LTC2000A - 16-/14-/11-Bit 2.7Gsps DACs

LTC2000A
16-/14-/11-Bit 2.7Gsps
DACs
Features
Description
80dBc SFDR at 50MHz fOUT
n>68dBc SFDR from DC to 1080MHz f
OUT
n40mA Nominal Full-Scale, ±1V Output Compliant
n10mA to 60mA Adjustable Full-Scale Current Range
n Single or Dual Port DDR LVDS and DHSTL Interface
n Low Latency (7.5 Cycles for Single Port,
11 Cycles for Dual Port)
n>78dBc 2-Tone IMD from DC to 1000MHz f
OUT
n–156dBc/Hz Additive Phase Noise at 1MHz Offset for
65MHz fOUT
n170-Lead (9mm × 15mm) BGA Package
The LTC®2000A is a family of 16-/14-/11-bit 2.7Gsps
current steering DACs with exceptional spectral purity.
Applications
A serial peripheral interface (SPI) port allows configuration and read back of internal registers. Operating from
1.86V and 3.3V supplies, the LTC2000A consumes 2.41W
at 2.7Gsps and 1.43W at 1.35Gsps.
n
n
n
n
n
n
n
The single (1.35Gsps mode) or dual (2.7Gsps mode) port
source synchronous LVDS interface supports data rates of
up to 1.35Gbps using a 675MHz DDR data clock, which
can be either in quadrature or in phase with the data. An
internal synchronizer automatically aligns the data with
the DAC sample clock.
Additional features such as pattern generation, LVDS
loopout and junction temperature sensing simplify system
development and testing.
Broadband Communication Systems
DOCSIS CMTS
Direct RF Synthesis
Radar
Instrumentation
Automatic Test Equipment
L, LT, LTC, LTM, Linear Technology and the Linear logo are registered trademarks of Linear
Technology Corporation. All other trademarks are the property of their respective owners.
Protected by U.S. Patents, including 8330633.
Block Diagram
TSTP/N
PD
JUNCTION
TEMPERATURE
PATTERN
GENERATOR
CS
SCK SDI
SVDD
SDO
SPI
SFDR vs fOUT, fDAC = 2.7Gsps
100
DAP/N[15:0]
DDR DATA FLIP-FLOPS
16-BIT DAC
4:1
50Ω
IOUTN
GAIN
ADJUST
DCKIP/N
CLK DIVIDER
÷2 OR ÷4
AVDD18
DVDD18
FSADJ
CLK
RECEIVER
AVDD33
90
DVDD33
GND
CKP/N
80
70
60
CLOCK
SYNC
DELAY
ADJUST
DCKOP/N
DIGITAL AMPLITUDE = 0dBFS
IOUTFS = 40mA
50Ω
SFDR (dBc)
LVDS RECEIVERS
DBP/N[15:0]
IOUTP
REFIO
10k
50
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A TA01b
REF
2000A BD
2000af
For more information www.linear.com/LTC2000A
1
LTC2000A
Table of Contents
Features...................................................... 1
Applications................................................. 1
Block Diagram............................................... 1
Description.................................................. 1
Absolute Maximum Ratings............................... 3
Pin Configuration........................................... 3
Order Information........................................... 4
Electrical Characteristics.................................. 5
Timing Characteristics..................................... 8
Typical Performance Characteristics.................... 9
Pin Functions............................................... 16
Block Diagram.............................................. 17
Timing Diagrams.......................................... 18
Operation................................................... 18
Introduction............................................................. 18
Dual-Port Mode....................................................... 19
Single-Port Mode..................................................... 19
Serial Peripheral Interface (SPI)..............................22
Power-On Reset.......................................................22
Power Down............................................................22
Reference Operation................................................22
Setting the Full-Scale Current..................................23
DAC Transfer Function............................................. 24
Analog Outputs (IOUTP/N)......................................... 24
DAC Sample Clock (CKP/N).....................................25
Divided Clock Output (DCKOP/N).............................25
LVDS Data Clock Input (DCKIP/N)...........................25
LVDS Data Input Ports (DAP/N, DBP/N)..................26
Clock Synchronizer.................................................. 27
Minimizing Harmonic Distortion..............................29
Measuring LVDS Input Timing Skew........................29
Measuring Internal Junction Temperature (TJ)........ 32
Pattern Generator.................................................... 32
2
SPI Register Summary.................................... 33
Applications Information................................. 34
Sample Start-Up Sequence......................................34
Output Configurations ............................................35
Generating the DAC Sample Clock...........................35
Synchronizing Multiple LTC2000As in Dual-Port
Mode.......................................................................36
Synchronizing Multiple LTC2000As in Single-Port
Mode.......................................................................38
PCB Layout Considerations.....................................40
Pin Locations (LTC2000A-16)............................ 45
Pin Locations (LTC2000A-14)............................ 47
Pin Locations (LTC2000A-11)............................ 49
Package Description...................................... 51
Typical Application........................................ 52
Related Parts............................................... 52
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Absolute Maximum Ratings
(Notes 1, 2)
AVDD33, DVDD33, SVDD.................................. –0.3V to 4V
AVDD18, DVDD18............................................ –0.3V to 2V
IOUTP, IOUTN..................–1.2V to Min (AVDD33 + 0.3V, 4V)
FSADJ, REFIO............. –0.3V to Min (AVDD33 + 0.3V, 4V)
DCKIP, DCKIN.............–0.3V to Min (DVDD33 + 0.3V, 4V)
DCKOP, DCKON...........–0.3V to Min (DVDD33 + 0.3V, 4V)
DAP/N, DBP/N.............–0.3V to Min (DVDD33 + 0.3V, 4V)
TSTP, TSTN................. –0.3V to Min (AVDD33 + 0.3V, 4V)
CKP, CKN.................... –0.3V to Min (AVDD18 + 0.3V, 2V)
CS, PD, SCK, SDI, SDO....–0.3V to Min (SVDD + 0.3V, 4V)
Operating Temperature Range
LTC2000AC............................................... 0°C to 70°C
LTC2000AI............................................–40°C to 85°C
Maximum Junction Temperature........................... 125°C
Storage Temperature Range................... –55°C to 125°C
Lead Temperature (Soldering, 10 sec).................... 260°C
Pin Configuration
LTC2000A-16
1
2
A
3
TOP VIEW
4 5 6 7
LTC2000A-14
8
9
1
10
A
GND
B
2
C
AVDD18
D
DVDD18
E
LTC2000A-11
8
9
10
1
C
AVDD18
D
DVDD18
F
G
G
H
H
H
J
J
L
L
M
M
N
N
N
P
P
R
S
AVDD33
R
GND
DVDD33
Q
GND
S
10
GND
K
M
DVDD33
9
DVDD18
J
GND
L
Q
8
E
F
AVDD33
TOP VIEW
5 6 7
C
G
K
4
D
F
K
3
GND
B
E
GND
2
A
GND
B
AVDD18
3
TOP VIEW
4 5 6 7
AVDD33
P
DVDD33
Q
R
S
GND
BGA PACKAGE
170-LEAD (9mm × 15mm × 1.54mm)
BGA PACKAGE
170-LEAD (9mm × 15mm × 1.54mm)
BGA PACKAGE
170-LEAD (9mm × 15mm × 1.54mm)
TJMAX = 125°C, θJA = 20°C/W, θJB = 8°C/W,
θJCtop = 9°C/W, θJCbottom = 3°C/W
TJMAX = 125°C, θJA = 20°C/W, θJB = 8°C/W,
θJCtop = 9°C/W, θJCbottom = 3°C/W
TJMAX = 125°C, θJA = 20°C/W, θJB = 8°C/W,
θJCtop = 9°C/W, θJCbottom = 3°C/W
2000af
For more information www.linear.com/LTC2000A
3
LTC2000A
Order Information
LTC2000A
C
Y
–16
#PBF
LEAD FREE DESIGNATOR
PBF = Lead Free
RESOLUTION
16 = 16-Bit Resolution
14 = 14-Bit Resolution
11 = 11-Bit Resolution
PACKAGE TYPE
170-Lead (15mm × 9mm × 1.54mm) BGA
TEMPERATURE GRADE
C = Commercial Temperature Range (0°C to 70°C)
I = Industrial Temperature Range (–40°C to 85°C)
PRODUCT PART NUMBER
PART NUMBER
BALL FINISH
PART MARKING*
PACKAGE TYPE
MSL RATING
LTC2000ACY-16#PBF
SAC305 (RoHS)
LTC2000ACY-14#PBF
SAC305 (RoHS)
LTC2000ACY-11#PBF
LTC2000AIY-16#PBF
TEMPERATURE RANGE
LTC2000Y-16
BGA
3
0°C to 70°C
LTC2000Y-14
BGA
3
0°C to 70°C
SAC305 (RoHS)
LTC2000Y-11
BGA
3
0°C to 70°C
SAC305 (RoHS)
LTC2000Y-16
BGA
3
–40°C to 85°C
LTC2000AIY-14#PBF
SAC305 (RoHS)
LTC2000Y-14
BGA
3
–40°C to 85°C
LTC2000AIY-11#PBF
SAC305 (RoHS)
LTC2000Y-11
BGA
3
–40°C to 85°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/
This product is only offered in trays. For more information go to: http://www.linear.com/packaging/
4
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Electrical Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. AVDD18, DVDD18 = 1.8V to 1.92V, AVDD33, DVDD33 = 3.135V to 3.465V,
SVDD = 1.71V to 3.465V, RFSADJ = 500Ω, 12.5Ω load from IOUTP/N to GND including internal 50Ω termination, unless otherwise
specified.
SYMBOL
PARAMETER
DC Performance
Resolution
DNL
INL
Differential Nonlinearity
Integral Nonlinearity
Offset Error
Offset Error Drift
Gain Error
Gain Error Drift
Power Supply Rejection Ratio
Analog Output
Full-Scale Output Current
Output Compliance Range
Output Resistance
Output Capacitance
Output Bandwidth
AC Performance
Maximum Update Rate
SFDR
Spurious Free Dynamic Range
fDAC = 1.25Gsps, 0dBFS
Spurious Free Dynamic Range
fDAC = 2.7Gsps, 0dBFS
IMD
2-Tone Intermodulation Distortion
fOUT2 = fOUT1 + 1.25MHz
fDAC = 1.25Gsps, –6dBFS
2-Tone Intermodulation Distortion
fOUT2 = fOUT1 + 1.25MHz
fDAC = 2.7Gsps, –6dBFS
CONDITIONS
LTC2000A-16
LTC2000A-14
LTC2000A-11
LTC2000A-16
LTC2000A-14
LTC2000A-11
LTC2000A-16
LTC2000A-14
LTC2000A-11
LTC2000A-16
LTC2000A-14
LTC2000A-11
MIN
l
l
l
MAX
16
14
11
±0.5
±0.2
±0.1
±1
±0.5
±0.2
l
l
l
l
l
l
l
l
l
±2.7
±1
±0.5
±4
±2
±1
±0.05
±0.06
±0.09
1
±0.5
5
69
Full-Scale; AVDD33 = 3.135V to 3.465V
RFSADJ = 500Ω
40
l
IOUTP/N to GND
TYP
l
–1
42
RIOUTP/N = 12.5Ω, –3dB Excluding sin(x)/x
l
Dual-Port Mode
l
Single-Port Mode
fOUT = 50MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 100MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 250MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 500MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 100MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 200MHz, LIN_DIS = 0, LIN_GN = 75% l
fOUT = 500MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 1000MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 500MHz, LIN_DIS = 1
fOUT = 1000MHz, LIN_DIS = 1
fOUT = 50MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 100MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 250MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 500MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 100MHz, LIN_DIS = 1
fOUT = 200MHz, LIN_DIS = 1
fOUT = 500MHz, LIN_DIS = 1
fOUT = 1000MHz, LIN_DIS = 1
fOUT = 500MHz, LIN_DIS = 0, LIN_GN = 75%
fOUT = 1000MHz, LIN_DIS = 0, LIN_GN = 75%
50
6
2.1
2.7
1.35
67
82
82
74
74
75
72
75
72
67
61
103
93
97
84
87
86
82
80
79
68
1
58
UNITS
Bits
Bits
Bits
LSB
LSB
LSB
LSB
LSB
LSB
% FSR
% FSR
% FSR
ppm/°C
% FSR
ppm/°C
dB
mA
V
Ω
pF
GHz
Gsps
Gsps
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
dBc
2000af
For more information www.linear.com/LTC2000A
5
LTC2000A
Electrical Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. AVDD18, DVDD18 = 1.8V to 1.92V, AVDD33, DVDD33 = 3.135V to 3.465V,
SVDD = 1.71V to 3.465V, RFSADJ = 500Ω, 12.5Ω load from IOUTP/N to GND including internal 50Ω termination, unless otherwise
specified.
SYMBOL
NSD
WCDMA
ACLR
Latency
PARAMETER
Noise Spectral Density
0dBFS Single Tone, fDAC = 2.7Gsps,
IOUTFS = 40mA
Phase Noise
fDAC = 2.7Gsps, fOUT = 65MHz
0dBFS Single Tone, IOUTFS = 40mA
WCDMA ACLR (Single Carrier)
Adjacent/Alternate Adjacent Channel
Latency (Note 5)
Aperture Delay
Settling Time
CONDITIONS
LTC2000A-16, fOUT = 100MHz
LTC2000A-16, fOUT = 350MHz
LTC2000A-16, fOUT = 550MHz
LTC2000A-16, fOUT = 950MHz
LTC2000A-14, fOUT = 100MHz
LTC2000A-14, fOUT = 350MHz
LTC2000A-14, fOUT = 550MHz
LTC2000A-14, fOUT = 950MHz
LTC2000A-11, fOUT = 100MHz
LTC2000A-11, fOUT = 350MHz
LTC2000A-11, fOUT = 550MHz
LTC2000A-11, fOUT = 950MHz
10kHz Offset
1MHz Offset
MIN
fDAC = 2.7Gsps, fOUT = 350MHz
fDAC = 2.7Gsps, fOUT = 950MHz
TYP
–164
–158
–155
–153
–163
–158
–155
–153
–156
–154
–153
–150
–141
–156
MAX
78/79
72/75
Single-Port Mode
Dual-Port Mode, DAP/N Data
Dual-Port Mode, DBP/N Data
CKP/N Rising to IOUTP/N Transition
±0.1% FSR, Full-Scale Step
UNITS
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBm/Hz
dBc/Hz
dBc/Hz
dBc
dBc
7.5
10
11
3
2.2
Cycles
Cycles
Cycles
ns
ns
Reference
Output Voltage
Input Voltage
Reference Temperature Coefficient
Output Impedance
DAC Clock Inputs (CKP, CKN)
Differential Input Voltage Range
Common-Mode Input Voltage
Sampling Clock Frequency
Input Impedance
LVDS Inputs (DCKIP, DCKIN, DAP/N, DBP/N)
Differential Input Voltage Range
Common-Mode Voltage Range
Differential Input Impedance
Maximum Data Rate
LVDS Clock Frequency
6
l
l
1.225
1.1
1.25
1.275
1.4
V
V
ppm/°C
kΩ
±1.8
V
V
MHz
kΩ
±25
10
l
±0.3
Set Internally
1
l
50
2700
5
l
l
l
±0.2
0.4
95
l
l
25
120
±0.6
1.8
145
1350
675
V
V
Ω
Mbps
MHz
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Electrical Characteristics
The l denotes the specifications which apply over the full operating
temperature range, otherwise specifications are at TA = 25°C. AVDD18, DVDD18 = 1.8V to 1.92V, AVDD33, DVDD33 = 3.135V to 3.465V,
SVDD = 1.71V to 3.465V, RFSADJ = 500Ω, 12.5Ω load from IOUTP/N to GND including internal 50Ω termination, unless otherwise
specified.
SYMBOL
PARAMETER
LVDS Output (DCKOP, DCKON)
Differential Output Voltage
Common-Mode Output Voltage
Internal Termination Resistance
CMOS Digital Inputs (CS, PD, SCK, SDI)
Digital Input High Voltage
VIH
VIL
Digital Input Low Voltage
ILK
Digital Input Leakage
Digital Input Capacitance
CIN
CMOS Digital Output (SDO)
Digital Output High Voltage
VOH
VOL
Digital Output Low Voltage
Hi-Z Output Leakage
Hi-Z Output Capacitance
Power Supply
AVDD33, DVDD33 Supply Voltage
VVDD33
AVDD18, DVDD18 Supply Voltage
VVDD18
SVDD SPI Supply Voltage
VSVDD
AVDD33 Supply Current, AVDD33 = 3.3V
IAVDD33
IDVDD33
IAVDD18
IDVDD18
ISVDD
CONDITIONS
100Ω Differential Load, DCKO_ISEL = 0
50Ω Differential Load, DCKO_ISEL = 1
MIN
TYP
MAX
UNITS
l
0.24
0.24
1.075
0.36
0.36
1.2
100
0.48
0.48
1.325
V
V
V
Ω
l
70
l
l
DCKO_TRM = 1
l
VIN = GND or SVDD
l
30
±10
%VSVDD
%VSVDD
µA
pF
15
±10
%VSVDD
%VSVDD
µA
pF
8
ISOURCE = 0.2mA
ISINK = 1.6mA
l
85
l
l
8
l
l
l
PD = SVDD
PD = GND
DVDD33 Supply Current, DVDD33 = 3.3V PD = SVDD
PD = GND
AVDD18 Supply Current, AVDD18 = 1.86V fDAC = 2700MHz
fDAC = 1350MHz
fDAC = 0Hz, PD = SVDD
fDAC = 0Hz, PD = GND
DVDD18 Supply Current, DVDD18 = 1.86V fDAC = 2700MHz
fDAC = 1350MHz
fDAC = 0Hz, PD = SVDD
fDAC = 0Hz, PD = GND
SVDD Supply Current (Note 4),
fSCK = 0Hz
SVDD = 3.3V
Total Power Dissipation
fDAC = 2700MHz
fDAC = 1350MHz
fDAC = 0Hz, PD = SVDD
fDAC = 0Hz, PD = GND
l
l
l
l
l
l
l
l
l
l
l
l
l
3.135
1.8
1.71
3.3
1.86
68
0.1
8
0.1
780
425
23
3
380
210
10
0.1
0.1
2408
1432
312
6
3.465
1.92
3.465
78
10
14
5
870
480
27
180
440
240
14
240
5
V
V
V
mA
µA
mA
µA
mA
mA
mA
µA
mA
mA
mA
µA
µA
mW
mW
mW
µW
2000af
For more information www.linear.com/LTC2000A
7
LTC2000A
Timing Characteristics
The l denotes the specifications which apply over the full operating temperature
range, otherwise specifications are at TA = 25°C. AVDD18, DVDD18 = 1.8V to 1.92V, AVDD33, DVDD33 = 3.135V to 3.465V, SVDD = 1.71V
to 3.465V, RFSADJ = 500Ω, output load 50Ω double terminated, unless otherwise specified.
SYMBOL
t1
t2
t3
t4
t5
t6
t7
t10
t13
t11
t12
PARAMETER
SDI Valid to SCK Setup
SDI Valid to SCK Hold
SCK High Time
SCK Low Time
CS Pulse Width
SCK High to CS High
CS Low to SCK High
CS High to SCK High
SCK Low to SDO Valid
SCK Frequency
LVDS DAP/N, DBP/N to DCKI Setup Time
(Note 3)
LVDS DAP/N, DBP/N to DCKI Hold Time
(Note 3)
CONDITIONS
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
(Note 3)
Unloaded (Note 3)
50% Duty Cycle (Note 3)
DCKI_Q = 1
DCKI_Q = 0, DCKI_TADJ = 000
DCKI_Q = 1
DCKI_Q = 0, DCKI_TADJ = 000
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 voltages are with respect to GND.
8
l
l
l
l
l
l
l
l
l
MIN
4
4
9
9
10
7
7
7
10
l
l
MAX
50
l
l
l
TYP
200
570
200
–170
UNITS
ns
ns
ns
ns
ns
ns
ns
ns
ns
MHz
ps
ps
ps
ps
Note 3: Guaranteed by design and not production tested.
Note 4: Digital inputs at 0V or SVDD.
Note 5: Latency is the delay from a transition on DCKIP/N until the CKP/N
transition which causes the sample on DAP/N or DBP/N to appear at the
DAC output IOUTP/N, as measured in DAC sample clock (CKP/N) cycles.
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, unless otherwise noted.
LTC2000A-16 Integral
Nonlinearity (INL)
LTC2000A-16 Differential
Nonlinearity (DNL)
4
2
2
1
6
INL vs IOUTFS and Temperature
4
INL (POS)
0
INL (LSB)
DNL (LSB)
INL (LSB)
2
0
0
–2
–1
–2
–4
–4
–32768
–16384
0
16384
–2
–32768
32768
0
–16384
16384
DNL vs IOUTFS and Temperature
0
0.02
0
–0.02
–2
16 BIT
14 BIT
11 BIT
–0.04
–4
DNL (NEG)
–25
0
25
50
TEMPERATURE (°C)
75
100
–0.06
–50
–25
0
25
50
TEMPERATURE (°C)
Reference Output Voltage
vs Temperature
1.255
–1.0
–50
100
I (mA)
1.250
30
25
AVDD18
400
DVDD18
2000A G07
0
100
DVDD18
AVDD18
SVDD
AVDD33
DVDD33
15
10
5
AVDD33
DVDD33
100
100
75
20
200
75
0
25
50
TEMPERATURE (°C)
Shutdown Current
vs Temperature
700
300
1.245
–25
2000A G06
35
500
0
25
50
TEMPERATURE (°C)
–0.5
800
600
–25
0
Supply Current vs fDAC
1.260
VREF (V)
75
0.5
2000A G05
2000A G04
1.240
–50
40mA
60mA
Gain Error vs Temperature
GAIN ERROR (% FSR)
DNL (POS)
–6
–50
100
2000A G03
1.0
I (µA)
DNL (LSB)
2
75
0.04
OFFSET ERROR (%FSR)
4
Offset Error vs Temperature
0.06
40mA
60mA
0
25
50
TEMPERATURE (°C)
10mA
20mA
2000A G02
2000A G01
10mA
20mA
INL (NEG)
–25
CODE
CODE
6
–6
–50
32768
0
500
1000
1500 2000
fDAC (MHz)
2500
0
–5
–50
–25
0
25
50
75
100
TEMPERATURE (°C)
2000A G08
2000A G09
2000af
For more information www.linear.com/LTC2000A
9
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
0
Single-Tone Spectrum at
fOUT = 990MHz, fDAC = 2.7Gsps
0
RBW = 2kHz, VBW = 5kHz
REF = 5dBm, ATTEN = 25dB
SWEEP = 620s (1001 PTS)
–10
–20
–20
–30
–40
10dB/DIV
10dB/DIV
–30
fDAC-fOUT
IMAGE
–50
–60
–70
–70
–80
–80
20
556
–90
2700
2164
1628
1092
FREQUENCY (MHz)
HD2 AND HD3
IN 2ND
NYQUIST BAND
–50
20
556
60
50
200
400
600
800
fOUT (MHz)
1000
80
80
50
0
200
400
600
800
fOUT (MHz)
1000
1200
0
200
400
600
800
fOUT (MHz)
2000A G16
10
1000
80
70
50
1200
0
200
400
600
800
fOUT (MHz)
60
100
90
80
70
0dBFS
–3dBFS
–6dBFS
–12dBFS
–16dBFS
60
200
400
600
800
fOUT (MHz)
1200
HD2 vs fOUT and Digital Amplitude
(dBFS), fDAC = 2.7Gsps
DIGITAL AMPLITUDE = 0dBFS
2.7Gsps
2.0Gsps
1.6Gsps
1.25Gsps
0
1000
2000A G15
HD3 vs fOUT and fDAC
70
50
1200
60
HD2 (dBc)
90
HD3 (dBc)
HD2 (dBc)
100
1000
60mA
40mA
20mA
10mA
90
SFDR (dBc)
70
90
60
600
800
fOUT (MHz)
2000A G14
HD2 vs fOUT and fDAC
DIGITAL AMPLITUDE = 0dBFS
2.7Gsps
2.0Gsps
1.6Gsps
1.25Gsps
400
100
0dBFS
–3dBFS
–6dBFS
–12dBFS
–16dBFS
80
50
1200
70
200
SFDR vs fOUT and IOUTFS,
fDAC = 2.7Gsps
60
0
0
2000A G12
SFDR vs fOUT and Digital
Amplitude (dBFS), fDAC = 2.7Gsps
2000A G13
100
50
2700
2164
1628
1092
FREQUENCY (MHz)
90
SFDR (dBc)
SFDR (dBc)
70
70
60
100
DIGITAL AMPLITUDE = 0dBFS
85°C
25°C
–40°C
80
80
2000A G11
SFDR vs fOUT and Temperature,
fDAC = 2.7Gsps
90
DIGITAL AMPLITUDE = 0dBFS
2.7Gsps
2.0Gsps
1.6Gsps
1.25Gsps
HD3
2000A G10
100
SFDR vs fOUT and fDAC
90
fDAC-fOUT
IMAGE
–40
–60
–90
RBW = 2kHz
VBW = 5kHz
REF = 5dBm
ATTEN = 25dB
SWEEP = 620s
(1001 PTS)
–10
100
SFDR (dBc)
Single-Tone Spectrum at
fOUT = 70MHz, fDAC = 2.7Gsps
1000
1200
2000A G17
50
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A G18
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
2-Tone IMD vs fOUT and fDAC
110
90
100
80
90
70
0dBFS
–3dBFS
–6dBFS
–12dBFS
–16dBFS
60
50
0
200
400
600
800
fOUT (MHz)
1000
80
60
0
200
400
600
800
fOUT (MHz)
2000A G19
90
80
200
400
600
800
fOUT (MHz)
1000
90
80
60
1200
0
200
400
600
800
fOUT (MHz)
200
400
600
800
fOUT (MHz)
1000
1200
2000A G25
1200
DIGITAL AMPLITUDE = 0dBFS
LIN_GN = 100%
LIN_GN = 75%
LIN_GN = 50%
LIN_DIS = 1
70
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A G24
–145
LTC2000A-16 Single-Tone NSD
vs fOUT and IOUTFS
DIGITAL AMPLITUDE = 0dBFS
fDAC = 2.7Gsps
12.5Ω TOTAL OUTPUT LOAD
–150
60mA
–155
40mA
–160
–160
–170
1000
SFDR vs fOUT and Linearization
Setting, fDAC = 2.7Gsps
NSD (dBm/Hz)
HD3 (dBc)
600
800
fOUT (MHz)
80
50
1200
–155
20mA
–165
–165
0
1000
DIGITAL AMPLITUDE = 0dBFS
40mA FULL-SCALE CURRENT
0dBm SINGLE-TONE
–150
NSD (dBm/Hz)
60
400
60
–145
DIGITAL AMPLITUDE = 0dBFS
LIN_GN = 100%
LIN_GN = 75%
LIN_GN = 50%
LIN_DIS = 1
70
200
90
DIGITAL AMPLITUDE = –6dBFS
F2 = F1 + 1.25MHz
LTC2000A-16 Single-Tone NSD
vs fOUT and fDAC
80
0
2000A G23
HD3 vs fOUT and Linearization
Setting, fDAC = 2.7Gsps
90
50
100
LIN_GN = 100%
LIN_GN = 75%
LIN_GN = 50%
LIN_DIS = 1
2000A G22
100
0dBFS
–6dBFS
–12dBFS
–16dBFS
2000A G21
70
LIN_DIS = 0
LIN_GN = 75%
0
80
60
1200
2-Tone IMD vs fOUT and
Linearization Setting,
fDAC = 2.7Gsps
100
70
60
90
70
SFDR (dBc)
IMD (dBc)
100
110
IMD (dBC)
DIGITAL AMPLITUDE = –6dBFS
F2 = F1 + 1.25MHz
2.7Gsps
2.0Gsps
1.6Gsps
1.25Gsps
F2 = F1 + 1.25MHz
LIN_DIS = 1
2000A G20
2-Tone IMD vs fOUT and fDAC with
Default Linearization
110
1000
2-Tone IMD vs fOUT and Digital
Amplitude, fDAC = 2.7Gsps
100
2.7Gsps
2.0Gsps
1.6Gsps
1.25Gsps
70
1200
110
DIGITAL AMPLITUDE = –6dBFS
F2 = F1 + 1.25MHz
LIN_DIS = 1
IMD (dBC)
100
IMD (dBC)
HD3 (dBc)
HD3 vs fOUT and Digital Amplitude
(dBFS), fDAC = 2.7Gsps
2.7Gsps
1.25Gsps
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A G26
–170
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A G27
2000af
For more information www.linear.com/LTC2000A
11
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
LTC2000A-16
Single Carrier DOCSIS Low Band
Wideband ACLR, fDAC = 2.7Gsps
–11.4dBm/6MHz
–30
–40
–60
–70
–80
–88.9dBm/6MHz
–90
–40
–50
–90.1dBm/6MHz
–70
–80
–110
–120
–120
50 150 250 350 450 550 650 750 850 950 1050
FREQUENCY (MHz)
–40
–50
–13.9dBm/6MHz
–60
–70
–80
–88.5dBm/6MHz
–83.4dBm/6MHz
–90
–100 –86.7dBm/6MHz
–110
–120
50 150 250 350 450 550 650 750 850 950 1050
FREQUENCY (MHz)
50 150 250 350 450 550 650 750 850 9501050
FREQUENCY (MHz)
2000A G28
2000A G29
2000A G30
Single Carrier DOCSIS Low Band
Narrowband ACLR, fDAC = 2.7Gsps
Single Carrier DOCSIS Mid Band
Narrowband ACLR, fDAC = 2.7Gsps
Single Carrier DOCSIS High Band
Narrowband ACLR, fDAC = 2.7Gsps
–20
RBW = 30kHz
VBW = 3kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 24s
–30
–40
–50
–30
–40
–50
10dB/DIV
–60
–70
–80
–20
RBW = 30kHz
VBW = 3kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 24s
–60
–70
–80
–40
–50
–60
–70
–80
–90
–90
–90
–100
–100
–100
–110
–110
–110
–120
173
–120
523
183
193
203
213
FREQUENCY (MHz)
223
2000A G31
CARRIER POWER = –11.40dBm, CENTER FREQ = 200MHz
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
12
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–99.97dBm
–95.69dBm
–96.89dBm
–96.68dBm
–96.19dBm
533
UPPER
–95.33dBm
–95.18dBm
–95.80dBm
–96.60dBm
–95.94dBm
543
553
563
FREQUENCY (MHz)
RBW = 30kHz
VBW = 3kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 24s
–30
10dB/DIV
–20
10dB/DIV
RBW = 20kHz
VBW = 2kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 60s
–30
–84.4dBm/6MHz
–100
–110
–20
RBW = 20kHz
VBW = 2kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 60s
–60
–90
–87.2dBm/6MHz
–100
–11.8dBm/6MHz
–30
10dB/DIV
10dB/DIV
–50
–20
RBW = 20kHz
VBW = 2kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 60s
Single Carrier DOCSIS High Band
Wideband ACLR, fDAC = 2.7Gsps
10dB/DIV
–20
Single Carrier DOCSIS Mid Band
Wideband ACLR, fDAC = 2.7Gsps
573
–120
953
963
2000A G32
973
983
993
FREQUENCY (MHz)
1003
2000A G33
CARRIER POWER = –11.85dBm, CENTER FREQ = 550MHz CARRIER POWER = –13.95dBm, CENTER FREQ = 980MHz
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–95.92dBm
–91.68dBm
–94.58dBm
–94.54dBm
–94.17dBm
UPPER
–92.58dBm
–91.81dBm
–94.33dBm
–94.36dBm
–94.16dBm
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–89.76dBm
–87.24dBm
–92.87dBm
–93.36dBm
–92.65dBm
UPPER
–90.00dBm
–86.24dBm
–92.16dBm
–92.29dBm
–91.52dBm
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
LTC2000A-16
32-Carrier DOCSIS Low Band
Wideband ACLR, fDAC = 2.7Gsps
–50
–60
–70
–40
–50
–60
–80
–90
–30
RBW = 20kHz
VBW = 2kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 60s
–40
–50
–60
–70
10dB/DIV
10dB/DIV
–30
RBW = 20kHz
VBW = 2kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 60s
–40
10dB/DIV
–30
32 Carrier DOCSIS High Band
Wideband ACLR, fDAC = 2.7Gsps
32-Carrier DOCSIS Mid Band
Wideband ACLR, fDAC = 2.7Gsps
–80
–90
–70
–80
–90
–100
–100
–100
–110
–110
–110
–120
–120
–120
–130
–130
50 150 250 350 450 550 650 750 850 950 1050
FREQUENCY (MHz)
2000A G34
2000A G35
32-Carrier DOCSIS Low Band
Narrowband ACLR, fDAC = 2.7Gsps
32 Carrier DOCSIS Mid Band
Narrowband ACLR, fDAC = 2.7Gsps
–40
–50
–60
–70
–40
–50
–60
10dB/DIV
10dB/DIV
–30
RBW = 30kHz
VBW = 3kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 24s
–80
–90
32 Carrier DOCSIS High Band
Narrowband ACLR, fDAC = 2.7Gsps
–30
RBW = 30kHz
VBW = 3kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 24s
–40
–50
–60
–70
–80
–90
–80
–90
–100
–100
–110
–110
–110
–120
–120
–120
–130
359
–130
473
379
389
399
FREQUENCY (MHz)
409
2000A G37
CARRIER POWER = –28.77dBm, CENTER FREQ = 386MHz
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–39.64dBm
–29.39dBm
–28.89dBm
–28.83dBm
–28.74dBm
483
UPPER
–103.72dBm
–94.79dBm
–94.96dBm
–95.34dBm
–95.75dBm
493
503
513
FREQUENCY (MHz)
523
RBW = 30kHz
VBW = 3kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 24s
–70
–100
369
50 150 250 350 450 550 650 750 850 950 1050
FREQUENCY (MHz)
2000A G36
10dB/DIV
–30
–130
50 150 250 350 450 550 650 750 850 950 1050
FREQUENCY (MHz)
RBW = 20kHz
VBW = 2kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 60s
–130
773
783
2000A G38
793
803
813
FREQUENCY (MHz)
823
2000A G39
CARRIER POWER = –28.59dBm, CENTER FREQ = 500MHz CARRIER POWER = –29.43dBm, CENTER FREQ = 800MHz
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–102.11dBm
–94.82dBm
–94.42dBm
–94.16dBm
–94.17dBm
UPPER
–39.71dBm
–28.18dBm
–28.96dBm
–29.11dBm
–29.04dBm
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–97.92dBm
–89.60dBm
–89.45dBm
–89.52dBm
–89.38dBm
UPPER
–40.25dBm
–29.56dBm
–29.23dBm
–29.40dBm
–29.56dBm
2000af
For more information www.linear.com/LTC2000A
13
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
LTC2000A-16
128 Carrier DOCSIS Low Band
Wideband ACLR, fDAC = 2.7Gsps
–30
–30
RBW = 20kHz, VBW = 2kHz
REF = –30dBm, ATTEN = 5dB
SWEEP = 60s
–40
–50
–40
–50
–30
RBW = 20kHz, VBW = 2kHz
REF = –30dBm, ATTEN = 5dB
SWEEP = 60s
–50
–70
–70
–90
10dB/DIV
–60
–70
–80
–80
–90
–80
–90
–100
–100
–100
–110
–110
–110
–120
–120
–120
–130
–130
50 150 250 350 450 550 650 750 850 950 1050
FREQUENCY (MHz)
–60
–30
–40
–50
10dB/DIV
–70
–80
–90
2000A G42
157 Carrier Tones with Gap Channel
Narrowband ACLR, fDAC = 2.7Gsps
–30
RBW = 30kHz, VBW = 3kHz
REF = –30dBm, ATTEN = 5dB
SWEEP = 24s
–40
–50
–60
–60
–70
–70
10dB/DIV
–50
30 130 230 330 430 530 630 730 830 9301030
FREQUENCY (MHz)
157 Carrier DOCSIS Gap Channel
Narrowband ACLR, fDAC = 2.7Gsps
RBW = 30kHz
VBW = 3kHz
REF = –30dBm
ATTEN = 5dB
SWEEP = 24s
–40
–130
30 130 230 330 430 530 630 730 830 930 1030
FREQUENCY (MHz)
2000A G41
128 Carrier DOCSIS Low Band
Narrowband ACLR, fDAC = 2.7Gsps
–30
RBW = 20kHz, VBW = 2kHz, REF = –30dBm,
ATTEN = 5dB, SWEEP = 60s, LIN_DIS = 1
–40
–60
2000A G40
10dB/DIV
157 Carrier Tones with Gap Channel
Wideband ACLR, fDAC = 2.7Gsps
–60
10dB/DIV
10dB/DIV
157 Carrier DOCSIS Gap Channel
Wideband ACLR, fDAC = 2.7Gsps
–80
–90
–80
–90
–100
–100
–100
–110
–110
–110
–120
–120
–120
–130
805
–130
481
–130
510
815
825
835
845
FREQUENCY (MHz)
855
CARRIER POWER = –37.77dBm, CENTER FREQ = 832MHz
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
14
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–47.75dBm
–38.07dBm
–37.57dBm
–37.39dBm
–37.27dBm
491
2000A G43
UPPER
–104.61dBm
–95.82dBm
–95.53dBm
–95.43dBm
–95.18dBm
501
511
521
FREQUENCY (MHz)
531
2000A G44
GAP CHANNEL POWER = –96.61dBm, CENTER FREQ = 508MHz
OFFSET FREQ
3.375MHz
6.375MHz
12.00MHz
18.00MHz
24.00MHz
BW
750kHz
5.25MHz
6MHz
6MHz
6MHz
LOWER
–47.17dBm
–36.79dBm
–36.58dBm
–36.76dBm
–36.76dBm
RBW = 30kHz, VBW = 3kHz
REF = –30dBm, ATTEN = 10dB
SWEEP = 24s, LIN_DIS = 1
520
515
FREQUENCY (MHz)
525
2000A G45
UPPER
–48.00dBm
–36.67dBm
–36.56dBm
–36.60dBm
–36.69dBm
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
LTC2000A-16 Single Carrier WCDMA
ACLR vs fOUT, fDAC = 2.7Gsps
LTC2000A-16 Single Carrier WCDMA
ACLR at 350MHz, fDAC = 2.7Gsps
RBW = 30kHz
VBW = 3kHz
REF = –20dBm
ATTEN = 5dB
SWEEP = 24s
–40
–50
–60
10dB/Hz
ACLR (dBc)
–60
–70
–80
–90
–100
–80
–90
–70
200
400
600
800
fOUT (MHz)
1000
–130
323
333
OFFSET FREQ
5.00MHz
10.00MHz
15.00MHz
20.00MHz
25.00MHz
343
353
363
FREQUENCY (MHz)
1
0.5
4096
8192
CODE
BW
3.84MHz
3.84MHz
3.84MHz
3.84MHz
3.84MHz
LOWER
–95.14dBm
–95.76dBm
–95.41dBm
–95.69dBm
–95.00dBm
100
0.1
1
10
OFFSET FREQUENCY (kHz)
2000A G48
LTC2000A-14 Single-Tone NSD
vs fOUT and fDAC
–145
DIGITAL AMPLITUDE = 0dBFS
40mA FULL-SCALE CURRENT
0dBm SINGLE TONE
–150
0
–1.0
–8192
–155
–160
–165
–4096
0
4096
1000
UPPER
–94.76dBm
–96.22dBm
–95.10dBm
–96.41dBm
–96.69dBm
8192
CODE
2000A G49
–160
2000A G47
–0.5
0
–150
–180
0.01
373
NSD (dBm/Hz)
1.0
DNL (LSB)
INL (LSB)
2
–4096
–140
LTC2000A-14 Differential
Nonlinearity (DNL)
LTC2000A-14 Integral Nonlinearity
(INL)
–2
–8192
–130
CARRIER POWER = –16.44dBm, CENTER FREQ = 350MHz
2000A G46
–1
–120
–170
–120
1200
0
–110
–110
ADJ CHANNEL
2ND ADJ CHANNEL
5TH ADJ CHANNEL
0
–100
PHASE NOISE (dBc/Hz)
–30
–50
Additive Phase Noise,
fOUT = 65MHz, fDAC = 2.7Gsps
2000 G50
–170
2.7Gsps
1.25Gsps
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A G51
2000af
For more information www.linear.com/LTC2000A
15
LTC2000A
Typical Performance Characteristics
IOUTFS = 40mA, TA = 25°C, AVDD18 = DVDD18 = 1.86V,
AVDD33 = DVDD33 = 3.3V, RLOAD = 12.5Ω, LIN_DIS = 0, LIN_ GN = 75% unless otherwise noted.
LTC2000A-11 Differential
Nonlinearity (DNL)
0.50
0.5
0.25
0
–145
–150
0
–155
–160
DIGITAL AMPLITUDE = 0dBFS
40mA FULL-SCALE CURRENT
0dBm SINGLE TONE
–0.25
–0.5
–1.0
–1024
LTC2000A-11 Single-Tone NSD
vs fOUT and fDAC
NSD (dBm/Hz)
1.0
DNL (LSB)
INL (LSB)
LTC2000A-11 Integral Nonlinearity
(INL)
–165
–512
0
512
1024
–1.50
–1024
–512
0
512
1024
CODE
CODE
2000A G53
2000A G52
–170
2.7Gsps
1.25Gsps
0
200
400
600
800
fOUT (MHz)
1000
1200
2000A G54
Pin Functions
AVDD18: 1.8V Analog Supply Voltage Input. 1.8V to 1.92V.
AVDD33: 3.3V Analog Supply Voltage Input. 3.135V to
3.465V.
CKP, CKN: DAC Sample Clock Inputs. Maximum clock
frequency (fDAC) is 2700MHz. Clock signal should be AC
coupled.
CS: Serial Interface Chip Select Input. When CS is low,
SCK is enabled for shifting data on SDI into the register.
When CS is taken high, SCK is disabled and SDO is high
impedance.
DAP[15:0], DAN[15:0]: Port A LVDS Data Inputs. Maximum
data rate is 1.35Gbps. Port A is used only in dual-port
mode. Connect to GND if not used. The data input format
is two’s complement.
DBP[15:0], DBN[15:0]: Port B LVDS Data Inputs. Maximum
data rate is 1.35Gbps. In single-port mode, only Port B is
used. In dual-port mode, the sample from Port B appears
at IOUTP/N one cycle after the sample from Port A. The data
input format is two’s complement.
DCKIP, DCKIN: LVDS Data Clock Inputs. Maximum
frequency (fDCKI) is 675MHz. In dual-port mode, fDCKI =
fDAC/4. In single-port mode, fDCKI = fDAC/2
16
DCKOP, DCKON: LVDS Data Clock Outputs. Maximum
frequency is 675MHz. Select frequency (fDAC/4 or fDAC/2),
output current (3.5mA or 7mA), and termination (none or
100Ω) using register 0x02.
DVDD18: 1.8V Digital Supply Voltage Input. 1.8V to 1.92V.
DVDD33: 3.3V Digital Supply Voltage Input. 3.135V to
3.465V.
FSADJ: Full-Scale Adjust Pin. The DAC full-scale current is
16 • (VREFIO/RFSADJ). Connect a 500Ω resistor from FSADJ
to GND to set the full-scale current to 40mA.
GND: Ground.
IOUTP, IOUTN: DAC Analog Current Outputs. Differential
output is nominally ±40mA. Maximum update rate is
2.7Gsps. The output current is evenly divided between
IOUTP and IOUTN when the two’s compliment DAC code is
set to mid-scale (all zeros).
PD (Pin S1): Active Low Power-Down Input. When PD is
low, the LTC2000A supply current is less than 440µA. To
exit power-down mode switch PD high to SVDD.
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Pin Functions
REFIO: Reference Voltage Input or Output. The 1.25V
internal reference is available at the pin through a 10k
internal resistor. May be overdriven with an external reference voltage between 1.1V and 1.4V.
SCK: Serial Interface Clock Input. Maximum frequency
is 50MHz.
SDI: Serial Interface Data Input. Data on SDI is clocked
in on the rising edge of SCK.
SDO: Serial Interface Data Output. Data is clocked out onto
SDO by the falling edge of SCK. SDO is high impedance
when CS is high.
SVDD: SPI Supply Voltage Input. 1.71V to 3.465V.
TSTP, TSTN: Test Output Pins. May be optionally used to
measure internal temperature or timing of LVDS inputs.
See Measuring Internal Junction Temperature and Measuring LVDS Input Timing Skew sections in Applications
Information. Use SPI internal registers 0x18 and 0x19 to
control TSTP/N. Connect to GND if not used.
Note: For pin locations, refer to the Pin Locations section
of this data sheet.
Block Diagram
TSTP/N
PD
JUNCTION
TEMPERATURE
CS
PATTERN
GENERATOR
SCK SDI
SVDD
SDO
SPI
DAP/N[15:0]
DDR DATA FLIP-FLOPS
LVDS RECEIVERS
DBP/N[15:0]
IOUTP
50Ω
16-BIT DAC
4:1
50Ω
IOUTN
GAIN
ADJUST
DCKIP/N
CLOCK
SYNC
DELAY
ADJUST
CLK DIVIDER
÷2 OR ÷4
DCKOP/N
AVDD18
DVDD18
FSADJ
CLK
RECEIVER
AVDD33
DVDD33
GND
CKP/N
REFIO
10k
REF
2000A BD
2000af
For more information www.linear.com/LTC2000A
17
LTC2000A
Timing Diagrams
t1
t2
SCK
t3
1
t6
t4
2
3
15
16
t10
SDI
t5
t7
CS
t13
Hi-Z
SDO
Hi-Z
2000A F01
Figure 1. Serial Interface Timing
t12
t12
t11
t11
DAP/N, DBP/N
DCKI
2000A F02
Figure 2. LVDS Interface Timing (DCKI_Q = 0, DCK_TADJ = 000)
t11
t11
DAP/N, DBP/N
DCKI
2000A F03
t12
t12
Figure 3. LVDS Interface Timing (DCKI_Q = 1)
Operation
Introduction
The LTC2000A is a family of 2.7Gsps current steering
DACs. Three resolutions (16-, 14-, 11-bit) are available
in a 170-lead BGA package. The LTC2000A features high
output bandwidth and output current, while maintaining
a clean output spectrum with low spurs, making it ideal
for generating high frequency or broadband signals. The
LTC2000A output current is nominally 40mA and is a
scaled (16x) replica of the current flowing out of the FSADJ
pin (nominally 2.5mA). The high output current allows
18
flexibility in the output impedance, and the high FSADJ
current and low scaling factor give excellent close-in phase
noise performance.
The LTC2000A has two 16-, 14-, 11-bit wide LVDS or
DHSTL-compatible parallel data input ports (DAP/N,
DBP/N). Each data input port is capable of receiving two’s
complement data at up to 1.35Gbps using a double data
rate (DDR) data input clock (DCKIP/N) at up to 675MHz.
The DDR data input clock may be either in quadrature or
in phase with the data arriving on the data input ports.
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
After incoming data is sampled by DCKIP/N, an internal
multiplexer interleaves the data for resampling by the
DAC sample clock (CKP/N). See Figures 4a and 4b. After
a pipeline delay (latency) of up to 11 DAC sample clock
cycles, the rising edges of CKP/N update the DAC code
and a proportional differential output current is steered
between the two outputs (IOUTP/N). Note it takes about 3ns
(aperture delay) from the CKP/N rising edge that updates a
DAC code to the actual IOUTP/N transition for that DAC code.
An internal clock synchronizer monitors the incoming
phase of DCKIP/N and chooses the appropriate phase for
the multiplexer control signals to ensure that the data is
sampled correctly by CKP/N. The LTC2000A also generates
an LVDS clock output (DCKOP/N) by dividing the sample
clock frequency to simplify clocking of the host FPGA
or ASIC. Additional features such as pattern generation,
LVDS loopout, and junction temperature sensing simplify
system development and testing.
The serial peripheral interface (SPI) port allows configuration and read back of the internal registers which control
the above functions.
Dual-Port Mode
In dual-port mode, data is written to both ports A and B
simultaneously and then subsequently interleaved inside
the LTC2000A, allowing DAC output sampling rates of up
to 2.7Gsps. Figures 4a and 4b show a simplified block
diagram and sample waveforms for dual-port operation.
The LVDS data input ports A and B are sampled on both
the falling and rising edges of the DDR data input clock
(DCKIP/N) by four groups of flip-flops. The contents of
these flip-flops are then interleaved by the 4:1 MUX and
sampled by the DAC sample clock (CKP/N) at frequencies
up to 2.7GHz, with data from port A (DAP/N) preceding
data from port B (DBP/N) at the DAC output.
Note that the sample clock (CKP/N) frequency is always
four times the DDR data input clock (DCKIP/N) frequency
in dual-port mode. For example, to use the DAC at 2.7Gsps,
apply a 2.7GHz clock to CKP/N and a 675MHz clock to
DCKIP/N and send data into both ports A and B (DAP/N,
DBP/N) at 1.35Gsps per port.
Latency is defined as the delay from the DCKIP/N transition
that samples a DAC code to the CKP/N rising transition
which causes that sample to appear at the DAC output
IOUTP/N. In dual-port mode the latency from DAP/N to
IOUTP/N is 10 sample clock cycles and the latency from
DBP/N to IOUTP/N is 11 cycles, starting from the CKP/N
rising edge that immediately follows the DCKIP/N transition that sampled the DAC code (Figure 4b).
Single-Port Mode
In single-port mode, data is written to port B (DBP/N) only,
allowing DAC output sampling rates of up to 1.35Gsps.
Figures 4c and 4d show a block diagram and sample
waveforms representing single-port operation. Samples are
written to port B (DBP/N) and sampled on both the falling
and rising edges of the DDR data input clock (DCKIP/N) by
two groups of flip-flops. The contents of these flip-flops
are then interleaved into a single data stream by the 2:1
MUX and sampled by the DAC sample clock (CKP/N) at
frequencies up to 1.35GHz.
Note that in single-port mode the sample clock (CKP/N) frequency is always twice the DDR data input clock (DCKIP/N)
frequency. For example, to use the DAC at 1.35Gsps, apply
a 1.35GHz clock to CKP/N and a 675MHz clock to DCKIP/N
and send data into port B (DBP/N) at 1.35Gsps. In singleport mode, port A (DAP/N) should be grounded. Due to
the design of the internal clock synchronizer in single port
mode, there is a half cycle shift in the single port latency.
The latency from DBP/N to IOUTP/N in single-port mode is
7.5 sample clock cycles, starting from the CKP/N falling
edge that immediately follows the DCKIP/N transition that
sampled the DAC code (Figure 4d).
2000af
For more information www.linear.com/LTC2000A
19
LTC2000A
Operation
DAP/N[15:0]
N
IOUTP
N+1
N+2
4:1 MUX
DBP/N[15:0]
50Ω
16-BIT DAC
50Ω
IOUTN
N+3
CLOCK
SYNC
DCKIP/N
CKP/N
2000A F04a
Figure 4a. Simplified Block Diagram – Dual-Port Operation
DAP/N[15:0]
N
N+2
N+4
N+6
N+8
N+10
DBP/N[15:0]
N+1
N+3
N+5
N+7
N+9
N+11
DCKIP/N
CKP/N
1
2
3
4
5
6
7
8
IOUTP/N
9
10
11
N
10 CYCLE LATENCY
11 CYCLE LATENCY
N+1
2000A F04b
Figure 4b. Sample Waveforms – Dual-Port Operation
20
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
DBP/N[15:0]
IOUTP
N
2:1 MUX
N+1
50Ω
16-BIT DAC
50Ω
IOUTN
CLOCK
SYNC
DCKIP/N
CKP/N
2000 F04c
Figure 4c. Simplified Block Diagram – Single-Port Operation
DBP/N[15:0]
N
N+1
N+2
N+3
N+4
N+5
N+6
N+6
N+7
DCKIP/N
CKP/N
0.5
1.5
2.5
3.5
4.5
5.5
6.5
IOUTP/N
7.5
N
7.5 CYCLE LATENCY
2000A F04d
Figure 4d. Sample Waveforms – Single-Port Operation
2000af
For more information www.linear.com/LTC2000A
21
LTC2000A
Operation
Serial Peripheral Interface (SPI)
Power-On Reset
The LTC2000A uses an SPI/MICROWIRE-compatible
3-wire serial interface to configure and read back internal
registers. The SVDD pin is the power supply for the SPI
interface (nominally 1.8V or 3.3V). The CS input is level
triggered. When this input is taken low, it acts as a chipselect signal, enabling the SDI and SCK buffers and the
SPI input register. After the falling edge of CS, the first
data byte clocked into SDI by the rising edges of SCK is
the command byte. The first bit of the command byte
signifies a read (R/W = 1) or write (R/W = 0) operation.
The next seven bits contain the register address, which
completes the command byte.
The internal power-on reset circuit will reset the LTC2000A
upon power up and clear the output to mid-scale when
power is first applied, making system initialization consistent and repeatable. All internal registers are reset to
0x00, with the exception of register address 0x08, which
resets to 0x08. A software reset can also be applied by
using the SPI interface to load 0x01 into register address
0x01, setting SW_RST to 1 (see Table 1). Note that the
SW_RST bit is automatically cleared when CS returns
high. It is recommended that users perform a software
reset once all power supplies are stable.
Power Down
The next byte transferred after the command byte is the
data byte. For write operations, the data byte is written to
the SPI register specified by the register address set in
the command byte. During read operations, the data byte
is ignored, and the contents of the selected SPI register
are clocked out onto the SDO pin by the falling edges of
SCK. During write operations, SDO will be low. When CS
goes high, SDO is high impedance. Figure 5 shows the
SPI command and data input.
Users wishing to save power when the DAC is not being
used may reduce the supply current to less than 440µA by
pulling the PD pin to GND or by writing to register 0x01
to set FULL_PD = 1. Alternatively, users may power down
unused portions of the chip individually using DAC_PD,
CK_PD, DCKO_DIS, DCKI_EN, DA_EN, and DB_EN in
registers 0x01, 0x02, 0x03, and 0x04 (see Table 1).
Reference Operation
Users wishing to transfer multiple bytes of data at once
may do so, with the address for each subsequent byte
automatically incremented internally. The address will
continue to increment until CS goes high or until address
bits A[4:0] reach 0x1F, after which subsequent bytes will
continue to be written to the same address.
The LTC2000A has a 1.25V internal bandgap voltage reference that drives the REFIO pin through a 10k internal
resistor, and should be buffered if driving any additional
external load. For noise performance, a 0.1µF capacitor
to GND is recommended on the REFIO pin, but is not
required for stability.
Reserved address and bit locations should not be written
with any value other than zero. Table 11 contains a full
description of all internal SPI registers and can be found
in the SPI Register Summary section.
In the case where an external reference would be preferred,
the external reference is simply applied to the REFIO pin
and overdrives the internal reference. The acceptable
external reference range is 1.1V to 1.4V.
COMMAND BYTE
READ/WRITE
R/W
A6
REGISTER ADDRESS
A5
A4
A3
A2
DATA BYTE
A1
A0
D7
D6
D5
D4
D3
D2
D1
D0
2000A F05
Figure 5. SPI Command and Data Input
22
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
Table 1. Power-On Reset and Power-Down SPI Registers
ADDRESS BIT
0x01
0x02
NAME
DESCRIPTION
0
SW_RST
Software Reset. Set SW_RST = 1 to restore all registers to their power-on reset state. SW_RST is automatically cleared
when CS returns high. All registers reset to 0x00, except address 0x08 which resets to 0x08.
1
DAC_PD
DAC Power Down. Set DAC_PD = 1 to power down the DAC and FSADJ bias circuits.
2
FULL_PD
Full Power Down. Set FULL_PD = 1 to power down all active circuits on the chip and reduce the supply current to less
than 100µA.
0
CK_PD
4
DCKO_DIS
CKP/N Clock Receiver Power Down. CKP/N clock receiver is powered down when CK_PD = 1.
0x03
0
DCKI_EN
0x04
0
DA_EN
DAC Data Port A LVDS Receiver Enable. Set DA_EN = 1 to enable port A (DAP/N) LVDS receivers. For DA_EN = 0,
port A LVDS receivers are powered down and port A data will be zeroes.
1
DB_EN
DAC Data Port B LVDS Receiver Enable. Set DB_EN = 1 to enable port B (DBP/N) LVDS receivers. For DB_EN = 0,
port B LVDS receivers are powered down and port B data will be zeroes.
DCKOP/N Output Disable. Set DCKO_DIS = 1 to power down the DCKO LVDS transmitter. For DCKO_DIS = 1,
DCKOP/N are high impedance.
DCKIP/N Clock Receiver Enable. Set DCKI_EN = 1 to enable the DCKI clock receiver.
Note: Registers 0x01 to 0x04 reset to 0x00 (default).
Setting the Full-Scale Current
The full-scale DAC output current (IOUTFS) is nominally
40mA, but can be adjusted as low as 10mA or as high as
60mA. The full-scale current is set by placing an external
resistor (RFSADJ) between the FSADJ pin and GND. An
internal reference control loop amplifier sets the current
flowing through RFSADJ such that the voltage at FSADJ
is equal to the voltage at REFIO, which is typically 1.25V.
IOUTFS is set as a scaled replica of the current flowing out
of the FSADJ pin (IFSADJ):
IFSADJ =
VREFIO
RFSADJ
IOUTFS =16 •IFSADJ •
256
256+GAIN_ ADJ
Changing GAIN_ADJ to 0x1F (+31) will decrease the current by 10.8% to 35.7mA. Changing GAIN_ADJ to 0x20
(–32) will increase the current by 14.3% to 45.7mA.
Note that GAIN_ADJ appears in the denominator of the
equation for IOUTFS, so the adjustment resolution varies
from 0.5% to 0.3% per step. The circuit shown in Figure
6 may be used to vary the full-scale output current beyond
the range of the GAIN_ADJ register.
DAC linearity and harmonic distortion may be degraded
when using full-scale currents other than 40mA. The fullscale current must not exceed 60mA, and is recommended
to be at least 10mA.
LTC2000A
REF
LTC2630-LM12
where GAIN_ADJ is a 6-bit two’s complement number from
–32 to 31 (nominally 0) which can be programmed using
SPI register 0x09 as shown in Table 2. For example, for
RFSADJ = 500Ω, VREFIO = 1.25V, and GAIN_ADJ = 0x00,
the control loop will force 1.25V at the FSADJ pin, causing
2.5mA to flow through RFSADJ. IOUTFS will then be set to
16 • 2.5mA = 40mA.
10k
0V TO 2.5V
1.25V
1k
REFIO
–
FSADJ
+
500Ω
2000A F06
Figure 6. LTC2000A Full-Scale Adjust from 20mA to 60mA
2000af
For more information www.linear.com/LTC2000A
23
LTC2000A
Operation
Table 2. Full-Scale Gain Adjustment
ADDRESS
BIT
NAME
0x09
[5:0]
GAIN_ADJ
DESCRIPTION
GAIN_ADJ (HEX)
GAIN_ADJ (DECIMAL)
GAIN ADJUSTMENT
FULL-SCALE CURRENT
(RFSADJ = 500Ω, VREFIO = 1.25V)
0x1F
+31
89.2%
35.68mA
0x1E
+30
89.5%
35.80mA
—
—
—
—
0x01
+1
99.6%
39.84mA
0x00
0
100.0%
40.00mA
0x3F
–1
100.4%
40.16mA
—
—
—
—
0x21
–31
113.8%
45.51mA
0x20
–32
114.3%
45.71mA
Note: Register 0x09 resets to 0x00 (default).
DAC Transfer Function
The LTC2000A contains an array of current sources
that are steered through differential switches to either
IOUTP or IOUTN, depending on the DAC code programmed
through the LVDS parallel interface. The LTC2000A uses a
16‑/14‑/11‑bit two’s complement DAC code. The complementary current outputs, IOUTP and IOUTN, source current
from 0mA to IOUTFS. For IOUTFS = 40mA (nominal), IOUTP
swings from 0mA (for zero-scale DAC code) to 40mA (for
full-scale DAC code). IOUTN is complementary to IOUTP.
When the DAC code is set to mid-scale (all zeros), IOUTFS
is evenly divided between IOUTP and IOUTN. IOUTP and IOUTN
are given by the following formulas:
The LTC2000A differential output currents typically drive a
resistive load either directly or drive an equivalent resistive
load through a transformer (see the Output Configurations
section). The voltage outputs generated by the IOUTP and
IOUTN outputs currents are then:
VOUTP = IOUTP • RLOAD
VOUTN = IOUTN • RLOAD
VDIFF = VOUTP – VOUTN = (IOUTP-IOUTN) • RLOAD
Substituting the values above gives:
LTC2000A-16:
VDIFF = VREFIO • (RLOAD/RFSADJ) • (2 • CODE + 1)/4096
LTC2000A-14:
VDIFF = VREFIO • (RLOAD/RFSADJ) • (2 • CODE + 1/4)/1024
LTC2000A-16:
IOUTP = IOUTFS • (CODE + 32768)/65536 + IOUTCM
IOUTN = IOUTFS • (32768 – CODE – 1)/65536 + IOUTCM
LTC2000A-11:
VDIFF = VREFIO • (RLOAD/RFSADJ) • (2 • CODE + 1/32)/128
LTC2000A-14:
IOUTP = IOUTFS • (CODE + 8192)/16384 + IOUTCM
IOUTN = IOUTFS • (8192 – CODE – 1/4)/16384 + IOUTCM
Note that the gain of the DAC depends on the ratio of RLOAD
to RFSADJ, and the gain error tempco is affected by the
temperature tracking of RLOAD with RFSADJ.
LTC2000A-11:
IOUTP = IOUTFS • (CODE + 1024)/2048 + IOUTCM
IOUTN = IOUTFS • (1024 – CODE – 1/32)/2048 + IOUTCM
Analog Outputs (IOUTP/N)
–2N-1
2N-1
The DAC code ranges from
to
– 1, with N
being the DAC resolution (16/14/11). IOUTCM is a small,
constant common-mode output current that is equal to
approximately 0.2% full-scale, or 80µA for IOUTFS = 40mA.
24
The two complementary analog outputs (IOUTP/N) have low
output capacitance that, with appropriate RLOAD values,
can achieve high output bandwidths of 2.1GHz. The analog
outputs also have an internal impedance of 50Ω to GND
that will affect the calculation of RLOAD and the output
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
voltage swing of the DAC. For example, loading both IOUTP
and IOUTN with external 50Ω resistors to GND will cause
RLOAD to equal 25Ω. Assuming an IOUTFS of 40mA, VDIFF
will swing between 1V and –1V.
The specified output compliance voltage range is ±1V.
Above 1V, the differential current steering switches will
start to approach the transition from saturation to linear
region and degrade DAC linearity. Below –1V protection
diodes will limit the swing of the DAC. Small voltage
swings and low common-mode voltages typically result
in the best distortion performance.
DAC Sample Clock (CKP/N)
The DAC sample clock (CKP/N) is used to update the
LTC2000A outputs at rates of up to 2.7Gsps. Provide a
clean, low jitter differential clock at up to 2.7GHz on pins
CKP/N (see Generating the DAC Sample Clock section).
The DC bias point of CKP/N is set internally through a
5kΩ impedance. A 0dBm DAC sample clock should be
sufficient to obtain the performance shown in the Typical
Performance Characteristics section. For best jitter and
phase noise, AC couple a differential clock onto CKP/N with
balanced duty cycle and the highest possible amplitude
and slew rate.
Use SPI register 0x02 to control the DAC sample clock
receiver (Table 3). The LTC2000A contains a clock detector
which sets CK_OK = 1 if the DAC sample clock is present
and fDAC > 50MHz. When the sample clock is not present
(CK_OK = 0), the DAC output is forced to mid-scale and
the internal data path is held at reset. Set CK_PD = 1 to
power down the clock receiver and save power when the
DAC is not being used. Note that at power-on reset, the
DAC sample clock receiver is on by default.
Divided Clock Output (DCKOP/N)
The LTC2000A contains a programmable clock divider
and LVDS transmitter which provide a divided version
(either fDAC/4 or fDAC/2) of the DAC sample clock for
use by the host FPGA or ASIC. Use SPI register 0x02 to
control DCKOP/N (Table 3). At power-on reset, the LVDS
transmitter will provide a clock signal at fDAC/4 with a
3.5mA differential output current.
If desired, set DCKO_DIV = 1 to change the divided clock
output frequency to fDAC/2. The output current can be increased to 7mA by setting DCKO_ISEL = 1, and an internal
100Ω differential termination can be enabled by setting
DCKO_TRM = 1. Set DCKO_DIS = 1 to disable the LVDS
transmitter and save power when not in use.
LVDS Data Clock Input (DCKIP/N)
The DAC code data written to the LTC2000A is captured
on both the rising and falling edges of DCKIP/N. For
single-port operation, provide a DDR clock at half the DAC
sample clock frequency (fDCKI = fDAC/2). To use a 1.35GHz
sample clock in single-port mode, provide a 675MHz
clock on DCKIP/N. For dual-port operation, provide a
DDR clock at one quarter the DAC sample clock frequency
(fDCKI = fDAC/4). To use a 2.7GHz sample clock in dual-port
mode, provide a 675MHz clock on DCKIP/N.
Table 3. DAC Sample Clock, and Divided Clock Output SPI Registers
ADDRESS
BIT
NAME
0x02
0
CK_PD
CKP/N Clock Receiver Power Down When CK_PD = 1
DESCRIPTION
1
CK_OK
CKP/N Clock Present Indicator. When CK_OK = 1, clock is present at CKP/N pins and fDAC > 50MHz.
When CK_OK = 0, DAC output is forced to mid-scale. CK_OK is read only.
4
DCKO_DIS DCKOP/N Output Disable. Set DCKO_DIS = 1 to power down the DCKO LVDS transmitter. For DCKO_DIS = 1,
DCKOP/N are high impedance.
5
DCKO_DIV DCKOP/N Divide Select. When DCKO_DIV = 0, fDCKOP/N = fDAC/4. When DCKO_DIV = 1, fDCKOP/N = fDAC/2.
6
DCKO_ISEL DCKOP/N Output Current Select. When DCKO_ISEL = 0, output current is 3.5mA. When DCKO_ISEL = 1,
output current is 7mA.
7
DCKO_TRM DCKOP/N Internal Termination On. When DCKO_TRM = 0, there is no internal termination at DCKOP/N.
When DCKO_TRM = 1, there is 100Ω between DCKOP and DCKON.
Note: Register 0x02 resets to 0x00 (default).
2000af
For more information www.linear.com/LTC2000A
25
LTC2000A
Operation
Table 4. LVDS Clock SPI Registers
ADDRESS
0x03
BIT
NAME
DESCRIPTION
0
DCKI_EN
DCKIP/N Clock Receiver Enable. DCKI_EN = 1 enables LVDS clock receiver.
1
DCKI_OK
DCKIP/N Clock Present Indicator. When DCKI_OK = 1, clock is present at DCKIP/N pins and fDCKIP/N > 25MHz.
When DCKI_OK = 0, DAC output is forced to mid-scale unless pattern generator is enabled (PGEN_EN = 1).
DCKI_OK is read only.
2
DCKI_Q
DCKIP/N Quadrature Phase Select. For DCKI_Q = 0, DCKIP/N should be in phase with DAP/N and DBP/N.
Set DCKI_Q = 1 to use DCKIP/N in quadrature with DAP/N and DBP/N.
[6:4]
DCKI_TADJ
DCKIP/N Delay Adjust. Use with DCKI_Q = 0 to adjust delay of DCKIP/N relative to DAP/N and DBP/N.
For DCKI_Q = 1, DCKIP/N delay matches DAP/N and DBP/N and is unaffected by DCKI_TADJ.
NOMINAL DCKIP/N DELAY
DCKI_TADJ
DCKI_Q = 1
110
0ps
DCKI_Q = 0
230ps
111
0ps
315ps
000
0ps
400ps (Default)
001
0ps
485ps
010
0ps
570ps
Note: Register 0x03 resets to 0x00 (default).
Use SPI register 0x03 to control the LVDS data clock input
(see Table 4). Setting DCKI_EN=1 will enable the LVDS receiver at DCKIP/N. The LTC2000A contains a clock detector
which sets DCKI_OK=1 if the data input clock is present
and has a frequency greater than 25MHz (fDCKI > 25MHz).
When the data clock is not present (DCKI_OK = 0), the
DAC output is forced to mid-scale and the internal data
path is held at reset.
For maximum setup/hold margin, set DCKI_Q = 1 and
provide DCKIP/N in quadrature (90° out of phase) with
the data on DAP/N and DBP/N (Figure 3 in the Timing
Diagrams section). For DCKI_Q = 1, the internal delays
on DCKIP/N, DAP/N, and DBP/N are nominally matched.
Alternatively, it is possible to leave DCKI_Q = 0 and provide
the clock at DCKIP/N in phase with the data on DAP/N
and DBP/N (see Figure 2 of the Timing Diagram section).
In this case, an internal 400ps delay on DCKIP/N is used
to provide setup/hold margin. Note that for DCKI_Q = 0,
supply and temperature variation may reduce the setup/
hold margin on the bus by up to 150ps. If desired, users
may use the DCKI_TADJ bits in register 0x03 to adjust
the 400ps internal DCKIP/N delay with a typical resolution of 85ps.
Board trace lengths on DCKIP/N, DAP/N, and DBP/N must
be carefully matched to ensure that phase alignment is
26
maintained on all inputs. If desired during development,
users may observe the relative timing of neighboring LVDS
inputs on the TSTP/N pins (refer to the Measuring LVDS
Input Timing Skew section).
LVDS Data Input Ports (DAP/N, DBP/N)
The LTC2000A-16/LTC2000A-14/LTC2000A-11 allow for
DAC Code Data to be applied through one or two parallel
16-/14-/11-bit LVDS ports (DAP/N, DBP/N). Each port can
run up to 1.35Gbps using a double-data-rate (DDR) LVDS
data clock (DCKIP/N) at frequencies up to 675MHz. The
data input format is two’s complement.
There are two modes of operation for applying the DAC
code to the LTC2000A — single-port mode and dualport mode. Single port operation uses only LVDS port B
(DBP/N) and allows sample rates of up to 1.35Gsps. Dual
port operation uses both LVDS ports (DAP/N and DBP/N)
and allows sample rates up to 2.7Gsps.
Use SPI register 0x04 to control the LVDS data input
ports (see Table 5). After the clocks have stabilized and
the synchronizer has initialized itself, set DATA_EN = 1
to allow the data from ports A and B to be used to update
the DAC code. Clear DATA_EN = 0 to mute the DAC and
force the DAC code to mid-scale as desired.
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
Table 5. LVDS Data Input SPI Registers
ADDRESS BIT
0x04
NAME
DESCRIPTION
0
DA_EN
DAC Data Port A LVDS Receiver Enable. DA_EN = 1 enables port A receivers. For DA_EN = 0, receivers are powered
down and port A data is 0x0000.
1
DB_EN
DAC Data Port B LVDS Receiver Enable. DB_EN = 1 enables port B receivers. For DB_EN = 0, receivers are powered
down and port B data is 0x0000.
2
DATA_SP
DAC Data Single Port Mode Select. DATA_SP = 1 sets single port mode and only port B data is used to update the DAC
code. DATA_SP = 0 sets dual-port mode and data from both ports A and B are used.
3
DATA_EN
DAC Data Enable. DATA_EN = 0 mutes the DAC output by forcing the DAC code to mid-scale. DATA_EN = 1 allows data
from data ports A and B to be used to update the DAC code.
Note: Register 0x04 resets to 0x00 (default).
For single port operation, set DATA_SP = 1, DA_EN = 0,
DB_EN = 1 and provide data to LVDS port B (DBP/N) only.
For dual port operation leave DATA_SP = 0, set DA_EN =
1 and DB_EN = 1, and provide interleaved data to LVDS
ports A and B (DAP/N, DBP/N). The data on port A will
precede the data on port B at the DAC output.
Clock Synchronizer
Figure 7 shows a simplified block diagram of the internal
clock synchronizer. The synchronizer monitors the incoming phase of DCKIP/N using a pair of internal phase comparators. The synchronizer then automatically adjusts the
DAP/N[15:0]
IOUTP
4:1 MUX
50Ω
DBP/N[15:0]
50Ω
IOUTN
SYNC_PS
SYNC_PH[7:4]
SYNC_PH[3:0]
LOGIC
16-BIT DAC
DCKIP/N
PHASE
CMP
0°
90°
180°
270°
PHASE
CMP
÷4
CKP/N
2000A F07
Figure 7. Simplified Block Diagram — Clock Synchronizer in Dual-Port Mode
2000af
For more information www.linear.com/LTC2000A
27
LTC2000A
Operation
phase of the MUX control signals as needed to track any
slow drift in the phase between the DCKIP/N and CKP/N
due to supply and temperature variation. This ensures that
data is sampled correctly by CKP/N.
Use SPI registers 0x05 and 0x06 (Table 6) to observe and
control the operation of the synchronizer. Upon power-up,
apply clocks to CKP/N and DCKIP/N and set DCKI_EN = 1
(register 0x03) to enable the LVDS data clock receiver.
Allow at least 1ms after the clocks have stabilized for the
synchronizer to initialize, after which the LTC2000A is
ready to accept LVDS input data.
The synchronizer uses phase comparators to monitor
the phase of the data input clock relative to the sample
clock divider which controls the MUX. The outputs of
these phase comparators (SYNC_PH) may be observed
in register 0x06.
The SYNC_PS bits control the phase of the data multiplexer.
For SYNC_MSYN = 0, the SYNC_PS bits are read-only and
are automatically adjusted by the synchronizer as needed,
based upon the phase of DCKIP/N indicated by SYNC_PH.
Users may choose to override the automatic synchronizer
by setting SYNC_MSYN = 1 and writing values manually
to SYNC_PS to set the phase of the internal multiplexer. When using SYNC_MSYN = 1, users must monitor
SYNC_PH and adjust SYNC_PS as needed according to
Table 6. For further details see the Synchronizing Multiple
LTC2000As section.
Table 6. Clock Synchronizer SPI Registers
ADDRESS
BIT
0x05
[1:0]
2
0x06
[7:0]
NAME
DESCRIPTION
SYNC_PS
Synchronizer Phase Select. Selects phase of internal data multiplexer. SYNC_PS is read-only when SYNC_MSYN = 0.
SYNC_MSYN Synchronizer Manual Mode Select. When SYNC_MSYN = 0, SYNC_PS is set automatically by the clock synchronizer
based upon SYNC_PH. When SYNC_MSYN = 1, SYNC_PS must be set by the user.
SYNC_PH
Synchronizer Phase Comparator Outputs. SYNC_PH indicates the phase of the LVDS data clock (DCKIP/N) relative to
the DAC sample clock (CKP/N) divider used to control the data multiplexer. SYNC_PH is read only.
OPTIMAL SYNC_PS SETTING
SYNC_PH
DUAL-PORT MODE
SINGLE-PORT MODE
0x03
10
00
0x04
10
00
0x05
10
00
0x15
10
00
0x25
10
00
0x35
00
10
0x45
00
10
0x55
00
10
0x54
00
10
0x53
00
10
0x52
01
10
0x51
01
10
0x50
01
10
0x40
01
10
0x30
01
10
0x20
11
00
0x10
11
00
0x00
11
00
0x01
11
00
0x02
11
00
Note: Registers 0x05 and 0x06 reset to 0x00 (default).
28
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
Minimizing Harmonic Distortion
The LTC2000A contains proprietary dynamic linearization
circuitry which dramatically reduces 3rd order harmonic
distortion in the DAC output. SPI registers 0x07 and 0x08
are used to control these circuits (see Table 7). Optimal
performance is normally achieved by setting LIN_VMX and
LIN_VMN (register 0x08) to correspond to the maximum
and minimum voltages expected at IOUTP/N. At power-on
reset the default values are 0b1000 and 0b0000, which
are appropriate for IOUTP/N swinging between 500mV and
GND. If an application requires a different voltage swing,
LIN_VMX and LIN_VMN can be programmed by writing
to register 0x08 (see Table 7). For applications in which
IOUTP/N swing below GND, use LIN_VMN = 0b0000.
In some applications where 2-tone intermodulation distortion (IMD) is a critical specification, it may be desired to
vary the amount of 3rd order harmonic correction. For high
sampling frequencies (fDAC > 2Gsps), adjusting LIN_GN
in register 0x07 (see Table 7) can improve 2-tone intermodulation distortion at the expense of higher 3rd order
harmonic distortion. For best IMD performance at high
sampling frequencies, users may also choose to disable
dynamic linearization by setting LIN_DIS = 1. SFDR and
IMD curves in the Typical Performance Characteristics
section show more detail regarding this effect. Note that
for fDAC < 2Gsps, it is recommended to leave the dynamic
linearization enabled.
Measuring LVDS Input Timing Skew
It is important to ensure that the LVDS inputs (DCKIP/N,
DAP/N, DBP/N) are well aligned. Skew between clock and
data lines, for example due to board trace length mismatch
or output timing mismatch inside the host FPGA or ASIC,
will degrade the setup and hold margin of the incoming
data. The LTC2000A includes an internal test multiplexer
which may be used during development to verify timing
alignment by comparing the timing of LVDS inputs one
pair at a time through the TSTP/N pins.
Use SPI register 0x18 to control this test multiplexer (see
Table 8). Be sure TDIO_EN = 0 in register 0x19 and then
set LMX_EN = 1 to enable the test multiplexer output.
The signal from the LVDS data input will be driven onto
TSTP/N by an NMOS differential pair steering a 6.6mA
sink current onto an external load. Connect a pair of 50Ω
resistors from TSTP/N to 3.3V and observe TSTP/N on a
high speed oscilloscope.
Apply clocks to CKP/N and DCKIP/N and apply the pattern
shown in Figure 8 to port B for single-port mode or ports
A and B for dual-port mode. This pattern is designed to
simplify comparison of rising-to-rising and falling-tofalling edge timing for each input pair. Set LMX_ADR to
select a pair of LVDS inputs for timing comparison. Set
LMX_MSEL = 0 to observe the first signal at TSTP/N. Set
LMX_MSEL = 1 to observe the second signal with inverted
output polarity.
For example, to compare DB15P/N to DCKIP/N, first write
0x60 to register 0x18 to set LMX_EN = 1, LMX_ADR =
10000, and LMX_SEL = 0. The signal from DB15P/N will
be driven onto TSTP/N. Write 0x61 to register 0x18 to set
LMX_SEL = 1 and cause DCKIP/N to appear at TSTP/N
with inverted polarity.
Record the skew between the two signals and repeat
this measurement for each pair of inputs. After all pairs
have been measured, add the skews to calculate the total
skew from DCKIP/N to each data input (DAP/N, DBP/N).
In this way the skew of all LVDS data inputs (DAP/N,
DBP/N) relative to DCKIP/N can be accurately measured
to within 100ps.
Note that due to internal delays inside the test multiplexer,
it is only valid to compare timing between neighboring
LVDS pairs using the same LMX_ADR setting. Similarly,
the multiplexer itself contains up to 400ps of skew between
rising and falling edges, so it is only valid to compare the
timing of a rising edge at TSTP/N to another rising edge,
and a falling edge to another falling edge.
Note that Figure 8 shows the suggested input pattern
for the LTC2000A-16. LTC2000A-14 users should apply
codes 0x1555 and 0x2AAA, and LTC2000A-11 users
should apply codes 0x555 and 0x2AA. Also note that for
the LTC2000A‑14 and LTC2000A-11 in dual-port mode,
the timing skew of LVDS port A (DAP/N) cannot be compared to that of the LVDS clock (DCKIP/N) and LVDS port
B (DBP/N), as there is no single test multiplexer address
(LMX_ADR) that enables a timing comparison between
signals DA0N/P and DCKIP/N (see Table 8). It is recommended to keep LMX_EN = 0 during normal operation.
For more information www.linear.com/LTC2000A
2000af
29
LTC2000A
Operation
Table 7. Dynamic Linearization SPI Registers
ADDRESS
0x07
BIT
NAME
DESCRIPTION
0
LIN_DIS
Dynamic Linearization Disable. Disabled when LIN_DIS = 1.
[3:1]
LIN_GN
Dynamic Linearization Gain Select. Changing LIN_GN varies the amount of 3rd order harmonic correction applied to the
DAC output. LIN_GN = 000 is normally optimal.
LIN_GN
0x08
LINEARIZATION PERCENTAGE
110
50%
111
63%
000
75% (default)
001
88%
010
100%
011
113%
100
125%
101
138%
[3:0]
LIN_VMX
Dynamic Linearization Max IOUTP/N Voltage Select. For optimal 3rd order harmonic performance, set LIN_VMX to
correspond to the maximum voltage expected at IOUTP/N. Reset state is LIN_VMX = 1000, which corresponds to 0.51V.
LIN_VMX must be greater than LN_VMN.
[7:4]
LIN_VMN
Dynamic Linearization Min IOUTP/N Voltage Select. For optimal 3rd order harmonic performance, set LIN_VMN to
correspond to the minimum voltage expected at IOUTP/N. Reset state is LIN_VMN = 0000, which corresponds to 0.0V.
LIN_VMN must be less than LN_VMX.
LIN_VMX/N
MAX/MIN VOLTAGE EXPECTED AT IOUTP/N
0000
0.00V (Default for LIN_VMN)
0001
0.16V
0010
0.19V
0011
0.22V
0100
0.25V
0101
0.31V
0110
0.38V
0111
0.44V
1000
0.51V (Default for LIN_VMX)
1001
0.63V
1010
0.75V
1011
0.87V
1100
1.00V
Note: Register 0x07 resets to 0x00 (default). Register 0x08 resets to 0x08 (default).
30
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Operation
Table 8. SPI Registers for Measuring LVDS Input Timing Skew
ADDRESS
BIT
NAME
DESCRIPTION
0x18
0
LMX_MSEL
LVDS Test MUX Select. Set LMX_MSEL high or low to select between a pair of neighboring LVDS signals for
comparison at TSTP/N.
[5:1]
LMX_ADR
LVDS Test MUX Address. Use LMX_ADR to select which pair of LVDS signals will be compared at TSTP/N
(See Below).
6
LMX_EN
LMX_ADR
LVDS Test MUX Enable. Set LMX_EN=1 to compare timing of neighboring signals at TSTP/N.
Ensure TDIO_EN = 0 when LMX_EN = 1.
LTC2000A-16
LTC2000A-14
LTC2000A-11
LMX_MSEL = 0
LMX_MSEL = 1
(INVERTED)
LMX_MSEL = 0
LMX_MSEL = 1
(INVERTED)
DA15N/P
DA12P/N
DA13N/P
DA9P/N
DA10N/P
DA[14:6]N/P
DA[11:3]P/N
DA[12:4]N/P
DA[8:0]P/N
DA[9:1]N/P
LMX_MSEL = 0
LMX_MSEL = 1
(INVERTED)
00000
DA14P/N
00001…01001
DA[13:5]P/N
01010
DA4P/N
DA5N/P
DA2P/N
DA3N/P
—
DA0N/P
01011
DA3P/N
DA4N/P
DA1P/N
DA2N/P
—
—
01100
DA2P/N
DA3N/P
DA0P/N
DA1N/P
—
—
01101
DA1P/N
DA2N/P
—
DA0N/P
—
—
01110
DA0P/N
DA1N/P
—
—
—
—
01111
DCKIP/N
DA0N/P
DCKIP/N
—
DCKIP/N
—
10000
DB15P/N
DCKIN/P
DB13P/N
DCKIN/P
DB10P/N
DCKIN/P
10001
DB14P/N
DB15N/P
DB12P/N
DB13N/P
DB9P/N
DB10N/P
10010…11010
DB[13:5]P/N
DB[14:6]N/P
DB[11:3]P/N
DB[12:4]N/P
DB[8:0]P/N
DB[9:1]N/P
11011
DB4P/N
DB5N/P
DB2P/N
DB3N/P
—
DB0N/P
11100
DB3P/N
DB4N/P
DB1P/N
DB2N/P
—
—
11101
DB2P/N
DB3N/P
DB0P/N
DB1N/P
—
—
11110
DB1P/N
DB2N/P
—
DB0N/P
—
—
11111
DB0P/N
DB1N/P
—
—
—
—
Note: Register 0x18 resets to 0x00 (default).
DAP/N[15:0]
0xAAAA
0x5555
0xAAAA
0x5555
DBP/N[15:0]
0x5555
0xAAAA
0x5555
0xAAAA
DCKI
2000A F08
Figure 8. Sample Pattern for Measuring LVDS Input Timing Skew (LTC2000A-16)
2000af
For more information www.linear.com/LTC2000A
31
LTC2000A
Operation
Measuring Internal Junction Temperature (TJ)
To use this feature, do the following:
The LTC2000A test multiplexer may also be used to connect
internal junction temperature measurement diodes to the
TSTP/N pins. Ensure LMX_EN = 0 (register address 0x18)
and use SPI register 0x19 to set TDIO_EN = 1 to enable
this function (Table 9). There are two methods the user can
choose from to measure internal junction temperature (TJ).
For TDIO_SEL = 0, an unbiased NPN transistor is diodeconnected between the TSTP/N pins (base/collector to TSTP,
emitter to TSTN). This diode is suitable for use with external
temperature sensors such as the LTC2991 or LTC2997.
1.Set DCKO_DIV = 0 in register 0x02, DATA_SP = 0 and
DATA_EN = 0 in register 0x04, and PGEN_EN = 0 in
register 0x1E.
If such a temperature sensor is not available, set
TDIO_SEL = 1 to directly observe a temperature dependent
voltage between TSTP and TSTN. The typical expected
voltage at TSTP is VTSTP = 2.02V – 5.5mV/°C • (TJ – 25°C).
The junction temperature can be calculated as TJ = 25°C +
(2.02V – VTSTP)/(5.5mV/°C). For best accuracy with
TDIO_SEL = 1, use TSTN to sense GND at the bottom of
the diode and calibrate the voltage at a known temperature. Typical uncalibrated accuracy is ±5°C.
Pattern Generator
A 64 sample deep pattern generator is included in the
LTC2000A to simplify system development and debug.
The pattern generator allows the user to send a repeating
64 sample pattern to the DAC, completely independent
of the presence or absence of valid signals on DCKIP/N,
DAP/N, and DBP/N.
Table 9. Internal Junction Temperature SPI Registers
ADDRESS
0x19
BIT
0
1
2.Write 128 bytes of pattern data to address 0x1F (PGEN_D)
to fill the pattern generator with 64 samples. Data is written MSB first, and will be applied to the DAC in the order
written. Data may be written one byte at a time or in larger
multi-byte words. For the LTC2000A‑14 and LTC2000A-11,
data should be left justified with zeros filling the remaining
two (LTC2000A-14) or five (LTC2000A-11) bits.
3.Set PGEN_EN = 1 to start the pattern generator.
4.Wait at least 1ms to ensure that the synchronizer has
initialized.
5.Set DATA_EN = 1 in register 0x04. The DAC will then
begin to output the 64 sample pattern.
The pattern generator will send the repeating 64 sample
pattern to the DAC until the user writes PGEN_EN = 0 or
DATA_EN = 0.
To read back the pattern, set DATA_EN = 0 and PGEN_EN
= 0 and then read 128 bytes from address 0x1F. Note that
the starting point of the pattern may have changed while the
pattern was running. To modify the pattern, set DATA_EN =
0 and PGEN_EN = 0 and write a new 64 sample pattern to
address 0x1F. Ensure PGEN_EN = 0 when reading or writing to address 0x1F, and always read or write an entire 64
sample pattern prior to setting PGEN_EN = 1. See Table 10.
NAME
TDIO_EN
DESCRIPTION
TSTP/N Junction Temperature Diode Enable. Set TDIO_EN = 1 to measure internal junction temperature (TJ) at TSTP/N.
Ensure LMX_EN = 0 when TDIO_EN = 1.
TDIO_SEL Selects which internal temperature diode is observable at TSTP/N.
For TDIO_SEL = 1, the typical voltage at TSTP with respect to TSTN is VTSTP = 2.02V – 5.5mV/°C • (TJ – 25ºC).
Junction temperature can be calculated as TJ = 25°C + (2.02V – VTSTP)/(5.5mV/°C). Typical accuracy is ±5°C.
For TDIO_SEL = 0, an unbiased diode is connected b/w TSTP/N for use with external temperature sensors.
Note: Register 0x19 resets to 0x00 (default).
Table 10 – Pattern Generator SPI Registers
ADDRESS
0x1E
NAME
DESCRIPTION
PGEN_EN Pattern Generator Enable. Set PGEN_EN = 1 to use the internal 64 sample pattern generator to provide data to the DAC.
Set DATA_SP = 0, DCKO_DIV = 0, and DATA_EN = 1 when PGEN_EN = 1.
0x1F
[7:0] PGEN_D
Pattern Generator Data. Write 128 bytes of data to this address to fill the pattern generator with 64 samples.
Data is written MSB first. Reading this location causes the pattern generator data to be shifted out through SDO.
Ensure PGEN_EN = 0 when reading or writing to address 0x1F. Read or write an even number of bytes to address
0x1F prior to setting PGEN_EN = 1 to avoid corrupting the data inside the pattern generator.
Note: Registers 0x1E and 0x1F reset to 0x00 (default).
2000af
32
BIT
0
For more information www.linear.com/LTC2000A
LTC2000A
SPI Register Summary
Table 11. SPI Register List
RESET
VALUE
R/W
Software Reset. SW_RST = 1 resets all registers.
0
R/W
DAC_PD
DAC Power Down. DAC_PD = 1 to power down DAC core.
0
R/W
FULL_PD
Full Power Down. FULL_PD = 1 to power down LTC2000A.
0
R/W
ADDRESS
BIT
NAME
0x00
[7:0]
Reserved
Reserved
0x01
0
SW_RST
1
2
3
Reserved
Reserved
[5:4]
DAC_RES
DAC Resolution Indicator. DAC_RES = 00 for LTC2000A-16. DAC_RES = 01 for LTC2000A-14. 00-16b
01-14b
DAC_RES = 11 for LTC2000A-11. Note that for PD = GND or FULL_PD = 1, DAC_RES = 00.
11-11b
DAC_RES is read only.
[7:6]
Reserved
Reserved
0
CK_PD
CKP/N Clock Receiver Power Down. CK_PD = 1 disables
0
R/W
CKP/N Clock Present Indicator. CK_OK = 1 clock present
0
R
0x02
0x03
0x04
0x05
DESCRIPTION
R
1
CK_OK
[3:2]
Reserved
Reserved
4
DCKO_DIS
DCKOP/N Output Disable. DCKO_DIS = 1 disables
0
R/W
5
DCKO_DIV
DCKOP/N Divide Select. (0 = fDAC/4, 1 = fDAC/2).
0
R/W
6
DCKO_ISEL
DCKOP/N Output Current Select. (0=3.5mA, 1 = 7mA)
0
R/W
7
DCKO_TRM
DCKOP/N Internal Termination On. DCKO_TRM = 1 enables internal 100Ω termination
0
R/W
0
DCKI_EN
DCKIP/N Clock Receiver Enable. DCKI_EN = 1 enables.
0
R/W
1
DCKI_OK
DCKIP/N Clock Present Indicator. DCKI_OK = 1 indicates clock present
0
R
2
DCKI_Q
DCKIP/N Quadrature Phase Select. (0 = In Phase, 1 = Quadrature)
0
R/W
3
Reserved
Reserved
[6:4]
DCKI_TADJ
000
R/W
7
Reserved
0
DA_EN
Port A LVDS Receiver Enable. DA_EN = 1 enables
0
R/W
1
DB_EN
Port B LVDS Receiver Enable. DB_EN = 1 enables
0
R/W
2
DATA_SP
Port Mode Select. (0 = Dual port, 1 = Single port)
0
R/W
0
R/W
00
R/W
0
R/W
DCKIP/N Delay Adjust. (See Table 4)
Reserved
3
DATA_EN
DAC Data Enable. DATA_EN = 0 forces DAC output to mid-scale.
[7:4]
Reserved
Reserved
[1:0]
SYNC_PS
Clock Synchronizer Phase Select.
2
SYNC_MSYN Clock Synchronizer Manual Mode Select.
SYNC_MSYN = 0: SYNC_PS is set automatically.
SYNC_MSYN = 1: SYNC_PS is set by the user.
[7:3]
Reserved
Reserved
0x06
[7:0]
SYNC_PH
Clock Phase Comparator Outputs. (See Table 6)
0x07
0
LIN_DIS
Dynamic Linearization Disable. LIN_DIS = 1 disables.
[3:1]
LIN_GN
Dynamic Linearization Gain Select. (See Table 7)
[7:4]
Reserved
Reserved
[3:0]
LIN_VMX
[7:4]
LIN_VMN
0x08
0x00
R
0
R/W
000
R/W
Dynamic Linearization Max IOUTP/N Voltage Select. (See Table 7)
1000
R/W
Dynamic Linearization Min IOUTP/N Voltage Select. (See Table 7)
0000
R/W
2000af
For more information www.linear.com/LTC2000A
33
LTC2000A
SPI Register Summary
Table 11. SPI Register List (cont)
ADDRESS
0x09
BIT
NAME
DESCRIPTION
RESET
VALUE
R/W
0x00
R/W
0
R/W
0x00
R/W
[5:0]
GAIN_ADJ
DAC Gain Adjustment. (See Table 2)
[7:6]
Reserved
Reserved
0x0A Thru
0x17
[7:0]
Reserved
Reserved
0x18
0
LMX_MSEL
LVDS Test MUX Select. (See Table 8)
[5:1]
LMX_ADR
LVDS Test MUX Address Select. (See Table 8)
6
LMX_EN
LVDS Test MUX Enable. LMX_EN = 1 enables LVDS text MUX. Ensure TDIO_EN = 0 when
LMX_EN = 1.
7
Reserved
Reserved
0
TDIO_EN
TSTP/N Junction Temperature Diode Enable.
TDIO_EN = 1 enables temperature (TJ) measurement.
Ensure LMX_EN = 0 when TDIO_EN = 1.
0
R/W
1
TDIO_SEL
Junction Temperature Select. TDIO_SEL = 0 uses a diode-connected unbiased NPN transistor.
TDIO_SEL = 1 outputs a voltage to calculate internal die temperature using:
TJ = 25°C + (2.02V – VTSTP)/(5.5mV/°C). (See Table 9)
0
R/W
[7:2]
Reserved
Reserved
0x1A Thru
0x1D
[7:0]
Reserved
Reserved
0x1E
0
PGEN_EN
Pattern Generator Enable. PGEN_EN = 1 enables.
0
R/W
0x00
R/W
0x19
[7:1]
Reserved
Reserved
0x1F
[7:0]
PGEN_D
Pattern Generator Data.
0x20 Thru
0x7F
[7:0]
Reserved
Reserved
Applications Information
Sample Start-Up Sequence
The following is an example of a common start-up
sequence.
1. Apply valid supply voltages to AVDD33, DVDD33,
AVDD18, DVDD18 and SVDD.
2. Write 0x01 to address 0x01 to perform a software
reset.
3. Apply a clock to CKP/N at the desired fDAC frequency.
The LTC2000A will generate a clock at DCKOP/N at
fDAC/4.
4. Apply a clock to DCKIP/N at fDAC/4 for dual-port mode
or fDAC/2 for single-port mode.
6. Write to address 0x03 to enable the DCKIP/N LVDS
receiver. Set address 0x03 to 0x01 if the LVDS clock
(DCKI) and data (DA, DB) are in phase with each other.
Set address 0x03 to 0x05 if they are in quadrature.
7. Write 0x06 to address 0x04 for dual-port mode, or
write 0x04 to address 0x04 for single-port mode to
enable the DAP/N and DBP/N LVDS receivers.
8. Wait at least 1ms for the synchronizer to finish
initializing.
9. Write 0x0B to address 0x04 for dual-port mode, or
write 0x0E to address 0x04 for single-port mode to
set DATA_EN = 1.
5. Apply zeroes to ports A and B (DAP/N, DBP/N) for
dual-port mode, or only to port B for single-port mode.
34
For more information www.linear.com/LTC2000A
2000af
LTC2000A
Applications Information
10. Apply desired data pattern to ports A and B (DAP/N,
DBP/N) for dual-port mode, or only to port B for
single-port mode. Port A samples will precede port B
samples at the DAC output when using dual-port
mode.
For any output configuration, any imbalances in the output
impedance between the IOUTP and IOUTN pins results in
asymmetrical signal swings that lead to distortion (mostly
even order). Careful consideration is needed to select the
best output configuration for a given application.
Output Configurations
Generating the DAC Sample Clock
The LTC2000A’s complementary current outputs (IOUTP/N)
source current into an external load referenced to GND.
Output load configuration, component selection, and layout
are critical to the performance of the LTC2000A. For best
AC performance, the output stages should be configured
for differential (or balanced) operation.
For best AC performance, it is important that the DAC
sample clock waveforms be clean, with low phase noise
and good jitter performance, as the phase noise and spurious content of the clock source will appear directly in the
DAC output spectrum.
A differential resistor loaded output is a very simple output
stage. Well matched resistors are connected between GND
and IOUTP/N, with the resistance values setting both the
output swing and non-zero output common-mode voltage
(Figure 9). While it is economical, this type of output stage
can drive only differential loads with impedance levels and
amplitudes appropriate for the DAC outputs.
Differential transformer-coupled output configurations
usually give the best AC performance and provide excellent rejection of common mode distortion and noise over
a broad frequency range. Figure 10 shows a transformer
output configuration that uses a Mini-Circuits TC1-1-13M
and a JTX-2-10T RF transformer for differential to singleended conversion.
LTC2000A
A differential clock should be AC coupled onto the CKP/N
pins, since the DC bias point of CKP/N is set internally to
1V through a 5kΩ impedance. Figure 11 shows the DAC
sample clock receiver input and common-mode voltage
control. While the differential input voltage range of the
clock receiver spans from ±300mV to ±1.8V, a signal with
the highest possible slew rate and amplitude and a balanced duty cycle is recommended. Traces that carry the
differential clock signal need to have accurately controlled
impedance and accurate termination as close to the CKP/N
pins of the LTC2000A as possible.
There are several ways to generate the DAC sample clock.
For lab evaluation and testing, a high quality RF signal
generator can provide a clean high frequency sine wave
that is converted to the DAC sample clock with a 1:1 RF
transformer or balun (see Figure 12).
IOUTP
LTC2000A
AVDD18
IOUTN
R
R
CKP
2000A F09
1V
Figure 9. Differential Resistor Output Load
IOUTP
LTC2000A
•
5k
CKN
MINI-CIRCUITS
TC1-1-13M
MINI-CIRCUITS
JTX-2-10T
5k
•
•
•
GND
2000A F11
IOUTN
2000A F10
Figure 10. Transformer-Based Output Configuration
for Differential to Single-Ended Conversion
Figure 11. DAC Sample Clock Receiver
For more information www.linear.com/LTC2000A
2000af
35
LTC2000A
Applications Information
50Ω
1nF
+
RF SIGNAL
GENERATOR
the default behavior is for the output of DAC Y to update
with sample N one cycle earlier than the output of DAC X.
It is possible to correct this misalignment and synchronize DACs X and Y by adjusting the clock synchronizer
settings to subtract one cycle of latency from DAC X, as
shown in the adjusted waveform at the bottom of Figure
15a. See the Clock Synchronizer section and Figure 7 for
more details on the operation of the clock synchronizer.
MINI-CIRCUITS
TC1-1-13M
•
•
CKP
50Ω
+
1nF
50Ω
LTC2000A
+
100pF
CKN
LTC2000A F12
Figure 12. DAC Sample Clock Generation with an
RF Signal Generator and a 1:1 Balun
A more integrated clock source is one based on a low
phase noise, low jitter PLL. Figure 13 shows how the
DAC sample clock can be generated from the LTC6946,
a high performance PLL with an internal VCO that can
provide output frequencies from 0.37GHz to 5.7GHz. See
the LTC6946 data sheet for details.
Synchronizing Multiple LTC2000As in Dual-Port Mode
In some applications, it is necessary to synchronize multiple LTC2000As to each other such that related samples
arrive at all DAC outputs simultaneously. Figures 14 and
15a show a block diagram and sample waveforms for
such a system in which two DACs (X and Y) are to be
synchronized in dual-port mode.
Note that in this example a small timing skew between
the two data signals at the DCKIP/N pins of DACs X and
Y has caused the DCKIP/N rising edges to arrive on opposite sides of a DAC sample clock (CKP/N) rising edge,
and thus within different CKP/N clock cycles. As a result
LOOP FILTER
CP
2700pF
RZ
453Ω
CI
0.022µF
LF(s)
25
CP
ICP =
11.2mA
VRF+ f
PFD
KPFD
In order to synchronize multiple DACs as shown in Figures 14 and 15a, distribute the DAC sample clock carefully
with matched delays so that it arrives at the CKP/N pins of
all DACs simultaneously. Any remaining timing mismatch
between sample clocks will appear directly as mismatch in
the DAC output timing. Ensure that the timing mismatch
between LVDS data clock signals at the DCKIP/N pins of all
DACs is less than 0.4 cycles of the DAC sample clock, minus any timing mismatch between the DAC sample clocks.
Be sure to maintain sufficient matching between the timing
of the LVDS data inputs (DAP/N, DBP/N) and DCKIP/N for
each DAC to meet the setup and hold time specifications
(t11, t12) in the Timing Characteristics section.
For example, let us consider a system using multiple DACs
at 2.7Gsps in which the sample clock is designed to arrive
at the CKP/N pins of all DACs within 30ps of one another.
The sample clock period is 370ps, so the maximum allowable timing mismatch between the data clock signals at
the DCKIP/N pins of all DACs will be (0.4 • 370ps) – 30ps
= 118ps. For a system using multiple DACs at 1.35Gsps,
the allowable mismatch between DCKIP/N pins will be (0.4
• 740ps) – 30ps = 266ps. In both cases, once the DACs
LTC6946
REF±
R_DIV
(fREF)
÷R = 10
L1
68nH
fREF*
L2
68nH
100pF
50Ω
÷N = 250
15
TUNE
O_DIV
f
KVCO VCO ÷O = 1
CKP
+
N_DIV
RF±
(fRF)
RF+
RF–
100pF 50Ω
+
*CRYSTEK CVHD-950-100.000 100MHz OSCILLATOR
100pF
+
LTC2000A
CKN
2000A F13
Figure 13. DAC Sample Clock Generation with the LTC6946
36
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Applications Information
MATCHED
DELAYS
CLOCK
SOURCE
MATCHED
DELAYS
CKP/N
DCKIP/N
DAC X
LTC2000A
DAP/N
LVDS CLK
LVDS DATA
2000A F14
DBP/N
FPGA
CKP/N
DCKIP/N
DAC Y
LTC2000A
DAP/N
LVDS CLK
LVDS DATA
DBP/N
Figure 14. System with Multiple LTC2000A DACs Synchronized
CKP/N
DAP/N[15:0]
N
N+2
N+4
N+6
N+8
N+10
DBP/N[15:0]
N+1
N+3
N+5
N+7
N+9
N+11
DAC X
DCKIP/N
0.9
CYCLES
DAC Y
DCKIP/N
0.1 CYCLES
10 CYCLES
DAC X
N
IOUTP/N
9 CYCLES
DAC X
IOUTP/N
(ADJUSTED)
N
N+1
10 CYCLES
DAC Y
N
IOUTP/N
N+1
2000A F15a
Figure 15a. Sample Waveforms — Synchronizing Multiple LTC2000As in Dual Port Mode
are synchronized the mismatch in the DAC output timing
will be limited to 30ps.
Once all the DAC sample clocks and LVDS data clocks are
aligned, determine whether any DACs are being updated
one cycle late (such as DAC X in Figure 15a) by determining
whether DCKIP/N is arriving at the DACs within the same
CKP/N clock cycle. To do this in dual-port mode, first use
the phase comparator outputs SYNC_PH and Table 12
to determine the delay from the DCKIP/N rising edge to
the next CKP/N rising edge for each DAC (measured in
sample clock cycles).
Recall that the DCKIP/N timing mismatch must be kept
below 0.4 cycles of the sample clock. If DCKIP/N arrives at both DACs within the same sample clock cycle,
the difference in DCKIP/N to CKP/N delays indicated by
SYNC_PH will equal the actual DCKIP/N timing mismatch,
and thus will be less than 0.4 cycles. If DCKIP/N arrives
at the DACs within different cycles, as in Figure 15a, the
difference in the delays indicated by SYNC_PH will equal
1 cycle minus the actual DCKIP/N timing mismatch, and
thus will be greater than 0.4 cycles. Thus if the difference
between the delays indicated by SYNC_PH is greater than
2000af
For more information www.linear.com/LTC2000A
37
LTC2000A
Applications Information
0.4 cycles, the DCKIP/N rising edges are arriving in different sample clock cycles.
For the example in Figure 15a, we might read 0x25 for
SYNC_PH on DAC X and 0x52 on DAC Y. Table 12 tells us
that the DCKIP/N to CKP/N delay is greater than 0.8 cycles
for DAC X and less than 0.2 cycles for DAC Y, and thus the
difference between them is at least 0.6 cycles. We conclude
that the DCKIP/N rising edge of DAC X must fall within
a later sample clock cycle than that of DAC Y, and thus
that DAC X is being updated one cycle later than DAC Y.
To correct any such misalignment and synchronize the
DACs, consult Table 12 and adjust the SYNC_PS settings
for those DACs which are being updated one cycle late
(DAC X in the above example) by setting the synchronizer
to manual mode (SYNC_MSYN = 1) and overwriting the
SYNC_PS value.
In this example, reading register 0x06 of DAC X shows
SYNC_PH = 0x25 and that the SYNC_PS setting needs to
change from the default (10) to the desired adjusted value
(00), subtracting one cycle from the latency of DAC X (refer
to Table 12). Write 0x04 to register 0x05 of DAC X to set
SYNC_MSYN = 1 and SYNC_PS = 00. The outputs of DAC
X should now align with DAC Y as shown in Figure 15a.
See Table 6 for details regarding the synchronizer registers
0x05 and 0x06. Sample verilog code implementing the
synchronization of multiple LTC2000As using Tables 12
and 13 can be found at:
http://www.linear.com/docs/44845
Synchronizing Multiple LTC2000As in Single-Port Mode
Figure 15b shows sample waveforms for synchronizing two
LTC2000As in single port mode. Synchronizing multiple
LTC2000As in single port mode is essentially the same
Table 12. Adjusting Latency in Dual-Port Mode
PHASE COMPARATOR OUTPUTS
SYNC_PH (REG 0x06)
SYNC_PS SETTING
DELAY FROM DCKIP/N RISING
EDGE TO NEXT CKP/N RISING EDGE
(CKP/N CYCLES)
(DEFAULT)
(ADJUSTED TO REDUCE LATENCY
BY 1 CYCLE)*
0x03
0 to 0.2
10
N/A
0x04
0.2 to 0.4
10
N/A
0x05
0.4 to 0.6
10
N/A
0x15
0.6 to 0.8
10
00
0x25
0.8 to 1.0
10
00
0x35
0 to 0.2
00
N/A
0x45
0.2 to 0.4
00
N/A
0x55
0.4 to 0.6
00
N/A
0x54
0.6 to 0.8
00
01
0x53
0.8 to 1.0
00
01
0x52
0 to 0.2
01
N/A
0x51
0.2 to 0.4
01
N/A
0x50
0.4 to 0.6
01
N/A
0x40
0.6 to 0.8
01
11
0x30
0.8 to 1.0
01
11
0x20
0 to 0.2
11
N/A
0x10
0.2 to 0.4
11
N/A
0x00
0.4 to 0.6
11
N/A
0x01
0.6 to 0.8
11
10
0x02
0.8 to 1.0
11
10
*N/A indicate SYNC_PH values that should not occur if the timing mismatch requirements described above are met. If such a case occurs,
keep SYNC_PS as the default value.
38
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Applications Information
CKP/N
DBP/N[15:0]
N
N+1
N+2
N+3
N+4
N+5
N+6
N+7
N+8
N+9
DAC X
DCKIP/N
0.8 CYCLES
DAC Y
DCKIP/N
0.1 CYCLES
7.5 CYCLES
DAC X
N
N+1
N
N+1
N+2
N
N+1
IOUTP/N
6.5 CYCLES
DAC X
IOUTP/N
(ADJUSTED)
7.5 CYCLES
DAC Y
IOUTP/N
N+2
2000A F15b
Figure 15b. Sample Waveforms — Synchronizing Multiple LTC2000As in Single Port Mode
procedure as when operating in dual port mode—DAC
sample clocks must all be aligned to arrive at the CKP/N
pins of all DACs simultaneously and timing mismatch
between LVDS data clock signals at the DCKIP/N pins of
all DACs must be less than 0.4 cycles of the DAC sample
clock, minus any timing mismatch between the DAC
sample clocks.
To determine whether any DACs are being updated one
cycle late in single port mode, first use the phase comparator outputs SYNC_PH and Table 13 to determine the delay
from the DCKIP/N rising edge to the next CKP/N falling
edge (as opposed to rising edge in dual port mode) for
each DAC. If the difference between the delays indicated
by SYNC_PH is greater than 0.4 cycles, the DCKIP/N rising edges are arriving in different sample clock cycles.
For the example in Figure 15b, we might read 0x15 for
SYNC_PH on DAC X and 0x20 on DAC Y. Table 13 shows
that the DCKIP/N to CKP/N delay is greater than 0.8 cycles
for DAC X and less than 0.1 cycles for DAC Y, and thus the
difference between them is at least 0.7 cycles. This indicates
that DAC X is being updated one cycle later than DAC Y.
Consult Table 13 and use the same procedure described
above in the dual-port mode case to correct the SYNC_PS
settings for those DACs that are updating one cycle late.
In this single port example, writing 0x06 to register 0x05
of DAC X would set SYNC_MSYN = 1 and SYNC_PS = 10,
reducing the latency of DAC X by one cycle and aligning
its output with DAC Y, as shown in Figure 15b.
Note that variations in system temperature or supply
voltage may cause the phase of the data clock (DCKIP/N)
and sample clock (CKP/N) to vary with time. When using
the LTC2000A with SYNC_MSYN = 1, it is recommended
that users monitor SYNC_PH and adjust SYNC_PS using
Tables 12 or 13 as needed to maintain proper alignment.
The synchronization procedures described above also work
for systems with more than two DACs. Simply determine
the minimum DCKIP/N to CKP/N delay of all DACs by
reading SYNC_PH, and then adjust the SYNC_PS settings
to subtract one cycle of latency to those DACs whose
DCKIP/N to CKP/N delays are at least 0.4 cycles more
than the minimum. Sample verilog code implementing the
synchronization of multiple LTC2000As using Tables 12
and 13 can be found at:
http://www.linear.com/docs/44845
2000af
For more information www.linear.com/LTC2000A
39
LTC2000A
Applications Information
Table 13. Adjusting Latency in Single-Port Mode
SYNC_PS SETTING
PHASE COMPARATOR OUTPUTS
SYNC_PH (REG 0x06)
DELAY FROM DCKIP/N RISING
EDGE TO NEXT CKP/N FALLING
EDGE (CKP/N CYCLES)
(DEFAULT)
(ADJUSTED TO REDUCE LATENCY
BY 1 CYCLE)*
0x03
0.5 to 0.6
00
N/A
0x04
0.6 to 0.7
00
10
0x05
0.7 to 0.8
00
10
0x15
0.8 to 0.9
00
10
0x25
0.9 to 1.0
00
10
0x35
0 to 0.1
10
N/A
0x45
0.1 to 0.2
10
N/A
0x55
0.2 to 0.3
10
N/A
0x54
0.3 to 0.4
10
N/A
0x53
0.4 to 0.5
10
N/A
0x52
0.5 to 0.6
10
N/A
0x51
0.6 to 0.7
10
00
0x50
0.7 to 0.8
10
00
0x40
0.8 to 0.9
10
00
0x30
0.9 to 1.0
10
00
0x20
0 to 0.1
00
N/A
0x10
0.1 to 0.2
00
N/A
0x00
0.2 to 0.3
00
N/A
0x01
0.3 to 0.4
00
N/A
0x02
0.4 to 0.5
00
N/A
*N/A indicate SYNC_PH values that should not occur if the timing mismatch requirements described above are met. If such a case occurs,
keep SYNC_PS as the default value.
PCB Layout Considerations
The close proximity of high frequency digital data lines
and high dynamic range, wideband analog signals make
clean printed circuit board design and layout an absolute
necessity. Figures 16 and 17 show a schematic and PCB
layers for an evaluation circuit for the LTC2000A. A single,
solid ground plane should be used, while separate supply
planes for AVDD18, DVDD18, AVDD33, and DVDD33 should
be kept all the way to the individual supply or LDO. All
LVDS input (DCKIP/N, DAP/N, DBP/N) board traces must
be carefully matched to ensure proper phase alignment.
These LVDS inputs should be kept far away from both the
IOUTP/N and CKP/N traces to avoid any data dependent
coupling into the analog output and DAC sample clock.
40
The CKP/N traces should be routed either over the analog
ground plane or over their own section on the ground plane.
These traces also need to have accurately controlled impedance and should be well terminated near the LTC2000A.
The IOUTP/N traces should also be carefully matched to
each other, routed over the ground plane, away from the
LVDS inputs and CKP/N signals.
Bypass capacitors are required on AVDD18, DVDD18, AVDD33,
and DVDD33, and should all be connected to the analog
ground plane. 2.2µF ceramic capacitors with low ESR
are recommended to be placed close to the LTC2000A
with minimum trace lengths. A sample PCB layout and
schematic can be found below.
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Applications Information
3.3V
MATCHED LVDS
DATA AND CLOCK
LINES FROM FPGA
CLOCK
SOURCE
L6
1nH
C66
1pF
C47
10pF
T1
ANAREN
B0430J50100AHF
6
1
GND
2
3
IN
GND
5
4
C41
100pF
SPI
PORTS
DAP[15:0]
DAN[15:0]
J10
DCKIP
J8
DCKIN
DBP[15:0]
DBN[15:0]
K8
DCKOP
K7
DCKON
S1
PD
S2
LTC2000A
CS
S3
SDO
S4
SDI
S5
SCK
R29
50Ω
C42
100pF
C43
0.01µF
R26
50Ω
CKP
47µF
SVDD
AVDD33
DVDD33
AVDD18
DVDD18
1.8V
47µF
R4
TSTP
R3
TSTN
H1
IOUTP
IOUTN
J1
M1
REFIO
M2
FSADJ
CKN
GND
R40
50Ω
R47
50Ω
R46
50Ω
R48
50Ω
2000A F16
R45
500Ω
C40
10µF
Figure 16
Layer 1
Layer 2
2000af
For more information www.linear.com/LTC2000A
41
LTC2000A
Applications Information
Layer 3
42
Layer 4
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Applications Information
Layer 5
Layer 6
2000af
For more information www.linear.com/LTC2000A
43
LTC2000A
Applications Information
Layer 7
Layer 8
Figure 17
44
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Pin Locations (LTC2000A-16)
LTC2000A-16 BGA Pinout
PIN ID
FUNCTION
PIN ID
FUNCTION
A1
GND
B1
GND
A2
CKN
B2
GND
A3
CKP
B3
GND
A4
GND
B4
GND
A5
DVDD18
B5
A6
GND
A7
DAN15
A8
DAP15
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
C1
AVDD18
D1
AVDD18
E1
AVDD18
F1
GND
C2
AVDD18
D2
AVDD18
E2
AVDD18
F2
GND
C3
AVDD18
D3
AVDD18
E3
DVDD18
F3
GND
C4
AVDD18
D4
DVDD18
E4
DVDD18
F4
GND
DVDD18
C5
DVDD18
D5
DVDD18
E5
DVDD18
F5
GND
B6
GND
C6
GND
D6
GND
E6
GND
F6
GND
B7
DAN14
C7
DAN13
D7
DAN12
E7
DAN11
F7
DAN10
B8
DAP14
C8
DAP13
D8
DAP12
E8
DAP11
F8
DAP10
A9
DBN15
B9
DBN14
C9
DBN13
D9
DBN12
E9
DBN11
F9
DBN10
A10
DBP15
B10
DBP14
C10
DBP13
D10
DBP12
E10
DBP11
F10
DBP10
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
G1
GND
H1
IOUTP
J1
IOUTN
K1
GND
L1
GND
M1
REFIO
G2
GND
H2
GND
J2
GND
K2
GND
L2
GND
M2
FSADJ
G3
GND
H3
GND
J3
GND
K3
GND
L3
GND
M3
GND
G4
GND
H4
GND
J4
GND
K4
GND
L4
GND
M4
AVDD33
G5
GND
H5
GND
J5
GND
K5
GND
L5
GND
M5
DVDD33
G6
GND
H6
GND
J6
GND
K6
GND
L6
GND
M6
GND
G7
DAN9
H7
DAN8
J7
DCKON
K7
DAN7
L7
DAN6
M7
DAN5
G8
DAP9
H8
DAP8
J8
DCKOP
K8
DAP7
L8
DAP6
M8
DAP5
G9
DBN9
H9
DBN8
J9
DCKIN
K9
DBN7
L9
DBN6
M9
DBN5
G10
DBP9
H10
DBP8
J10
DCKIP
K10
DBP7
L10
DBP6
M10
DBP5
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
N1
GND
P1
AVDD33
Q1
AVDD33
R1
GND
S1
PD
N2
GND
P2
AVDD33
Q2
AVDD33
R2
GND
S2
CS
N3
GND
P3
AVDD33
Q3
AVDD33
R3
TSTN
S3
SDO
N4
AVDD33
P4
AVDD33
Q4
AVDD33
R4
TSTP
S4
SDI
N5
DVDD33
P5
DVDD33
Q5
DVDD33
R5
GND
S5
SCK
N6
GND
P6
GND
Q6
GND
R6
GND
S6
SVDD
N7
DAN4
P7
DAN3
Q7
DAN2
R7
DAN1
S7
DAN0
N8
DAP4
P8
DAP3
Q8
DAP2
R8
DAP1
S8
DAP0
N9
DBN4
P9
DBN3
Q9
DBN2
R9
DBN1
S9
DBN0
N10
DBP4
P10
DBP3
Q10
DBP2
R10
DBP1
S10
DBPO
2000af
For more information www.linear.com/LTC2000A
45
LTC2000A
Pin Locations (LTC2000A-16)
1
2
3
CKN
CKP
TOP VIEW
4 5 6 7
8
9
10
A
B
GND
C
AVDD18
D
GND
DVDD18
E
F
GND
G
H
J
K
IOUTP
IOUTN
M
REFIO
FSADJ
GND
DVDD33
P
AVDD33
Q
GND
R
GND
S
PD
46
CS
TSTN
TSTP
SDO
SDI
DBN15 DBP15
DAN14 DAP14
DBN14 DBP14
DAN13 DAP13
DBN13 DBP13
DAN12 DAP12
DBN12 DBP12
DAN11 DAP11
DBN11 DBP11
DAN10 DAP10
DBN10 DBP10
DAN9
DAP9
DBN9
DBP9
DAN8
DAP8
DBN8
DBP8
DCKIN
DCKIP
DCKON DCKOP
GND
L
N
DAN15 DAP15
SCK
SVDD
DAN7
DAP7
DBN7
DBP7
DAN6
DAP6
DBN6
DBP6
DAN5
DAP5
DBN5
DBP5
DAN4
DAP4
DBN4
DBP4
DAN3
DAP3
DBN3
DBP3
DAN2
DAP2
DBN2
DBP2
DAN1
DAP1
DBN1
DBP1
DAN0
DAP0
DBN0
DBP0
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Pin Locations (LTC2000A-14)
LTC2000A-14 BGA Pinout
PIN ID
FUNCTION
PIN ID
FUNCTION
A1
GND
B1
GND
A2
CKN
B2
GND
A3
CKP
B3
GND
A4
GND
B4
GND
A5
DVDD18
B5
A6
GND
A7
DAN13
A8
DAP13
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
C1
AVDD18
D1
AVDD18
E1
AVDD18
F1
GND
C2
AVDD18
D2
AVDD18
E2
AVDD18
F2
GND
C3
AVDD18
D3
AVDD18
E3
DVDD18
F3
GND
C4
AVDD18
D4
DVDD18
E4
DVDD18
F4
GND
DVDD18
C5
DVDD18
D5
DVDD18
E5
DVDD18
F5
GND
B6
GND
C6
GND
D6
GND
E6
GND
F6
GND
B7
DAN12
C7
DAN11
D7
DAN10
E7
DAN9
F7
DAN8
B8
DAP12
C8
DAP11
D8
DAP10
E8
DAP9
F8
DAP8
A9
DBN13
B9
DBN12
C9
DBN11
D9
DBN10
E9
DBN9
F9
DBN8
A10
DBP13
B10
DBP12
C10
DBP11
D10
DBP10
E10
DBP9
F10
DBP8
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
G1
GND
H1
IOUTP
J1
IOUTN
K1
GND
L1
GND
M1
REFIO
G2
GND
H2
GND
J2
GND
K2
GND
L2
GND
M2
FSADJ
G3
GND
H3
GND
J3
GND
K3
GND
L3
GND
M3
GND
G4
GND
H4
GND
J4
GND
K4
GND
L4
GND
M4
AVDD33
G5
GND
H5
GND
J5
GND
K5
GND
L5
GND
M5
DVDD33
G6
GND
H6
GND
J6
GND
K6
GND
L6
GND
M6
GND
G7
DAN7
H7
DAN6
J7
DCKON
K7
DAN5
L7
DAN4
M7
DAN3
G8
DAP7
H8
DAP6
J8
DCKOP
K8
DAP5
L8
DAP4
M8
DAP3
G9
DBN7
H9
DBN6
J9
DCKIN
K9
DBN5
L9
DBN4
M9
DBN3
G10
DBP7
H10
DBP6
J10
DCKIP
K10
DBP5
L10
DBP4
M10
DBP3
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
N1
GND
P1
AVDD33
Q1
AVDD33
R1
GND
S1
PD
N2
GND
P2
AVDD33
Q2
AVDD33
R2
GND
S2
CS
N3
GND
P3
AVDD33
Q3
AVDD33
R3
TSTN
S3
SDO
N4
AVDD33
P4
AVDD33
Q4
AVDD33
R4
TSTP
S4
SDI
N5
DVDD33
P5
DVDD33
Q5
DVDD33
R5
GND
S5
SCK
N6
GND
P6
GND
Q6
GND
R6
GND
S6
SVDD
N7
DAN2
P7
DAN1
Q7
DAN0
R7
GND
S7
GND
N8
DAP2
P8
DAP1
Q8
DAP0
R8
GND
S8
GND
N9
DBN2
P9
DBN1
Q9
DBN0
R9
GND
S9
GND
N10
DBP2
P10
DBP1
Q10
DBP0
R10
GND
S10
GND
2000af
For more information www.linear.com/LTC2000A
47
LTC2000A
Pin Locations (LTC2000A-14)
2
3
CKN
CKP
1
A
TOP VIEW
4 5 6 7
B
E
GND
DAN12 DAP12 DBN12 DBP12
AVDD18
DVDD18
DAN7 DAP7 DBN7 DBP7
IOUTP
DAN6 DAP6 DBN6 DBP6
IOUTN
DCKON DCKOP DCKIN DCKIP
DAN4 DAP4 DBN4 DBP4
DVDD33
DAN1 DAP1 DBN1 DBP1
DAN0 DAP0 DBN0 DBP0
GND
PD
48
DAN2 DAP2 DBN2 DBP2
AVDD33
Q
S
DAN3 DAP3 DBN3 DBP3
REFIO FSADJ
P
R
DAN5 DAP5 DBN5 DBP5
GND
L
N
DAN8 DAP8 DBN8 DBP8
GND
K
M
DAN10 DAP10 DBN10 DBP10
DAN9 DAP9 DBN9 DBP9
G
J
10
DAN11 DAP11 DBN11 DBP11
F
H
9
DAN13 DAP13 DBN13 DBP13
C
D
8
CS
TSTN
TSTP
SDO
SDI
GND
SCK
SVDD
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Pin Locations (LTC2000A-11)
LTC2000A-11 BGA Pinout
PIN ID
FUNCTION
PIN ID
FUNCTION
A1
GND
B1
GND
A2
CKN
B2
GND
A3
CKP
B3
GND
A4
GND
B4
GND
A5
DVDD18
B5
A6
GND
A7
DAN10
A8
DAP10
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
C1
AVDD18
D1
AVDD18
E1
AVDD18
F1
GND
C2
AVDD18
D2
AVDD18
E2
AVDD18
F2
GND
C3
AVDD18
D3
AVDD18
E3
DVDD18
F3
GND
C4
AVDD18
D4
DVDD18
E4
DVDD18
F4
GND
DVDD18
C5
DVDD18
D5
DVDD18
E5
DVDD18
F5
GND
B6
GND
C6
GND
D6
GND
E6
GND
F6
GND
B7
DAN9
C7
DAN8
D7
DAN7
E7
DAN6
F7
DAN5
B8
DAP9
C8
DAP8
D8
DAP7
E8
DAP6
F8
DAP5
A9
DBN10
B9
DBN9
C9
DBN8
D9
DBN7
E9
DBN6
F9
DBN5
A10
DBP10
B10
DBP9
C10
DBP8
D10
DBP7
E10
DBP6
F10
DBP5
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
G1
GND
H1
IOUTP
J1
IOUTN
K1
GND
L1
GND
M1
REFIO
G2
GND
H2
GND
J2
GND
K2
GND
L2
GND
M2
FSADJ
G3
GND
H3
GND
J3
GND
K3
GND
L3
GND
M3
GND
G4
GND
H4
GND
J4
GND
K4
GND
L4
GND
M4
AVDD33
G5
GND
H5
GND
J5
GND
K5
GND
L5
GND
M5
DVDD33
G6
GND
H6
GND
J6
GND
K6
GND
L6
GND
M6
GND
G7
DAN4
H7
DAN3
J7
DCKON
K7
DAN2
L7
DAN1
M7
DAN0
G8
DAP4
H8
DAP3
J8
DCKOP
K8
DAP2
L8
DAP1
M8
DAP0
G9
DBN4
H9
DBN3
J9
DCKIN
K9
DBN2
L9
DBN1
M9
DBN0
G10
DBP4
H10
DBP3
J10
DCKIP
K10
DBP2
L10
DBP1
M10
DBP0
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
PIN ID
FUNCTION
N1
GND
P1
AVDD33
Q1
AVDD33
R1
GND
S1
PD
N2
GND
P2
AVDD33
Q2
AVDD33
R2
GND
S2
CS
N3
GND
P3
AVDD33
Q3
AVDD33
R3
TSTN
S3
SDO
N4
AVDD33
P4
AVDD33
Q4
AVDD33
R4
TSTP
S4
SDI
N5
DVDD33
P5
DVDD33
Q5
DVDD33
R5
GND
S5
SCK
N6
GND
P6
GND
Q6
GND
R6
GND
S6
SVDD
N7
GND
P7
GND
Q7
GND
R7
GND
S7
GND
N8
GND
P8
GND
Q8
GND
R8
GND
S8
GND
N9
GND
P9
GND
Q9
GND
R9
GND
S9
GND
N10
GND
P10
GND
Q10
GND
R10
GND
S10
GND
2000af
For more information www.linear.com/LTC2000A
49
LTC2000A
Pin Locations (LTC2000A-11)
2
3
CKN
CKP
1
A
TOP VIEW
4 5 6 7
B
C
D
DAN9 DAP9 DBN9 DBP9
AVDD18
DAN8 DAP8 DBN8 DBP8
DVDD18
DAN7 DAP7 DBN7 DBP7
DAN6 DAP6 DBN6 DBP6
DAN5 DAP5 DBN5 DBP5
GND
G
DAN4 DAP4 DBN4 DBP4
IOUTP
DAN3 DAP3 DBN3 DBP3
IOUTN
DCKON DCKOP DCKIN DCKIP
K
N
DAN1 DAP1 DBN1 DBP1
REFIO FSADJ
DANO DAPO DBNO DBPO
DVDD33
P
AVDD33
Q
GND
R
GND
S
PD
50
DAN2 DAP2 DBN2 DBP2
GND
L
M
10
GND
F
J
9
DAN10 DAP10 DBN10 DBP10
E
H
8
CS
TSTN
TSTP
SDO
SDI
SCK
SVDD
2000af
For more information www.linear.com/LTC2000A
LTC2000A
Package Description
Please refer to http://www.linear.com/designtools/packaging/ for the most recent package drawings.
BGA Package
170-Lead (15.00mm × 9.00mm × 1.54mm)
(Reference LTC DWG# 05-08-1890 Rev B)
A
aaa Z
E
Y
A2
X
Z
SEE NOTES
DETAIL A
SEE NOTES
10
9
8
7
6
5
4
3
2
7
1
PIN 1
3
A
A1
PIN “A1”
CORNER
B
ccc Z
C
4
b
D
E
b1
MOLD
CAP
F
SUBSTRATE
H
F
Z
// bbb Z
D
G
H1
H2
J
DETAIL B
K
L
M
e
Øb (170 PLACES)
N
ddd M Z X Y
eee M Z
P
Q
R
S
aaa Z
DETAIL A
3.60
2.80
2.00
1.20
0.40
0.00
0.40
1.20
2.55
4.80
4.00
3.20
2.40
1.60
0.80
0.00
0.80
1.60
2.40
3.20
4.00
4.80
5.60
6.40
2. ALL DIMENSIONS ARE IN MILLIMETERS
DIMENSIONS
5.60
SUGGESTED PCB LAYOUT
TOP VIEW
PACKAGE BOTTOM VIEW
NOTES:
1. DIMENSIONING AND TOLERANCING PER ASME Y14.5M-1994
6.40
0.40 ±0.025 Ø 170x
b
G
DETAIL B
PACKAGE SIDE VIEW
2.00
2.80
3.05
PACKAGE TOP VIEW
3.60
e
5.35
5.85
SYMBOL
A
A1
A2
b
b1
D
E
e
F
G
H1
H2
aaa
bbb
ccc
ddd
eee
MIN
1.39
0.35
1.04
0.45
0.35
0.39
0.65
NOM
1.54
0.40
1.14
0.50
0.40
15.00
9.00
0.80
12.80
7.20
0.44
0.70
MAX
1.69
0.45
1.24
0.55
0.45
0.49
0.75
0.15
0.10
0.12
0.15
0.08
TOTAL NUMBER OF BALLS: 170
NOTES
3
BALL DESIGNATION PER JESD MS-028 AND JEP95
4
DETAILS OF PIN #1 IDENTIFIER ARE OPTIONAL,
BUT MUST BE LOCATED WITHIN THE ZONE INDICATED.
THE PIN #1 IDENTIFIER MAY BE EITHER A MOLD OR
MARKED FEATURE
5. PRIMARY DATUM -Z- IS SEATING PLANE
6. SOLDER BALL COMPOSITION CAN BE 96.5% Sn/3.0% Ag/0.5% Cu
OR Sn Pb EUTECTIC
7
!
PACKAGE ROW AND COLUMN LABELING MAY VARY
AMONG µModule PRODUCTS. REVIEW EACH PACKAGE
LAYOUT CAREFULLY
LTXXXXXX
COMPONENT
PIN “A1”
TRAY PIN 1
BEVEL
PACKAGE IN TRAY LOADING ORIENTATION
BGA 170 1112 REV B
2000af
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 its information
circuits as described
herein will not infringe on existing patent rights.
For more
www.linear.com/LTC2000A
51
LTC2000A
Typical Application
LTC2000A High Speed DAC Driving LT5579 Mixer as an Upconverting Transmitter with Low Noise Power Supply Solution
5V
1µF
5V
1µF
5
8
3
6
7
5
8
3
6
7
LT1763CS8-3.3
SHDN
IN
GND
GND
GND
1
OUT
2
SEN
0.01µF
BYP
47µF
L6
1nH
C65
10pF
47µF
10µF
4.7µF
3.24k
1k
SHDN
IN
GND
GND
GND
1
OUT
2
SEN
47µF
0.01µF
BYP
47µF
4
T1
ANAREN
B0430J50100AHF
6
1
GND
3
IN
GND
5
C41
100pF
C42
100pF
4
SPI
PORTS
C43
0.01µF
AVDD33 DVDD33
DAP[15:0]
SVDD
AVDD18
DAN[15:0]
J10
DVDD18
DCKIP
J8
DCKIN
DBP[15:0]
DBN[15:0]
K8
DCKOP
TSTP
K7
DCKON
TSTN
S1
PD
IOUTN
S2
LTC2000A-16
CS
S3
SDO
S4
SDI
S5
SCK
IOUTP
A3
CKP
R29
50Ω
REFIO
R26
50Ω
A2
FSADJ
CKN
LT3080EDD
3
7
V
VIN
2 OUT
8
VOUT
VIN
1
VOUT
5
9
PAD
VCTRL
SET
7.15k
LT3080EDD
3
7
V
VIN
2 OUT
8
V
VIN
1 OUT
V
5
9 OUT
PAD
VCTRL
SET
LT1763CS8-3.3
2
C66
1pF
1µF
4
MATCHED LVDS
DATA AND CLOCK
LINES FROM FPGA
CLOCK
SOURCE
7
1
SHDN OUTF
6
2
V
OUTS
8
3 IN
GND
GND
5
4
GND
GND
LTC6655CH
MS8-2.048
3.3V
4.7µF
3.24k
3.3V
3.3V
1µF
5V
3.3V
1µF
5V
S6
3.3V
C74
10µF
R4
R3
H1
R52
20Ω
J1
R49
0Ω
C26
0.1µF
R53
11.5Ω
C66
47pF
C68
82pF
R54
11.5Ω
C65
47pF
C67
82pF
1
L8
2.7nH
M1
M2
GND
R51
20Ω
R45
500Ω
C72
100pF
0.1dB CHEBYCHEV, BW = 535MHz
L11
L9
1.5nH
2.7nH
1
2 1
2
C25
0.1µF
R50
0Ω
C73
1µF
2
1
C71
10pF
4
IF –
C68
22pF
2
L10
1.5nH
3
VCC
LO
LT5579
IF
+
GND
RF
LO
INPUT
22
15
RF
OUTPUT
L12
1.8nH
C69
1.2pF
2000A TA01
C40
10µF
Related Parts
PART NUMBER
DESCRIPTION
COMMENTS
LTC1666/LTC1667/
LTC1668
12-/14-/16-Bit 50Msps DACs with 10mA Full Scale
VCC = ±5V, –1V to 1V Output Compliance, 28-Pin SSOP Package
LTC2000
16-/14-/11-Bit 2.5Gsps DAC
68dBc SFDR from DC to 1000MHz, 170-Lead BGA Package
LTC2153/LTC2158
Single/Dual 14-/12-Bit 310Msps ADCs
88dB SFDR, 1.25GHz Bandwidth Sample-and-Hold
LTC2630
Single 12-/10-/8-Bit Rail-to-Rail DACs with Internal Reference VCC = 2.7V to 5.5V, SC70 Package
LTC2991
Octal I2C Voltage, Current, and Temperature Monitor
VCC = 3V to 5.5V, 16-Lead MSOP Package
LTC2997
Remote/Internal Temperature Sensor
VCC = 2.V to 5.5V, 170µA, 6-Lead 2mm × 3mm DFN Package
LT®5521
Very High Linearity Active Mixer
10MHz to 3.7GHz, 24.2dBm OIP3 at 1.95GHz
LT5579
High Linearity Upconverting Mixer
1.5GHz to 3.8GHz, 27.3dBm OIP3 at 2.14GHz
LT5578
High Linearity Upconverting Mixer
400MHz to 2.7GHz, 24.3dBm IIP3 at 1.95GHz
LTC6406
3GHz, Low Noise, Rail-to-Rail Input Differential
Amplifier/Driver
Low Noise: 1.6nV/√Hz RTI , 18mA at 3V, Low Distortion
LTC6430-15
High Linearity Differential RF/IF Amplifier
20MHz to 2GHz Bandwidth, 50dBm OIP3 at 240MHz, 15.2dB Gain
LTC6946
Ultralow Noise and Spurious Integer-N Synthesizer with
Integrated VCO
0.37GHz to 5.7GHz, –226dBc/Hz Normalized In-Band Phase Noise,
–274dBc/Hz Normalized In-Band 1/f Noise
52 Linear Technology Corporation
1630 McCarthy Blvd., Milpitas, CA 95035-7417
For more information www.linear.com/LTC2000A
(408) 432-1900 ● FAX: (408) 434-0507
●
www.linear.com/LTC2000A
2000af
LT 0415 • PRINTED IN USA
 LINEAR TECHNOLOGY CORPORATION 2015