Encoder Communications Hanbook

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16
At a glance
Powered by AI
The document discusses different types of encoders and their communication protocols. It also talks about issues with signal transmission and ways to minimize distortion.

The two main types of encoders are incremental encoders and absolute encoders. Incremental encoders generate pulses in response to motion while absolute encoders generate digital words that directly indicate position.

Absolute encoders have the inherent ability to remember position if power is lost, while incremental encoders do not. Absolute encoders can be used for both position and speed sensing and are useful where it is impractical to home machinery each time power is brought up.

Encoder Communications

A reference guide for


understanding the basics of
encoder communications

Dynapars characterizations of the Hiperface,


Sine/Cos and EnDat protocols are based on the
form of these products being offered commercially
at the time of this publication. These products are
subject to periodic revisions, so that interested
parties should check the following web sites for
additional and more current product information

20 7 D nap r, G rne , IL
ed
v

s m y be

y.

ed
.
duc s an co pan
n na

m
ww

control .com

ade ark

ade am

heir

o,
s

Modern factories are among the most wired places on earth, with multiple layers
of computer networks handling a broad range of communications tasks. There
are enterprise-level networks connecting different buildings and locations, plantlevel networks within the facility, and multiple machine-level networks, which
may be based on any of about a dozen network protocols.
The rising use of factory networks for automation has an impact on many
devices, among them are rotary encoders. Encoders may use any one of a
number of communication protocols. Other equipment may use still different
protocols. Integrating the different encoder networks so that diagnostic and
status information on individual encoders anywhere can be accessed by the
plant-level network is a challenge.

Encoders
Encoders are sensors that generate digital signals in response to movement. Both
shaft encoders, which respond to rotation, and linear encoders, which respond
to motion in a line, are available. When used in conjunction with mechanical
conversion devices such as rack-and-pinions, measuring wheels, or spindles, shaft
encoders can also be used to measure linear movement, speed, and position.
In general, there are two main types of encoders: incremental and absolute.
Incremental encoders generate a series of pulses in response to motion. These
pulses can be used to measure speed or be fed to a counter to keep track of
position. On the other hand, absolute encoders generate multi-bit digital words
that indicate actual position directly. A major benefit of absolute encoders is
that if the application loses power the encoder remembers its position prior to
the power loss. An incremental encoder, on the other hand, does not have this
inherent memory feature and so must be driven to home or a reference position
in order to reset itself.
Incremental encoders are used mainly for position control and also for speed
measurement where the lack of a zero point doesnt effect performance. Absolute
encoders can be used for both position and speed sensing and are useful where
it is impractical or inconvenient to home the machinery each time the power is
brought up.
Among encoders, there is a clear trend: the percentage of applications using
absolute encoders has increased. There are several key reasons for this. Recently,

Encoder Communications

Introduction

Encoder Communications Handbook


the price gap compared to incremental encoders has shrunk. This price decrease
has meant that absolute encoders are not just for critical applications anymore.
Other reasons are more industry wide and have to do with technological
changes. One of these is the increased need for information as factories get
more wired to factory software systems such as ERP (Enterprise Resource
Planning) and SPC (Statistical Process Control). This increases the need
for communication between machines as more devices and components are
connected to fieldbuses in retrofits and equipment upgrades. Another is a
desire and need for more diagnostic information from equipment to increase
preventative maintenance and prevent unexpected failure and downtime.
Many absolute encoders have the ability to change parameters (including the
zero point) without opening the case, and have internal diagnostics that monitor
and report the condition of the unit and help anticipate or prevent breakdowns.
For more detailed information on encoder basics, see the Encoder Application
Handbook from Dynapar available at www.dynapar.com.

Encoder Communications
The output of an encoder can be transmitted in either parallel or serial form.

Parallel Output
Parallel output makes all output bits
available simultaneously. It may be provided
as straight binary or transformed into gray
code. Gray code produces only a singlebit change at each step, which can reduce
errors.
Some parallel-output encoders also can
accept inputs from the outside output
latching commands, for example, and
direction sense setting.
The advantage of parallel output is that
its fast: all the data is available in real time,
all the time. Disadvantages include bulky
and expensive cables and limited cable

Decimal Gray Code


0
0000
1
0001
2
0011
3
0010
4
0110
5
0111
6
0101
7
0100
8
1100
9
1101
10
1111
11
1110
12
1010
13
1011
14
1001
15
1000

Binary
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
1100
1101
1110
1111

Serial Output
The alternative to parallel communication is serial. There are several dedicated
serial buses available, as well as standard industrial buses. Tradeoffs among these
include bandwidth, update rate, hardware requirements, wire count, proprietary
vs. nonproprietary nature, and availability.

SSI
SSI (or Synchronous Serial Interface) is an all-digital point-to-point interface
popular in Europe. It provides unidirectional communication at speeds up to
1.5MHz and uses two twisted pair wires plus two wires for power.
Some encoders also provide a 1V p-to-p sin/cos output for real-time control,
since the on-demand absolute encoder data can come in too slowly for many
control loops.
SSI extended can also provide parity or an alarm that can represent any
condition.
Encoders may also be supplied with additional incremental outputs for speed
feedback. For serial communications, one pair of wires is for a differential clock
signal and the other pair is for data feedback from the sensor.
Data rate depends on both resolution and cable length. Clock frequencies
can be as high as 1.5MHz. However, as clock frequency increases, the maximum
cable run decreases, a limitation common to all protocols.
Cable Length vs. Data Rate for SSI
Cable Length
Data Rate
50 m
400 kHz
100 m
300 kHz
200 m
200 kHz
400 m
100 kHz
The data frame length depends on the device and its resolution. In an SSI
interface, there is one slave per master, referred to as a point-to-point connection.
The clock remains high until the master needs information from the sensor.
It then sends a stream of clock pulses equivalent to the number of bits of

Encoder Communications

length. Most encoders come with cables a meter or two long. Open-collector
(sinking or sourcing) outputs can go roughly a third that far. However, using a
push-pull output configuration, cable lengths can reach 100m.

Encoder Communications Handbook


information from the sensor. Samples of each bit are usually taken on the falling
edge of the clock pulse. This insures that the propagation and process delays are
accounted for. Sensor manufacturers are free to insert alarm bits if required, but
the requirement must be provided before the product is shipped, and a single
alarm bit may have several meanings.
1

Clock pulse

tm

tv
Serial data

Dn (MSB)

Dn1

D1

D0 (LSB)

Special

Monostable P/S
Figure 1

SSI Communication Format

EnDat
EnDat (Encoder Data) is a proprietary protocol developed by Heidenhain of
Germany. Like SSI, it is synchronous and features a point-to-point connection,
with clock signals fed to the encoder by the controller with speeds to 4MHz.
EnDat can carry more information than SSI because it provides for internal
memory in the encoder that can be read and written to by the controller. This
data can include encoder diagnostics, identification, and alarm status. It can also
contain information about the motor in which it was mounted to such as model
and serial numbers.
Varying clock pulse stream lengths are also required from the varying data
that may be received or sent to the device. Also, using a function called a Datum
Shift, an encoder can easily be reset to a new zero or reference point. The Datum
Shift is a value that is added to the physical position of an encoder. SSI provides
a direct reading of the physical position, so it requires the end user to rotate the
shaft to the zero position.
As with SSI, EnDat encoders transmit absolute position data on demand.
Depending on the version, EnDat can include an analog 1-V p-to-p sin/cos
output that electronics in the controller interpolate to derive incremental data
for real-time control.
The hardware level minimally requires six wires for communications with
distances up to 150m. Depending on the version of EnDat, there may be an

For more information on the EnDat protocol refer to the Heidenhain website
at www.heidenhain.com.

HIPERFACE
HIPERFACE is a proprietary protocol developed by Max Stegmann GmbH. It
uses an eight-wire cable (two for the RS-485 communications, two for power and
four for 1V p-to-p incremental sin/cos). Absolute position data is transmitted
via the RS-485 link at power up, and the system uses the incremental signal
after that. This solves the problem of not knowing position at startup, but it
does not self-correct for momentary data dropouts during operation. With only
incremental data coming into the controller, a glitch can cause the controller

Open Protocol
Connection

Dedicated Serial Interfaces


SSI +
Hiperface
EnDat
BiSS
Sine/ Cos
No (License
No
No
Yes
available)
RS-485: Bus or
Point-to-Point
Bus or
Point-to-Point Point-to-Point
Point-to-Point
Analog: Pointto-Point

Analog Signals
Yes
Yes
No
No
Required
Transmission Mode Bidirectional, Unidirectional, Bidirectional, Bidirectional,
(Digital)
asynchronous synchronous synchronous synchronous
Digital Data
38.4 kBaud
1.5 MHz
8 MHz*
10 MHz
Transmission Rate
Cable Length
No
No
Yes
Yes
Compensation
Protocol Length
No
No
Yes
Yes
Adjustable
No. of Wires
8
68
6 to 12
6
Hardware
Compatible
Alarm/Warning Bit
No
Definable
Yes
Definable
* Heidenhain cables recommended for transmission rates above 2MHz.

Encoder Communications

additional wire that carries an analog incremental output for faster speed
sampling. This is useful for speed-controlling drives that can be more demanding
than the rate at which serial feedback is normally provided.

Encode Communi ations Handbook


to l se racc
n

po

. nli e S I, n at nd iS , t e
-

IP RF
F C di
co tro er.
u
us

e
z
sig all t 12,00
0 rp
p , ltth ug a sol te osi io da a c nn t b se t r lia ly
l
beyy nd ab
a ut alff th
ha sp
p ed.
HIPERFA
FAC
FA
AC c n be intt o- oi t li e S I a d EnDa ut t can also se us
connec io s in wh
w ch eve
v ral nc de s c n be w red to eth r a d dd ess d b a
ve
sin le aster. IPERFA
FAC
FA
AC is the onll o e o th fo
f r aj
a o se ial ro oc ls t at
is a yn hronous.
Lik En a it as inte na memo y t at an ol m to inff r ati n s ch
as v lta e, ur en an ot er param ter . U in
en ed ma ter,
r yo
r,
y ca also
de ne n bso
zer p
I ER A E c n
t e e co er
m ry
ma on an so on. n
on,
.
At t e hard
d ar le
eq

ig t wi

eq
ent a

.
ro rie ar Si Co
a r,
r w t t o w res req r

or

.
F r o e i fo
Fo
f r at on n
ww .s -st gman . e.

IPERFA
FAC
FA
AC , r fe
f r to he te ma n eb ite t

BiSS
BiS is he ate t otocol es n d a d eve
v o d I -H us, Germa . It
ve
was cre te to e n o en no -p opr eta y prot col alt rn tiv
ive
ve to En at and

HIPERFA
FA E ut es gn d ith he ap bil ties of he ro rie ary pr toc ls in
mind. Combined features such as alarms, warnings, diagnostics, and the
ability to store motor information to the encoder are part of the BiSS
protocol.

BiSS uses four data lines, one pair carrying data from the encoder and one
carrying clock data to it, plus two power conductors. It also uses a variable clock
rate of up to 10MHz.
BiSS can address internal registers in the encoder that can be read by and
written to by the master with data about the encoder itself (identification, device
data, resolution, etc.) It can also carry other digital data (temperature, etc.)
and transmit it to the master on demand, without interfering with real-time
operation.
BiSS, like the other serial protocols, can be connected either point-to-point or
via a bus.
BiSS has two modes; sensor mode and register mode. In sensor mode, the
sensor or encoder communicates in a manner similar to SSI. The master begins
to send a stream of clock pulses. Eventually the data line level will drop low
and data sampling will begin. The data can be received and clocked at 10MHz.
Due to the transmission speed, many drives may not require additional analog
incremental outputs to control motor speed. Since SSI has a slower max transfer
rate of 1.5MHz, the propagation and calculation delay is less than a full pulse
width. This is how BiSS is capable of hardware compatibility with SSI. Only the
data length and framework need to be changed in the software. Figure 2 shows
the communication format and a data frame map.
In register mode, the protocol modulates the clock pulse width to address
specific slaves and parameters. This mode is unlike any other protocol. If in the
sensor mode a warning or alarm bit is set high by the sensor or encoder, the enduser may access the register mode and find specifics on the alarm or warning.
This might be an over-temperature warning or, in the case of an encoder, a
disk pollution alarm. Dynapar Group has developed the ability to provide
single step alarms in case improper position is being provided in sequence.

Encoder Communications

BiSS sends full absolute position data whenever the controller polls the
encoder, rather than just at startup. It allows easy recovery from momentary
data dropouts during operation. Since it is an all-digital system, it eliminates the
cost of A/D converters needed in drive systems that connect to encoders using
some proprietary protocols. It is also hardware compatible with SSI and EnDat,
requiring only software changes.

An important distinction between protocols is whether they are


open or closed. Closed communication protocols are proprietary,
meaning that the protocol use is controlled by the originators of the
protocol. Slave device manufacturers are limited to those that are given
permission by the originator, and can often be single source. In some
cases detailed specifications of a proprietary protocol are restricted as
well: they are available to license holders and not the general public.

10

Closed Communications Protocol


There is one good reason to use a closed proprietary protocol; simplicity. For
instance, the communication protocols slave configuration software is usually
pre-designed, saving time and development costs. Another reason may be that a
user prefers a particular vendors product and agrees to pay the licensing fee in
exchange for the ability to completely rely on that manufacturer for all levels of
required support for the encoder or sensor. The drawback to this approach is that
the user limits his/her options in the future, when a replacement device is needed.

Open Communications Protocol


There are several key benefits to using an open communication protocol. These
include availability. Encoder buyers have more selection options and alternative
manufacturers to choose from.
Another is cost. Manufacturers of products using closed communication
protocols can charge any fee that the market will bear for a replacement encoder,
since there is virtually no competition from other manufacturers. On the other
hand, competition among suppliers of open-communication-based products
promotes price and product alternatives.
Closed protocol developers limit the amount of information that is published
and available to buyers. However, if the need exists, an educated user can check
the compliance of an open protocol from a specific manufacturer. Information
regarding that protocol is freely posted on the Internet.
Understanding the operating capabilities and functional limitations of each
of the available feedback control protocols helps in selecting the right encoder
protocol for a given application.

11

Encoder Communications

Open communication protocols, on the other hand, are non-proprietary and


non-restrictive. The protocol developers freely share specifications regarding data
format and electrical design. Designers of products that use serial communications,
such as sensors and encoders, can create their own interfaces without paying a
licensing fee. This means that when you are shopping for a replacement serial
communication device youll find many more encoders available on the market.
The most popular open communications protocol is SSI. SSI is a well accepted
and time-tested protocol. More recently, the BiSS interface open serial protocol
has come into the market. Using SSI as a basis, it represents an alternative to bidirectional communication protocols such as EnDat and HIPERFACE.

Encoder Communications Handbook

Topology
Communication
System
Data Exchange
Max. Length
Max. Nodes
Data Packet
Transmission
Speed
Transmission
Media

Bus Network Comparison


DeviceNet
Profibus
Interbus
Linear (trunkline/
Linear (trunkline/
Closed Loop
dropline)
dropline)
Master/Slave
Multimaster
Master/Slave
(Producer/Consumer)
Polled, Change of
Polled
Polled
State, Cyclic
500 m
1200 m (w/repeaters)
13 km
64
126
512
08 bytes
244 bytes
Flexible
125 Kbps @ 500 m
9.6 Kbps to 12 Mbps
500 Kbps
250 Kbps @ 250 m
500 Kbps @ 100 m
2-wire twisted pair
2-wire twisted pair
Local: 3-pair
with 2-wire bus power
w/shield
twisted w/drain
cable w/drain wire
Remote: 5-pair
twisted w/drain

Industrial Buses
Industrial buses have proliferated in recent years, with many kinds of proprietary
and non-proprietary buses on the market. For feedback applications, there
are three general-purpose industrial buses that are most commonly used with
encoders.

DeviceNet

DeviceNet Cable vs. Baud Rate


Max. Trunk Line
Max. Cumm.
Based on the Controller
Baud Rate Length (thick cable) Drop Length
Area Network, or
125 kbps
1640 ft
512 ft.
CAN bus, which was
250
kbps
820
ft.
256
ft.
developed by Bosch
500
kbps
328
ft
128
ft.
for use in automobiles
and commercialized
for industrial use by Allen-Bradley/Rockwell Automation, this bus is now
administered by the Open DeviceNet Vendor Association (ODVA).
The systems basic trunkline-dropline topology provides separate twistedpair wires for both signal and power distribution, enabling 24 VDC devices to be
powered directly from the bus. End-to-end network distance varies with data rate

12

Device Net supports baud rates to 500kbps maximum at a maximum


trunk line length of 328 ft or a maximum cumulative drop length of 128 ft.
Encoders with DeviceNet have programmable scale, direction, and preset with
position data.

Profibus

Profibus Cable vs. Baud Rate


Max. Segment Max Expansion
This open communicaBaud Rate
Length
(with repeaters)
tion standard devel500kbps
1311
ft.
13,114 ft.
oped by the European
1.5Mbps

655
ft.

6,557 ft.
Community (Europe12Mbps

327
ft.

3278 ft.
an Common Standard
EC50170) comes in
two variations: FMS, which is used for upper level cell-to-cell communication,
and Profibus DP, which is optimized for data transfer with local field devices like
valves, drives, and encoders. There are specific device profiles defined, including
one for encoders. DP is good for applications that require high-speed transmission of fairly large amounts of information (512 bits of input data and 512 bits of
output data over 32 nodes in 1msec).
Initially developed by Siemens, Profibus boasts baud rates to 12Mbps with
a maximum segment length of 327 ft and features programmable resolution,
preset, and direction. Profibus encoders can communicate speed, direction, and
on-time diagnostics.

Interbus
Designed by Phoenix Contact in the mid 1980s, Interbus is the longest-standing
open industrial network. A true token-ring topology, Interbus is actually divided
into two buses. The remote bus is an RS-485 transmission medium with length
capabilities up to 13km. The local or peripheral bus enables connection of up to
eight devices within a 10m range.

CAN
The CAN network (Controller Area Network) was first used in the automotive
industry in the 1980s. The current CAN L2 and CANOpen protocols feature
baud rates up to 1Mbps. CANOpen has programmable preset and resolution

13

Encoder Communications

and cable size. The 0 to 8 byte data packet is ideal for low-end devices with small
amounts of I/O that must be exchanged frequently.

Encoder Communications Handbook


with L2 having programmable direction and limit values. Both protocols offer
speed, acceleration, position, and limit data.
CAN is closely related to DeviceNet with virtually identical hardware and
wiring configurations, differing only in the message formatting.
For many of these protocols, the encoder transmits the absolute position only
when interrogated. In others, the encoder sends incremental data as it turns, and
provides absolute data either on startup or when interrogated. These methods
cut down on bus traffic, but also mean that the control system must know enough
to demand an absolute position update whenever theres a momentary glitch.
The update may take only a few milliseconds, but its a factor to keep in mind
when designing a control system. Some encoders also provide a 1V p-p sin/cos
output for real-time control, since the on-demand absolute encoder data can
come in too slowly for many control loops.

General Wiring and Installation Guidelines


The most frequent problems encountered in transmitting an encoders signals to
receiving electronics are signal distortion and electrical noise. Either can result
in gain or loss of encoder counts. Many problems can be avoided with good
wiring and installation practices.
For starters, take reasonable care when connecting and routing power
and signal wiring on a machine or system. Radiated noise from nearby relays,
transformers, and electronic drives may be induced into the signal lines causing
undesired signal pulses. Likewise, the encoder may induce noise into sensitive
equipment lines adjacent to it.
Route machine power and signal lines separately. Signal lines should be
shielded, twisted and routed in separate conduits or harnesses spaced at least 12
inches from power leads. Power leads are defined here as transformer primary
and secondary leads, motor armature leads, and any 120VAC or above control
wiring for relays, fans, and thermal protection.
Maintain continuity of wires and shields from the encoder through to the
controller avoiding the use of terminals in a junction box. This helps minimize
radiated and induced noise and ground loops.
In addition, transients in the encoder power supply can adversely influence
operation. Typically, encoder power should be regulated to within 5%, and it
should be free of induced transients.

14

Signal distortion can be eliminated with complementary encoder signals (line


drivers), used with differential receivers (line receivers or comparators) at the
instrument end. Also, an encoder using twisted-pair shielded cable ensures that
wire-induced currents will self-cancel.
In industrial environments, high current fluxes are created by motors,
remote control switches, and magnetic fields. This can result in varying electrical
potentials at different ground points. To avoid problems, ground the shield,
together with all other parts of the system requiring grounding, from a single
point at the instrument end.
Grounding requirements, conventions, and definitions are contained in the
National Electric Code. Local codes will usually dictate the particular rules and
regulations for system safety grounds.

Signal Distortion
The majority of signal
transmission problems
involve electrical noise, with
the severity of the problem
increasing with greater
transmission distance. Good
shielding practices help
reduce noise.
The primary cause of
signal distortion is cable
length, or more specifically, cable capacitance. The longer the cable, the greater
the potential for signal distortion. Generally, the receiving electronics will
respond to an input signal that is either a logical 0 or 1. The region between 0
and 1 is undefined, and the transition though this region must be less than about
1 microsecond. As the leading edge of the waveform is distorted, the transition
time increases. At some point, the receiver becomes unstable and encoder counts
may be gained or lost.

15

Encoder Communications

The encoder case must also be grounded to insure proper and reliable
operation. Some encoders have provisions for a case ground connection through
the connector/cable if a ground cannot be secured through the mounting
bracket or machine ground. Use high quality shielded wire and connect the
shield only at the instrument end.

To minimize distortion, use low capacitance cable, typically less than 40


picofarads per foot. Squarewave distortion is not usually significant for cable
lengths less than 50 feet (capacitance up to about 1,000 picofarads). Encoders
supplied with differential line drivers are recommended for applications with
cable length requirements of hundreds of feet.

About Dynapar
Dynapar is a world-leading manufacturer of optical and magnetic encoders
and resolvers with more than 50 years of experience in engineering and
manufacturing rotary feedback devices. Dynapar was founded in Gurnee,
Illinois in 1955 and has been expanded through the acquisitions to include
Hengstler, Acuro, NorthStar, and Harowe product lines. Dynapar is
now uniquely a provider of technologies spanning optical, magnetic, and resolver
based feedback.
Pioneering the first true vector-duty hollow-shaft encoder launched Dynapars
strong presence in several industries, including steel, paper, elevator, oil and gas,
wind energy, medical, material handling, and industrial servo manufacturers.
From small kit encoders to large mill-duty tachometers, Dynapar has the industry
covered.
Dynapar customers rely on expertise and support provided from the U.S. sales
and manufacturing location in Gurnee, Illinois: Phone +1 800.873.8731 or
+1 847.662.2666; FAX +1 847.662.6633; Website: www.dynapar.com. In addition,
Dynapar supports global customers with local sales and production locations in
Germany, Japan, China, West Indies, and Brazil.

Customer Service: +1 800.873.8731


Application Support: +1 800.234.8731
www.feedbackdevices.com/hb905
www.danaherindustrialcontrols.com
2007 Dynapar
Encoder Communications Handbook
P/N# ECH150 (10/07)

You might also like