Encoder Basics

Autotech offers one of the broadest lines of encoders in the industry, from a tiny size 15 (housing diameter to be 1.5”), industry standard size 25 (2.5” dia), to explosion proof encoders and everything in between including built-in gear trains, along with wide variety of position sensing technologies, electrical characteristics and communication options. This section is written to provide a reference guide to selecting the right encoder for your application.

Many industrial control systems need position and speed feedback. In the initial stages, the encoders consisted of potentiometers, brush encoders, magnetic encoders and rarely optical encoders and resolvers. Each device had certain limitations. The potentiometers and magnetic encoders had limited resolution. The brush encoders required frequent maintenance. The optical encoders used incandescent lamps, which were large in size and had limited life expectancy. The resolvers could offer better resolution and accuracy, but were very expensive due to the decoding electronics required.

The recent technological developments have brought significant improvements in the initial models. Today optical encoders and resolvers are more commonly used in industry. And with the introduction of cost effective Smart-Encoders by Autotech, there will be a paradigm shift in the selection and use of encoders.

Types of Encoders

Absolute and Incremental:

The incremental encoder, when it rotates, generates pulses, which are counted to give position information relative to a known point, whereas an absolute encoder provides a unique value at each position and retains actual shaft position even if power fails and the shaft moves. Incremental encoders are less complex and have fewer outputs, 2 or 3, whereas Absolute encoders typically have 12 outputs, and are generally more expensive.

Incremental encoder applications typically require a reset input to zero out the count and start a fresh cycle whereas absolute encoders do not need a reset input as the output is always unique and absolute.

Absolute encoders are also available as multi-turn units with built-in gear trains suitable for linear applications where it takes several revolutions of the encoder shaft to complete one machine or process cycle.

Choice of Optical or Resolver:

Optical Encoders

The Optical Encoders typically consist of a rotating and a stationary member. The rotor is usually a metal or glass disc mounted on its shaft. The disc has an optical pattern.

The stator has an LED block and phototransistors arranged so that the LED light shines through the transparent sections of the rotor disc and received by phototransistors on the other side.

Incremental Optical Encoder

The incremental optical encoders uses a simple disc pattern. This slotted rotor disc alternately interrupts the light beam between the LED & phototransistor and thus produces a pulse output. The number of pulses depends on the number of slots on the disc. The pulses are then fed to a counter, where they are counted to give position information. The pulse rate indicates shaft speed. An additional phototransistor can also determine the direction of rotation. Some models also provide a marker pulse output, which is generated once every revolution at a fixed shaft position and can be used to mark a zero reference point. Many different pulse configurations are available, but the most commonly known is called the “quadrature”, where two square wave pulses 90° apart from each other are generated.

Absolute Optical Encoders

As you can see from the picture of the disc used in an absolute optical encoder, it is much more complex than the simple disc used in incremental encoders. Since the absolute encoder needs to encode a unique value for the shaft position, the number of tracks on the disc and corresponding phototransistors depend on the resolution sought and number of bits used. For example, for a 12 bit absolute encoder with resolution of 4096, you need 12 tracks on the disc. Depending upon the shaft position, the phototransistor output is modulated in a gray-code pattern, which can be converted internally to binary or BCD. The size, complexity and cost of absolute optical encoders increases exponentially with resolution, as the pattern gets increasingly complex with increased number of bits.

Resolver Encoders

Resolvers, invented during World War II for military applications are by far the most rugged position transducers available. Resolver is essentially a rotary transformer, having one rotor winding and two stator windings. The stator windings are located 90° apart. Either rotor or stator winding can be used as primary. Typically, the rotor winding is driven by a reference voltage at a frequency ranging from 400 Hz to several KHz.

As the shaft rotates, the output voltages of the stator windings vary as the sine and cosine of the shaft angle. See figure 2. The two induced stator voltages are a measure of the shaft angle and are converted to a digital signal in resolver-todigital decoder.

Ratiometric Tracking Converter

(A typical block diagram for a Ratiometric Tracking Converter is shown in figure 3.) The circuit features a Type II servo-loop that comprises of sine/cosine multiplier and an error amplifier together with phase sensitive demodulator, error processor, voltage controlled oscillator (VCO) and an up/down counter. Since the VCO is controlled by an error integrator, the greater the lag between the actual shaft angle and the digital angle in the counter, faster will the counter be called upon to “catch-up” or “track” and eliminate the error. The information produced by this type of converter is always “fresh”, being continually updated and always available at the output. As an added bonus, additional outputs, such as, an analog output proportional to the shaft RPM to eliminate external tachometers and a busy signal pulse for incremental pulse applications, are also available. The basis of determining the shaft angle in a ratiometric converter is the ratio between the two stator signals: From this relationship it can be noted that the angle is no longer a function of the induced rotor voltage Vr, but rather the ratio of VS1 and VS2. Therefore, variations in the rotor voltage Vr, frequency and temperature are no longer factors in a ratiometric converter. This results in a highly accurate and repeatable resolver-to-digital converter.