IEEE-P1451.2 Smart Transducer Interface Module
Stan P. Woods
Janusz Bryzek, Ph.D.
Steven Chen
Jeff Cranmer
Edwin Vivian El-Kareh
Mike Geipel
Fernando Gen-Kuong
John Houldsworth
Norm LeComte
Kang Lee
Michael F. Mattes
David E. Rasmussen
Hewlett-Packard Company
Intelligent MicroSensor Technology
Aeptec Microsystems
Lucas Control Systems Products
AB Networks
Eurotherm Controls
Endevco
Eurotherm Controls
Texas Instruments
National Institute of Standards and Technology
SSI-Controls Technology
Hewlett-Packard Company
Abstract
This paper provides a technical overview of the smart transducer interface module (STIM), the key
element of the proposed IEEE-P1451.2 Draft Standard for Transducer to Microprocessor Communication
Protocols and Transducer Electronic Data Sheets (TEDS) Formats. The draft standard is released for
balloting as of August, 1996. Objectives and genealogy of this standard are summarized. Key technical
innovations such as the TEDS, representation of physical units, general calibration model, triggering of
sensors and actuators, variable transfer rate between a host and the STIM, and support for multivariable
transducers are briefly discussed. Detailed descriptions of the STIM, TEDS, digital interface, and plug-
and-play operation are also provided. The specifics of physical units encoding, an example of a TEDS,
and an example of timing requirements for taking a sensor reading are also included to aid the overview.
1. Objectives of IEEE-P1451
The Draft Standard for A Smart Transducer Interface for
Sensors and Actuators, IEEE-P1451, aims at simplifying
transducer connectivity to existing networks.
IEEE-
P1451 consists of two parts:
•
P1451.1, developing a network independent common
object model for smart transducers.
•
P1451.2, enabling connection of transducers to net-
work microprocessors.
These are goals the IEEE-P1451.2 Smart Transducer
Interface
1
is helping to meet.
This paper describes the progress made to date by the
IEEE-P1451.2 Transducer to Microprocessor Working
Group to facilitate the ease of connecting transducers to
microprocessors.
At the core of this effort is a
transducer electronic data sheet (TEDS), which is a data
structure stored in a small amount of nonvolatile
memory, physically associated with the transducer. The
TEDS is used to store parameters which describe the
transducer to the network capable application processor
(NCAP), making self-identification of the transducer to a
system possible.
The working group has defined the contents of the TEDS
and a digital hardware interface to access the TEDS, read
sensors, and set actuators. The resulting hardware
partition encapsulates the measurement aspects
2
in a
smart transducer interface module (STIM) on one side of
2. Introduction
Imagine that you can:
•
Select the transducer best suited to solve the
measurement or control problem independently of
the selected control network.
•
Use the same transducers on multiple control
networks.
•
Select the control network best suited for the
application without transducer compatibility
constraints.
•
Achieve automatic self-configuration when a trans-
ducer is connected to a network microprocessor.
1
These goals are being met in combination with IEEE-
1451.1 which is not covered in this paper.
2
Covered by P1451.2.
the digital interface, and the application related aspects
3
on the NCAP.
This paper describes the hardware block diagram of the
STIM, including the TEDS and the digital interface.
4. Key technical features
Figure 1A. depicts a STIM and the associated digital
interface as described in the P1451.2 draft. The STIM is
shown here under the control of a network-connected
microprocessor. In addition to their use in control
networks, STIMs can be used with microprocessors in a
variety of applications such as portable instruments and
data acquisition cards as shown in Figure 1B.
The STIM embodies specific unique features of this
proposed standard, which are briefly described below.
3. Background
Control networks provide many benefits for transducers,
4
such as:
•
Significant reduction of
installation costs by
eliminating long and large numbers of analog wires.
•
Acceleration of control loop design cycles, reduction
of commissioning time, and reduction of downtime.
•
Dynamic configuration of measurement and control
loops via software.
•
Addition of intelligence by leveraging the micro-
processors used for digital communication.
One major problem for analog transducer manufacturers
is the large number of networks on the market today. It
is currently too costly for many transducer manufacturers
to make unique smart transducers tailored for each
network on the market.
In September 1993, the proposal of developing a smart
sensor communication interface standard was accepted by
IEEE-TC9.
5
In March, 1994, the National Institute of
Standards and Technology (NIST) and the Institute of
Electrical and Electronics Engineers (IEEE), hosted a
first workshop to discuss smart sensor interfaces and the
possibility of developing a standard interface that would
simplify connectivity of smart transducers to networks.
Since then, a series of four more workshops have been
held and two technical working groups formed in
February, 1995:
•
The P1451.1 working group concentrating on a
common object model for smart transducers along
with interface specifications to the model [1] [2] [3].
•
The P1451.2 working group concentrating on
defining the TEDS, the STIM, and the digital
interface including connector pin allocation and a
com-munication protocol between the STIM and the
NCAP [1] [4].
4.1 Single general purpose TEDS
The TEDS as presently defined supports a wide variety of
transducers with a single general purpose TEDS
structure.
6
This approach makes the rest of the system
easier to implement and the implementation scaleable. If
specific fields are not required for a given transducer,
these fields have zero width, saving the required
memory.
4.2 Representation of physical units
The P1451.2 draft adopts a general method for describing
physical units sensed or actuated by a transducer. The
method, described in the table in Appendix A, employs a
binary sequence of ten bytes to encode physical units. A
unit is represented as a product of the seven SI
7
base
units and the two SI supplementary units, each raised to a
rational power. This structure encodes only the
exponents; the product is implicit. Appendix A contains
examples for distance, pressure, acceleration, and strain.
The U/U forms (enumerations one and three in Appendix
A) are for expressing “dimensionless” units such as
strain (meters per meter) and concentration (moles per
mole). The numerator and denominator units are
identical, each being specified by subfields two through
ten [5].
6
3
4
Covered by P1451.1.
“Transducer” is defined by IEEE-P1451 working
groups as either a sensor or an actuator.
5
Technical Committee 9 (TC-9) on Sensor Technology
of the Instrumentation and Measurement Society
As opposed to creating a unique TEDS structure for
each kind of transducer: for example, one TEDS
structure for temperature sensors and another TEDS
structure for servo actuators, each structure with unique
required fields.
7
Le Système International d’Unités
Network
DCLK
DOUT
DIN
NIOE
Interface
logic
TEDS
Signal
conditioning and
conversion
Transducer
#1
Networked sensors
Network
Capable
Application
Processor
(NCAP)
or
Host
Microprocessor
NTRIG
NTRACK
NIO_INT
.
.
.
Transducer
#255
Data acquisition cards
+5 V
Common
STIM
Portable instruments
(Networking / P1451.2 digital
application)
interface
A)
(Encapsulated
measurement)
B)
= P1451.2 transducer
Figure 1: Hardware partition proposed by P1451.2 and possible uses for the interface
4.3 General calibration model
The P1451.2 draft provides a general model to optionally
specify the transducer calibration. It is very flexible yet
can collapse to an acceptable size for a simple linear
relationship. The scheme supports a multi-variable, piece-
wise polynomial with variable segment widths and variable
segment offsets.
4.5 Variable transfer rate between host
and STIM
The hardware data clock line is driven by the NCAP.
There is a field in the TEDS which specifies the maximum
data transport rate that the STIM can support. This
provides a flexible mechanism to match NCAPs and
STIMs.
4.4 Triggering of sensors and actuators
The proposed digital interface has hardware trigger lines to
allow the NCAP to initiate sensor measurements and
actuator actions, and to allow the STIM to report the
completion of the requested actions. The NCAP can trigger
an individual channel, or all transducer channels at once.
In the latter case, there are TEDS fields provided to specify
timing offsets between the STIM’s channels and to
determine when each measurement or actuation has
occurred relative to the single trigger acknowledgment.
The draft proposes that the slowest channel be the
reference channel and that all the offsets be specified
relative to this channel.
4.6 Support for multi-variable
transducers
P1451.2 includes support for multi-variable transducers in
a single STIM. A STIM may have up to 255 inputs or
outputs allowing the creation of multi-variable sensors,
actuators, or combinations thereof. Several multi-variable
STIM examples are shown in Figure 3.
5. STIM
Figure 2 shows the block diagram for a STIM, along with
the interfaces between each module. A TEDS is
incorporated into a STIM. In addition to the TEDS, the
STIM contains logic to implement the P1451.2 interface,
the transducer, and any necessary signal conditioning and
signal conversion.
8
Only interface “A” is defined by P1451.2. Interfaces “B”,
“C” “D”, and “E” allow transducer manufacturers to
continue to obtain competitive differentiation in areas of
performance, quality, feature set, and cost by choosing how
these interfaces are implemented. At the same time
P1451.2 offers the opportunity to design transducers to a
common interface between a STIM and NCAPs enabling
the use of the same transducer across many networks and
applications [6].
The first example in Figure 3A demonstrates a single
channel analog sensor implementation.
A)
Temperature sensor STIM
P1451.2
9
Micro-controller
ADC
TEDS
Temperature sensor
STIM
B)
Eight channel digital I/O STIM
P1451.2
9
Micro-controller
TEDS
In
Out
STIM
P1451.2
9
P1451.2
logic
block
TEDS
Signal
conversion
Signal
Transducer
conditioning (Sensor or
Actuator)
C)
Four channel sensor STIM
P1451.2
9
Micro-controller
TEDS
ADC
Pressure sensor
Flow sensor
Temperature sensor
C
A
B
D
STIM
E
STIM
Figure 2: STIM block diagram
D)
Sensor and actuator STIM
The P1451.2 logic block shown in Figure 2 may be
implemented in several ways. The working group has now
implemented STIMs using a field programmable gate array
(FPGA) and a low-cost microcontroller to serve as the
logic block. These methods demonstrate that P1451.2
STIMs can be built today using off-the-shelf parts. The
microcontroller option provides the additional advantage of
potentially combining all the logic, TEDS, and signal
conversion into one integrated circuit, where the P1451.2
logic block is implemented using microcontroller
firmware.
Figure 3 shows four examples of STIM configurations
using a low-cost microcontroller.
These examples
demonstrate the flexibility in STIM design provided by
P1451.2.
8
pH sensor
P1451.2
9
Micro-controller
TEDS
ADC
Temperature sensor
Pressure sensor
DAC
Digital
Proportional valve
Relay
STIM
Figure 3: STIM examples
The second example in Figure 3B demonstrates the
implementation of a digital input/output (I/O) module with
four digital inputs and four digital outputs. The TEDS
model in P1451.2 allows this STIM to be described as an
eight-channel STIM or alternatively it could be described
as a two-channel STIM with one input channel and one
output channel each with a length of four. This flexibility
Signal conditioning and signal conversion are not
covered by P1451.2.
in the model allows digital I/O modules with thousands of
inputs/outputs to be implemented if such a product were
needed.
The third example in Figure 3C shows a STIM with
multiple analog sensors. These four sensors could be
measuring a process liquid.
Figure 3D illustrates that combinations of sensors and
actuators can be combined into one STIM to support all the
transducers used in control system solutions. The code
implementing the control loop could reside either in the
NCAP or the microcontroller used to implement the
P1451.2 interface.
Meta-TEDS
One per STIM:
Contains the overall description
of the TEDS data structure,
worst case STIM timing
parameters, and channel
grouping information.
One per STIM channel:
Contains upper/lower range
limits, physical units, warm up
time, presence of self-test,
uncertainty, data model,
calibration model, and
triggering parameters.
One per STIM channel:
Contains the last calibration
date, calibration interval and all
the calibration parameters
supporting the multi-segment
model.
Multiple per STIM:
For application specific use.
Multiple per STIM:
Used to implement future and
industry extensions to P1451.2.
Channel TEDS
6. TEDS
The TEDS is one of the main technical innovations
introduced in P1451.2. A TEDS, which carries information
about the transducer and its performance, is not a new
concept. Companies have been embedding data structures
in memory associated with their products for many years
[7] [8] [9] [10]. What is new is the general model of a
transducer behind the P1451.2 TEDS which supports a
wide variety of sensors and actuators.
The TEDS contains fields that fully describe the type,
operation, and attributes of the transducer.
If the
transducer is moved to a new location, it is moved with the
TEDS. This way the information necessary for using the
transducer in a system is always present.
Figure 4 shows the main addressable sections of the TEDS
along with examples of the content for each segment. The
sections shown with dotted lines (calibration-TEDS,
application specific-TEDS and extension-TEDS) are
optional.
The calibration specification in the TEDS permits the
sensor manufacturer to describe a multi-dimensional
calibration for each channel. To eliminate high order
polynomials it is possible to specify a segmented
calibration where each segment can have a variable width
and offset. It is expected that a general correction engine
will be present in the NCAP that understands this
calibration scheme so that it can be run “blindly” no matter
which transducer is attached. An example of a multi-
segment calibration curve with simple linear segments is
shown in Figure 5.
Calibration TEDS
Application
specific TEDS
Extension TEDS
Figure 4: Overview of the TEDS structure
Appendix B contains an example of the complete P1451.2
TEDS for a single channel pressure sensor. It is a ceramic
pressure sensor with an analog output between 0 to 5 V dc
corresponding to 0 to 20,684,190 Pa (3000 lb/in
2
) pressure
input. The sensor has a 10 ms response time, no
appreciable warm-up requirement and the maximum non-
linearity is measured to be 0.56% of V
supply
. The primary
components of the single channel STIM are a serial 12-bit
analog-to-digital converter (ADC) for data conversion (75
µs
conversion cycle), and an 8-bit PIC-type processor with
4K by 12-bit on-chip EEPROM (8 MHz operation).
Calibration is fixed and is specified using five equal
segments with non-zero offsets for each segment. This
allows first order calibration functions to be used to reduce
the non-linearity in the analog output (0.56% reduced to
0.03%).
京公网安备 11010802033920号
Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved
评论