The vast majority of data acquisition (DAQ) and control I/O channels are fairly generic A/D inputs, D/A outputs, and parallel digital I/O. System requirements vary greatly, but the old 80-20 rule holds as well in the DAQ arena as anywhere. Eighty percent of the I/O channels typically are addressed by 20% of the available I/O products. However, nobody wants an 80% solution, so the remaining 20% of the I/O channels also must be addressed.
Much has been written about the 80% products. It’s time the remaining 20% received their day in the sun.
Most second-tier I/O is related to monitoring and sometimes controlling motion.
Synchros and Resolvers
Synchros and resolvers have been used to measure and control shaft angles in various applications for more than 50 years. These units became extremely popular during World War II in fire/gun control applications as indicators/controllers for aircraft control surfaces and even to synchronize the sound and video in early motion picture systems. In the past, these units also were called selsyns.
At a first glance, synchros and resolvers don’t look too different from electric motors. They share the same rotor, stator, and shaft components. The primary difference between a synchro and a resolver is that a synchro has three stator windings installed at 120-degree offsets while the resolver has windings installed at 90-degree angles.
To monitor rotation via a synchro or resolver, the DAQ needs to provide an AC excitation signal and an analog input capable of digitizing the corresponding AC output. In some applications, the synchro/resolver excitation is provided by the DUT itself. In these cases, it is important to ensure that the DAQ interface is capable of synchronizing to the external excitation. This typically is accomplished by using an additional analog input channel.
Although it is possible to create such a system using standard analog input and output devices, it is a fairly complicated process, and most people opt for a dedicated synchro/resolver interface. These DAQ products not only provide appropriate signal conditioning, they also typically take care of most of the math required to turn the analog input into rotational information. It always is a good idea to check the software support of any synchro/resolver interface to ensure it does provide results in a format you can use.
Most synchros/resolvers require an excitation of roughly 26 Vrms at frequencies of either 60 Hz or 400 Hz. It is important to check the requirements of the actual device you are using. Some units need 120 Vrms. Also, some synchro/resolver devices, particularly those used in applications where rotational speed is high, require higher excitation frequencies although you infrequently will see a system requiring anything higher than a few kilohertz.
Finally, some synchro/resolver interfaces can use the excitation outputs as simulated synchro/resolver signals. This capability is very helpful while developing aircraft or ground vehicle simulators as well as providing a way to test and calibrate synchro/resolver interfaces without requiring an actual synchro or resolver.
LVDT and RVDT
Linear/rotary variable differential transformer (LVDT) and (RVDT) devices are similar to synchros/resolvers since they use transformer coils to sense motion. However in an RVDT/LVDT, the coils are fixed in location, and the movement of the ferromagnetic core relative to the coils induces the desired signal. The primary difference between the LVDT and a synchro/resolver is that it is used to measure linear motion, not rotation.
Unlike synchros/resolvers, the RVDT has a limited angular measurement range. However, the synchro/resolver can be used for multiturn rotational measurement with rated accuracy for the entire 0 to 360-degree spectrum.
When connecting an RVDT/LVDT to a DAQ system, most of the concerns are similar to those of synchros. First, you may build an RVDT/LVDT interface out of generic A/D and D/A interfaces, but it’s not a trivial exercise. Most people opt for a special-purpose interface designed for the task. In addition to eliminating the need for complex signal conditioning, the specifically designed interface usually will convert the various signals into either rotation in degrees or percent of scale or, in the case of the LVDT, in percentage of full scale.
The LVDT/RVDT interface also will provide the necessary excitation, which typically is in the 2 to 7 Vrms range at frequencies of 100 Hz to 5 kHz. Some systems may supply their own excitation, and in this case, be sure the LVDT/RVDT interface you choose has a means to synchronize to it.
Finally, like the synchro/resolver, LVDT/RVDT interfaces can use the excitation outputs as simulated LVDT/RVDT signals. This capability is very helpful when developing aircraft or ground vehicle simulators as well as providing a way to test and calibrate RVDT/LVDT interfaces without requiring the actual hardware.
String pots are designed to measure linear displacement. They are typically lower cost than LVDTs and can offer much longer measurement distances.
As the name implies, the basis for a string pot is a string or cable and a potentiometer. Basically, a string and a spring are attached to the wiper screw of the potentiometer, and as the string is pulled, the potentiometer resistance changes. The string pot provides a calibration factor that describes what displacement is represented by a percentage of resistance change.
As a simple variable resistance device with a linear output, most string pots are interfaced to standard A/D boards. The most common configuration connects a voltage reference to the wiper, then to an A/D input channel. With the string completely retracted, the measured voltage will be either the reference voltage or zero. With the string completely extended, the voltage measured will be the opposite, either zero or the reference voltage. At any intermediate string extension, the voltage measured will be proportional to the percentage of string out.
Be sure your voltage reference has the output current capacity to drive the string pot resistance. Your measurement will be in error by the same percentage as any voltage reference error.
In some cases, it may be beneficial to drive the string pot with a higher capacity, lower accuracy voltage source. Should you need higher accuracy than the voltage source provides, you may always dedicate an A/D channel to measure the voltage source. This makes the system virtually immune to inaccuracies in the voltage source.
String pots are single-ended isolated devices. When connecting a string pot to a differential input, be sure to connect the string pot/reference ground and the A/D channel’s low or negative input. Failing to make this connection in some way will likely cause unreliable and even odd behavior as the input negative terminal floats in and out of the input amplifier’s common-mode range.
Quadrature encoders also are used to measure angular displacement and rotation. These products provide a digital output. There are two primary digital outputs, which are in the form of 90-degree out-of-phase digital pulse trains. The frequency of the pulses determines the angular velocity while the relative phase between the two (+90 degrees or -90 degrees) describes the direction of rotation.
These pulse trains can be monitored by many generic DAQ counter systems with one of the outputs being connected to a counter clock while the other is connected to an up/down pin. However, the quadrature encoder is such a common part of many DAQ systems that many vendors provide an interface specifically developed for quadrature measurements.
One thing that cannot be determined from the pulse counts alone is the absolute position of the shaft. For this reason, most quadrature encoder systems also provide an index output. This index signal generates a pulse at a known position. Once a known position is identified, the absolute position can be determined by adding or subtracting the relative rotation to the known index position.
Many encoders provide differential outputs, but the differential noise immunity is seldom required unless the electrical environment is very difficult or the runs from the encoder to the DAQ system are hundreds of feet or more.
Piezoelectric Crystal Controllers
When considering piezoelectric crystals in the DAQ and control arena, most people think about vibration and accelerometer sensors because these crystals are the basis for the ubiquitous integrated circuit piezoelectric/integrated electronics piezoelectric (ICP®/IEPE) sensors. When you exert a force on a piezoelectric crystal, it causes the crystal to deform slightly, and this deformation induces a measurable voltage across the crystal.
A voltage placed across an unstressed piezoelectric crystal causes the crystal to deform. This deformation actually is very small but also very well behaved and predictable.
Piezoelectric crystals have become a very common motion-control device in systems that require very small deflections. In particular, they are used in a wide variety of laser control systems and a host of other optical control applications. In these applications, a mirror is attached to the crystal, and as the voltage applied to the crystal is changed, the mirror moves. The movement typically is not detectable by the human eye, but compared to the wavelengths of light, the movement is substantial.
Driving these piezoelectric devices presents two interesting challenges. First, achieving the desired movement from a piezoelectric crystal often requires large voltages, though mercifully at low DC currents. Second, although the crystals have high DC impedance, they also have very high capacitance, and driving them at high rates is not trivial. Special drivers often are required because the typical analog output board does not offer the output voltage or capacitive drive capability required.
Communications is an often forgotten part of many DAQ and control systems. We’re not talking about the communications interface between the I/O device and the host computer. We’re referring to various devices which we either need to acquire data or issue control commands. Examples of this type of device might be the ARINC-429 interface in a commercial aircraft or ship or the controller area network (CAN) bus in an automobile.
ARINC-429 is the avionics interface used by almost all commercial aircraft although 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 to monitoring more simplistic devices such as airspeed sensors or flap deflection indicators.
In test systems, it’s often critical to coordinate data from ARINC-429 devices with more typical DAQ products such as pressure sensors and strain gauges. When studying stress placed on a wing spar, you’d certainly like 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 DAQ system to be able to differentiate between the labels. If your system is only interested in air speed, you want to ignore other parameters. 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.
MIL-STD-1553 is the military’s equivalent to ARINC-429 although structurally it is very different. First and most obvious, most 1553 links are designed with dual redundant channels. Commercial aircraft typically don’t get wires cut by bullets or flak, but military aircraft usually are designed so that a single cut wire or wiring harness won’t cause a loss of system control. If you are looking to hook to a MIL-STD-1553 device, be sure your interface has both channels.
Also, a MIL-STD-1553 device can serve as a bus controller, bus monitor, or remote terminal. Not all interfaces support all three functions. Be sure the interface you select has the capability you require.
Similar to the ARINC-429 bus, when operating as a bus controller, the device must handle detailed transmission scheduling including major and minor frame timing. This is best performed in hardware rather than via software timing.
The CAN bus is the standard communications interface for automotive and truck systems. Gone are the days when your car was controlled by mechanical linkages, gears, and high-current switches. Your transmission now shifts gears based on CAN commands sent from a computer. Even such things as raising/lowering the windows and adjusting the outside rearview mirror now are done via CAN sensors and actuators. Vehicle speed, engine rpm, and even internal temperature are all available on the CAN bus.
When running tests in a car or truck, it’s very useful to coordinate the data available on the various CAN networks with any more conventional DAQ measurements you may be making. If you are measuring internal vibration, you’ll want to coordinate it with engine rpm and speed.
Like any DAQ system, one of the first things you need to be aware of when specifying a CAN interface system is how many CAN ports you will need. Sometimes there are 50 or more CAN sensors and actuators in a given vehicle. Be sure your system has enough channels to grab all the data you need. The CAN specification supports data rates up to 1 megabaud. Be sure the system you specify is capable of matching the speed of the network you wish to monitor.
Predicting the demise of RS-232 began in the 1980s. Of course RS-232 is still around and kicking. In fact, the RS-series ports remain extremely common in the DAQ and control arena.
RS-232 is older and slower than its 422/423/485 family mates but still very much in use. There is not too much to consider when specifying an RS series interface, but a few words may be in order. First, not all serial devices operate at the same speed. Specify a device that will handle the baud rate. Second, for stable and consistent operation, especially at higher speeds, select a device with a substantial FIFO.
RS-232 ports, and in particular those on older devices, use hardware handshaking signals such as Ready to Send or Clear to Send. Many newer RS-232 interfaces do not support these handshaking signals so make sure your serial interface supports what you need.
Another common series of questions arises when considering the differences of RS-422, -423, and -485. RS-422 uses a two-wire, fully differential signal interface. RS-423 uses the same signal levels but only one of the two wires. RS-422 and RS-485 are almost identical. However, an RS-485 is networkable and can be connected to multiple serial devices; an RS-485 interface will almost always be perfectly suitable for talking to an RS-422 device.
Timing and Synchronization
One final aspect of nonstandard DAQ and control systems is how larger systems are synchronized. Often it is critical to know not only what happened but also when it happened. In small systems, this usually is 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 spread over large areas require careful attention to timing.
Simple wiring of clock and trigger signals often is 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. The propagation of an electronic signal in a wire is very close to the speed of light. A thousand feet of wire typically would only introduce about a microsecond of delay.
Most people think of GPS as an inexpensive way to find the nearest gas station or pizza restaurant. However, GPS also is an excellent technology for providing very precise time information. In fact, the entire basis for GPS is extremely accurate clocks as well as satellites at known locations.
Even a relatively inexpensive GPS can provide absolute timing accuracy to better than 1 µs. Although the GPS on your boat or car may not have a time output signal, many GPS devices provide a 1- or 5-pulse/s signal accurate to within 1 µs of absolute UTC. Using these simple and inexpensive devices, it becomes straightforward to synchronize samples anywhere in the world.
Inter-Range Instrumentation Group (IRIG) is not so much a timing technology as it is a timing protocol. However, many of the high-accuracy timing and synchronization devices are generically referred to as IRIG interfaces.
The underlying timing of an IRIG device can be based upon many things including GPS, WWV synchronization, or a highly stable and accurate on-board clock. The key to using an IRIG device is simply to look to the device’s rated timing accuracy and ensure that it will provide the synchronization your system requires.
IEEE 1588 is a fairly new Ethernet-based timing and synchronization protocol based on a network’s capability to automatically identify the most accurate timing device online and then synchronize all the various network nodes to the master clock. In this way, an application can be developed where a single high-performance clock provides accurate and reliable timing information throughout the network.
As you can see, there often is much more to specifying a DAQ and control system than simply selecting the appropriate A/D, D/A, and digital I/O devices. Hopefully, I have provided a useful introduction to some of the more common second-tier interfaces. If you wish more information on any of these items, it should not be much more than a few key-clicks away on your favorite Web-based search engine.
About the Author
Bob Judd is director of sales and marketing at United Electronic Industries. Prior to joining UEI, he was general manager and vice president of marketing and hardware engineering at Measurement Computing and previously vice president of marketing at MetraByte. Mr. Judd, who has been involved in the PC-based DAQ market for more than 20 years, holds a bachelor’s degree in engineering from Brown University and a master’s degree in management from MIT. United Electronic Industries, 27 Renmar Ave., Walpole, MA 02081, 508-921-4557, e-mail: [email protected]