Studies with a TPC
and development of a procedure
to measure ion feedback
James Inman – Radford University
Dan Peterson - Cornell University
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
1 of 20
Equipment, Amplifiers and Electronics
High voltage system:
-20 kV module, 2 channels
-2 kV module, 4 channels
Wires
(Multi-Wire-Proportional-Chamber)
Micromegas
Readout:
VME crate
Struck FADC
GEM
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
2 of 20
Sample data taken with Cornell TPC
ArCO2 (10%) , 300V/cm
Gas gain: ~ 300
Micromegas event – raw (no smoothing)
tracks
25 MHz , 40 ns
2048 time buckets (81.92 ms)
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
3 of 20
Implementing Program in Visual C++
track
Design Goals:
• Maintain functionality
of old program
• Allow for analysis of
ion feedback
• Interactive control
• Optimize run speed
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
4 of 20
TPC Data Analysis Program – Block Diagram
Form1 - GUI
Initialize event class
Load channel pad configuration data file
Load analysis configuration data file
Get TPC data
file name
Create ROOT File?
or
Open data-file-list file,
load list of TPC data file names
Initialize ROOT components
Open TPC data file
Load event from
TPC data file
Create ROOT file
Pulses & Event
Analyze event
Fill ROOT tree
Tracks
Ion Feedback
Plot event
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
5 of 20
TPC Data Analysis Program - Data Files
Channel Pad Configuration
•
Contains information about the layout of the
TPC (# rows, #pads, …)
•
Maps the channels of data in the TPC data
files to the pad board layout
(channel# -> (row, pad))
•
Contains the geometry of the padboard, used
for track fitting and plotting
MAP!
N
Analyze Configuration
•
Has 3 sets of 35 input parameters that are
used to analyze data
•
Data smoothing (smoothwidth)
•
Pulse identity (thresh, recoverwidth)
•
Track fitting (pad pulse sharing)
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
6 of 20
TPC Data Analysis Program - ROOT
Create an ntuple tree
• A tree stores data
• TTree object consists of branches
that can hold any type of data from
integers to classes
• Branches are filled for each event
Create ROOT file
• Contains ROOT objects
(in particular my TTree)
• ROOT is run in “interactive”
mode (root.exe) and file is
opened to make cuts on the
data and then plot histograms
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
7 of 20
Ion Feedback Measurement
Positive ions are created in the amplification
and
drift back into the field cage.
Ion feedback is expected to be
suppressed with the
GEM or Micromegas devices
relative to MWPC.
We measured ion feedback
on the field cage termination plane,
for individual tracks.
This is a different method than used by other groups
and will allow
a more direct comparative measurement
of ion feedback in the various devices.
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
8 of 20
Ion Feedback Measurement
Wire Amplifier
Wire amplifier produces large feedback signal
this is why it is unsuitable for use in the ILC
It is used as a starting point for measuring ion
feedback
Field cage termination plane
Located between field cage
and amplifier (next slide)
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
9 of 20
Field Cage Termination
Single Field Cage Termination
The field cage termination provides a close
to uniform potential plane while still being
transparent
It would not collect ion signal; ions drift
past the plane into the field cage
Double Field Cage Termination
Allows for an area to be defined with a
reverse electric field
This reduces transmission of elections from
track and allows for ions drifting back into
the field cage to be collected on wires
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
10 of 20
Partial Transmission Mode
Transmission coefficient
of electrons through
double field cage
termination wires using
Micromegas
Transmission Coefficient
160
Single Grid
(maximum
transmission)
120
100
40
20
0 V/cm
60
+300 V/cm
80
-300 V/cm
Pulse Height (counts)
140
0
0
100
200
300
400
500
~40% transmission at
bias voltage of -450 V
~60% of the ion
feedback should be
detected by the field
cage termination wires
Voltage
(-Volt)
Bias
Voltage
(-V)
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
11 of 20
40% Transmission Mode, 5mm Drift
Pulse observed on field
cage termination, is this
ion feedback?
X-axis is in units of data buckets
Top 6 traces are from pad board, data collected at 20ns/bucket (82 µs)
Bottom traces are from FCT, data collected at 320ns/bucket (655 µs)
Logic pulses identify pulse threshold crossings
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
12 of 20
40% Transmission Mode, 5mm Drift
14
Average: 5.5%
12
Average pulse delay time
is 202 µs with σ = 8.7 µs
Average relative pulse
height is 6.0%
# Events
10
8
6
4
Initial Results
calculated by
hand, not fun
2
0
1
2
3
4
5
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
6
7
8
9
10
11
12
13
14
Relative Pulse Height (%)
13 of 20
“Full” Transmission Mode, 5mm Drift
Pulse still observed on field cage termination
By observation pulse looks smaller
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
14 of 20
“Full” Transmission Mode, 5mm Drift
Average pulse delay time
is 207 µs with σ = 9.3 µs
Average relative pulse
height is 4.2%
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
15 of 20
“>Full” Transmission Mode, 5mm Drift
Pulse still observed on field cage
termination; decreased signal to noise
ratio makes pattern-recognition difficult
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
16 of 20
“>Full” Transmission Mode, 5mm Drift
Average delay time is 216 µs with σ = 21 µs
Average relative pulse height is 2.6% (reduced by 43%)
Pulse Height is effectively controlled by bias voltage
applied to field cage termination
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
17 of 20
40% Transmission Mode, 7mm Drift
Change the ion drift distance to see if the pulse
changes accordingly
Pulse observed on field cage termination with a
noticeably later time
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
18 of 20
40% Transmission Mode, 7mm Drift
Average pulse delay time is 245 µs with σ = 7.9 µs (5mm – 202 µs)
Pulse delay time increase of 21% (40% drift distance increase)
Average relative pulse height is 7.4% (5mm – 6.0%)
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
19 of 20
Results / Next
Created a new display and analysis program in Microsoft C++ that can be used
to analyze ion feedback
Ion feedback was measured from a wire amplifier which has a large gain (x300)
An attempt to measure ion feedback from the double layer GEM amplifier was
made; however, the GEM has an expected gain that is 10% of the wire amplifier
and only 5% of the ion feedback
Plan to make a pulsing voltage bias on the field cage termination to allow full
transmission of electrons and then pulse field to collect ions, need pulsing circuit
and electronic amplifiers that can withstand pulsing bias
Smaller diameter wires on the field cage termination would increase field near
wire increasing feedback signal (20 µm -> 8 µm)
A triple layer GEM would provide much more signal amplification
Make more measurements of ion feedback ;
measure ion feedback for various amplifier devices.
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
20 of 20
Acknowledgements
I would like to thank Dan Peterson of Cornell University for
being my mentor and guiding me through my research expeirence.
Thanks to Tarek Anous for his encouragement, programming
assistance, and daily entertainment. Thanks to Chris Macklin for
getting me started with ROOT and to LEPP Computing. Thanks to all
the other REU students for adding plenty of excitement and support.
And much thanks to Rich Galik for his dedication to the REU program
and genuine interest in the success of all the REU students.
J. Inman, “Studies with a TPC … measure ion feedback” , REU presentation, 08-Aug-2006
21 of 20