Transcript Signal features vs # of instrumented RO and HV lines & with

Signal features vs # of instrumented
RO and HV lines
& with w/o shielded cover plates
Ferretti, Levin
UM April 9, 2012
Signal vs Shielding panels
• Characterize the signal with these metrics:
– Amplitude
– FWHM (or Width at 10% of base later on)
– Rise time
• First test: Measure baseline panel, no
sheilding
• Then cover top plate, then bottom with thin Al
foil
Red: no shielding
Green top plate shielding
Blue: Full shielding
Red: no shielding
Green top plate shielding
Blue: Full shielding
Red: no shielding
Green top plate shielding
Blue: Full shielding
Signal vs # of readoutlines
• Characterize the signal with these metrics:
– Amplitude
– Width at 10% of base
– Rise time
• First test: instrument one HV line, crossing N
readout lines.
• A readout line is instrumented if it is connected to
ground via a ~100 ohm termination resisitor.
– If it is left floating, it is not considered instrumented
because a discharge cannot occurr
• Runs where all RO lines were masked with tape
applied directly on the PDP RO end, but 1 (then 2,
3, 4, 5, 6, 18, 24): the goal is to see if the numbers
of channels connected to ground matters to the
signal.
• HV configuration: 11 instrumented lines 100-110
with the new HV board (R=200 MOhm, plus the
circuitry to readout the HV signal, not used yet)
running at the nominal 860V for “VPC”.
• Test on the signal on one single line (#6 at the
connector = PDP line#9) when 1->N lines are
connected to the readout (GND).
• Same HV configuration, but it was needed to rise the
HV by 3 volts since we wanted runs with higher
statistics and the rate of hits seemed lower than before
• The HV lines used are the ones powered also during
the test beam (sign of degradation ?)
• The plots will show the same quantities as before:
|amplitude|, FWHM, Q=|A|*FWHM, rising edge time
and slope (10%->90%)
The amplitude and the FWHM distributions are bi-modal,
even when only one RO line is connected.
Trying to understand where this behavior could come from,
the events have divided in two classes: High=events with
|A|>330mV and Low=events with |A|<330mV. 2 lines
connected
• Now we back up, and instrument a single HV
line – then increase the # of readout lines
April 4, 2012
Started a new set of run with only one new HV line (#111 with 22
MOhm resistor on it) at 860V.
The amplitude of the signal from 1->6 lines connected is not
anymore bi-modal
If we use the symmetry between HV and RO electrodes and the previous
result that the amplitude of a pulse (capacitance of the pixel/line) depends on
the electrical configuration of the nearby lines. we can make the hypothesis
that the different pulse amplitudes seen on those runs could depend on the
position of the source respect to the HV lines hooked up. If the source is close
to an edge of the set of the HV lines, the line at the extreme has HV
instrumented lines only on one side, while the next one (as example) has one
close line with HV on each side.
With this in mind three runs have been taken with just one RO line connected
and the previous 11 HV lines (#100->110), one with the source centered on
the left edge of the HV lines, one on the middle and one on the right edge. If
the hypothesis is correct we should see double peaked distributions for the
two “extreme” runs and not for the centered one
In all of the three runs (black=center, red=left, green=right) the distribution
shows a double peak more or less evident (different weights), one around
250 mV and the other around 300 mV.
It seems that there are always two kind of pulses: in order to clarify if it is
each pixel producing these two classes of hits (something relative to the
physics of the discharge) or is a variation pixel-to-pixel since we have
many of them ORed in each RO line, I think we need the 2D readout so
that we can really discriminate between pulses coming from different
pixels and so factor out eventual different intrinsic capacitance and/or
efficiency of each pixel.