Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Modular multi-wire drift chambers for the TRIUMF QQD-spectrometer Forster, Brigitta Monica 1986

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1987_A6_7 F67.pdf [ 3.84MB ]
Metadata
JSON: 831-1.0096901.json
JSON-LD: 831-1.0096901-ld.json
RDF/XML (Pretty): 831-1.0096901-rdf.xml
RDF/JSON: 831-1.0096901-rdf.json
Turtle: 831-1.0096901-turtle.txt
N-Triples: 831-1.0096901-rdf-ntriples.txt
Original Record: 831-1.0096901-source.json
Full Text
831-1.0096901-fulltext.txt
Citation
831-1.0096901.ris

Full Text

MODULAR MULTI-WIRE DRIFT CHAMBERS FOR THE TRIUMF QQD-SPECTROMETER by BRIGITTA MONICA FORSTER B.Sc, U n i v e r s i t y of B r i t i s h Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of Physics We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1986 © B r i g i t t a Monica Forster 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 or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Physics  The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date Dec. 19, 1986 DE-6(3/81) - i i -ABSTRACT A modular, multiwire d r i f t chamber was b u i l t and tested i n the M13 area. The re s o l u t i o n obtained with s i x sense wires, was i n the hori z o n t a l d r i f t d i r e c t i o n a(mean) < (79.9± .2)Lim, with an angular re s o l u t i o n of (9.37±.02)mrads. With the d r i f t chamber mounted i n the f i r s t wire chamber p o s i t i o n of the QQD-spectrometer, t h i s corresponds to a re s o l u t i o n of 1.97mm FWHM at the p o s i t i o n of the s c a t t e r i n g target. In the v e r t i c a l d i r e c t i o n , using a charge d i v i s i o n method, the best achieved r e s o l u t i o n was < 2mm FWHM per wire, which gives an angular r e s o l u t i o n of 71mrads or 15mm FWHM at the target p o s i t i o n . The chamber was run at rates i n excess of 10 6 p a r t i c l e s / s e c . A multi-chamber system was tested, c o n s i s t i n g of two modular four c e l l d r i f t chambers f o r the h o r i z o n t a l d i r e c t i o n , and one eight c e l l modular d r i f t chamber, located i n between and rotated by 90°, was used f o r the v e r t i c a l p o s i t i o n determination. The re s o l u t i o n for one four c e l l chamber was a(mean) < (108.5± .9)uim, and for the eight c e l l chamber a(mean) < (79.9± .2 ) L i m . The angular r e s o l u t i o n f o r t h i s system was 1.37 mrad h o r i z o n t a l l y and 9.37mrads v e r t i c a l l y ; i n connection with the QQD-spectrometer, t h i s corresponds to FWHM of 0.267mm and 1.12mm resp e c t i v e l y at the target spot. - i i i -TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES v i LIST OF FIGURES v i i ACKNOWLEDGEMENTS ix CHAPTER I. INTRODUCTION 1 A. Motivation for this project 1 CHAPTER II. GENERAL PRINCIPLES OF A DRIFT CHAMBER 4 A. The single d r i f t c e l l 4 B. Basic phases 6 1. Thermalization 7 2. Drift phase 7 3. Avalanche multiplication 8 C. Basic Processes 9 1. Drift of charged particles in gases 9 2. Electron diffusion 10 3. Drift of electrons i n the presence of magnetic fields .. 12 - iv -TABLE OF CONTENTS (continued) CHAPTER III. PRINCIPLES OF OPERATION 14 A. Electric Field 14 B. Gas mixtures 15 C. Resolution 18 D. High rates 18 E. Charge division 20 CHAPTER IV. THE MODULAR MULTI-WIRE DRIFT CHAMBER (MMDC) 23 A. The concept 23 B. Construction of modules 24 C. Assembly 26 CHAPTER V. MMDC VERSIONS BUILT FOR THE TRIUMF QQD SPECTROMETER 29 A. Design considerations 29 B. The 8 wire MMDC 29 C. The 4 wire MMDC 31 D. Electronics 33 - v -TABLE OF CONTENTS (continued) CHAPTER VI. CALIBRATION SETUPS 38 A. Bench testing 38 1. Laser velocity calibration 38 2. 6-source velocity calibration 45 3. Laser efficiency tests 45 B. Beam testing 47 1. Velocity calibrations 47 2. Multiplicity tests for anodes 47 3. Data taking for resolution 50 CHAPTER VII. DATA ANALYSIS & RESULTS 53 CHAPTER VIII DISCUSSION AND FUTURE CONSIDERATIONS 71 APPENDIX 75 REFERENCES 76 - v i -LIST OF TABLES I. TDC Correction Coefficients 58 II. Resolution of the Eight Wire Chamber 60 III. Resolution with Changing Drift Field 61 IV. Resolution with Changing Gas Gain 62 V. Resolution of the Four Wire Chamber 63 VI. The Horizontal Angular Resolution in the Three Chamber System 64 - v i i -LIST OF FIGURES CHAPTER I. INTRODUCTION 1. QQD Spectrometer at Ml3 Focus 2 CHAPTER II. GENERAL PRINCIPLES OF A DRIFT CHAMBER 1. A Single Drift Cell 5 CHAPTER IV. THE MODULAR MULTI-WIRE DRIFT CHAMBER (MMDC) 1. The Two Basic MMDC frames 25 2. Stringing Jig 27 CHAPTER V. MMDC VERSIONS BUILT FOR THE TRIUMF QQD SPECTROMETER 1. Modular 8-Wire Drift Chamber 30 2. Mounted 4- and 8-Wire Chambers 32 3. Electronics Block Diagram 34 4. Preamplifier Circuit Diagram 35 5. Eurocard Circuit Diagram 36 - v i i i -LIST OF FIGURES (continued) CHAPTER VI. CALIBRATION SETUPS 1. Setup for Laser Calibration 40 2. Laser Calibration of Drift Velocity 41 3. Laser Velocity Calibration 42 4* Comparison of Beta-Source and Laser Calibration 44 5. Laser Efficiency Calibration for 6 Central Wires 46 6. Setup for "in-beam" Calibration 48 7. Velocity Calibration with Wire Chambers 49 8- Wire Multiplicity as Function of Anode Voltage 51 9. Setup for Three Chamber System 52 CHAPTER VII. DATA ANALYSIS & RESULTS 1. Track Fitting: Chisquare Distribution 55 2. Chisquare Distribution for 6 Wires of MMDC8 59 3. Charge Division Method with 55 F e source 67 4- Charge Division Method: Laser Calibration of y-Resolution . 68 5. Calculated Field Maps for MMDC8 69 6« Signs of Deterioration on Anode Wires 70 CHAPTER VIII DISCUSSION AND FUTURE CONSIDERATIONS 1. Calculated Field Map for Proposed Chamber 73 - ix -ACKNOWLEDGEMENTS I would like to express my thanks to my supervisors Dr. R.R. Johnson and Dr. K. Erdman as well as Dr. D. G i l l for encouragement and guidance during this project. Many thanks go to John Stewart and Eddie Knight for their expert advice and continual technical assistance. I am further indebted to Dr. D. Hutcheon for inciting discussions and suggestions for the data analysis. Thanks are also due to the members of the PISCAT group. Finally my deep appreciation goes to my husband Rolf, for his support and assistance in typing this manuscript. - 1 -CHAPTER I - INTRODUCTION I.A. MOTIVATION FOR THIS PROJECT At the front end of the low energy pion QQD -spectrometer at TRIUMF are presently provisions for three standard type Multi-Wire Proportional Chambers (MWPC). The spectrometer is usually running with one MWPC at the front of the f i r s t quadrupole magnet in the WC1 position (Fig. 1.1), and an other MWPC at the WC3 position between the second quadrupole and the dipole magnet. The horizontal (x) and the vertical (y) readout from each chamber is used to project the position of an incoming pion back to the target. The resolution of the spectrometer i s limited partly by the intr i n s i c resolution of these two detectors, as well as the multiple scattering induced mainly by their window f o i l s , and to a lesser degree the gas-contents of the chambers and the spectrometer i t s e l f . The int r i n s i c resolution of the MWPC's depends on their wire-spacings which are 1mm in the horizontal and 2mm in the vertical direction and the ch a r a c t e r i s t i c s of the delay-lines which are used for the identification of the Individual wires. A second limitation is imposed by the incoming particle flux which can be tolerated by MWPC's. As a consequence, the QQD spectrometer does not operate ef f i c i e n t l y at small forward angles. Fig. 1.1 QQD spectrometer at M13 focus. A l l available positions for wire chambers are shown. - 3 -Single wire readout chambers could c e r t a i n l y overcome the rate problem, but could not simultaneously Improve the r e s o l u t i o n . Dr. K. Erdman suggested the replacement of the two MWPCs with a d r i f t chamber, which would o f f e r improvements i n both r e s o l u t i o n and rates. It i s the i n t e n t i o n of t h i s work to document the bu i l d i n g and t e s t i n g of driftchambers based on h i s design and to comment on possible implementations. - 4 -CHAPTER I I - GENERAL PRINCIPLES OF A DRIFT CHAMBER D r i f t chambers are used extensively in high energy physics. They are best known for very high spatial resolutions as well as high rate capabilities. A d r i f t chamber consists of one or several d r i f t c e l l s . An individual d r i f t c e l l measures a single spatial coordinate of a charged particle which traverses a suitable medium. This is achieved by exploiting the physical message provided by the electromagnetic interaction of the particle with the medium. This chapter gives a brief review of the general principles which make d r i f t chambers possible. It is based mainly on three review articles [1-3]. I I . A . THE SINGLE DRIFT CELL The most basic c e l l consists of an anode wire and a cathode plane which are kept at high voltage in the gas f i l l e d d r i f t space. The electrons produced through ionization by a charged particle traversing the d r i f t space d r i f t towards the anode wire and produce an electrical signal. The d r i f t time, that is the time difference between the anode signal and a reference time signal (given for example by plastic s c i n t i l l a t i o n counters outside the d r i f t cell) is then proportional to the d r i f t distance Ax. A schematic picture of a single d r i f t c e l l is given in Fig. II.1. - 5 -Fig. II.1 A single d r i f t c e l l - 6 -For the reconstruction of the actual t r a j e c t o r y of a charged p a r t i c l e , two or more d r i f t c e l l s can be stacked behind each other. A l t e r n a t i v e l y , a multi-anode d r i f t chamber can be used. II.B. BASIC PHASES A charged p a r t i c l e traversing the gas i n a d r i f t chamber ion i z e s along i t s track. The primary electrons created i n t h i s process lose t h e i r energy by further i o n i z a t i o n and e x c i t a t i o n of gas atoms and/or molecules u n t i l they become s e n s i t i v e to the applied e l e c t r i c f i e l d and s t a r t to " d r i f t " . This means the electrons are accelerated by the e l e c t r i c f i e l d towards the anode, having multiple c o l l i s i o n s with gas atoms and/or molecules along the way. A global motion i n the d i r e c t i o n of the e l e c t r i c f i e l d i s superimposed on the random thermal motion. The average displacement of the electron swarm per unit time i s c a l l e d the d r i f t v e l o c i t y . F i n a l l y , i n the immediate v i c i n i t y of an anode wire, the e l e c t r i c f i e l d r i s e s r a p i d l y . This results i n an avalanche m u l t i p l i c a t i o n which i s necessary f o r a measurable e l e c t r i c s i g n a l . There are therefore three d i s t i n c t phases which have to be taken into consideration when designing or operating a d r i f t chamber: - 7 -II.B.l. Thermalization The f i r s t phase encountered is the thermalization of the electrons along the path of the incident charged particle. It is this physical extension and non-uniform density of the ionized t r a i l - due mostly to emission of delta electrons - which can limit the position accuracy of a d r i f t c e l l . This phase is dominated by the original energy of the ionizing particle, and the composition as well as the pressure of the d r i f t gas. II.B.2. Drift Phase After the thermalization process the electrons d r i f t in the direction of the applied electric f i e l d . A constant value of the d r i f t velocity makes i t possible to translate the measured time difference into a spatial distance. The value and s t a b i l i t y of the d r i f t velocity have direct bearing on the resolving power of the d r i f t c e l l and on the maximum rate at which the c e l l can be operated. The d r i f t velocity depends mainly on the homogeneity and strength of the electric f i e l d , and on the gas components and the gas pressure. - 8 -II.B.3. Avalanche Multiplication A v a l a n c h e m u l t i p l i c a t i o n i s i n d u c e d by the r a p i d l y r i s i n g e l e c t r i c f i e l d around the anode w i r e . T h i s e f f e c t s e t s the performance l i m i t s f o r the time-measuring e l e c t r o n i c s , which u l t i m a t e l y determines the e f f i c i e n c y o f the d e t e c t o r . The a n g u l a r s p r e a d of the e l e c t r o n a v a l a n c h e s depends s t r o n g l y on the c o u n t e r g a i n and - to a s m a l l e r e x t e n t - on the gas m i x t u r e and anode d i a m e t e r [1]. D e t a i l e d measurements have shown t h a t t h i s spread can be r a t h e r narrow a t moderate a v a l a n c h e g a i n s and i n w e l l quenched g a s e s . A narrow a n g u l a r s p r e a d of the a v a l a n c h e s not o n l y a l l o w s the d e t e c t o r to be run a t h i g h e r r a t e s , but a l s o to a c q u i r e an e x t r a space c o o r d i n a t e through the c h a r g e d i v i s i o n method (See c h a p t e r I I I . E ) . A n o t h e r b a s i c p r o c e s s t a k i n g p l a c e throughout the l e n g t h of t h e d r i f t p ath i s the d i s p e r s i o n due to d i f f u s i o n . I t depends on the k i n d of gas m i x t u r e , the p r e s s u r e , and the f i e l d s t r e n g t h and i n c r e a s e s w i t h the d r i f t d i s t a n c e . A s h o r t b a s i c i n t r o d u c t i o n to d i f f u s i o n and d r i f t o f charged p a r t i c l e s f o l l o w s i n the next two s e c t i o n s , w h i l e more d e t a i l s on e l e c t r i c f i e l d c o n s i d e r a t i o n s , s p e c i f i c gas m i x t u r e s and o p t i m i z a t i o n o f r a t e s and r e s o l u t i o n w i l l be d i s c u s s e d i n the r e s p e c t i v e s e c t i o n s of c h a p t e r I I I . - 9 -I I . C . BASIC PROCESSES I I . C l . D r i f t of Charged P a r t i c l e s i n Gases For ions the d r i f t velocities w+ increase linearly with the applied electric f i e l d up to values of several kV/cm.atm as [4]: w+ = u"hE/p (II. 1) where p is the gas pressure and u + is the ion mobility, which is specific for each ion in a given gas. The average ion energy In this region remains equal to the thermal value kT [4]. The electrons, on the other hand, can much more easily be accelerated by the applied f i e l d between collisions with gas molecules. They reach average energies far exceeding the thermal energy even at moderate f i e l d s . Electron theory gives a simple formula for the electron d r i f t velocity w~ [4]: w~ = eET/2m (II.2) where T is the mean time between collisions. Unfortunately i t was found that for some gases the co l l i s i o n cross-section and therefore T varies strongly with E, going through maxima and minima. This anomaly - 10 -(Ramsauer Effect) was f i r s t observed in Argon (incidentally a very popular d r i f t gas component) and later also in Krypton and Xenon as well as several organic compounds like Methane, but not in Helium and Neon. In addition i t was found [5] that even the addition of a very small fraction of another gas can modify the average electron energy and so dramatically change the d r i f t properties. II.C.2. Electron Diffusion A group of electrons in a d r i f t gas w i l l suffer dispersion due to diffusion. A charge distribution which at time t=0 is described by a delta function, w i l l be described at time t (along a given direction x) by a Gaussian distribution with the following standard deviation [3] : a = /2Dt" (II.3) x where D i s a f i e l d dependent diffusion coefficient and t is the time of d r i f t . It i s customary to define a characteristic energy as the ratio between the diffusion coefficient and the mobility [6]: - 11 -e k = eE - 5 - (II.4) where w is the d r i f t velocity. In the ideal case, where the electron energy i s not modified with increasing values of E, the d r i f t velocity increases linearly with the f i e l d . Then equals i t s classic value kT and the space diffusion for a d r i f t length x Is given by [6]: / 2kTx / T T c x a x = / — i E " <n'5> This quantity i s often called the thermal limit to electron diffusion. For any gas at a given temperature T, the average electron d r i f t velocity i s given by [7]: 4TT e E °° v 2 df w" = - ^ - — — / (II.6) 3 mN P 3 a (v) dv o m where N Is the number of gas molecules per cm3 at 1 torr, is the momentum transfer cross section, f i s the spherically symmetric term in the expansion of the electron velocity distribution function. If one combines II.3 and II.4 one obtains: - 12 -a x 2e, x P E 7T" 1 (II.7) It can be seen that the diffusion in any given d r i f t gas depends not only on E/P (since does) but also on ^ = . At a given gas pressure, the diffusion coefficient Is smaller for "cool" gases, i . e . gases in which the characteristic energy i s small even at high E/P values. It should be noted that the given a x represents the width of the charge distribution, and not the error on i t s localization. Furthermore, the d i f f u s i o n i s not always symmetric, e s p e c i a l l y at high f i e l d s . The longitudinal diffusion coefficient D Li (with respect to the dr i f t velocity) tends to be smaller than the transverse diffusion coefficient D^ . This classical approach does not work that well when applied to a gas mixture. Only calculations based on rigorous transport theory have accurately reproduced the experimental data. II.C.3. Drift of Electrons i n Presence of Magnetic Fields d r i f t direction modifies both the d r i f t velocity and the d r i f t path of the electron swarm. In noble gases the drifting electrons show a broad spectrum of velocities even at moderate fields, since most of their The presence of a magnetic f i e l d other than parallel to the - 13 -collisions are elastic. Consequently, In high transverse magnetic fi e l d s , the Lorentz force v x B w i l l cause a wide range of deflection angles and thus impair the performance of the detector. Organic admixtures to the noble gases allow the drifting electrons to transfer energy quickly to rotations and vibrations of complex molecules. This leads to the desired narrowing of the velocity distribution for the electrons. Drift chamber operation In strong magnetic fields w i l l not be discussed any further here. Although a l l the d r i f t chambers constructed in this thesis were tested and intended to be used in the fringe f i e l d of the TRIUMF QQD spectrometer, i t has been shown that similar d r i f t c e lls f i l l e d with various gas mixtures could tolerate fields as high as 0.13 Tesla before corrections to the data were necessary [8]. - 14 -CHAPTER III - PRINCIPLES OF OPERATION III.A. ELECTRIC FIELD A homogeneous E-field is essential for a constant d r i f t velocity throughout a d r i f t c e l l . However, in the immediate vicinity of an anode, the homogeneity w i l l cease since the E-field varies rapidly as 1/r, where r is the distance from the wire center. This leads to an avalanche multiplication which is necessary for a measurable signal. The amplification increases exponentially with applied voltage above threshold. Threshold occurs when an electron can gain enough energy between collisions to ionize a molecule in the next c o l l i s i o n . Since the velocity distribution is a function of the electric f i e l d and the pressure through the ratio E/P, i t is extremely important to use a gas in which the d r i f t velocity saturates at a moderate threshold. Saturation should extend to the highest values of the f i e l d , making the chamber response almost independent from local imperfections and mechanical tolerances in the placing of the cathode and the f i e l d defining wires. In multi-wire d r i f t chamber designs, there are often f i e l d wires at ground potential alternating with anodes (see Fig. V . l ) . Their function is to help to separate individual d r i f t cells and thus to focus incoming electrons. Those f i e l d wires also prevent electrostatic coupling between anodes. - 15 -III.B. GAS MIXTURES The composition of the gas mixture used in a driftchamber w i l l depend very much on the specific experimental requirements; such as high rate capabilities, low working voltage, long lifetime,etc. Noble gases are often chosen as the main component since avalanche multiplication occurs at much lower fields than with complex molecules. However, during the avalanche process excited and ionized atoms are formed. The excited noble gases can return to the ground state only through a radiative process, and the energy difference between the f i r s t excited state and the ground state, (11.6eV for Argon, 16.6eV for Neon) is well above the ionization potential of any metal used for wires or cathode planes (7.7eV for copper, 6.0eV for aluminium). This puts limits on the allowed gain for the chamber before i t enters a regime of permanent discharge. On the other hand, most organic compounds (hydrocarbons, alcohols) have many non-radiative excited states due to rotations and vibrations. Because this allows the absorption of photons in a wide energy range, the quenching efficiency of a polyatomic gas increases with the number of atoms in the molecule. But large, complex molecules with very high quenching a b i l i t i e s w i l l at the same time reduce dramatically the gas-gain at the - 16 -anode wire. The problem with very simple molecules, like carbon dioxide, i s that secondary emission which may lead to a discharge, has occasionally been observed. The polymerization of hydrocarbons, especially at high counting rates, makes i t mandatory to operate the chamber in an open gas flow configuration. Non-polymerizing quenchers l i k e methylal [ (OCH.J )^CYi.^ ] are not effective against photo-ionization and secondary emission, but, i f added even in small quantities w i l l neutralize hydrocarbon-ions into a non-polymerizing species. It is advisable to choose a gas mixture in which the d r i f t velocity reaches an extended plateau at a specific electric f i e l d . It i s then extremely important that the chamber be operated above this f i e l d threshold at a l l times. Characteristics of different gases and gas mixtures as well as their associated velocities, are readily available in the literature. A compilation of many works is found in Ref. [ 9 ] . The general requirements for the performance of a d r i f t chamber depend on the detector design. A single d r i f t c e l l is characterized by the applied electric f i e l d , i t s length to depth ratio and the potential on the anodes. As far as the d r i f t gas i s concerned, a very high d r i f t velocity i s required i f the emphasis of the chamber is towards handling high rates. Better spatial resolution i s achieved, i f f i r s t of a l l the d r i f t gas is kept at high pressure and by chosing a - 17 -slow velocity "cool" gas, where the characteristic energy e of the electrons is close to the thermal limit e ~ kT Unfortunately, many of those very low velocity gases, like Di-methyl-ether, do not have a saturation velocity. Although very good resolution can be attained with i t , the chamber has to be constantly and carefully monitored [10]. While testing the chambers built for this project i t was found, that they worked well with either Argon- or Neon-hydrocarbon mixtures, as long as the noble gas content was not dropped below 50% in volume. In mixtures with more then 50% hydrocarbons, the signal to noise ratio in the electronics no longer allowed the chambers to be operated at maximum efficiencies. Except when stated, a l l d r i f t gases were mixed on site using the TRIUMF mobile mixing units. A l l tests were done at 1 atmosphere pressure with continuous flow. A l l flow meters have been individually calibrated for the pertinent gases. Methylal was only needed for the high rate tests. At these times the noble gas was bubbled through the liquid methylal which was kept at zero degrees Celsius. This corresponds to a (3 - 5)% methylal content In the noble gas. The gas doping for the laser calibration runs is discussed in section VI.A.l. - 18 -III.C. RESOLUTION The spatial resolution ln a d r i f t chamber Is limited mainly by: - the physical width and the s t a t i s t i c a l spread of the original ionization track - the value and sta b i l i t y of the electron d r i f t velocity - the dispersion due to diffusion, which depends on the gas mixture as well as the gas pressure and the electric f i e l d strength. Dispersion always increases with d r i f t distance. Both the physical extension of the ionization track and the diffusion can be reduced by increasing the gas pressure to several atmospheres. The actual depth of the d r i f t c e l l can further limit the resolution i f there is a time difference between ionization clusters belonging to the same track arriving at the same wire. III.D. HIGH RATES The count rate i s limited most severely by space charge accumulation in the d r i f t region, and an efficiency loss caused by the fi n i t e duration of the signals, usually referred to as dead time loss. - 19 -III.D.l. Space charge The positive ions which are generated in the gas amplification w i l l lower the electric f i e l d strength near the anode. Due to their greater mass, ions w i l l have a slower d r i f t velocity than electrons. This results in a pulse height drop and could consequently affect the chambers efficiency. Since this effect depends on the extension of the avalanche near the anode wire, the most important design parameters w i l l be the radius of the sense wire, gas gain and gas composition. III.D.2. Dead time At high rates, the signals occasionally overlap and can therefore not be counted individually. For a high rate chamber, the following design c r i t e r i a should be observed: - small wire spacing to reduce the rate per wire, - small gap width to reduce space charge effects and reduce signal t a i l s caused by late arriving ionization, - low noise preamplifier to operate at low gas amplification, - small anode radius to provide fast anode signals and to reduce local space charge, - 20 -large bandwidth amplifier to allow fast clipping of signals, - dead time of discriminator matched to width at base of clipped signal [11]• Finally, i f the chamber i s run at high rates over an extended period of time, additional care should be taken to preserve the condition of the wires in the chamber. Only highly purified gases can be used. The hydrocarbon content in the gas mixture may have to be reduced and a non-polymerizing quencher, such as Methylal, must be added. III.E. CHARGE DIVISION In the charge division method, the position of impacting particles i s determined along the wire, through the ratio of resistances. If Q and Q, are the amounts of charge collected at the up down up and the down end of the sense wire respectively, the position i s then given by: ^up ^down wire length Q + Qj ' 2 up down - 21 -In a d r i f t c e l l , the charge division method can be used to measure a second space coordinate in the y direction. If applied to the Ion signal on highly resistive cathode wires, i t gives the y coordinate, but due to the low mobility of the ions the rate capability of the d r i f t chamber would drop significantly. A more promising method would be to place resistive wires close to the anodes and process the electrostatically induced electron signal. It has been shown by several authors [12,13,14] that d r i f t time measurements and charge division can be performed simultaneously on the same resistive anode. Both ends of the wire are connected through capacitive decoupling directly to the input of low Impedance preamplifiers. The signals are then processed separately for charge division and for timing. The value of the anode wire resistance should be high in order to keep the noise level low and to improve the position resolution along the wire. On the other hand, a small resistance is required in order to have good linearity in the position determination and high resolution in the d r i f t time measurement. The resolving time for the charge division method is about one electrode time constant T for optimum position resolution and linearity [15]. (T = RC, where R, C are the resistance and capacitance per unit length of the line). A - 22 -detailed analysis [16] has shown that the best compromise for simultaneous position and d r i f t time resolution i s achieved by [13]: R = 2 T T ( L / C ) 1 / 2 where L stands for the inductance per unit length of wire. This corresponds to the c r i t i c a l damping condition for transmission lines. The charge measurement is seriously limited by the input impedance of the amplifier, the input stage of which should be a current amplifier [16]. - 23 -CHAPTER IV. - THE MODULAR MULTIWIRE DRIFT CHAMBER (MMDC) IV.A. THE CONCEPT The f i r s t attempts at the new QQD front end detector were the construction and the testing of a more conventional multi-wire d r i f t chamber. Out of the collected experiences during that project grew the design of modular d r i f t c e l l s . Rather than being limited to a single instrument which i s fixed by i t s rigid dimensions and design specifications, modular type d r i f t cells are capable of providing versatile detector combinations which can be easily built, maintained, and repaired whenever the need arises. They can also be adapted to different tasks which may require the change of the characteric depth of a d r i f t c e l l , as well as the replacements of wires by new ones of different materials or/and diameters. With the exception of the actual printing of the circuit boards, a l l of these d r i f t chambers were strung, assembled, altered, and calibrated at TRIUMF, without requiring sophisticated tools or equipment. - 24 -IV.B. CONSTRUCTION OF MODULES The modular dr i f t chambers were constructed entirely out of 1.5 mm thick printed circuit boards. The G-10 material is cut into identical frames with a 50 mm x 50 mm opening. 12 positioning holes are dr i l l e d symmetrically throughout the border. Only two circuit designs are necessary: The type A boards (see Fig. IV.la) contain the f i e l d defining wire cage and are used as the two end boards for the chamber. Field wires are 100 um Cu-Be, strung at a tension of 120 grams and are 5 mm apart from each other. The voltage drops evenly from the two cathodes on either side of the frame towards the centre wire, which i s grounded. Potential differences between wires are obtained through a chain of 12 MOhm resistors which are soldered directly onto the boards. The inside edges of the circuit boards, which define the cathode planes, are painted with a conducting graphite solution. The type B boards (see Fig. IV.lb) have the centre electrical contact pins offset towards one side. They are used for both the sense wires (anodes) and the potential wires, which are kept at ground potential. Anode wires are 20 um thick gold plated tungsten and are strung at a tension of 50 grams. The two anodes used for the charge division measurements are 20 ym thick high resistance (3.5 kfi/m) NiCr wires at a tension of 10 grams. type A board type B board Fig. IV.1 The two basic MMDC frames - 26 -The potential wires are 100 um thick CurBe at a tension of 120 grams. Care must be taken that the contact pin offset on the boards which are strung with potential wires are a l l towards the same side, while for a l l the sense wires the boards have to be rotated by 180 degrees. This gives a maximum distance between the contact pins of the grounded potential wires and the pins for the anodes (which carry h i g h positive potential), thus reducing the chance of sparking between them. As before, the inside edges of the frames which define the cathode planes are painted with a conducting graphite solution. For stringing the wires, a stringing j i g was used (Fig.IV.2). The printed board is positioned with four locating pins onto a sled which moves laterally. The wire sits in V-grooves located at either end of the circuit board and is brought to the required tension. A three position toggle determines the wire position either at centre or 200 iim to the right or l e f t of i t . The tolerance of the j i g was measured to be < 10 um. IV.C. ASSEMBLY The chamber is assembled by pushing threaded nylon rods through the positioning holes a l l around the ci r c u i t boards. Distances between anodes and potential wires, which determine the characteristics of a single d r i f t c e l l , are set by nylon washers which were machined to whlch to8gl e s , P°si c i 6 ft « t 4 s * t s , P O s i ^ o n [°cati Pins at °nto c e n t r e tne Three - 28 -the desired thickness of 0.5mm, rendering a c e l l depth of 4mm. A modular type driftchamber is built starting with one type A wire cage board, followed by as many sequential d r i f t c e l l s as are desired and finished with another wirecage board. Nylon bolts are used on either end of the threaded rods to fasten the frames together. The finished chamber is very rigid, even i f only three individual d r i f t c e l l s are used, which is the minimum of cells needed in order to solve the left-right ambiguity. This method of assembly not only allows for an easy exchange of wire planes, but also the addition of extra d r i f t cells for better resolution as well as the removal of cells i f there are, for example, space limitatons. Wire planes can be pre-fabricated with any number of different wire diameters and/or wire materials. This permits not only fast replacement of a damaged wire, but also the complete exchange of specific wires with different ones which are more suited to a new task or another set of circumstances. The characteristic depth of some or a l l of the consecutive d r i f t c e l l s can be changed simply by installing a set of washers with appropriate thicknesses. The assembled chamber was mounted inside a leak proof container with 0.5 mil thick Mylar windows. Gas tight feedthroughs have to be installed for the high voltages, the preamplifier bias voltage, the signals, and the d r i f t gas flow tubes. Pictures of mounted MMDC's are shown in Fig V.2. CHAPTER V. - MMDC VERSIONS BUILT FOR THE TRIDMF QQD SPECTROMETER V.A. DESIGN CONSIDERATIONS The specifications for the new chambers asked for high rate capability, high angular spatial resolution in the x-direction (width of the incident particle beam), and to a lesser degree angular re s o l u t i o n i n the y-d i r e c t i o n (height of the p a r t i c l e beam). Furthermore, the chambers should be run with the d r i f t gas at atmospheric pressure. To achieve this, a compromise was necessary between high spatial resolution and high rate capability in the design of the d r i f t c e l l . Two different systems were developed and tested. They are described in the following sections. V.B. THE 8 WIRE MMDC This system consisted of one single chamber, which in turn comprised eight consecutive d r i f t c e l l s , which were to measure particle trajectories in the x-z-plane. The third dimension y was to be determined via charge division read-out off the f i r s t and last wire. The layout for this modular 8-wire d r i f t chamber is shown in Fig. V . l . The dimension from cathode to cathode is 50 mm, i.e. the maximum d r i f t distance is 25 mm. The anode wires are set off-centre alternatively by ±200um in order to solve the left-right side - 30 -RESISTOR CHAIN to c < c E-H a as < o c -y/,//'/-/•;//A -\Y •\////7/?//7A -Y///7/ '7777A •Y7/7/7'7/77A  AV////T///7A - Y / / / / A •Y//////7S//A • Y7//7//7777A -\/////?////A • Y//////V/777\ -Y/////-//;;A -Ys 7s 7, ,'7/7/A •Y 7//77.7/777A •X//.A-/////A • \7//77/7/777\~-/A -Y////777//Y 25mm XL • ^ ^ o - ^ v//7///////\  V / / ' A Y / / / A Y7/7/7777/7A  &7/7/7////A Y/777777777A  U////7////A Y///;;7y//7Zi \///7~/7777\  177/7?/7/7A  \///,-77////A  \///7/~77/7A  V///>>//7//ZA  Y7/7//7v7777\ Y77/77T7777A '/////7V///A o-»»»^ i7/V ///7//7A LL I I FIELD DEFINING WIRES UZZZZZZZZZA-o CATHODE PLANE -HV1 ANODE +HV2 CORRECTION POTENTIAL -HV3 POTENTIAL WIRE AT GROUND Fig. V.l Modular 8-wire d r i f t chamber. Max. d r i f t distance i s 25 mm. Anode wires are offset ±200 ym to resolve left/right ambiguity. Two f i e l d correction wires. - 31 -ambiguity. Two f i e l d correction wires are connected to a separate power supply. This provides the option of applying any independent voltage which may be required to correct f i e l d aberrations specific to the special position of the f i r s t and last d r i f t c e l l in the chamber. The mounted MMDC8 is shown in Fig. V.2a. The high voltages for the cathodes, the correction wires, and the anodes are fed through an RF choke. The electric signals are picked up directly form the pins on the boards and decoupled through 1000 pF capacitors. They pass through a preamplifier and then leave the gas container. V.C. THE 4 WIRE MMDC The second system which was built for the QQD spectrometer used three chambers. The trajectories in the xz-plane were given by the f i r s t and third chamber, each of which consisted of four drift cells (MMDC4). The less important y trajectory was measured only in the second chamber, the previously tested eight-wire model (MMDC8), which was rotated for this purpose by 90 degree with respect to the other chambers. Since there is no need for a charge division measurement in this three-chamber system, a l l anode wires were made from 20 \im thick gold-plated tungsten wire. Fig. V.2b shows a mounted MMDC4. For this system, the feedthroughs are the same as described for the MMDC8. The gas container - 32 -B Fig. V.2 a) Mounted 8-wire chamber MMDC8 b) Mounted 4-wire chamber MMDC4 Copper shield for electronics i s removed i n both pictures - 33 -shown in Fig. V.2b was made specifically to hold one MMDC4. The other MMDC4 for this three-chamber system was simply mounted into the spare gas container used in the MMDC8 testing. V.D. ELECTRONICS The data acquisition block diagram is shown in Fig. V.3. The signals from each wire are fed through a 1000 nF decoupling capacitor into a fast rise-time (< 5 ns) hybrid preamplifier (see Fig. V.4 for a circu i t diagram). Regular anodes are read out from one side only, anodes used for charge division measurements are read out from both sides. The signals are brought outside the gas container, where they are further amplified and discriminated in the "Eureocard" (see Fig. V.5). Both the "Eurocard" and the hybrid preamplifiers were fabricated in the TRIUMF electronics shop. Out of the "Eurocard" comes a logic timing pulse for each wire. These pulses are fed through a twisted pair cable into the counting room, converted from ECL to NIM and provide the stop signals for an eight channel time to di g i t a l converter (TDC, LeCroy Model 2228). For the charge division measurements, the analogue signals from top and bottom of each wire are transmitted in a similar way from the Eurocard via twisted pair cable directly to an analog to di g i t a l converter (ADC, LeCroy Model 2249A). A separate ADC unit was used for - 34 -ZAA HKEAMP1JFIER . 0 DISCKMINATOB EVENT GATE FAN Fig. V.3 Electronics block, diagram - 35 -->+6V INPUT <-BFR31 100: 0.1 BAV99 GROUND < — 100 0.1 T^2 o.i v -BFR93 1 BFR93 -> + OUTPUT 50 -www-1/ BFR93 50 K BFR31 -> - OUTPUT 0.1 147 GROUND MMBF4416 -> -6V Fig. V.4 Preamplifier c i r c u i t diagram A l l capacitors in units of uF A l l resistors in units of fi - 36 -+ 12V<-INPUT*— BAV99 GKOUNW— r 0 1¥ 2N4416 10 0.1 -4 2K BFR93 BFR93 100 0.1 30 2K 2N4416 0.1 0.1 0.1 47 9> o m X + m w l £ ' t -o CO CO + + 450 450 0.1 Hh —I (-0.1 100 10 100 CM S 10 0.1 Hf-ECL OUTPUT Q 5 Fig. V.5 Eurocard c i r c u i t diagram A l l capacitors in units of uF A l l resistors in units of ft - 37 -each wire. One of the two signals was fanned out, discriminated, and used in coincidence with the "event gate" signal as input gate for the ADC. The event gate i s given by one discriminated s c i n t i l l a t o r output or the coincidence of the discriminated outputs of two sci n t i l l a t o r s for the bench testing. For the in-beam tests the "good event" strobe signal of the QQD spectrometer was used. The ADC's and TDC's are read via CAMAC into the memory of a PDP-11 process computer and logged onto magnetic tape. The f i n a l data analysis was done on the TRIUMF VAX-cluster. - 38 -CHAPTER VI. - CALIBRATION SETDPS VI.A. BENCH TESTING VI.A.l. Laser velocity calibrations The velocity calibration of a chamber includes not only the measurement of the d r i f t velocity i t s e l f , but also the determination of the variation of this velocity throughout a l l areas of the detector. Simulating tracks of ionizing particles with a pulsed laser gives the advantage of having a narrow, well collimated beam whose position can be defined with great accuracy. Laser calibrations have been the subject of much debate. The source of the ionization i s a two photon absorption in some low ionization potential impurity in the chamber gas [17]. Such impurities are d i f f i c u l t to control since they could come from different sources such as the outgassing of materials from which the detector was constructed, pump o i l vapour, or contaminated gas mixtures. This randomness of molecules available for ionization has lead to inconsistent data for different chambers, or even for the same chamber, tested at different times. A pulsed N^-Laser was used for this calibration which has a wavelength of 337nm (hv=3.68 eV). Ionization can occur whenever there are molecules present with an ionization potential I < 2hv (< 7.36 eV). - 39 -In order to obtain a controlled and reproducible intensity of ionization, the chambers were doped with small quantities (~60 ppm) of N,N-diethylaniline CgH^N(C2H^)^, which has an ionization potential of 6.99 eV. The gas induced ionization in the chamber i s expected to build up as the concentration of the dopant increases with time. The build-up was exponential over the f i r s t five to ten minutes and levelled out after ~ 45 min, which i s consistent with Ref. [17]. The setup for the laser calibrations of the QQD d r i f t chambers i s shown in Fig. VI.1. The laser beam was collimated with two quartz lenses and the beamspot cleaned up by a pinhole. A mirror, mounted on a micrometer transport (± 2jjm) deflects the beam through the chamber. The chamber i t s e l f could be displaced vertically (±200 mm) on i t s stand. An exposed plastic s c i n t i l l a t o r , mounted on a photomultiplier tube was used to trigger the electronic readout. During the tests the lights in the laboratory room were dimmed in order to protect the photomultiplier which was operating at a low Voltage (700V). A typical raw TDC histogram for a single wire is shown in Fig. VI.2. Varying peak heights have no significance as the exposure time changed with each position. Fig. VI.3 shows TDC channel number versus laser beam position for the Argon(70%)-Isobutane(30%) d r i f t gas mixture. The linearity of this plot demonstrates a constant d r i f t velocity throughout the chamber. - 40 -55 PH < o PH DRIFTCH AMBER ( mounted on u vertical mobile stage) SCINTILLATOR 55 i-J Nl E-" < PULSED N 2 LASER Fig. VI.1 Setup for laser calibration of d r i f t velocity and efficiency - 41 -HUH WIKE « 6 HIUlOCfcAM 01 01 *l A-DCT-4000 -3900 3600 3700 34,00 3500 3400 3300 2 3200 3100 3(>00 -2900 2800 2700 2600 2500 I 2400 N 2300 2200 C 2100 0 2000 u 1900 N 1B00 T 1700 S lfaOG X X IbOO 1400 1300 iroo 1 100 1000 900 BOO * iX 700 X XX 600 X X K 500 4 A X IX 400 XX X XX 300 XI XX XX 200 5XX XX 1 XX 100 7XIX XX6445XXX 500 OO *OG 00 tax X X X X 5 X X X X XX 1 X X X 9 4 x x ; i X X X 4 4 x >:x X X. X X XXX G X 3 I X X X XXX X X X X X X X XXX X X X X X X X X X X XXI X X X 6 7 X X X X X X kX X X X 7 X X X X 9 X '.iXXX X X X X t x X X X X X X X X9 XX XX XXX X XV i'l XX X X X X X X XX4 3 X X X X 2 8 X X X 5 41 (XX X X X X 1 X X X 7 2 X X X 7 X l X X X X X X X X 7 4 4 r > X X X X X 9 7 7 X X t X X B 6 3 5 X K X X X 4 4 3 5 X X X X S 3 2 4 X X X X X Fig. VI.2 Laser calibration of d r i f t velocity (Example for wire #6 of MMDC8) Peak separation corresponds to 2 mm shift of laser beam position. - 42 -" 1 4 0 0 PI cd o CJ Q E-1 2 0 0 1 0 0 0 8 0 0 6 0 0 <0 4 0 0 MMDC8 wire #6 • 1 1 1 < • • • • • • • • • - • -• • • • • • • • F i - 3 0 - 2 0 - 1 0 0 10 20 Distance to Anode Wire (mm) 30 Fig. VI.3 Laser velocity calibration. Linearity demonstrates constant d r i f t velocity throughout the chamber - 43 -The following observations were made concerning laser calibrations for d r i f t chambers: - No data should be taken within the f i r s t hour after doping. - Direct contact between the laser beam and any wire should be avoided, since the photo-electrons ejected from the wire surface, w i l l result in gigantic anode signals. - For chambers which need windows thicker than 0.5mil, i t w i l l be necessary to replace the Kapton or Mylar window with a material which i s transparent to UV radiation (i.e. Aclar). - Chambers which require aluminized windows are not suitable for laser calibrations. Due to the intrinsic width of the ionizing beam, laser calibrations cannot be used to find the true resolution of the detector. They are, however, a clean and fast method to determine the magnitude and uniformity of d r i f t velocities throughout a chamber. The laser beam was easily collimated over large distances (50-100 cm), which allows several chambers to be tested and calibrated in coincidence. A further advantage of laser calibrations is important i f chambers are to be used in strong magnetic fields: As the laser light path, which liberates the d r i f t electrons, is not affected by magnetic forces, the influence of any magnetic f i e l d on a d r i f t chamber can be studied directly by using this method. - 44 -a) 1 0 6Ru Source b)Laser Fig. VI.4 Velocity calibration for MMDC8 (raw TDC histograms for wires #1,4,8) Neighbouring peaks in the histograms correspond to 5 mm shift of the calibrating beam. Deterioration of the spatial definition of the electron trajectory with increasing penetration into the chamber i s clearly v i s i b l e . - 45 -VI.A.2. B-Source Velocity Calibration A further velocity calibration method was attempted using a source of the 3 emitter 1 0 6Ru (100 uCi). The set-up from the laser testing (see Fig. VI.1) was used, with the collimated 3-source in the place where the mirror had deflected the Laser beam through the chamber. In the sequence of Fig. VI.4, i t can be seen that the 3-beam is starting out well collimated. When i t reaches wire #4, however, and even worse at the position of wire #8, i t s wide dispersion make i t unsuitable for the calibration of multi-wire d r i f t chambers. VI.A.3. Laser Efficiency Tests for Wires A graph which shows the efficiency of each wire as a function of the position of the ionizing track (Fig. VI.5) was made from the same data which were collected for the laser velocity calibration. The number of "triggers" received from the s c i n t i l l a t o r was compared to how many times each wire had fired. Since only one s c i n t i l l a t o r was used (the UV-laser beam was stopped in the plastic) rather than the coincidence of at least two, which would help to guard against noise triggering, the plotted efficiency results are only to be taken as lower limits. - 46 -O 1 0 0 , 8 0 . 1 0 0 . 8 0 . 1 0 0 . 8 0 . 1 0 0 . 8 0 . 1 0 0 . 1 0 0 . 8 3 . 0 , MMDC8 efficiency calibration (Laser) 1 O o ° o ° o o ° O 0 1 0 1 0 1 • • 1 • 1 • 1 1 • 1 • • • • • • • • 1 • 1 • 1 • 1 • 1 1 0 1 1 « 1 « 1 B R B B - B B B B H 1 B 1 B 1 B 1 B 1 1 • • 1 • • 1 , 1 I 1 1 0 5 10 15 20 2 5 30 Distance Beam-Anode (mm) Fig. VI.5 Laser calibration: efficiency of the 6 central wires of MMDC8 - 47 -VLB. BEAM TESTING VI.B.l. Velocity Calibrations For the pion beam velocity calibrations, the MMDC8 was set up between two conventional MWPC's (see Fig. VI.6). The location of the pion in the d r i f t chamber was calculated by projection, assuming a straight line trajectory through the two MWPC's space coordinates. For the purpose of testing a l l the d r i f t space, i.e. the f u l l width of the chamber, the beam had to be completely defocussed. In the scatterplot (Fig. VI.7) the traceback position in the dri f t chamber i s plotted against the TDC channel number. There is a similar scatterplot from each wire, and again the linearity of the d r i f t velocity extends throughout the chamber. For the purpose of calculating an exact d r i f t velocity, very narrow calibration s l i t s were defined at regular distance intervals. The resulting histograms were projected onto the TDC-axis, producing a peak for every calibration window. A straight line f i t through a l l the calibration peaks was used for calculating the d r i f t velocities. VI.B.2. Multiplicity Test for Anodes The MMDC8 was then mounted into the WC1 position (Fig. 1.1) at the front of the QQD spectrometer. In order to determine the gas - 48 -O u 00 K Ed u Ed K Ed CQ u Ed OS SS 1 PION B E A M 130 mm- 215mm-Flg. VI.6 Setup for "in beam" calibration of d r i f t velocity. - 49 -• i 1 1 1 1 1 1 1 1 1 490 0 1 480 0 1 . 470.0 I . . • : «< • 4*0 0 1...::.. . : «B. : 490.0 1..::::. . . . . . . :Xi. :. 440. 0 1 . . : . . : . . . . . . . : iX. . . 430.0 1 . . : . . : . . . XX.. . . 420. 0 1....::.:... . . .8X: 410 0 1 . . i : . . . . .SS: . . . 400. 0 - : : i : : . . .... XX 390.0 1. .::.:... . .Bi: 380 0 1 : . . . : • • 370.0 1 . . . • : . . . . ... .:»•:... 3*0 0 I. . . . . i : XX 3*0.0 1 ::•:... . . . : . 4X 340. 0 1 . . . : aa: • • • • • B : 330 0 I : . X : . . . lt«: 320 0 1 .a: . . . X6X: 310.0 I : . i . «X 300. 0 - . : : ! : . . . . : .BS: . . . . 890 0 1 . . . . X#X: SSO. 0 1 . . . . . X . • : * • « : 270. 0 1. • : XSX: T 2*0. 0 1 . . . .8: 8*i: I 290. 0 1 . :. X: : . . : i t i : « 240. 0 1 X . . : . • « : . . . . E 230. 0 1 . . :. X: :. . . : iXX: . . . 220. 0 1 . . IX. . . . :8Si: 210. 0 1 : iX . : *• 200 0 - . : . . • • • :B4i 1*0. 0 1 . : . Xl. • X*: : 180. 0 1 . . X. : . I B I : . . J70. 0 1 . . . . X :BS. : 1*0. 0 1 . . :. 8. : . . XB 190. 0 1 . . ,»X. . . . . 140. 0 1 Bi: : X«:. . . . . 130 0 1 . ,S. : . . . . :6S 120 0 1 . . 1 6 . : . . . . XX 110 0 1 . . . . : XX. . . . . . XX. . . ... ... 100. O - . : XB. : : Bi. *0. 0 I . . . . : XX: : . . 80. O I :SS: B i : 70 0 I . . : XX. : . . . : |XX: *0 0 I . . : X I I: . . . . . B. : 90 0 I Si: : aXi: : 40 0 I : BSX. XXi: . . . . 30.0 I. aOM.: . . . 20. 0 I . . . : XX,. . . 10. O I . . . 0 0 - • ' | | -2.90 -2.00 -1.90 -1.00 -0 90 0.00 0.90 1 OO 1.90 2 00 2 • 10*i KD1STX) Fig. VI.7 Velocity calibration with wire chambers - 50 -gain needed for an e f f i c i e n t operation of the chamber, the multiplicity, that i s the number of wires which had fired in response to a "good event" trigger, was monitored. A histogram of multiplicity versus the anode voltage i s shown In Fig. VI.8. VLB.3. Data Taking for Resolution Data has been taken with the MMDC8 chamber, when i t was mounted in the WC1 position, as i t was intended to be during experiments. At times the rate through the chamber was in excess of 10s particles/sec. The three chamber system was set up as shown i n Fig. VI.9 for both the velocity calibrations and the data taking. B 09 cr • M < M O i-h OO to 0) c 3 O o 3 8 HIRES FIRED 7 WIRES FIRED 6 HIRES FIRED 5 HIRES FIRED 4 HIRES FIRED 3 HIRES FIRED 2 HIRES FIRED 1 HIRE FIRED •0 O < a o. <! O rt QJ 00 lt> - 52 -X o S i - H H 55 I—f O C/2 -I I Ed m as o Ed X u o co u O u Q 2 K Ed 03 as o Ed OH PION BEAM — 32mm 126mm 79mm X O H 55 O CO 34 mm Fig. VI.9 Test setup for three-chamber system - 53 -CHAPTER VII. - DATA ANALYSIS AND RESULTS When the measured dr i f t times are to be converted into distances, the following points have to be considered: Each TDC channel has a specific delay time and the position of each anode wire in the chamber is only known within mechanical tolerances. While the electronic delays certainly could be calibrated, the exact determination of any wire position i s very d i f f i c u l t without having a detector available with at least similar or better spatial resolution than the chamber which is to be calibrated. Both the electronics delay times and the wire positions can be calculated, using a simple model for the chamber response. The TDC data depend on two types of parameters: a) Variables which vary from event to event, which are essentially due to track position and angle, and b) Variables which are constant during a run, such as the wire offsets and TDC channel delays. For each event, corrected times t ^ with respect to the "type b)" parameters can be computed: Cnk = 'nk + T D C k e l + £ n k W k f f del where t , i s the raw TDC reading for wire number k, TDC . allows for nk ° nk - 54 -differences i n delays in the el e c t r o n i c s , allows for the wire offset relative to the median plane of the anodes, and £ r+1 i f the track is to the l e f t n k S ' -1 i f the track is to the right This implies that e . , _ = - e , . v nk+1 nk The following is a step by step description of how the correction coefficients TDcf6^ and W.°^ were determined for the 6 inner k k wires of the MMDC8. The data set was restricted to events where a l l six wires had fired. The left-right ambiguity was resolved by f i r s t assuming that a l l particle trajectories were in the le f t half of the chamber. Then a least squares f i t for the trajectory to the data points (raw TDC values) was calculated. Due to the ±200 um wire offsets, the chi-square values for the tracks which actually did go through the l e f t half of the chamber were much lower than the values for the right half. Fig. VII. 1 shows that events can be clearly separated by this method. A similar histogram is obtained by assuming that a l l tracks went through the right half of the chamber. Assuming no wire offset, the particle track can be described by pos i t i o n and angle parameters ^ , T A n « The centre of the chamber was chosen as reference in order to make the error matrix diagonal. - 55 -0 3 H : H 14-jok WS CHI I • 0 0 o -3*0 0 I 300 0 I 370 • 14 3*0 O II 330 O II 340 O II 330 . O i i i a n o I I 310 O II 3O0 0 -I 8 *0 0 I I aao o II 370 O II 8 *0 O II 340 O II 340 e I I ax o II 830 O 810 0 o o o 0 0 o o 0 o 0 . o o o o o o o 0 o o 0 O  i « o II 170 1*0 I K 140 130 180 110 100 « 0 • 0 70 40 30 ao io e i n - I I i n i n i n i n i n i n - • • II II II CO w • H I O 1 CO E- I X. o OS I CO ja co E-H S 00 33411 4 m n •34 s i e i n i i I 9 447III4XI11IIII 7 4 I 9 I 4 7 I I I I I I I I E I I I I I 1 1 » I 114*1138 I I I I I I I I 3 4 I I I I I I I I I I 7 S III1IK11IIII4I4 1 I I I I I I I I I I I I I I I 4 1 4 4 I •IIIlIlllllI»IIIIII * iaS7»4433333l33iai 4 I I 1 30 00 40 00 •0.00 •O 00 TO 00 100 00 Fig. VII.l Track f i t t i n g : Chisquare distribution obtained assuming a l l trajectories in right half of chamber. Similar histogram i s obtained i f a l l trajectories are assumed in l e f t half of chamber. - 56 -Then the corrected time t ^ i s given by + (k - 3.5)T. (VII.2) nk 3.5n An The correction c o e f f i c i e n t s TDC^6^" and can now be determined by the maximum likelihood method (see Appendix). During the f i r s t pass through the data, compute the corrected times t', . TDcf^can be set to zero and the i n i t i a l guesses for W.°^^ nk k k o o can be the same for each wire. A least squares f i t i s made to determine the track position and angle. The 2x2 normal matrix i s computed, where the data set are the six t*. and the variables are T ~ ..and T . , the nk 3.5 An diagonal terms are the squares of uncertainties in T ~ c and T . . Now J. o An the least square value x 2 f ° r the event can be calculated. If o^ is the position resolution (in unit of d r i f t times), then X 2 = \Kn~ T3.5 " <k-3'5>TAn>2/ °l ' < Ak»'°k> 2 ( V I I ' 3 ) k=l k=l The i n i t i a l guess for a can be common to a l l wires, x 2 i s binned into a common histogram, as well as accumulated separately for each wire. The contribution from this event to the 2x2 error matrix for each wire can be calculated. The inverted error matrix is used to find corrections to TDC^ 6''' and W°^^ at the end of this pass, i.e. c^e^and - 57 -c,°^ are determined such that k A d e l - e c 0 f f I ( nk ' Ck - £k Ck }2 ( V I I > 4 ) n k (summed over a l l events), Is a minimum. At the end of the pass Oq i s multiplied by /y2/N (where N is the number of degrees of freedom, in this case N=4n), in order to get a new estimate for the resolution per wire for the next pass. Also make the substitutions: T D c £ e l + c £ e l -> T D c f 1 and „off , off v ..off a . , w. + c. ==> w. for use with the next pass, k k k It was found that only 2 to 3 passes were needed to find the correction coefficients. As shown In Table I, only very few events have to be analyzed to give reliable coefficients. Fig. VII.2 shows the chisquare distribution for a l l six wires after the f i r s t pass (without any TDC corrections) . The second column shows the chisquare distribution for each wire with the f i n a l TDC coefficients. - 58 -TABLE I TDC Correction C o e f f i c i e n t s // of Events c o r r e c t i o n to i n d i v i d u a l wires (unit s : TDC channels) analyzed #1 #2 #3 #4 #5 #6 375 -18.97 14.53 16.63 1.72 -16.60 2.69 3648 -19.00 14.03 17.16 1.73 -16.06 2.14 18547 -18.94 14.36 16.55 1.91 -16.18 2.31 a f t e r f i r s t pass of analysis with f i n a l TDC c o e f f i c i e n t s F i g . VII.2 Chisquare d i s t r i b u t i o n f o r a l l s i x wires of MMDC8. - 60 -TABLE II R e s o l u t i o n o f the E i g h t Wire Chamber (MMDC8 6 w i r e s c o n s i d e r e d ) GAS VELOCITY (um/nsec) a(mean) (um) ANGULAR (mrads) TARGET FWHM(mm) I 51.8 145+1 17.1±.l 3.59 I I 51.9 114±1 13.4±.l 2.81 I I I 40.9 77.4±.2 9.11±.02 1.91 IV 54.8 151±1 17.8±.l 3.74 V 60.8 143±1 16.8±.l 3.53 VI 64.4 126±1 14.8±.l 3.11 Key t o the gas m i x t u r e s : I ARGON 80% - METHANE 20% (PREMIX) I I ARGON 50% - ETHANE 50% (PREMIX) I I I ARGON 50% - ETHANE 50% (IN HELIUM)* IV NEON 80% - ETHANE 20% V NEON 65% - ETHANE 35% VI NEON 50% - ETHANE 50% Helium from the s p e c t r o m e t e r l e a k e d through M y l a r window i n t o the chamber. - 61 -TABLE III Resolution with Changing Drift Field (MMDC8 6 wires considered) FIELD (KV/cm) a(mean) (ym) a(wire #2) (ym) ANGULAR (mrads) TARGET (mm) FWHM 1.20 119.Oil.4 160.4±4.4 14.0±.2 2.94 1.40 113.7±0.6 148.8±1.9 13.4±.l 2.81 1.60 115.2±0.9 153.3+2.8 13.5±.2 2.84 (AR 50% - ETH 50%) PREMIX - 62 -TABLE IV Resolution with Changing Gas Gain (MMDC8 6 wires considered) ANODE (KV) a(mean) (um) a(wire #2) (ym) ANGULAR mrads) TARGET (mm) FWHM 1.90 113.9±0.5 148.0±1.7 13.41.1 2.81 1.85 114.0H.6 150.815.2 13.41.2 2.81 1.80 113.7±0.6 148.8H.9 13.41.1 2.81 1.75 113.7±1.8 155.715.7 13.41.2 2.81 1.70 116.0±1.6 166.615.6 13.61.2 2.86 (AR 50% - ETH 50%) PREMIX - 63 -TABLE V Resolution of the Four Wire Chamber (MMDC4 4 wires considered) GAS VELOCITY (um/nsec) a(mean) (um) ANGULAR (mrads) I 68.9 138.111.2 27.11.2 II 43.2 104.610.8 20.51.2 III 66.8 124.5+1.2 24.41.2 IV 67.6 130.010.9 25.51.2 Key to the gasmixtures: I (ARGON + METHYLAL)* 70% - IS0BUTANE 30% II ARGON 80% - ETHANE 20% III (ARGON + METHYLAL)* 80% - ETHANE 20% IV (ARGON + METHYLAL)* 50% - ETHANE 50% The Argon was bubbled through liquid Methylal wich was kept at 0° C. Estimated Methylal content ~(3-4)% of Argon content. - 64 -TABLE VI THE HORIZONTAL ANGULAR RESOLUTION IN THE THREE CHAMBER SYSTEM ( 2 MMDC4 positioned 200mm apart) GAS ANGULAR (mrads) TARGET FWHM (mm) I 1.63 0.34 II 1.29 0.26 III 1.47 0.31 IV 1.53 0.32 Key to the gasmixtures: I (ARGON + METHYLAL)* 70% - ISOBUTANE 30% II ARGON 80% - ETHANE 20% III (ARGON + METHYLAL) 80% - ETHANE 20% IV (ARGON + METHYLAL) 50% - ETHANE 50% The Argon was bubbled through liquid Methylal which was kept at 0° C. Estimated Methylal content i s ~(3-4)% of Argon content - 65 -To convert the TDC channels Into actual d r i f t distances and resolutions of wires, the calibrated d r i f t velocities were needed. A l l TDC measurements were done with the same unit (LeCroy Mod. 2228). Its time range was extended to 740 nsec to allow electrons to dr i f t the f u l l 25 mm maximum distance even In very slow gas mixtures. This TDC unit was then calibrated and found to measure 0.38 nsec/channel. The results of the beam tests, done with 50 MeV/c pions, are summarized in the following tables: Table II shows the resolution of the eight wire MMDC with different gas mixtures. Only the six inner wires were considered. The high resistance wires in the f i r s t and last positions did not give consistent responses, due mostly to electric f i e l d aberrations (see below). Most significant i s the improvement in the resolution due to the addition of Helium to the Argon-Ethane mixture. Table III shows the resolution for the same chamber with different electric fields, while Table IV summarizes how the resolution changes with changes in the gas gain, i.e. different anode voltages. The test results for the four wire MMDC are shown in Table V. while the horizontal resolution of the three chamber system tests are li s t e d in Table VI. The vertical dimension was determined by the previously tested 8-wire MMDC which was rotated by 90 degrees for this purpose. Therefore the vertical resolution for the three chamber system is also contained in Table II. - 66 -The results for the charge division measurements are shown in Fig. VII.4. The resolution is approximately 2 mm. As seen in the map of the equipotential lines for the MMDC8 (Fig. VII.5a), both wires used for the charge division read-out (In c e l l #1 and c e l l #8) are not very efficient for particles passing more than 10 mm away from the anode. On the other hand, the applied correction voltage was not able to correct the f i e l d near the anode (Fig. VII.5b). So for tracks close to the centre of the chamber, the f i r s t and last d r i f t c e l l had an effective c e l l depth of 6mm, which is much more than the 4 mm standard depth for a l l other c e l l s . It follows that the resolution in the f i r s t and last c e l l i s worse. A l l the chambers have been running with incoming pion rates in excess of 106 particles/sec. Testing of the chambers has been severely limited by frequent break-downs of the newly developed preamplifiers and amplifier -discriminators. The sense wires were damaged during the testing, when a premixed d r i f t gas was found to be heavily contaminated with propene. In Fig. VII.6 the magnified wire surface i s shown. - 67 a) Pulse height spectrum of 5 5 F e b) Oscilloscope signals from both ends of anode wire (top s i g n a l inverted, source at centre of wire) c l ) y=+20mm c.2) y=+10mm c.3) y= Omm c.4) y=-10mm c.5) y=-20mm c) Difference of signals from b) fo r d i f f e r e n t y-positions of the 5 5 F e source. V e r t i c a l : lOOmV/div. Horizontal. lOnsec/div. F i g . VII.3 Charge d i v i s i o n method with 5 5 F e source - 68 -FWHM = 2.0mm J I Fig. VII.4 Laser calibration of y-resolution with the charge division method. Peak shifts correspond to 5 mm. - 69 -A B Fig. VII.5 Calculated f i e l d maps for MMDC8 a) without correction voltage b) with correction voltage of 300 V. - 70 -B Fig. VII.6 a) Signs of deterioration on anode wire surface due to extensive sparking caused by contaminated d r i f t gas. b) Magnified section of wire from a). On the right a new wire for comparison - 71 -CHAPTER VIII. - DISCUSSION AND FUTURE CONSIDERATIONS The tested multi-wire d r i f t chambers (MMDC), due to the modular design, have proven to be a capable and flexible system. "In beam" tests, performed under conditions of a "real" experiment, showed the MMDC's to have high rate capability and very good spatial resolution for the coordinate perpendicular to the sense wires. Even though the charge-division read out techniques were not pursued very far, mainly because of f i e l d deficiencies in the respective d r i f t cells and the lack of suitable amplifiers, the achieved resolution along the sense wires is comparable to the second-coordinate resolution of the presently used MWPC's. Experiments using the QQD spectrometer, which require l i t t l e or no vertical resolution at the target spot, can be run with the existing MMDC8 chamber exactly as i t was set up and mounted for the testing in this thesis. There are big advantages over the 2 front MWPC set-up: much higher rate capability, reduced multiple scattering as two window fo i l s are eliminated and the spectrometer i t s e l f can run in vacuum. With the existing modules, one can also build a detector for experiments which do need vertical resolution by installing two 6-wire MMDC's, one of which is rotated by 90 degrees, at the front end of the spectromenter. This w i l l provide a system which, in addition to the advantages outlined above, offers significantly improved resolution at - 72 -the target spot both in the horizontal and vertical direction. Considerations would have to be given to the space available between the spectrometer and the target chamber as well as to the decoupling of the electric fields from the two MMDC's inside the gas container. A logical extension of the present MMDC system kit would be the construction of larger sized module frames. Multiple anode wire planes make i t possible to build d r i f t chambers of any size. A suggestion with regard to QQD spectrometer is to build modules of a size suitable to be placed into the WC3 position. As long as the maximum d r i f t space is kept at 25 mm and the depth of each d r i f t c e l l remains at 4 mm, MMDC's built with these units w i l l have a similar performance to the ones which have now been tested. The QQD spectrometer could then be run with e.g. a four-wire MMDC at the front end (WC1) and another four-wire MMDC at the WC3 position. Extremely high horizontal and vertical resolutions could be obtained, since horizontally a trajectory can be fitted through eight data points. Vertically, four data points would be available over a large baseline i f again only the f i r s t and last wire of each chamber is read out via charge division. The e l e c t r i c d r i f t f i e l d in the new modules can be considerably improved by f i r s t changing the wire spacings on the type "A" boards from 10 mm to 2 mm, and secondly, by placing two closely spaced wires in front of each anode plane. These wires have to be at the same potential as the anodes (see Fig. VIII.l). - 73 -Fig. VIII.l Calculated f i e l d map for the improved proposed modular system - 74 -The big improvement in the resolution due to the addition of Helium to the Argon-Ethane gas mixture should be further investigated. Helium is normally not used as a component in d r i f t gas mixtures. Therefore, systematic measurements of chamber resolution as a function of Helium contents of three component gas mixtures are necessary to establish and optimize the findings of this work. Finally, the most significant improvement for any of those d r i f t chambers would be the implementation of Charge Coupled Devices (CCD's) for a l l wire read outs, since this would render the preamplification of the anode signals obsolete. - 75 -APPENDIX - MAXIMUM LIKELIHOOD METHOD Call t= C J(a,,a_....,a ) the "theoretical" expression for x., where i i 1 2 n i the a are a set of parameters. In this case £. are linear functions of m 1 the parameters a : v m 2 C i = ^ c i m a m 1 = 1 " " 6 m=l Then build a "data vector" X N c, m i=l °i and a "measurement Matrix" M N c . c.. M - I - M. ml lm Then solve for a = M ^ "X (M i s now a 2x2 matrix) - 76 -REFERENCES [I] F. Sauli, Principles of Operation of Multi Wire Proportional and Drift Chambers, CERN Report 77-09 (1979) [2] G. Charpak, F. Sauli, High Resolution Electronic Particle Detecors CRN-EP (1984) [3] F. Sauli, Limiting Accuracies in Multiwire Proportional and Drift Chambers, Wire Chamber Conference, Vienna (1978) [4] E.W. McDaniel, E.A. Mason, The Mobility and Diffusion of Ions in Gases, Wiley & Sons, New York (1973) [5] E. Townsend, Electrons in Gases, Hutchinson, London (1947) [6] V. Palladino, B. Sadoulet, Nucl.Inst.Meth. 128, 323 (1979) [7] L.G. Christophorou, D.L. McCorkle, D.V. Maxey, J.G. Carter, Nucl.Inst.Meth. 163, 141 (1979) [8] A. Breskin, G. Charpak, B. Gabioud, F. Sauli, N. Trautner, W. Duniker, W. Schultz, Nucl.Inst.Meth. 119_, 23 (1974) [9] A. Peisert, F. Sauli, Drift and Diffusion of Electrons In Gases: A Compilation, CERN Report 84-08 (1984) [10] F. V i l l a , Nucl.Inst.Meth. 217, 273 (1983) [II] A.H. Walenta, Nucl.Inst.Meth. 217, 65 (1983) - 77 -[12] CW. Fabjan, J. Lindsay, F. Piuz, F. Ranjard, E. Rosso, A. Rudge, S. Serednyakov, W.J. W i l l i s , H.B. Jensen, F.O. Petersen, Nucl.Inst.Meth. 156, 267 (1978) [13] S. Bartalucci, R.Bertani, S. Berolucci, M. Cordelli, R. Dini, P. Giromini, H. Pallotta, A. R u t i l i , A. Sermoneta, M. Spadoni Nucl.Inst.Meth. 192, 223 (1982) [14] D.M. Binnie, Nucl.Inst.Meth. 192, 231 (1982) [15] J.L. Alberi, V. Rakeka, IEEE Trans.Nucl.Sci., NS-23, Nol, 251 (1976) [16] F. Schneider, CERN Report 82-06 (1982) [17] K.W.D. Ledingham, C. Raine, K.M. Smith, A.M. Campbell, M. Towrie, C. Trager, CM. Houston, Nucl.Inst.Meth. 225, 319 (1984) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096901/manifest

Comment

Related Items