TMS320C26
|
|
|
Top Searches for this datasheet
ULN2803 - ULN2803
tms320c2x - tms320c2x
TMS320C26 starter - TMS320C26 starter
TMS320C26 schematic - TMS320C26 schematic
TMS320C26* - TMS320C26*
tms27c256 - tms27c256
sensor to control blood variation - sensor to control blood variation
ram generator control circuit 220v - ram generator control circuit 220v
automatic heat detector report - automatic heat detector report
TMS320C26 - TMS320C26
Implementing TMS320C26 Counter
Authors: Normand, Cordesses, Tournayre, J.L. Damez
ESIEE, Paris September 1996 SPRA325
IMPORTANT NOTICE
Texas Instruments (TI) reserves right make changes products discontinue semiconductor product service without notice, advises customers obtain latest version relevant information verify, before placing orders, that information being relied current. warrants performance semiconductor products related software specifications applicable time sale accordance with TI's standard warranty. Testing other quality control techniques utilized extent deems necessary support this warranty. Specific testing parameters each device necessarily performed, except those mandated government requirements. Certain application using semiconductor products involve potential risks death, personal injury, severe property environmental damage ("Critical Applications"). SEMICONDUCTOR PRODUCTS DESIGNED, INTENDED, AUTHORIZED, WARRANTED SUITABLE LIFE-SUPPORT APPLICATIONS, DEVICES SYSTEMS OTHER CRITICAL APPLICATIONS. Inclusion products such applications understood fully risk customer. products such applications requires written approval appropriate officer. Questions concerning potential risk applications should directed through local sales office. order minimize risks associated with customer's applications, adequate design operating safeguards should provided customer minimize inherent procedural hazards. assumes liability applications assistance, customer product design, software performance, infringement patents services described herein. does warrant represent that license, either express implied, granted under patent right, copyright, mask work right, other intellectual property right covering relating combination, machine, process which such semiconductor products services might used.
Copyright 1997, Texas Instruments Incorporated
TRADEMARKS
trademark Texas Instruments Incorporated. Other brands names property their respective owners.
CONTACT INFORMATION
TMS320 HOTLINE TMS320 TMS320 TMS320 email
(281) 274-2320 (281) 274-2324 (281) 274-2323 dsph@ti.com
Contents
Abstract Product Support World Wide Introduction. Beef Cutting Project. Half Carcass Beef Splitting Prototype. Cutting Cage Procedure Used From Analog Digital Processing Stand Alone Card Specification Sheet. Board Design. Modification Existing Software. Results. Graphs Comments Limits Defaults Software Approach Overview. Amplitude Demodulator Detector. Model Detector Main Drawbacks Results Discussion Conclusion References
Figures
Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Fore Hindquarters. Current Signal Signals During Digital Processing Counter Environment Board Architecture. Typical Graph. Block Diagram Detector Demodulator Detector. Output Demodulators. Input Outputs Software.
Implementing TMS320C26 Counter
Abstract
This application report describes development automated system control cutting ribs during boning process bovine carcasses using Texas Instruments (TITM) TMS320C26 digital signal processor (DSP). Operating costs meat processing plants encourage automated equipment such robot saws. precisely forequarter hindquarter side beef, robot must stop movement when cuts desired number ribs. This document part first European Education Research Conference that took place September 1996 Paris. information encourages students from around world find innovative ways DSPs, TI's World Wide site www.ti.com.
Implementing TMS320C26 Counter
SPRA325
Product Support World Wide
World Wide site www.ti.com contains most date product information, revisions, additions. Users registering with TI&ME build custom information pages receive product updates automatically email.
Implementing TMS320C26 Counter
SPRA325
Introduction
Clermont-Ferrand/Theix Centre centres National Institute Agricultural Research (INRA). currently employs people, including graduates (Scientists, Researchers, Engineers, Lecturers). weight originality studies conducted Centre based wide range approaches, from molecular cellular biology systems, organism biology. Likewise, research production meat milk, quality human foodstuffs seeds plants, taken increasing amount needs food farm industry: development quality materials, production cost control, processing techniques, product usage health value foodstuffs man. Meat quality measured, thanks development methods objective product characterisation: instrumental measurements, sensory analysis, modalities consumer's sensory perception texture, consumer behaviour analysis. control transformation processes, studies conducted slaughtering physiology cell death, solve ethical problem facing industry, accordance with economic constraints process engineering: robotics smart tools, heat mass transfer (process product modelling), which have principal role team.
Implementing TMS320C26 Counter
SPRA325
Beef Cutting Project
Half Carcass Beef Splitting Prototype
Within European Eureka project 1032:"ABCD", INRA, Durand International, Interbev develop industrial prototypes: will separate automatically beef side into forequarter hindquarter, second will make three pieces from forequarter (shoulder, neck, breast). This project results from previous projects INRA1, with cooperation ADIV Cemagref, aimed first proving feasibility robotized cutting beef forequarter. first prototype, cutting path (Figure achieved special tool designed around rotative handled robot. used ribs other part special tool used seam meat.
Figure Fore Hindquarters
really complex perform this action, incidence biological variability tremendous. have made some trials using home-made morphometric models linked anatomic parameters only solve this problem seems feedback control from parameters coming from tool itself.
Implementing TMS320C26 Counter
SPRA325
Cutting Cage
During forequarter hindquarter splitting, necessary stop movement cutting when desired number ribs reached. achieve this, tried video image analysis ribs cannot detected each carcass because presence blood, fat, etc. Therefore judicious signal directly linked saw, like sound frequency variations, vibrations amplitude variations feeding current saw. sound frequency variations seem signal from which easily extract position very dependent surrounding installations, especially noisy plant like abattoir. vibrations also used necessary adapt specific sensors equipment. current very easy obtain, even industrial equipment, have chosen this signal. Moreover, dependent surrounding equipment contrary sound signal. This signal directly proportional torque asynchronous motor. increasing amplitude represents cutting while meat inter space represented lower amplitude. this signal (Figure very noisy must significantly processed.
Figure Current Signal
Implementing TMS320C26 Counter
SPRA325
Procedure Used
From Analog Digital Processing
first approach check feasibility detector consisted designing analog prototype. This gave promising results quickly reached limits, necessary perform some manual adjustments. decided investigate possibilities digital signal processing. Texas Instruments gave opportunity evaluate coast solution.2 obtained affordable TMS320C26 Starter (DSK) developed first software based primarily analog solution previously tested.3 Three block functions were realised:
first A.M. demodulation. second threshold which extracted parts signal containing ribs. third detector using sliding windows comparators.
Figure illustrates results after each step process. have been limited amount available memory lack digital Input/Output lines. add-on board including extra memories interface circuits developed. Moreover this card aimed running without external computer.4 whole system (hardware software) gave such interesting results that chose design stand alone single board.
Implementing TMS320C26 Counter
SPRA325
Figure Signals During Digital Processing
Implementing TMS320C26 Counter
SPRA325
Stand Alone Card
Specification Sheet
counter must seen Input/Output black shown Figure previous software huge into internal memory TMS320C26, added some fast external static RAM. amount fixed knowledge sampling frequencies spectrum incoming signal. typical signal lasted about seconds sampling frequency 1000 Hertz, that gave memory size chose EPROM, TMS27C256 DSP, program memory because price availability. However long access time regard memory access timing designed wait state generator. classical glue logic been preferred high integration circuit (like PAL) because easier maintain repair industrial context. digital Output lines latched buffered using Darlington array (ULN2803). these lines used send stopping signal hardware that drove movement saw. digital Input lines allows enter number ribs adjust sensibility algorithm. analog uses same analog interface circuit (AIC) (TLC32040) compatibility.5
Implementing TMS320C26 Counter
SPRA325
Figure Counter Environment
Asynchronous motor driving Algorithm parameters Synchronisation
counter
feeding current 220V Stopping signal
Board Design
build single size board with classical tools small laboratory. bought software schematic design printed circuit board, including auto-route. This single size board, with layers, main advantages:
design allows evolution final system. Indeed, EPROM allows change easily software board. another detection method chosen (for example sound detection), hardware will remain functional because flexibility. memory capacity obtained with allows more powerful algorithms.
Moreover, system easy connect cutting device. power supply current sensor stand separate card avoid problems.
Implementing TMS320C26 Counter
SPRA325
Figure Board Architecture
Algorithm parameters Wait state generator Analog output Analog input Synchro interrupt Clock 40MHz Output port
EPROM
Input port
Synchronisation signal
Modification Existing Software
Stoping signal
order extended possibilities hardware, write some other functions: auto-boot, memory greedy procedures. auto-boot self-initialise TMS320C26 load EPROM content into fast internal RAM. program then DSK. external fast SRAM used data memory internal memory free program code. Furthermore, standalone system, program been modified send pulse unused output easy control scope.
Implementing TMS320C26 Counter
SPRA325
Results
Graphs
tests were conducted forequarter with only five ribs cut. Moreover, could make great number tests because expensive cost bovine carcasses. During these tests, four ribs were each time detected. fifth always detected. example typical graph obtained shown Figure
Figure Typical Graph
Implementing TMS320C26 Counter
SPRA325
Comments
study these graphs different tests highlight four essentials points:
last four pulses correspond four ribs. detection with this software quite good. Sometimes ribs detected algorithm detects some which does exist (for instance third pulse). This problem lower higher sensitivity software. first pulse corresponds starting engine also considered rib. second pulse corresponds beginning cutting meat surrounding rib, raising torque detected software rib.
consequence, detection must only begin when meat carcass, which could avoid appearance first pulses. This synchronisation signal must planned formal automatic cutting device. hardware counter already designed receive this signal. first observations show limits defaults software, even detection most time, quite good.
Limits Defaults Software
detection systematic carcasses. Sometimes need modify sensitivity algorithm. This essential default software. Indeed, fact that user settles sensitivity, does allow automatic industrial utilisation detection algorithm. Moreover, tests were carried industrial environment. input signal quite clear. However, industrial environment, signal could more noisy could perturb detection because demodulation. fact, demodulation gives accurate reproduction cover signal without decreasing noise. perturbations present each part processing. industrial utilisation counter software ought improved.
Implementing TMS320C26 Counter
SPRA325
Approach
Overview
first software based upon analog system previously designed. order achieve better results with same hardware, decided study signal produced electric saw. Then optimized each part software (Figure 7)67.
Figure Block Diagram Detector
Amplitude Demodulator
decided (Finite Impulse Response) filters instead mainly because they stable8, even fixedpoint device. block diagram Figure illustrates demodulator.
Figure Demodulator Detector
taps (f0=50 filters incoming electric amplitude-modulated signal9. Then software detects this signal, taking absolute value rejects carrier using taps notch filter. chose taps lowpass filter thoroughly with trial error process order kill glitches noise, allow waveform through demodulator. Kaiser window provides excellent results fc=2.1 comparison between demodulators shown Figure
Implementing TMS320C26 Counter
SPRA325
Figure Output Demodulators
NOTE: vertical axis proportional output voltage demodulator.
Detector
Model Detector
kind model developed under Matlab then translated into TMS320C26 assembly language. directions have investigated are:
shape waveform-based detector
Implementing TMS320C26 Counter
SPRA325
memory-based method storing parameters previous probability function prediction
first model rejects "wrong" shapes; second model saves into shape last rib. last function indicates likely come away not. found that single method does produce accurate results. This reason formal algorithm takes into account three methods.
Main Drawbacks
were limited possibilities assembler: assembly source code long (more than Kb). Moreover, faced problems during this project:
central arithmetic logic unit (CALU) TMS320C26 fixed-point device, were lacking sufficient dynamic calculations thus used filters. TMS320C26 does have hardware division, modify model cope with multiplication shift.
Results Discussion
time writing, software provided very good results very accurate time localisation ribs. also detected ribs software unable find (Figure 10).
Implementing TMS320C26 Counter
SPRA325
Figure Input Outputs Software
upper window typical electric current. lower window shows processed signal input detector (smooth curve) pulses generated detector when found. noticed that last (t=5400 corresponding last pulse detected software. mentioned that simulation (Figure implementations produce same results. Moreover, have some more ideas improve this system. algorithm will store samples intensity signal. Thus, will able process data twice:
Once real time, described above second time after fourth fifth rib. This will allow validate results detection algorithm, making complete shape instead beginning shape.
Another research area wavelet processing incoming data10. Some experiments conducted under Matlab proved that wavelets more powerful signal than classical digital filtering (this time-frequency description signal). Unfortunately, lacked time power processing implement such method TMS320C26.
Implementing TMS320C26 Counter
SPRA325
Conclusion
automatic system control cutting ribs during boning process bovine carcasses discussed this paper developed five months engineer training period. This system based real time analysis feeding current rotative electric saw. choice carry this operation focused TMS320C26. versatile single size board been designed able support strongest algorithm TMS320C26 handle with assembler. Discussion done filters instead filters openings with wavelets processing. Such improvements will carried using stronger assembler maybe floatingpoint device such TMS320C31.
Implementing TMS320C26 Counter
SPRA325
References
DAMEZ J.L, SALE (1994), Recherches l'automatisation quartiers avants bovins. Viandes Prod. VOL.15 103-107 1994. Texas Instruments, (1993), TMS320C2x User's Guide, Digital Signal Processing Products. THIRY J.L., (1995) Analyse mise forme signaux analogiques l'aide DSP, Rapport stage Electrique CUST. NORMAND (1996) Etude, mise point d'une carte traitement signaux l'utilisation d'un TMS320C26, l'obtention titre Electrique. Texas Instruments, (1995), Data Acquisition Circuits, Data Conversion Analog Interface, Mixed Signal Products. MATLAB®, version 4.2c.1, Math Works, Inc. CORDESSES (1996), Etude, conception d'un code calcul pour permettant filtrage l'analyse signaux fortements Rapport stage Electrique CUST MARVEN EWERS (1994), simple approach digital signal processing, Alden Press Limited, Oxford Northampton, Great Britain. SELESNICK I.W., LANG BURRUS C.S., (1995), Constrained Least Square Design Filters Without Specified Transition Bands, Department Electrical Computer Engineering 366, Rice University, Houston, Texas, USA. MISITI MISITI OPPENHEIN POGGI J.M., (1996), Wavelet Toolbox. Math Works, Inc.
Implementing TMS320C26 Counter
Other recent searches
TMCM-110 - TMCM-110 TMCM-110 Datasheet
SGA-4186 - SGA-4186 SGA-4186 Datasheet
SA9616 - SA9616 SA9616 Datasheet
S12OSCLCPV1 - S12OSCLCPV1 S12OSCLCPV1 Datasheet
PD44165082 - PD44165082 PD44165082 Datasheet
LTC4219 - LTC4219 LTC4219 Datasheet
LTC4219-5 - LTC4219-5 LTC4219-5 Datasheet
LTC4219-12 - LTC4219-12 LTC4219-12 Datasheet
KD2003-DC10A - KD2003-DC10A KD2003-DC10A Datasheet
BA782 - BA782 BA782 Datasheet
BA783 - BA783 BA783 Datasheet
Privacy Policy | Disclaimer
© 2013 Datasheets.org.uk