UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An on-line computer assisted mass spectrometer 1971

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1971_A6_7 M58.pdf
UBC_1971_A6_7 M58.pdf [ 3.43MB ]
UBC_1971_A6_7 M58.pdf
Metadata
JSON: 1.0053430.json
JSON-LD: 1.0053430+ld.json
RDF/XML (Pretty): 1.0053430.xml
RDF/JSON: 1.0053430+rdf.json
Turtle: 1.0053430+rdf-turtle.txt
N-Triples: 1.0053430+rdf-ntriples.txt
Citation
1.0053430.ris

Full Text

AN ON-LINE COMPUTER ASSISTED MASS SPECTROMETER by DAVID LAURIE MITCHELL B.Sc, King's College London, 1968 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of GEOPHYSICS We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Ap r i l , 1971 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver 8, Canada 0 ABSTRACT An analog data a c q u i s i t i o n system incorporating an Interdata Model 4 d i g i t a l computer has been designed and b u i l t for a mass spectrometer. This system has been conceived with the primary objectives of improving a n a l y t i c a l p r e c i s i o n and production. Automated mass spectrometer operation allows for the c o l l e c t i o n of larger quantities of data while decreasing operator involvement and consequently diminishing operator bias and fatigue. The analog signal from the mass spectrometer measur- ing system i s d i g i t i z e d using a d i g i t a l voltmeter and trans- mitted, v i a an in t e r f a c e , to the processor where the d i g i t a l information i s manipulated i n accordance with a computer pro- gram. An additional f a c i l i t y i s provided whereby d i g i t a l data from the processor can be displayed, i f desired, on a 5 decade numerical readout situated at the mass spectrometer console. Hardware i s also available i n the interface to provide control of the magnetic f i e l d scan rate. A function control switch at the mass spectrometer console allows the operator to convey a variety of predetermined instructions to the computer at any time during the course of a run. Thus, the system provides for on-line (real time) data processing of mass spectra as well as limi t e d computer control over the mass spectrometer. This thesis i s primarily concerned with the design and construction of the l o g i c hardware for t h i s system together with a demonstration of i t s operating a b i l i t y . i i i TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS i i i LIST OF FIGURES v LIST OF TABLES v i ACKNOWLEDGEMENTS v i i CHAPTER 1. INTRODUCTION 1.1 General Background 1.2 Design O b j e c t i v e s CHAPTER 2 GENERAL DESCRIPTION 2.1 The Mass Spectrometer 2.2 The Computer 2.3 Programming the I n t e r d a t a Model 4 CHAPTER 3 THE INTERFACE HARDWARE DESIGN 3.1 I n t r o d u c t i o n 3.2 Device A d d r e s s i n g 3.3 Data and Status Input 3.4 Data and Command Output 3.5 I n t e r r u p t C o n t r o l 3.6 Read/Write Sequencing 3.6.1 Read Op e r a t i o n 3.6.2 Write O p e r a t i o n 3.7 The A n a l o g - t o - D i g i t a l Converter 3.8 The Numerical D i s p l a y 3.9 C o n s t r u c t i o n o f the I n t e r f a c e CHAPTER 4 ON-LINE FILTERING OF DATA CHAPTER 5 SYSTEM PERFORMANCE 5.1 I n t r o d u c t i o n 5.2 P r e p a r a t i o n o f Strontium Samples 5.3 A n a l y s i s of Eimer § Amend SrC03 I n t e r l a b o r a t o r y Standard 5.4 Analysis of Strontium i n a Rock Sample CHAPTER 6 CONCLUSIONS BIBLIOGRAPHY APPENDIX I Ion Current Amplifier and Output Attenuator C i r c u i t II Scan Drive C i r c u i t III Summary of Interdata Programming Instructions IV Mass Spectrometer Interface Programming Guide V LIST OF FIGURES Page FIGURE 1. COMPUTER TO MASS SPECTROMETER INTERFACE 4 FIGURE 2. ANALOG DATA ACQUISITION SYSTEM BLOCK DIAGRAM 8 FIGURE 3. THE MULTIPLEXOR CHANNEL 10 FIGURE 4. BASIC LOGIC FUNCTIONS AND TRUTH TABLES 16 FIGURE 5. DEVICE ADDRESSING, LOGIC DIAGRAM 21 FIGURE 6. DATA AND STATUS INPUT, LOGIC DIAGRAM 23 FIGURE 7. DATA AND COMMAND OUTPUT, LOGIC DIAGRAM 24 FIGURE 8. INTERRUPT CONTROL, LOGIC DIAGRAM 26 FIGURE 9. COMPUTER/MASS SPECTROMETER INTERFACE READ/WRITE LOGIC 29 FIGURE 10. EXTERNAL CONNECTIONS TO READ/WRITE SEQUENCING LOGIC 30 FIGURE 11. READ SEQUENCING INSTRUCTIONS 32 FIGURE 12. WRITE SEQUENCING INSTRUCTIONS 32 FIGURE 13. ANALOG-TO-DIGITAL CONVERTER BLOCK DIAGRAM 35 FIGURE 14. DUAL SLOPE WAVEFORMS FOR DIGITAL VOLTMETER 36 FIGURE 15. THE NUMERICAL DISPLAY 38 FIGURE 16. MASS SPECTROMETER FILTER -DISPLAY PROGRAM 43 FIGURE 17. FREQUENCY RESPONSE OF DIGITAL FILTER 46 v i LIST OF TABLES TABLE I DATA PROCESSING EQUIPMENT PURCHASED TABLE II SPECIFICATIONS OF QUAD 2-INPUT NAND FUNCTION FOR THE FIVE LOGIC FAMILIES TABLE III INTEGRATED CIRCUIT PACKAGES USED IN CONSTRUCTION OF INTERFACE LOGIC CIRCUITRY TABLE IV RESULTS OF ANALYSES OF EIMER AND AMEND INTERLABORATORY STANDARD SrCOj TABLE V RESULTS OF ANALYSES OF GREY ARGILLITE Page 7 18 40 54 54 ACKNOWLEDGEMENTS The success of this project owes a great deal to the various people,who assisted the writer. The project was i n i t i a t e d by Dr. R. D. Russell whose thoughtful guidance, enthusiasm and encouragement throughout were always appreciated. Special thanks go to J. Blenkinsop who wrote the necessary computer programs, imparted much useful advice, and was responsible for the well-functioning of the mass spectrom- eter. The writer is indebted to the technical staff of the Department of Geophysics, U.B.C, in particular C. Croucher, R. D. Meldrum, and E. J. B e l l i s whose specialized s k i l l s made a large contribution to the project. Thanks must also go to B. D. Ryan who provided the writer with much needed geological enlightenment and to Dixie Pidgeon for typing this thesis. The Data Technology Corporation and Interdata Incor- porated, very kindly gave permission for the reproduction of several pages from their equipment manuals. The entire computer system was purchased from funds granted to Dr. R. D. Russell by the National Research Council of Canada, Mobil Oil Canada, Ltd., and the Standard O i l (Indi- ana) Foundation, Incorporated. Final l y , the writer wishes to thank Dr. T. J. Ulrych whose c r i t i c a l appraisal of the manuscript in i t s f i n a l stages made for a greatly improved thesis. CHAPTER 1 INTRODUCTION 1.1 General Background During the past few years a number of important advances have been made in the f i e l d of isotope geophysics. New analytical techniques, such as the double spiking pro- cedures of Compston and Oversby (1969) and the triple-filament technique for lead (Catanzaro, 1967), benefit from a high order of precision in mass spectrometer analysis. In order to do f u l l justice to these improved analytical techniques i t i s necessary to increase the measurement precision of mass spectrometer ion-currents. One way in which this can be achieved is by means of d i g i t a l data collection and analysis procedures. Using an on-line d i g i t a l computer, Wasserburg and his associates at the California Institute of Technology have demonstrated the value of these procedures by achieving results of outstanding precision on analyses of strontium (Papanastassiou and Wasserburg, 1969) and gadolinium (Albee et a l , 1970) . At the University of B r i t i s h Columbia, research of d i g i t a l data collection systems commenced in 1963, using a gas-source mass spectrometer. A servo-voltmeter ion-current amplifier (Stacey et a l , 1965) was u t i l i z e d which had, as a primary output, the shaft rotation of a motor-driven potent- 2. iometer. Thus analog — t o - d i g i t a l conversion was easily achieved using a shaft position encoder (Weichert et a l , 1967) . Digital data was i n i t i a l l y stored on punched paper tape but, as the system was developed, the paper tape punch was replaced by an incremental magnetic tape recorder. In both cases, the data was processed using the f a c i l i t i e s of the University's Computing Centre which presently include a duplex IBM 360/67 computer. To maintain completely independent data acquisition systems for the three mass spectrometers operated by the Department would have required the purchase of two more tape recorders, and the construction of suitable interfaces. A small d i g i t a l computer was therefore purchased in 1969 and work began on the design of a suitable d i g i t a l logic interface to a mass spectometer. The on-line data acquisition system that was subsequently b u i l t i s capable of immediately and sim- ultaneously supervising the operation and processing the data from the three mass spectrometers, which are used at different times for isotopic analysis of Pb, Th, U, Rb, Sr, Gd, Eu and Sm. The system description presented in this thesis is intended as an example of what is possible and no claim is made that the design is optimum in any sense. 3. 1.2 Design Objectives The system was conceived with three main objectives i n mind: 1. A decrease i n the o v e r a l l time required to process the data from the mass spectrometer. 2. A reduction i n operator involvement and hence a reduction of operator bias and operator s t r a i n . 3. An increase i n a n a l y t i c a l p r e c i s i o n . These objectives have been r e a l i z e d by using an on-line d i g i t a l computer interfaced to the mass spectrometer. The computer should be able to f i l t e r the incoming d i g i t a l data i n r e a l time and display the f i l t e r e d data points at the mass spectrometer console for the operator's convenience. In ad d i t i o n , i t should be possible to reduce the mass spectra from the input data and display the r e s u l t s at his request. The computing system should have the c a p a b i l i t y of automatically adjusting the operating conditions of the mass spectrometer. The system b u i l t has the f a c i l i t y for c o n t r o l l i n g the magnetic scan rate but addit i o n a l control features may be added at a l a t e r time. The writer's research was p r i m a r i l y concerned with the design and construction of the necessary d i g i t a l e l e c t r o n i c s for the interface together with a demonstration of i t s operating a b i l i t y . A s i m p l i f i e d block diagram of the computer to mass spectrometer interface i s shown i n Figure 1. SCAN READ PARAMETERS FUNCTION SELECT SWITCH READ SCAN RATE CONTROL WRITE ION EEATT ION CURRENT AMPLIFIER 4-DIGIT DIGITAL VOLTMETER DATA READY ± CHART RECORDER DISPLAY^ WRITE INTERFACE " H s E C T I O N-2 4 3 f e e t CABLE INTERFACE ION-1 ^SECT] INTERDATA MODEL-4 DIGITAL OMPUTER % V PEC 7-TRACK SYNCHR. TAPE RECORDER FIGURE 1. COMPUTER TO MASS SPECTROMETER INTERFACE 5. CHAPTER 2 GENERAL DESCRIPTION 2.1 The Mass Spectrometer The mass spectrometer used in this research project was designed principally by R. D. Russell in cooperation with F. Kollar, J. S. Stacey and T. J. Ulrych, and was brought to i t s present state of operating efficiency by J. Blenkinsop. It is a 90-degree sector, 30 cm radius, solid source machine, with single order focussing and a variable magnetic f i e l d scan. In the past and at present this machine has been used for rubidium-strontium analyses but i t i s in no way limited to these elements. The analog section of the mass spectrometer measuring system incorporates a hybrid amplifier using an electrometer vacuum tube and integrated c i r c u i t components. The voltage output from this amplifier can be attenuated by factors of 1, 1/3, 1/9, 1/27, 1/81 using a shunt switching network. The schematic c i r c u i t s for the ion current amplifier and output attenuator, both designed by R. D. Russell, can be found in Appendix I. The analog output of the measuring system, which i s generally of the order of one volt for most isotope peaks, i s fed to a d i g i t a l voltmeter which displays the input voltages and also functions as an analog - t o - d i g i t a l converter for the computer interface. 6. 2.2 The Computer An Interdata Model 4 computer was chosen chiefly because of i t s v e r s a t i l i t y at a reasonable cost. In addition, i t uses a halfword length of 16 bits and a programming language that i s similar to the University of B.C. IBM 360/67 computer with which i t can be easily interfaced i f required. The 8k 8-bit bytes of memory supplied with the Model 4 Pro- cessor are marginally adequate for sophisticated mass spectrom- eter control and data reduction programs, however additional memory modules may be added, up to a maximum of 65k bytes, at any time. A Teletype Model ASR33 and Peripheral Equipment Corporation PEC 3520-72 synchronous tape transport were pur- chased with the Model 4 Processor (see Table I ) . The teletype was supplied ready interfaced to the computer and the tape transport was interfaced in our laboratory by R. D. Meldrum using logic c i r c u i t r y of his own design. Figure 2 shows the complete analog data acquisition system for one mass spectrometer. The processor can communicate with the teletype and mass spectrometer via a multiplexor channel which has the capacity to handle a total of 256 devices. Each peripheral device is connected via a device controller to the multiplexor bus, i t i s this controller that provides the interface between the computer and an external device. The only essential requirement of the device i s that i t can supply data to the device controller in an acceptable d i g i t a l format. In 7. TABLE I DATA PROCESSING EQUIPMENT PURCHASED ITEM MODEL NUMBER MANUFACTURER APPROXIMATE COST (1969) PROCESSOR INTERDATA INC., $7,800 * 8,192 BYTE MEMORY INTERDATA INC., $6,000 * SELECTOR CHANNEL INTERDATA INC., $2,900 * HIGH SPEED ARITHMETIC § INPUT/OUTPUT INTERDATA INC. , $1 ,500 * TYPEWRITER $ PAPER TAPE READER/PUNCH ASR33 TELETYPE CORPORATION $1,900 MAGNETIC TAPE TRANSPORT PEC PERIPHERAL EQUIP- $4,000 3520-72 MENT CORPORATION DIGITAL VOLTMETER NUMERICAL DISPLAY POWER SUPPLY DT-344-2 DATA TECHNOLOGY $ 500 CORPORATION DS-103-5 DISPLAY GENERAL $ 300 INCORPORATED PS-200 DISPLAY GENERAL $ 60 INCORPORATED * - 1971 PRICES ARE ABOUT 401 LOWER THAN 1969 PRICES 8. CORE MEMORY HIGH SPEED MEMORY BUS PROCESSOR MULTIPLEXOR CHANNEL ~2W MULTIPLEXOR BUS SE- TS 7K DEVICE CONTROLLER 3 2 . TELETYPE DISPLAY SCAIM RATE CONTROL J D E V I C E " ^ CONTROLLER*1 DATA READY DVM SHUNT SELECTOR ION CURRENT AMPLIFIER MASS SPECTR- OMETER. SELECTOR CHANNEL SELECTOR BUS DEVICE ^ONTROLLEF; READ M.S. PARAMETERS MAGNETIC TAPE Figure 2. Analog Data Acquisition System Block Diagram the case of the mass spectrometer this is achieved by using the binary-coded-decimal (BCD) outputs of a d i g i t a l voltmeter. Figure 3 shows the 27 lines that constitute the multiplexor bus; 16 lines are reserved for data input and output which i s matched to the 8-bit byte. The 8 control lin e s , with one exception, are used to enable data onto the data lines. The control designated ACKO, together with the attention line (ATNO) and the system clear line (SCLRO), are associated with the processor interrupt f a c i l i t y which i s an optional feature of every device controller. The synchronize line (SYNO) i s used to notify the processor when a signal on one of the control lines is accepted by a device controller. A typical sequence of operations to service the mass spectrometer over the multiplexor channel would be: 1. A pulse from the d i g i t a l voltmeter, signifying that data i s available at the mass spectrometer, causes a signal to be sent along the attention line (ATNO) which interrupts the processor. The processor acknowledges the interrupt and sends a signal over the ACKO line which i n i t i a t e s a hardware scan cycle to determine which device caused the ATNO signal. The mass spectrometer automatically returns i t s device number to the processor over the data request lines (DRL's). The processor is now ready to service the mass spectrometer. 2. The mass spectrometer i s addressed by the processor over the 8 data available lines (DAL's). This address appears on the bus to a l l device controllers. INTERDATA PROCESSOR V INIT. 8 DAL 1s - 8 DRL's MULTIPLEXOR BUS • — 8 CL's • ACKO ^ o o SYNO ^ o < EH R AC KO  TA CK O RA CK O TA CK O ATNO SCLRO ' INI FAC1 INr FAC1 2-2 [NT! FACE - 3 M.S.I M.S. 2 M.S.3 FIGURE 3. THE MULTIPLEXOR CHANNEL 3. The processor then activates the address control l i n e which s i g n i f i e s that the DAL's now provide an address (as opposed to data). 4. The mass spectrometer interface decodes i t s address, sets a f l i p - f l o p memory, and sends a signal back to the proc- essor along the SYNO l i n e . The mass spectrometer remains addressed u n t i l another device c o n t r o l l e r i s addressed or u n t i l a system clear signal (SCLRO) i s received. 5. The processor places an "output command" on the DAL's. 6. The processor then activates the command control l i n e ; t h i s causes the data from (for instance) one p a r t i c u l a r decade of the d i g i t a l voltmeter, s p e c i f i e d by the output command, to be made a v a i l a b l e . A SYNO signal i s sent back to the processor to indicate the command has been stored i n the device c o n t r o l l e r . 7. The mass spectrometer i s again addressed by the proc- essor, as described i n steps 2, 3 and 4. 8. The processor then activates the data request (DR) control l i n e which enables the byte of data, made available i n step 6, from the d i g i t a l voltmeter to the processor, along the DRL's. A synchronize signal (SYNO) i s generated to indicate that the data has arrived at the processor. 12. 2.3 Programming the Interdata Model 4 The Interdata Model 4 computer possesses a repertoire of 75 basic instructions which can manipulate data between core memory, 16 general r e g i s t e r s , and up to 256 external devices. In addition, the Interdata system also provides for the d i r e c t transfer of a block of data between core memory and a peripheral device under control of an optional selector channel. Once i n i t i a t e d by the processor, t h i s d i r e c t transfer takes place i n - v i s i b l y without inter r u p t i o n to normal processing. Data of three d i f f e r e n t word lengths; the 8-bit byte, 16-bit halfword, and 32-bit fullword, can be manipulated by the i n s t r u c t i o n set. In the Interdata system hexadecimal notation (base 16) i s used to express binary information, so that, for example, a byte of data can be represented by two hexadecimal d i g i t s . Three i n s t r u c t i o n formats are available i n the Inter- data system: r e g i s t e r to r e g i s t e r (RR), r e g i s t e r to indexed memory (RX), and r e g i s t e r to storage (RS). A t o t a l of sixteen 16-bit general r e g i s t e r s , numbered Oto F i n hexadecimal notation, function as accumulators or index registers i n arithmetic and l o g i c a l operations. In a l l three i n s t r u c t i o n formats, b i t s 0-7 specify the machine operation to be performed (th 8-bit OP code); b i t s 8-11 specify the address of the f i r s t operand, which i s normally a general r e g i s t e r . In the RR format the address of the second operand i s s p e c i f i e d by b i t s 12-15 and i s always a general r e g i s t e r . In the RX i n s t r u c t i o n formats, b i t s 12-15 always specify the address of a general r e g i s t e r whose content i s used as an index value. The remaining 16 b i t s (bits 16-31) specify a memory address i n the RX format, and, i n the case of the RS format, an integer value for use as an immediate operand. A summary of the Interdata Model 4 programming in s t r u c - tions i s given i n Appendix I I I . The information necessary for program execution i s contained i n the 32-bit program status word (PSW). Bits 0-11 of the PSW define the status of the current user program; b i t s 12-15 constitute the 4-bit condition code (CC) which i s set a f t e r execution of input/output, l o g i c a l , s h i f t or arithmetic i n s t r u c t i o n s . The memory address of the next i n s t r u c t i o n to be executed i s s p e c i f i e d by the 16-bit i n s t r u c t i o n address f i e l d (bits 16-31 of the PSW). In instances of machine malfunctions, divide f a u l t s , i l l e g a l i n s t r u c t i o n s , and external device service requests, system interrupts are generated. When an interrupt i s recognised, the current PSW, which defines the present operating status of the processor, i s placed i n a reserved storage area (the old PSW) and a new PSW re-defines the status of the machine. On com- p l e t i o n of the interrupt service sub-program the previous machine status, stored i n the old PSW, i s restored. Input/output data transfer in the Interdata system can be either program controlled or interrupt c o n t r o l l e d . The former method interrogates the device to ascertain i f i t i s ready to transfer data, and waits i f necessary u n t i l transfer can occur. The interrupt method allows the device to demand service when i t i s ready for the transfer of data. This l a t t e r method i s the one employed for the mass spectrometer i n t e r f a c e . CHAPTER 3 THE INTERFACE HARDWARE DESIGN 3.1 Introduction. The construction of a successful computer int e r f a c e involves the design of suitable l o g i c c i r c u i t r y which w i l l enable the computer to communicate with the p a r t i c u l a r p e r i - pheral device. The hardware design i s evolved using the standard l o g i c elements a v a i l a b l e , namely gates, inverters and f l i p - f l o p s . Figure 4 summarises the most commonly used l o g i c elements together with t h e i r respective truth tables. In general, gates are used to d i r e c t the signals to and from the computer and f l i p - f l o p s are used to store information, acting as one-bit memories. Logic c i r c u i t design i s i n general easier than most analog c i r c u i t designs because only two voltage l e v e l s are employed, and the only major concern i s the duration and timing of these voltage s i g n a l s . There are two classes of s i g n a l , the steady voltage l e v e l and the pulse, the l a t t e r having a duration of a few tens of nanoseconds ( t y p i c a l l y ) . The timing of both types of signal are controlled by the processor and the peripheral device. 16. INVERTER &MC834P) AND GATE &MC1806P) NAND GATE &MC849P) OR GATE &MC1809P) A_ B NOR GATE A_ B CLOCKED FLIP-FLOP (MC845P) C_ J-K FLIP- FLOP &MC855P) t J Q C K TRUTH TABLES A B f 0 1 1 0 0 0 0 0 1 0 1 0 0 1 1 1 0 0 1 0 1 1 1 0 1 1 1 0 0 0 0 0 1 1 1 0 1 1 1 1 0 0 1 0 1 0 1 0 0 1 1 0 S-n, R* C> 0 0 0 1 0 1 0 1 1 1 0 0 Cv 0 1 0 1 0 1 I 1 FIGURE 4. BASIC LOGIC FUNCTIONS $ TRUTH TABLES Once the interface logic c i r c u i t r y has been designed, by suitably matching the computer and peripheral device spec- if i c a t i o n s to perform the required functions, i t is then only necessary to select appropriate integrated c i r c u i t logic packages (chips) to perform the required task consistently. When choosing logic packages there i s normally a choice between five different logic families namely, resistor- transistor logic (RTL), diode-transistor logic (DTL) , transistor- transistor logic (TTL), high-threshold logic (HTL), and emitter- coupled logic (ECL); Table II compares the 2-input NAND function for the five logic families. ECL i s a non-saturating form of logic which eliminates transistor storage time as a speed limiting characteristic, i t i s used where extremely high speed operation i s required. HTL was developed for applications such as in industry requiring a higher degree of inherent e l e c t r i c a l noise immunity than is available with the more standard integrated c i r c u i t logic families. Disadvantages of HTL are a re l a t i v e l y high supply voltage (15±1 volts) and slow speed. TTL i s a medium speed, high noise-immunity family of saturating integrated logic c i r c u i t s , i t is presently the most commonly employed logic family. DTL offers moderate speed and good noise immunity, i t i s somewhat inferior to TTL and used to be less expensive. RTL i s slow, has poor noise immunity and a large power dissipation, i t is no longer used commercially in high quality d i g i t a l electronics. TABLE II SPECIFICATIONS OF QUAD 2-INPUT NAND FUNCTION FOR THE FIVE LOGIC FAMILIES LOGIC FAMILY TYPICAL MOTOROLA NUMBER SUPPLY VOLTAGE VOLTS OUTPUT LOADING FACTOR RELATIVE NOISE IMMUNITY PROPAGATION DELAY ns TYPICAL TOTAL POWER DISSIPATION mW TYPICAL/PKG. APPROX. COST/ PKG. RTL MC9714P +3±.3 WORST 55 145 $1.55 oo DTL MC849P +5±l/2 7 MODERATE 25 66 $1.65 TTL MC7400P +5±l/2 10 GOOD 13 40 $1.29 HTL MC672P +15±1 10 BEST 110 114 $1.40 ECL MC1048P -5.2±l/2 25 BAD 130 $3.90 Diode-transistor logic was chosen for the mass spectrometer interface because i t s specifications are more than adequate for the relatively slow speeds involved and because, at the time of purchase, i t was s l i g h t l y less ex- pensive than the equivalent transistor - transistor logic, i t is also the logic family most used in the Interdata Model 4 Processor. Where a particular logic function was not av a i l - able in DTL then the appropriate TTL function was employed. An aggravating problem often encountered with d i g i t a l logic c i r c u i t r y is e l e c t r i c a l noise pickup, either, from adja- cent lines running in close proximity (interline crosstalk) or, from other sources. The best way of eliminating i t i s to use high-threshold logic (HTL) which operates at a 7.5 volt threshold level. This logic family i s somewhat inconvenient to use, un- fortunately, since a separate 15 volt regulated power supply i s required together with HTL/DTL level converters on a l l lines to and from the interface. In addition, HTL i s about five times slower than DTL. Using diode-transistor l o g i c , with careful c i r c u i t design, i t is generally possible to keep noise pickup at a negligible l e v e l . Where a c i r c u i t element i s required to drive a long l i n e , for instance between the processor and a per- ipheral device, a power gate and appropriate pull-up resistor are always used. This helps to provide good noise immunity, primarily by lowering the impedance level of the lines. In addi- tion, a l l long lines are made false-active, i . e . , a line i s active 20. when i t i s at ground p o t e n t i a l (zero v o l t s ) , to further reduce the p o s s i b i l i t y of noise pickup. The following pages describe the l o g i c c i r c u i t r y i n some d e t a i l . It has been found convenient, for c i r c u i t descrip- t i o n , to divide the interface into f i v e sections, namely, device addressing, data and status input, data and command output, interrupt c o n t r o l , and read/write sequencing c i r c u i t r y . 3.2 Device Addressing The Interdata device addressing l o g i c diagram i s shown i n Figure 5. The mass spectrometer address (hexadecimal D) i s wired into the device number s e l e c t i o n board. When t h i s address appears on the data available l i n e s (DALOO through DAL07) the decoded device output (DDO) goes low. A signal on the address control l i n e (ADRSO), i n conjunction with DDO, sets the address f l i p - f l o p so that i t s output (DENB1) i s made high. During the presence of ADRS1, a synchronize signal i s returned to the processor v i a the address synchronize l i n e (ADSYO). A delay of about 200 nano- seconds i s produced by capacitor Cl on the synchronize l i n e (SYNO). This prevents the processor from lowering the ADRSO l i n e before the address f l i p - f l o p has been set. The device en- able l i n e (DENB1) gates a l l other input/output control l i n e s to the i n t e r f a c e . When the address of another device appears on the data available l i n e s , DDI goes low, causing ADRS1 to reset the address DALO 0[> ATS Y N O 0- (r- 70- A D R S O o S Y N O <r 0 DEVICE NUMBER SELECTION -O -o o o -o DDO DDI AP_RS_1~ :470pF S Q T R ATSYN1 G O G7 0 DENB1 -DAO -D> -i> 4> -0 -0A7 -0 A D D R E S S FLIP-FLIP A D S Y O A T S Y N O -DRSYO ;470pF C D S Y O SRSYO —DASYO Figure 5. Device Addressing, Logic Diagram f l i p - f l o p , and disable the mass spectrometer i n t e r f a c e . The eight NAND gates, GO through G7, together with the l i n e s ATSYNO and ATSYN1, are part of the interrupt control c i r c u i t r y described i n section 3.5 3.3 Data and Status Input Figure 6 shows the Interdata input gating l o g i c diagrams. When the mass spectrometer i s addressed, DENB1 i s high, enabling the data request (DRO) or status request (SRO) control l i n e . The data or status byte i s thus enabled onto the data request l i n e s (DRLOO through DRL07). A return syn- chronize s i g n a l , DRSYO or SRSYO, i s automatically generated when ei t h e r of the control l i n e s is enabled. The l i n e s AO through A7 connect to the eight gates, GO through G7, shown i n Figure 5 and form part of the interrupt c o n t r o l l e r described i n section 3.5. 3.4 Data and Command Output The c i r c u i t shown i n Figure 7 i s used to control the flow of data and commands from the processor. When the mass spectrometer i s addressed, DENB1 i s high, enabling the data available (DAO) or command (CMDO) control l i n e . DAO or CMDO, i n turn, enable the data or command byte onto the data available l i n e s (DALOPO through DAL0P7). In addi- t i o n , control pulses are sent to the read/write l o g i c (Figure 9) 23. DENB1 D SRO 0— DRO 0 — <r <r <r <r 7<r SRSYO DRLO oo a p +5v DRSYO < 3 +5v 1470.0. AO A T A A A Q 9- ® » 4 A <& ^ A7 —© & SIIMOP —Oo DINOP —<]o -0 -0 -0 •0 -0 •0 -0 -0 -<3 -0 -0 -0 -07 -07 Figure 6. Data and Status Input, Logic Diagram Figure7 Data and Command Output, Logic Diagram v i a the DAGOP or CMGOP l i n e . The duration of these pulses i s shortened, from about 800 ns to about 400 ns, by the use of a one-shot multivibrator in each l i n e ; t h i s ensures that the control pulses are removed before the data disappears from the DALOP l i n e s . The li n e s DASYO and CDSYO return the respective synchronize signals to the processor. The data available and command control l i n e s are OR ed so that there cannot be any data or commands on the DALOP l i n e s except when one of the two control l i n e s i s a c t i v e . Thus t h i s OR gate eliminates extraneous noise on the data transmission cable, an advantage when several interface cables are run i n close proximity. 3.5 Interrupt Control The l o g i c c i r c u i t for the Interdata interrupt c o n t r o l - l e r i s shown i n Figure 8. The d e t a i l e d operation of t h i s c i r c u i t i s described in the Interdata"Systems Interface Manual" and only a b r i e f explanation w i l l be given here. The enable/disable switch i s set i n the enable p o s i t i o n to activate the interrupt c o n t r o l l e r . When a byte of data i s availab l e at the d i g i t a l voltmeter a data ready pulse i s gener- ated which causes the queue f l i p - f l o p to be d i r e c t s e t. The output from the queue f l o p - f l o p sends an attention signal (ATNO) to the processor v i a G12. The processor responds by returning 26. TACKO RACKO +5v4 | V 47Cvn| P G13 +5v d*70ii 0- ATSYNO ATIMO G12 * * +5v< +5v 470pF Q J T > K — >470.n. G i l 0 & C2 QUEUE FLIP-FLOP I 470pF .+5v DATA - 0 READY SCLROA <j ATSYN1 0 Figure 8. Interrupt Cont ro l , Logic Diagram 27. a receive acknowledge signal (RACKO) to the c o n t r o l l e r . The gate G8 disables G13 and prevents the transmit acknowledge signal (TACKO) from being sent to the next device. Gates G9 and G10 generate attention synchronize (ATSYNO) which sends a synchronize signal (SYNO) to the processor as well as causing the mass spectrometer address to appear on the inputs of GO through G7 (see Figure 5). This address i s then enabled onto the data request l i n e s by the ATSYNI output from G i l . On receiving the SYNO s i g n a l , the processor raises RACKO, causing the output of G i l to drop and the queue f l i p - f l o p to r e s e t . If the processor i s busy s e r v i c i n g another device i n t e r r u p t , when a data ready pulse i s generated, RACKO i s low and the mass spectrometer interrupt i s disabled. However, t h i s l a t t e r interrupt i s stored i n the queue f l i p - f l o p and i s serviced immediately a f t e r the previous interrupt has been serviced. A push-button switch situated at the processor (the i n i t i a l i z e switch) i s connected v i a the system c l e a r l i n e (SCLRO) to each interrupt c o n t r o l l e r such that a l l queue f l i p - flops can be d i r e c t reset simultaneously. The interrupt acknowledge control l i n e (ACKO) shown in Figure 3 i s divided up into a series of short l i n e s to form the daisy-chain p r i o r i t y system. Clearly the acknowledge signal must pass through every interface equipped with an interrupt c o n t r o l l e r , and the device situated closest ( e l e c t r i c a l l y ) to the processor, along the daisy-chain, has highest p r i o r i t y . 28. 3.6 Read/Write Sequencing The l o g i c c i r c u i t shown i n Figure 9 i s used to control the sequencing of the read and write operations c a l l e d for by the computer program. Data and command bytes from the processor arrive along the DALOP l i n e s and are fed to the output command (0C) memory and display l o g i c (Figure 15) v i a the four DALIP l i n e s . Three of the outputs from the 0C memory are used to drive a l-of-8 decoder which provides the sequencing signals for both read and write operations. The remaining output i s used to provide a read or write enable signal to a series of AND gates. Both the data bytes received from the d i g i t a l voltmeter and the data bytes to be written upon the display have to be sequenced i n a p a r t i c u l a r order. This i s accomplished p a r t l y by the software and p a r t l y by the sequencing l o g i c . The external connections to the read-write sequencing l o g i c (interface section 2) are shown i n Figure 10. 3.6.1 Read Operation Binary-coded-decimal (BCD) information i s available on four l i n e s from each decade of the d i g i t a l voltmeter. This data, together with the overrange d i g i t , i s gated onto the li n e s DIN0P4 through DIN0P7; the li n e s DINOP0 through DIN0P3 being unused. Additional information pertaining to shunt number, scan d i r e c t i o n and display function switch p o s i t i o n , are also gated onto the DINOP l i n e s . p L>n FIGURE 9. COMPUTER/MASS SPECTROMETER INTERFACE READ/WRITE LOGIC 4DAL0P- 5 : 6 : 7 : DAGOP: CMGor: 4 5 6 7 <D.U. DIGITAL VOLTMETER 17 LINES INTERFACE SECTION-2 READ/WRITE SEQUENCING LOGIC SHUNT NUMBER SCAN DIRECTION 1 FUNCTION 8 LINES SELECT i SWITCH 4 LINES 14 LINES DISPLAY AUTO RANGED TO SHUNT 'SELECT SWITCH STOP. ,GANGED TO SCAN UP /DIRECTION SWITCH SCAN STOP UP DOWN SEP. SEP. ~ REJ. ODOWN II it II >*<REJ, MAN, SCAN RATE CONTROL 2 LINES A SCAN V DRIVE FIGURE 10. EXTERNAL CONNECTIONS TO READ/WRITE SEQUENCING LOGIC Figure 11 shows a typical sequence of instructions necessary to read the d i g i t a l voltmeter, shunt number, scan direction, and display function switch position. It i s assumed here that the d i g i t a l voltmeter has generated an interrupt and the processor i s now ready to service the mass spectrometer. An "output command (0C)" i s sent from the processor, arrives at the flC memory (Figure 9), and is promptly stored on" receipt of a command strobe pulse (CMGOP). This f i r s t 0C contains the coded information requesting that the overrange d i g i t of the d i g i t a l voltmeter be placed on the DINOP line s . The next instruction i s "read data (RD)" which causes the data on the DINOP lines to be read into a specified location in core memory. The "add halfword register (AHR)", "compare logical halfword immediate (CLHI)", and "branch on low (BL)", instructions cause the 0C index register to be repetitively incremented by one u n t i l a l l available information has been read, one byte at a time, into separate locations in core memory. 3.6.2 Write Operation The sequence of instructions necessary to write (out- put) data from core memory (Figure 12) resemble those used to read (input) data from the mass spectrometer (Figure 11) . The "output command (0C)" sequences the output bytes in a similar manner to that described in Section 3.6.1, however b i t 4 of the OOOOR 0002R 0004R 0006R 0008R OOIOR 0012R 001 6R 001AR 001ER 0022R 0024R 0026R 0809 OAOB OCOD OEOF 0755 C820 0001 C830 OOOD DE35 OOOOR DB35 0008R 0A52 C550 0008 4280 001AR RCMD DC DC DC DC RDATA DS XHR LHI LHI RSTART OC RD AHR X ' 0 8 0 9 ' X'OAOB' X'OCOD' X'OEOF' 8 5, 5 2, H' 1 ' 3>RCMD<5) 3. » RDATAC 5 > 5* 2 CLHI 5*8 BL RSTART DEFINE CONSTANTS DEFINE STORAGE ZERO REG#5 LOAD 1 INTO REG#2 LOAD 13 INTO REG# 3 OUTPUT COMMANDC READ) READ ONE BYTE DATA INCREMENT REG#5 BY 1 CONTENTS REG#5<8? YES; BRANCH TO RSTART FIGURE 11. READ SEQUENCING INSTRUCTIONS OOOOR 0002R 0004R 0006R OOOCR OOOER 0012R 001 6R 001AR 001 ER 0020R 0024R 0001 0203 0400 0755 C820 0001 C830 OOOD DE3 5 OOOOR DA35 0006R 0A52 C550 0 0 0 5 4280 001 6R WCMD DC DC DC WDATA DS XHR L H I LHI WSTART OC WD AHR CLHI BL X'OOOl • X '0203 ' X '0400 * 6 5> 5 2#H'1 ' 3*H"1 3' 3*WCMD(5) 3>WDATAC5) 5* 2 5> 5 WSTART DEFINE CONSTANTS DEFINE STORAGE ZERO REG#5 LOAD 1 INTO REG#2 LOAD 13 INTO REG#3 OUTPUT COMMANDCWRITE) WRITE ONE BYTE DATA INCREMENT REG#5 BY 1 CONTENTS REG#5<5? YES; BRANCH TO WSTART FIGURE 12. WRITE SEQUENCING INSTRUCTIONS 0C i s now a zero causing the write l i n e to be made ac t i v e . When a "write data (WD)" i n s t r u c t i o n i s executed, a byte of data i s fetched from a s p e c i f i e d l o c a t i o n i n core memory and placed on the DAL's. A pulse on the DAGOP control l i n e strobes t h i s data byte into one decade of the display desig- nated by the previous 0C. The location of the decimal point on the display i s c o n t r o l l e d by a separate memory and decoder, and can be up- dated when required by a s u i t a b l y coded 0C and WD i n s t r u c t i o n . S i m i l a r l y the mass spectrometer magnetic scan rate can be varied using another memory. Each output of t h i s l a t t e r memory i s connected to a simple t r a n s i s t o r switching c i r c u i t which controls the frequency of a unijunction t r a n s i s t o r o s c i l l a t o r . This, i n turn, determines the magnetic scan rate v i a a stepping motor and potentiometer. The complete scan drive c i r c u i t , designed by R. D. Russell, i s given i n Appendix I I . In the present system there are four possible scan speeds which are considered ample for the programmed control envisaged A programming guide for the mass spectrometer i n t e r - face i s given i n Appendix IV. 3.7 The Analog-To-Digital Converter A Model 344-2 d i g i t a l voltmeter (DVM) manufactured by the Data Technology Corporation i s used as an analog-to- d i g i t a l (A/D) converter i n the data a c q u i s i t i o n system. This 34. instrument, u t i l i z e s the dual slope integration technique for A/D conversion. Figure 13 shows a block diagram of the DVM and Figure 14 i l l u s t r a t e s some t y p i c a l dual slope waveforms. The analog voltage output from the mass spectrometer measuring system i s applied to the input amplifier of the DVM. A pulse from the reset o s c i l l a t o r (frequency: 5Hz) i n i t i a t e s a 10,000 count such that the input signal i s integrated for a period of 50ms. This integration time i s co n t r o l l e d by a200KHz o s c i l l a t o r (clock). The integrating capacitor (Cl) i s d i s - charged u n t i l the 10,000 count i s completed ( f u l l s c a l e ) , leav- ing a voltage on Cl which i s proportional to the input s i g n a l . Upon reaching f u l l scale the input current (Ijjyj) i s switched o f f and a constant current source ( I ^ p ) i s switched to C l . The capacitor i s then charged at a constant rate while the counter continues to run. When the voltage across the capac- i t o r reaches the s t a r t voltage (15 v o l t s ) , a zero detect (ZD) pulse i s generated which resets the ZD f l i p - f l o p and disables the clock. The number of counts (N), accumulated by the counter, i s proportional to the input voltage. The output of the ZD f l i p - f l o p triggers a one-shot pulse (data ready) which enables the quatch-latch memories to accept the new BCD numbers into storage. The BCD outputs from these memories are used to provide the d i g i t a l information for the computer i n t e r f a c e . +200V V l N UN ^ P O S N E C i ) ( "1000" ~ ^ (" "100" ~~^) ^ "10" ^ "1" " " " ^ 5 r J K F/F JK F/F F U L L S C A L E S T A R T - F U L L S C A L E FLIP-FLOP Z E R O D E T E C T ( Z D I Q U A D L A T C H 8 4 2 1 D E C O D E R D R I V E R Q U A D L A T C H 4 2 1 C O U N T E R 200 kHr OSC RESET OSC ONE SHOT EXT S T A R T DATA READY FIGURE 13. ANALOG-TO-DIGITAL CONVERTER BLOCK DIAGRAM (Reproduced, by permission, from Data Technology Corp., Manual#18915-10) RESET to tfg tzo 1 1 { START FULL SCALE I FLIP-FLOP ^ 50ms ^ c 200KHz CLOCK ZERO DETECT DATA READY FIGURE 14. DUAL SLOPE WAVEFORMS-FOR DIGITAL VOLTMETER An important feature of the dual slope technique i s that the accuracy of the A/D conversion i s not dependent on the d r i f t of the 200KHz o s c i l l a t o r . Since the 10,000 count remains constant, the integrating time varies i n accordance with any o s c i l l a t o r d r i f t such that the t o t a l count, N, remains constant for a given input voltage. The Model 344-2 DVM has a f u l l scale voltage range of 0-1.0000 v o l t s with a 40% overrange c a p a b i l i t y . The manu- facturers quoted accuracy i s ±(.011 reading + .0001) v o l t s . 3.8 The Numerical Display The numerical output display for the mass spectrom- eter interface consists of 5 Datecon DS-103 display modules. Each module consists of a quad-latch memory which stores the input data (in BCD format) on application of a strobe pulse. The output from the memory i s fed to a BCD-to-Decimal decoder which drives a cold-cathode decade display tube. Each display tube incorporates a decimal point which can be c o n t r o l l e d by a separate quad-latch memory and decoder (Figure 9). Power requirements, in addition to the +5 v o l t s l o g i c supply, are a +200 vo l t s supply for the display tube anodes. Figure 15 shows the complete display l o g i c c i r c u i t . 3.9 Construction of the Interface The interface was constructed in two sections to try 38. O- 1 O- 2 O- 3 t>- 4 o- DAL IP 2 8 MEMORY 1 2 4 8 MEMORY BCD/DEC DECODER DRIVER COLD-CATHODE A+200V DECADE DISPLAY TUBE 0 MEMORY BCD/DEC DECODER DRIVER MEMORY BCD/DEC DECODER DRIVER BCD/DEC DECODER DRIVER MEMORY A A A A A DECIMAL POINTS FIGURE 15. THE NUMERICAL DISPLAY 39. and minimize the number of communication l i n e s between the computer and mass spectrometer. Section 1 consists of the address l o g i c , the data and status input l o g i c , the data and command output l o g i c , and the interrupt c i r c u i t r y b u i l t on two double-sided printed c i r c u i t boards which s l i d e into a rack situated inside the computer cabinet. Section 2 contains the read/write l o g i c c i r c u i t r y i l l u s t r a t e d i n Figure 9; the prototype unit constructed by the writer was b u i l t on 6 single-sided printed c i r c u i t boards which could communicate with each other through 22 pin connectors. The 6 printed c i r c u i t boards, numerical display, d i g i t a l voltmeter, and power supply are mounted on an aluminum chassis which f i t s into the mass spectrometer console. An aluminum front panel incorporates the display and voltmeter bezels and a l l the necessary switches. The integrated c i r c u i t (i.e.) l o g i c packages (chips) used i n the construction of the interface are mostly Motorola DTL p l a s t i c types with the exception of the complex functions (decoders and quad lat c h memories) which are of the TTL family (see Table I I I ) . Each 14 or 16 pin p l a s t i c package contains from 1 to 6 l o g i c elements depending upon the p a r t i c u l a r l o g i c function desired. A l l the i . e . packages employed i n the mass spectrometer interface were designed for operation from a 5±1/2 v o l t s regulated power supply (±5%). The power supply (PS-200) i n section 2 supplies the 200 v o l t s for the cold-cathode decade display tubes and a s t a b i l i z e d 5 v o l t s for the read/write l o g i c c i r c u i t r y . The i . e . packages i n section 1 u t i l i z e the Interdata Processor 5 v o l t TABLE III MOTOROLA NUMBER MC834P MC845P MC849P MC855P MC858P MC1803P MC1806P MC1809P MC4038P MC7442P MC7475P INTEGRATED CIRCUIT PACKAGES USED IN CONSTRUCTION OF INTERFACE LOGIC CIRCUITRY FUNCTION HEX INVERTER CLOCKED FLIP-FLOP QUAD-2 INPUT NAND J-K FLIP-FLOP QUAD-2 INPUT POWER 8-INPUT NAND QUAD-2 INPUT AND QUAD-2 INPUT OR l-OF-8 DECODER BCD/DEC DECODER QUAD LATCH MEMORY FAMILY DTL DTL DTL DTL DTL DTL DTL DTL TTL TTL TTL OUTPUT LOAD- ING FACTOR /OUTPUT 8 12 7 11 27 7 8 7 11 11? 10 PROPAGA- TOTAL POW- COST QUANTITY TION DELAY ER DISSIPA- /PKG USED IN ns TYPICAL TION mW TYP/PKG $ INTERFACE 30 40 25 40 30 25 35 30 45 45? 30 66 60 66 140 130 16.5 72 115 240 105? 160 2.00 1.62 1.65 2.10 2.70 1.55 1.90 1.90 6.65 6.75 4.50 8 1 17 1 17 2 6 1 1 1 3 o supply which serves to eliminate one l i n e between sections 1 and 2. An unshielded cable, approximately 43 feet i n length, i s used to transfer information on 13 li n e s between section 1 and 2; t h i s cable should be kept as short as po s s i b l e . No noise or crosstalk problems were experienced provided a l l the l i n e s were made f a l s e - a c t i v e and power gates, with a suitable pull-up r e s i s t o r , were used at the transmission end of each l i n e . In addition great care was taken to avoid ground loops which can e a s i l y arise during the construction of complex e l e c t r o n i c equipment. After the prototype interface had proved i t s e l f r e l i a b l e over several months of t e s t i n g , a double-sided, one piece printed c i r c u i t board was designed (by E. J . B e l l i s ) for the section 2 l o g i c c i r c u i t r y . Using t h i s new printed c i r c u i t board, two more complete interfaces were constructed and i n s t a l l e d i n the remaining two 30 cm radius mass spectrometers at the University of B.C., isotope geophysics laboratory. A l l three interfaces are now operating and are being used for isotope analyses, and i t i s intended that the one Interdata computer w i l l service a l l three mass spectrometers on a time sharing basis. CHAPTER 4 4.1 On-Line F i l t e r i n g of Data The simplest mode of operation of the data a c q u i s i t i o n system i s the on-line f i l t e r i n g of data from the d i g i t a l v o l t - meter and the display of the f i l t e r e d data on the numerical readout. Previous computer o f f - l i n e programs designed to pro- cess mass sp e c t r a l data have always incorporated some form of d i g i t a l f i l t e r i n g to reduce higher frequency noise. It therefore seemed reasonable to design a d i g i t a l f i l t e r program for the Interdata computer which would display the f i l t e r e d data points at the mass spectrometer console immediately, as well as storing a smoothed version of the mass spectrum i n a memory buff e r , from which i t could be transferred to magnetic tape v i a the Selector Channel. A suitable low-pass d i g i t a l f i l t e r had been designed by R. D. Russell and J . Blenkinsop for an IBM 360/67 and t h i s program was rewritten (by J . Blenkinsop) i n the Interdata programming language. A flow-diagram of the f i l t e r program i s shown i n Figure 16. The d i g i t a l voltmeter produces 5 data points/second which corresponds to a Nyquist frequency of 2.5Hz. The fund- amental sampling theorem requires that i n order to completely recover the o r i g i n a l s i g n a l , the sampling frequency must be 43. READ 5 CHARACTERS FROM DIGITAL VOLTMETER \ / CONVERT DECIMAL TO BINARY 7 POINT AVERAGE SKIP EVERY 2ND POINT 3 POINT AVERAGE 5 POINT AVERAGE > SKIP 2 OUT OF 3 POINTS STORE FILTERED POINT CONVERT BIN . TO DEC. WRITE 5 CHARACTERS ON DISPLAY FIGURE 16. MASS SPECTROMETER FILTER-DISPLAY PROGRAM at l e a s t twice the highest occurring frequency i n the sampled s i g n a l . Physical s i g n a l s , however, do not have a f i n i t e frequency content. The part of the signal spectrum l y i n g above the Nyquist frequency w i l l be r e f l e c t e d and superimposed (folded back) onto lower frequencies, and the o r i g i n a l s i g n a l can only be recovered approximately. The best one can do i s to choose the Nyquist (folding) frequency (and therefore the sampling rate) high enough to include a l l frequencies l y i n g i n the passband of the ion current measuring system. However, f r e - quencies above 2.5Hz do not contribute s i g n i f i c a n t l y to the mass spe c t r a l records encountered and a sampling rate of 5 point second should therefore be quite adequate. When sele c t i n g a f i l t e r for mass spectral data i t i s important that the width of the t o t a l averaging function (the data window) i s less than or equal to the mininum width of the peak tops. For the f i l t e r used, the width of the data window i s 3.8 seconds which necessitates dwelling on a peak top for a period of at least 3.8 seconds. D i g i t a l f i l t e r i n g i s accomplished i n the Processor by adding together points to form a moving average by sevens and applying every other averaged point to a three point moving average, followed by a f i v e point moving average. One out of every three points from the f i v e point average i s stored i n a memory bu f f e r , converted from binary to decimal notation, and then written on the display. The 7-point, 3-point, and 5-point averages a l l have tapered endpoints, which i s to say that the end-points have weighting c o e f f i c i e n t s of one h a l f (in our case) 45. This has the e f f e c t of reducing the amplitude of side lobes on the f i l t e r response (Figure 17) . It should be evident from the above description that there i s one f i l t e r e d point a v a i l a b l e , at the output of the f i l t e r , for every six raw input data points, hence the display i s updated once every 1.2 seconds. The Nyquist frequency i s lowered to 2.5/6 Hz aft e r f i l t e r i n g , but since there i s l i t t l e s i g n a l or noise s t i l l present above 0.1Hz, i t i s cl e a r that no information i s being discarded i n t h i s way. a FREQUENCY-CPS FIGURE 17. FREQUENCY RESPONSE OF DIGITAL FILTER CHAPTER 5 SYSTEM PERFORMANCE 5.1 Introduction i In order to test the accuracy and r e l i a b i l i t y of the computer interface two strontium isotope analyses were performed using both an interlaboratory standard and a rock sample of unknown composition. The mass spectrometer which has been inter f a c e d to the computer i s p r i n c i p a l l y used for the rubidium-strontium age dating of rock samples. Natural rubidium has two isotopes, Rb 8 5 which i s stable, and Rb 8 7 which i s radioactive and decays to S r 8 7 with a h a l f - l i f e of approximately 5x10 1 0 years. There are four stable isotopes of strontium, namely Sr 8 1*, S r 8 6 , S r 8 7 and S r 8 8 , and of these, i n a closed chemical system, only the abundance of S r 8 7 increases with time. I f the i n i t i a l amount of S r 8 7 i n the sample can be determined, and the present abundances of S r 8 7 and Rb 8 7 i n the sample are measured, the age of the sample can be determined. 5 .2 Preparation of Strontium Samples The method devised by B. D. Ryan (1971) for the chemical spearation of strontium from a rock sample i s used. The rock sample i s crushed to a fine powder and 48. approximately 0.25 gms i s \tfeighed out into a t e f l o n beaker. The sample i s completely dissolved i n H 2S0i t and HF, heated, and evaporated to dryness. The residue i s dissolved by warming with HC1 and allowed to cool. The sample i s c e n t r i - fuged and the soluti o n transferred to the top of a cation exchange column, of length 21 cms, containing Dowex 50W-X8 Resin (200-400 Mesh). The column i s eluted with 6NHC1 and the eluate c o l l e c t e d i n a measuring c y l i n d e r . The f i r s t 25 ml of eluate are discarded and the next 40 ml c o l l e c t e d . The cut of 40 ml i s evaporated dry and taken up i n 2 ml 2NHC1. Mean- while, the columns are back aspirated with 2NHC1. The sample (2 ml) i s added to the column and eluted with2NHCl. The f i r s t 90 ml of eluate are discarded and the next 40 ml c o l l e c t e d . This cut of 40 ml i s evaporated to dryness and the residue dissolved i n 3 drops 2NHC1. 5.3 Analysis of Eimer and Amend SrC0 3 Interlaboratory Standard The Eimer and Amend SrC03 was chosen because i t has been analysed at a large number of d i f f e r e n t laboratories, i n - cluding The University of B r i t i s h Columbia,where i t has been analysed a number of times previously using the same mass spectrometer. For the purposes of interlaboratory comparison the S r 8 7 / S r 8 6 r a t i o i s taken as 0.70800. The SrC0 3 i s dissolved i n 2NHC1 to produce a solution containing approximately 400ug strontium per ml of 2NHC1. Two drops of the solution are deposited on each of the outgassed rhenium side-filaments* and evaporated to dryness by passing a current of about one Amp through each filament while they are exposed to the atmosphere. The two side-filaments and a rhenium centre-filament are mounted i n a s t a i n l e s s s t e e l block which, i n turn, i s positioned inside the mass spectrometer. The whole system i s evacuated to a low pressure (< 2xl0" 7 mm Hg). The centre-filament i s heated by passing a current of about 4.0 Amps through i t . C o l l i s i o n with the hot centre- filament produces e f f i c i e n t i o n i z a t i o n of the sample that i s being gently evaporated from the r e l a t i v e l y cool side-filaments. The p o s i t i v e ions produced are accelerated by a p o t e n t i a l of 5000 v o l t s and can be focussed into a Faraday cup using a variable f i e l d strength electromagnet. The charges that c o l l e c t i n the cup, constitute the ion current which i s measured, the magnitude of t h i s current i s d i r e c t l y proportional to the abundance of the p a r t i c u l a r isotope ion beam focussed into the Faraday cup. The centre-filament current i s slowly increased and a search i s made for a Rb 8 5 contamination peak. I f any rubidium i s present, i t i s burnt o f f the side-filaments at a filament temperature just below that required to produce an appreciable strontium spectrum. When the height of the Rb 8 5 *The t r i p l e - f i l a m e n t technique of solid-source mass spectrometry was used for a l l the analyses. 50. peak i s n e g l i g i b l e , which implies that the height of the Rb 8 7 peak i s also n e g l i g i b l e (since the r a t i o Rb 8 5/Rb 8 7 = 2.593 i s constant), the centre-filament current i s increased to a value such that a S r 8 8 peak height of about 1 Volt i s obtained with the output attenuator set on shunt 3. A check i s made for rubidium contamination at t h i s new centre-filament temper- ature and i f a Rb 8 5 peak i s s t i l l detectable i t i s burnt away before the strontium spectrum i s scanned. The strontium 86, 87 and 88 peaks are scanned i n sequence about 12 times by varying the magnet f i e l d i n t e n s i t y using a peak-hopping technique. The peak heights and baselines are read from the f i l t e r display and are written by hand on the chart recorder which provides a v i s u a l record of the spec- trum. In order to calc u l a t e the S r 8 7 / S r 8 6 r a t i o for each scan i t i s necessary to correct the measured r a t i o for the growth or decay of the peak heights, then t h i s r a t i o i s corrected for mass discrimination at the ion source. Fortunately the peak heights grow or decay almost l i n e a r l y and hence a correction i s e a s i l y applied. Discrepancies i n the isotope abundances due to mass discrimination (fractionation) can be simply correc- ted for i n the case of strontium since the r a t i o S r 8 7 / S r 8 6 i s constant (=0.1194). The S r 8 6 / S r 8 8 r a t i o i s calculated for each scan and the discrimination per unit mass determined. An average value for the corrected (normalised) S r 8 7 / S r 8 6 r a t i o over a l l the scans i s calculated, together with a value for the standard deviation of the mean. The r e s u l t , given i n Table IV, indicates about a twofold increase i n p r e c i s i o n over previous values obtained by J . Blenkinsop, using the same mass spectrometer, without the computer i n t e r f a c e . 5.4 Analysis of Strontium i n a Rock Sample As a further test of the performance of the computer interface two analyses were performed on a rock sample. The sample .selected was a l i g h t grey a r g i l l i t e ( s l i g h t l y metamor- phosed claystone) from the Creston formation outcropping i n S.E. B r i t i s h Columbia. This formation i s part of the P u r c e l l Series which i s s t r a t i g r a p h i c a l l y equivalent to the Belt Series which outcrops i n western Montana and northern Idaho. Smith and Barnes (1966) r e f e r to these two series as the B e l t - P u r c e l l Supergroup. The Supergroup, which crops out over an area of more than 50,000 square miles, consists l a r g e l y of metamorphosed sediments which have not undergone intense deformation. Thick- nesses of greater than 40,000 feet have been attained. The sediments contain no useful datable f o s s i l s and therefore pro- vide excellent material for isotope dating studies of Precambrian rocks (1400 - 900 m.y. [ m i l l i o n years]). Obradovich and Peterman (1968) have dated rocks of the Belt Series using Rb-Sr and K-Ar techniques. Their deter- minations y i e l d ages ranging from about 900 m.y. to around 1300 m.y. Rb-Sr isotope measurements performed by Ryan and Blenkinsop (1971) on the Hellroaring Creek Stock i n the P u r c e l l Mountains indicate an approximate age of 1260 m.y. This stock, which intrudes the lowest known formation of the P u r c e l l Series (the Aldridge Formation) , i s the oldest recognized i n B r i t i s h Columbia. The sample of a r g i l l i t e was c o l l e c t e d by Dr. W. C. Barnes of the Geology Department, University of B.C., and was made available to the writer by Mr. B. D. Ryan. X-ray fluorescence measurements, performed by C. Croucher, have shown that t h i s sample contains 80 ± 2% ppm (parts per m i l l i o n ) rubidium and 248 ± 2% ppm strontium. The chemical spearation described i n Section 5.2 was c a r r i e d out twice to provide two strontium samples (A and B) from the one specimen of a r g i l l i t e . This duplicate separa- t i o n was used to provide a check on the r e p r o d u c i b i l i t y of the chemistry since, i d e a l l y , samples A and B should y i e l d i d e n t i c a l isotope abundance r a t i o s . Mass spectrometric analyses of samples A and B were performed i n a s i m i l a r manner to that described i n Section 5.3 for the Eimer and Amend standard. The much lower concentration of strontium i n the a r g i l l i t e samples necessitates much greater care i n the filament preparation. Consequently i t i s important to ensure that a l l the sample i s deposited on the side-filaments, which normally e n t a i l s p i p e t t i n g three or more drops of solution onto each side-filament. In order to perform this task, and s t i l l maintain a l l the sample within the middle one-third upper- surface of each side-filament ribbon, the following technique has been found s a t i s f a c t o r y . Place one drop of solution i n the centre of each side-filament, slowly evaporate the drop to dryness by passing a current of about one Amp through each filament, cool completely, add another drop to the middle of each ribbon, evaporate to dryness, and so on u n t i l a l l the sample solution i s shared equally between the side- filament. The current through each filament i s then increased u n t i l the filaments glow with a d u l l red color and are then l e f t to "cook" for about one minute. This l a t t e r procedure helps produce a more stable ion beam. Table V gives the re s u l t s of the two mass spectrom- eter runs for the samples A and B and the concentrations of the four isotopes of strontium i n the a r g i l l i t e sample. These concentrations were calculated using the X-ray fluorescence data and the r a t i o S r 8 7 / S r 8 6 derived from the mass spectrometer runs. The percentage deviation from the mean of the r a t i o S R 8 7 / S r 8 6 at a 95% confidence l e v e l i s about 0.05% which i s as good or better than previous analyses performed on rock samples using the same chemical separation procedure and mass spectrom- eter without the d i g i t a l f i l t e r i n g of data. O H . TABLE IV RESULTS OF ANALYSES OF EIMER S AMEND INTERLABORATORY STANDARD SrCQ 3 LABORATORY UBC (RUSSELL ET AL, 1971) UBC (RYAN § BLENKINSOP, 1971) USGS (STACEY ET AL, 1971) USGS (STACEY ET AL, 1971) MIT (SPOONER $ FAIRBAIRN, 1970) YALE (DASCH, 1969) U of T (PURCY § YORK, 1968) RATIO ( S r 8 7 / S r 8 6 ) n COMMENTS 0.7080±0.0002* 0.7082±0.0004 0.7080±0.0002 0.7079±0.0006 0.7082±0.0008 0.7075±0.0012 0.7080±0.0012 ON-LINE FILTER, HAND CALCULATED (THIS THESIS) HAND CALCULATED DIGITALLY RECORD- ED $ COMPUTED HAND CALCULATED HAND CALCULATED HAND CALCULATED HAND CALCULATED ACCEPTED VALUE 0.70800 TABLE V RESULTS OF ANALYSES OF GREY ARGILLITE SAMPLE A B MEAN OF A 5 B CONSTANT RATIOS: S r 8 V S r 8 6 - 0.0568 RATIO ( S r 8 7 / S r 8 6 ) n 0.7419±0.0003* 0.7420±0.0005 0.7419±0.0004 S r 8 6 / S r 8 8 = 0.1194 ISOTOPE CONCENTRATIONS (parts per m i l l i o n ) ±2% R K 1 Q r S 1 ' 9 T 8 6 < ! r 8 7 ^ r 8 8 (TOTAL) (TOTAL-' b r & r b r b T 80 248 1.3 23.9 17.9 204.8 * - ALL UNCERTAINTIES QUOTED ARE TWO STANDARD DEVIATIONS (95% CONFIDENCE LEVEL) CHAPTER 6 6.1 Conclusions The analyses described i n Chapter 5 were not expected to show any appreciable increase i n p r e c i s i o n over previous analyses, performed on the same mass spectrometer, without the computer i n t e r f a c e , but only to test the design and operating r e l i a b i l i t y of the system. With t h i s end i n mind one can only be very s a t i s f i e d with the res u l t s so far obtained. The f u l l p o t e n t i a l of t h i s on-line data a c q u i s i t i o n system w i l l only be r e a l i z e d when the processor i s programmed to determine peak heights and baselines, control scan rates and apply the various a n a l y t i c a l corrections. J . Blenkinsop has completed a prototype program to provide on-line processing of data from the mass spectrometer and his i n i t i a l r e s u l t s , from several t r i a l runs, indicate about a threefold increase i n p r e c i s i o n over previous o f f - l i n e analyses. No further improvements i n the hardware are envisaged i n the immediate future, although possible areas of investiga- t i o n may l i e with the computer control of source conditions and the d i r e c t reading of the magnetic f i e l d i n t e n s i t y for each peak. There remains great scope for improvements i n the software and i t i s i n thi s f i e l d that much e f f o r t i s being channelled. BIBLIOGRAPHY Albee, A.L., Burnett, D.S., Chodos, A.A., Eugster, O.J., Hun^eke,' J.C. , Papanastassiou, D.A., Podosek, F.A., Russ, G.P., Sanz, H.G., Tera, F., and Wasserburg, G.J. (1970). Ages, i r r a d i a t i o n h i s t o r y , and chemical composition of lunar rocks from the Sea of Tran- q u i l i t y . Science 167, 463. Catanzaro, E.J. (1967). T r i p l e filament method for s o l i d - sample lead isotope an a l y s i s . J . Geophys. Res., 72, 1325. Compston, W., and Oversby, V.M. (1969). Lead i s o t o p i c analysis using a double spike. J . Geophys. Res., 74_, 4338. Dasch, E.J. (1969). Sr isotopes i n weathering p r o f i l e s , deep- sea sediments , and sedimentary rocks . Geochim.. Cosmochim., 3_3, 1521. Data Technology Corp. Instruction Manual 344 4 Digit/Systems Meter (part No. 18915-10). Interdata, Inc. (1969). Reference Manual (publication No. 29- 004R02). Systems Interface Manual (publication No. 29-003R02). Motorola Semiconductor Products Inc. (1968). The Integrated C i r c u i t Data Book. Obradovich, J.D., and Peterman, Z.E. (1968). Geochronology of the Belt Series, Montana. Can. J . Earth S c i . , 5_, #3, part 2, 737. Papanastassiou, D.A., and Wasserburg, G.J. (1969). I n i t i a l Sr i s o t o p i c abundances and the re s o l u t i o n of small time differences i n the formation of planetary objects. Earth § Planet. S c i . Let t e r s , 5_, 361. Purdy, J.W., and York, D. (1968). Rb-Sr whole rock and K-Ar mineral ages of rocks from the Superior Province near Kirkland Lake, northeastern Ontario, Canada. Can. J . Earth S c i . , 5_, 699. Ryan, B.D., and Blenkinsop J . (1971). Geology and geochronology of the Hellroaring Creek stock, B.C. Can. J . Earth S c i . , 8, 85. 57. Russell, R.D., Blenkinsop, J . , Meldrum, R.D., and M i t c h e l l , D.L. (1971). On-line computer assisted mass spectrometry for geological research. Journal of the Physics of the Earth (in press). Smith, A.G., and Barnes, W.C. (1966). Correlations of and facies changes i n the carbonaceous , calcareous, and dolomitic formations of the Pre-cambrian B e l t - P u r c e l l Supergroup. B u l l . Geol. Soc. Amer.,77, 1399. Spooner, CM., and Fairbairn, H.W. (1970). S r 8 7 / S r 8 6 i n i t i a l r a t i o s i n pyroxene granulite terranes. J . Geophys. Res., 75, No. 32 , 6706. Stacey, J.S., Russell, R.D., and K o l l a r , F. (1965). Servo-am- p l i f i e r s for ion current measurement i n mass spectrom- etry . J . S c i . Inst., 42, 390. Stacey, J.S., Wilson, E.E., Peterman, Z.E., and Terrazas, R. (1971). D i g i t a l recording of mass spectra i n geologic studies, I. Can. J . Earth S c i . , 8_, 371. Wasserburg, G.J., Papanastassiou, D.A., Nenow, E.V., and Bauman, C A . (1969). A programmable magnetic f i e l d mass spectrometer with on-line data processing. Rev. S c i . Inst. , 40 , 288. Weichert, D.H. (1965). D i g i t a l analysis of mass spectra (Ph.D. Thesis, University of B r i t i s h Columbia). Weichert, D.H., Russell, R.D., and Blenkinsop J . (1967). A method for d i g i t a l recording f o r mass spectra. Can. J . Physics, 45, 2609. APPENDIX I. ION CURRENT AMPLIFIER $ OUTPUT ATTENUATOR CIRCUIT. ( Designed by R.D.Russell, June 1969. ) STEPPING MOTOR -28v L O nrrYTYn. APPENDIX II. SCAN DRIVE CIRCUIT (designed by R.D.Russel1 , June 1966) 60. APPENDIX I I I . SUMMARY OF INTERDATA PROGRAMMING INSTRUCTIONS* • CODE TYPE MNEMONIC INSTRUCTION 01 RR BALR Branch and Link 02 RR BTCR Branch on True Condition 03 RR BFCR Branch on False Condition 04 RR NHR AND Halfword 05 RR CLHR Compare Halfword 06 RR OHR OR Halfword 07 RR XHR Exclusive OR Halfword 08 RR LHR Load Halfword OA RR AHR Add Halfword OB RR SHR Subtract Halfword OC RR MHR Multiply Halfword OD RR DHR Divide Halfword OE RR ACHR Add with Carry Halfword OF RR SCHR Subtract with Carry Halfword 28 RR LER Floating-Point Load 29 RR CER Floating-Point Compare 2A ,RR AER Floating-Point Add 2B RR SER Floating-Point Subtract 2C RR MER Floating-Point Multiply 2D RR DER Floating-Point Divide 40 RX STH Store Halfword 41 RX BAL Branch and Link 42 RX BTC Branch on True Condition 43 RX BFC Branch on False Condition 44 RX NH AND Halfword 45 RX CLH Compare Logical Halfword 46 RX OH OR Halfword 47 RX XH Exclusive OR Halfword 48 RX LH Load Halfword 4A RX AH Add Halfword 4B RX SH Subtract Halfword 4C RX MH Multiply Halfword 4D RX DH ' Divide Halfword 4E RX ACH Add with Carry Halfword 4F RX SCH Subtract with Carry Halfword 60 RX STE Floating-Point Store O P C O D E - T Y P E M N E M O N I C I N S T R U C T I O N 68 R X L E F l o a t i n g - P o i n t Load 69 R X C E F l o a t i n g - P o i n t Compare 6A R X A E F l o a t i n g - P o i n t A d d 6B R X SE F l o a t i n g - P o i n t Subtract 6C R X M E F l o a t i n g - P o i n t M u l t i p l y 6D R X D E F l o a t i n g - P o i n t D i v i d e 90 R R U N C H Unchain 92 R R S T B R Store Byte 93 R R L B R L o a d Byte 96 R R W B R W r i t e B l o c k 97 R R R B R Read B l o c k 9 A R R W D R W r i t e Data 9B R R R D R Read Data 9D R R SSR Sense Status 9E R R O C R Output Command 9 F R R A I R Acknowledge Interrupt CO R S B X H . B r a n c h on Index High C l R S B X L E B r a n c h on Index L o w or E q u a l C2 R X L P S W L o a d P r o g r a m Status W o r d C 4 R S N H I A N D Ha l fword Immediate C5 R S C L H I Compare L o g i c a l Ha l fword Immediate C6 RS O H I O R Hal fword Immediate C7 R S X H I E x c l u s i v e O R Hal fword Immediate C8 R S L H I L o a d Ha l fword Immediate C A RS A H I A d d Ha l fword Immediate C B R S SHI Subtract Hal fword Immediate C C RS S R H L Shift Right L o g i c a l C D RS S L H L Shift Left L o g i c a l C E RS S R H A Shift Right A r i t h m e t i c C F RS S L H A Shift Left A r i t h m e t i c DO R X S T M Store M u l t i p l e D l R X L M L o a d M u l t i p l e D2 R X S T B ' Store Byte D3 R X L B L o a d Byte D5 R X A L Autoload D6 R X W B W r i t e B l o c k D7 R X R B Read B l o c k D A R X WD Wr i t e Data D B R X R D Read Data D D R X SS Sense Status D E R X O C Output Command D F R X A l Acknowledge Interrupt * Reproduced, b y permission, from Interdata Reference Manual #29-004R02. 62. APPENDIX IV MASS SPECTROMETER INTERFACE PROGRAMMING GUIDE ADDRESSES HEX D - M.S.2 HEX E - M.S.I HEX F - M.S.3 STATUS AND COMMAND BYTE DATA BIT NUMBER 0 1 2 3 4 5 6 7 STATUS BYTE DU COMMAND BYTE ( .-REAl )-WRIr 3 rE si •QUENi :ING DU - The device unavailable power supply i s o f f . READ - This command sets the data can be read from b i t i s set when the M.S. 5v output command memory so that the M.S. WRITE - This command sets the OC memory so that data can be written on the display. The scan rate i s also co n t r o l l e d in the write mode. SEQUENCING Bits 5,6 and 7 are used for sequencing the data i n both the read and write modes. READ SEQUENCING - COMMAND BYTE BIT NUMBER 4 5 6 7 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 •Read DVM overrange •Read DVM decades in sequence -Read shunt number -Read scan d i r e c t i o n code -Read function select switch Data i s read using the OC i n s t r u c t i o n followed by the read data (RD) i n s t r u c t i o n . WRITE SEQUENCING COMMAND BYTE BIT NUMBER 4 S 6 7 1 1 1 1 1 1 1 1 1 1 1 1 •Write on display decades i n sequence -Write decimal point -Set scan rate -Not used Data i s written using the OC i n s t r u c t i o n followed by the write data (WD) i n s t r u c t i o n . 64. DVM DECADE NUMBERING (Front view). OVER- RANGE 1 2 3 4 DISPLAY DECADE NUMBERING (Front view) 0 1 2 3 4

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 4 2
China 3 0
France 2 0
Canada 1 0
City Views Downloads
Redmond 3 1
Beijing 3 0
Unknown 2 15
Vancouver 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items