Slew Rate 265 V/s with 150 pF Load
Available in 48-Lead LQFP
APPLICATIONS
LCD Analog Column Driver
DB (0:9)
STBY
BYP
E/O
R/L
CLK
STSQ
XFR
FUNCTIONAL BLOCK DIAGRAM
10
AD8381
BIAS
SEQUENCE
CONTROL
VREFHIVREFLOINV VMID
10
10
10
10
10
10
CONTROL
2-STAGE
LATCH
2-STAGE
LATCH
2-STAGE
LATCH
2-STAGE
LATCH
2-STAGE
LATCH
2-STAGE
LATCH
SCALING
10
DAC
10
DAC
10
DAC
10
DAC
10
DAC
10
DAC
VID0
VID1
VID2
VID3
VID4
VID5
PRODUCT DESCRIPTION
The AD8381 provides a fast, 10-bit latched decimating digital
input, which drives six high voltage outputs. Ten-bit input
words are sequentially loaded into six separate high speed, bipolar
DACs. Flexible digital input format allows several AD8381s to be
used in parallel for higher resolution displays. STSQ synchronizes
sequential input loading, XFR controls synchronous output
updating and R/L controls the direction of loading as either
left-to-right or right-to-left. Six channels of high voltage
output drivers drive to within 1.3 V of the rail in rated settling
time. The output signal can be adjusted for brightness, signal
inversion, and contrast for maximum flexibility.
REV. B
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties that
may result from its use. No license is granted by implication or otherwise
under any patent or patent rights of Analog Devices. Trademarks and
registered trademarks are the property of their respective owners.
The AD8381 is fabricated on ADI’s proprietary, fast bipolar
24 V process, providing fast input logic, bipolar DACs with
trimmed accuracy and fast settling, high voltage precision drive
amplifiers on the same chip.
The AD8381 dissipates 570 mW nominal static power. The
STBY pin reduces power to a minimum, with fast recovery.
The AD8381 is offered in a 48-lead 7 mm ¥ 7 mm ¥ 1.4 mm
LQFP package and operates over the commercial temperature
range of 0∞C to 85∞C.
DIGITAL INPUT CHARACTERISTICSCLK Rise and Fall Time = 5 ns
Input Data Update RateNRZ100Ms/s
CLK to Data Setup Time: t
CLK to STSQ Setup Time: t
CLK to XFR Setup Time: t
CLK to Data Hold Time: t
CLK to STSQ Hold Time: t
CLK to XFR Hold Time: t
t
CLK HIGH
t
CLK LOW
C
IN
I
IH
I
IL
V
IH
V
IL
V
TH
1
3
5
2
4
6
E/O = HIGH4.5ns
Threshold Voltage1.4V
VIDEO OUTPUT CHARACTERISTICS
Output Voltage SwingAVCC – VOH, VOL – AGND11.3V
CLK to VID Delay4: t
7
50% of VIDx13.515.517.5ns
INV to VID Delay50% of VIDx121416ns
Output Current3075mA
Output Resistance29W
VIDEO OUTPUT DYNAMIC PERFORMANCET
MIN
to T
, VO = 5 V Step, CL = 150 pF
MAX
Data Switching Slew Rate265V/ms
Invert Switching Slew Rate410V/ms
Data Switching Settling Time to 1%2732ns
Data Switching Settling Time to 0.25%5075ns
Invert Switching Settling Time to 1%3340ns
Invert Switching Settling Time to 0.25%55100ns
CLK and Data Feedthrough
All-Hostile Crosstalk
Operating Temperature Range . . . . . . . . . . . . . . 0∞C to 85∞C
Storage Temperature Range . . . . . . . . . . . . –65∞C to +125∞C
Lead Temperature Range (Soldering 10 sec) . . . . . . . . 300∞C
NOTES
1
Stresses above those listed under the Absolute Maximum Ratings may cause
permanent damage to the device. This is a stress rating only; functional operation
of the device at these or any other conditions above those indicated in the
operational section of this specification is not implied. Exposure to the absolute
maximum ratings for extended periods may reduce device reliability.
The AD8381 employs a two-stage overload protection circuit
that consists of an output current limiter and a thermal shutdown.
The maximum current at any one output of the AD8381 is
internally limited to 100 mA average. In the event of a momentary short circuit between a video output and a power supply rail
(VCC or AGND), the output current limit is sufficiently low to
provide temporary protection.
The thermal shutdown debiases the output amplifier when the
junction temperature reaches the internally set trip point. In the
event of an extended short circuit between a video output and a
power supply rail, the output amplifier current continues to
switch between 0 mA and 100 mA typ with a period determined by
the thermal time constant and the hysteresis of the thermal trip
point. The thermal shutdown provides long term protection by
limiting the average junction temperature to a safe level.
Recovery from a momentary short circuit is fast, approximately
100 ns. Recovery from a thermal shutdown is slow and is
dependent on the ambient temperature.
MAXIMUM POWER DISSIPATION
The maximum power that can be safely dissipated by the AD8381
is limited by its junction temperature. The maximum safe junction temperature for plastic encapsulated devices is determined
by the glass transition temperature of the plastic, approximately
150∞C. Exceeding this limit temporarily may cause a shift in the
parametric performance due to a change in the stresses exerted
on the die by the package. Exceeding a junction temperature of
175∞C for an extended period can result in device failure.
To ensure proper operation within the specified operating temperature range, it is necessary to limit the maximum power
dissipation as follows:
P
DMAX
= (T
JMAX
– TA)/q
JA
where:
T
= 150∞C.
JMAX
3.5
3.0
2.5
2.0
1.5
1.0
MAXIMUM POWER DISSIPATION – W
0.5
0
102030405060708090
AMBIENT TEMPERATURE – ⴗC
Figure 4. Maximum Power Dissipation vs. Temperature
ORDERING GUIDE
Temperature PackagePackage
ModelRangeDescriptionOption
AD8381JST0∞C to 85∞C48-Lead LQFPST-48
AD8381-EB
Evaluation Board
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection. Although
the AD8381 features proprietary ESD protection circuitry, permanent damage may occur on
devices subjected to high-energy electrostatic discharges. Therefore, proper ESD precautions are
recommended to avoid performance degradation or loss of functionality.
–4–
REV. B
PIN CONFIGURATION
AGNDDAC
AV CC DAC
NC
NC
NC
STSQ
XFR
CLK
48
47 46
45 4439 38 3743 42 41 40
1
NC
DB0
DB1
DB2
DB3
DB4
DB5
DB6
DB7
DB8
DB9
NC
NC = NO CONNECT
PIN 1
2
IDENTIFIER
3
4
5
6
7
8
9
10
11
12
13 14
E/O
(Not to Scale)
15 16 17 18
R/L
INV
DGND
AD8381
TOP VIEW
19 20
DVC C
AV CCBIAS
NC
STBY
PIN FUNCTION DESCRIPTIONS
VREFHI
21 22
BYP
AGND0
VMID
VREFLO
23 24
NC
NC
AGNDBIAS
36
VID0
35
AV CC0, 1
34
VID1
33
AGND1, 2
32
VID2
31
AV CC2, 3
30
VID3
29
AGND3, 4
28
VID4
27
AV CC4, 5
26
VID5
25
AGND5
AD8381
Pin No.MnemonicFunctionDescription
1, 12, 19, 23, NCNo Connect
24, 43–45
2–11DB (0:9)Data Input10-Bit Data Input MSB = DB (9).
13E/OEven/Odd SelectThe active CLK edge is the rising edge when this input is held high,
and it is the falling edge when this input is held low.
Data is loaded sequentially on the rising edges of CLK when this input
is high and loaded on the falling edges when this input is low.
14R/LRight/Left SelectA new data loading sequence begins on the left, with Channel 0, when this
input is low, and on the right, with Channel 5, when this input is high.
15INVInvertWhen this pin is high, the analog output voltages are above VMID.
When low, the analog output voltages are below VMID.
16DGNDDigital Supply ReturnThis pin is normally connected to the analog ground plane.
17DVCCDigital Power SupplyDigital Power Supply.
18, 27, 31AVCCxAnalog Power SuppliesAnalog Power Supplies.
35, 42
20STBYStandbyWhen high, the internal circuits are debiased and the power
dissipation drops to a minimum.
21BYPBypassA 0.1 mF capacitor connected between this pin and AGND ensures
optimum settling time.
22, 25, 29AGNDxAnalog Supply ReturnsThese pins are normally connected to the analog ground plane.
33, 37, 41
26, 28, 30,VID5, VID4, VID3,Analog OutputsThese pins are directly connected to the analog inputs of the LCD panel.
32, 34, 36VID2, VID1, VID0
38VMIDMidpoint ReferenceThe voltage applied between this pin and AGND sets the midpoint
reference of the analog outputs. This pin is normally connected to VCOM.
39VREFLOFull-Scale ReferenceThe voltage applied between Pins 39 and 40 sets the full-scale output voltage.
40VREFHIFull-Scale ReferenceThe voltage applied between Pins 39 and 40 sets the full-scale output voltage.
46STSQStart SequenceA new data loading sequence begins on the rising edge of CLK when
this input was high on the preceding rising edge of CLK and the E/O
input is held high.
A new data loading sequence begins on the falling edge of CLK when
this input was high on the preceding falling edge of CLK and the E/O
input is held low.
47 XFRData TransferData is transferred to the outputs on the immediately following falling
edge of CLK when this input is high on the rising edge of CLK.
48CLKClockClock Input.
REV. B
–5–
AD8381
–Typical Performance Characteristics
12V
VMID = 7V
VFS = 5V
VIDx
2V
C
L
150pF
20ns/DIV
TPC 1. Invert Switching 10 V Step Response (Rise) at C
7V
VMID = 7V
VFS = 5V
VIDx
2V
C
L
150pF
12V
VMID = 7V
VFS = 5V
VIDx
2V
L
TPC 4. Invert Switching 10 V Step Response (Fall) at C
7V
VMID = 7V
VFS = 5V
VIDx
2V
C
L
150pF
20ns/DIV
C
L
150pF
L
10ns/DIV
TPC 2. Data Switching 5 V Step Response (Rise)
at CL, INV = L
12V
VMID = 7V
VFS = 5V
VIDx
7V
C
L
150pF
20ns/DIV
TPC 3. Data Switching 5 V Step Response (Rise)
at CL, INV = H
10ns/DIV
TPC 5. Data Switching 5 V Step Response (Fall)
, INV = L
at C
L
12V
VMID = 7V
VFS = 5V
VIDx
7V
C
L
150pF
20ns/DIV
TPC 6. Data Switching 5 V Step Response (Fall)
at CL, INV = H
–6–
REV. B
AD8381
20ns/DIV
+10mV
VMID = 7V
–10mV
DB (0:9)
0.25%
7V
0.00%
–0.25%
–0.50%
–0.75%
–1.00%
t = 0
VMID = 7V
VFS = 5V
VIDx
C
L
150pF
10ns/DIV
TPC 7. Output Settling Time (Rising Edge) at CL,
5 V Step, INV = Low
0.00%
12V
–0.25%
–0.50%
–0.75%
–1.00%
t = 0
VMID = 7V
VFS = 5V
VIDx
C
L
150pF
10ns/DIV
1.00%
0.75%
0.50%
0.25%
2V
0.00%
–0.25%
–0.50%
–0.75%
–1.00%
t = 0
VMID = 7V
VFS = 5V
VIDx
C
L
150pF
10ns/DIV
TPC 10. Output Settling Time (Falling Edge) at CL,
5 V Step, INV = Low
VMID = 7V
1.00%
0.75%
0.50%
0.25%
0.00%
–0.25%
–0.50%
–0.75%
7V
t = 0
VFS = 5V
VIDx
C
L
150pF
10ns/DIV
TPC 8. Output Settling Time (Rising Edge) at CL,
5 V Step, INV = High
+30mV
+20mV
+10mV
VMID = 7V
–10mV
–20mV
TPC 9. All-Hostile Crosstalk at C
VID5
VID0 – VID4
5V
20ns/DIV
L
TPC 11. Output Settling Time (Falling Edge) at CL,
5 V Step, INV = High
TPC 12. Data Switching Transient (Feedthrough) at C
L
REV. B
–7–
AD8381
0.5
0.4
0.3
0.2
0.1
0.0
DNL – LSB
–0.1
–0.2
–0.3
–0.4
–0.5
0
2565127681023
INPUT CODE
TPC 13. Differential Nonlinearity (DNL) vs. Code, INV = H
0.5
0.4
0.3
0.2
0.1
0.0
INL – LSB
–0.1
–0.2
–0.3
–0.4
–0.5
0
2565127681023
INPUT CODE
TPC 14. Integral Nonlinearity (INL) vs. Code, INV = H
0.5
0.4
0.3
0.2
0.1
0.0
DNL – LSB
–0.1
–0.2
–0.3
–0.4
–0.5
0
2565127681023
INPUT CODE
TPC 16. Differential Nonlinearity (DNL) vs. Code, INV = L
0.5
0.4
0.3
0.2
0.1
0.0
INL – LSB
–0.1
–0.2
–0.3
–0.4
–0.5
0
2565127681023
INPUT CODE
TPC 17. Integral Nonlinearity (INL) vs. Code, INV = L
5
0
–5
–10
VFS = 4V
–15
–20
NORMALIZED VDE AT CODE 0 – mV
–25
5
VFS = 5V
VFS = 5.75V
671110
VMID – V
VFS = 5.75V
89
VFS = 4V
VFS = 5V
TPC 15. Normalized VDE at Code 0 vs. VMID, AVCC = 15.5 V
–8–
0
–20
–40
PSRR – dB
–60
–80
10k
CODE 512, INV = LOW
CODE 512, INV = HIGH
100k1M5M
FREQUENCY – Hz
TPC 18. AVCC Power Supply Rejection vs. Frequency
REV. B
7.5
INPUT CODE
3.50
0
VCME – mV
1.75
0.00
–1.75
5127681023256
–3.50
TEMPERATURE – ⴗC
3.50
0
VCME – mV
1.75
0.00
–1.75
406010020
–3.50
80
CODE 512
5.0
2.5
0.0
VDE – mV
–2.5
–5.0
–7.5
0
5127681023256
INPUT CODE
TPC 19. Differential Error Voltage (VDE) vs. Code
7.5
5.0
AD8381
TPC 21. Common-Mode Error Voltage (VCME) vs. Code
2.5
CODE 512
0.0
VDE – mV
–2.5
–5.0
–7.5
0
406010020
TEMPERATURE – ⴗC
80
TPC 20. Differential Error Voltage (VDE) vs. Temperature
TPC 22. Common-Mode Error (VCME) vs. Temperature
REV. B
–9–
AD8381
FUNCTIONAL DESCRIPTION
The AD8381 is a system building block designed to directly
drive the columns of LCD panels of the type popularized for use
in data projectors. It comprises six channels of precision 10-bit
digital-to-analog converters loaded from a single, high speed,
10-bit-wide input. Precision current feedback amplifiers, providing well-damped pulse response and rapid voltage settling into
large capacitive loads, buffer the six outputs. Laser trimming at
the wafer level ensure low absolute output errors and tight channelto-channel matching. In addition, tight part-to-part matching
in high channel count systems is guaranteed by the use of an
external voltage reference.
Input Data Loading (STart SeQuence Control—STSQ)
A valid STSQ control input initiates a new six-clock pulse
loading cycle, during which six input data words are loaded
sequentially into six internal channels. A new loading sequence
begins on the current active CLK edge only when STSQ was
held high at the preceding active CLK edge.
Data Loading—Expanded Systems (Even/Odd Control)
To facilitate expanded, even/odd systems, the active CLK edge, at
which input data is loaded, is set with the E/O control input.
Input data is loaded on rising CLK edges while the E/O input is
held high and loaded on falling CLK edges while the E/O input
is held low.
Data Loading—Inverted Images (Right/Left Control)
To facilitate image mirroring, the order in which input data is
loaded is set with the R/L input.
A new loading sequence begins at Channel 0 and proceeds
to Channel 5 when the R/L input is held high and begins at
Channel 5 and proceeding to Channel 0 when the R/L input
is held low.
Data Transfer to Outputs (XFR Control)
Data transfer to all outputs is initiated by the XFR control input.
When XFR is held high during a rising CLK edge, data is
simultaneously transferred to all outputs on the immediately
following falling CLK edge.
VCOM Reference (VMID Reference Input)
An external analog reference voltage connected to this input
sets the reference level at the outputs. This input is normally
connected to VCOM.
The difference between two external analog reference voltages,
connected to these inputs, sets the full-scale output voltage at
the outputs. VREFLO is normally tied to VMID.
Analog Voltage Inversion (INVert Control)
To facilitate systems that use column, row or pixel inversion,
the analog output voltage inversion is controlled by the INV
control input. While INV is high, the analog voltage equivalent
of the input code is subtracted from (VMID + VFS) at each
output. While INV is low, the analog voltage equivalent of the
input code is added to (VMID – VFS) at each output.
Standby Mode (STBY Control)
A high applied to the STBY input debiases the internal
circuitry, dropping the quiescent power dissipation to a few
milliwatts. Since both digital and analog circuits are debiased,
all stored data will be lost. Upon returning STBY to low, normal operation is restored.
–10–
REV. B
AD8381
VCMEVOUTN nVOUTP nVMID=¥ ¥+
()
Ê
Ë
Á
ˆ
¯
˜
121
2
()()–
AV CC
AV CC/2
4.3
VFS (V)
5.75
AV CC/2–1.3
2
VAL ID VMID
0
5.37AVCC–7AVCC–3
AV CC /2
VMID (V)
TRANSFER FUNCTION
The AD8381 has two regions of operation, selected by the INV
input, where the video output voltages are either above or below
a reference voltage, applied externally at the VMID input.
The transfer function defines the analog output voltage as the
function of the digital input code as follows:
VOUTVMID VFS
=±¥
Ê
1
Á
Ë
–
1023
ˆ
n
˜
¯
where:
n = input code
VFS = 2 ¥ (VREFHI – VREFLO)
VOUT (V)
AV CC
(VMID + VFS)
INV = HIGH
INV = LOW
INPUT CODE
1023
VMID
(VMID – VFS)
AGND
VOUTN(n)
VOUTP(n)
0
Figure 5. Transfer Function
The region over which the output voltage varies with input code
is selected by the INV input. When INV is low, the output voltage increases from (VMID – VFS), (where VFS = the full-scale
output voltage), to VMID as the input code increases from 0 to
1023. When INV is high, the output voltage decreases from
(VMID + VFS) to VMID with increasing input code.
For each value of input code there are then two possible values
of output voltage. When INV is low, the output is defined as
VOUTP(n) where n is the input code and P indicates the operating region where the slope of the transfer function is positive.
When INV is high, the output is defined as VOUTN(n) where N
indicates the operating region where the slope of the transfer
function is negative.
VDE, the differential error voltage, measures the deviation of the
rms value of the output from the rms value of the ideal. It is dependent on the difference between the output amplitudes VOUTN(n)
and VOUTP(n) at a particular code. The defining expression is
VDEVOUTN nVOUTP nVFS
1
=¥
()
2
()–()––
Ê
Á
Ë
Ê
¥
1
Á
Ë
n
1023
ˆ
ˆ
˜
˜
¯
¯
where:
1
VOUTN nVOUTP n()–()
¥
()
2
is the rms value of the output.
(VFS¥ (1 – n/1023)) is the rms value of the ideal.
VCME, the common-mode error voltage, measures the deviation of the average value of the output from the average value of
the ideal. It is dependent on the average between the output
amplitudes VOUTN(n) and VOUTP(n) at a particular code.
The defining expression is:
where:
1
VOUTN nVOUTP n()()
¥+
()
2
is the average value of the output.
VMID is the average value of the ideal.
MAXIMUM FULL-SCALE OUTPUT VOLTAGE
The following conditions limit the range of usable output voltages:
∑The internal DACs limit the minimum allowed voltage at
the VMID input to 5.3 V.
∑The scale factor control loop limits the maximum full-scale
output voltage to 5.75 V.
∑The output amplifiers settle cleanly at voltages within 1.3 V
from the supply rails.
∑The common-mode range of the output amplifiers limit the
maximum value of VMID to AVCC – 3.
At any given valid value of VMID, the voltage required to reach
any one of the above limits defines the maximum usable fullscale output voltage VFSMAX.
VFSMAX is the envelope in Figure 6. The valid range of VMID
is the shaded area.
ACCURACY
To best correlate transfer function errors to image artifacts, the
overall accuracy of the AD8381 is defined by two parameters,
VDE and VCME.
REV. B
Figure 6. VFSMAX vs. VMID
–11–
AD8381
S
3
9
S
Operating Modes—6-Channel Systems
The simplest full color LCD based system is characterized by an
image processor with a single 10-bit-wide data bus and a 6-channel
LCD per color.
Such systems usually have VGA or SVGA resolution and require a
single AD8381 per color.
The INV input facilitates column and row inversion for
these systems.
Single and dual data bus type 12-channel systems are commonly in use.
The single data bus 12-channel system is characterized by an
image processor with a single, 10-bit data bus and a 12-channel
LCD per color. The maximum resolution of such a system is
usually up to 85 Hz XGA or 75 Hz SXGA and requires two
AD8381s per color.
One AD8381 is set to run in even mode while the other is in
odd mode. Both AD8381s share the same data bus and CLK.
The timing diagram of such system is shown in Figure 8.
The dual data bus 12-channel system is characterized by an
image processor with two 10-bit parallel data buses and a
12-channel LCD. The maximum resolution of such system is
usually up to 75 Hz UXGA and requires two AD8381s per color.
Both AD8381s may be set to run in Even mode and may share
the same CLK. The timing diagram of each AD8381 in such
system is identical to that of the 6-channel system.
The INV input facilitates column, row, and pixel inversion for
both types of 12-channel systems.
PIXEL CLK
–3 –212345 678910 121314
DB (0:9)
CLK
STSQ
EVEN
STSQ
ODD
XFR
INPUTS
R/L
E/O
EVEN
E/O
ODD
CH0
CH1
CH2
CH3
CH4
INTERNAL LATCHE
CH5
VID0
AD8381 EVEN
VID1
VID2
VID3
OUTPUT
VID4
VID5
CH0
CH1
CH2
CH3
CH4
INTERNAL LATCHE
CH5
VID0
AD8381 ODD
VID1
VID2
VID3
OUTPUT
VID4
VID5
0–1
0
2
4
6
–2
–12
–10
–8
–6
–4
–2
1
3
5
–3
–1
–11
–9
–7
–5
–3
–1
11
8
10
7
9
15 16 17 18 19 20 21 22 23 24
12
14
16
0
2
4
6
8
10
13
15
11
1
5
7
11
18
20
22
12
14
16
18
20
22
17
19
21
23
13
15
17
19
21
23
Figure 8. Twelve-Channel Even/Odd System
Timing Diagram
Operating Modes—Large Channel Count Systems
To facilitate 18, 24, or higher channel systems, any number of
required AD8381s may be cascaded.
–12–
REV. B
AD8381
H. REVERSE
HSYNC
VSYNC
IMAGE PROCESSOR
STSQ2
STSQ1
CLK
CLK
CLK
CLK
REFERENCES
DB(0:9)
CLK
XFR
R/L
STSQ
INV
E/O
VREFHI
VMID
VREFLO
DB(0:9)
CLK
XFR
R/L
STSQ
INV
E/O
VREFHI
VMID
VREFLO
AD8381
AD8381
PIXEL CLK
ⴜ6 COUNTER
ⴜ6 COUNTER
HSTART
INV1
INV2
VREFHI
VCOM
DB(0:9)
CLK
+2
XFR
R/L
STSQ1
INV1
E/O1
STSQ2
INV2
E/O2
Figure 9. Single Data Bus 12-Channel Even/Odd System Block Diagram
VID0
VID1
VID2
VID3
VID4
VID5
VID0
VID1
VID2
VID3
VID4
VID5
CH 0
CH 2
CH 4
CH 6
CH 8
CH 10
12-CHANNEL
LCD
CH 1
CH 3
CH 5
CH 7
CH 9
CH 11
H. REVERSE
HSYNC
VSYNC
IMAGE PROCESSOR
D(0:9) ODD
D(0:9) EVEN
CLK
D(0:9) EVEN
D(0:9) ODD
REFERENCES
DB(0:9)
CLK
XFR
R/L
STSQ
INV
E/O
VREFHI
VMID
VREFLO
DB(0:9)
CLK
XFR
R/L
STSQ
INV
E/O
VREFHI
VMID
VREFLO
AD8381
AD8381
PIXEL CLK
ⴜ6 COUNTER
HSTART
INV1
INV2
VREFHI
VCOM
“1”
+2
DB1(0:9)
CLK
XFR
R/L
STSQ
INV1
E/O
INV2
DB2(0:9)
Figure 10. Dual Parallel Data Bus 12-Channel System Block Diagram
VID0
VID1
VID2
VID3
VID4
VID5
VID0
VID1
VID2
VID3
VID4
VID5
CH 0
CH 2
CH 4
CH 6
CH 8
CH 10
12-CHANNEL
LCD
CH 1
CH 3
CH 5
CH 7
CH 9
CH 11
REV. B
–13–
AD8381
LAYOUT CONSIDERATIONS
The AD8381 is a mixed-signal, high speed, very accurate
device. In order to realize its specifications, it is essential to use
a properly designed printed circuit board.
Layout and Grounding
The analog outputs and the digital inputs of the AD8381 are
pinned out on opposite sides of the package. When laying out
the circuit board, keep these sections separate from each other
to minimize crosstalk and noise and the coupling of the digital
input signals into the analog outputs.
All signal trace lengths should be made as short and direct as
possible to prevent signal degradation due to parasitic effects.
Note that digital signals should not cross or be routed near
analog signals.
It is imperative to provide a solid analog ground plane under
and around the AD8381. All of the ground pins of the part
should be connected directly to the ground plane with no extra
signal path length. For conventional operation this includes the
pins DGND, AGNDDAC, AGNDBIAS, AGND0, AGND1,2,
AGND3,4, and AGND5. The return traces for any of the
signals should be routed close to the ground pin for that section
to prevent stray signals from coupling into other ground pins.
Power Supply Bypassing
All power supply and reference pins of the AD8381 must be
properly bypassed to the analog ground plane for optimum
performance.
All analog supply pins may be connected directly to an analog
supply plane located as close to the part as possible. A 0.1 mF
chip capacitor should be placed as close to each analog supply
pin as possible and connected directly between each analog
supply pin and the analog ground plane.
A minimum of 47 mF tantalum capacitor should be placed near
the analog supply plane and connected directly between the
supply and analog ground planes.
A minimum of 10 mF tantalum capacitor should be placed near the
digital supply pin and connected directly to the analog ground
plane. A 0.1 mF chip capacitor should be connected between the
digital supply pin and the analog ground.
VREFHI, VMID, VREFLO Reference Distribution
To ensure well-matched video outputs, all AD8381s must operate from equal reference voltages.
Each reference voltage should be distributed to each AD8381
directly from the source of the reference voltage with approximately equal trace lengths.
A 0.1 mF chip capacitor should be placed as close to each refer-
ence input pin as possible and directly connected between the
reference input pin and the analog ground plane.