The Constant Fraction Discriminators CFD8c, CFD7x, CFD4c, CFD1c and CFD1x (11.0.1701.1)
Page 19 of 25
Figure 3.22b: Typical image artefact caused by pre-trigger events (left picture). These events may appear only on
certain parts of the detector: Around the position x = 13 mm some events seem to be “relocated” (see red arrow).
If the time sum (as in Figure 3.21b right) is plotted as function of position (middle picture) the localized
contribution of pre-trigger events is revealed. Setting a narrow time sum gate can remove the pre-trigger
events but the image artefact (missing data) remains, see right picture.
3b.5.3
Cross talk problem
A common problem of all CFD units with more than one channel in a housing is the cross-influencing of input signal timing
by output signals from other channels. This can occur if several CFD channels are simultaneous operating on input signals
above threshold. The cross talk is especially remarkable when the threshold settings are very low and when the CFD timing
output signals are very short.
Although greatest care was taken to reduce cross-influences between channels, even the
CFD8c/7x
and
CFD4c
can show a
certain amount of time cross talk between channels under some conditions. This can affect signal timing and thus the imaging
performance of a delay-line detector.
If thresholds are set to very low values, it can be of advantage to mix the delay-line outputs between channels so that signals
from the same delay-line layer (i.e. from x1 and x2 outputs) are not processed on neighbouring channels that share the same
internal board. In this sense channels 4 and 5 in case of the
CFD8c
and channels 3 and 4 in case of the
CFD7x
do not have
to be considered as “neighbours”.
3b.6
The CFDx pulse height determination option
The
CFDx
function is an add-on circuit that allows measuring the signal’s pulse height. The pulse height information is coded
as a time delay of a NIM-logic level transition with respect to the CFD timing signal from the standard CFD out socket and
can be recorded with a TDC or TAC unit. Recording pulse height information with the time and/or position of a
particle/photon can be beneficial for reducing background and improving spatial or temporal resolution of a detector.
Additionally, an external signal’s pulse height (“energy”) information can be included into the data stream in coincidence.
Two of the standard
RoentDek
CFD
modules contain a
CFDx
circuit, the
CFD1x
and the
CFD7x
, the latter having the
same form factor as the
CFD8c
modules with one CFD channel “sacrificed” to give room for a
CFDx
circuit: Effectively, the
CFD7x
has the functionality of a
CFD1x
+
CFD6c
module with 6 standard CFD channels plus one with an additional
CFDx
circuit. For the CFD channel with additional
CFDx
circuit, the pulse-height is coded into the length of the “
CFDx
signal”
(from the two sockets labelled “
CFDx
Out”) and into the delay of the “
stop
” signal’s leading edge from the likewise labelled
socket. The time delay between the leading edge of the “CFD out” (and
CFDx)
signal and the leading edge of the
stop
signal
(and the trailing transition of the
CFDx
signal) are an almost linear function of the input signal’s pulse height, assuming proper
settings of additional controls.
The
CFDx
signal can be used with a multi-stop TDC that can record both falling and leading edges of the same signal (e.g. the
RoentDek
TDC8HP
). It then allows recording timing AND pulse height by using only one TDC channel. The
stop
signal
can be used to determine the pulse height on a separate TDC channel or TAC which does not have to be able of detecting
multiple hits.
