Transcript Document

APV25s1 STATUS
Testing started beginning September
1 wafer cut, others left for probing
10 chips mounted on test boards
9 work well
all I2C registers read/writeable
no obvious pipeline defects (run with
pseudo-random trigger after single reset)
pulse shape OK for one channel
1 chip with obvious analogue problem
1 channel pedestal stuck high and low gain
in others
Detailed results here confined to few chips so far
Outline
Performance of S1 chip version (early results)
Pulse shape, linearity, noise,
pipeline uniformity, radiation tests
Conclusions
Mark Raymond
[email protected]
Oct, 2000
CMS Tracker Electronics
1
Wafer layout
~ 390 viable APV25s1 sites / wafer
Oct, 2000
CMS Tracker Electronics
2
Wafer detail showing reticule
reticule consists of
4 APVs
1 APVMUX/PLL
test structures
Oct, 2000
CMS Tracker Electronics
3
7.1 mm
APV25s1 layout
8.1 mm
design changes
input edge layout changes to reduce track resistance
(input pad layout changed)
calibration capacitor layout improved
gain increased (resistor values at mux input stage)
added internal master current reference for bias generator
(backend pad layout changed from s0 version)
digital bug in pipeline logic fixed
minor tweaks to component values to achieve greater
margins for operation (shaper feedback resistance)
Oct, 2000
CMS Tracker Electronics
4
Output data frame
Raw data frame
4
2
Digital header
1 mip
0
Current [mA]
-2
-4
Analogue part of data
frame software reordered
4
Actual channel order
2
0
-2
-4
Samples at 50 nsec. intervals
differential current output (+ve o/p only shown above)
nominal (mid-range) gain now 1.2 mA/mip
variable between 0.8 and 1.5 in 5 steps
Oct, 2000
CMS Tracker Electronics
5
APV25s1 biasing
internal/external option determined by 2 pads (IREF & IREFBIAS)
mode
IREF
IREFBIAS IVDD[mA]
external
internal
128mA (R to VDD)
disconnect
VDD
GND
94
89
IVSS[mA]
156
143
(VDD=2.5, GND=1.25, VSS=0)
recommended bias settings (decimal) (preliminary)
if external bias at 128 mA
IPRE
IPCASC
IPSF
ISHA
ISSF
IPSP
IMUXIN
VFP
VFS
85
45
30
~ 30 tune for optimum pulse rise time
30
48
30
~ 30
~ 60 tune for optimum pulse fall time
NOTE: These current values are ~70% less than those for s0
(current mirroring ratio changed)
Oct, 2000
CMS Tracker Electronics
6
Pulse shape dependence on input capacitance
2pF
4p1
8p1
10p7
14p5
17p5
20p5
125
100
75
50
25
0
0
50
100
150
200
250
time [nsec.]
Peak mode pulse shape tuned (shaper amplifier bias settings)
for each value of input capacitance
ISHA: 21 -> 65
VFS: 63 -> 50
Oct, 2000
(I2C register setting ~ roughly = mA)
CMS Tracker Electronics
7
Pulse shape dependence on shaper feedback
resistor control voltage VFS
200
VFS value (decimal)
180
160
160
ADC units
140
120
120
100
80
80
60
40
20
0
40
0
0
25
50
75 100 125 150 175 200 225 250
time [nsec]
Feedback FET length slightly reduced from S0 version
to ensure short enough time constant achievable with
margin
optimum value here ~ 60 (decimal)
Oct, 2000
CMS Tracker Electronics
8
Linearity
500
400
ADC units
Input signal
charge injected
in 0.5 mip steps
from 0.5 to 7
mips.
300
200
100
0
0
7
6
5
4
3
2
1
0
50 100 150 200 250
time [nsec.]
peak mode
deconvolution
0
1
2
3
4
5
6
input signal injected [mips]
7
output normalised to input at 1 mip point
linearity good up to 3 mips, gradual fall off beyond
Oct, 2000
CMS Tracker Electronics
9
Gain/calibration uniformity
Calibration capacitor layout improved
Calibration response of
all 128 channels
superimposed
ADC units
150
100
50
0
0
20
40
60
80 100
3.125 nsec steps
Pulse peak height dependence on channel number
ADC units
150
100
50
0
0
32
64
96
128
20
entries:128
16 mean:131.0
1.91
12 sigma:
min:127.3
8 max: 134.6
4
0
100
120
channel no.(mux order)
140
peak ht. [ADC units]
Good channel matching
Oct, 2000
CMS Tracker Electronics
10
Noise
Dependence on input capacitance
ENC [rms electrons]
2000
1600
closed symbols: peak mode: 270 + 38/pF
open symbols:deconvolution: 430 + 61/pF
1200
800
chan 2
chan 43
chan 107
400
0
-10
-5
0
5
10
15
20
25
Input capacitance [pF]
No dependence on channel number
< 2000 electrons achievable for detectors < 25 pF
Oct, 2000
CMS Tracker Electronics
11
Pipeline - pedestals
Pedestal dependence on pipeline location
ADC units
2.0
channel 19, peak mode
1.0
0.0
-1.0
-2.0
0
32
64
96
128
pipeline location
160
192
Measure for every channel, take rms value, convert to electrons
and histogram
frequency
20
peak mode
deconvolution
15
10
5
0
0
25 50 75 100 125 150
rms electrons
No significant noise contribution
Oct, 2000
CMS Tracker Electronics
12
Pipeline – gain uniformity
How well do pipeline capacitors match?
Signal retrieved as charge so mismatch => gain dependence on
pipeline capacitance
Measure by storing/retrieving signal to/from every cell
50
192 entries
mean 122.7
sd 0.86
min 120.6
max 125.7
frequency
40
30
20
10
0
0
50
100
ADC units
150
Very good matching between pipeline cells
Oct, 2000
CMS Tracker Electronics
13
Radiation tests
1chip irradiated to 10 Mrads using 50 kV X-rays
ADC units
before
after 10 Mrads
500
400
300
200
100
0
500
400
300
200
100
0
0
50 100 150 200 250
0
50 100 150 200 250
time [nsec.]
minor retuning of pulse shape required after irradiation
Noise – histogram all channels
peak mode
deconvolution
before
after 10 Mrads
0.0
0.5
1.0
1.5
2.0
rms ADC units
2.5
3.0
Almost no observable degradation after 10 Mrads (confirms
results from s0 run)
Oct, 2000
CMS Tracker Electronics
14
Probe testing plans for APV25s1
New probe card made (only needs active and decoupling
components to be added)
similar (functionally) to s0 version except
multiple (3) probes for power on front edge
1 amplifier input probed allowing true chip gain
to be determined
Additions to test software
measure pulse height for probed channel
test multi-mode operation
sweep one bias setting measuring power supply currents
Tested die should be available early November
Oct, 2000
CMS Tracker Electronics
15
APVMUX/PLL status
Oct, 2000
CMS Tracker Electronics
16
Conclusions
Early APV25s1 results indicate good performance consistent
with results from APV25s0. Minor problems fixed.
Oct, 2000
CMS Tracker Electronics
17