Transcript DCDRewx

DCD
ASIC Review 2014
1
•
•
•
•
DCD is implemented in UMC 0.18 um CMOS technology
3.2mm x 5mm
DCD-B uses bump bonding on the UMC technology provided by EuroPractice.
DCD has 256 analog channels each housing an input stage and ADCs (1 pipeline or 2 cyclic).
ASIC Review 2014
2
DCDBPip
TIA
ADC
200 µm
5 mm
ASIC Review 2014
3
•
•
•
•
The analog input stage performs various tasks.
Keeps voltage constant, amplification, shaping, CM correction, pedestal correction
The analog signal is digitized using current-mode cyclic or pipelined ADCs
A large synthesized digital block decodes and derandomizes the ADC raw data which are then
transmitted in a well sorted sequence to the DHP chips using fast parallel 8-bit digital outputs.
ASIC Review 2014
4
•
DCD Block
ASIC Review 2014
5
•
Analog Part with cyclic ADC
ASIC Review 2014
6
•
Analog Part with pipelined ADC
ASIC Review 2014
7
•
•
•
•
Current-receiver based on a trans-impedance amplifier with a resistor on its output that amplifies
the DEPFET current.
Two current-mode cyclic ADCs (ADCR and ADCL) based on current-memory cells (CMC).
Two-bit DAC for pedestal correction.
Decoder for generating of the control signals for the ADC.
ASIC Review 2014
8
•
•
•
•
The hearth of the block is a trans-impedance amplifier
A ”bleeding resistor” Rs is connected between Vout and the input of the ADC. Since the ADC
holds its input at a constant potential, the current flowing into the ADC is proportional to Iin
The amplifier can operate in the mode to suppress common mode variations and amplify only the
difference from the common mode signal.
Novel voltage-drop insensitive current sources with enclosed NMOS transistors have been used
for sensible currents like ISF
ASIC Review 2014
9
•
Signal receiver
ASIC Review 2014
10
Common mode operation
ASIC Review 2014
11
Single-Input Amplifier
•
•
The single input amplifier is not able to distinguish between signal and crosstalk
Amplifier bandwidth quite high ~ 100Mhz/2
5cm
Crosstalk sources:
Switching of DEPFET rows
Fluctuations in power supply voltages
EM environment
Assumption: crosstalk affects all channels equally – “common mode noise”
ADC
ASIC Review 2014
12
Differential Amplifier
•
Differential amplifiers are not sensitive to common signals.
ADC
Output common mode feedback
ADC
+
Input common mode feedback
+
ASIC Review 2014
13
Differential Amplifier
•
•
•
Problem: DEPFET signal is not differential.
Idea: feed two DEPFET signals to a differential amplifier.
Drawback: double hits are not amplified.
ADC
ADC
Hit
+
+
ASIC Review 2014
14
New: Multi-differential Amplifier
•
•
Idea: extend the differential scheme to n inputs. The inputs are connected
to n DEPFET rows.
The amplifier is not sensitive common signals.
ADC
ADC
ADC
ADC
ADC
+
ADC
+
ASIC Review 2014
15
New: Multi-differential Amplifier
•
•
•
The amplifier amplifies the difference from the mean value.
In the case of large n, the mean value is not influenced by single hits.
Behaves as the single input amplifier for the case of sparse signals.
ADC
ADC
ADC
ADC
ADC
+
ADC
+
ASIC Review 2014
16
•
•
•
•
•
•
•
•
•
The ADC uses redundant signed-digit (RSD) conversion.
The algorithm starts with the comparison of the input signal with two thresholds, one positive and
one negative.
If the input signal is larger than the positive threshold, the pair of output code bits is set to 10,
meaning +1, and a reference current is
subtracted.
If the input signal is lower than the negative threshold, the output code is set to 01 (-1) and the
reference is added.
If the input signal value is between the thresholds, the bits are set to 00 (0) and no arithmetical
operation is carried out.
The residue signal is multiplied by two and the result undergoes the same operation for the next
bits.
The conversion is not inuenced by the comparator osets, providing the osets are not larger than
half of the threshold.
Current-mode memory cells are used to implement the described A/D conversion algorithm.
ASIC Review 2014
17
Cf
Wr
1
Iin/Iout
I1
Wr+Rd
2
Sw1
Iin/Iout
1
Sw1
4
Sw2
ASIC Review 2014
2
Sw2
C
T1
A
3
Wr+Rd
Wr
A
IO
Ref
VGS
TC
Ref2
sample state 1
sample state 2
S
wr
S
r
r
1.
nc
state3
lt
lt
3.
nc
nc
2.
1.
nc
nc
4.
3.
ck1
c
c
h0
wr
lt
memory cell
l0
lt
comparator
rd
rd
lt
lt
rd
2.
lt
lt
nc
wr
r
4.
ck2
state1
rd
state4
nc
2(S-h0R+l0R)
ck3
state2
r
nc
h2
c
wr
2(S-h0R+l0R)
ck4
state3
2(2(S-h0R+l0R)-h1R+l1R)
c
state4
States:
rd – read
wr – write
nc – not connected
r – reset
c – compare
lt - latched
sample state 1
S‘
l2
wr
r
r
nc
nc
c
c
wr
rd
lt
lt
rd
rd
lt
lt
rd
wr
r
r
nc
rd
lt
lt
rd
rd
lt
lt
rd
wr
r
r
nc
nc
c
c
wr
nc
lt
lt
nc
h1
l1
ck5
ck6
ck7
2(2(2(S-h0R+l0R)-h1R+l1R)–h2R+l2R)
ck8
h3
ck9
2(2(2(2(S - h0R + l0R) - h1R + l1R) – h2R + l2R) – h3R + l3R) = Res Res
ASIC Review 2014
l3
•
ADC
ASIC Review 2014
20
•
Current memory cell
ThHi
ThLo
Cf
Cmp1 Cmp2
Rd
WrB*
Wr*
Sw1
NotRd AND Not Wr
4
Sw5
NotWr
Sub
WrB
Sw2
Add
2
VPSource
VPSourceCasc
Sw3
A
3
RefNWELL
Iin
Sw4
VAmpPBias
DAC
1
NotRd
En Logic
SF
AmpLow
24 μA
VFBPBias
RefIn
RefIn
VFBNCasc
RefFB
12 μA
VFBNBias (VPSource2)
24 μA
TC
RefIn
ASIC Review 2014
21
•
Comparator
ResB
Or th
VPMOS
LB
VNMOS
ResB
RefIn
6X2 μA
Th
Comp In
AmpLow
IFBP
24 μA
TP1 TP2
12 μA
IFBN
Vbias
24 μA
Gate
ASIC Review 2014
22
•
Global bias block
ASIC Review 2014
23
•
Slow control
ASIC Review 2014
24
•
•
•
•
•
•
•
•
•
•
•
•
•
TCUM1 first implementation of the cyclic ADC based on current memory cells (CMC) – too slow
and too large
TCUM3 improved (for speed) implementation of the cyclic and pipelined ADC with CMCs – too
large
DCD1 72-channel readout chip for DEPFET – small cyclic ADCs (2/channel) and regulated
cascode as receiver – 80ns sampling time and proper size – too high noise due to crosstalk
DCD2 – fixed crosstalk problem (constant current consumption) – INL ~3.5 in some channels and
noise ~50nA at 100ns sampling time
DCDB1 256-channel chip with Belle II size – cyclic ADCs from DCD2 and transimpedance
amplifers – works but noise a bit high at 100ns sampling time (120nA)
DCDB2 – noise improved ~ 60nA at full speed – INL about 4 LSB in some channels
DCDB4 cyclic – improved DCDB2 not tested yet
DCDB pipelined – noise ~ 45nA at full speed – INL about 2.5 LSB (sometimes some channels
have long codes)
We believe, the problem has been understood and can be solved by resizing of transistors
Still to be decided – should we use cyclic or pipelined version or both
We need more measurements
Measurements of DCD cyclic
DC characteristics measurements
ASIC Review 2014
25
Test all ADCs
•
Test results on DCD pipeline
ADC gain 72nA/LSB (~110e @ gq 650pA/e)
Noise: ~0.55LSB (60e @ gq 650pA/e)
90
80
1.4
1.2
70
DNL of All ADCs
5
DNL [ADU]
100
Peak-to-Peak INL of All ADCs
INL [ADU]
Mean Noise of All ADCs
Noise [ADU]
Gain [nA/ADU]
Gain of All ADCs
4.5
10
9
4
8
3.5
7
1
60
3
50
40
0.6
30
220e
6
275e
0.8
2.5
5
2
4
1.5
3
1
2
0.5
1
0.4
20
0.2
10
0
0
20
40
60
ASIC Review 2014
80
100
0
0
20
40
60
80
100
0
0
20
40
60
80
100
0
0
20
40
60
80
100
26