Swipe to the left

How to select an encoder

By Yun Shen May 7, 2019 No comments
Selecting the right encoder for your application can be a stressful task. With a variety of outputs, reading types and encoder body options to choose from, how can one get the best sensor for his application?

Let us first explain the difference between incremental and absolute encoders.

What is an encoder?
In general, an encoder is a device that converts data from one format to another. In position sensing, an encoder is a device which can detect and convert mechanical motion to an analogue or digital coded output signal.

Incremental and absolute encoders

The difference between incremental and absolute encoders is analogous to the difference between a stop watch and a clock. A stop watch measures the incremental time that elapses between its start and stop, much as an incremental encoder will provide a known number of pulses relative to an amount of movement. If you knew the actual time when you started the watch, you can tell what time it is later by adding the elapsed time value from the stop watch. For position control, adding incremental pulses to a known starting position will measure the current position. When an absolute encoder is used, the actual position will constantly be transmitted, just as a clock will tell you the current time.

See incremental and absolute encoders.

Rotary and linear encoders

More specifically, an encoder measures position, while velocity, acceleration and direction can be derivated from position in either linear or rotary movement. Different functionality of encoders comes from different physical principles of operation, outputs, communication protocols etc.

See rotary and linear encoders.

Resolution

Resolution is the smallest movement detected by the encoder. It is related to precision with which the measurement is made and it's measured differently according to the type of encoder:

Absolute Rotary encoder: resolution is the number of measuring segments or units in one revolution.13 bit resolution of an absolute rotary encoder means that there are 213 = 8192 steps within 360° rotation.
Incremental Rotary encoders: define resolution in PPR and CPR. Various manufacturers may interpret those terms differently so it's recommended to check in advance what specifically they mean. PPR usually stands for pulses per revolution which means a signal pulse from one rising edge to the next. This number is the same as the number of lines on a disk of a conventional optical encoder. CPR on the other hand, usually stands for counts per revolution and is achieved by using both the rising and the falling edges on both channels (CPR = PPR*4). Resolution of a rotary encoder can also be expressed in arcseconds, arcminutes, degrees, grads or radians.
Linear encoder: resolution is the length of the measuring step. It can be expressed in micro meters (µm) or nano meters (nm). It is also possible to express resolution in number of increments per unit of length. In printing applications it's commonly expressed in the form of DPI (dots per inch).

Accuracy

Accuracy is a measure of how close the output is to where it should be – the deviation between actual position and from the encoder's reported position.

For rotary encoders it is usually expressed in arcseconds or degrees. On the other hand, the most common unit for expressing accuracy of linear encoders is µm per scale unit of length. Accuracy of an encoder is a combination of the scale accuracy and errors introduced by the readhead.

Errors introduced by the scale include:

  • Uniformity of scale pattern – quality of pole magnetisation, length and position.
  • Stability of scale pattern carrier – thermal expansion.

Errors introduced by the readhead include:

  • SDE - sub divisional error also known as interpolation error is a consequence of imperfectly dividing one sin/cos cycle into measuring steps. It is a result of imperfect sin/cos input signals. SDE is periodic with period equal to the length of sin/cos period. Usually it is better than 0,1% of sin/cos period length.

Perfect sin/cos input signals result in equal distribution of increments within one period (after intepolation process) -> low SDE (good accuracy) :


Distorted sin/cos input signal result in unequally long increments within one period (after interpolation process) -> high SDE (bad accuracy):

  • Hysteresis - the difference in reported position when coming to the same position from either direction of rotation or linear movement. It is composed of magnetic, mechanical and electronic components (more on this later).

Errors introduced by installation of an encoder:

  • For rotary measurement:
    1. eccentricity of the scale – offset of geometric center of magnet or ring relative to centre of rotation
    2. inclination of rotational axis relative to scale axis
    3. relative position of readhead to scale – yaw, pitch, roll, ride height (for ring encoders)
    4. quality of bearings (for contact-less encoders)
    5. quality of couplings (for shafted encoders)
  • For linear measurement
    1. emperature of material and environment – thermal expansion.
    2. Abbe error – magnification of angular error over distance. For example, if the axis of linear movement is not straight, the difference between actual movement and reported movement will include error which is proportional to offset and tangent of angle Θ.May 28, 2019 10:11:07 AM
    3. Cosine error – misalignment of motion axis relative to relative to measurement axis. It is eliminated when the axis of motion and the measurement axis are parallel.May 28, 2019 10:12:08 AM
    4. Relative position of readhead to scale – yaw, pitch, roll, ride height, lateral offset.May 28, 2019 10:13:07 AM

    Additionally environmental factors such as dust and liquids for optical encoders or external magnetic fields for magnetic encoders can significantly reduce encoder functionality.

    Encoders are often grouped in accuracy grades which can be defined differently according to each manufacturer or end user.

    • For linear systems, extreme error values for any max. one meter section of the measured length lie within the specified accuracy class of of ±a μm with respect to their mean value. For measuring length up to 1 m, the tolerance (±a μm) refers to the actual measuring lengths. The accuracy applies to a reference temperature of 20°C. Examples of RLS accuracy grades for linear encoders could be ±20 µm, ±40 µm, ±100 µm,...
    • For rotary systems, extreme error values for full circle of rotation (360°) lie within the specified accuracy class of of ±a ° with respect to their mean value. The accuracy applies to a reference temperature of 20°C.

Generally, magnetic encoders are less accurate than optical encoders, however there are areas where both types of encoders overlap in performance and price. It should be noted that higher resolution does not automatically provide higher accuracy.

#encoder Posted in: How to

How to select an encoder

By Yun Shen May 7, 2019 No comments
Selecting the right encoder for your application can be a stressful task. With a variety of outputs, reading types and encoder body options to choose from, how can one get the best sensor for his application?

Let us first explain the difference between incremental and absolute encoders.

What is an encoder?
In general, an encoder is a device that converts data from one format to another. In position sensing, an encoder is a device which can detect and convert mechanical motion to an analogue or digital coded output signal.

Incremental and absolute encoders

The difference between incremental and absolute encoders is analogous to the difference between a stop watch and a clock. A stop watch measures the incremental time that elapses between its start and stop, much as an incremental encoder will provide a known number of pulses relative to an amount of movement. If you knew the actual time when you started the watch, you can tell what time it is later by adding the elapsed time value from the stop watch. For position control, adding incremental pulses to a known starting position will measure the current position. When an absolute encoder is used, the actual position will constantly be transmitted, just as a clock will tell you the current time.

See incremental and absolute encoders.

Rotary and linear encoders

More specifically, an encoder measures position, while velocity, acceleration and direction can be derivated from position in either linear or rotary movement. Different functionality of encoders comes from different physical principles of operation, outputs, communication protocols etc.

See rotary and linear encoders.

Resolution

Resolution is the smallest movement detected by the encoder. It is related to precision with which the measurement is made and it's measured differently according to the type of encoder:

Absolute Rotary encoder: resolution is the number of measuring segments or units in one revolution.13 bit resolution of an absolute rotary encoder means that there are 213 = 8192 steps within 360° rotation.
Incremental Rotary encoders: define resolution in PPR and CPR. Various manufacturers may interpret those terms differently so it's recommended to check in advance what specifically they mean. PPR usually stands for pulses per revolution which means a signal pulse from one rising edge to the next. This number is the same as the number of lines on a disk of a conventional optical encoder. CPR on the other hand, usually stands for counts per revolution and is achieved by using both the rising and the falling edges on both channels (CPR = PPR*4). Resolution of a rotary encoder can also be expressed in arcseconds, arcminutes, degrees, grads or radians.
Linear encoder: resolution is the length of the measuring step. It can be expressed in micro meters (µm) or nano meters (nm). It is also possible to express resolution in number of increments per unit of length. In printing applications it's commonly expressed in the form of DPI (dots per inch).

Accuracy

Accuracy is a measure of how close the output is to where it should be – the deviation between actual position and from the encoder's reported position.

For rotary encoders it is usually expressed in arcseconds or degrees. On the other hand, the most common unit for expressing accuracy of linear encoders is µm per scale unit of length. Accuracy of an encoder is a combination of the scale accuracy and errors introduced by the readhead.

Errors introduced by the scale include:

  • Uniformity of scale pattern – quality of pole magnetisation, length and position.
  • Stability of scale pattern carrier – thermal expansion.

Errors introduced by the readhead include:

  • SDE - sub divisional error also known as interpolation error is a consequence of imperfectly dividing one sin/cos cycle into measuring steps. It is a result of imperfect sin/cos input signals. SDE is periodic with period equal to the length of sin/cos period. Usually it is better than 0,1% of sin/cos period length.

Perfect sin/cos input signals result in equal distribution of increments within one period (after intepolation process) -> low SDE (good accuracy) :


Distorted sin/cos input signal result in unequally long increments within one period (after interpolation process) -> high SDE (bad accuracy):

  • Hysteresis - the difference in reported position when coming to the same position from either direction of rotation or linear movement. It is composed of magnetic, mechanical and electronic components (more on this later).

Errors introduced by installation of an encoder:

  • For rotary measurement:
    1. eccentricity of the scale – offset of geometric center of magnet or ring relative to centre of rotation
    2. inclination of rotational axis relative to scale axis
    3. relative position of readhead to scale – yaw, pitch, roll, ride height (for ring encoders)
    4. quality of bearings (for contact-less encoders)
    5. quality of couplings (for shafted encoders)
  • For linear measurement
    1. emperature of material and environment – thermal expansion.
    2. Abbe error – magnification of angular error over distance. For example, if the axis of linear movement is not straight, the difference between actual movement and reported movement will include error which is proportional to offset and tangent of angle Θ.May 28, 2019 10:11:07 AM
    3. Cosine error – misalignment of motion axis relative to relative to measurement axis. It is eliminated when the axis of motion and the measurement axis are parallel.May 28, 2019 10:12:08 AM
    4. Relative position of readhead to scale – yaw, pitch, roll, ride height, lateral offset.May 28, 2019 10:13:07 AM

    Additionally environmental factors such as dust and liquids for optical encoders or external magnetic fields for magnetic encoders can significantly reduce encoder functionality.

    Encoders are often grouped in accuracy grades which can be defined differently according to each manufacturer or end user.

    • For linear systems, extreme error values for any max. one meter section of the measured length lie within the specified accuracy class of of ±a μm with respect to their mean value. For measuring length up to 1 m, the tolerance (±a μm) refers to the actual measuring lengths. The accuracy applies to a reference temperature of 20°C. Examples of RLS accuracy grades for linear encoders could be ±20 µm, ±40 µm, ±100 µm,...
    • For rotary systems, extreme error values for full circle of rotation (360°) lie within the specified accuracy class of of ±a ° with respect to their mean value. The accuracy applies to a reference temperature of 20°C.

Generally, magnetic encoders are less accurate than optical encoders, however there are areas where both types of encoders overlap in performance and price. It should be noted that higher resolution does not automatically provide higher accuracy.

#encoder Posted in: How to