Temperature Effects in Strain Measurement

Temperature adversely impacts strain measurements in many ways, though three are of primary concern:
• The device or object studied will almost always have a non-zero coefficient of thermal expansion. Unless compensated for, changes in temperature will cause the item to which the strain gauge is attached to expand or contract, which is then indicated as a change in strain.
• The materials of the strain gauge itself have a non-zero coefficient of thermal expansion. Changes in temperature will cause the strain gauge itself to expand or contract, independent of any strain in the part to which it is attached.
• The wiring and the strain gauge itself will have a non-zero Temperature Coefficient of Resistance. That is, as the temperature changes, the resistance of the strain gauge and connecting wires will change independently of any change in strain. (For example, copper wire resistance changes at approximately 3,900 ppm per °C (.393% /°C).)

Some texts treat the first two items as the same effect. After all, if the coefficients of expansion of the gauge and the item under test are the same, they will contract or expand at the same rates in response to a temperature change. In this case, a change in system temperature would not cause any change in the indicated strain, except that based on the gauge’s temperature coefficient of resistance.

It’s important to note that in some applications, it may be desirable or even critical that strain induced by temperature changes be noted. Imagine an application where a “hot section” turbine blade is being tested to ensure proper clearance between the blade tip and the surrounding shroud. It’s important to know how much the blade has elongated based upon temperature in addition to the centrifugal force of rotation. On the other hand, if the parameter of interest is really stress, or its close relative, force, any strain caused by temperature changes would induce a true error in the result. A strain gauge used to measure the “g” forces on a supersonic aircraft wing skin might see temperatures from -45°C to +200°C. If the g-force information was critical to not overstressing the wing, you’d certainly not want significant temperature-induced error. In a more simple case, the load cell used to measure the force placed on a postal scale should not induce errors simply because the scale is next to the window on a sunny summer day!

Most applications fall into the second category, where the key measurement parameter is really stress, and the ideal system would be not to recognize any changes caused by thermal expansion or contraction. Like most engineering challenges, there is more than one way to skin this proverbial cat. They are: (1) Calculate the error and eliminate it mathematically, (2) Match the strain gauge to the part, (3) Use an identical strain gauge in another leg of the bridge. We’ve previously covered how to eliminate it mathematically, so lets take a closer look at choices 2 and 3.

Match the Strain Gauge to the Part Tested
The use of different alloys/metals allows manufacturers to provide strain gauges designed to match the thermal expansion/contraction behavior of a wide variety of materials commonly subject to strain (and stress) testing. This type of gauge is referred to as a “Self Temperature Compensated” (or STC) strain gauge. These STC gauges are available from a variety of manufacturers and are specified for use on a wide assortment of part materials. As you might imagine, the more common a metal, the better the chances are there is an STC gage that matches. However, you may count on being able to find a good match for such materials as aluminum, brass, cast iron, copper, carbon steel, stainless steel, titanium and many more. Though the match between the STC gauge and the part under test may not be perfect, it will typically be accurate enough from freezing to well past the boiling point of water. For more details on the precise accuracy to expect, you should contact your strain gauge manufacturer.

Use an Identical Strain Gauge in Another Leg of the Bridge
Due to the ratiometric nature of the Wheatstone bridge, a second, unstrained gauge (often referred to as a “dummy” gauge) placed in another leg of the bridge will compensate for temperature induced strain. Note that the dummy gauge should be identical to the “measuring” gauge and should be subject to the same environment.

Strain gauges tend to be small, and have short thermal time constants (i.e., their temperature changes very quickly in response to a temperature change around them), while the part under test may have substantial thermal mass and may change temperature slowly. For this reason, it is good practice to mount the dummy gauge adjacent to gauge being measured. However, it should be attached in such a way as not to be subjected to the induced strain of the tested part. In some cases, with relatively thin subjects and when measuring bending strain (as opposed to pure tensile or compressive strain), it may be possible to mount the dummy gauge on the opposite side of a bar or beam. In this case, the temperature impact of the gauges is eliminated and the scale factor of the output is effectively doubled.

Quarter, Half and Full Bridges
Strain gauges and measurement devices based upon strain gauges (e.g., load cells) can be configured in three different configurations. These are referred to as Quarter, Half and Full Bridges.

Quarter Bridge Strain Gauge

The quarter bridge gauge shown above is the simplest and probably most common strain gauge con-figuration (though some devices “based” on strain gauges are more likely to be provided in half or full bridge). The name “quarter” comes from the fact that in this configuration, the strain gauge represents one out of four, or one quarter of the resistors in the Wheatstone bridge. In this configuration, the user must supply the other three resistors.

Half Bridge Strain Guage

In the half-bridge configuration, two resistors or half of the bridge are provided in the strain gauge itself. Half Bridge configurations have two advantages over the single bridge. First, they simply require the user to provide one less resistor. Second and more important, however, is the fact that most half bridge sensors automatically provide temperature compensation, made possible by having two identical gauges in the same side of the bridge.

As you might expect, the full bridge sensor shown below provides all four resistors, in effect, providing the entire bridge. All the measurement system needs to provide is an excitation voltage and a differential analog input. Like the half-bridge configuration, most full bridge gauges are temperature compensated.

Learn more about measuring output from a strain gauge.

Full Bridge Strain Guage

All Systems Go at AUTOTESTCON 2013

United Electronic Industries – Booth #501 at AUTOTESTCON 2013 featuring the new MIL-CUBE DAQ platform!

Be sure to stop by to see us at AUTOTESTCON Booth #501 at the Schaumburg Hotel & Convention Center. We’re introducing the MIL-CUBE, a highly advanced data acquisition and intelligent I/O platform that’s designed to be compliant with MIL-STD-810 and MIL-STD-461 (power supply requirements). In contrast to some of our neighboring booths, ours features no bells, no whistles, just the best DAQ systems available and a couple of engineering experts to answer your questions!

UEIDAQ.com | Facebook | LinkedIn | Youtube

High Resolution Does Not Guarantee High Accuracy in Data Acquisition Systems – Part 1

Data Acquisition (DAQ) system accuracy is often associated with resolution, but they’re not the same specification. Just because an A/D input can resolve a one microvolt signal doesn’t mean the input is accurate to one microvolt. In fact, it’s common that systems with sub-microvolt resolution specify overall accuracy on the order of millivolts. Take for example a 24-bit audio input with an input range of ±2 volts and 0.238 microvolt resolution. It might only provide overall DC accuracy of ±20 millivolts. And, while 20 millivolts of accuracy may be suitable for most audio applications, it is not likely to be good enough for precision temperature or vibration measurement systems.

The key is to remember that high resolution does not always ensure high accuracy. Resolution is specified in bits; accuracy specifications are provided in a variety of ways. Depending on the application, some accuracy (i.e., error) specifications might provide more insight than others. An in-depth discussion of input accuracy specifications would be quite long, so we will only touch on a few of the key issues here.

One more thought before explaining the key components that contribute to system inaccuracy – be reasonable in specifying your input system! If your sensor provides an overall accuracy of ±1%, it’s unlikely that you’ll need a 24-bit A/D input. Over-specifying your system will simply add to system cost and complexity. The one caveat is to consider the future. You may want a system that can deliver the performance you’ll need down the road.

The four largest error contributors of an analog input system are typically: input offset, gain error, non-linearity and system noise. Parts 2 and 3 of this BLOG will describe these in greater detail. There are additional contributors of course, but for most applications and in most systems, they tend to be insignificant compared to the four primary error sources. The Figure shown here depicts the relationship between a “perfect” input system and the effect that each of these errors has on the measurement. Stay tuned for Parts 2 and 3 where we’ll dive deeper into…

Error Sources: Input Offset and Gain – Part 2

Non Linearity and Noise Error Sources – Part 3

UEIDAQ.com | Facebook | LinkedIn | Youtube

ARINC Communication Interfaces

Communications is an often overlooked aspect of data acquisition and control systems. Note that we’re not talking about the communications interface between the I/O device and the host computer. We’re referring to various devices to/and from which we either need to acquire data or issue control commands. Consider the ARINC-429 interface, commonly used in either a commercial aircraft or ship.

ARINC-429 is the avionics interface used by almost all commercial aircraft (though 429 is not the primary interface on the Boeing 777 and 787 and the Airbus A-380). It is used for everything from communicating between various complex systems such as flight directors and autopilots as well as for monitoring more simplistic devices such as airspeed sensors or flap position indicators.

In test systems, it’s often critical to coordinate data from ARINC-429 devices with more typical DAQ devices such as pressure sensors and strain gauges. When studying stress placed on a wing spar, you’d certainly like to be able to coordinate the stress results with such parameters as air-speed, altitude, and any turn or climb/descent induced g-forces.

While the ARINC-429 bus is well defined, computer based interfaces for the 429 bus are very different. The 429 bus defines functionality in terms of labels, with each label representing a different parameter. It’s important for the data acquisition system to be able to differentiate between the labels. If your system is only interested in airspeed, you want to ignore other parameters. Note that some ARINC-429 interfaces allow you to make these selections in interface hardware, while others place the burden of effort on the software.

Many ARINC-429 devices run on a definitive schedule. For example, the magnetic heading may be transmitted every 200 mS. Some ARINC interfaces count on software-based scheduling while others build the scheduling into an FPGA in the hardware. The more factors and parameters a given ARINC interface builds into hardware the better, as you may be counting on those precious host CPU cycles for other things.

UEIDAQ.com | Facebook | LinkedIn | Youtube | Twitter

Thoughts on Data Acquisition Timing and Synchronization

One important aspect of “non-standard” data acquisition and control systems is how larger systems are best synchronized. Often, it is critical that you know not only what happened, but also when it happened. In small systems, this is usually easy to accomplish as the analog inputs and even the output excitation, are on the same board. However, systems with high channel counts and, in particular, applications distributed over large areas require careful attention to timing. An in-depth discussion of this topic is well beyond the scope of this BLOG, but the following may help you get started in the right direction immediately.

Simple wiring of clock and trigger signals is often the quickest, easiest and most accurate way to synchronize events in different places. Most DAQ devices have one or more trigger/clock inputs and it is frequently possible to simply synchronize systems by connecting these signals. Note that the propagation of an electronic signal in a wire is very close to the speed of light. A thousand feet of wire would typically only introduce about a microsecond of delay.

Most people think of GPS as an inexpensive way to find the nearest gas station or Starbucks. However, GPS is also an excellent technology for providing very precise time information. It is a little known fact that the entire basis for the GPS system is extremely accurate clocks (as well as satellites at known locations). Even a relatively inexpensive GPS can provide absolute timing accuracy better than 1 microsecond. Though the GPS on your boat or in your car may not have a time output signal, many inexpensive GPS devices provide a 1 or 5 pulse per second signal accurate to within 1 microsecond of absolute UTC. Using these simple and inexpensive devices, it becomes straight-forward to synchronize data samples anywhere in the world. Check out our GPS and IRIG (Inter-range instrumentation Group) timing generation and synchronization boards.

The DNx-IRIG-650 provides inputs for standard analog, modulated IRIG signals as well as non-modulated DC, DCLS and Manchester II inputs. In addition to the IRIG inputs, the board also allows the user to provide an external 10 MHz master clock and/or a 1 PPS synchronization pulse. A generic digital input may also be used to directly capture event timing.

Data Acquisition Sample Rate Considerations

Always be certain to examine your analog input systems carefully and determine whether the sample rate specification really meets your needs. Many multi-channel DAQ input boards use a multiplexer connected to a single A/D converter. Most data sheets will specify the total sample rate of the board or system and leave you to calculate the “per channel” sample rate. Take for example a100 kilosample per second (kS/s), 8-channel, analog-to-digital (A/D) board. It will most certainly sample one channel at 100 kS/s. But if two or more channels are used, the 100 kS/s may be shared and sampled at 50 kS/s (max) each. Similarly, five channels may be sampled at 20 kS/s each. If the data sheet does not specify the sample rate as “per-channel”, assume that the sample rate must be divided among all of the channels sampled.
This becomes important when two or more input signals contain widely varying frequency content. For example, an automotive test system may need to monitor vibration at 20 kS/s and temperature at 1 S/s. If the analog input only samples at a single rate, the system will be forced to sample temperature at 20 kS/s and will waste a great deal of memory/disk space with the 19,999 temperature S/s that aren’t needed. Some systems, including all of UEI’s “Cube” based products, allow inputs to be sampled at different rates, while products from many vendors do not.
Another sampling rate concern is the need to sample fast enough, or provide filtering to prevent aliasing. If signals included in the input signal contain frequencies higher than the sample rate, there is the risk of aliasing errors. Without going into the mathematics of aliasing, lets just say that these higher frequency signals will manifest themselves as a low frequency error. The accompanying Figure provides a graphical representation of the aliasing phenomenon. A visual example of aliasing can be seen in video where the blades of a helicopter or the spokes of a wheel appear to be moving slowly and/or backwards. In the movies it doesn’t matter, but if the same phenomenon appears in the measured input signal, it’s a critical error!
UEIDAQ.com | Facebook | LinkedIn | Youtube | Twitter

Board- vs. Box-based Systems

PC-based data acquisition (DAQ) systems are available with a wide variety of interfaces. Ethernet, PCI, USB, PXI, PCI Express, Firewire, Compact Flash and even the venerable GPIB, RS-232/485, and ISA bus are all popular. Determining which one is the best fit for a given application may be far from obvious.

Perhaps the first question to address when considering a new DAQ project is whether the application is best served by a plug-in board system (e.g., UEI’s PD2 series of multifunction boards) or an external “box” based system (e.g., UEI’s PowerDNA Cubes or various other USB devices). This issue has been a source of much confusion (and competition) over the years, and the decision may be less well defined today than ever.

In the early days of PC-based DAQ, the rule of thumb was: High speed measurements were performed by board solutions, high accuracy was the domain of the external box. Of course, there was a “gray” area that could be addressed by either form factor.

Today’s gray area is much larger than ever before. Board level solutions offering 24-bit resolution are now available as are 6.5 digit DMM boards. On the box side, USB 2.0 is theoretically capable of delivering 30 million 16-bit conversions per second and Gigabit Ethernet will handle more than twice that. Though internal plug-in slot data transfer rates have increased 10 fold in recent years, the typical data acquisition system sample rate has not. Planes and cars don’t go much faster now than they did in 1980, and temperatures and pressures are still relatively slow changing phenomena.

Since most application accuracy and sample rates are perfectly within the capabilities of both board and box level solutions, other considerations will ultimately determine which solution is best for a given application.

UEIDAQ.com | Facebook | LinkedIn | Youtube | Twitter

When form, fit & function matter to Ethernet-based measurement and control

Control and automation engineers use real-time Ethernet to create high-density distributed I/O measurement and control systems. Such platforms are constantly evolving in form, fit and function to accommodate various application needs, including those related to footprint and workspace requirements. Fortunately for these engineers, there’s a plethora of different systems, models and configurations from which to choose for bench-top I/O, highly integrated workstations, and data/control center (rack-mount) projects.

For data/control center applications that require a standard 1U rack-mountable solution, the choices have been limited…that is until now!

Check out this new Ethernet-based measurement and control system platform. The DNF-4-1G FLATRACK™ is a highly integrated, low-profile (1U) chassis providing access to a wide array of I/O including analog, digital, avionics, communications and more. The 6” x 1.75” x 17.5” unit was developed in response to key customer requests for a 1U version of the company’s RACKtangle™.

Designed for a variety of high-density applications that cannot tolerate external signal conditioning, the FLATRACK provides two Gigabit Ethernet (100/10 Base-T compatible) interfaces and four front-loading slots that allow I/O boards to be quickly and easily installed, retrofitted or removed. The Flat-RACK plays host to dual-channel NICs, a PowerPC CPU, two USB 2.0 controller/slave ports, timing/trigger interface circuitry, configuration ports, power supply and operational software – all in a standard 1U (6″ x 1.75″ x 17.5″) rack-mountable chassis.

The beauty of the FLATRACK is its configurable versatility for deployment as a high-density Power DNA (Distributed Networked Automation) system, Programmable Automation/Embedded Controller, MODBUS TCP interface, or a Simulink I/O platform. These high-density I/O deployments require precise, real-world measurement and control capabilities. The FLATRACK’s PowerDNA® (Distributed Networked Automation) architecture hosts a rugged, Ethernet-based data acquisition (DAQ) interface, ideally suited for various industrial, aerospace and laboratory data acquisition and control environments. A MODBUS messaging protocol is used to establish master-slave/client-server communication between devices that measure voltage, current, strain gages, thermocouples and more. The UEIPAC (Programmable Automation Controller) is nicely suited for embedded data acquisition (DAQ) applications, as it allows systems to be developed without the cost or the additional space required by an external host computer.

Finally, UEISIM offers Simulink developers a powerful and flexible I/O target. Models built in Simulink are deployed directly on the UEISIM using Real-Time Workshop to create, for example, an efficient tuning solution for real-time and non-real-time applications, including simulation model verification, rapid prototyping, and hardware-in-the-loop testing. With I/O interfaces for analog I/O, digital I/O, counter/timer, ARINC-429 and ARINC-664, MIL-STD-1553, quadrature encoder, CAN-bus, serial I/O and more, the DNF-4-1G FLATRACK stands protocol-ready.

Contact UEI Sales for a demonstration.

UEIDAQ.com | Facebook | LinkedIn | Youtube | Twitter