Control unit for two-dimensional robot
Citation for published version (APA):
Retraint, E. (1988). Control unit for two-dimensional robot. (TH Eindhoven. Afd. Werktuigbouwkunde, Vakgroep Produktietechnologie : WPB; Vol. WPA0542). Technische Universiteit Eindhoven.
Document status and date: Published: 01/01/1988 Document Version:
Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.
• The final author version and the galley proof are versions of the publication after peer review.
• The final published version features the final layout of the paper including the volume, issue and page numbers.
Link to publication
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal.
If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:
www.tue.nl/taverne Take down policy
If you believe that this document breaches copyright please contact us at: openaccess@tue.nl
CONTROL UNIT
loa
TWO-DIMENSIONAL ROBOTEric RETRAINT Universite de
Techno1ogie de Compiegne France
September 1987 - February 1988 WPA-Rapport nr.0542
ACKNOWLEDGEMENTS
I want to express my gratitude to my coach Mr P.C. MULDERS, for the way he introduced me to robotics and
for helping me to progress in my work.
Special thanks are due to Henk SMIT and Gerard
KREFFER for their competence, kindness and valuable
suggestions, and also to the staff of the international relations/external training proj ects for the perfect organisation of the traineeship.
We will remember, for a long time, the nice
atmosphere created in a nice country by the following
team: Miroslaw BOBER, Tom BRUGMAN, Marek FLORKOWSKI,
John JANSSEN. Han VAN DAL ...
EINDHOVEN. February 1988
Contents
§ I. Summary . . . 5
§ 2. The work enyironment... 6
§ 2.1. Structure and lenght of study . . . 6
§ 2 . 2 . The training of engineer . . . 7
§ 2.3. Eindhoven University of Technology . . . . 7
§ 2.4. The Department of Mechanics . . . , " , . 8 § 2.5. The FL.A.I.R. project . . . 9
§ 2.6. My training period . . . 9
§ 3, Defining the Robot , . " . , . . . 11
§ 4. Architecture system . . . r • • • • • • • • • • • • • • • • • 15 § 4.1. Single board computer ISBC 186/03
.
.. .. .. .. 15§ 4.2. Multibus system .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. II 18 § 4.3. RAM memory card ISBC 028A II .. .. .. .. .. .. .. .. II .. .. II 19 § 4.4. Single board computer ISBC 86/05 .. .. II .. .. .. 20
! 21
Sen!uu~i§.l ~l~m~nts .. .. II II .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. II .. .. .. .. 21§ 5.1. Closed loop .. .. .. .. .. .. .. .. .. .. ..
..
.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 21§ 5.1.1. Interface Master/pow. amplifier. 22 § 5.1.2. Position's counter .. .. .. .. .. .. .. .. .. .. .. .. .. 24
§ 6. Interfaces . . . 34 § 6.1. Multibus arbitration
.
.
. .
.. .. .. .. .. .. .. .. .. .. .. .. .. .. 34 § 6 . 2 . Synchronization .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 35 § 6.3. Memory space .. .. .. .. .. .. .. .. .. ....
.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 36 § 6.3.1. RAM memory .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 36 § 6.3.2. EPROM memory .. .. .. .. .. .. .. .. .. .. .. .. ....
.. .. .. .. .. 36 § 6.4. Communication Operator-System ....
.. .. .. .. .. .. .. 38 § 6 . 5 . Interrupt system .. .. ,. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 39§ 6.6. Communication Masters-Torque sensor
. . .
41§ 7. The work atmosphere . . . 44
§ 8. Conclusion . . . 45
§ 9. Appendix . . . 46
Appendix contents
I. Resume en Fran~ais . . . 47
II. Master/power amplifier sheet . . . 48
III. Counter sheet . . . 49
IV. End switches sheet . . . • . . . 50
V. Torque sensor sheet . . . 51
VI. Electronic specifications . . . 52
VII. DC motor, power amplifier •.. specifications .. 58
VIII. Listing of the torque program . . . 62
IX. Listing of the transformation program . . . 67
X. Listing of the torque program extension . . . 68
XI. Multibus arbitration . . . 70
XI I. RAM memory . . . 73
XIII. Operator/system sheet . . . 75
XIV. Interrupt system sheet . . . 76
11.
SUMMARYAs a part of program FL.A. I .R .•
the robot research and application Eindhoven Uni versi ty of Technology is developping a modular robot system. The first module
of that system is a linear actuator. Its major
spec ifica tions are: pos i tion accurcy better than .01
mm, speed up to 1 mls and a maximum acceleration of 5
mls for a mass of 50 Kg.
In order to extend the system to another module, researchers and students are developping an angular
actuator. This module must have the following
characteristics: positionning accurcy better than 5.10- 40 , speed up to 900 /s and a maximum acceleration
of 900
/s for a mass of 50 Kg.
The following phases can be distinguished in the design of the actuator:
- defining the construction; - production of the module; - static analysis;
- dynamic analysis;
- implementation of the control system.
The different elements of the second actuator is developping: mechanical, e1ectronical, electrical and strategy of control.
In chapter §2. of this report, we wil definite the construction in its conception and the teaching model
wi th a torque sensor. The next chapter presents the
architecture of the computer unit which drives the
actuators. Chapter §4. discusses the realization of the
,
sensorial elements which our mechanical robot needs with the outside environment and the computer unit. The
last part presents all the interfaces between
12. THE WORK ENVIRONMENT
De Technische Hogeschool Eindhoven was founded the 23 June 1956 and inaugurated 19 September 1957. Since 1 September 1986, T.H.E. is officialy called Technische
Universiteit Eindhoven, the designation in English,
however, Eindhoven University of Technology, will the same.
12.1. Structure and lengh of study
A full University course in the Netherlands used to take at least 5 years. Moreover. under this old system there was no limit to the actual time spent on
completing the course; the maj ority of the students
needed a longer time to complete their studies.
From 1 September 1982 onwards a new act on the
structure of University Education will take effect.
This act, which applies to all University courses,
restricts both the duration of a course and the period of time permitted for its completion.
The new system divides the study into two phases: - The first phase has a duration of 4 years
and comprises two examinations: the first or
preliminary' propaedeutisch I examination at the end
of the fourth year. Students are allowed 2 extra years to complete the first phase which has to be finished at the end of these 2 years at the latest. After 6 years
of study one may no longer register as a regular
student for first-phase courses in any field of study at any Dutch university.
I
This new system of studies is not very well received by all the students. In fact, after spending five, six months at T.U.E., a lot of students I know are in the old system. They think that this system gave the students almost complete freedom to study or not to study. They could choose more or less freely when they wanted to take a particular examstet.
This system allowed the students to manage their
own life ( jobs with or without any educational
interest, ), to pratice sport or to be involved in
a student association. But the part that is mostly
affected is the extra training period.
For five months, several student meetings were
organized to denounce the new system because many of them are touched by the restriction.
62.2. The training of engineers in the Netherlands.
The training of engineers at University level in the Netherlands takes place at a separate institutes
called Universiteit at these institutes only
academic engineering degrees are conferred.
There are three technological institutes in that sense: - T.U. Delft - T.U. Eindhoven - T.U. Twente founded in 1842; founded in 1956; founded in 1961. Together they have about 21.000 students. A student who
passes the doctoraal examination at one of these
Universities acquires the title of 'ingenieur '( ir. ).
12.3. Eindhoven University of Technology
University. founded in 1956, offers 9 courses of
study in which students can qualify as graduate
• mathematics; - computing science; · technical physics; • mechanical engineering; - electrical engineering; · chemical engineering;
- building and architecture; - industrial engineering.
The University has about 6.000 enrolled students
and an academic and administrative staff of 2200
persons. Since the university opened, about 8600
students have graduated.
12.4. The Department of Mechanics
The Departement is one of the larger ones at
T.U.E .. Design and production are the two main sections
into the which the highly varied tasks of the
mechanical engineers are divided. Their task varies
from scientific research and development to industrial organisation.
The courses reflect the eight topics on which the Department focusses its research efforts:
- control and simulation of mechanical manufacturing processes;
• biomechanical technology of vital human biological functions;
design of process equipment for maximum reduction of life cycle costs;
- non linear dynamics and random vibrations; - high efficiency power transmissions and
I have worked for six months in the W.P.A.
division ( production engineering and production
automation) as part of the FL.A.I.R. project.
12.5. The FL.A.LR. project: FLexible Automation and Industrial Robots.
The research project FL.A.l.R is financed and
directed by the Dutch government. Its aim is to get
some experience in flexible automation and industrial robot.
In this project each University has its own task.
At University of Eindhoven, this project is called
F. A. I. R. . Both the mechanical and electrical
Departements are invo 1 ved in this proj ec t as we 11 as several private companies.
The project is divided into five sections: - the general aspect of automation; - the handling of parts;
- cinematics and dynamics of mechanical
structures;
- the drive systems. the control systems and aplications of those systems;
- the arc-welding and the sensory system.
12.6. The project of my training period
After having developed a one axis robot arm, a new subject of research was defined in the laboratory with
the aim of extending the robot to two degrees of
The subject has been defined as follows: - Design
actuators.
a computer unit driving the two
- Equip this latter of its sensory elements which allow i t to interface with the outside environment and its mechanical framework.
The mechanical framework was designed by another student at the same time of my project.
My project was split into three phases:
- During the first six weeks, 1 analysed the
existing system and the possibilities for extension.
- The second part was to design a new
architecture.
- The last phase was to add to this control unit
sensory elements composed of hardware and software
13.
DEFINING THE ROBOTThe extension of the robot consists to bring an angular actuator to the linear robot arm. Each one is constitued of two parts:
7
- A mechanical part which is composed of two elements: - one motionless part (immobile)
- one mobile part with a screw and the aotor.
1: Support
2: Table
3: Arm
4: DC aotors
5: Hall effect switches
6: Bncoders
7: Torque sensor
- A drive .ystem containing a contro1er ( computer
unit ) and for each degree: a power amplifier, an
actuator ( DC motor ) and an encoder:
The following figure shows the general
-
DC MOTOR -CioMI:>I I ... ~B
~
~ ~-
OCMOTOR •-~
~ O~ • ~ JO:I:NTIn order to teach our modular robot we used a method which integrates a 6-axis torque sensor in the robot. This method allows immediate programming of the
robots either for paths or, at the same time, for
forces and couples to be applied to the robot's
environment. It avoids all the dificulties inherent to
any kind of off-line or CAD ( Computer Aided Design )
programming caused by uncertainties in the geometry of the robot and the material to be processed.
This method is based on a torque sensor ball
through which the human hand forces/couples are
•
During the instruction mode the robot is used as a
sensory controlled robot. The control is then a
function of the information from the end effector. Each actuator is driven by a procesor unit while a memory allows storage of the recorded path.
-~~"",=_~,
...
- - - I "m:
.-,.th-onl,-te.Ch-in with robot .aunttd sensor bill
Therefore, the control of our system is not
simple. In fact, the dynamic equations are highly non-linear as well as interdependent.
A model of the dynamic equations:
•
('IX + '2 X + '3)d 0/dt - '4X dX/dt d0/dt +
'SdX/dt d0/dt
+
'6 d0 / dt + '7 X + '8 + '9 U2with: - ( d X/dt dX/dt, X ) linear state space
- ( d 0/dt d0/dt, 0 ) angular state space
First, the control will be composed of an adaptive
trajectory control to treat the load disturbandes
acting on the motor's shaft. adaptive trajectory control control.
Next, i t will contain an and an adaptative force
In this regard, a new architecture has been
designed to obtain these characteristics and to keep a modular concept.
14.ARCHITECTURE SYSTEM
In order to develop a modular robot system, each degree of our system is controlled by one single board computer.
extension
. This operation permits a two
without modifying the first
architecture.
degrees degree
Before the extension, the first degree was
controlled by one single board computer, ISBC 86/05
from Intel's system architecture. Therefore, the second
degree has b~en equiped with the ISBC 186/03 which is
similar in architecture to the first one. 14.1. Sinale board computer ISBC 186/03
The ISBC 186/03 single board computer is a general
purpose 16-bit computer system ( 8 MHz IAPX 80186
microprocessor ) on a multibus-compatible printed
circuit board.
A brief list of key features:
- IAPX 186 high-integration microprocessor;
- IAPX 86/30 ( 80130 ) opreating system processor;
- Eight byte-wide memory sites for ROM, RAM and their derivatives;
- 27 interrupt sources on-board using 80186,
80130, and 8259A interrupt controllers, and the
I
8274 serial controller;
- Two serial Input/Output channels controlled by an 8274 serial controller;
....
-
...
--_
...
-
....
--_
.. .
.
.....
-
...
-
...
--~·
l
·
·
·
! !:
... _-: !!
i
I ! : ! ... __ ... _____ . . . _ ... __ .... - ____ .1 H...L T:tBlJS SYSTEM ~-...
---
..
.. ... _ ... -...
__ .._--_
...
--: : : : !!
12errs
!
!
!
.
.
! !;
v
i
i!
; = ! ~v
!L.~
...
~
..
=~~
__ ... _
... _
...
~~
.
.1
:-...
~.-~....
: : ! ! : : ! : ! :. .. - - . -. . . oo __ oo ... _ ... ;I/O expension;
- Master capability on the multibus interface .
.t"
e",. OIIA . . . tIT . .
I""
lIIOIty• 'OUR.m : . . "A ... IO.. •
I •
"'---MUL MUI. '''TIM .U,
Therefore, equations, the as seen control previously in of one degree the dynamic affects the
second. In fact, each master needs to know the state space of the complete system.
Consequently, both masters have been inserted in
the multibus sytem which allows a multimaster
14.2. Multibus system
The multibus interface is a general purpose system bus structure. It contains all the necessary signals to
allow the system components to interact with one
another. This system is based on the Master Slave
concept. The handshaking protocol between master
and slave devices allows masters of different speeds to use the multibus interface and also allows data rates of up to 5 million transfers per second. The multibus sytem bus can support multiple master device on a back plane and can directly address up to 16 megabytes of memory.
The multibus system is designed to perform 8-bit and 16-bit transfers between single board computers,
memory and 1/0 expension boards. Its interface
structure consists of 24 address lines, 16 data lines, 12 control lines, 9 interrupt lines and 6 bus exchange lines.
We have just seen that the multibus is built upon
the Master - Slave concept, using the ' handshaking'
protocol to transfer data. Therefore. the two masters are not equiped with a multibus accessible memory. We then inserted a RAM memory card ( ISBC 028A ) into the multibus system in order to transfer the immediate
control values and to stoke the teaching robot
14.3. ISBC 028A; IAH memory board
The ISBC 028A Random Access Hemory board provides a dynamic memory storage capacity of 128 K bytes. This RAH memory interfaces directly with the bus master via
the multibus interface in any 8 or 16-bit ISBC
operating board. - Access time: Read: Vrite: - Cycle times: 500 nanoseconds maximum 343 nanoseconds maximum.
Read, write and refresh in 608
nanoseconds maximum
Being part of the research project FL.A. I .R. a torque sensor has been developped. Ve have then adapted i t for teaching operations.
The analysis of an appli.d torque n.eds to be comput.d. The compl.t architecture has been realized by adding a single board computer ISBC 86/12.
14.4. Sinsle board computer ISBC 86/12
This last part of the system is characterized by its dual port memory architecture. Dual port access allows the on-board CPU to address the RAH directly via the on-board bus and the other CPUs to do so via the multibus interface.
The ISBC 86/12 also includes a 16 bit Central Processing Unit ( 8086 16·bit HMOS with a clock rate of 5 MHz ), 32 K bytes of dynamic RAM ( dual port ), a serial communication interface, programmable timers, a
priority interrupt control and three programmable
parallel I/O ports ( 3*8 lines ).
In conclusion. our multi·processor architecture
shares the memory, thus keeping the whole multibus
control for the two masters. As a result. we
significantly reduced the amount of system bus traffic and the waiting times.
15.
SENsoaIAL ELEHENTSOur mechanical robot needs sensorial elements in order to communicate with the outside environment and the computer unit.
'5.1. Closed loop
Each actuator of our robot is driven by a DC motor ( MC 19 p from the Electro - Mecanique Compagnie: CEM ) which needs a power amplifier unit ( Axodyn Series 05
LV power amplifier from Brown Boveri Company: BBC ).
Two encoder system give the position of two
degrees. A discrimator of the sent signals makes it possible to know the direction.
I
I
..
,
t
..-,
~ ~_l_"''''''''.."
1---'"
II
t
_ft J..
,-...
:r.:~c. l V ' l \ T \!~
...
*--...
•----is.1.1. The interface master I motor power amplifier
An interface has been designed to drive the motor
of the table. A digital-to-analog converter (
DAC8llKP ), an operational amplifier ( OPA27 ) and an invert gate have been used to realize this interface.
The DACal1 is a complete single-chip
integrated circuit microcomputer-compatible 12 bit
digital-analog converter. The chip includes a precision
vol tage reference, microcomputer interface logic,
double buffered latch, and a l2bit D/A converter with a
voltage output amplifier. Fast current switches and
laser-trimmed thin film resistor net work provide a
highly accurate and fast D/A converter. The DAC is
fully specified for operation on +12 V and -12 V power
supplies. Then the DAC can produce bipolar output
ranges from +10 V to -10 V.
Remarks:
- For optimum performance and noise rej ection. power supply decoupling capacitors have been added.
- Offset and gain have been trimmed by
installing external offset and gain potentiometers.
The invert gate has been
input post tive NAND gates ( 74LSOO
obtained by a ). It is used
2-to invert the most significant bit of the sent l2-bit.
This has been done to obtain a nice conversion
characteristic, according to the two's complement
notation for negative numbers.
Two other NAND gates ( AND gate ) have been used to select the DAC.
OUTPUT DATA ( FFF )
---1-+ Vmaxo
- Vmax DAC Conversion ( FFF ) ---1---.-- Vmaxo
+ VmaxThe interface must allow us to command the robot power supply by inputing in its drive system a voltage
from +10 V to ~10 V. An operatonal amplifier OPA27 has
been used to obtain a good precision vol tage
amplification. The OPA27 is an ultra-low noise, high
precision monolithic operation amplifier.
Laser-trimmed thin film resistors provide exellent long term
voltage offset stability and allow superior voltage
offset compared to common Zener-Zap technique.
The ISBX ( connector J7 ) has been used also for this interface, because i t provides the necessary power
lines: +12 V, -12 V, +5V.
Software:
A word is sent to the on~board output address 80 H
§5.2.1. Position's counter.
In order to know with precisely the angular path, the system needs an encoder ( LIDA 360 . HEIDENHAIN ), which sends a signal to the single board. This encoder has the same features as a linear one with a grating pitch of incremental track ( .1 mm ) and a scale tape ( circle of diameter .6 m ).
A system permits to transform and to improve the accuracy of the signal coming from the encoder. Indeed, this signal is sinusoidal and in order for i t to be used by the single board, i t must be transformed in
square rate by an interpolation system ( EXE
710-HEIDENHAIN ) which improves the accuracy by 25.
An interface has been realized to count the number of pulses. The device must count per rotation
[ circumference ( *.6 m )/ accuracy ( 100 m / .3 )
] 472 000 pulses. We thus need at least 20 bits to code the robot's position. The maximum angular speed for the table is 900
/S. Then, the frequency of this signal is
[ maximum angular speed ( / 2 rad/s ) / angle of the
accuracy ( 100 m / .3 m / 25) ] 0.12 MHz. We have
designed this device with two counters in cascade. The first is a 16 bit . direction discriminator ( SN74LS2000 ). The device we designed has required to
evaluate the outputs signals, received from an
incremental length measurement systems or any others
incremental transducer such as robots. It is
recommanded to use a clock frequency which is at least 4 times higher than the frequency of the signals ( ».48 MHz ). We use the clock of the ISBX bus which is of 6 MHz.
feeding the borrow and carry outputs to the count-down and count-up respectively of a second counter.
The second counter is a 4 bit synchronous
reversible ( up/down) counter ( SN74LS193 ).
A 4-bit D-type registers with 3 state outputs (
SN74LS173 ) has also been used as storage for binary
information between the ISBX bus ( connector J6 ) of
the single board and the outputs of the 4 bit • binary counter.
Selection signals:
IORD/ and IOWR/ give the signal respectively to read and load the counter.
AO allows to select the least-significant byte ( high level ) and the most-significant byte ( low level
) on the direction discrimator. and gives the signal to load the latches.
lORD, IWRD and the address line are common all the bus on the single board, we use MeSO/ of the ISBX bus to select the two counters.
Read/Write protocol:
Writing: to load the counter we use two on-board output addresses:
- l2-upper bit: sent the 12 lower-bit of one word to the address : AO H
- lower byte: sent a byte to the
address: A2H
Reading: to latch and to read we use two on-board input addresses:
- to latch and to read the l2-upper bit with the address: AO H
- to read the lower byte with the
addresss: A2H
§S.l.). Two end switches
In order to deliminate the ends of movement, each degree is equiped with two hall effect switches. Two Nand Schmitt Trigger integrated circuits have been used to fit up the wave form.
The two outputs of each degree, translation and rotation, are respectively connected to the interrupt matrix, IRO and IR1, of the ISBC 86/05 and 186/03.
The following scheme gives the end switch positions for one degree:
- - - - I
• Releasing input Molor { connection .5 LV .1 i_._._
...
Emergency 011 S21
--~
I
---+.
'
rt ;15.2. Torgue sensor
The 3wdimensionna1 torque sensor is composed
because of its mechanical structure in a11uminium and of its 8 strain gauges.
The calculation of the 6 components. one force and
one moment per axis, are based on the following
relation: wi th:
-v -
A F F - [ Fl' F 2 • F3 • M1 • M2 • M3 ]t F: applied torque A - [ all' . . . a16 a8l' . . . aa6 A: conversion torquewtension ,V - [V l ••..•.••..••.•. Va ]V: tension from the 8 strain gauges
As we know the matrix A. we can obtain the applied torque:
with:
F - A-
l
V
u
Each tension Vi is amplified by an amplifier INA 101 ( gain 86 ). See appendix V for schema.
This entire electronic part is integrated in the case of the torque sensor.
~.4.l. Electronic card
In order to interface between case and the single board computer ISBC 86/12, an hybrid data acquisition system has been used: SDM 186. This makes i t possible to select one tension by a multiplexer and to obtain an analogical to digital conversion.
An INA 101 amplifier is put between the output of the multiplexer and the converter to adapt the tensions to the range of the converter ( gain of this amplifier is 25 ).
A digital to analogical converter, DAC 08 CBI-V, allows to eliminate the offset of each channel.
---
IJm:H---
- .
5 t 4 t 2, SO,ftwareIn our case, only three components of the torque
needs to be computed, In fact. the control of each
degree needs to know the respective joint torque Tt,Our system can be configured as follows:
T1
/
F1
N3
-I
l'
T
2 [ N3
+
r/\
F 2 ]
IThe prismatic j oint torque. T l , is a function of
the applied force along axis 1 ( Fl ). The revolute
joint, T2 • needs to know the arm postion. force along axis 2 ( F2 ) and the moment along axis 3 ( N3 ).
We can derive the equation as follows: S
-
11 V with: S - [ Sl' S2 ]t-
[ Fl' ( F2 • N3 ) ]t-
[ Ek • 1t V - [ Vi' ]t i-
1,.
. .
.
. .
.
8 k-
1 ,. .
.
. .
.
. .
.
3 in another form:if we consider the time t, we can write:
with
These two equations are the basic algorithm.
Al&orithmj
In order to minimize the OE k calculation time. hardware and software work in parallel:
- The software is decomposed in three phases: selection channel i;
reading Ui ;
calculation of the variation: Aki Uit+Ot - Aki Uit
and the Ek k 1, 2, 3.
- The hardware in two phases:
stabilization of the analogical tension; analogical to digital conversion.
Algorithm timing
SELECTION
CHANNEL
I
+
1
M I..
J
ICALCULATIONS
STABILIZATION
CHANNEL
I - 1
CHANNEL I
+
1
-uki Ui
- Bk
k= 1,2,3
c
I .. I IREADING
CONVERSION
CANNEL
I
CHANNEL
I
+
1c..J
I :L:
0,1, •••••••• 7:JEvery 8 loops, we have renewed the torque S.
Remark:
The offset of each channel must be relieved before getting in this infinite loop. The program comprises a
first part which makes the average of several
statements in order to decrease the influence of the noise.
Variables:
- NOFF2S1:
,
contains the value of the offset,LSByte: bit 0 - 0 bit 1 to 3 bit 4 to 7
• CFlSI, CF2SI,
address of the channel MSByteoffset
CM3SI are respectively the
coefficient lines of the force I, force 2 and moment 3. - FlSI, F2S1 and M3SI respectively correspond to the product Aki Ui with k - 1, 2, 3.
M FORCEl, FORCE2, MOMENT3 make up the 3 elements
of the torque.
Remarks:
The biggest coefficients Aki of the kline ( k -I, 2, 3 ) respectively need to be coded with I-I, 10, 5
and one signed bit each. The digital tension Ui are
coded with ll-bit and plus one signed bit.
The two forces F l , F2 need to be coded with 3
N3 with 2 bytes and 1 bit. The bytes and
LSByte of LSBit.
moment
and are eliminated and for the
This manipulation increases the number of errors:
OF 1 -
±
51 mNOF2 - ± 51 mN
ON 3 - ±.2 mNm
M Relation between the decimal and hexadecimal
values: Fl
-
Fl [ D H*
.0256 ] ( N ) F2-
F2 [*
.0256 ] ( N ) D H 10- 4 N3 N3*
2 ] ( Nm ) D H- The program has been written in Assemblor86 in
order to ~ptimize and minimize the calculation time. An
extension of this program to six degrees is given in appendix X.
6. INTERFACES
This paragraph gives a survey of the
communications. interactions. synchronizations and
articulations among all the elements.
6.1. Hu1tibus arbitration
We have 8een that one master can effect data
transfers via the mu1tibus through a generation of
command 8igna18 and. memory or I/O address. But our architecture system possesses more than one bus master. In this case. the mu1tibus must be arbitrated so that each master can take control of the bus as it
needs to effect data transfers. The multi-master
configuration requests bus control through a bus
exchange sequence.
All the techniques are based on the priority
concept. At a given time one master will have priority above all the rest. This occurs when more than one master requests control of the bus at the same time.
We have chosen to use a serial priority
resolution. This technique is accomplished with a daisy chain. The priority output of the highest priotity master is connected to the priority input of the next
lowest one, and. so on.
Therefore, a limited number of
accomodated by this technique, due
through the daisy chain. The current
mas ters can be to gate delays Intel mul tibus controller chip on the master boards up to 3 masters may be accomodated.
This configuration defines in our system, higher
( ISBC 186/03 ) and lower ( ISBC 86/05 ) priority
masters to access to the multibus. The lower priority master obtains and keeps the bus when a higher priority master is not accessing to the system bus.
Remarks:
- In control loop the necessary number of bytes for each master is small. Consequently, the lower priority master can access to the system bus very quickly. So the higher priority master accesses to the bus when the former completes its present transfer.
- Data transfers occur with a maximum
bandwith of 5 MHz for single or multiple read/write
transfers. Due to the bus arbitration and the memory access time, a typical maximum transfer rate is often in the order of 2 MHz.
See appendix XI for the jumper comfiguration and the arbitration's protocol.
6.2. Synchronization Haster-Haster
Two interrupt lines of the multibus are reserved for the communication links between the masters. In fact, arbitration and synchronization are necessary for
the evolution, in parallel, of the two control
programs. The one to direction of another; the the bit two interrupts l-port CC ( are opposed Programmable Peripheral Interface: PPI ) are respectively connected
Remark:
before we use the interrupts, each PPI must be initialized by the control byte ******1* to the on-board ouput address OCEH.
6.3. Memory space
6.3.1. RAM memory
The RAM memory card, ISBC 028A, is used to stoke
the control elements of a recorded trajectory ( state
space, tension, etc) at each sample time. The time of this traj ectory is defined by the number of sample
times we can stoke. A part of this memory,
nevertheless, is reserved to exchange the immediate
values for replayed trajectory.
The on-board RAM memory of ISBC 86/12 which is accessible via the multibus, holds the recent value of
the applied torque. 3 words of 16 K bytes are reserved
for this application. The part only accessible by the
on-board CPU contains the necessary data for the
torque-sensor program.
The on-board memory of the two masters is also used in this latter case to contain the DMA transfer block from the ISBC 028A.
See appendix XII for the hardware configuration.
6.3.2. EPROM memory
For the ISBC 86/12, 4 K bytes of Erase Programming Read Only Memory suffices for the torque-sensor progam.
IISBe
o28Al
RAM
128 K Bytes
AOOOO - A3FFF
I-tRAM
16 KBytcs
a4000.07FFF
6.4. Communications Operator-System
The communication between the operator and the system is effected by two different proceedings:
- the first one uses a serial Input/Output
communcation between the system and the system console. One master of the system is only interfaced with the console, the ISBC 186/03 via its RS232 port ( connector
J2 ).
- The second is based on the inteI1"upt
technique. Each master is connected to a mechanical
switch which gives a signal to the on-board interrupt
controller.
The former allows a choice of options with the console and the latter does to start or to stop each option when the operator wishes it. This decomposition avoids reading of the console, via the RS232 line, to know if a key has been pressed.
Remark:
IT appeared that the switch contact did not open or close cleanly: at closure there are several seperate contacts over a period of few micro- seconds. It gave more than one interrupt. A switch debouncer removes the
bounce by responding to only one of the voltage
excursions during each switch operation.
6.5. Interrupt system
The two master computers need to interact with some different off-board and on-board elements in using the interrupt proceeding. For those applications. each single board is equipped with one master controller
( Programmable Interrupt Controller: PIC ). which can be extented to other slave interrupt controllers.
Remark:
All interrupts. execpt the Non-Maskable Interrupt NMI are handled by the controllers.
The 186/03 and 86/05 single board computers are
respectively controlled by the 80130 operating system
processor and the 8259A master controller which
provides 8 interrupt levels.
Interrupt sources with priority levels:
- IRO and IRli off-board
Those two interrupts respectively allow to define the two end positions of each degree: right and hall effect switches.
off-board
Hardware key for the Operator-system communications.
on-board
This line comes from the interval timer 0 and is used to define the sample time.
off-board
This signal comes from the other master, via the multibus, to synchronize the contol programs.
off-board
This line defines the zero-position which is obtained by the reference signal from the encoder.
+ 2 off-board lines for the linear degree:
l&i
and lR2:These two lines respectively correspond to the borrow and carry signals from the counter.
REAL
6.6. Communication Masters-Torque sensor
For the control. before beginning a new loop. the two masters need to relieve the sensorial elements. These values must approach the robot's state at this time. For the torque sensor there exists a time between
the request and the calculated elements. This one
increases the effect-answer time.
Traditional channels.
synchronized interaction: 8 latched
sq q
,
,
.'I~,
EF'f"'ECT~,
9,
i t t ~ TD1E :,
.
.
1
:
: :.
.
~HOTORVOL TAGES yq ii:~ y, ,-~ II
:
Y'i~·
I
i :.
,
,
!
i
I
l'
i
A
·
A
l'
.
!!
C()N1"ROl.. LOOP F"ORCE SENSOfiit :!
:,
·
.
! i !I
·
·
j
I
l q &2..
q)·
l qt1~ ! !!
:~
!
/:'
·
·
i
/:'
~
'V
V
.
i
·
i!
; !·
i
U!+11
t
c't&1
.
Fp?I
·
Ii.•
•
In order to minimize this time. our system is not synchronized with the two masters. This is possible if
we can immediat1y answer to the request from those
later.
This is realized without latching the 8 analog channels. but measuring each channel in proportion to
HOTORVOL TAGES FtEAL UF"ECT-ANS$-ER T:tt£: OONTI'«>L LOOP F"ORCE SE:N:SOR
Ok;
Uj
I
i
=
1. ...
8
I
~~i
y~,
~~13
fll+
A
!i
I :
I :
! P-! II ! ' P.
!
!/:'
-1'
!.
.. I I II
eMWt
11~1~1:
( i ) Comparison:•
~ ~.r-....
I I "I;
II
I',
sel IS
~B,
A
A
!
: :L
LPtlllI
I4'
.
r~1>
I
1+2)
...
ti
m
(? 1E
k=[rn
~
,-
U
'I1
t' I, -~ t (h
The information transfer of the applied torque from the program to the masters is realized via the
mu1tibus. That, composed of Force 1, Force 2 and
Moment 3, is placed into the mu1tibus accessible memory
part of the ISBC 86/12.
Addresses: MSByte-LSbyte
-
Force 1: AOOOl-
AOOOO-
Force 2 : AOOO3-
AOOO2-
Moment 3 : AOOO5-
AOOO4Remark:
When both the CPU and the controlling bus master need to effect data transfers to or from the on-board 86/12 RAM their operations are interleaved. The dual
port controller cares about word or byte transfer
17. THE WORK ATHOSPHERE
I would like to mention in this
friendly was the work atmosphere during period.
report how
my training
After only a few days, I was very integreted not
only into the labotory but also with the people I lived wi th. In this short period of time. we did a lot of things together not only in our work. I want also to mention that the french are really apreciated in the
Netherlands. As a consequence, our integration is
really easy because a lot of people here like to work
with the french. At the beginning people were so
friendly that I was really surprised by their attitude but now, I am used to it and I appreciate very much that attitude.
After these six months. I consider that I have
very good friends in the Netherlands. not only dutch people. but also foreigners. In fact, Eindhoven is a real international town.
As a conclusion,I just want to say that I enjoyed very much these six months in the Netherlands and I will have a lot of difficulties to leave all my friends and this marvellous life.
18. CONCLUSION
My training period, in a laboratory of the
Eindhoven University of Technology, has allowed me to
understand what was the development world and the
necessity to communicate between the menbers of a same team.
In a thechnical field, the conception of a control unit and its means of communication has brought to me
one aspect of the robotic which was, until n~w,
unknown. This last part comes to complement my
knowledges about mechanics, electricity, electronic and theory control.
For the complete realization of this project i.e. a 2D adaptive contol robot, it remains to realize the
connection between the actuator and the control unit,
and the most important part, the implementation of the adaptive control.
But what
difference of
realization of studies.
I remember of this period, is the
approach existing between planning,
I. Resume en franQais
Mon deroule
stage professionnel de longue duree s'est
a
l'Universite de Technologie d'Eindhoven ausein du Groupe Production et Automatisme rattache au Genie Mecanique. Ce dernier travail dans Ie cadre d'un projet gouvernemental. FL.A.I.R. ( FLexible Automation
and Industrial Robots ). visant
a
rapprocher lesentreprises privees du milieu universitaire.
Le projet
groupe a pour
condui t par un
d'un module de
attribue
a
un des laboratoires de cebut de realiser un modulaire robot
auto-adaptatif contr6le. Sur la base translation existant et un module de rotation en constuction. mon travail consistait en la conception d'une unite contr61ant ces deux actuateurs.
Le stage s'est decompose en trois phases:
- une premiere partie de recherche afin
d' acquerir les elements necessaires dans les domaines du contr61e adaptatif et des ordinateurs de contr61e;
- la deuxieme partie
l'architecture de l'unite de
d'elements du INTEL's system;
fut de
contr6le
realiser composee
- la derniere partie fut d'implanter
elements sensoriels
communication entre
et les differentes voies
l'unite, les actuateurs.
structure mecanique et l'operateur.
les de la
HeSl/ 20 V
kU~c ~
~~
'-FV 13 V 74LS00~~
HAl 9 V MACa 11 V ~U~A !3 I'Dll 42 Vt...2..
74LS00f
IS "1>10 41 V~
UIBi
44 V 74LS00 4 43 V : 19 Vl
: : 21 Vl
t-b4
23 V'-..
'-..
'-..
'-..t
2!5 V 1,
"
"-~ I;l; ~ ~ ~ ~~
f
N N Nk
~B
c B A~
OACal11<P C§
2
~
v
~ R + 0 ~~
0B
~ ~B !
1 :I:'t
N't
~
1-5( M 2 N M CO(
r
,/ ,/ ,/,/ Ht!
I!InF'/N6
N6
«::'::1 ~ GN> 17?
Mp0 33>
31"
R:3 SH)i
RU1 : as ~t-ba
27>
... + ...
i" ...R10'0k
I
R100.<
GIN:) 3"
::}: CILF*
CluF"::!=
Cl!..F -.12 V 2 +IS V Ie"
"
...
-
...
_...
--
...
--"'
...
-
...
_--
...
-
...
-_
...
--:t:eax sue JS-~
+~
~
~
~
i
:t:tt
~
i
~.-...
~~
~.-.-.~ 6...
-
...
0 V t;I 0,1
...
3,
,
~ ~ ~5
Y
~ ~ ~g
f
6 5 ~~a1~ ~a1~~
"
') ') ') ') 1 ') ') ')
"
"
"
"
~~ ~~1-~
12l"~
~~UIA
( ,.. ute
74LS02 '741.Ji!H'7~ 74LS02Y
4
CLR G2 L...! ~ GJ. ) N~
H CLf<~~
QA :3 1.~ D4 Q4Etl
V
Ule ~ 1~ 74LSC!t4 QB ~ 1.3 03 Q31-.a.
5
QC QO '7 1.4 02 OJ. GI2 GIl ~5 UP co
5ii:
Ie> 4I
1.1. ONeo
J I I) ) /
1 4 ' L<>t=tD J.4)) ) ) ) ) )
/~ 74LS.193 CLR~ ~
i
~ ~ ~ ~ ~
s
~
'l
8
RM
~
0~
Ya
-
-
:3 !~8
(3~ ~
tt
121
~ ~ ~
9
i~ ~ ~
-
E
\/ute
~ '74LS04 1.5"~111 ...,
'" '"
'1
1 ...
~1
29 )f4
.... ""...
_to:!I
~
-1<1 I : : 74LS2000 Dl:R£CT:ION O:ISCR:IM.l:NATOR I !-
5 V C2.2 t.F ~ :-_ ..._-~
:!
1 R2.2 K·
I-
i H :t
i 2 ! :~
!
R2.2 K·
l
j 1,
i I j : 2I
i ! ...~
::-_
.._--1
·
i : i§
i I 2 !~
!
R2.2 K ! 1 ! :~
!
·
~
,'i!o.j'f.J'
11 TI
I
I
/I
'
.
.
+.
% IV .-:s 4 ~'"
G
!-"
~ ~ i21 :z::';; 1-;;; a
~ 4:d ~ 22 SO 261
u
;. ..•
~~___________________
~nHL-~1 ~ r _ _ _ _ _ _ _ _ _ _ -'1l~(,"_ _ _ ' ;.-<, Inr' li15II\ .'I
"
I
n
,
"
I r-_~---~n~'~-~~? 38"q
, :>, ,,8 J...---D 'l:L 4 - :I
~I
~b 1) 7 ~s ]) 6 I "D3,'t)f1~; ~
Z
I
,...;
_
_____
....::
l)
""2
i
~
n
.!o:1.
0
"_
e<o:'
: ;~; ~~
1
~
Dl/\)~
, f I' IU~~~ _ _ ~_' _"_·_· ---~=··===···~· ~'~· · --~----~~~ ---~O~i ! 29 D <t> (LSS) I .. , h":~8 I fo! :z ;"
W
E
[!
Ng3L
__ ,.... _ _ _
.:.:..
_ _
r---...(
2~
1
Hi~
_________
l"'S;,;:\l;.:./:.:MsB:;!;.._· 1~:
,~
Ei<2J:
i
I
.
f\~
«
: ; : ' NI
LS~
'/,r'
I
::
:
i
:~
20 A' IN I1
.-I--___
...!:.>.L.
_~
21 A.2 IN (HS~\ . ... . _ 1 r--.. _-:--_ _ , . .c.'.~~
_ _ _ I _ _ _ _ ._, .. _ ..---_
.11'
.
I 6,,,,,,,,,,, p24~I---ii1-"llJIlI.T.J. '.!!"~l>··".ll:~\\k 1-_ _ --<> ' 12 ;,
'--_+--1 i6 'C\I' .',',n .. \(,~ Nt .. 11,1 / 2 ~l; 2 4. 0, T ' 6 ~I' 6 ; 'bIT 1 8 ~ITll 9 &IT.9 1.0 SIT 10 i t ?IT 11'
,,,
"" ."', ' 8 40 , ." if.,
...
,..",. ) ( ;,:J3 T'I
&:JL 't cc,J 6:11 r.I
b./L• .. ,,
7
~
b.-If : . J; J"f~ c •• ] 6:13 0," p o,
u
/l
- ceO;
u f -6dl .:;"
.,'11 ~ j:r 5 l"l .I,:lf 1,/ ....:... "',"' T" -~,fj f1 J:/( 11 ~ ;.15 r 9 •. if [<IJI
LN 7't7 1)~ ,
,
.
",
.
"
.
~
;::
'_.=
\.=
~
'.
~':..'
.~
":..
.
~A:..M~~I~N=T
:
,
E=R=f=/<.=(=E~T
.
:
.
B=y::
.
,
::
3:=D=K:R:A~(fI~1:5=
.
E=N:S=D~R~
;"U'"(;I''' 'lG~
"IO;
:
!
"
T
.vW .. llta .. · .... OI'. VIC.I N, .. <10\JU:2S,. ... .f"L
~
![(H~IIC EI~DHOVEHVI. Electronic component specifications
VII.A. Di&ital to analo&ical converter: DAC 811 KP
Mark: Burr-Brown Type: DAC 811 KP
, ..
...
~UST l-+-tM~I.r. II Ian Il-If-+-<r--4-0-f ", 1,1Nonlinearity:
±
~ LSB over temperature.Slew rate: 12
VI
sVII.B. Operationnal amplifier: OPA 27 Mark; Type: oil
..
Burr-Brown OPA27 FZ Offset yo1tage:Input bia§ current:
Slew rate: 2.8V/ ~Mii; 126 dB at OI/Tl'UT 10 V and .2 V/oC 15 nA s V of ±11 V
VII.C. Direction discrimator: 74 LS 2000 ... N\ "2 UOI 11 Uo2 - - - - t MEASUREMENT LOGIC t---+-... , ----~~~T_-T~~~ Rm
a
AO 10 CQtfTROl lOGIC iinliiLli_---_..J
Mark: Texas Instruments
Type: 74LS2000
%)0 • D7
Capacity: 16-bit cascadable up/down counter
Mode: frequency measurement with either forward or
YIII.D. Hybrid data acquisition system: SDM 856 KG
Mark: Burr-Brown
Type: SDM-SS6 KG
R.esolution: 12-bit
Linearity error: ± .0012 %
Conversion time ADC; 25 s
VIII.E. Digital to analogical converter: DAC 80-CBI-V
Mark; Burr-Brown
Type: DAC 80-CBI-V
Non linearity:
±
.012 ,Time conversion: 3 s
VIII.F. Power amplifier: INA 101 AM .
-<!XD
Mark: Burr-Brown Type: INA 101 AM Volta&e drift: .25 Offset yoltase: Nonlinearity: Input impedance: 1 --25 V .002 ,CMRR:
106 dB at 60 Hz Settlins time: 50 sVII.A. Power amplifier
Axodyn-Power Servo Amplifier 05 LV 05 B.B.C.
Table 1 Summary of Specification
Axod~
Power supplies power servo amplifier
Tfl)e Order No. Supply Power GJV 160 .. Voltage Con-sumption V VA 05 LV01...-E ... 1001 R1 220Vll- 200 05 LV02 ... -E ... 1002 Rl 22OVll- 250 05LV03 ... -E ... 1003R1 22OVll- 550 05 LV04 ... -E ... 1004 Rl 380 VI 3- 1200 05 LV05 ... -E ... looS Rl 38OVl3- 1200 05 LV06 ... ·E ... l006Rl 220Vll- 500 05 LV07 ... ·E ... 1007 Rl 220Vll- 400 05 LV08 ... ·E ... 1008Rl 220Vll- 550 05 LV09 ... ·E ... 1009 Rl 220V/l- 550 G) at rated d c. current
~ can be edJusted from 10'11. to 100"10 of rated d. c. currenl
Table 2
... tlng and Control Section
Operating ranges
Tolerance of connection voltage Frequency
+
10%-10% 50 ... 60 Hz OOC ... 400C d. c. voltage ratingeD V-±13 ±24 ±52 +80/-12 ±45 ±40 ±24 70/-12 ±24Guaranted temperature range
Operational temperature range from -100C ... +600C
at.at.lhad current auppIy
positive output voltage
negative output voltage Tamperature drift
max.
additional loading ... tor +14.1V ... +1S.9V (speCimen scatter) -14.1 V ... -15.9V<
2.25mVloC ±20mABrown Bovery company
Output
d. c. current Dynam-peak Outpul Power Dynam-rating® current@ impedance rating~} peak
power A- I.- e w w ± 8 ±16 0,30 100 210 ± 6 ±15 0,20 145 360 ± 6 ±15 0,30 360 780 ±10 +30/-20 0,30 800 2400 ±20 ±40 0,30 900 1800 ±10 ±20 0,30 400 800 ±10 ±20 0,20 240 480 ± 6 + 15/-10 0,30 420 1050 ±15 ±30 0,25 360 720
ell max. 2 sec. decay time constant can be set
(!!) reduction 2%/oe from 400e ambient temperature increase
Tacho control Adjustment range Control range Control error IR Compenaatlon Actuating range Control range Control error Current limiting 0 ... 100% 1: 5000 ±0.5% 10 ... 100% 1:100 ±5%
Current limiting (continuous current) 10 ... 100% Dynamic peak current see table 1 Max. duration of dynamic peak
current see Fig. 11
Note weighl kg 6,0 8,5 18,0 37,0 37,0 18,0 10,0 18,0 18,0
B B C
BROWN BOVERI NEDERLAND B.V.Elektroweg 22 Poslbus 301
BROWN BOVERI Rotterdam TeL:Ol0-180 260
E
tPAR V E X
27·29, rue lucien Juy. 21007 - DlJON·CEDEX • FRANCEDEpt VITESSE VARIABLE Phone (IO) 41.81.18 - Telex 350653
Me 19 P Me 19 P
SYMBOLS Unc.oled Coole .. (2)
1 - MOTOR RATINGS (I) II. Rated Torque.
10 liter/sec
12. Rated Speed ..
13. Rated Output .
lAt. Rated Voltage.
15. Rated Current.
16. Maximum at Very Low Speed *
17. Pulse Torque (intermittent operation) (3)
la.
Moxlmum Speed with no external load ** • 6tall motor : esk us for ma •. eurrenl• • For other duly cycles : ask us
2 - MOTOR CONSTANTS 21. t.M.F./IOOO RPM ..
22. Torque Constant/Ampere .•
23. Regulation Constant Voltagelcm.N
2At. friction Torque
25. Damping ConstontllOOO RPM ..
26. Terminal resislance (4) .•
27. Armotur. Inductance ..
• 21. Total Inertia ••
29. Mechanical Time Constant
2.
to
Power late (II)* "
parallel loyers Cn Nn Pn Un In Icc Cimp Nmax Ke Kr KN Tf KO R L J.,.
p,3 - PERFORMANCE CHARACTERISTICS. CONTINUOUS OPERATION
I
,
18 4., c 3. 1. N 8 iii _ .... '~#v' I I ~'I\ I' ,I
''1-
!
... iI
.-
.. ..:.... j ..
'1 J .. dI> i i ! -""~...-:Hi
i·' . ! II
I
,~~~_,~·~·~:]_;-~.~t·~_'·~·-_·~~'
~·~·~i_-·~··_-~'-+.~
___
·i_·~-~-~T_"
..
o 200 4'00 600 100 Torque (em. N)Dele grven .re on pr.ctrce inClepenCIenl 01 Ihe .'mat .... e temperalure
cm.N 320 r.p,m. 3000 W 1000 V 83 A
.4 ...
A 16,5 cm.N 2440 r.p.m . 5000 V 25,5 cm.N 24,4 f.p.m. 0,3 cm.N 10 cm.N 8 12 0,46 ~H <100 g.cm2 12000 ms 9,2 KW/s 500" Mounting on metal plate with thermal insulalion 1400.400.10 mm). and pure DC supply. Ambient temperalure O' C 10 40- C
21 Lou In motor: 13 mm 01 H20 510 3000 1600 87 22,2 22,3 2440 SOOO 25,5 24,4 0,3 10 8 0,46 <100 12000 9,2 500
VII.C. Encoder
Winkelme~gerat
Inkrementales offenes WinkelmeBgerat mit AURODUR-StahlbandTeilungsperiode 100 !-1m
MeBschritt bis O.OOOO~
Bendeuflagedurchmesaer ;;: 600 mm
HEIDENHAIN
,
Strichzahlen [z] Ihcham.che Kennwerte Bandauf/agedurchmesser MaBband-TeilstOcke Wlrmeausdehnungskoeffizient des AURODUR-Stahlbandes Referenzmarken Mal3band-Genauigkeitsklasse GrMter Unterteilungsfehler zuUissige Beschleunigung zuUissige StoBbelastung Schutzart Korrosionsschutz Arbeitstemperatur -Bereich lagertemperatur-Bereich relative Feuchte Gewicht Lichtquelle Impulsformer-Elektronik Ausgangssignale Raferenz-Signal SignalgroBeHOchste zullssige Drehzahl "max
abNingig
yom
Bandauflagedurchmesser (D} •. da Teilungsperiode des AU~ODLJR· Stahl-MaBbandes immer 100 ~mz .. Int [(Dmm + 0.3 mm) • n 0.1mm
Int: Ganzzah/-Anteil des in Klammern stehenden Ausdrucks
:5 3000 mm
Stah/-MaBband-TeilstOcke konnen aber Spannschlosser miteinander verbunden werden.
10.5 • 10-41 K-'
In der Mitte (Toleranzbereich ± 10 mm) des Stahl-MaBbandteilstOckes und davon ausgehend im 50-mm-Raster.
Damit die erste und letzte Referenzmarke nicht naher als 5 mm am Stahl-MaB-band-StoB liegt wird der Toleranzbereich von ± 10 mm ausgenutzt.
± 2 jJITl (± 1 jJITl nur mit EXE 700 nach Abgleich)
± 5 jJITl an den StoBstellen
Abtasteinheit staub- und spritzwassergeschOtzt nach IP54 (DIN 40050) Abtasteinheit: eloxiertes Aluminium
Stahl-MaBband. Spannschlosser und Endspannstucke: rostfreier Stahl Abtasteinheit: rP C bis 450 C
Stahl-MaBband und MaBbandtrager: rP C bis 5rP C
Nur wenn der MaBbandtrager aus einem Material besteht dessen Warmeaus-dehnungskoeffizient zwischen 9 . 10-6 und 12 . 10-6 K-1 (z. B. GuBeisen oder
ferritischer Stahl) betragt. Bei hoheren Warmeausdehnungskoeffizienten (z. B.
Aluminium) gilt ein eingeschrankter Temperaturbereich von 1rP C bis 3rP C. Abtasteinheit: - 3rP C bis 7rP C
Stahl-MaBband und Mal3bandtrager: rP C bis 5rP C max. 80%
Abtasteinheit 350 9 SpannschloB 300 9 EndspannstOck 300 9 Stahl-MaBband 60 g/m
LED mit Vorwiderstand: 5 V ± 10%.
<
120 mA a) in zahler eingebautb) extern. siehe EXE-Druckschrift
• 2 annahernd sinusformige Signale 1.1 und 1.2
V'
1\ 1\ 1\•
1 Signal lea pro Umdrehung1.2 ca. 11 JJAss bei Last 1 kOhm
1.1 ca. 11 ~s
I
lea ca. 5,0
J:IA.
• Nutzantell
[ . -" fmax [kHz] 1~ 60
n max min J - Z .• u-·
z: Anzahl dar Teilungsperioden von 100 jJITl auf dem Umfang (Strichzahl)