- Email: [email protected]
High-accuracy, two-dimensional read-out in multiwire proportional chambers
NUCLEAR
INSTRUMENTS
AND
METHODS
HIGH-ACCURACY,
II 3
([973) 381-385; ©
TWO-DIMENSIONAL PROPORTIONAL
NORTH-HOLLAND
PUBLISHING
CO.
READ-OUT IN MULTIWIRE
CHAMBERS
G. C H A R P A K and F. S A U L I
European Organization for Nuclear Research, Nuclear Physics Division, Geneva, Switzerland
Received 13 April 1973 By measuring the center of gravity of the pulses induced on cathode strips in a multiwire proportional chamber it is shown that a good two-dimensional accuracy can be reached, of the
order of 150/~m. Applications for two-dimensional read-out in drift chambers or low energy X-ray mapping are discussed.
1. Introduction
pulses are induced on the c a t h o d e ; they are coincident with the negative pulses collected on the sense wires. S o m e o f the properties o f these pulses have been extensively studied at an early stage o f the multiwire c h a m b e r s ' development. It was shown that one can d e t e r m i n e the o r t h o g o n a l
In m o s t a p p l i c a t i o n s o f p r o p o r t i o n a l chambers, especially in high-energy physics, separate c h a m b e r s are used for m e a s u r i n g different coordinates. F o r some applications, such as for X - r a y or neutron m a p p i n g s , this is a l m o s t impossible a n d it is necessary to r e a d out two c o o r d i n a t e s f r o m a single-gap chamber1). In general one c o o r d i n a t e is o b t a i n e d by recording the pulses from the a n o d e wires a r o u n d which avalanches have grown. Several m e t h o d s have been i m a g i n e d for o b t a i n i n g the position o f an a v a l a n c h e a l o n g a wire. One type o f a p p r o a c h relies on the finite i m p e d a n c e o f the a m p l i f y i n g wire; either the a m o u n t o f charge o r the rise-time o f the pulse flowing at one end o f the wire is p o s i t i o n - d e p e n d e n t . The best accuracy o b t a i n e d in small c h a m b e r s by the rise-time m e t h o d 2) has been 150/lm, using very high resistivity a n o d e wires. A n i m p o r t a n t field o f a p p l i c a t i o n s for X - r a y c r y s t a l l o g r a p h y seems to be o p e n for c h a m b e r s with such accuracies, even if they are limited to one direction only, the accuracy in the o r t h o g o n a l direction being given by the wire spacing. A n o t h e r type o f a p p r o a c h , d e v e l o p e d by the PerezM e n d e z group3'4), is based on external delay lines c o u p l e d to c a t h o d e wires, o r t h o g o n a l to the sense wires. These are c a p a b l e o f locating the centre o f gravity o f the induced pulse with an accuracy o f 150/~m. C h a m b e r s for X - r a y m a p p i n g s have been successfully built by this method. I n this article we p r o p o s e a n o t h e r m e t h o d , which leads to the same range o f accuracies a n d m a y be preferred in some cases. W e also discuss the p r o b l e m o f accurate m e a s u r e m e n t s for large-size c h a m b e r s .
2. Properties of the induced pulses W h e n an a v a l a n c h e occurs a r o u n d a wire, positive 381
/ 1/o
Pulse height /
\\ o~.~
----
group of 10wwes FWHM=16cm single wwe FWHM:llcrn
Ililt/
I
/ / "\ l///
I,
10
\\
~---O o I 11
112
I 13
I 14
80
60
40
20 b) 11
12
13
1/* (cm)
Source position
Fig. 1. Variation of the height of the induced pulses as a function of source position: (a) Distance from the central wire to sensing wires = 2 mm. Curve l: pulse on a single wire. Curve If: pulse on a group of 10 wires connected together. (b) Distance from central wire to sensing wire = 7 mm. (From ref. 1.)
382
G. C H A R P A K
coordinate by using, for the cathode, wires whose directions are perpendicular to the sense wires1). In contrast to the situation existing for the anode wires, the pulse is not localized at the wire facing the avalanche but is spread over large distances. Fig. 1 shows the pulse-height distribution of the induced pulses as a function of the distance from the avalanche (from ref. 1) under very different conditions: gaps of 2 m m and 7 m m ; pulse in a single wire or in groups of wires. The striking fact is the extension of the induced pulses over one or a few centimetres. Similar observations have been made by Fisher and PlchS), who have studied the localization of the induced pulses on wide bands of 5 cm width, and have shown that simple calculations agree with the observations. In order to obtain the accurate position of an avalanche, it is necessary to measure the centre of gravity of thi's distribution. We will show that the shape of the distribution is so sensitive to the avalanche position that it is sufficient to measure the ratio of the pulses induced on two adjacent pick-up electrodes to get the centre of gravity of the avalanche with a high accuracy.
A N D F. S A U L I
The wires of one cathode plane were parallel to the sense wires, and those of the other cathode plane were orthogonal to them. The gas filling was argon + + isobutane + methylal (75%, 21%, 4%), and the chamber operated at a high voltage between 3 and 4 kV. For all measurements we used a well-collimated 55Fe source emitting 5.9 keV X-rays. Sets of several consecutive wires on the cathode plane were connected together and used as pick-up electrodes for the induced pulses. We have tried several different dimensions and distances for the strips, but the best results were obtained with strips of 4 mm, 8 mm apart. The positive pulses were amplified by a simple fast amplifier (Rin -~ 2 k~?), and the ratio of their pulse height was obtained in a fast divider.* The big advantage of operating the current division on the cathode electrodes is that high-impedance amplifiers could be used. The intrinsic noise of the system was measured, connecting the same pick-up electrode to the two * Developed at C E R N by J. Olsfors. F! 35+
1
This method was experimented on in a chamber having the following characteristics: chamber gap: anode wire diameter: anode wire spacing: cathode wire diameter: cathode wire spacing:
8 mm, 30/~m, 3 mm, 100/tin, 1 mm.
./
i
3. Localization by the ratio of induced pulses
/ /
/ /
2.5.
[13 "-. ,,q:
2.0'
/
b
8mm
/
//
,/ /
,0~E
•
,c r
, s~
'
/7 /
0 '
I0"
/ ~0.5
"'o "o
0 %
%
c
~
o
05-A/B
J 0
5
I0
mm
Fig. 2. Localization by the ratio of induced pulses. Pick-up strips of 4 mm, 8 m m apart, C h a m b e r gap L = 8 mm. Collimated source o f SSFe, separated by a step o f 2 m m along the s e n s e wires in the region between the two strips.
Fig. 3. Variation of the ratio of induced pulses. Full curve: ratio as a function of the avalanches along the sense wires; the error bars represent the fwhm o f the peaks. Dashed curve: the equivalent standard deviation as a function of position. The width of the source ( ~ 0 . 1 mm) and the noise ( ~ 3 0 % o f the peak width) are not subtracted.
MULTIWlRE
PROPORTIONAL
amplifiers; it was always about 30% of the measured peak's width (see next section) and was not subtracted from the data. 4. Space resolution 4.1. STRIPS ORTHOGONALTO THE SENSE WIRES Fig. 2 shows the separation of the ratios of the pulse heights as the source, well-collimated, is displaced by a step of 2 m m in the region between two neighbouring strips. The fwhm of the distribution corresponds to an accuracy of ~r = 200/~m. In fig. 3 the results of a complete scanning are shown. The ratio A/B of the pulse height on the two strips, whose centres lie 8 m m apart, is given on the left scale together with the fwhm of the peaks. On the right scale, the equivalent standard deviation in millimetres is presented. The width of the source (,-~ 100/~m) and the noise (30%) have not been subtracted. 4.2.
383
CHAMBERS
wire in a uniform way. Walenta et al.6), have observed that a right-left effect exists in the time of the appearance of an induced pulse in the adjacent wires; notice that e difference in pulse height could simulate a difference in time when threshold circuits are used. The accuracy in position determination is such that one could clearly separate two wires only 0.5 m m apart by current division on the induced pulses, and perhaps interpolate between them. 5. Two-dimensional read-out of chambers 5.1. IN SMALL CHAMBERS For small-size chambers of the order of 10 cm, one can adopt the following read-out scheme. The
0
•
STRIPS PARALLEL TO THE WIRES
We observe two phenomena: (a) The sense wires are clearly separated from one another. Fig. 4 shows the ratio A/B provided by two pick-up electrodes, 8 m m apart, when the source was collimated on two anode wires 3 mm apart. The resolution corresponds to about 150/,m. (b) There is a clear right-left effect, depending on the source position in respect to the anode wire. Fig. 5a shows the measured ratio A/B for two source positions, 1 m m at the left and 1 m m at the right of the same anode wire; the results of a complete scanning are shown in fig. 5b. This proves that the avalanches do not surround the
,•,.
•°
..
,,°,,%°•.,.
"...~..,•..•
.,,
.....
(a)
/" I
I
,~
I0
~
9 ~
O ~
09r
~ "
08L_
~
l
©
_
1
I
(b) m ~ f
. . . . .
8.O
~ 0 ~ t O ~ 0 Q m Q ~ Q I ~ q ~ Q ~
~ . . . . . .
~
lll~Q~
. . . . . . . .
Fig. 4. L o c a l i z a t i o n o f the sense wires by the ratio of induced pulses. R a t i o of the pulses i n d u c e d on two strips 8 m m apart, parallel to the sense wires. Wire distance: 3 mm.
Fig. 5. L o c a l i z a t i o n o f the r i g h t - l e f t side o f an avalanche development. W i t h the two pick-up strips on each side o f a sense wire the ratio is sensitive to the position with respect to the wire. (a) R a t i o o f the pick-up pulses f o r the two positions o f the source at 1 m m f r o m the wire. (b) R a t i o as a function o f the distance to t h e wire•
384
G. C H A R P A K
A N D F. S A U L I
cathodes are made of two layers of strips orthogonal to each other, the strips being 5 mm wide and separated by a 5 mm spacing. The pulse height in each of the 20 strips can be digitized to a reasonable accuracy, say a few percent; the central wires can be connected together and give the pulse gating the system. The position of an ionizing event is given roughly by the Anode
centre of the distribution, while the ratio of the pulse heights in two adjacent strips--or better, a best fit to the distribution--gives the accurate position 7). The attractive feature of the method is that the high accuracy is obtained with a very inaccurate measurement of the pulse heights. In this respect it is easier to operate than any current division method.
wires
5.2. I N LARGE CHAMBERS
Recently we have discussed the properties of drift chambers providing one coordinate with a very high accuracy (of the order of 100 pm) 8). [t is clear that a method giving the localization of the avalanche along the wires has the considerable advantage of avoiding the ambiguities arising when many particles are detected and two independent orthogonal coordinates are measured. Current division was discussed as a possible solution. For fast measurements required by high rates, the current division has to be done on wires with a low resistance. Accuracies of the order of 3 mm have been obtained
Pickup Electrodes
A
B
Fig. 6. Principle of a vernier m e t h o d for high-accuracy measu rements along a wire.
1,5
! /B
•
8m
r.
t,0 <:~
v
E E
b
1,0 ??-
,
,
~0.5
,,o "
L 0
¢
-¢
[
~-
5
I0 m
I
o
15
J
m
Fig. 7. R a t i o of the pulses induced in groups of alternate strips. Full curve: ratio as a function of the a va l a nc he s a l ong the sense wires; the error bars represent the f w h m of the peaks. D a s h e d curve: the equivalent s t a n d a r d deviation as a function of position. The width of the source ( ~ 0 . 1 mm) and the noise ( ~ 30% of the p e a k width) are not subtracted.
MULTIWIRE
PROPORTIONAL
by us and by other authors under such conditions for wire lengths of the order of 30 cm. For lengths of the order of metres the accuracy would be about a centimetre. The method we describe offers the possibility of a vernier method (fig. 6). Suppose that in a chamber such as the one described, the pick-up strips orthogonal to the anode wires are connected together alternately and provide the pulse heights A and B. The ratio A/B gives then, as before, the fine position between two strips, and the normal current division on the anode wire provides the gross coordinate. Fig. 7 shows the results of the measurement for a repetitive structure such as the one described. Here the pick-up electrodes were groups of four cathode wires, 8 m m apart, connected alternately to lines A and B. The accuracy is of course rather poor when the source is just in front of a strip, and improves to about 200/lm in the centre of the separation. As before, noise and source width were not subtracted. In a drift chamber of the type we have proposed a) it is not possible to have cathode wires perpendicular to the anode wires. However, one could imagine a long, thin, metallized strip facing the anode wires repeating the alternative pick-up electrodes structure we have described. 6. Conclusion
This report describes a method of allowing the
CHAMBERS
385
localization, with a very high accuracy, of the position of avalanches in an MWPC. The accuracy is of the order of ~ = 150 p m and has so far been shown to be limited only by the electronics. For small chambers of the order of 10 cm, as required for applications as X-ray crystallography, moderately simple systems based on our method can be envisaged. Although at this stage it does not pretend to be more accurate or more simple than the other existing methods, it may be preferable in some cases--for instance when non-delayed analogue information is needed--giving the position of a particle with the highest precision.
References 1) G. Charpak, D. R a h m and H. Steiner, Nucl. Instr. and Meth. 80 (1970) 13. 2) C. J. Borkowski and M. K. Kopp, IEEE Trans. Nucl. Sci. 17 (1970) No. 3, 340. 3) R. Grove, K. Lee, V. Perez-Mendez and J. Sperinde, Nucl. Instr. and Meth. 89 (1970) 257. 4) R. Grove, V. Perez-Mendez and J. Sperinde, Nucl. Instr. and Meth. 106 (1973) 407. 5) G. Fischer and J. Plch, Nucl. Instr. and Meth. 100 (1972) 515. 6) A. H. Walenta, J. Heintze and B. Schurlein, Nucl. Instr. and Meth. 92 (1971) 373. 7) G. Charpak, A. Jeavons, F. Sauli and R. Stubbs, Highaccuracy measurements o f the center of gravity o f avalanches in proportional chambers, C E R N 73-11 (Sept. 1973). 8) G. Charpak, F. Sauli and W. Duinker, High-accuracy drift chambers and their use in strong magnetic fields. Submitted to Nucl. Instr. and Meth.
INSTRUMENTS
AND
METHODS
HIGH-ACCURACY,
II 3
([973) 381-385; ©
TWO-DIMENSIONAL PROPORTIONAL
NORTH-HOLLAND
PUBLISHING
CO.
READ-OUT IN MULTIWIRE
CHAMBERS
G. C H A R P A K and F. S A U L I
European Organization for Nuclear Research, Nuclear Physics Division, Geneva, Switzerland
Received 13 April 1973 By measuring the center of gravity of the pulses induced on cathode strips in a multiwire proportional chamber it is shown that a good two-dimensional accuracy can be reached, of the
order of 150/~m. Applications for two-dimensional read-out in drift chambers or low energy X-ray mapping are discussed.
1. Introduction
pulses are induced on the c a t h o d e ; they are coincident with the negative pulses collected on the sense wires. S o m e o f the properties o f these pulses have been extensively studied at an early stage o f the multiwire c h a m b e r s ' development. It was shown that one can d e t e r m i n e the o r t h o g o n a l
In m o s t a p p l i c a t i o n s o f p r o p o r t i o n a l chambers, especially in high-energy physics, separate c h a m b e r s are used for m e a s u r i n g different coordinates. F o r some applications, such as for X - r a y or neutron m a p p i n g s , this is a l m o s t impossible a n d it is necessary to r e a d out two c o o r d i n a t e s f r o m a single-gap chamber1). In general one c o o r d i n a t e is o b t a i n e d by recording the pulses from the a n o d e wires a r o u n d which avalanches have grown. Several m e t h o d s have been i m a g i n e d for o b t a i n i n g the position o f an a v a l a n c h e a l o n g a wire. One type o f a p p r o a c h relies on the finite i m p e d a n c e o f the a m p l i f y i n g wire; either the a m o u n t o f charge o r the rise-time o f the pulse flowing at one end o f the wire is p o s i t i o n - d e p e n d e n t . The best accuracy o b t a i n e d in small c h a m b e r s by the rise-time m e t h o d 2) has been 150/lm, using very high resistivity a n o d e wires. A n i m p o r t a n t field o f a p p l i c a t i o n s for X - r a y c r y s t a l l o g r a p h y seems to be o p e n for c h a m b e r s with such accuracies, even if they are limited to one direction only, the accuracy in the o r t h o g o n a l direction being given by the wire spacing. A n o t h e r type o f a p p r o a c h , d e v e l o p e d by the PerezM e n d e z group3'4), is based on external delay lines c o u p l e d to c a t h o d e wires, o r t h o g o n a l to the sense wires. These are c a p a b l e o f locating the centre o f gravity o f the induced pulse with an accuracy o f 150/~m. C h a m b e r s for X - r a y m a p p i n g s have been successfully built by this method. I n this article we p r o p o s e a n o t h e r m e t h o d , which leads to the same range o f accuracies a n d m a y be preferred in some cases. W e also discuss the p r o b l e m o f accurate m e a s u r e m e n t s for large-size c h a m b e r s .
2. Properties of the induced pulses W h e n an a v a l a n c h e occurs a r o u n d a wire, positive 381
/ 1/o
Pulse height /
\\ o~.~
----
group of 10wwes FWHM=16cm single wwe FWHM:llcrn
Ililt/
I
/ / "\ l///
I,
10
\\
~---O o I 11
112
I 13
I 14
80
60
40
20 b) 11
12
13
1/* (cm)
Source position
Fig. 1. Variation of the height of the induced pulses as a function of source position: (a) Distance from the central wire to sensing wires = 2 mm. Curve l: pulse on a single wire. Curve If: pulse on a group of 10 wires connected together. (b) Distance from central wire to sensing wire = 7 mm. (From ref. 1.)
382
G. C H A R P A K
coordinate by using, for the cathode, wires whose directions are perpendicular to the sense wires1). In contrast to the situation existing for the anode wires, the pulse is not localized at the wire facing the avalanche but is spread over large distances. Fig. 1 shows the pulse-height distribution of the induced pulses as a function of the distance from the avalanche (from ref. 1) under very different conditions: gaps of 2 m m and 7 m m ; pulse in a single wire or in groups of wires. The striking fact is the extension of the induced pulses over one or a few centimetres. Similar observations have been made by Fisher and PlchS), who have studied the localization of the induced pulses on wide bands of 5 cm width, and have shown that simple calculations agree with the observations. In order to obtain the accurate position of an avalanche, it is necessary to measure the centre of gravity of thi's distribution. We will show that the shape of the distribution is so sensitive to the avalanche position that it is sufficient to measure the ratio of the pulses induced on two adjacent pick-up electrodes to get the centre of gravity of the avalanche with a high accuracy.
A N D F. S A U L I
The wires of one cathode plane were parallel to the sense wires, and those of the other cathode plane were orthogonal to them. The gas filling was argon + + isobutane + methylal (75%, 21%, 4%), and the chamber operated at a high voltage between 3 and 4 kV. For all measurements we used a well-collimated 55Fe source emitting 5.9 keV X-rays. Sets of several consecutive wires on the cathode plane were connected together and used as pick-up electrodes for the induced pulses. We have tried several different dimensions and distances for the strips, but the best results were obtained with strips of 4 mm, 8 mm apart. The positive pulses were amplified by a simple fast amplifier (Rin -~ 2 k~?), and the ratio of their pulse height was obtained in a fast divider.* The big advantage of operating the current division on the cathode electrodes is that high-impedance amplifiers could be used. The intrinsic noise of the system was measured, connecting the same pick-up electrode to the two * Developed at C E R N by J. Olsfors. F! 35+
1
This method was experimented on in a chamber having the following characteristics: chamber gap: anode wire diameter: anode wire spacing: cathode wire diameter: cathode wire spacing:
8 mm, 30/~m, 3 mm, 100/tin, 1 mm.
./
i
3. Localization by the ratio of induced pulses
/ /
/ /
2.5.
[13 "-. ,,q:
2.0'
/
b
8mm
/
//
,/ /
,0~E
•
,c r
, s~
'
/7 /
0 '
I0"
/ ~0.5
"'o "o
0 %
%
c
~
o
05-A/B
J 0
5
I0
mm
Fig. 2. Localization by the ratio of induced pulses. Pick-up strips of 4 mm, 8 m m apart, C h a m b e r gap L = 8 mm. Collimated source o f SSFe, separated by a step o f 2 m m along the s e n s e wires in the region between the two strips.
Fig. 3. Variation of the ratio of induced pulses. Full curve: ratio as a function of the avalanches along the sense wires; the error bars represent the fwhm o f the peaks. Dashed curve: the equivalent standard deviation as a function of position. The width of the source ( ~ 0 . 1 mm) and the noise ( ~ 3 0 % o f the peak width) are not subtracted.
MULTIWlRE
PROPORTIONAL
amplifiers; it was always about 30% of the measured peak's width (see next section) and was not subtracted from the data. 4. Space resolution 4.1. STRIPS ORTHOGONALTO THE SENSE WIRES Fig. 2 shows the separation of the ratios of the pulse heights as the source, well-collimated, is displaced by a step of 2 m m in the region between two neighbouring strips. The fwhm of the distribution corresponds to an accuracy of ~r = 200/~m. In fig. 3 the results of a complete scanning are shown. The ratio A/B of the pulse height on the two strips, whose centres lie 8 m m apart, is given on the left scale together with the fwhm of the peaks. On the right scale, the equivalent standard deviation in millimetres is presented. The width of the source (,-~ 100/~m) and the noise (30%) have not been subtracted. 4.2.
383
CHAMBERS
wire in a uniform way. Walenta et al.6), have observed that a right-left effect exists in the time of the appearance of an induced pulse in the adjacent wires; notice that e difference in pulse height could simulate a difference in time when threshold circuits are used. The accuracy in position determination is such that one could clearly separate two wires only 0.5 m m apart by current division on the induced pulses, and perhaps interpolate between them. 5. Two-dimensional read-out of chambers 5.1. IN SMALL CHAMBERS For small-size chambers of the order of 10 cm, one can adopt the following read-out scheme. The
0
•
STRIPS PARALLEL TO THE WIRES
We observe two phenomena: (a) The sense wires are clearly separated from one another. Fig. 4 shows the ratio A/B provided by two pick-up electrodes, 8 m m apart, when the source was collimated on two anode wires 3 mm apart. The resolution corresponds to about 150/,m. (b) There is a clear right-left effect, depending on the source position in respect to the anode wire. Fig. 5a shows the measured ratio A/B for two source positions, 1 m m at the left and 1 m m at the right of the same anode wire; the results of a complete scanning are shown in fig. 5b. This proves that the avalanches do not surround the
,•,.
•°
..
,,°,,%°•.,.
"...~..,•..•
.,,
.....
(a)
/" I
I
,~
I0
~
9 ~
O ~
09r
~ "
08L_
~
l
©
_
1
I
(b) m ~ f
. . . . .
8.O
~ 0 ~ t O ~ 0 Q m Q ~ Q I ~ q ~ Q ~
~ . . . . . .
~
lll~Q~
. . . . . . . .
Fig. 4. L o c a l i z a t i o n o f the sense wires by the ratio of induced pulses. R a t i o of the pulses i n d u c e d on two strips 8 m m apart, parallel to the sense wires. Wire distance: 3 mm.
Fig. 5. L o c a l i z a t i o n o f the r i g h t - l e f t side o f an avalanche development. W i t h the two pick-up strips on each side o f a sense wire the ratio is sensitive to the position with respect to the wire. (a) R a t i o o f the pick-up pulses f o r the two positions o f the source at 1 m m f r o m the wire. (b) R a t i o as a function o f the distance to t h e wire•
384
G. C H A R P A K
A N D F. S A U L I
cathodes are made of two layers of strips orthogonal to each other, the strips being 5 mm wide and separated by a 5 mm spacing. The pulse height in each of the 20 strips can be digitized to a reasonable accuracy, say a few percent; the central wires can be connected together and give the pulse gating the system. The position of an ionizing event is given roughly by the Anode
centre of the distribution, while the ratio of the pulse heights in two adjacent strips--or better, a best fit to the distribution--gives the accurate position 7). The attractive feature of the method is that the high accuracy is obtained with a very inaccurate measurement of the pulse heights. In this respect it is easier to operate than any current division method.
wires
5.2. I N LARGE CHAMBERS
Recently we have discussed the properties of drift chambers providing one coordinate with a very high accuracy (of the order of 100 pm) 8). [t is clear that a method giving the localization of the avalanche along the wires has the considerable advantage of avoiding the ambiguities arising when many particles are detected and two independent orthogonal coordinates are measured. Current division was discussed as a possible solution. For fast measurements required by high rates, the current division has to be done on wires with a low resistance. Accuracies of the order of 3 mm have been obtained
Pickup Electrodes
A
B
Fig. 6. Principle of a vernier m e t h o d for high-accuracy measu rements along a wire.
1,5
! /B
•
8m
r.
t,0 <:~
v
E E
b
1,0 ??-
,
,
~0.5
,,o "
L 0
¢
-¢
[
~-
5
I0 m
I
o
15
J
m
Fig. 7. R a t i o of the pulses induced in groups of alternate strips. Full curve: ratio as a function of the a va l a nc he s a l ong the sense wires; the error bars represent the f w h m of the peaks. D a s h e d curve: the equivalent s t a n d a r d deviation as a function of position. The width of the source ( ~ 0 . 1 mm) and the noise ( ~ 30% of the p e a k width) are not subtracted.
MULTIWIRE
PROPORTIONAL
by us and by other authors under such conditions for wire lengths of the order of 30 cm. For lengths of the order of metres the accuracy would be about a centimetre. The method we describe offers the possibility of a vernier method (fig. 6). Suppose that in a chamber such as the one described, the pick-up strips orthogonal to the anode wires are connected together alternately and provide the pulse heights A and B. The ratio A/B gives then, as before, the fine position between two strips, and the normal current division on the anode wire provides the gross coordinate. Fig. 7 shows the results of the measurement for a repetitive structure such as the one described. Here the pick-up electrodes were groups of four cathode wires, 8 m m apart, connected alternately to lines A and B. The accuracy is of course rather poor when the source is just in front of a strip, and improves to about 200/lm in the centre of the separation. As before, noise and source width were not subtracted. In a drift chamber of the type we have proposed a) it is not possible to have cathode wires perpendicular to the anode wires. However, one could imagine a long, thin, metallized strip facing the anode wires repeating the alternative pick-up electrodes structure we have described. 6. Conclusion
This report describes a method of allowing the
CHAMBERS
385
localization, with a very high accuracy, of the position of avalanches in an MWPC. The accuracy is of the order of ~ = 150 p m and has so far been shown to be limited only by the electronics. For small chambers of the order of 10 cm, as required for applications as X-ray crystallography, moderately simple systems based on our method can be envisaged. Although at this stage it does not pretend to be more accurate or more simple than the other existing methods, it may be preferable in some cases--for instance when non-delayed analogue information is needed--giving the position of a particle with the highest precision.
References 1) G. Charpak, D. R a h m and H. Steiner, Nucl. Instr. and Meth. 80 (1970) 13. 2) C. J. Borkowski and M. K. Kopp, IEEE Trans. Nucl. Sci. 17 (1970) No. 3, 340. 3) R. Grove, K. Lee, V. Perez-Mendez and J. Sperinde, Nucl. Instr. and Meth. 89 (1970) 257. 4) R. Grove, V. Perez-Mendez and J. Sperinde, Nucl. Instr. and Meth. 106 (1973) 407. 5) G. Fischer and J. Plch, Nucl. Instr. and Meth. 100 (1972) 515. 6) A. H. Walenta, J. Heintze and B. Schurlein, Nucl. Instr. and Meth. 92 (1971) 373. 7) G. Charpak, A. Jeavons, F. Sauli and R. Stubbs, Highaccuracy measurements o f the center of gravity o f avalanches in proportional chambers, C E R N 73-11 (Sept. 1973). 8) G. Charpak, F. Sauli and W. Duinker, High-accuracy drift chambers and their use in strong magnetic fields. Submitted to Nucl. Instr. and Meth.