June - 2008 - issue > Technology
Swaminathan Kumaragurubaran
Friday, May 30, 2008
Imagine the exhilaration of speeding on the German Autobahn where there are no speed limits at most stretches. You feel the acceleration all along until the speed limiter announces its presence loudly indicating that you have reached the electronically set speed limit of 250 km/hr. Exhilaration turns to apprehension as you start seeing a trail of red lights from the rapidly decelerating traffic in front, you feel the seat belts tighten as the brakes are applied, the ABS kicks in and the car stops where it has to – a few meters before the car in front.

The functionality, performance, and reliability of every product need to be tested to know how the product or device would perform in real world use. Test and Measurement Engineering focuses on this part of a product’s lifecycle. Thanks to the advancements in Test and Measurement Engineering, the cars we drive today are safer, our cell phones are reliable, and quality of products is enhanced across industry.

Test and Measurement Fundamentals
Typically, testing a product involves measuring the states and responses of various physical parameters. It is also required to control states, conditions, and responses of stimuli used to simulate real world conditions. Thousands of measurements are made in each stage of testing and a sequence of tests is carried out to simulate various real world conditions. Then the test data is analyzed to ensure compliance to various standards of design, quality, performance, reliability, and statutory requirements.

Each stage of testing would involve interfacing the product via electrical, mechanical, and software interfaces to measure the states and responses of various physical parameters and ensuring that they are within the required specifications. Most of the physical parameters such as speed, force, torque, temperature, stress, voltage, current, amplitude, frequency, power and phase noise are typically measured by using appropriate sensors that convert physical response to an electrical signal. This electrical signal is digitized by a data acquisition device. The data is available for further analysis in various forms depending on the technology used.

Typically, a measurement instrument is characterized by parameters such as its range (lower and upper limits of measurement), resolution (how small a change it can measure), accuracy (how close to calibrated value it can measure), repeatability (how repeatable it is at various measurement ranges), and speed (how fast it can sample a measurement).

Virtual Instrumentation
Traditional measurement instruments were mostly application specific devices with a display to show the measured values. The oscilloscopes and multimeters that are used in laboratories are classic examples of this type. The advanced versions of these came with a computer interface via a digital interface bus such as the GPIB (General Purpose Interface Bus), which was originally developed by HP in the late 1960s when it was widely known as the HP-IB. Conventional instrumentation was hardware-centric, and was typically meant for a specific measurement application.

With the phenomenal growth in computing power, newer technologies emerged in the test and measurement domain. Software took the center stage and spawned a new class of instrumentation known as ’virtual instrumentation’. Unlike a vendor-defined instrument, a virtual instrument is user defined, and typically is a combination of modular hardware, good computing power, and powerful software that controls the entire functionality. The modularity of virtual instrumentation enables the user to change a specific component and leverage the rest of the components for an entirely different application, providing flexibility that cannot be matched by other technology in measurement. Since virtual instrumentation is software centric, it enables a wide range of applications, greatly improves productivity, provides powerful methods of analyzing and displaying information, excellent connectivity, and contributes to lowering costs of testing products.

Enablers of Virtual Instrumentation
The modular hardware used in virtual instrumentation, which is available for a wide range of measurement applications, includes data acquisition cards that acquire electrical signals from sensors, high speed digitizers that acquire signals at millions of samples per second, RF instruments that acquire radio frequency signals, motion control, sound and vibration analyzers, and image acquisition modules. These modules are available from a variety of manufacturers such as National Instruments, Agilent, Tektronix, Keithley, and others. While conventional software such as VC++ and Visual Basic are used to develop test and measurement applications, the leader in this category is LabVIEW from National Instruments. LabVIEW has excellent features and tools that make it ideally suited for developing the engineering applications for test automation and control.

Developing a Virtual Instrumentation Solution
The interplay of various engineering disciplines makes developing a solution for testing a product a delightful engineering challenge. It requires strong engineering fundamentals, a quick understanding of the product being tested, good knowledge of measurement and control technologies, and excellent software engineering skills. The software is typically developed as a deployable application. After development of software, it is tested with the hardware, integrated with sensors and signals and deployed as a system. Most solutions require connectivity to a storage network, database, and secured availability of information at multiple locations via the internet.

Virtual instrumentation systems find wide range of applications across the industry spectrum, primarily to test functionality, performance, and reliability. This technology is also used in industrial control.

The automotive industry uses this technology to test engines, electronic ECUs (engine control unit), drive trains, brakes, tires, horns, vehicle electronics, cable harnesses, instrument panels, motors, seat belts, and almost every other component in the car. In addition to component level testing, system level tests such as NVH (Noise, Vibration, and Harshness), crash tests, ride quality tests, on road tests, etc. are carried out to check the performance of the vehicle as a whole. Some Formula 1 racing teams have extensively adopted this technology to test and validate the performance of their cars.

Apart from the automotive industry, the military and aerospace industry uses this to test commercial and military aircraft, and to conduct various missile tests. Medical equipment such as ultrasound machines, CT scanners, MRI tables, and X-ray machines are tested worldwide using this technology. The semiconductor industry uses virtual instrumentation for various measurement and control applications. Electronic devices such as mobile phones, LCD TVs, X-Boxes, and i-Pods are tested for various parameters including signal quality, audio quality, video quality, display defects, compliance to standards, and mechanical reliability. Research and Development labs worldwide utilize virtual instrumentation technology to validate their designs and prototypes by conducting reliability tests, highly accelerated stress tests (HAST), and accelerated life tests (ALT). In addition, virtual instrumentation also enables visual inspection that helps in detecting physical defects that cannot be determined by electrical measurements.

The recent trends in virtual instrumentation are a reflection of the advancements in computing technology. Compact sizes, highly rugged designs, portable displays such as PDAs, and FPGA based reconfigurable hardware are some of the recent advancements in this domain that promise to take the realm of virtual instrumentation to newer application areas at even lower costs. Advanced modular hardware today incorporate field programmable gate arrays (FPGA) as a core component that enables the user to reconfigure the hardware functionality and performance to address highly sophisticated requirements. Evolution in software technology has enabled systems to run on real time and within embedded targets. Apart from test and measurement applications, this technology is becoming increasingly popular in graphical embedded design and industrial control applications.

The author is General Manager, Soliton Technologies
Share on LinkedIn

Previous Magazine Editions