Converting angular rotation to an electrical signal is the job of an AC transducer. Types of transducers include synchros, resolvers, and linear/rotary variable differential transformers (LVDTs/RVDTs). They can be used in a variety of applications, such as an inertial navigation reference unit (gyro or compass), an automatic direction finder (ADF), an omnirange system, distance measurement equipment, cockpit indicators, and landing-gear positioning and control.
Synchros have been used in a variety of military and commercial systems for many years. Traditionally, they have been the transducer of choice where reliability is important and difficult environment conditions exist. The simplicity of their connection and today’s synchro-to-digital and digital-to-synchro converter boards make the synchro a very attractive component.
In an ADF, for example, the resolver or synchro is used to drive an indicator. As the aircraft turns, the amount of coupling in the transducer changes proportionally. But how do synchros and resolvers work, and what techniques must be used to ensure they produce accurate results?
Theory of Operation
Synchros and resolvers, World War II-era technology, still are widely used in modern-day electronic motion-control applications. Essentially, they are transformers. Just like a traditional transformer, they have a primary winding and multiple secondary windings. And just like a transformer, their primary is driven by an AC signal.
Synchros and resolvers are very similar; however, there are some differences. As shown in Figure 1, a synchro has one primary winding and three secondary windings, with each secondary winding mechanically oriented 120º apart. In contrast, as shown in Figure 2, a resolver has two primary windings and two secondary windings oriented at 90º to each other.
While a synchro and a resolver are electrically very similar to a transformer, they are mechanically more like a motor. The primary winding in a synchro or a resolver can be physically rotated with respect to the secondary windings. For this reason, the primary winding is called the rotor. The secondary windings, which are fixed, are called stators.
Synchros are often used to track the rotary output angle of a closed-loop system, which uses feedback to achieve accuracy and repeatability. A synchro can be turned continuously and, since its secondary winding outputs are analog signals, provide infinite resolution output.
As the shaft of a synchro turns, the angular position of its rotor winding changes with respect to its secondary (stator) windings. The relative amplitude of the resulting AC output signals from the secondary windings indicates the rotary position of the synchro’s shaft.
A synchro is excited by an AC reference voltage applied to its rotor winding. Typically, this reference signal is 115 Vrms @ 60 Hz or 400 Hz or 26 Vrms @ 400 Hz.
Synchro-to-Digital Conversion
Synchros generate analog output signals, and those outputs typically must be converted to digital form. This can be accomplished by using a synchro-to-digital converter.
It might seem logical to use a set of conventional analog-to-digital converters to simultaneously sample the AC voltages on the outputs of a synchro. Then, the relative magnitudes of these samples could be used to determine the rotary position of the synchro’s shaft at the time the samples where taken.
However, this does not work well for several reasons:
Synchros have inductive characteristics that must be taken into account.
Synchro output signals can be heavily distorted due to nonlinearities in the synchro and phase-shift of the transducer.
Synchro output signals often contain a lot of noise because of the environment they work in.
In addition, if the inputs and the outputs of a synchro are not galvanically isolated from each other and from signal ground, common-mode noise can dramatically affect the accuracy of conversion. Galvanic isolation implies magnetic, transformer-coupled isolation as opposed to resistive or solid-state isolation.
To avoid these problems, a synchro-to-digital converter must use transformer-isolated inputs and outputs. This significantly reduces the effects of common-mode noise and ground loops.
In terms of signal processing, tracking conversion (closed-loop) is more desirable than successive approximation or direct digitization because of system nonlinearities. Tracking conversion implies a servo mechanism for continuous tracking of the input. On the other hand, direct digitization, such as analog-to-digital (A/D) conversion, and successive approximation are suspect to instantaneous errors that may corrupt the output.
The major benefits of tracking conversion include an error-free output value that tracks the synchro’s angular position up to the maximum tracking speed. Another advantage is high noise immunity and quadrature rejection (the voltage component of an AC signal that is 90° out of phase with the reference and at the fundamental frequency).
A third benefit is the capability to provide tracking over a wide range of excitation frequencies (from 47 Hz to 20 kHz). The reference frequency impacts the overall system dynamic response: the higher the frequency, the higher the bandwidth.
Tracking converters can be classified into three types:
Type 0—with a finite position error even when the rotor is stationary.
Type I—with zero positional error when the rotor is stationary, but a finite positional lag error when the rotor is spinning at a constant angular velocity.
Type II—with zero positional error when the rotor is stationary and when it is spinning at a constant velocity.
Modern resolver-to-digital converters use Type II conversion methodology. It has the best tracking characteristics and minimizes velocity errors accumulating in the position data. As shown in Figure 3, a typical Type II tracking converter consists of the following elements:
An input isolation transformer.
A digital-to-analog converter that is used to multiply the analog SIN input by a digitally generated COS function and the analog COS input by a digitally generated SIN function.
A summing amplifier which still contains harmonics and quadrature at its output.
A phase-sensitive synchronous demodulator to clean the error voltage.
An integrator so there is no lag error associated with the rotor revolving at a constant angular velocity.
A voltage-controlled oscillator to generate a constant frequency to track the input.
An up-down counter with function generator ROMs on its output to evaluate polarity to count up for forward rotation and down for backward rotation.
A phase shifter and reference squarer that drive the demodulator.
This methodology is similar to that used by a phase-locked loop, which generates a local frequency and then adjusts that frequency to track the frequency of an input signal. As such, the Type II converter is a self-contained servo-mechanism for data acquisition and conversion, uniquely tailored to the properties of the input signals.
LVDTs and RVDTs
LVDTs and RVDTs are electromagnetic displacement transducers designed to provide output voltages proportional to linear and rotary displacement, respectively. These transducers consist of a primary winding, two secondary windings, and a movable armature made of a soft ferromagnetic material, as illustrated in Figure 4.
This transducer type would be typical of flight-surface measurement, such as flaps, slats, rudder, and aileron, or landing-gear positioning and control. As differential or ratiometric devices, they are particularly insensitive to common-mode noise and temperature effects, making them ideal for use in harsh environments.
Like synchros, LVDTs have front-end configuration and inductive characteristics that must be considered. The output signals can be heavily distorted because of nonlinearities and phase-shift in the transducer. And, output signals often contain a lot of noise because of the environment they work in.
Like synchros, if the inputs and the outputs of an LVDT are not galvanically isolated from each other and from signal ground, common-mode noise can dramatically affect the accuracy of conversion. Multichannel implementations are typical, and hardware is tailored to meet voltage and frequency requirements. Such instruments are essential tools to the measurement, control, simulation, and test of motion electronics.
Testing and Troubleshooting
Synchro and LVDT signals require specialized test equipment for accurate measurement and calibration. Many parameters must be considered since all of these invariably will affect the performance and integrity of the system as a whole. The list includes accuracy, phase shift and quadrature effects, resolution, dynamic characteristics, and distortion and noise.
Accuracy of the synchro/resolver signals are most easily verified using an angle position indicator (API). These instruments can read the synchro signal directly in degrees. APIs automatically compensate for phase shift, noise, and distortion. Due to their transformer-coupled inputs, common-mode effects are minimized.
For calibration-grade accuracy up to 2 arc seconds, a synchro/resolver bridge and phase angle voltmeter (PAV) are required. The PAV also measures phase shift and total rms and fundamental, quadrature, and in-phase voltages. It is designed to measure these parameters accurately in the presence of highly distorted and noisy signals.
To accurately simulate a synchro/resolver signal or to drive a synchro/resolver, a synchro/resolver standard is required. It is important to understand the load requirements of the unit-under-test and to make sure that these fall within the output drive capability and rated accuracy of the simulator.
These simulators are designed to generate signals with enough resolution, speed, and drive capability to dynamically and statically test most common devices that accept three (synchro) or four (resolver) wire inputs. Typically, accuracy will be 2 arc seconds and traceable to the National Institute of Standards and Testing (NIST).
Conclusion
Synchros, resolvers, and LVDT/RVDTs are important and reliable elements of avionics systems. The operation and interface of these transducers enable navigation and flight-surface control feedback. Consequently, the operational characteristics, typical failure mechanisms, and available equipment for measurement, control, simulation, and test of these transducer types are important elements to the design of aircraft systems, flight simulators, and ATE. Transformer isolation of input signals and tracking conversion techniques are essential to the accurate handling of data acquisition and signal processing.
References
1. Johnson, M. and Salz, K., “Synchros and Resolvers, Motion Electronics in Avionics,” Avionics Test Equipment Handbook and Directory, ISBN1-885544-09-X, 1998, pp. 1-19 to 1-24.
2. Salz, K., “How Synchros and Resolvers Work, and How You Can Use Them to Build High-Performance Motion Control Systems,” VMEbus Systems, February 1998/1.
About the Author
Michael W. Johnson is a regional sales manager at North Atlantic Instruments. He has written several articles on topics including motion control, electronics, and chemical processing. Mr. Johnson holds a B.S. in engineering chemistry and an M.A. in electrical engineering and industrial management from the State University of New York at Stony Brook. North Atlantic Instruments, 170 Wilbur Place, Bohemia, NY 11716, (516) 218-1233.
Copyright 1999 Nelson Publishing Inc.
March 1999
|
|