LTC2348-18 - Octal, 18-Bit, 200ksps Differential ±10.24V Input SoftSpan ADC with Wide Input Common Mode Range

Features
LTC2348-18
Octal, 18-Bit, 200ksps
Differential ±10.24V Input SoftSpan ADC
with Wide Input Common Mode Range
Description
200ksps per Channel Throughput
nn Eight Simultaneous Sampling Channels
nn ±3LSB INL (Maximum, ±10.24V Range)
nn Guaranteed 18-Bit, No Missing Codes
nn Differential, Wide Common Mode Range Inputs
nn Per-Channel SoftSpan Input Ranges:
±10.24V, 0V to 10.24V, ±5.12V, 0V to 5.12V
±12.5V, 0V to 12.5V, ±6.25V, 0V to 6.25V
nn 96.7dB Single-Conversion SNR (Typical)
nn −109dB THD (Typical) at f = 2kHz
IN
nn 118dB CMRR (Typical) at f = 200Hz
IN
nn Rail-to-Rail Input Overdrive Tolerance
nn Guaranteed Operation to 125°C
nn Integrated Reference and Buffer (4.096V)
nn SPI CMOS (1.8V to 5V) and LVDS Serial I/O
nn Internal Conversion Clock, No Cycle Latency
nn 140mW Power Dissipation (Typical)
nn 48-Lead (7mm x 7mm) LQFP Package
The LTC®2348-18 is an 18-bit, low noise 8-channel simultaneous sampling successive approximation register
(SAR) ADC with differential, wide common mode range
inputs. Operating from a 5V low voltage supply, flexible
high voltage supplies, and using the internal reference
and buffer, each channel of this SoftSpanTM ADC can be
independently configured on a conversion-by-conversion
basis to accept ±10.24V, 0V to 10.24V, ±5.12V, or 0V to
5.12V signals. Individual channels may also be disabled
to increase throughput on the remaining channels.
nn
The wide input common mode range and 118dB CMRR of
the LTC2348-18 analog inputs allow the ADC to directly
digitize a variety of signals, simplifying signal chain design. This input signal flexibility, combined with ±3LSB
INL, no missing codes at 18 bits, and 96.7dB SNR, makes
the LTC2348-18 an ideal choice for many high voltage
applications requiring wide dynamic range.
The LTC2348-18 supports pin-selectable SPI CMOS (1.8V
to 5V) and LVDS serial interfaces. Between one and eight
lanes of data output may be employed in CMOS mode,
allowing the user to optimize bus width and throughput.
Applications
Programmable Logic Controllers
Industrial Process Control
nn Power Line Monitoring
nn Test and Measurement
nn
L, LT, LTC, LTM, Linear Technology and the Linear logo are registered trademarks and
SoftSpan is a trademark of Linear Technology Corporation. All other trademarks are the property
of their respective owners. Protected by U.S. Patents, including 7705765, 7961132, 8319673.
Other Patents pending.
nn
Typical Application
15V
0.1µF
5V
0.1µF
2.2µF
1.8V TO 5V
0.1µF
Integral Nonlinearity vs
Output Code and Channel
CMOS OR LVDS
I/O INTERFACE
0V
0V
0V
–10V
–10V
VDD
VDDLBYP
UNIPOLAR
DIFFERENTIAL INPUTS IN+/IN– WITH
WIDE INPUT COMMON MODE RANGE
S/H
S/H
1.5
LTC2348-18
1.0
SDO0
MUX
18-BIT
SAR ADC
SDO7
SCKO
SCKI
SDI
CS
BUSY
CNV
S/H
S/H
IN7+
S/H
IN7–
2.0
OVDD LVDS/CMOS
PD
S/H
–5V
+10V
VCC
S/H
0V
• • •
TRUE BIPOLAR
+10V
IN0+ S/H
IN0–
• • •
–10V
FULLY
DIFFERENTIAL
+5V
VEE REFBUF
REFIN
INL ERROR (LSB)
+10V
ARBITRARY
GND
0.1µF
47µF
0.5
0
–0.5
–1.0
SAMPLE
CLOCK
234818 TA01a
EIGHT SIMULTANEOUS
SAMPLING CHANNELS
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
ALL CHANNELS
0.1µF
–15V
–1.5
–2.0
–131072
–65536
0
65536
OUTPUT CODE
131072
234818 TA01b
234818fa
For more information www.linear.com/LTC2348-18
1
LTC2348-18
Absolute Maximum Ratings
Pin Configuration
(Notes 1, 2)
TOP VIEW
48
47
46
45
44
43
42
41
40
39
38
37
IN7+
IN7–
GND
VEE
GND
VDD
VDD
GND
VDDLBYP
CS
BUSY
SDI
Supply Voltage (VCC)......................–0.3V to (VEE + 40V)
Supply Voltage (VEE)................................. –17.4V to 0.3V
Supply Voltage Difference (VCC – VEE).......................40V
Supply Voltage (VDD)...................................................6V
Supply Voltage (OVDD).................................................6V
Internal Regulated Supply Bypass (VDDLBYP).... (Note 3)
Analog Input Voltage
IN0+ to IN7+,
IN0– to IN7– (Note 4).......... (VEE – 0.3V) to (VCC + 0.3V)
REFIN..................................................... –0.3V to 2.8V
REFBUF, CNV (Note 5).............. –0.3V to (VDD + 0.3V)
Digital Input Voltage (Note 5)...... –0.3V to (OVDD + 0.3V)
Digital Output Voltage (Note 5)... –0.3V to (OVDD + 0.3V)
Power Dissipation............................................... 500mW
Operating Temperature Range
LTC2348C................................................. 0°C to 70°C
LTC2348I..............................................–40°C to 85°C
LTC2348H........................................... –40°C to 125°C
Storage Temperature Range................... –65°C to 150°C
IN6– 1
IN6+ 2
IN5– 3
IN5+ 4
IN4– 5
IN4+ 6
IN3– 7
IN3+ 8
IN2– 9
IN2+ 10
IN1– 11
IN1+ 12
SDO7
SDO–/SDO6
SDO+/SDO5
SCKO–/SDO4
SCKO+/SCKO
OVDD
GND
SCKI–/SCKI
SCKI+/SDO3
SDI–/SDO2
SDI+/SDO1
SDO0
IN0– 13
IN0+ 14
GND 15
VCC 16
VEE 17
GND 18
REFIN 19
GND 20
REFBUF 21
PD 22
LVDS/CMOS 23
CNV 24
36
35
34
33
32
31
30
29
28
27
26
25
LX PACKAGE
48-LEAD (7mm × 7mm) PLASTIC LQFP
TJMAX = 150°C, θJA = 53°C/W
Order Information
LEAD FREE FINISH
TRAY
PART MARKING*
PACKAGE DESCRIPTION
TEMPERATURE RANGE
LTC2348CLX-18#PBF
LTC2348CLX-18#PBF
LTC2348LX-18
48-Lead (7mm × 7mm) Plastic LQFP
0°C to 70°C
LTC2348ILX-18#PBF
LTC2348ILX-18#PBF
LTC2348LX-18
48-Lead (7mm × 7mm) Plastic LQFP
–40°C to 85°C
LTC2348HLX-18#PBF
LTC2348HLX-18#PBF
LTC2348LX-18
48-Lead (7mm × 7mm) Plastic LQFP
–40°C to 125°C
Consult LTC Marketing for parts specified with wider operating temperature ranges. *The temperature grade is identified by a label on the shipping container.
For more information on lead free part marking, go to: http://www.linear.com/leadfree/
234818fa
2
For more information www.linear.com/LTC2348-18
LTC2348-18
Electrical Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. (Note 6)
SYMBOL
PARAMETER
CONDITIONS
VIN+
Absolute Input Range
(IN0+ to IN7+)
VIN–
Absolute Input Range
(IN0– to IN7–)
VIN+ – VIN– Input Differential Voltage
Range
VCM
TYP
MAX
UNITS
(Note 7)
VEE
VCC – 4
V
(Note 7)
l
VEE
VCC – 4
V
SoftSpan 7: ±2.5 • VREFBUF Range (Note 7)
SoftSpan 6: ±2.5 • VREFBUF/1.024 Range (Note 7)
SoftSpan 5: 0V to 2.5 • VREFBUF Range (Note 7)
SoftSpan 4: 0V to 2.5 • VREFBUF/1.024 Range (Note 7)
SoftSpan 3: ±1.25 • VREFBUF Range (Note 7)
SoftSpan 2: ±1.25 • VREFBUF/1.024 Range (Note 7)
SoftSpan 1: 0V to 1.25 • VREFBUF Range (Note 7)
l
–2.5 • VREFBUF
l –2.5 • VREFBUF/1.024
l
0
l
0
l
–1.25 • VREFBUF
l –1.25 • VREFBUF/1.024
l
0
2.5 • VREFBUF
2.5 • VREFBUF/1.024
2.5 • VREFBUF
2.5 • VREFBUF/1.024
1.25 • VREFBUF
1.25 • VREFBUF/1.024
1.25 • VREFBUF
V
V
V
V
V
V
V
Input Common Mode Voltage (Note 7)
Range
VIN+ – VIN– Input Differential Overdrive
Tolerance
MIN
l
(Note 8)
IIN
Analog Input Leakage Current
CIN
Analog Input Capacitance
Sample Mode
Hold Mode
CMRR
Input Common Mode
Rejection Ratio
VIN+ = VIN− = 18VP-P 200Hz Sine
VIHCNV
l
VEE
VCC – 4
V
l
−(VCC − VEE)
(VCC − VEE)
V
l
–1
1
µA
l
100
CNV High Level Input Voltage
l
1.3
VILCNV
CNV Low Level Input Voltage
l
IINCNV
CNV Input Current
VIN = 0V to VDD
50
10
pF
pF
118
dB
V
–10
l
0.5
V
10
μA
Converter Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL
PARAMETER
CONDITIONS
MIN
MAX
UNITS
l
18
Bits
No Missing Codes
l
18
Bits
Transition Noise
SoftSpans 7 and 6: ±10.24V and ±10V Ranges
SoftSpans 5 and 4: 0V to 10.24V and 0V to 10V Ranges
SoftSpans 3 and 2: ±5.12V and ±5V Ranges
SoftSpan 1: 0V to 5.12V Range
INL
Integral Linearity Error
SoftSpans 7 and 6: ±10.24V and ±10V Ranges (Note 10)
SoftSpans 5 and 4: 0V to 10.24V and 0V to 10V Ranges (Note 10)
SoftSpans 3 and 2: ±5.12V and ±5V Ranges (Note 10)
SoftSpan 1: 0V to 5.12V Range (Note 10)
DNL
Differential Linearity Error (Note 11)
ZSE
Zero-Scale Error
(Note 12)
1.3
2.6
2.0
4.0
(Note 12)
l
–3
–4
–2.5
–2.5
±1
±1.5
±0.75
±0.75
l
−0.9
l
−550
l
l
l
l
Zero-Scale Error Drift
FSE
TYP
Resolution
Full-Scale Error
LSBRMS
LSBRMS
LSBRMS
LSBRMS
3
4
2.5
2.5
LSB
LSB
LSB
LSB
±0.2
0.9
LSB
±80
550
μV
±2
Full-Scale Error Drift
−0.1
±0.025
±2.5
μV/°C
0.1
%FS
ppm/°C
234818fa
For more information www.linear.com/LTC2348-18
3
LTC2348-18
Dynamic Accuracy
The l denotes the specifications which apply over the full operating temperature range,
otherwise specifications are at TA = 25°C. AIN = –1dBFS. (Notes 9, 13)
SYMBOL PARAMETER
CONDITIONS
MIN
TYP
SINAD
Signal-to-(Noise +
Distortion) Ratio
l
SoftSpans 7 and 6: ±10.24V and ±10V Ranges, fIN = 2kHz
SoftSpans 5 and 4: 0V to 10.24V and 0V to 10V Ranges, fIN = 2kHz l
l
SoftSpans 3 and 2: ±5.12V and ±5V Ranges, fIN = 2kHz
l
SoftSpan 1: 0V to 5.12V Range, fIN = 2kHz
93.0
87.6
90.0
84.2
96.5
90.6
93.2
87.3
dB
dB
dB
dB
SNR
Signal-to-Noise Ratio
SoftSpans 7 and 6: ±10.24V and ±10V Ranges, fIN = 2kHz
SoftSpans 5 and 4: 0V to 10.24V and 0V to 10V Ranges, fIN = 2kHz
SoftSpans 3 and 2: ±5.12V and ±5V Ranges, fIN = 2kHz
SoftSpan 1: 0V to 5.12V Range, fIN = 2kHz
l
l
l
l
93.7
87.7
90.2
84.3
96.7
90.7
93.2
87.3
dB
dB
dB
dB
THD
Total Harmonic Distortion
SoftSpans 7 and 6: ±10.24V and ±10V Ranges, fIN = 2kHz
SoftSpans 5 and 4: 0V to 10.24V and 0V to 10V Ranges, fIN = 2kHz
SoftSpans 3 and 2: ±5.12V and ±5V Ranges, fIN = 2kHz
SoftSpan 1: 0V to 5.12V Range, fIN = 2kHz
l
l
l
l
SFDR
Spurious Free Dynamic
Range
SoftSpans 7 and 6: ±10.24V and ±10V Ranges, fIN = 2kHz
SoftSpans 5 and 4: 0V to 10.24V and 0V to 10V Ranges, fIN = 2kHz
SoftSpans 3 and 2: ±5.12V and ±5V Ranges, fIN = 2kHz
SoftSpan 1: 0V to 5.12V Range, fIN = 2kHz
l
l
l
l
Channel-to-Channel
Crosstalk
One Channel Converting 18VP-P 200Hz Sine in ±10.24V Range,
Crosstalk to All Other Channels
–109
–111
–113
–114
101
105
105
105
–3dB Input Bandwidth
Aperture Delay
Aperture Delay Matching
Aperture Jitter
Transient Response
MAX
–101
–104
–104
–103
dB
dB
dB
dB
110
112
114
115
dB
dB
dB
dB
−109
dB
7
MHz
1
ns
150
ps
3
Full-Scale Step, 0.005% Settling
UNITS
psRMS
360
ns
Internal Reference Characteristics
The l denotes the specifications which apply over the full
operating temperature range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL
PARAMETER
VREFIN
Internal Reference Output Voltage
CONDITIONS
Internal Reference Temperature Coefficient
(Note 14)
Internal Reference Line Regulation
VDD = 4.75V to 5.25V
MIN
TYP
MAX
2.043
2.048
2.053
5
20
l
0.1
Internal Reference Output Impedance
VREFIN
REFIN Voltage Range
1.25
V
ppm/°C
mV/V
20
REFIN Overdriven (Note 7)
UNITS
kΩ
2.2
V
Reference Buffer Characteristics
The l denotes the specifications which apply over the full
operating temperature range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL
PARAMETER
CONDITIONS
VREFBUF
Reference Buffer Output Voltage
REFIN Overdriven, VREFIN = 2.048V
REFBUF Voltage Range
REFBUF Overdriven (Notes 7, 15)
REFBUF Input Impedance
VREFIN = 0V, Buffer Disabled
REFBUF Load Current
VREFBUF = 5V, 8 Channels Enabled (Notes 15, 16)
VREFBUF = 5V, Acquisition or Nap Mode (Note 15)
IREFBUF
MIN
TYP
MAX
UNITS
l
4.091
4.096
4.101
V
l
2.5
5
V
13
l
1.5
0.39
kΩ
1.9
mA
mA
234818fa
4
For more information www.linear.com/LTC2348-18
LTC2348-18
Digital Inputs and Digital Outputs
The l denotes the specifications which apply over the
full operating temperature range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL
PARAMETER
CONDITIONS
MIN
TYP
MAX
UNITS
CMOS Digital Inputs and Outputs
VIH
High Level Input Voltage
VIL
Low Level Input Voltage
IIN
Digital Input Current
CIN
Digital Input Capacitance
l 0.8 • OVDD
V
l
VIN = 0V to OVDD
l
–10
0.2 • OVDD
V
10
μA
5
pF
VOH
High Level Output Voltage
IOUT = –500μA
l OVDD – 0.2
VOL
Low Level Output Voltage
IOUT = 500μA
l
V
IOZ
Hi-Z Output Leakage Current
VOUT = 0V to OVDD
l
ISOURCE
Output Source Current
VOUT = 0V
–50
mA
ISINK
Output Sink Current
VOUT = OVDD
50
mA
0.2
–10
V
10
μA
LVDS Digital Inputs and Outputs
VID
Differential Input Voltage
RID
On-Chip Input Termination
Resistance
CS = 0V, VICM = 1.2V
CS = OVDD
l
200
350
600
mV
l
90
106
10
125
Ω
MΩ
1.2
2.2
V
10
μA
mV
VICM
Common-Mode Input Voltage
l
0.3
IICM
Common-Mode Input Current
VIN+ = VIN– = 0V to OVDD
l
–10
VOD
Differential Output Voltage
RL = 100Ω Differential Termination
l
275
350
425
VOCM
Common-Mode Output Voltage
RL = 100Ω Differential Termination
l
1.1
1.2
1.3
V
IOZ
Hi-Z Output Leakage Current
VOUT = 0V to OVDD
l
–10
10
μA
Power Requirements
The l denotes the specifications which apply over the full operating temperature
range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL PARAMETER
CONDITIONS
MIN
TYP
MAX
UNITS
VCC
Supply Voltage
l
0
38
V
VEE
Supply Voltage
l
–16.5
0
V
VCC − VEE Supply Voltage Difference
l
10
VDD
Supply Voltage
l
4.75
IVCC
Supply Current
200ksps Sample Rate, 8 Channels Enabled
Acquisition Mode
Nap Mode
Power Down Mode
l
l
l
l
IVEE
Supply Current
200ksps Sample Rate, 8 Channels Enabled
Acquisition Mode
Nap Mode
Power Down Mode
l
l
l
l
–2.8
–4.9
–1.1
–15
l
1.71
38
V
5.00
5.25
V
1.8
3.8
0.7
1
2.2
4.5
0.9
15
mA
mA
mA
μA
–2.2
–4.0
–0.8
–1
mA
mA
mA
μA
CMOS I/O Mode
OVDD
Supply Voltage
IVDD
Supply Current
200ksps Sample Rate, 8 Channels Enabled
200ksps Sample Rate, 8 Channels Enabled, VREFBUF = 5V (Note 15)
Acquisition Mode
Nap Mode
Power Down Mode (C-Grade and I-Grade)
Power Down Mode (H-Grade)
l
l
l
l
l
l
15.2
13.4
1.6
1.4
65
65
5.25
V
17.5
15.4
2.1
1.9
175
450
mA
mA
mA
mA
μA
µA
234818fa
For more information www.linear.com/LTC2348-18
5
LTC2348-18
Power Requirements
The l denotes the specifications which apply over the full operating temperature
range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL PARAMETER
CONDITIONS
MIN
IOVDD
Supply Current
200ksps Sample Rate, 8 Channels Enabled (CL = 25pF)
Acquisition or Nap Mode
Power Down Mode
l
l
l
PD
Power Dissipation
200ksps Sample Rate, 8 Channels Enabled
Acquisition Mode
Nap Mode
Power Down Mode (C-Grade and I-Grade)
Power Down Mode (H-Grade)
l
l
l
l
l
TYP
MAX
UNITS
1.6
1
1
2.6
20
20
mA
μA
μA
140
125
30
0.36
0.36
169
152
40
1.4
2.8
mW
mW
mW
mW
mW
5.25
V
LVDS I/O Mode
OVDD
Supply Voltage
IVDD
Supply Current
200ksps Sample Rate, 8 Channels Enabled
200ksps Sample Rate, 8 Channels Enabled, VREFBUF = 5V (Note 15)
Acquisition Mode
Nap Mode
Power Down Mode (C-Grade and I-Grade)
Power Down Mode (H-Grade)
l
l
l
l
l
l
17.7
16.1
3.2
3.0
65
65
20.4
18.5
3.8
3.7
175
450
mA
mA
mA
mA
μA
µA
IOVDD
Supply Current
200ksps Sample Rate, 8 Channels Enabled (RL = 100Ω)
Acquisition or Nap Mode (RL = 100Ω)
Power Down Mode
l
l
l
7
7
1
8.5
8.0
20
mA
mA
μA
PD
Power Dissipation
200ksps Sample Rate, 8 Channels Enabled
Acquisition Mode
Nap Mode
Power Down Mode (C-Grade and I-Grade)
Power Down Mode (H-Grade)
l
l
l
l
l
166
151
55
0.36
0.36
199
180
69
1.4
2.8
mW
mW
mW
mW
mW
l
2.375
ADC Timing Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL
PARAMETER
CONDITIONS
MIN
TYP
fSMPL
Maximum Sampling Frequency
8 Channels Enabled
7 Channels Enabled
6 Channels Enabled
5 Channels Enabled
4 Channels Enabled
3 Channels Enabled
2 Channels Enabled
1 Channel Enabled
l
l
l
l
l
l
l
l
tCYC
Time Between Conversions
8 Channels Enabled, fSMPL = 200ksps
7 Channels Enabled, fSMPL = 225ksps
6 Channels Enabled, fSMPL = 266ksps
5 Channels Enabled, fSMPL = 300ksps
4 Channels Enabled, fSMPL = 375ksps
3 Channels Enabled, fSMPL = 450ksps
2 Channels Enabled, fSMPL = 625ksps
1 Channel Enabled, fSMPL = 1000ksps
l
l
l
l
l
l
l
l
5000
4444
3750
3333
2666
2222
1600
1000
tCONV
Conversion Time
N Channels Enabled, 1 ≤ N ≤ 8
l
450•N
500•N
tACQ
Acquisition Time
(tACQ = tCYC – tCONV – tBUSYLH)
8 Channels Enabled, fSMPL = 200ksps
7 Channels Enabled, fSMPL = 225ksps
6 Channels Enabled, fSMPL = 266ksps
5 Channels Enabled, fSMPL = 300ksps
4 Channels Enabled, fSMPL = 375ksps
3 Channels Enabled, fSMPL = 450ksps
2 Channels Enabled, fSMPL = 625ksps
1 Channel Enabled, fSMPL = 1000ksps
l
l
l
l
l
l
l
l
570
564
420
553
436
542
470
420
980
924
730
813
646
702
580
480
MAX
UNITS
200
225
266
300
375
450
625
1000
ksps
ksps
ksps
ksps
ksps
ksps
ksps
ksps
ns
ns
ns
ns
ns
ns
ns
ns
550•N
ns
ns
ns
ns
ns
ns
ns
ns
ns
234818fa
6
For more information www.linear.com/LTC2348-18
LTC2348-18
ADC Timing Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. (Note 9)
SYMBOL
PARAMETER
CONDITIONS
tCNVH
CNV High Time
l
40
ns
tCNVL
CNV Low Time
l
420
ns
tBUSYLH
CNV↑ to BUSY Delay
tQUIET
Digital I/O Quiet Time from CNV↑
l
20
ns
tPDH
PD High Time
l
40
ns
tPDL
PD Low Time
l
40
ns
tWAKE
REFBUF Wake-Up Time
CL = 25pF
MIN
TYP
MAX
30
l
CREFBUF = 47μF, CREFIN = 0.1μF
200
UNITS
ns
ms
CMOS I/O Mode
tSCKI
SCKI Period
l
10
ns
tSCKIH
SCKI High Time
l
4
ns
tSCKIL
SCKI Low Time
l
4
ns
tSSDISCKI
SDI Setup Time from SCKI↑
(Note 17)
l
2
ns
tHSDISCKI
SDI Hold Time from SCKI↑
(Note 17)
l
1
ns
tDSDOSCKI
SDO Data Valid Delay from SCKI↑
CL = 25pF (Note 17)
l
tHSDOSCKI
SDO Remains Valid Delay from SCKI↑
CL = 25pF (Note 17)
l
1.5
tSKEW
SDO to SCKO Skew
(Note 17)
l
–1
tDSDOBUSYL SDO Data Valid Delay from BUSY↓
CL = 25pF (Note 17)
l
0
tEN
Bus Enable Time After CS↓
(Note 17)
l
15
ns
Bus Relinquish Time After CS↑
(Note 17)
l
15
ns
tDIS
(Notes 17, 18)
7.5
ns
1
ns
ns
0
ns
LVDS I/O Mode
tSCKI
SCKI Period
(Note 19)
l
4
ns
tSCKIH
SCKI High Time
(Note 19)
l
1.5
ns
tSCKIL
SCKI Low Time
(Note 19)
l
1.5
ns
tSSDISCKI
SDI Setup Time from SCKI
(Notes 11, 19)
l
1.2
ns
tHSDISCKI
SDI Hold Time from SCKI
(Notes 11, 19)
l
–0.2
ns
tDSDOSCKI
SDO Data Valid Delay from SCKI
(Notes 11, 19)
l
tHSDOSCKI
SDO Remains Valid Delay from SCKI
(Notes 11, 19)
l
1
tSKEW
SDO to SCKO Skew
(Note 11)
l
–0.4
(Note 11)
l
0
tDSDOBUSYL SDO Data Valid Delay from BUSY↓
6
ns
ns
0
0.4
ns
ns
tEN
Bus Enable Time After CS↓
l
50
ns
tDIS
Bus Relinquish Time After CS↑
l
15
ns
234818fa
For more information www.linear.com/LTC2348-18
7
LTC2348-18
ADC Timing Characteristics
Note 1: Stresses beyond those listed under Absolute Maximum Ratings
may cause permanent damage to the device. Exposure to any Absolute
Maximum Rating condition for extended periods may affect device
reliability and lifetime.
Note 2: All voltage values are with respect to ground.
Note 3: VDDLBYP is the output of an internal voltage regulator, and should
only be connected to a 2.2μF ceramic capacitor to bypass the pin to GND,
as described in the Pin Functions section. Do not connect this pin to any
external circuitry.
Note 4: When these pin voltages are taken below VEE or above VCC, they
will be clamped by internal diodes. This product can handle input currents
of up to 100mA below VEE or above VCC without latch-up.
Note 5: When these pin voltages are taken below ground or above VDD or
OVDD, they will be clamped by internal diodes. This product can handle
currents of up to 100mA below ground or above VDD or OVDD without
latch-up.
Note 6: –16.5V ≤ VEE ≤ 0V, 0V ≤ VCC ≤ 38V, 10V ≤ (VCC – VEE) ≤ 38V,
VDD = 5V, unless otherwise specified.
Note 7: Recommended operating conditions.
Note 8: Exceeding these limits on any channel may corrupt conversion
results on other channels. Refer to Absolute Maximum Ratings section for
pin voltage limits related to device reliability.
Note 9: VCC = 15V, VEE = –15V, VDD = 5V, OVDD = 2.5V, fSMPL = 200ksps,
internal reference and buffer, true bipolar input signal drive in bipolar
SoftSpan ranges, unipolar signal drive in unipolar SoftSpan ranges, unless
otherwise specified.
Note 10: Integral nonlinearity is defined as the deviation of a code from a
straight line passing through the actual endpoints of the transfer curve.
The deviation is measured from the center of the quantization band.
Note 11: Guaranteed by design, not subject to test.
Note 12: For bipolar SoftSpan ranges 7, 6, 3, and 2, zero-scale error
is the offset voltage measured from –0.5LSB when the output code
flickers between 00 0000 0000 0000 0000 and 11 1111 1111 1111 1111.
Full-scale error for these SoftSpan ranges is the worst-case deviation of
the first and last code transitions from ideal and includes the effect of
offset error. For unipolar SoftSpan ranges 5, 4, and 1, zero-scale error is
the offset voltage measured from 0.5LSB when the output code flickers
between 00 0000 0000 0000 0000 and 00 0000 0000 0000 0001. Fullscale error for these SoftSpan ranges is the worst-case deviation of the
last code transition from ideal and includes the effect of offset error.
Note 13: All specifications in dB are referred to a full-scale input in the
relevant SoftSpan input range, except for crosstalk, which is referred to
the crosstalk injection signal amplitude.
Note 14: Temperature coefficient is calculated by dividing the maximum
change in output voltage by the specified temperature range.
Note 15: When REFBUF is overdriven, the internal reference buffer must
be disabled by setting REFIN = 0V.
Note 16: IREFBUF varies proportionally with sample rate and the number of
active channels.
Note 17: Parameter tested and guaranteed at OVDD = 1.71V, OVDD = 2.5V,
and OVDD = 5.25V.
Note 18: A tSCKI period of 10ns minimum allows a shift clock frequency of
up to 100MHz for rising edge capture.
Note 19: VICM = 1.2V, VID = 350mV for LVDS differential input pairs.
CMOS Timings
0.8 • OVDD
tWIDTH
0.2 • OVDD
tDELAY
tDELAY
0.8 • OVDD
0.8 • OVDD
0.2 • OVDD
0.2 • OVDD
50%
50%
234818 F01
LVDS Timings (Differential)
+200mV
tWIDTH
–200mV
tDELAY
tDELAY
+200mV
+200mV
–200mV
–200mV
0V
0V
234818 F01b
Figure 1. Voltage Levels for Timing Specifications
234818fa
8
For more information www.linear.com/LTC2348-18
LTC2348-18
Typical Performance Characteristics
TA = 25°C, VCC = +15V, VEE = –15V, VDD = 5V,
OVDD = 2.5V, Internal Reference and Buffer (VREFBUF = 4.096V), fSMPL = 200ksps, unless otherwise noted.
Integral Nonlinearity
vs Output Code and Channel
2.0
Integral Nonlinearity
vs Output Code and Channel
2.0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
ALL CHANNELS
1.5
–0.5
0.5
0
–0.5
–1.0
–1.0
–1.5
–1.5
–2.0
–131072
–2.0
–131072
0
65536
OUTPUT CODE
131072
2.0
–65536
0.5
0
65536
OUTPUT CODE
–0.5
131072
0
–0.5
2.0
0.5
0
–0.5
0
–0.5
–1.5
–1.5
–1.5
–2.0
–131072
131072
–65536
0
65536
OUTPUT CODE
234818 G04
1.0
DC Histogram (Zero-Scale)
90000
±10.24V RANGE
COUNTS
0.5
–0.5
–1.0
ARBITRARY DRIVE
IN+/IN– COMMON MODE
SWEPT –10.24V to 10.24V
–1.5
–2.0
–131072
–65536
0
65536
OUTPUT CODE
131072
234818 G07
65536
131072
196608
OUTPUT CODE
70000
60000
60000
50000
40000
50000
40000
30000
30000
20000
20000
10000
10000
–6
–4
–2
0
CODE
2
4
±10.24V RANGE
σ = 1.4
80000
70000
0
262144
DC Histogram (Near Full-Scale)
Full–Scale)
90000
±10.24V RANGE
σ = 1.3
80000
TRUE BIPOLAR DRIVE (IN– = 0V)
0
0
234818 G06
COUNTS
1.5
0V TO 10.24V AND 0V TO 10V
RANGES
234818 G05
Integral Nonlinearity
vs Output Code
2.0
–2.0
131072
0V TO 5.12V
RANGE
0.5
–1.0
0
65536
OUTPUT CODE
262144
1.0
–1.0
–65536
131072
196608
OUTPUT CODE
UNIPOLAR DRIVE (IN– = 0V)
ONE CHANNEL
1.5
–1.0
–2.0
–131072
65536
Integral Nonlinearity
vs Output Code and Range
±10.24V, ±10V,
±5.12V, AND ±5V
RANGES
1.0
±5.12V AND ±5V
RANGES
0
234818 G03
FULLY DIFFERENTIAL DRIVE (IN– = –IN+)
ONE CHANNEL
1.5
INL ERROR (LSB)
INL ERROR (LSB)
±10.24V AND ±10V
RANGES
–0.2
Integral Nonlinearity
vs Output Code and Range
TRUE BIPOLAR DRIVE (IN– = 0V)
ONE CHANNEL
1.5
0.0
–0.1
234818 G02
Integral Nonlinearity
vs Output Code and Range
2.0
0.1
–0.4
INL ERROR (LSB)
–65536
0.2
–0.3
234818 G01
INL ERROR (LSB)
0.3
DNL ERROR (LSB)
INL ERROR (LSB)
INL ERROR (LSB)
0
ALL RANGES
ALL CHANNELS
0.4
1.0
0.5
1.0
0.5
±10.24V RANGE
FULLY DIFFERENTIAL DRIVE (IN– = –IN+)
ALL CHANNELS
1.5
1.0
Differential Nonlinearity
vs Output Code and Channel
6
234818 G08
0
131056
131059
131062
CODE
131065
131068
234818 G09
234818fa
For more information www.linear.com/LTC2348-18
9
LTC2348-18
Typical Performance Characteristics
TA = 25°C, VCC = +15V, VEE = –15V, VDD = 5V,
OVDD = 2.5V, Internal Reference and Buffer (VREFBUF = 4.096V), fSMPL = 200ksps, unless otherwise noted.
32k
FFT fSMPLFFT
=
32k Point
Point Two-Tone
Arbitrary Two-Tone
+ = –7dBFS
32k Point FFT fSMPL = 200kHz,
= 200kHz,
200kHz,
IN+ = –7dBFS
2kHz Sine,
32k Point FFT fSMPL
= 200kHz,
SMPL = 200kHz,
fSMPL
= 200kHz,
IN
2kHz
– = –7dBFS
fIN = 2kHz
fIN = 2kHz
= 2kHz
IN
3.1kHz3.1kHz
Sine Sine
= 2kHz
Sine,
IN– = –7dBFS
–40
SNR = 96.7dB
THD = –109dB
SINAD = 96.5dB
SFDR = 110dB
–60
–80
–100
–120
SNR = 96.7dB
THD = –119dB
SINAD = 96.6dB
SFDR = 120dB
–40
–60
–80
–100
–120
–80
–100
–160
–160
40
60
FREQUENCY (kHz)
80
–180
100
0
20
40
60
FREQUENCY (kHz)
80
32k Point FFT fSMPL = 200kHz,
fIN = 2kHz
= 2kHz
100.0
±5.12V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
–40
SNR = 93.2dB
THD = –113dB
SINAD = 93.2dB
SFDR = 114dB
–60
–80
–100
–120
SNR, SINAD vs VREFBUF,
fSNR,
SINAD vs VREFBUF, fIN = 2kHz
IN = 2kHz
SNR
96.0
SINAD
94.0
92.0
40
60
FREQUENCY (kHz)
80
90.0
2.5
100
–70.0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
–80.0
SNR
94.0
90.0
–115.0
SINAD
86.0
3
3.5
4
4.5
REFBUF VOLTAGE (V)
5
–130.0
2.5
3
3.5
4
4.5
REFBUF VOLTAGE (V)
–110.0
THD
–130.0
100
2ND
1k
10k
FREQUENCY (Hz)
–40
–60
100k
234818 G17
±10.24V RANGE
2VP-P FULLY DIFFERENTIAL DRIVE
–14.5V ≤ VCM ≤ 10.5V
–80
–100
–120
–140
3RD
5
THD, Harmonics vs Input
Common Mode, fIN = 2kHz
–20
–100.0
78.0
100
234818 G16
0
–90.0
–120.0
100k
3RD
234818 G15
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
82.0
1k
10k
FREQUENCY (Hz)
2ND
–120.0
THD, Harmonics
vs Input Frequency
THD, HARMONICS (dBFS)
98.0
THD
234818 G14
SNR, SINAD
vs Input Frequency
100
–125.0
234818 G13
102.0
80
±2.5 • VREFBUF RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
–110.0
THD, HARMONICS (dBFS)
20
40
60
FREQUENCY (kHz)
THD, Harmonics vs VREFBUF,
fIN = 2kHz
= 2kHz
–105.0
–160
0
–100.0
±2.5 • VREFBUF RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
98.0
–140
–180
20
234818 G12
THD, HARMONICS (dBFS)
–20
0
234818 G11
SNR, SINAD (dBFS)
0
–180
100
6.2kHz
–120
–140
234818 G10
SNR, SINAD (dBFS)
–60
–160
20
SFDR = 119dB
SNR = 96.7dB
–40
–140
0
±10.24V RANGE
ARBITRARY DRIVE
–20
–140
–180
AMPLITUDE (dBFS)
0
±10.24V RANGE
FULLY DIFFERENTIAL DRIVE (IN– = –IN+)
–20
AMPLITUDE (dBFS)
–20
AMPLITUDE (dBFS)
0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
AMPLITUDE (dBFS)
0
–160
–15
THD
3RD
2ND
–10
–5
0
5
10
INPUT COMMON MODE (V)
15
234818 G18
234818fa
10
For more information www.linear.com/LTC2348-18
LTC2348-18
Typical Performance Characteristics
TA = 25°C, VCC = +15V, VEE = –15V, VDD = 5V,
OVDD = 2.5V, Internal Reference and Buffer (VREFBUF = 4.096V), fSMPL = 200ksps, unless otherwise noted.
135.0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
CMRR (dB)
SNR, SINAD (dBFS)
97.0
SINAD
105.0
95.0
85.0
96.5
–30
–20
–10
INPUT LEVEL (dBFS)
65.0
0
SINAD
96.0
95.5
95.0
10
100
1k
10k
FREQUENCY (Hz)
100k
94.5
–110.0
2ND
–115.0
3RD
–125.0
–55 –35 –15
5 25 45 65 85 105 125
TEMPERATURE (°C)
CH7
10
2.0
0.075
0.000
–0.025
–0.050
5 25 45 65 85 105 125
TEMPERATURE (°C)
1.0
0.5
0
–0.5
–0.025
–0.050
–0.100
–55 –35 –15
234818 G25
MAX INL
MAX DNL
MIN DNL
–1.0 MIN INL
–2.0
–55 –35 –15
5 25 45 65 85 105 125
TEMPERATURE (°C)
234818 G24
5
±10.24V RANGE
ALL CHANNELS
0.000
–0.100
–55 –35 –15
5 25 45 65 85 105 125
TEMPERATURE (°C)
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
Zero-Scale Error vs
Temperature
and vs
Channel
Zero-Scale Error
Temperature
0.025
–0.075
1M
–1.5
0.050
–0.075
100k
INL/DNL
INL,
DNLvs
vsTemperature
Temperature
1.5
4
ZERO-SCALE ERROR (LSB)
0.100
±10.24V RANGE
ALL CHANNELS
0.025
1k
10k
FREQUENCY (Hz)
234818 G21
Full-Scale Error
Negative Full–Scale
Error vs
Temperature
and Channel
vs
Temperature
0.050
100
234818 G23
FULL–SCALE ERROR (%)
FULL–SCALE ERROR (%)
–135.0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
THD
Full-Scale Error
Positive Full–Scale
Error vs
Temperature
and Channel
vs
Temperature
0.075
–115.0
THD, Harmonics vs Temperature,
fIN = 2kHz
2kHz
234818 G22
0.100
1M
–105.0
–120.0
94.0
–55 –35 –15
CH2
–110.0
INL, DNL ERROR (LSB)
SNR
96.5
–105.0
–130.0
–100.0
THD, HARMONICS (dBFS)
97.0
SNR, SINAD (dBFS)
–95.0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
97.5
–100.0
234818 G20
SNR, SINAD vs Temperature,
fIN = 2kHz
= 2kHz
CH1
–125.0
234818 G19
98.0
–90.0
–120.0
75.0
96.0
–40
±10.24V RANGE
IN0+ = 0V
IN0– = 18VP-P SINE
ALL CHANNELS CONVERTING
–85.0
–95.0
115.0
SNR
–80.0
±10.24V RANGE
P-P SINE
ALL CHANNELS
IN+ = IN– = 18V
125.0
97.5
Crosstalk vs Input Frequency
and
Channel
Crosstalk
vs Input Frequency
CROSSTALK (dB)
98.0
CMRR vs Input Frequency
and
Channel
CMRR
vs Input Frequency
SNR, SINAD vs Input Level,
fIN = 2kHz
= 2kHz
±10.24V RANGE
ALL CHANNELS
3
2
1
0
–1
–2
–3
–4
5 25 45 65 85 105 125
TEMPERATURE (°C)
234818 G26
–5
–55 –35 –15
5 25 45 65 85 105 125
TEMPERATURE (°C)
234818 G27
234818fa
For more information www.linear.com/LTC2348-18
11
LTC2348-18
Typical Performance Characteristics
TA = 25°C, VCC = +15V, VEE = –15V, VDD = 5V,
OVDD = 2.5V, Internal Reference and Buffer (VREFBUF = 4.096V), fSMPL = 200ksps, unless otherwise noted.
Power-Down Current
vs Temperature
Supply Current vs Temperature
POWER-DOWN CURRENT (µA)
IVDD
16
14
12
10
8
6
4
IVCC
2
0
IOVDD
IVEE
–2
–4
–55 –35 –15
IVDD
100
130
10
1
–0.5
–1.0
VCC = 21.5V, VEE = –16.5V
VCM = –16.5V to 17.5V
–1.5
–2.0
–16.5
0
17.5
INPUT COMMON MODE (V)
2.049
2.048
2.047
40
WITH NAP MODE
t CNVL = 420ns
20
0
0
200
400
600
800
SAMPLING FREQUENCY (kHz)
12
10
8
6
4
–4
5 25 45 65 85 105 125
TEMPERATURE (°C)
234818 G34
IOVDD
IVEE
0
40
80
120
160
SAMPLING FREQUENCY (kHz)
250
65536
0
–32768
200
234818 G33
Step Response
(Fine Settling)
±10.24V RANGE
IN+ = 200.0061kHz SQUARE WAVE
IN– = 0V
–65536
–98304
1000
IVCC
2
–2
32768
1M
IVDD
0
2.046
98304
60
100k
WITH NAP MODE
t CNVL = 1µs
14
131072
80
1k
10k
FREQUENCY (Hz)
234818 G32
N=1
100
100
Supply Current vs Sampling Rate
16
Step Response
(Large-Signal Settling)
N=8
120
18
2.050
2.045
–55 –35 –15
34
OUTPUT CODE (LSB)
POWER DISSIPATION (mW)
N=4
140
N=2
10
234818 G30
15 UNITS
Power Dissipation vs Sampling
Rate, N-Channels Enabled
160
50
5 25 45 65 85 105 125
TEMPERATURE (°C)
234818 G31
180
VDD
60
IVCC
SUPPLY CURRENT (mA)
INTERNAL REFERENCE OUTPUT (V)
OFFSET ERROR (LSB)
VCC = 38V, VEE = 0V
VCM = 0V to 34V
0
70
IOVDD
–IVEE
2.051
±10.24V RANGE
0.5
90
Internal Reference Output
vs Temperature
1.0
VEE
100
234818 G29
Offset Error
vs Input Common Mode
1.5
110
80
0.1
234818 G28
2.0
VCC
120
0.01
–55 –35 –15
5 25 45 65 85 105 125
TEMPERATURE (°C)
IN+ = IN– = 0V
OVDD
140
–131072
–100 0 100 200 300 400 500 600 700 800 900
SETTLING TIME (ns)
234818 G35
DEVIATION FROM FINAL VALUE (LSB)
18
SUPPLY CURRENT (mA)
PSRR vs Frequency
150
1000
PSRR (dB)
20
200
150
100
±10.24V RANGE
IN+ = 200.0061kHz
SQUARE WAVE
IN– = 0V
50
0
–50
–100
–150
–200
–250
–100 0 100 200 300 400 500 600 700 800 900
SETTLING TIME (ns)
234818 G36
234818fa
12
For more information www.linear.com/LTC2348-18
LTC2348-18
Pin Functions
Pins that are the Same for All Digital I/O Modes
IN0+ to IN7+, IN0− to IN7− (Pins 1, 2, 3, 4, 5, 6, 7, 8, 9,
10, 11, 12, 13, 14, 47, and 48): Positive and Negative
Analog Inputs, Channels 0 to 7. The converter simultaneously samples and digitizes (VIN+ – VIN–) for all channels.
Wide input common mode range (VEE ≤ VCM ≤ VCC – 4V)
and high common mode rejection allow the inputs to accept a wide variety of signal swings. Full-scale input range
is determined by the channel’s SoftSpan configuration.
GND (Pins 15, 18, 20, 30, 41, 44, 46): Ground. Solder
all GND pins to a solid ground plane.
VCC (Pin 16): Positive High Voltage Power Supply. The
range of VCC is 0V to 38V with respect to GND and 10V to
38V with respect to VEE. Bypass VCC to GND close to the
pin with a 0.1μF ceramic capacitor. In applications where
VCC is shorted to GND this capacitor may be omitted.
VEE (Pins 17, 45): Negative High Voltage Power Supply.
The range of VEE is 0V to –16.5V with respect to GND and
–10V to –38V with respect to VCC. Connect Pins 17 and 45
together and bypass the VEE network to GND close to Pin
17 with a 0.1μF ceramic capacitor. In applications where
VEE is shorted to GND this capacitor may be omitted.
REFIN (Pin 19): Bandgap Reference Output/Reference
Buffer Input. An internal bandgap reference nominally
outputs 2.048V on this pin. An internal reference buffer
amplifies VREFIN to create the converter master reference
voltage VREFBUF = 2 • VREFIN on the REFBUF pin. When
using the internal reference, bypass REFIN to GND (Pin
20) close to the pin with a 0.1μF ceramic capacitor to filter
the bandgap output noise. If more accuracy is desired,
overdrive REFIN with an external reference in the range
of 1.25V to 2.2V.
REFBUF (Pin 21): Internal Reference Buffer Output. An
internal reference buffer amplifies VREFIN to create the
converter master reference voltage VREFBUF = 2 • VREFIN on
this pin, nominally 4.096V when using the internal bandgap
reference. Bypass REFBUF to GND (Pin 20) close to the
pin with a 47μF ceramic capacitor. The internal reference
buffer may be disabled by grounding its input at REFIN.
With the buffer disabled, overdrive REFBUF with an external reference voltage in the range of 2.5V to 5V. When
using the internal reference buffer, limit the loading of any
external circuitry connected to REFBUF to less than 10µA.
Using a high input impedance amplifier to buffer VREFBUF
to any external circuits is recommended.
PD (Pin 22): Power Down Input. When this pin is brought
high, the LTC2348-18 is powered down and subsequent
conversion requests are ignored. If this occurs during a
conversion, the device powers down once the conversion
completes. If this pin is brought high twice without an
intervening conversion, an internal global reset is initiated, equivalent to a power-on-reset event. Logic levels
are determined by OVDD.
LVDS/CMOS (Pin 23): I/O Mode Select. Tie this pin to OVDD
to select LVDS I/O mode, or to ground to select CMOS I/O
mode. Logic levels are determined by OVDD.
CNV (Pin 24): Conversion Start Input. A rising edge on
this pin puts the internal sample-and-holds into the hold
mode and initiates a new conversion. CNV is not gated
by CS, allowing conversions to be initiated independent
of the state of the serial I/O bus.
BUSY (Pin 38): Busy Output. The BUSY signal indicates
that a conversion is in progress. This pin transitions lowto-high at the start of each conversion and stays high until
the conversion is complete. Logic levels are determined
by OVDD.
VDDLBYP (Pin 40): Internal 2.5V Regulator Bypass Pin. The
voltage on this pin is generated via an internal regulator
operating off of VDD. This pin must be bypassed to GND
close to the pin with a 2.2μF ceramic capacitor. Do not
connect this pin to any external circuitry.
VDD (Pins 42, 43): 5V Power Supply. The range of VDD
is 4.75V to 5.25V. Connect Pins 42 and 43 together and
bypass the VDD network to GND with a shared 0.1μF
ceramic capacitor close to the pins.
234818fa
For more information www.linear.com/LTC2348-18
13
LTC2348-18
Pin Functions
CMOS I/O Mode
LVDS I/O Mode
SDO0 to SDO7 (Pins 25, 26, 27, 28, 33, 34, 35, and 36):
CMOS Serial Data Outputs, Channels 0 to 7. The most
recent conversion result along with channel configuration
information is clocked out onto the SDO pins on each rising edge of SCKI. Output data formatting is described in
the Digital Interface section. Leave unused SDO outputs
unconnected. Logic levels are determined by OVDD.
SDO0, SDO7, SDI (Pins 25, 36, and 37): CMOS Serial
Data I/O. In LVDS I/O mode, these pins are Hi-Z.
SCKI (Pin 29): CMOS Serial Clock Input. Drive SCKI with
the serial I/O clock. SCKI rising edges latch serial data in
on SDI and clock serial data out on SDO0 to SDO7. For
standard SPI bus operation, capture output data at the
receiver on rising edges of SCKI. SCKI is allowed to idle
either high or low. Logic levels are determined by OVDD.
OVDD (Pin 31): I/O Interface Power Supply. In CMOS I/O
mode, the range of OVDD is 1.71V to 5.25V. Bypass OVDD
to GND (Pin 30) close to the pin with a 0.1μF ceramic
capacitor.
SCKO (Pin 32): CMOS Serial Clock Output. SCKI rising
edges trigger transitions on SCKO that are skew-matched
to the serial output data streams on SDO0 to SDO7. The
resulting SCKO frequency is half that of SCKI. Rising and
falling edges of SCKO may be used to capture SDO data at
the receiver (FPGA) in double data rate (DDR) fashion. For
standard SPI bus operation, SCKO is not used and should
be left unconnected. SCKO is forced low at the falling edge
of BUSY. Logic levels are determined by OVDD.
SDI (Pin 37): CMOS Serial Data Input. Drive this pin with the
desired 24-bit SoftSpan configuration word (see Table 1a),
latched on the rising edges of SCKI. If all channels will be
configured to operate only in SoftSpan 7, tie SDI to OVDD.
Logic levels are determined by OVDD.
CS (Pin 39): Chip Select Input. The serial data I/O bus is
enabled when CS is low and is disabled and Hi-Z when
CS is high. CS also gates the external shift clock, SCKI.
Logic levels are determined by OVDD.
SDI+, SDI– (Pins 26 and 27): LVDS Positive and Negative
Serial Data Input. Differentially drive SDI+/SDI– with the
desired 24-bit SoftSpan configuration word (see Table
1a), latched on both the rising and falling edges of SCKI+/
SCKI–. The SDI+/SDI– input pair is internally terminated
with a 100Ω differential resistor when CS is low.
SCKI+, SCKI– (Pins 28 and 29): LVDS Positive and Negative
Serial Clock Input. Differentially drive SCKI+/SCKI– with
the serial I/O clock. SCKI+/SCKI– rising and falling edges
latch serial data in on SDI+/SDI– and clock serial data out
on SDO+/SDO–. Idle SCKI+/SCKI– low, including when
transitioning CS. The SCKI+/SCKI– input pair is internally
terminated with a 100Ω differential resistor when CS = 0.
OVDD (Pin 31): I/O Interface Power Supply. In LVDS I/O
mode, the range of OVDD is 2.375V to 5.25V. Bypass OVDD
to GND (Pin 30) close to the pin with a 0.1μF ceramic
capacitor.
SCKO+, SCKO– (Pins 32 and 33): LVDS Positive and
Negative Serial Clock Output. SCKO+/SCKO– outputs a
copy of the input serial I/O clock received on SCKI+/SCKI–,
skew-matched with the serial output data stream on SDO+/
SDO–. Use the rising and falling edges of SCKO+/SCKO–
to capture SDO+/SDO– data at the receiver (FPGA). The
SCKO+/SCKO– output pair must be differentially terminated
with a 100Ω resistor at the receiver (FPGA).
SDO+, SDO– (Pins 34 and 35): LVDS Positive and Negative Serial Data Output. The most recent conversion result
along with channel configuration information is clocked
out onto SDO+/SDO– on both rising and falling edges of
SCKI+/SCKI–, beginning with channel 0. The SDO+/SDO–
output pair must be differentially terminated with a 100Ω
resistor at the receiver (FPGA).
CS (Pin 39): Chip Select Input. The serial data I/O bus is
enabled when CS is low, and is disabled and Hi-Z when
CS is high. CS also gates the external shift clock, SCKI+/
SCKI–. The internal 100Ω differential termination resistors
on the SCKI+/SCKI– and SDI+/SDI– input pairs are disabled
when CS is high. Logic levels are determined by OVDD.
234818fa
14
For more information www.linear.com/LTC2348-18
LTC2348-18
Configuration Tables
Table 1a. SoftSpan Configuration Table. Use This Table with Table 1b to Choose Independent Binary SoftSpan Codes SS[2:0] for Each
Channel Based on Desired Analog Input Range. Combine SoftSpan Codes to Form 24-Bit SoftSpan Configuration Word S[23:0]. Use
Serial Interface to Write SoftSpan Configuration Word to LTC2348-18, as shown in Figure 19
BINARY SoftSpan CODE
SS[2:0]
ANALOG INPUT RANGE
FULL SCALE RANGE
BINARY FORMAT OF
CONVERSION RESULT
111
110
101
100
011
010
001
000
±2.5 • VREFBUF
±2.5 • VREFBUF/1.024
0V to 2.5 • VREFBUF
0V to 2.5 • VREFBUF/1.024
±1.25 • VREFBUF
±1.25 • VREFBUF/1.024
0V to 1.25 • VREFBUF
Channel Disabled
5 • VREFBUF
5 • VREFBUF/1.024
2.5 • VREFBUF
2.5 • VREFBUF/1.024
2.5 • VREFBUF
2.5 • VREFBUF/1.024
1.25 • VREFBUF
Channel Disabled
Two’s Complement
Two’s Complement
Straight Binary
Straight Binary
Two’s Complement
Two’s Complement
Straight Binary
All Zeros
Table 1b. Reference Configuration Table. The LTC2348-18 Supports Three Reference Configurations. Analog Input Range Scales with
the Converter Master Reference Voltage, VREFBUF
REFERENCE CONFIGURATION
Internal Reference with
Internal Buffer
VREFIN
VREFBUF
2.048V
BINARY SoftSpan CODE
SS[2:0]
ANALOG INPUT RANGE
111
±10.24V
110
±10V
101
0V to 10.24V
100
0V to 10V
011
±5.12V
4.096V
1.25V
(Min Value)
2.5V
External Reference with
Internal Buffer
(REFIN Pin Externally
Overdriven)
2.2V
(Max Value)
4.4V
010
±5V
001
0V to 5.12V
111
±6.25V
110
±6.104V
101
0V to 6.25V
100
0V to 6.104V
011
±3.125V
010
±3.052V
001
0V to 3.125V
111
±11V
110
±10.742V
101
0V to 11V
100
0V to 10.742V
011
±5.5V
010
±5.371V
001
0V to 5.5V
234818fa
For more information www.linear.com/LTC2348-18
15
LTC2348-18
Configuration Tables
Table 1b. Reference Configuration Table (Continued). The LTC2348-18 Supports Three Reference Configurations. Analog Input Range
Scales with the Converter Master Reference Voltage, VREFBUF
REFERENCE CONFIGURATION
VREFIN
0V
VREFBUF
BINARY SoftSpan CODE
SS[2:0]
2.5V
(Min Value)
External Reference
Unbuffered
(REFBUF Pin
Externally Overdriven,
REFIN Pin Grounded)
0V
5V
(Max Value)
ANALOG INPUT RANGE
111
±6.25V
110
±6.104V
101
0V to 6.25V
100
0V to 6.104V
011
±3.125V
010
±3.052V
001
0V to 3.125V
111
±12.5V
110
±12.207V
101
0V to 12.5V
100
0V to 12.207V
011
±6.25V
010
±6.104V
001
0V to 6.25V
234818fa
16
For more information www.linear.com/LTC2348-18
LTC2348-18
Functional Block Diagram
CMOS I/O Mode
IN0+
IN0–
VCC
VDD
VDDLBYP
S/H
2.5V
REGULATOR
+
IN1
SDO0
S/H
• • •
IN1–
OVDD
LTC2348-18
IN2+
S/H
IN3+
IN3–
S/H
IN4+
IN4–
S/H
IN5+
IN5–
S/H
18-BIT
SAR ADC
8-CHANNEL MULTIPLEXER
IN2–
SCKO
SDI
CS
S/H
2.048V
REFERENCE
IN7+
IN7–
SDO7
CMOS
SERIAL
I/O
INTERFACE
SCKI
IN6+
IN6–
18 BITS
S/H
VEE
20k
GND
REFERENCE
BUFFER
2×
REFIN
REFBUF
CONTROL
LOGIC
BUSY
CNV PD
LVDS/CMOS
234818 BD01
LVDS I/O Mode
IN0+
IN0–
VCC
VDD
VDDLBYP
S/H
SDO+
2.5V
REGULATOR
IN1+
IN1–
OVDD
LTC2348-18
SDO–
S/H
SCKO+
IN2+
S/H
IN3+
IN3–
S/H
IN4+
IN4–
S/H
IN5+
IN5–
S/H
18-BIT
SAR ADC
8-CHANNEL MULTIPLEXER
IN2–
SDI+
SDI–
SCKI–
CS
S/H
2.048V
REFERENCE
IN7+
IN7–
SCKO–
SCKI+
IN6+
IN6–
18 BITS
LVDS
SERIAL
I/O
INTERFACE
S/H
VEE
GND
20k
REFERENCE
BUFFER
2×
REFIN
REFBUF
CONTROL
LOGIC
BUSY
CNV PD
LVDS/CMOS
234818 BD02
234818fa
For more information www.linear.com/LTC2348-18
17
LTC2348-18
Timing Diagram
CMOS I/O Mode
CS = PD = 0
SAMPLE N
SAMPLE N + 1
CNV
BUSY
CONVERT
ACQUIRE
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
SCKI
SDI
DON’T CARE
S23 S22 S21 S20 S19 S18 S17 S16 S15 S14 S13 S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 S0
SoftSpan CONFIGURATION WORD FOR CONVERSION N + 1
SCKO
DON’T CARE
SDO0
D17 D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 C2 C1 C0 SS2 SS1 SS0 D17
• • •
CONVERSION RESULT
CHANNEL ID SoftSpan
CONVERSION RESULT
CHANNEL 0
CONVERSION N
SDO7
CHANNEL 1
CONVERSION N
D17 D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 C2 C1 C0 SS2 SS1 SS0 D17
DON’T CARE
CONVERSION RESULT
CHANNEL ID SoftSpan
CHANNEL 7
CONVERSION N
CONVERSION RESULT
CHANNEL 0
CONVERSION N
234818 TD01
LVDS I/O Mode
CS = PD = 0
SAMPLE
N+1
SAMPLE N
CNV
(CMOS)
BUSY
(CMOS)
CONVERT
SCKI
(LVDS)
SDI
DON’T CARE
(LVDS)
• • •
ACQUIRE
1
2
3
4
5
6
7
8
• • •
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
186 187 188 189 190 191 192
• • •
S23 S22 S21 S20 S19 S18 S17 S16 S15 S14 S13 S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 S0
• • •
SoftSpan CONFIGURATION WORD FOR CONVERSION N + 1
SCKO
(LVDS)
• • •
SDO
(LVDS) DON’T CARE D17 D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 C2 C1 C0 SS2 SS1 SS0 D17 D16 D15 • • • D0 C2 C1 C0 SS2 SS1 SS0 D17
CONVERSION RESULT
CHANNEL ID SoftSpan
CHANNEL 0
CONVERSION N
CHANNEL 1
CONVERSION N
CHANNEL ID SoftSpan
CHANNEL 7
CONVERSION N
CONVERSION
RESULT
CHANNEL 0
CONVERSION N
234818 TD02
234818fa
18
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
Overview
Converter Operation
The LTC2348-18 is an 18-bit, low noise 8-channel simultaneous sampling successive approximation register
(SAR) ADC with differential, wide common mode range
inputs. The ADC operates from a 5V low voltage supply and
flexible high voltage supplies, nominally ±15V. Using the
integrated low-drift reference and buffer (VREFBUF = 4.096V
nominal), each channel of this SoftSpan ADC can be independently configured on a conversion-by-conversion
basis to accept ±10.24V, 0V to 10.24V, ±5.12V, or 0V to
5.12V signals. The input signal range may be expanded
up to ±12.5V using an external 5V reference. Individual
channels may also be disabled to increase throughput on
the remaining channels.
The LTC2348-18 operates in two phases. During the acquisition phase, the sampling capacitors in each channel’s
sample-and-hold (S/H) circuit connect to their respective
analog input pins and track the differential analog input
voltage (VIN+ – VIN–). A rising edge on the CNV pin transitions all channels’ S/H circuits from track mode to hold
mode, simultaneously sampling the input signals on all
channels and initiating a conversion. During the conversion
phase, each channel’s sampling capacitors are connected,
one channel at a time, to an 18-bit charge redistribution
capacitor D/A converter (CDAC). The CDAC is sequenced
through a successive approximation algorithm, effectively
comparing the sampled input voltage with binary-weighted
fractions of the channel’s SoftSpan full-scale range (e.g.,
VFSR/2, VFSR/4 … VFSR/262144) using a differential
comparator. At the end of this process, the CDAC output
approximates the channel’s sampled analog input. Once
all channels have been converted in this manner, the ADC
control logic prepares the 18-bit digital output codes from
each channel for serial transfer.
The wide input common mode range and high CMRR
(118dB typical, VIN+ = VIN– = 18VP-P 200Hz Sine) of the
LTC2348-18 analog inputs allow the ADC to directly digitize
a variety of signals, simplifying signal chain design. The
absolute common mode input range is determined by
the choice of high voltage supplies, which may be biased
asymmetrically around ground and include the ability for
either the positive or negative supply to be tied directly to
ground. This input signal flexibility, combined with ±3LSB
INL, no missing codes at 18-bits, and 96.7dB SNR, makes
the LTC2348-18 an ideal choice for many high voltage
applications requiring wide dynamic range.
The LTC2348-18 supports pin-selectable SPI CMOS (1.8V
to 5V) and LVDS serial interfaces, enabling it to communicate equally well with legacy microcontrollers and
modern FPGAs. In CMOS mode, applications may employ
between one and eight lanes of serial output data, allowing
the user to optimize bus width and data throughput. The
LTC2348-18 typically dissipates 140mW when converting
eight analog input channels simultaneously at 200ksps per
channel throughput. Optional nap and power down modes
may be employed to further reduce power consumption
during inactive periods.
Transfer Function
The LTC2348-18 digitizes each channel’s full-scale voltage
range into 218 levels. In conjunction with the ADC master
reference voltage, VREFBUF, a channel’s SoftSpan configuration determines its input voltage range, full-scale range,
LSB size, and the binary format of its conversion result, as
shown in Tables 1a and 1b. For example, employing the
internal reference and buffer (VREFBUF = 4.096V nominal),
SoftSpan 7 configures a channel to accept a ±10.24V
bipolar analog input voltage range, which corresponds
to a 20.48V full-scale range with a 78.125μV LSB. Other
SoftSpan configurations and reference voltages may be
employed to convert both larger and smaller bipolar and
unipolar input ranges. Conversion results are output in
two’s complement binary format for all bipolar SoftSpan
ranges, and in straight binary format for all unipolar
SoftSpan ranges. The ideal two’s complement transfer
function is shown in Figure 2, while the ideal straight
binary transfer function is shown in Figure 3.
234818fa
For more information www.linear.com/LTC2348-18
19
LTC2348-18
OUTPUT CODE (TWO’S COMPLEMENT)
Applications Information
pseudo-differential true bipolar, and fully differential,
simplifying signal chain design.
011...111
BIPOLAR
ZERO
011...110
000...001
000...000
111...111
111...110
100...001
FSR = +FS – –FS
1LSB = FSR/262144
100...000
–FSR/2
–1 0V 1
FSR/2 – 1LSB
LSB
LSB
INPUT VOLTAGE (V)
234818 F02
OUTPUT CODE (STRAIGHT BINARY)
Figure 2. LTC2348-18 Two’s Complement Transfer Function
111...111
111...110
100...001
100...000
011...111 UNIPOLAR
ZERO
011...110
000...001
FSR = +FS
1LSB = FSR/262144
000...000
0V
FSR – 1LSB
INPUT VOLTAGE (V)
234818 F03
Figure 3. LTC2348-18 Straight Binary Transfer Function
The wide operating range of the high voltage supplies
offers further input common mode flexibility. As long as
the voltage difference limits of 10V ≤ VCC – VEE ≤ 38V
are observed, VCC and VEE may be independently biased
anywhere within their own individual allowed operating
ranges, including the ability for either of the supplies to be
tied directly to ground. This feature enables the common
mode input range of the LTC2348-18 to be tailored to the
specific application’s requirements.
In all SoftSpan ranges, each channel’s analog inputs can
be modeled by the equivalent circuit shown in Figure 4.
At the start of acquisition, the 40pF sampling capacitors
(CIN) connect to the analog input pins IN+/IN– through the
sampling switches, each of which has approximately 600Ω
(RIN) of on-resistance. The initial voltage on both sampling
capacitors at the start of acquisition is approximately equal
to the sampled common-mode voltage (VIN+ + VIN–)/2
from the prior conversion. The external circuitry connected
to IN+ and IN– must source or sink the charge that flows
through RIN as the sampling capacitors settle from their
initial voltages to the new input pin voltages over the
course of the acquisition interval. During conversion, nap,
and power down modes, the analog inputs draw only a
small leakage current. The diodes at the inputs provide
ESD protection.
VCC
Analog Inputs
Each channel of the LTC2348-18 simultaneously samples
the voltage difference (VIN+ – VIN–) between its analog
input pins over a wide common mode input range while
attenuating unwanted signals common to both input
pins by the common-mode rejection ratio (CMRR) of the
ADC. Wide common mode input range coupled with high
CMRR allows the IN+/IN– analog inputs to swing with
an arbitrary relationship to each other, provided each
pin remains between (VCC – 4V) and VEE. This unique
feature of the LTC2348-18 enables it to accept a wide
variety of signal swings, including traditional classes of
analog input signals such as pseudo-differential unipolar,
RIN
600Ω
IN+
CIN
40pF
VEE
VCC
RIN
600Ω
IN–
CIN
40pF
BIAS
VOLTAGE
234818 F04
VEE
Figure 4. Equivalent Circuit for Differential Analog
Inputs, Single Channel Shown
234818fa
20
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
Bipolar SoftSpan Input Ranges
Input Drive Circuits
For channels configured in SoftSpan ranges 7, 6, 3,
or 2, the LTC2348-18 digitizes the differential analog
input voltage (VIN+ – VIN–) over a bipolar span of
±2.5 • VREFBUF, ±2.5 • VREFBUF/1.024, ±1.25 • VREFBUF, or
±1.25 • VREFBUF/1.024, respectively, as shown in Table
1a. These SoftSpan ranges are useful for digitizing input
signals where IN+ and IN– swing above and below each
other. Traditional examples include fully differential input
signals, where IN+ and IN– are driven 180 degrees out-ofphase with respect to each other centered around a common
mode voltage (VIN+ + VIN–)/2, and pseudo-differential true
bipolar input signals, where IN+ swings above and below
a ground reference level, driven on IN–. Regardless of the
chosen SoftSpan range, the wide common mode input
range and high CMRR of the IN+/IN– analog inputs allow
them to swing with an arbitrary relationship to each other,
provided each pin remains between (VCC – 4V) and VEE.
The output data format for all bipolar SoftSpan ranges is
two’s complement.
The initial voltage on each channel’s sampling capacitors
at the start of acquisition must settle to the new input
pin voltages during the acquisition interval. The external
circuitry connected to IN+ and IN– must source or sink
the charge that flows through RIN as this settling occurs.
The LTC2348-18 sampling network RC time constant of
24ns implies an 18-bit settling time to a full-scale step of
approximately 13 • (RIN • CIN) = 312ns. The impedance and
self-settling of external circuitry connected to the analog
input pins will increase the overall settling time required.
Low impedance sources can directly drive the inputs of
the LTC2348-18 without gain error, but high impedance
sources should be buffered to ensure sufficient settling
during acquisition and to optimize the linearity and distortion performance of the ADC. Settling time is an important
consideration even for DC input signals, as the voltages on
the sampling capacitors will differ from the analog input
pin voltages at the start of acquisition.
Unipolar SoftSpan Input Ranges
For channels configured in SoftSpan ranges 5, 4, or 1, the
LTC2348-18 digitizes the differential analog input voltage
(VIN+ – VIN–) over a unipolar span of 0V to 2.5 • VREFBUF, 0V
to 2.5 • VREFBUF/1.024, or 0V to 1.25 • VREFBUF, respectively,
as shown in Table 1a. These SoftSpan ranges are useful
for digitizing input signals where IN+ remains above IN–.
A traditional example includes pseudo-differential unipolar
input signals, where IN+ swings above a ground reference
level, driven on IN–. Regardless of the chosen SoftSpan
range, the wide common mode input range and high
CMRR of the IN+/IN– analog inputs allow them to swing
with an arbitrary relationship to each other, provided each
pin remains between (VCC – 4V) and VEE. The output data
format for all unipolar SoftSpan ranges is straight binary.
Most applications should use a buffer amplifier to drive the
analog inputs of the LTC2348-18. The amplifier provides
low output impedance, enabling fast settling of the analog
signal during the acquisition phase. It also provides isolation between the signal source and the charge flow at the
analog inputs when entering acquisition.
Input Filtering
The noise and distortion of an input buffer amplifier and
other supporting circuitry must be considered since they
add to the ADC noise and distortion. Noisy input signals
should be filtered prior to the buffer amplifier with a lowbandwidth filter to minimize noise. The simple one-pole
RC lowpass filter shown in Figure 5 is sufficient for many
applications.
At the output of the buffer, a lowpass RC filter network
formed by the 600Ω sampling switch on-resistance (RIN)
and the 40pF sampling capacitance (CIN) limits the input
bandwidth on each channel to 7MHz, which is fast enough
to allow for sufficient transient settling during acquisition
234818fa
For more information www.linear.com/LTC2348-18
21
LTC2348-18
Applications Information
TRUE BIPOLAR
INPUT SIGNAL
LOWPASS
SIGNAL FILTER
160Ω
BUFFER
AMPLIFIER
0V
10nF
IN0+
IN0–
LTC2348-18
BW = 100kHz
ONLY CHANNEL 0 SHOWN FOR CLARITY
234818 F05
Figure 5. True Bipolar Signal Chain with Input Filtering
while simultaneously filtering driver wideband noise. A
buffer amplifier with low noise density should be selected
to minimize SNR degradation over this bandwidth. An
additional filter network may be placed between the buffer output and ADC input to further minimize the noise
contribution of the buffer and reduce disturbances to the
buffer from ADC acquisition transients. A simple one-pole
lowpass RC filter is sufficient for many applications. It is
important that the RC time constant of this filter be small
enough to allow the analog inputs to completely settle to
18-bit resolution within the ADC acquisition time (tACQ),
as insufficient settling can limit INL and THD performance.
Also note that the minimum acquisition time varies with
sampling frequency (fSMPL) and the number of enabled
channels.
High quality capacitors and resistors should be used in
the RC filters since these components can add distortion.
NPO/COG and silver mica type dielectric capacitors have
excellent linearity. Carbon surface mount resistors can
generate distortion from self-heating and from damage
that may occur during soldering. Metal film surface mount
resistors are much less susceptible to both problems.
Buffering Arbitrary and Fully Differential Analog Input
Signals
The wide common mode input range and high CMRR of
the LTC2348-18 allow each channel’s IN+ and IN– pins
to swing with an arbitrary relationship to each other,
provided each pin remains between (VCC – 4V) and VEE.
This unique feature of the LTC2348-18 enables it to accept
a wide variety of signal swings, simplifying signal chain
design. In many applications, connecting a channel’s IN+
and IN– pins directly to the existing signal chain circuitry
will not allow the channel’s sampling network to settle to
18-bit resolution within the ADC acquisition time (tACQ).
In these cases, it is recommended that two unity-gain
buffers be inserted between the signal source and the ADC
input pins, as shown in Figure 6a. Table 2 lists several
amplifier and lowpass filter combinations recommended
for use in this circuit. The LT1469 combines fast settling,
high linearity, and low offset with 5nV/√Hz input-referred
noise density, enabling it to achieve the full ADC data sheet
SNR and THD specifications, as shown in the FFT plots in
Figures 6b to 6e. In applications where slightly degraded
Table 2. Recommended Amplifier and Filter Combinations for the Buffer Circuits in Figures 6a and 9. AC Performance Measured
Using Circuit in Figure 6a, ±10.24V Range
INPUT SIGNAL DRIVE
SNR
(dB)
THD
(dB)
SINAD
(dB)
SFDR
(dB)
1000
FULLY DIFFERENTIAL
96.7
−119
96.6
120
270
FULLY DIFFERENTIAL
96.5
−119
96.4
120
49.9
1000
TRUE BIPOLAR
96.7
−109
96.5
110
½ LT1355
100
270
TRUE BIPOLAR
96.5
−106
96.1
108
½ LT1469
0
0
TRUE BIPOLAR
95.7
−109
95.5
110
AMPLIFIER
RFILT
(Ω)
CFILT
(pF)
½ LT1469
49.9
½ LT1355
100
½ LT1469
½ LT1355
0
0
TRUE BIPOLAR
95.9
−106
95.5
108
½ LT1358
100
270
TRUE BIPOLAR
96.7
−108
96.5
110
234818fa
22
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
+10V
FULLY
DIFFERENTIAL
+5V
ARBITRARY
0V
0V
–10V
–5V
TRUE BIPOLAR
+10V
+10V
0V
0V
–10V
–10V
–
15V
AMPLIFIER
IN–
UNIPOLAR
0.1µF
RFILT
+
IN+
15V
OPTIONAL
LOWPASS FILTERS
VCC
IN0+
IN0–
CFILT
LTC2348-18
+
CFILT
AMPLIFIER
–
VEE REFBUF
RFILT
0.1µF
47µF
0.1µF
–15V
REFIN
–15V
ONLY CHANNEL 0 SHOWN FOR CLARITY
234818 F06a
Figure 6a. Buffering Arbitrary, Fully Differential, True Bipolar, and Unipolar Signals. See Table 2 For
Recommended Amplifier and Filter Combinations
Arbitrary Drive
0
±10.24V RANGE
–20
–80
–100
6.2kHz
–120
–60
–80
–100
–120
–140
–140
–160
–160
–180
0
20
40
60
FREQUENCY (kHz)
80
SNR = 96.7dB
THD = –119dB
SINAD = 96.6dB
SFDR = 120dB
–40
AMPLITUDE (dBFS)
–60
±10.24V RANGE
–20
SFDR = 119dB
SNR = 96.7dB
–40
AMPLITUDE (dBFS)
Fully Differential Drive
0
–180
100
0
20
40
60
FREQUENCY (kHz)
80
234818 F06b
Figure 6b. Two-Tone Test. IN+ = –7dBFS 2kHz Sine, IN– = –7dBFS
3.1kHz Sine, 32k Point FFT, fSMPL = 200ksps. Circuit Shown in
Figure 6a with LT1469 Amplifiers, RFILT = 49.9Ω, CFILT = 1000pF
234818 F06c
Figure 6c. IN+/IN– = –1dBFS 2kHz Fully Differential Sine,
VCM = 0V, 32k Point FFT, fSMPL = 200ksps. Circuit Shown in
Figure 6a with LT1469 Amplifiers, RFILT = 49.9Ω, CFILT = 1000pF
True Bipolar Drive
0
–80
–100
–120
–60
–80
–100
–120
–140
–140
–160
–160
–180
0
20
40
60
FREQUENCY (kHz)
80
SNR = 90.8dB
THD = –111dB
SINAD = 90.7dB
SFDR = 112dB
–40
AMPLITUDE (dBFS)
–60
0V to 10.24V RANGE
–20
SNR = 96.7dB
THD = –109dB
SINAD = 96.5dB
SFDR = 110dB
–40
AMPLITUDE (dBFS)
Unipolar Drive
0
±10.24V RANGE
–20
100
100
234818 F06d
Figure 6d. IN+ = –1dBFS 2kHz True Bipolar Sine, IN– = 0V, 32k
Point FFT, fSMPL = 200ksps. Circuit Shown in Figure 6a with
LT1469 Amplifiers, RFILT = 49.9Ω, CFILT = 1000pF
–180
0
20
40
60
FREQUENCY (kHz)
80
100
234818 F06e
Figure 6e. IN+ = –1dBFS 2kHz Unipolar Sine, IN– = 0V,
32k Point FFT, fSMPL = 200ksps. Circuit Shown in Figure 6a
with LT1469 Amplifiers, RFILT = 49.9Ω, CFILT = 1000pF
234818fa
For more information www.linear.com/LTC2348-18
23
LTC2348-18
Applications Information
SNR and THD performance is acceptable, it is possible
to drive the LTC2348-18 using the lower-power LT1355.
The LT1355 combines fast settling, good linearity, and
moderate offset with 10nV/√Hz input-referred noise density, enabling it to drive the LTC2348-18 with only 0.2dB
SNR loss and 3dB THD loss compared with the LT1469.
As shown in Table 2, both the LT1469 and LT1355 may be
used without a lowpass filter at a loss of ≤1dB SNR due
to increased wideband noise. For sampling frequencies
with minimum acquisition times (tACQ) under 500ns, use
either the LT1469 or LT1355 without lowpass filtering, or
the LT1358 with lowpass filtering, for the best settling,
linearity, and THD performance.
response approaches this ideal, with 119dB of SFDR
limited by the converter's second harmonic distortion
response to the 3.1kHz sine wave on IN–.
The ability of the LTC2348-18 to accept arbitrary signal
swings over a wide input common mode range with high
CMRR can simplify application solutions. In practice,
many sensors produce a differential sensor voltage riding
on top of a large common mode signal. Figure 7a depicts
one way of using the LTC2348-18 to digitize signals of
this type. The amplifier stage provides a differential gain
of approximately 10V/V to the desired sensor signal while
the unwanted common mode signal is attenuated by the
ADC CMRR. The circuit employs the ±5V SoftSpan range of
the ADC. Figure 7b shows measured CMRR performance
of this solution, which is competitive with the best commercially available instrumentation amplifiers. Figure 7c
shows measured AC performance of this solution.
The two-tone test shown in Figure 6b demonstrates the
arbitrary input drive capability of the LTC2348-18. This test
simultaneously drives IN+ with a −7dBFS 2kHz single-ended
sine wave and IN− with a −7dBFS 3.1kHz single-ended sine
wave. Together, these signals sweep the analog inputs
across a wide range of common mode and differential
mode voltage combinations, similar to the more general
arbitrary input signal case. They also have a simple spectral representation. An ideal differential converter with no
common-mode sensitivity will digitize this signal as two
−7dBFS spectral tones, one at each sine wave frequency.
The FFT plot in Figure 6b demonstrates the LTC2348-18
IN+
ARBITRARY
+
–
24V
In Figure 8, another application circuit is shown which
uses two channels of the LTC2348-18 to simultaneously
sense the voltage on and bidirectional current through a
sense resistor over a wide common mode range. In many
applications of this type, the impedance of the external
circuitry is low enough that the ADC sampling network
can fully settle without buffering.
31V
½ LT1124
LOWPASS FILTERS
18pF
0.1µF
2.49k
COMMON MODE
INPUT RANGE
31V
49.9Ω
6.6nF
2.49k
DIFFERENTIAL MODE
INPUT RANGE: ±500mV
IN–
–
+
LTC2348-18
6.6nF
18pF
0V
VCC
IN0+
IN0–
549Ω
49.9Ω
½ LT1124
BW ~ 500kHz
VEE REFBUF
0.1µF
–5V
ONLY CHANNEL 0 SHOWN FOR CLARITY
47µF
REFIN
0.1µF
–5V
234818 F07a
Figure 7a. Digitize Differential Signals Over a Wide Common Mode Range
234818fa
24
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications
Information
CMRR vs Input
Frequency
120
15V
±5V RANGE
110
0.1µF
CMRR (dB)
100
90
80
IN+ = IN– = 24VP–P SINE
OP–AMPS SLEW fIN > 30kHz
70
60
50
RSENSE
10
100
1k
10k
FREQUENCY (Hz)
–15V
V – VS2
ISENSE = S1
RSENSE
SNR = 89.8dB
THD = –116dB
SINAD = 89.8dB
SFDR = 119dB
AMPLITUDE (dBFS)
–60
–80
While the circuit shown in Figure 6a is capable of buffering
single-ended input signals, the circuit shown in Figure 9 is
preferable when the single-ended signal reference level is
inherently low impedance and doesn't require buffering.
This circuit eliminates one driver and lowpass filter, reducing part count, power dissipation, and SNR degradation
due to driver noise. Using the recommended driver and
filter combinations in Table 2, the performance of this
circuit with single-ended input signals is on par with the
performance of the circuit in Figure 6a.
–120
–140
–160
0
10
20
30
FREQUENCY (kHz)
40
50
234818 F07c
Figure 7c. IN+/IN– = 450mV 2kHz Fully Differential Sine,
0V ≤ VCM ≤ 24V, 32k Point FFT, fSMPL = 100ksps. Circuit
Shown in Figure 7a
TRUE BIPOLAR
+10V
15V
IN+
0V
+10V
0V
+
AMPLIFIER
–10V
–
–15V
UNIPOLAR
–10.24V ≤ VS1 ≤ 10.24V
–10.24V ≤ VS2 ≤ 10.24V
Buffering Single-Ended Analog Input Signals
–100
–180
234818 F08
Figure 8. Simultaneously Sense Voltage (CH0) and
Current (CH1) Over a Wide Common Mode Range
±5V RANGE
FULLY DIFFERENTIAL DRIVE (IN– = –IN+)
–40
0.1µF
ONLY CHANNELS 0 AND 1 SHOWN FOR CLARITY
Figure 7b. CMRR vs Input Frequency. Circuit Shown in Figure 7a
–20
47µF
0.1µF
100k
234818 F07b
0
LTC2348-18
IN1+
–
IN1
VEE REFBUF
REFIN
ISENSE
VS2
IN+ = IN– = 1VP–P SINE
VCC
IN0+
IN0–
VS1
15V
OPTIONAL
LOWPASS FILTER
0.1µF
RFILT
IN0+
IN0–
CFILT
VCC
LTC2348-18
IN–
VEE REFBUF
–10V
0.1µF
47µF
REFIN
0.1µF
–15V
ONLY CHANNEL 0 SHOWN FOR CLARITY
234818 F09
Figure 9. Buffering Single-Ended Input Signals. See Table 2 For Recommended Amplifier and Filter Combinations
234818fa
For more information www.linear.com/LTC2348-18
25
LTC2348-18
Applications Information
ADC Reference
LTC2348-18
As shown previously in Table 1b, the LTC2348-18 supports
three reference configurations. The first uses both the internal bandgap reference and reference buffer. The second
externally overdrives the internal reference but retains the
internal buffer, which isolates the external reference from
ADC conversion transients. This configuration is ideal
for sharing a single precision external reference across
multiple ADCs. The third disables the internal buffer and
overdrives the REFBUF pin externally.
Internal Reference with Internal Buffer
The LTC2348-18 has an on-chip, low noise, low drift
(20ppm/°C maximum), temperature compensated bandgap reference that is factory trimmed to 2.048V. The
reference output connects through a 20kΩ resistor to
the REFIN pin, which serves as the input to the on-chip
reference buffer, as shown in Figure 10a. When employing
the internal bandgap reference, the REFIN pin should be
bypassed to GND (Pin 20) close to the pin with a 0.1μF
ceramic capacitor to filter wideband noise. The reference
buffer amplifies VREFIN to create the converter master
reference voltage VREFBUF = 2 • VREFIN on the REFBUF pin,
nominally 4.096V when using the internal bandgap reference. Bypass REFBUF to GND (Pin 20) close to the pin with
at least a 47μF ceramic capacitor (X7R, 10V, 1210 size or
X5R, 10V, 0805 size) to compensate the reference buffer,
absorb transient conversion currents, and minimize noise.
20k
REFIN
0.1µF
REFBUF
BANDGAP
REFERENCE
REFERENCE
BUFFER
6.5k
47µF
6.5k
GND
234818 F10a
Figure 10a. Internal Reference with Internal Buffer Configuration
External Reference with Internal Buffer
If more accuracy and/or lower drift is desired, REFIN can
be easily overdriven by an external reference since 20kΩ
of resistance separates the internal bandgap reference
output from the REFIN pin, as shown in Figure 10b. The
valid range of external reference voltage overdrive on the
REFIN pin is 1.25V to 2.2V, resulting in converter master reference voltages VREFBUF between 2.5V and 4.4V,
respectively. Linear Technology offers a portfolio of high
performance references designed to meet the needs of
many applications. With its small size, low power, and high
accuracy, the LTC6655-2.048 is well suited for use with the
LTC2348-18 when overdriving the internal reference. The
LTC6655-2.048 offers 0.025% (maximum) initial accuracy
LTC2348-18
20k
REFIN
2.7µF
REFBUF
LTC6655-2.048
47µF
BANDGAP
REFERENCE
REFERENCE
BUFFER
6.5k
6.5k
GND
234818 F10b
Figure 10b. External Reference with Internal Buffer Configuration
234818fa
26
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
and 2ppm/°C (maximum) temperature coefficient for high
precision applications. The LTC6655-2.048 is fully specified over the H-grade temperature range, complementing
the extended temperature range of the LTC2348-18 up to
125°C. Bypassing the LTC6655-2.048 with a 2.7µF to 100µF
ceramic capacitor close to the REFIN pin is recommended.
External Reference with Disabled Internal Buffer
The internal reference buffer supports VREFBUF = 4.4V
maximum. By grounding REFIN, the internal buffer may
be disabled allowing REFBUF to be overdriven with an
external reference voltage between 2.5V and 5V, as shown
in Figure 10c. Maximum input signal swing and SNR are
achieved by overdriving REFBUF using an external 5V
reference. The buffer feedback resistors load the REFBUF
pin with 13kΩ even when the reference buffer is disabled.
The LTC6655-5 offers the same small size, accuracy, drift,
and extended temperature range as the LTC6655-2.048,
and achieves a typical SNR of 97.5dB when paired with
the LTC2348-18. Bypass the LTC6655-5 to GND (Pin 20)
close to the REFBUF pin with at least a 47μF ceramic capacitor (X7R, 10V, 1210 size or X5R, 10V, 0805 size) to
absorb transient conversion currents and minimize noise.
The LTC2348-18 converter draws a charge (QCONV) from
the REFBUF pin during each conversion cycle. On short
time scales most of this charge is supplied by the external
REFBUF bypass capacitor, but on longer time scales all of
the charge is supplied by either the reference buffer, or
when the internal reference buffer is disabled, the external
reference. This charge draw corresponds to a DC current
equivalent of IREFBUF = QCONV • fSMPL, which is proportional
LTC2348-18
20k
REFIN
REFBUF
47µF
LTC6655-5
BANDGAP
REFERENCE
REFERENCE
BUFFER
6.5k
6.5k
GND
234818 F10c
Figure 10c. External Reference with Disabled
Internal Buffer Configuration
to sample rate. In applications where a burst of samples
is taken after idling for long periods of time, as shown in
Figure 11, IREFBUF quickly transitions from approximately
0.4mA to 1.5mA (VREFBUF = 5V, fSMPL = 200kHz). This
current step triggers a transient response in the external
reference that must be considered, since any deviation in
VREFBUF affects converter accuracy. If an external reference
is used to overdrive REFBUF, the fast settling LTC6655
family of references is recommended.
Internal Reference Buffer Transient Response
For optimum performance in applications employing burst
sampling, the external reference with internal reference
buffer configuration should be used. The internal reference
buffer incorporates a proprietary design that minimizes
movements in VREFBUF when responding to a burst of
conversions following an idle period. Figure 12 compares
CNV
IDLE
PERIOD
IDLE
PERIOD
234818 F11
Figure 11. CNV Waveform Showing Burst Sampling
234818fa
For more information www.linear.com/LTC2348-18
27
LTC2348-18
Applications Information
the burst conversion response of the LTC2348-18 with an
input near full scale for two reference configurations. The
first configuration employs the internal reference buffer
with REFIN externally overdriven by an LTC6655-2.048,
while the second configuration disables the internal reference buffer and overdrives REFBUF with an external
LTC6655-4.096. In both cases REFBUF is bypassed to
GND with a 47µF ceramic capacitor.
DEVIATION FROM FINAL VALUE (LSB)
20
±10.24V RANGE
IN+ = 10V
IN– = 0V
15
10
EXTERNAL REFERENCE ON REFBUF
5
0
–5
Signal-to-Noise Ratio (SNR)
The signal-to-noise ratio (SNR) is the ratio between the
RMS amplitude of the fundamental input frequency and
the RMS amplitude of all other frequency components
except the first five harmonics and DC. Figure 13 shows
that the LTC2348-18 achieves a typical SNR of 96.7dB in
the ±10.24V range at a 200kHz sampling rate with a true
bipolar 2kHz input signal.
Total Harmonic Distortion (THD)
Total harmonic distortion (THD) is the ratio of the RMS sum
of all harmonics of the input signal to the fundamental itself.
The out-of-band harmonics alias into the frequency band
between DC and half the sampling frequency (fSMPL/2).
THD is expressed as:
INTERNAL REFERENCE BUFFER
0
100
200
300
TIME (µs)
400
500
THD = 20log
V22 + V32 + V42 ...VN2
V1
234818 F12
Dynamic Performance
Fast Fourier transform (FFT) techniques are used to test
the ADC’s frequency response, distortion, and noise at the
rated throughput. By applying a low distortion sine wave
and analyzing the digital output using an FFT algorithm,
the ADC’s spectral content can be examined for frequencies outside the fundamental. The LTC2348-18 provides
guaranteed tested limits for both AC distortion and noise
measurements.
Signal-to-Noise and Distortion Ratio (SINAD)
The signal-to-noise and distortion ratio (SINAD) is the
ratio between the RMS amplitude of the fundamental input
frequency and the RMS amplitude of all other frequency
components at the A/D output. The output is band-limited
to frequencies below half the sampling frequency, excluding DC. Figure 13 shows that the LTC2348-18 achieves a
typical SINAD of 96.5dB in the ±10.24V range at a 200kHz
sampling rate with a true bipolar 2kHz input signal.
where V1 is the RMS amplitude of the fundamental frequency and V2 through VN are the amplitudes of the second
through Nth harmonics, respectively. Figure 13 shows
that the LTC2348-18 achieves a typical THD of –109dB
(N = 6) in the ±10.24V range at a 200kHz sampling rate
with a true bipolar 2kHz input signal.
0
±10.24V RANGE
TRUE BIPOLAR DRIVE (IN– = 0V)
–20
–40
AMPLITUDE (dBFS)
Figure 12. Burst Conversion Response of the LTC2348-18,
fSMPL = 200ksps
SNR = 96.7dB
THD = –109dB
SINAD = 96.5dB
SFDR = 110dB
–60
–80
–100
–120
–140
–160
–180
0
20
40
60
FREQUENCY (kHz)
80
100
234818 F13
Figure 13. 32k Point FFT fSMPL = 200ksps, fIN = 2kHz
234818fa
28
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
Power Considerations
Timing and Control
The LTC2348-18 requires four power supplies: the positive and negative high voltage power supplies (VCC and
VEE), the 5V core power supply (VDD) and the digital input/
output (I/O) interface power supply (OVDD). As long as
the voltage difference limits of 10V ≤ VCC – VEE ≤ 38V
are observed, VCC and VEE may be independently biased
anywhere within their own individual allowed operating
ranges, including the ability for either of the supplies to be
tied directly to ground. This feature enables the common
mode input range of the LTC2348-18 to be tailored to the
specific application’s requirements. The flexible OVDD supply allows the LTC2348-18 to communicate with CMOS
logic operating between 1.8V and 5V, including 2.5V and
3.3V systems. When using LVDS I/O mode, the range of
OVDD is 2.375V to 5.25V.
Power Supply Sequencing
The LTC2348-18 does not have any specific power supply
sequencing requirements. Care should be taken to adhere
to the maximum voltage relationships described in the
Absolute Maximum Ratings section. The LTC2348-18 has
an internal power-on-reset (POR) circuit which resets the
converter on initial power-up and whenever VDD drops
below 2V. Once the supply voltage re-enters the nominal
supply voltage range, the POR reinitializes the ADC. No
conversions should be initiated until at least 10ms after
a POR event to ensure the initialization period has ended.
When employing the internal reference buffer, allow 200ms
for the buffer to power up and recharge the REFBUF bypass
capacitor. Any conversion initiated before these times will
produce invalid results.
CNV Timing
The LTC2348-18 sampling and conversion is controlled by
CNV. A rising edge on CNV transitions all channels’ S/H
circuits from track mode to hold mode, simultaneously
sampling the input signals on all channels and initiating
a conversion. Once a conversion has been started, it
cannot be terminated early except by resetting the ADC,
as discussed in the Reset Timing section. For optimum
performance, drive CNV with a clean, low jitter signal and
avoid transitions on data I/O lines leading up to the rising
edge of CNV. Additionally, to minimize channel-to-channel
crosstalk, avoid high slew rates on the analog inputs for
100ns before and after the rising edge of CNV. Converter
status is indicated by the BUSY output, which transitions
low-to-high at the start of each conversion and stays high
until the conversion is complete. Once CNV is brought high
to begin a conversion, it should be returned low between
40ns and 60ns later or after the falling edge of BUSY to
minimize external disturbances during the internal conversion process. The CNV timing required to take advantage
of the reduced power nap mode of operation is described
in the Nap Mode section.
Internal Conversion Clock
The LTC2348-18 has an internal clock that is trimmed to
achieve a maximum conversion time of 550•N ns with N
channels enabled. With a minimum acquisition time of
570ns when converting eight channels simultaneously,
throughput performance of 200ksps is guaranteed without
any external adjustments.
t CNVL
CNV
tCONV
BUSY
NAP
NAP MODE
tACQ
234818 F14
Figure 14. Nap Mode Timing for the LTC2348-18
234818fa
For more information www.linear.com/LTC2348-18
29
LTC2348-18
Applications Information
Nap Mode
The LTC2348-18 can be placed into nap mode after a conversion has been completed to reduce power consumption
between conversions. In this mode a portion of the device
circuitry is turned off, including circuits associated with
sampling the analog input signals. Nap mode is enabled
by keeping CNV high between conversions, as shown in
Figure 14. To initiate a new conversion after entering nap
mode, bring CNV low and hold for at least 420ns before
bringing it high again. The converter acquisition time (tACQ)
is set by the CNV low time (tCNVL) when using nap mode.
ends based on an internal timer. Reset clears all serial data
output registers and restores the internal SoftSpan configuration register default state of all channels in SoftSpan 7.
If reset is triggered during a conversion, the conversion
is immediately halted. The normal power down behavior
associated with PD going high is not affected by reset. Once
PD is brought low, wait at least 10ms before initiating a
conversion. When employing the internal reference buffer,
allow 200ms for the buffer to power up and recharge the
REFBUF bypass capacitor. Any conversion initiated before
these times will produce invalid results.
Power Dissipation vs Sampling Frequency
When PD is brought high, the LTC2348-18 is powered
down and subsequent conversion requests are ignored. If
this occurs during a conversion, the device powers down
once the conversion completes. In this mode, the device
draws only a small regulator standby current resulting in a
typical power dissipation of 0.36mW. To exit power down
mode, bring the PD pin low and wait at least 10ms before
initiating a conversion. When employing the internal reference buffer, allow 200ms for the buffer to power up and
recharge the REFBUF bypass capacitor. Any conversion
initiated before these times will produce invalid results.
When nap mode is employed, the power dissipation of
the LTC2348-18 decreases as the sampling frequency is
reduced, as shown in Figure 16. This decrease in average power dissipation occurs because a portion of the
LTC2348-18 circuitry is turned off during nap mode, and
the fraction of the conversion cycle (tCYC) spent napping
increases as the sampling frequency (fSMPL) is decreased.
18
WITH NAP MODE
t CNVL = 1µs
16
14
SUPPLY CURRENT (mA)
Power Down Mode
Reset Timing
A global reset of the LTC2348-18, equivalent to a poweron-reset event, may be executed without needing to cycle
the supplies. This feature is useful when recovering from
system-level events that require the state of the entire system to be reset to a known synchronized value. To initiate
a global reset, bring PD high twice without an intervening
conversion, as shown in Figure 15. The reset event is triggered on the second rising edge of PD, and asynchronously
I VDD
12
10
8
6
4
IVCC
2
0
IOVDD
–2
–4
I VEE
0
40
80
120
160
SAMPLING FREQUENCY (kHz)
200
234818 F16
Figure 16. Power Dissipation of the LTC2348-18
Decreases with Decreasing Sampling Frequency
tPDH
t WAKE
PD
CNV
BUSY
RESET
tPDL
tCNVH
tCONV
SECOND RISING EDGE OF
PD TRIGGERS RESET
RESET TIME
SET INTERNALLY
234818 F15
Figure 15. Reset Timing for the LTC2348-18
234818fa
30
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
CS = PD = 0
SAMPLE N
tCNVL
CNV
tCONV
BUSY
tACQ
tBUSYLH
RECOMMENDED DATA TRANSACTION WINDOW
tSCKI
tSCKIH
SCKI
1
SDI
SAMPLE N + 1
tCYC
tCNVH
DON’T CARE
S23
tDSDOBUSYL
2
3
4
5
6
7 8
tSCKIL
tSSDISCKI
tQUIET
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
tHSDISCKI
S22 S21 S20 S19 S18 S17 S16 S15 S14 S13 S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 S0
tHSDOSCKI
SOFTSPAN CONFIGURATION WORD FOR CONVERSION N + 1
tSKEW
SCKO
tDSDOSCKI
SDO0
DON’T CARE
D17
D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 C2 C1 C0 SS2 SS1 SS0 D17
CONVERSION RESULT
CHANNEL ID SOFTSPAN
CONVERSION RESULT
• • •
CHANNEL 0
24-BIT PACKET
CONVERSION N
SDO7
DON’T CARE
234818 TD01
D17
CHANNEL 1
24-BIT PACKET
CONVERSION N
D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 C2 C1 C0 SS2 SS1 SS0 D17
CONVERSION RESULT
CHANNEL 7
24-BIT PACKET
CONVERSION N
CHANNEL ID SOFTSPAN
CONVERSION RESULT
CHANNEL 0
24-BIT PACKET
CONVERSION N
Figure 17. Serial CMOS I/O Mode
Digital Interface
The LTC2348-18 features CMOS and LVDS serial interfaces,
selectable using the LVDS/CMOS pin. The flexible OVDD
supply allows the LTC2348-18 to communicate with any
CMOS logic operating between 1.8V and 5V, including
2.5V and 3.3V systems, while the LVDS interface supports
low noise digital designs. In CMOS mode, applications
may employ between one and eight lanes of serial data
output, allowing the user to optimize bus width and data
throughput. Together, these I/O interface options enable
the LTC2348-18 to communicate equally well with legacy
microcontrollers and modern FPGAs.
Serial CMOS I/O Mode
As shown in Figure 17, in CMOS I/O mode the serial data
bus consists of a serial clock input, SCKI, serial data
input, SDI, serial clock output, SCKO, and eight lanes of
serial data output, SDO0 to SDO7. Communication with
the LTC2348-18 across this bus occurs during predefined
data transaction windows. Within a window, the device
accepts 24-bit SoftSpan configuration words for the next
conversion on SDI and outputs 24-bit packets containing
conversion results and channel configuration information
from the previous conversion on SDO0 to SDO7. New
data transaction windows open 10ms after powering up
or resetting the LTC2348-18, and at the end of each conversion on the falling edge of BUSY. In the recommended
use case, the data transaction should be completed with
a minimum tQUIET time of 20ns prior to the start of the
next conversion, as shown in Figure 17. New SoftSpan
configuration words are only accepted within this recommended data transaction window, but SoftSpan changes
take effect immediately with no additional analog input
settling time required before starting the next conversion.
It is still possible to read conversion data after starting the
next conversion, but this will degrade conversion accuracy
and therefore is not recommended.
234818fa
For more information www.linear.com/LTC2348-18
31
LTC2348-18
Applications Information
Just prior to the falling edge of BUSY and the opening of
a new data transaction window, SCKO is forced low and
SDO0 to SDO7 are updated with the latest conversion
results from analog input channels 0 to 7, respectively.
Rising edges on SCKI serially clock conversion results
and analog input channel configuration information out
on SDO0 to SDO7 and trigger transitions on SCKO that
are skew-matched to the data on SDO0 to SDO7. The
resulting SCKO frequency is half that of SCKI. SCKI
rising edges also latch SoftSpan configuration words
provided on SDI, which are used to program the internal
24-bit SoftSpan configuration register. See the section
Programming the SoftSpan Configuration Register in
CMOS I/O Mode for further details. SCKI is allowed to
idle either high or low in CMOS I/O mode. As shown in
Figure 18, the CMOS bus is enabled when CS is low and
is disabled and Hi-Z when CS is high, allowing the bus
to be shared across multiple devices.
The data on SDO0 to SDO7 are grouped into 24-bit
packets consisting of an 18-bit conversion result, 3-bit
analog channel ID, and 3-bit SoftSpan code, all presented
MSB first. As suggested in Figures 17 and 18, each SDO
lane outputs these packets for all analog input channels
in a sequential, circular manner. For example, the first
24-bit packet output on SDO0 corresponds to analog
input channel 0, followed by the packets for channels 1
through 7. The data output on SDO0 then wraps back
to channel 0, and this pattern repeats indefinitely. Other
SDO lanes follow a similar circular pattern, except the
first packet presented on each lane corresponds to its
associated analog input channel.
When interfacing the LTC2348-18 with a standard SPI
bus, capture output data at the receiver on rising edges of
SCKI. SCKO is not used in this case. Multiple SDO lanes
are also usually not useful in this case. In other applications, such as interfacing the LTC2348-18 with an FPGA
or CPLD, rising and falling edges of SCKO may be used
to capture serial output data on SDO0 to SDO7 in double
data rate (DDR) fashion. Capturing data using SCKO adds
robustness to delay variations over temperature and supply.
Full Eight Lane Serial CMOS Output Data Capture
As shown in Table 3, full 200ksps per channel throughput
can be achieved with a 45MHz SCKI frequency by capturing
the first packet (24 SCKI cycles total) from all eight serial
data output lanes SDO0 to SDO7. This configuration also
allows conversion results from all channels to be captured
using as few as 18 SCKI cycles if the 3-bit analog channel
ID and 3-bit SoftSpan code are not needed and the device
SoftSpan configuration is not being changed. Multi-lane
data capture is usually best suited for use with FPGA
or CPLD capture hardware, but may be useful in other
application-specific cases.
PD = 0
BUSY
CS
SCKI DON’T CARE
SDI DON’T CARE
SCKO
SDO7
NEW SoftSpan CONFIGURATION WORD
(OVERWRITES INTERNAL CONFIG REGISTER)
TWO ALL-ZERO WORDS AND ONE PARTIAL WORD
(INTERNAL CONFIG REGISTER RETAINS CURRENT VALUE)
DON’T CARE
Hi-Z
Hi-Z
Hi-Z
CHANNEL 0 PACKET
CHANNEL 1 PACKET
CHANNEL 2 PACKET
CHANNEL 3 PACKET
(PARTIAL)
tEN
• • •
SDO0
DON’T CARE
Hi-Z
t DIS
CHANNEL 7 PACKET
CHANNEL 0 PACKET
CHANNEL 1 PACKET
CHANNEL 2 PACKET
(PARTIAL)
Figure 18. Internal SoftSpan Configuration Register Behavior. Serial CMOS Bus Response to CS
32
Hi-Z
For more information www.linear.com/LTC2348-18
Hi-Z
234818 F18
234818fa
LTC2348-18
Applications Information
Fewer Than Eight Lane Serial CMOS Output Data Capture
Applications that cannot accommodate the full eight lanes
of serial data capture may employ fewer lanes without
reconfiguring the LTC2348-18. For example, capturing
the first two packets (48 SCKI cycles total) from SDO0,
SDO2, SDO4, and SDO6 provides data for analog input
channels 0 and 1, 2 and 3, 4 and 5, and 6 and 7, respectively, using four output lanes. Similarly, capturing the first
four packets (96 SCKI cycles total) from SDO0 and SDO4
provides data for analog input channels 0 to 3 and 4 to
7, respectively, using two output lanes. If only one lane
can be accommodated, capturing the first eight packets
(192 SCKI cycles total) from SDO0 provides data for all
analog input channels. As shown in Table 3, full 200ksps
per channel throughput can be achieved with a 90MHz
SCKI frequency in the four lane case, but the maximum
CMOS SCKI frequency of 100MHz limits the throughput
to less than 200ksps per channel in the two lane and one
lane cases. Finally, note that in choosing the number of
lanes and which lanes to use for data capture, the user is
not restricted to the specific cases mentioned above. Other
choices may be more optimal in particular applications.
Programming the SoftSpan Configuration Register in
CMOS I/O Mode
The internal 24-bit SoftSpan configuration register controls the SoftSpan range for all analog input channels of
the LTC2348-18. The default state of this register after
power-up or resetting the device is all ones, configuring
each channel to convert in SoftSpan 7, the ±2.5 • VREFBUF
range (see Table 1a). The state of this register may be
modified by providing a new 24-bit SoftSpan configuration
word on SDI during the data transaction window shown
in Figure 17. New SoftSpan configuration words are only
accepted within this recommended data transaction window, but SoftSpan changes take effect immediately with
no additional analog input settling time required before
starting the next conversion. Setting a channel’s SoftSpan
code to SS[2:0] = 000 immediately disables the channel,
resulting in a corresponding reduction in tCONV on the next
conversion. Similarly, enabling a previously disabled channel requires no additional analog input settling time before
starting the next conversion. The mapping between the
serial SoftSpan configuration word, the internal SoftSpan
configuration register, and each channel’s 3-bit SoftSpan
code is illustrated in Figure 19.
If fewer than 24 SCKI rising edges are provided during a
data transaction window, the partial word received on SDI
will be ignored and the SoftSpan configuration register will
not be updated. If exactly 24 SCKI rising edges are provided,
the SoftSpan configuration register will be updated to
match the received SoftSpan configuration word, S[23:0].
The one exception to this behavior occurs when S[23:0] is
all zeros. In this case, the SoftSpan configuration register
will not be updated, allowing applications to retain the
current SoftSpan configuration state by idling SDI low. If
more than 24 SCKI rising edges are provided during a data
transaction window, each complete 24-bit word received
on SDI will be interpreted as a new SoftSpan configuration
word and applied to the SoftSpan configuration register
as described above. Any partial words are ignored.
Table 3. Required SCKI Frequency to Achieve Various Throughputs in Common Output Bus Configurations with Eight Channels
Enabled. Shaded Entries Denote Throughputs That Are Not Achievable In a Given Configuration. Calculated Using fSCKI = (Number of
SCKI Cycles)/(tACQ(MIN) – tQUIET)
I/O MODE
CMOS
LVDS
REQUIRED fSCKI (MHz) TO ACHIEVE THROUGHPUT OF
100ksps/CHANNEL
50ksps/CHANNEL
200ksps/CHANNEL
(tACQ = 5570ns)
(tACQ = 15570ns)
(tACQ = 570ns)
NUMBER OF SDO
LANES
NUMBER OF SCKI
CYCLES
8
18
35
4
2
8
24
45
5
2
4
48
90
9
4
2
96
Not Achievable
18
7
1
192
Not Achievable
35
13
1
96
180 (360Mbps)
18 (36Mbps)
7 (14Mbps)
234818fa
For more information www.linear.com/LTC2348-18
33
LTC2348-18
Applications Information
CMOS I/O MODE
tSCKIH
tSCKI
SCKI
1
SDI
DON’T CARE
2
S23
3
4
5
6
tSSDISCKI
7 8
tSCKIL
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
tHSDISCKI
S22 S21 S20 S19 S18 S17 S16 S15 S14 S13 S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 S0
SoftSpan CONFIGURATION WORD
LVDS I/O MODE
tSCKI
SCKI
(LVDS)
1
2
tSCKIH
3
4
5
6
7
8
9
tSCKIL
SDI
(LVDS)
DON’T CARE
S23
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
tSSDISCKI
tSSDISCKI
tHSDISCKI
tHSDISCKI
S22 S21 S20 S19 S18 S17 S16 S15 S14 S13 S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 S0
SoftSpan CONFIGURATION WORD
INTERNAL 24-BIT SoftSpan CONFIGURATION REGISTER
(SAME FOR CMOS AND LVDS)
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
CHANNEL 7 SoftSpan CHANNEL 6 SoftSpan CHANNEL 5 SoftSpan CHANNEL 4 SoftSpan CHANNEL 3 SoftSpan CHANNEL 2 SoftSpan CHANNEL 1 SoftSpan CHANNEL 0 SoftSpan
CODE SS[2:0]
CODE SS[2:0]
CODE SS[2:0]
CODE SS[2:0]
CODE SS[2:0]
CODE SS[2:0]
CODE SS[2:0]
CODE SS[2:0]
234818 F19
Figure 19. Mapping Between Serial SoftSpan Configuration Word, Internal SoftSpan
Configuration Register, and SoftSpan Code for Each Analog Input Channel
Typically, applications will update the SoftSpan configuration register in the manner shown in Figures 17 and 18.
After the opening of a new data transaction window at the
falling edge of BUSY, the user supplies a 24-bit SoftSpan
configuration word on SDI during the first 24 SCKI cycles.
This new word overwrites the internal configuration register
contents following the 24th SCKI rising edge. The user then
holds SDI low for the remainder of the data transaction
window causing the register to retain its contents regardless
of the number of additional SCKI cycles applied. SoftSpan
settings may be retained across multiple conversions by
holding SDI low for the entire data transaction window,
regardless of the number of SCKI cycles applied.
Serial LVDS I/O Mode
In LVDS I/O mode, information is transmitted using positive and negative signal pairs (LVDS+/LVDS−) with bits
differentially encoded as (LVDS+ − LVDS−). These signals
are typically routed using differential transmission lines
with 100Ω characteristic impedance. Logical 1’s and 0’s
are nominally represented by differential +350mV and
−350mV, respectively. For clarity, all LVDS timing diagrams
and interface discussions adopt the logical rather than
physical convention.
As shown in Figure 20, in LVDS I/O mode the serial data
bus consists of a serial clock differential input, SCKI, serial
data differential input, SDI, serial clock differential output,
SCKO, and serial data differential output, SDO. Communication with the LTC2348-18 across this bus occurs during
predefined data transaction windows. Within a window,
the device accepts 24-bit SoftSpan configuration words
for the next conversion on SDI and outputs 24-bit packets
containing conversion results and channel configuration
information from the previous conversion on SDO. New
data transaction windows open 10ms after powering up
or resetting the LTC2348-18, and at the end of each conversion on the falling edge of BUSY. In the recommended
use case, the data transaction should be completed with
a minimum tQUIET time of 20ns prior to the start of the
next conversion, as shown in Figure 20. New SoftSpan
configuration words are only accepted within this recommended data transaction window, but SoftSpan changes
234818fa
34
For more information www.linear.com/LTC2348-18
LTC2348-18
Applications Information
CS = PD = 0
SAMPLE N + 1
SAMPLE N
CNV
(CMOS)
BUSY
(CMOS)
t CYC
tCNVH
t CNVL
tCONV
t ACQ
tBUSYLH
RECOMMENDED DATA TRANSACTION WINDOW
t SCKI
SCKI
(LVDS)
SDI
(LVDS)
1
2
3
4
t SCKIL
DON’T CARE
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 185 186 187 188 189 190 191 192
t SSDISCKI
t HSDISCKI
t SSDISCKI
t HSDISCKI
S23 S22 S21 S20 S19 S18 S17 S16 S15 S14 S13 S12 S11 S10 S9 S8 S7 S6 S5 S4 S3 S2 S1 S0
t DSDOBUSYL
SoftSpan CONFIGURATION WORD FOR CONVERSION N + 1
t SKEW
t HSDOSCKI
SCKO
(LVDS)
SDO
(LVDS)
tQUIET
t SCKIH
t DSDOSCKI
DON’T CARE
D17
D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 C2 C1 C0 SS2 SS1 SS0 D17 D16 D15 D0 C2 C1 C0 SS2 SS1 SS0 D17
CONVERSION RESULT
CHANNEL ID SoftSpan
CHANNEL 0
24-BIT PACKET
CONVERSION N
CHANNEL 1
24-BIT PACKET
CONVERSION N
CHANNEL ID SoftSpan
CHANNEL 7
24-BIT PACKET
CONVERSION N
CONVERSION
RESULT
CHANNEL 0
24-BIT PACKET
CONVERSION N
234818 F20
Figure 20. Serial LVDS I/O Mode
take effect immediately with no additional analog input
settling time required before starting the next conversion.
It is still possible to read conversion data after starting the
next conversion, but this will degrade conversion accuracy
and therefore is not recommended.
Just prior to the falling edge of BUSY and the opening
of a new data transaction window, SDO is updated with
the latest conversion results from analog input channel
0. Both rising and falling edges on SCKI serially clock
conversion results and analog input channel configuration
information out on SDO. SCKI is also echoed on SCKO,
skew-matched to the data on SDO. Whenever possible, it
is recommended that rising and falling edges of SCKO be
used to capture DDR serial output data on SDO, as this will
yield the best robustness to delay variations over supply
and temperature. SCKI rising and falling edges also latch
SoftSpan configuration words provided on SDI, which are
used to program the internal 24-bit SoftSpan configuration register. See the section Programming the SoftSpan
Configuration Register in LVDS I/O Mode for further
details. As shown in Figure 21, the LVDS bus is enabled
when CS is low and is disabled and Hi-Z when CS is high,
allowing the bus to be shared across multiple devices.
Due to the high speeds involved in LVDS signaling, LVDS
bus sharing must be carefully considered. Transmission
line limitations imposed by the shared bus may limit the
maximum achievable bus clock speed. LVDS inputs are
internally terminated with a 100Ω differential resistor when
CS is low, while outputs must be differentially terminated
with a 100Ω resistor at the receiver (FPGA). SCKI must
idle in the low state in LVDS I/O mode, including when
transitioning CS.
The data on SDO are grouped into 24-bit packets consisting of an 18-bit conversion result, 3-bit analog channel
ID, and 3-bit SoftSpan code, all presented MSB first. As
suggested in Figures 20 and 21, SDO outputs these packets for all analog input channels in a sequential, circular
manner. For example, the first 24-bit packet output on
SDO corresponds to analog input channel 0, followed by
the packets for channels 1 through 7. The data output
on SDO then wraps back to channel 0, and this pattern
repeats indefinitely.
234818fa
For more information www.linear.com/LTC2348-18
35
LTC2348-18
Applications Information
Serial LVDS Output Data Capture
As shown in Table 3, full 200ksps per channel throughput
can be achieved with a 180MHz SCKI frequency by capturing eight packets (96 SCKI cycles total) of DDR data from
SDO. The LTC2348-18 supports LVDS SCKI frequencies
up to 250MHz.
Programming the SoftSpan Configuration Register in
LVDS I/O Mode
The internal 24-bit SoftSpan configuration register controls the SoftSpan range for all analog input channels of
the LTC2348-18. The default state of this register after
power-up or resetting the device is all ones, configuring
each channel to convert in SoftSpan 7, the ±2.5 • VREFBUF
range (see Table 1a). The state of this register may be
modified by providing a new 24-bit SoftSpan configuration
word on SDI during the data transaction window shown
in Figure 20. New SoftSpan configuration words are only
accepted within this recommended data transaction window, but SoftSpan changes take effect immediately with
no additional analog input settling time required before
starting the next conversion. Setting a channel’s SoftSpan
code to SS[2:0] = 000 immediately disables the channel,
resulting in a corresponding reduction in tCONV on the next
conversion. Similarly, enabling a previously disabled channel requires no additional analog input settling time before
starting the next conversion. The mapping between the
serial SoftSpan configuration word, the internal SoftSpan
configuration register, and each channel’s 3-bit SoftSpan
code is illustrated in Figure 19.
If fewer than 24 SCKI edges (rising plus falling) are
provided during a data transaction window, the partial
word received on SDI will be ignored and the SoftSpan
configuration register will not be updated. If exactly 24
SCKI edges are provided, the SoftSpan configuration
register will be updated to match the received SoftSpan
configuration word, S[23:0]. The one exception to this
behavior occurs when S[23:0] is all zeros. In this case,
the SoftSpan configuration register will not be updated,
allowing applications to retain the current SoftSpan configuration state by idling SDI low. If more than 24 SCKI
edges are provided during a data transaction window, each
complete 24-bit word received on SDI will be interpreted
as a new SoftSpan configuration word and applied to the
SoftSpan configuration register as described above. Any
partial words are ignored.
Typically, applications will update the SoftSpan configuration register in the manner shown in Figures 20 and 21.
After the opening of a new data transaction window at
the falling edge of BUSY, the user supplies a 24-bit DDR
SoftSpan configuration word on SDI during the first 12
SCKI cycles. This new word overwrites the internal configuration register contents following the 12th SCKI falling
edge. The user then holds SDI low for the remainder of
the data transaction window causing the register to retain
its contents regardless of the number of additional SCKI
cycles applied. SoftSpan settings may be retained across
multiple conversions by holding SDI low for the entire
data transaction window, regardless of the number of
SCKI cycles applied.
PD = 0
BUSY
(CMOS)
CS
(CMOS)
tEN
tDIS
SCKI
DON’T CARE
(LVDS)
SDI
DON’T CARE
(LVDS)
SCKO
(LVDS)
SDO
(LVDS)
DON’T CARE
NEW SoftSpan CONFIGURATION WORD
(OVERWRITES INTERNAL CONFIG REGISTER)
TWO ALL-ZERO WORDS AND ONE PARTIAL WORD
(INTERNAL CONFIG REGISTER RETAINS CURRENT VALUE)
DON’T CARE
Hi-Z
Hi-Z
Hi-Z
CHANNEL 0 PACKET
CHANNEL 1 PACKET
CHANNEL 2 PACKET
CHANNEL 3 PACKET
(PARTIAL)
Figure 21. Internal SoftSpan Configuration Register Behavior. Serial LVDS Bus Response to CS
36
For more information www.linear.com/LTC2348-18
Hi-Z
234818 F21
234818fa
LTC2348-18
Board Layout
To obtain the best performance from the LTC2348-18, a
four-layer printed circuit board (PCB) is recommended.
Layout for the PCB should ensure the digital and analog
signal lines are separated as much as possible. In particular, care should be taken not to run any digital clocks or
signals alongside analog signals or underneath the ADC.
Also minimize the length of the REFBUF to GND (Pin 20)
bypass capacitor return loop, and avoid routing CNV near
signals which could potentially disturb its rising edge.
Supply bypass capacitors should be placed as close as
possible to the supply pins. Low impedance common returns for these bypass capacitors are essential to the low
noise operation of the ADC. A single solid ground plane
is recommended for this purpose. When possible, screen
the analog input traces using ground.
Reference Design
For a detailed look at the reference design for this converter, including schematics and PCB layout, please refer
to DC2094A, the evaluation kit for the LTC2348-18.
234818fa
For more information www.linear.com/LTC2348-18
37
LTC2348-18
Package Description
Please refer to http://www.linear.com/product/LTC2348-18#packaging for the most recent package drawings.
LX Package
48-Lead Plastic LQFP (7mm × 7mm)
(Reference LTC DWG # 05-08-1760 Rev A)
7.15 – 7.25
9.00 BSC
5.50 REF
7.00 BSC
48
0.50 BSC
1
2
48
SEE NOTE: 4
1
2
9.00 BSC
5.50 REF
7.00 BSC
7.15 – 7.25
0.20 – 0.30
A
A
PACKAGE OUTLINE
C0.30 – 0.50
1.30 MIN
RECOMMENDED SOLDER PAD LAYOUT
APPLY SOLDER MASK TO AREAS THAT ARE NOT SOLDERED
1.60
1.35 – 1.45 MAX
11° – 13°
R0.08 – 0.20
GAUGE PLANE
0.25
0° – 7°
11° – 13°
0.09 – 0.20
1.00 REF
0.50
BSC
0.17 – 0.27
0.05 – 0.15
0.45 – 0.75
SECTION A – A
COMPONENT
PIN “A1”
TRAY PIN 1
BEVEL
XXYY
LTCXXXX
LX-ES
Q_ _ _ _ _ _
e3
NOTE:
1. PACKAGE DIMENSIONS CONFORM TO JEDEC #MS-026 PACKAGE OUTLINE
2. DIMENSIONS ARE IN MILLIMETERS
3. DIMENSIONS OF PACKAGE DO NOT INCLUDE MOLD FLASH. MOLD FLASH
SHALL NOT EXCEED 0.25mm ON ANY SIDE, IF PRESENT
4. PIN-1 INDENTIFIER IS A MOLDED INDENTATION, 0.50mm DIAMETER
5. DRAWING IS NOT TO SCALE
LX48 LQFP 0113 REV A
PACKAGE IN TRAY LOADING ORIENTATION
234818fa
38
For more information www.linear.com/LTC2348-18
LTC2348-18
Revision History
REV
DATE
DESCRIPTION
A
02/16
Updated the ADC Timing Characteristics section
PAGE NUMBER
6
Inserted new graphs: PSRR vs Frequency and Power Dissipation vs Sampling Rate
12
Updated Table 2
22
Updated the Application Information section
24
Updated Figure 16
30
Updated Table 3
33
Updated the Board Layout section
37
234818fa
Information furnished by Linear Technology Corporation is believed to be accurate and reliable.
However, no responsibility is assumed for its use. Linear Technology Corporation makes no representation that the interconnection
of itsinformation
circuits as described
herein will not infringe on existing patent rights.
For more
www.linear.com/LTC2348-18
39
LTC2348-18
Typical Application
Digitize Differential Signals Over a Wide Common Mode Range
31V
+
–
IN+
ARBITRARY
½ LT1124
LOWPASS FILTERS
18pF
24V
0.1µF
2.49k
COMMON MODE
INPUT RANGE
6.6nF
2.49k
–
+
IN–
LTC2348-18
6.6nF
18pF
0V
VCC
IN0+
IN0–
549Ω
DIFFERENTIAL MODE
INPUT RANGE: ±500mV
31V
49.9Ω
49.9Ω
½ LT1124
BW ~ 500kHz
–5V
ONLY CHANNEL 0 SHOWN FOR CLARITY
VEE REFBUF
0.1µF
47µF
REFIN
0.1µF
–5V
234818 F07a
Related Parts
PART NUMBER
ADCs
LTC2348-16
DESCRIPTION
COMMENTS
16-Bit, 200ksps, 8-Channel Simultaneous
Sampling, ±1LSB INL, Serial ADC
LTC2378-20/LTC2377-20/ 20-Bit, 1Msps/500ksps/250ksps,
±0.5ppm INL Serial, Low Power ADC
LTC2376-20
LTC2338-18/LTC2337-18/ 18-Bit, 1Msps/500ksps/250ksps, Serial,
LTC2336-18
Low Power ADC
LTC2328-18/LTC2327-18/ 18-Bit, 1Msps/500ksps/250ksps, Serial,
LTC2326-18
Low Power ADC
LTC2373-18/LTC2372-18 18-Bit, 1Msps/500ksps, 8-Channel, Serial ADC
±10.24V SoftSpan Inputs with Wide Common Mode Range, 94db SNR,
Serial CMOS and LVDS I/O, 7mm × 7mm LQFP-48 Package
2.5V Supply, ±5V Fully Differential Input, 104dB SNR, MSOP-16 and
4mm × 3mm DFN-16 Packages
5V Supply, ±10.24V Fully Differential Input, 100dB SNR, MSOP-16 Package
5V Supply, ±10.24V Pseudo-Differential Input, 95dB SNR,
MSOP-16 Package
5V Supply, 8 Channel Multiplexed, Configurable Input Range, 100dB SNR,
DGC, 5mm × 5mm QFN-32 Package
LTC2379-18/LTC2378-18/ 18-Bit,1.6Msps/1Msps/500ksps/250ksps, Serial, 2.5V Supply, Differential Input, 101.2dB SNR, ±5V Input Range, DGC, Pin
LTC2377-18/LTC2376-18 Low Power ADC
Compatible Family in MSOP-16 and 4mm × 3mm DFN-16 Packages
LTC2380-16/LTC2378-16/ 16-Bit, 2Msps/1Msps/500ksps/250ksps, Serial, 2.5V Supply, Differential Input, 96.2dB SNR, ±5V Input Range, DGC, Pin
Compatible Family in MSOP-16 and 4mm × 3mm DFN-16 Packages
LTC2377-16/LTC2376-16 Low Power ADC
LTC2389-18/LTC2389-16 18-Bit/16-Bit, 2.5Msps, Parallel/Serial ADC
5V Supply, Pin-Configurable Input Range, 99.8dB/96dB SNR, Parallel or
Serial I/O 7mm × 7mm LQFP-48 and QFN-48 Packages
LTC1859/LTC1858/
16-/14-/12-Bit, 8-Channel, 100ksps, Serial ADC ±10V, SoftSpan, Single-Ended or Differential Inputs, Single 5V Supply,
LTC1857
SSOP-28 Package
LTC1609
16-Bit, 200ksps Serial ADC
±10V, Configurable Unipolar/Bipolar Input, Single 5V Supply, SSOP-28 and
SO-20 Packages
DACs
±1LSB INL/DNL, Software-Selectable Ranges,
LTC2756/LTC2757
18-Bit, Serial/Parallel IOUT SoftSpan DAC
SSOP-28/7mm × 7mm LQFP-48 Package
LTC2668
16-Channel 16-/12-Bit ±10V VOUT SoftSpan DACs ±4LSB INL, Precision Reference 10ppm/°C Max, 6mm × 6mm QFN-40 Package
References
LTC6655
Precision Low Drift Low Noise Buffered Reference 5V/2.5V/2.048V/1.25V, 2ppm/°C, 0.25ppm Peak-to-Peak Noise, MSOP-8 Package
LTC6652
Precision Low Drift Low Noise Buffered Reference 5V/2.5V/2.048V/1.25V, 5ppm/°C, 2.1ppm Peak-to-Peak Noise, MSOP-8 Package
Amplifiers
LT1468/LT1469
Single/Dual 90MHz, 22V/µs, 16-Bit Accurate Op Amp Low Input Offset: 75µV/125µV
LT1354/LT1355/LT1356 Single/Dual/Quad 1mA, 12MHz, 400V/µs Op Amp Good DC Precision, Stable with All Capacitive Loads
LT1357/LT1358/LT1359 Single/Dual/Quad 2mA, 25MHz, 800V/µs Op Amp Good DC Precision, Stable with All Capacitive Loads
234818fa
40 Linear Technology Corporation
1630 McCarthy Blvd., Milpitas, CA 95035-7417
For more information www.linear.com/LTC2348-18
(408) 432-1900 ● FAX: (408) 434-0507
●
www.linear.com/LTC2348-18
LT 0216 REV A • PRINTED IN USA
 LINEAR TECHNOLOGY CORPORATION 2015