurement system consists of a video camera to view the test bed and instruments for gathering temperature, humidity, pH, flow, and gas concentration.
The monitoring application accepts streaming video from the test bed while receiving real-time updates of instrument readings. The monitoring application also can send new set-point and control data directly to the remote test site.
Distributed Execution
With network access to various measurement systems, you can develop software that uses each computer on the network to complete a portion of the application. Using distributed execution, you can remotely operate another computer to control the execution of the application.
For example, Ford Motor Company uses distributed execution to control and monitor in-vehicle testing in harsh environmental conditions. Part of Ford’s test requirements includes 12 to 14 hours of vehicle testing in triple-digit temperatures in Arizona. Requiring personnel to monitor and change tests in this environment is not practical. Instead, cellular modems allow test engineers to reconfigure systems and obtain results remotely.
Another, and perhaps a more popular, example of distributed measurements is the SETI@Home project. The Search for Extraterrestrial Intelligence (SETI) organization collects radio telescope data from locations around the world. It has developed a screen saver to analyze the data to look for patterns that may represent signals from an intelligent civilization.
Each person who downloads SETI@Home uses a computer to analyze data from the radio telescope when the screen saver is active. This application distributes the analysis execution across all the computers that have installed the software. Since April 1999, the SETI@Home project has collectively executed 350,000 years of CPU time—far more than SETI could have processed without the distributed application. For large data sets, distributed execution can provide significant time savings.
Conclusion
The Ethernet and Web allow you to share information and create networked measurement systems. By using online report generation, streaming measurement data, and distributed execution, you can save time and increase product quality by integrating engineering processes with measurement information.
About the Author
Brian Sierer is the LabVIEW product manager at National Instruments. Before his involvement with LabVIEW, Mr. Sierer oversaw the R&D of hardware configurations, OPC, FieldPoint, and instrument driver software. He received a B.S. in electrical engineering from Virginia Tech. National Instruments, 11500 N. Mopac Expressway, Austin, TX 78759, 512-683-5339, e-mail: [email protected].
Return to EE Home Page
Published by EE-Evaluation Engineering
All contents © 2000 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.
September 2000