- Email: [email protected]
Tunnel diode pulse shape discriminator
NUCLEAR
INSTRUMENTS
AND
METHODS
50
(I967) 7~-76; ©
NORTH-HOLLAND
PUBLISHING
CO.
T U N N E L D I O D E P U L S E S H A P E DISCRIMINATOR* B. S O U ~ E K * * and R.L. C H A S E
Brookhaven National Laboratory, Upton, N. Y. Received 19 October 1966 A pulse shape discriminator Is described which is suitable for n e u t r o n - g a m m a ray separation with several d~fferent detector types It uses three tunnel diode dlscrlmmator clrcuUs, one for setting an energy threshold, one for marking the leading edge of the input signals and one for marking their zero-crossover
The signal shape Js characterized by the time from inception to zero-crossover, which ~s converted to a proportional output pulse amphtude Measured szgnal width variations o f 5 : 2 nsec are observed with input signals varying from 50 mV to 5 V.
1. Introduction Scintillation counters have been used for a long time for measuring the specific ionization of exciting particles through variations in signal pulse shape. A considerable number of circuits have been developed for distinguishing particles of different specific ionization~'~2). Many of them employ the zero-crossing technique, used to provide a time invariant fiduoal point from a variable amplitude pulse of a constant shape. In the nanosecond time spectrum at least two possible sources of timing error are present. First, under overload conditions, the input pulse will be distorted by the photomultipher or the discriminator circuitry, causing the output pulse to shift in time. Secondly, if a photomultiplier is used as the event detector, the trailing edge of the signal is not of constant shape, due to statistical processes in the tube, and this wdl shift the zero-crossing point in time. The pulse shape fluctuations depend on the energy of particles detected, and hence on the number of photoelectrons emitted by the photo-cathode. The fluctuation is small at high energies, increasing gradually with decreasing energy. The present instrument employs cwcuit techmques similar to those used in fast coincidence systems12"18). It incorporates three tunnel diode discriminators in coincidence. The first discriminator is used to reject pulses of very small amplitude which could not be measured accurately. The second discriminator fires on the fast leading edge of accepted pulses. The third discriminator fires at the zero-crossing point of pulses which are previously integrated and differentiated to diminish statistical fluctuations. Discriminator outputs are connected to a simple nanosecond time-to-amplitude converter.
2. Circuit description Fig. 1 shows the circuit diagram of the discriminator. The discriminator has three channels, connected to the input through common base stages T Z, T4 and Tlo. Channel I is an amplitude discriminator. If the input pulse is larger than the triggering level preset with potentiometer P1, tunnel diode monostable D1 fires and removes biases from tunnel diodes D2 and D3. The period of the monostable is controlled by the value of the small inductor LI. Channel lI receives a delayed input signal, which triggers tunnel diode monostable D2, determining the leading edge of the output. Channel III integrates and differentiates the signal in a filtering network between transistors T~ (high output impedance) and Ta (high input impedance). The zero-crossing point of the signal triggers tunnel diode monostable D3, determining the trailing edge of the output. Such an arrangement has the following advantages: diodes D2 and D3 are quiescently biased into a noise insensitive state. They are unbiased to the sensitive state, very near to the peak current point, only a few nsec before receiving signals whose timing information is to be extracted. The current in the diode D3, after unbiasing, can be equal to the diode peak current, or even slightly higher, since it is compensated by a negative part of the signal before zero-crossover point. Such biasing enables high timing accuracy. Signals from the three channels are amplified and connected to the simple time-to-amplitude converter. When tunnel diodes D2 and D3 fire, the bases of transistors T a and T~ ~ receive an immediate flood of 5 mA, allowing the transistors to saturate fast. Fast pulses of 6 V amplitude are connected to a diode "'and" gate which controls the flow of a current from constant current source T, 3 into condenser C 3, changing time
* This work was performed under the auspices of the U.S. Atormc Energy C o m m i s s m n . ** On leave from lnstitut Ruder Boskovic, Zagreb, Jugoslavia.
71
72
a. S O U C E K A N D R.L. CHASE
m~r o~
u ~ m ~_ oo~_
:E
z ~.~tn
~
>
+
L
_ II
m
c,'J t::: 7-
",,^-", "~-- ->
o
o ~
o:
~- I ~ , ,
L_il ,
~V ~''
Ft"
E A
~JD
@-~---
~ ~:" ,
~II
:> ~J
o:
O_
~-~: ~,, i~
+
+
0 ,,._:
~o
. . L :L. ~:
o I~11
i
x
P
! °'~" j,, ,,~.~~-~ J
TUNNEL
DIODE
PULSE
SHAPE
73
DISCRIMINATOR
b i
C
d
e
b Fig 3.
Measured variation o f pulse width with amplitude
a. the signal anaphtude (ordinate was varied from 0.5 to 5 V); b. from 50 m V to 0 5 V. T h e h n e s correspond (from left to right) to pulse widths o f 80, 110, 120, 150 a n d 180 nsec
0 Fig. 2 (conttnued) Fig 2 Pulse shapes observed when triggered by a n a n o s e c o n d pulse generator T i m e scale 50 nsec/dlv. a. input to channel 1; b. o u t p u t from channel I;
c d. e. f. g.
i n p u t to channel 11; o u p u t from channel II; integrated and d~fferentlated signal at the base o f T 2 ; o u t p u t f r o m channel Ill, o u t p u t o f t h e time-to-amplitude converter.
74
B. S O U ~ E K A N D R.L. CHASE
information into amplitude information. The main pulse shapes and their time relationships can be seen in fig. 2. 3. T i m i n g a c c u r a c y
The relative time "walk'" of discriminators I1 and IlI was measured by observing the variatton of the output
of
the
time-to-helght
converter
wtth
input
pulse amphtude. The output of a fast rising pulse generator,
simulating
a photomultiplier
signal,
was
connected to the Y-input of a two-parameter pulse height height
analyzer, while converter was
the o u t p u t o f t h e t i m e - t o c o n n e c t e d to t h e X - i n p u t .
iii!! :!!:
b F~g 4 Separation of neutrons and gamma rays using a stllbene scintillation counter The t~o-parameter analyzer displays show the distribution of events classified by pulse width (abscissa) and pulse height (ordinate). a. was obtained w~th m~xed neutrons and gamma rays from Pu-Be; b. was obtained with gamma rays from Z2Na.
Fig 5 Separation of gamma rays and thermal neutrons m boron glass beads mixed with plastic scintillator Unblankmg levels a. 1000counts, b 4000counts; c 8000counts.
TUNNEL
DIODE
PULSE
SHAPE
DISCRIMINATOR
75
Fig. 3 shows the a n a l y z e r d i s p l a y as the pulse g e n e r a t o r a m p h t u d e was varied over a 100 to 1 range. Straight vertical lines would have indicated no walk. Over most o f the range the timing error ~ a s observed to be less than -4- 2 nsec.
4. Experiments The d i s c r i m i n a t o r was intended for d l s t m g m s h m g neutrons from g a m m a rays, using three kinds o f detectors" stdbene scintdlators, b o r o n glass beads mixed in a plastic scintdlator, a n d p r o p o r t i o n a l counters~9"2°)_ Results o f a few experiments are shown in figs. 4, 5 a n d 6. Fig. 4 shows the separation o f neutrons and g a m m a rays using stllbene and fast p h o t o m u l t t p h e r R C A 7264 (anode pulse rose time 3 nsec), at 1800 V a n o d e voltage. R a d i a t i o n sources used were Pu-Be (fig. 4a) a n d 22Na (fig. 4b). Fig. 5 shows s e p a r a t i o n o f g a m m a rays and thermal neutrons, using a scintillator c o m p o s e d o f b o r o n glass beads mixed into plastic. Design and o p e r a t i o n o f such a scintillator will be described elsewhere2'). The p h o t o m u l t l p h e r used was a 5 6 A V P (anode rose t i m e 2 n s e c ) at 2 1 0 0 V . The r a d m t l o n source was Pu-Be. Fig. 6 shows the s e p a r a t i o n o f g a m m a rays a n d neutrons using a h y d r o g e n filled p r o p o r t i o n a l counter. Pulses f r o m the c o u n t e r were amphfied m a chargesensitive, low noise, fast F E T preamplifier. The r a d i a t i o n source was Pu-Be. F o r the e x p e r i m e n t s with the p r o p o r t i o n a l counter, t i m i n g elements in the d i s c r i m m a t o r were increased to m a t c h the longer rise times from the counter. It is a pleasure to a c k n o w l e d g e the help recewed f r o m W . A . H i g i n b o t h a m in designing the c i r c m t r y and from K. Jones, D . L . Price and C. Sastre m a r r a n g i n g the experiments.
Fig 6 Separation of neutrons and gamma rays due to the different Jomzatlon track lengths produced m a proportional counter
a is a one-parameter frequency dlstmbutlon of pulse widths measured with the shape discriminator, using a gamma ra} source; b. is a s i m i l a r d i s t r i b u t i o n u s i n g a m i x e d n e u t r o n - g a m m a r a ) source ; c ~s a two-parameter plot obtained with the m~xed source. The vertical group to the left is produced by neutrons and the homzontal smear by gamma rays
76
B. S O U ( ~ E K A N D R.L.
CHASE
References 1) F . D Brooks, Nucl lnstr and Meth 4(1959) 151 2} R B. Owen, Nucleonics 17 (1959) 92 al G C Nellson, W K D a w s o n and F . A J o h n s o n , Rex' Scl lnstr 30 {19591 963 4) R Batchelor, W B Gflbo3, A . D . Purnell and J H. Towle, Nucl. lnstr and Meth 8(19601 146 5) M Forte, A K o n s t a and C M a r a n z a n a , Proc. Nucl Eleetr C o n f {Belgrade) 2 (19611 277. 6) j p Crettez, F C a m b o n and G A m b r o s l n o , ~b~d 207 7) L G d e V r l e s a n d F Udo, lbld 305 8) E G a i n and F de M a t u r e , lb~d 265. 9) M L. R o u s h , M A Wilson and W F H o r n y a k , Nucl. Instr. and Meth. 31 (1964) 112. 10) R. Fulle, G y Math6 and D N e t z b a n d , Nucl lnstr and Meth 35 (1965) 250.
11) G y Math6, Nucl Instr. a n d Meth 39 (1966) 356
12) E N a d a v and B K a u f m a n , Nucl Instr and Meth. 33 (19651 289 1.3) p R. O r m a n , Nucl Instr. and Meth 21 (1963) 121 14J H Verwekl, Nucl Instr and Meth 41 (1966) 181 15) j. G r u n b e r g and L Tepper, IEEE T r a n s Nuel Scl NS-13, no I (19661 389 x61 B. M Shoffner a n d E F. Sharder, 1bid 394 17} j K. Whittaker, ib~d 399 1~) D L Wleber and H W Lefevre, lbld 406 191 E . F Bennett, ANL-6897(19641 20) A S a y r e s a n d M Coppola, Rev Scl Instr 35(19641 431. -'t) H Palevsky, R J Sutter, H . R M e u t h e r and D k Price, to be p u b h s h e d
INSTRUMENTS
AND
METHODS
50
(I967) 7~-76; ©
NORTH-HOLLAND
PUBLISHING
CO.
T U N N E L D I O D E P U L S E S H A P E DISCRIMINATOR* B. S O U ~ E K * * and R.L. C H A S E
Brookhaven National Laboratory, Upton, N. Y. Received 19 October 1966 A pulse shape discriminator Is described which is suitable for n e u t r o n - g a m m a ray separation with several d~fferent detector types It uses three tunnel diode dlscrlmmator clrcuUs, one for setting an energy threshold, one for marking the leading edge of the input signals and one for marking their zero-crossover
The signal shape Js characterized by the time from inception to zero-crossover, which ~s converted to a proportional output pulse amphtude Measured szgnal width variations o f 5 : 2 nsec are observed with input signals varying from 50 mV to 5 V.
1. Introduction Scintillation counters have been used for a long time for measuring the specific ionization of exciting particles through variations in signal pulse shape. A considerable number of circuits have been developed for distinguishing particles of different specific ionization~'~2). Many of them employ the zero-crossing technique, used to provide a time invariant fiduoal point from a variable amplitude pulse of a constant shape. In the nanosecond time spectrum at least two possible sources of timing error are present. First, under overload conditions, the input pulse will be distorted by the photomultipher or the discriminator circuitry, causing the output pulse to shift in time. Secondly, if a photomultiplier is used as the event detector, the trailing edge of the signal is not of constant shape, due to statistical processes in the tube, and this wdl shift the zero-crossing point in time. The pulse shape fluctuations depend on the energy of particles detected, and hence on the number of photoelectrons emitted by the photo-cathode. The fluctuation is small at high energies, increasing gradually with decreasing energy. The present instrument employs cwcuit techmques similar to those used in fast coincidence systems12"18). It incorporates three tunnel diode discriminators in coincidence. The first discriminator is used to reject pulses of very small amplitude which could not be measured accurately. The second discriminator fires on the fast leading edge of accepted pulses. The third discriminator fires at the zero-crossing point of pulses which are previously integrated and differentiated to diminish statistical fluctuations. Discriminator outputs are connected to a simple nanosecond time-to-amplitude converter.
2. Circuit description Fig. 1 shows the circuit diagram of the discriminator. The discriminator has three channels, connected to the input through common base stages T Z, T4 and Tlo. Channel I is an amplitude discriminator. If the input pulse is larger than the triggering level preset with potentiometer P1, tunnel diode monostable D1 fires and removes biases from tunnel diodes D2 and D3. The period of the monostable is controlled by the value of the small inductor LI. Channel lI receives a delayed input signal, which triggers tunnel diode monostable D2, determining the leading edge of the output. Channel III integrates and differentiates the signal in a filtering network between transistors T~ (high output impedance) and Ta (high input impedance). The zero-crossing point of the signal triggers tunnel diode monostable D3, determining the trailing edge of the output. Such an arrangement has the following advantages: diodes D2 and D3 are quiescently biased into a noise insensitive state. They are unbiased to the sensitive state, very near to the peak current point, only a few nsec before receiving signals whose timing information is to be extracted. The current in the diode D3, after unbiasing, can be equal to the diode peak current, or even slightly higher, since it is compensated by a negative part of the signal before zero-crossover point. Such biasing enables high timing accuracy. Signals from the three channels are amplified and connected to the simple time-to-amplitude converter. When tunnel diodes D2 and D3 fire, the bases of transistors T a and T~ ~ receive an immediate flood of 5 mA, allowing the transistors to saturate fast. Fast pulses of 6 V amplitude are connected to a diode "'and" gate which controls the flow of a current from constant current source T, 3 into condenser C 3, changing time
* This work was performed under the auspices of the U.S. Atormc Energy C o m m i s s m n . ** On leave from lnstitut Ruder Boskovic, Zagreb, Jugoslavia.
71
72
a. S O U C E K A N D R.L. CHASE
m~r o~
u ~ m ~_ oo~_
:E
z ~.~tn
~
>
+
L
_ II
m
c,'J t::: 7-
",,^-", "~-- ->
o
o ~
o:
~- I ~ , ,
L_il ,
~V ~''
Ft"
E A
~JD
@-~---
~ ~:" ,
~II
:> ~J
o:
O_
~-~: ~,, i~
+
+
0 ,,._:
~o
. . L :L. ~:
o I~11
i
x
P
! °'~" j,, ,,~.~~-~ J
TUNNEL
DIODE
PULSE
SHAPE
73
DISCRIMINATOR
b i
C
d
e
b Fig 3.
Measured variation o f pulse width with amplitude
a. the signal anaphtude (ordinate was varied from 0.5 to 5 V); b. from 50 m V to 0 5 V. T h e h n e s correspond (from left to right) to pulse widths o f 80, 110, 120, 150 a n d 180 nsec
0 Fig. 2 (conttnued) Fig 2 Pulse shapes observed when triggered by a n a n o s e c o n d pulse generator T i m e scale 50 nsec/dlv. a. input to channel 1; b. o u t p u t from channel I;
c d. e. f. g.
i n p u t to channel 11; o u p u t from channel II; integrated and d~fferentlated signal at the base o f T 2 ; o u t p u t f r o m channel Ill, o u t p u t o f t h e time-to-amplitude converter.
74
B. S O U ~ E K A N D R.L. CHASE
information into amplitude information. The main pulse shapes and their time relationships can be seen in fig. 2. 3. T i m i n g a c c u r a c y
The relative time "walk'" of discriminators I1 and IlI was measured by observing the variatton of the output
of
the
time-to-helght
converter
wtth
input
pulse amphtude. The output of a fast rising pulse generator,
simulating
a photomultiplier
signal,
was
connected to the Y-input of a two-parameter pulse height height
analyzer, while converter was
the o u t p u t o f t h e t i m e - t o c o n n e c t e d to t h e X - i n p u t .
iii!! :!!:
b F~g 4 Separation of neutrons and gamma rays using a stllbene scintillation counter The t~o-parameter analyzer displays show the distribution of events classified by pulse width (abscissa) and pulse height (ordinate). a. was obtained w~th m~xed neutrons and gamma rays from Pu-Be; b. was obtained with gamma rays from Z2Na.
Fig 5 Separation of gamma rays and thermal neutrons m boron glass beads mixed with plastic scintillator Unblankmg levels a. 1000counts, b 4000counts; c 8000counts.
TUNNEL
DIODE
PULSE
SHAPE
DISCRIMINATOR
75
Fig. 3 shows the a n a l y z e r d i s p l a y as the pulse g e n e r a t o r a m p h t u d e was varied over a 100 to 1 range. Straight vertical lines would have indicated no walk. Over most o f the range the timing error ~ a s observed to be less than -4- 2 nsec.
4. Experiments The d i s c r i m i n a t o r was intended for d l s t m g m s h m g neutrons from g a m m a rays, using three kinds o f detectors" stdbene scintdlators, b o r o n glass beads mixed in a plastic scintdlator, a n d p r o p o r t i o n a l counters~9"2°)_ Results o f a few experiments are shown in figs. 4, 5 a n d 6. Fig. 4 shows the separation o f neutrons and g a m m a rays using stllbene and fast p h o t o m u l t t p h e r R C A 7264 (anode pulse rose time 3 nsec), at 1800 V a n o d e voltage. R a d i a t i o n sources used were Pu-Be (fig. 4a) a n d 22Na (fig. 4b). Fig. 5 shows s e p a r a t i o n o f g a m m a rays and thermal neutrons, using a scintillator c o m p o s e d o f b o r o n glass beads mixed into plastic. Design and o p e r a t i o n o f such a scintillator will be described elsewhere2'). The p h o t o m u l t l p h e r used was a 5 6 A V P (anode rose t i m e 2 n s e c ) at 2 1 0 0 V . The r a d m t l o n source was Pu-Be. Fig. 6 shows the s e p a r a t i o n o f g a m m a rays a n d neutrons using a h y d r o g e n filled p r o p o r t i o n a l counter. Pulses f r o m the c o u n t e r were amphfied m a chargesensitive, low noise, fast F E T preamplifier. The r a d i a t i o n source was Pu-Be. F o r the e x p e r i m e n t s with the p r o p o r t i o n a l counter, t i m i n g elements in the d i s c r i m m a t o r were increased to m a t c h the longer rise times from the counter. It is a pleasure to a c k n o w l e d g e the help recewed f r o m W . A . H i g i n b o t h a m in designing the c i r c m t r y and from K. Jones, D . L . Price and C. Sastre m a r r a n g i n g the experiments.
Fig 6 Separation of neutrons and gamma rays due to the different Jomzatlon track lengths produced m a proportional counter
a is a one-parameter frequency dlstmbutlon of pulse widths measured with the shape discriminator, using a gamma ra} source; b. is a s i m i l a r d i s t r i b u t i o n u s i n g a m i x e d n e u t r o n - g a m m a r a ) source ; c ~s a two-parameter plot obtained with the m~xed source. The vertical group to the left is produced by neutrons and the homzontal smear by gamma rays
76
B. S O U ( ~ E K A N D R.L.
CHASE
References 1) F . D Brooks, Nucl lnstr and Meth 4(1959) 151 2} R B. Owen, Nucleonics 17 (1959) 92 al G C Nellson, W K D a w s o n and F . A J o h n s o n , Rex' Scl lnstr 30 {19591 963 4) R Batchelor, W B Gflbo3, A . D . Purnell and J H. Towle, Nucl. lnstr and Meth 8(19601 146 5) M Forte, A K o n s t a and C M a r a n z a n a , Proc. Nucl Eleetr C o n f {Belgrade) 2 (19611 277. 6) j p Crettez, F C a m b o n and G A m b r o s l n o , ~b~d 207 7) L G d e V r l e s a n d F Udo, lbld 305 8) E G a i n and F de M a t u r e , lb~d 265. 9) M L. R o u s h , M A Wilson and W F H o r n y a k , Nucl. Instr. and Meth. 31 (1964) 112. 10) R. Fulle, G y Math6 and D N e t z b a n d , Nucl lnstr and Meth 35 (1965) 250.
11) G y Math6, Nucl Instr. a n d Meth 39 (1966) 356
12) E N a d a v and B K a u f m a n , Nucl Instr and Meth. 33 (19651 289 1.3) p R. O r m a n , Nucl Instr. and Meth 21 (1963) 121 14J H Verwekl, Nucl Instr and Meth 41 (1966) 181 15) j. G r u n b e r g and L Tepper, IEEE T r a n s Nuel Scl NS-13, no I (19661 389 x61 B. M Shoffner a n d E F. Sharder, 1bid 394 17} j K. Whittaker, ib~d 399 1~) D L Wleber and H W Lefevre, lbld 406 191 E . F Bennett, ANL-6897(19641 20) A S a y r e s a n d M Coppola, Rev Scl Instr 35(19641 431. -'t) H Palevsky, R J Sutter, H . R M e u t h e r and D k Price, to be p u b h s h e d