MuPix 8 Results
Jan Hammerich, Lennart Huth for the Mu3e Collaboration
Physikalisches Institut Heidelberg
DPG Frühjahrstagung Würzburg
MuPix8
First large HV-MAPS prototype for Mu3e 19.5×10.8 mm2 80×81 µm2 pixel with 1 diode each 200×128 pixel
submatrix A
submatrix B
sub- matrix
C
digital periphery pixel matrix
bias blocks & pads 10.8 mm
19.5mm
128 columns
48 48 32
200 rows
MuPix8
3 submatrices with different line drivers
→ A with Source Follower
→ B+C with Current Driver 4 LVDS links with 1.25 Gbit/s (A,B,C and copy or Mux)
Only small sample size so far due to production issues
submatrix A
submatrix B
sub- matrix
C
digital periphery pixel matrix
bias blocks & pads 10.8 mm
19.5mm
128 columns
48 48 32
200 rows
Submatrices
Source Follower (A) like previous MuPix prototypes Current Driver (B+C) new A worked out of the box, B+C did not as good Worse timing and no good default working point
→ Focus on A, B+C postponed
−1000 −500 0 500 1000
0 1000 2000 3000 4000 5000 6000 7000 8000
9000 TS-Trigger Matrix A+B+C
TS-Trigger Matrix A TS-Trigger Matrix B+C
TS Mupix - TS Ref [8ns]
Entries
0 20 40 60 80 100 120
0 20 40 60 80 100 120 140 160 180
0 50 100 150 200 250 300 350 400
Column Address
Row Address
Hits
Setup
Sources: Fe55, Sr90, Injection (1 or more Pixel) Most measurements done with an oscilloscope
Biased by scope trigger and measurement setup
Amplifier output of 1 pixel in col 0 studied
Hitbus (OR of comparator outputs in one col) for ToT
MuPix 8 Lab Results Testbeam Results
Amplifier Output
Nice triangle pulses Fe55 (1600 e−) ≈100mV
→ Cfb ≈2.5fF
Fe55 equivalent injection voltage ≈250mV
→ Cinj ≈1fF
Peak to Peak noise≈20mV
→ SNR≈17 (assuming uniform noise)
Fe55
Amplifier Output
Nice triangle pulses Fe55 (1600 e−) ≈100mV
→ Cfb ≈2.5fF
Fe55 equivalent injection voltage ≈250mV
→ Cinj ≈1fF
Peak to Peak noise≈20mV
→ SNR≈17 (assuming uniform noise)
Injection
AmpOut Saturation
Injection Voltage [V]
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
AmpOut Amplitude [mV]
0 50 100 150 200 250 300 350 400 450
ToT [ns]
1000 1500 2000 2500 3000 3500 4000
Sr90 @−2V HV
Error bars = standard deviation
AmpOut saturates, no longer triangular ToT less affected
Line Crosstalk
Row 100→ Row 101
Injection Voltage [V]
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
AmpOut Amplitude Injected Pixel[mV]
100 150 200 250 300 350 400
AmpOut Amplitude Neighbor Pixel [mV]
0 10 20 30 40 50
Row
n-2 n-1 n n+1 n+2
Amplifier Output Ratio
0 0.2 0.4 0.6 0.8
1 row 10
row 50 row 70 row 100 row 135 row 150 row 170 row 197
Crosstalk small compared to signal (Fe55 5.9keV≈100mV)
Should be small for low momentum electrons (≈8−12keV for 20−30µm depletion)
HV Dependence
Fe55
HV [-V]
0 10 20 30 40 50 60
Hit Rate [kHz]
40 45 50 55 60 65 70 75 80
Injection
HV [-V]
0 10 20 30 40 50 60
AmpOut Amplitude [mV]
0 20 40 60 80 100 120 140 160 180 200
ToT [ns]
0 500 1000 1500 2000 2500
Depletion volume and pixel capacity depend on bias HV Signal shows only little dependence on HV/pixel capacity
Row Dependence Signal
0 50 100 150 200
Fe55 Amplitude [mV]
80 90 100 110 120 130 140 150 160 170
Row Address 90 100 110 120 130 Fe55 Amplitude [mV]140
Entries/2mV
0 5 10 15 20
25 Entries 200
Mean 119.2 RMS 7.624
No dependence on row for signals
MuPix Telescope with 4GeV Positrons @DESY Tile MuPix Tile Beam
x y
z
Efficiency and Noise
10 20 30 40
col / pixel 20
40 60 80 100 120 140 160 180
row / pixel
0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1
efficiency
10 20 30 40
col / pixel 20
40 60 80 100 120 140 160 180
row / pixel
−2
10
−1
10 1
noiserate per pixel / hz
Overall efficiency: 99.7%
Chip untuned
Noise rate≈0.5 Hz per pixel
Noisy pixels ignored HV = - 60 V
Low threshold (45 mV)
Thresholdscan at -60V
40 60 80 100 120 140 160 180
Efficiency
0.8 0.85 0.9 0.95 1
40 60 80 100 120 140 160 180 0
1 2 3 4 5 6 7 8 9 10
Threshold [mV]
Noise rate per pixel [Hz]
Up to 25 noisy pixels ignored in analysis Broad operation region of over 40 mV
Timing Dependence
Delay
col/row address
0 50 100 150 200
peakpos [8 ns]
2 2.5 3 3.5 4 4.5 5 5.5 6
row column
Likely related to Clock Distribution
Resolution
row/col address
0 50 100 150 200
[8 ns]σTime resolution
2.8 3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 4.8
row column
Variation probably due to line capacities
Corrected Time Resolution
Correction for delay gives significant improvement Analysis limited by sample size
Possible improvement by tuning
trigger - hit time / 8 ns
−20 0 20 40 60 80
Entries
0 0.005 0.01 0.015 0.02 0.025 0.03
22 ns σ raw data
17 ns σ pixel corrected
Summary
MuPix8 as first large scale prototype in operation
High efficiency and wide range without much optimization Time resolution ok but can be improved Lots of testbeam data left to analyze
40 60 80 100 120 140 160 180
Efficiency
0.8 0.85 0.9 0.95 1
40 60 80 100 120 140 160 180 0
1 2 3 4 5 6 7 8 9 10
Threshold [mV]
Noise rate per pixel [Hz]
Outlook
Improvements by Tuning
More measurements in the lab
Lots of new features in the MuPix8 barely explored
Measurements with B+C
20 40 60 80 100 120
0 20 40 60 80 100 120 140 160 180
0 100 200 300 400 500 600
Column Address
Row Address
Hits
BACKUP
Crosstalk Absolute
row
n-2 n-1 n n+1 n+2
amplifier output [mV]
0 50 100 150 200 250 300 350 400 450
500 row 10
row 50 row 70 row 100 row 135 row 150 row 170 row 197
Row Dependence Injection
row
0 50 100 150 200
Fe55 equivalent Injection [mV]
230 235 240 245 250 255 260 265 270
Injection shows a row dependence Likely due to line capacity
Masked pixels
threshold / mV 40 60 80 100 120 140 160 180
maksed pixel
5 10 15 20 25
Pixels with a factor 10 more rate
as the average and at least 2 Hz rate are masked
Timing with higher Current
−20 −10 0 10 20 30 Trigger ts - hit time [8ns]
0 1000 2000 3000 4000 5000 6000 7000 8000
= 12.8 ns σ
PRELIMINARY
Tuning
muhist Entries 9323 Mean 0.02462 RMS 0.008751
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0 200 400 600 800 1000 1200
muhist Entries 9323 Mean 0.02462 RMS 0.008751
Mean Untuned - Mean Tuned [V]
Entries
MuPix8 readout
Pixel Periphery State Machine
Pixel Periphery State Machine
Pixel Periphery State Machine
readout state machine
VCO
&
PLL
8b/10b
encoder serializer LVDS ...
other pixels
LVDSLVDS
MUX LVDS
sensor CSA
comparator 1&2
tune DAC
threshold baseline
test-pulse injection
readout
x3
integrate charge
amplification
line driver per pixel digital output
threshold adjustment
Crosstalk and clustering
Particle induced clusters fire up to 2x2 pixels
Analog lines between active pixel and digital periphery are crosstalk affected→ row dependence
Only track assigned clusters
≤3 considered in crosstalk analysis
Difference between horizontal and vertical cluster sizes correspond to
crosstalk sharingcharge charge
sharing / crosstalk
crosstalk
Crosstalk Row Dependence
vertical pos of cluster / pixel
50 100 150
Entries
0 0.05 0.1 0.15
0.2 0.25 0.3
double cluster triple cluster
Crosstalk Thresholdscan
Average crosstalk over sensor Ignores row dependence
Heavily dependent on threshold
Ncluster,vertical−Ncluster,horizontal
Ncluster threshold / mV
40 60 80 100 120 140 160 180
crosstalk / %
0 2 4 6 8
10 Preliminary
Increasing Supply Voltages
VDD/VDDA = 1.9 V VSSA = 1.1 V
Time resolution improves by 15%
td0
Entries 2587518
Mean 5.774
RMS 7.739
HitTime-TriggerTime [8 ns]
0 10 20 30 40
0 20 40 60 80 100 120
103
×
td0
Entries 2587518
Mean 5.774
RMS 7.739
= 23.8 ns σ Default voltages
= 20.5 ns σ Increased voltages
ToT and timewalk maesurements
Timewalk Correction
Thr1 Comp Signal
Thr2 Comp Signal
Voltage
Time Threshold1
Threshold2 Signal
Hold TimeStamp
Store Timestamp
& Hit Flag
Voltage
Time Dynamic Threshold
ToT Threshold
Baseline
Thr Comp Signal
Dynamic Threshold
Start Ramp
& Store Timestamp
Dyn Thr Comp Signal
Store ToT Timestamp
& Hit Flag
Timing with higher supply Voltages
Before
reference time [8 ns]
0 5 10 15 20 25 30
MuPix8 timestamp [8 ns]
0 5 10 15 20 25 30
600 800 1000 1200 1400 1600 Correlation of 5 LSBs in Sensor 0 and Trigger 0
After
reference time [8 ns]
0 5 10 15 20 25 30
MuPix8 timestamp [8 ns]
0 5 10 15 20 25 30
3500 4000 4500 5000 5500 6000 6500 7000 7500
Timing with higher supply Voltages
Before
tsbits Entries 3.735316e+07 Mean 4.418 RMS 2.892
gray encoded ts bits
0 1 2 3 4 5 6 7 8 9
probability to be 1
0.5 0.51 0.52 0.53 0.54 0.55
tsbits Entries 3.735316e+07 Mean 4.418 RMS 2.892
Timestamp Bits After
gray encoded time stamp bit LSB=0
0 1 2 3 4 5 6 7 8 9
probability of ts bit to be 1
0.496 0.498 0.500 0.502 0.504 0.506 0.508