AD8380 [ADI]

Fast, High-Voltage Drive, 6-Channel Output DecDriver Decimating LCD Panel Driver; 快速,高电压驱动器, 6声道输出DecDriver抽取LCD面板驱动
AD8380
型号: AD8380
厂家: ADI    ADI
描述:

Fast, High-Voltage Drive, 6-Channel Output DecDriver Decimating LCD Panel Driver
快速,高电压驱动器, 6声道输出DecDriver抽取LCD面板驱动

驱动器 CD
文件: 总16页 (文件大小:212K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
Fast, High-Voltage Drive, 6-Channel Output  
a
DecDriver Decimating LCD Panel Driver  
AD8380  
FEATURES  
FUNCTIONAL BLOCK DIAGRAM  
High-Voltage Drive to Within 1.3 V of Supply Rails  
24 V Supply for Fast Output Voltage Drivers  
High Update Rates: Fast 75 Ms/s 10-Bit Input Word Rate  
Low Power Dissipation, 550 mW with Power-Down  
Voltage Controlled Video Reference and Full-Scale  
(Contrast) Output Levels  
INV Bit Reverses Polarity of Video Signal  
Nominal 3.3 V Logic and 15 V Analog Supplies  
Flexible Logic  
AD8380  
10  
10  
10  
10  
10  
10  
10  
2-STAGE  
LATCH  
DB [0:9]  
VID0  
VID1  
VID2  
DAC  
DAC  
DAC  
2-STAGE  
LATCH  
CLK  
STSQ/CS  
XFR  
CHANNEL  
SELECTOR  
2-STAGE  
LATCH  
E/O  
Addressable or Sequential Channel Loading  
STSQ/CS Allow Parallel AD8380 Operation for XGA  
and Greater Resolution  
Drives Capacitive Loads  
26 ns Settling Time to 1% Up to 150 pF Load  
Slew Rate 270 V/s  
R/L  
10  
10  
10  
10  
10  
10  
2-STAGE  
LATCH  
VID3  
VID4  
VID5  
DAC  
DAC  
DAC  
A[0:2]  
3
2-STAGE  
LATCH  
STBY  
BYP  
BIAS  
2-STAGE  
LATCH  
Available in 44-Lead MQFP  
VREFHI  
SCALING  
CONTROL  
APPLICATIONS  
Poly Si LCD Panel Analog Column Driver  
VREFLO  
INV  
VMID  
PRODUCT DESCRIPTION  
The AD8380 is fabricated on ADI’s XFCB26 fast bipolar 26 V  
process, providing fast input logic, trimmed accuracy bipolar  
DACs and fast settling, high voltage precision drive amplifiers  
on the same chip.  
The AD8380 provides a fast, 10-bit latched decimating digital  
input that drives 6-channel high voltage drive outputs. The 10-  
bit input word is sequentially muxed into six separate high speed,  
bipolar DACs. Flexible digital input formats allow several  
AD8380s to be used in parallel for higher resolution displays.  
STSQ/CS, in conjunction with 3-bit addressable channel-loading  
pins, allows loading of the digital words either sequentially or  
randomly, and R/L control sets loading as either left to right, or  
vice versa. 6-channel high voltage output drivers drive to within  
1.3 V of the rails to rated settling time. The output signal can be  
adjusted for dc signal reference, signal inversion or contrast for  
maximum flexibility.  
The AD8380 dissipates nominally 0.55 W of static power. STBY  
pin reduces power to a minimum, with fast recovery.  
The AD8380 is offered in a 44-lead 10 × 10 × 2.0 mm MQFP  
package and operates over the commercial temperature range of  
0°C to 85°C.  
DecDriver is a trademark of Analog Devices, Inc.  
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, norforanyinfringementsofpatentsorotherrightsofthirdpartiesthat  
may result from its use. No license is granted by implication or otherwise  
under any patent or patent rights of Analog Devices.  
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.  
Tel: 781/329-4700  
Fax: 781/326-8703  
www.analog.com  
© Analog Devices, Inc., 2001  
(@ 25؇C, AVCC = 15 V, DVCC = 3.3 V, TMIN = 0؇C, TMAX = 85؇C, unless  
otherwise noted)  
AD8380–SPECIFICATIONS  
Model  
Conditions  
Min  
Typ  
Max  
Unit  
VIDEO DC PERFORMANCE1  
TMIN to TMAX  
VDE  
VCME  
Scale Factor Error  
Offset Error  
DAC Code = 450 to 800  
DAC Code = 450 to 800  
DAC Code = 0 to 1023  
DAC Code = 0 to 1023  
–7.5  
–3.5  
–0.25  
–7  
+1  
+0.5  
+7.5  
+3.5  
+0.25  
+7  
mV  
mV  
%
+1  
mV  
REFERENCE INPUTS  
VMID Range2  
6
7
7.5  
V
VMID Bias Current  
VFS Range  
VREFHI  
3
5
µA  
V
V
VFS = 2 × (VREFHI–VREFLO)  
1
6
VREFLO +0.5  
VMID – 0.5  
AVCC – 2.5  
VREFHI – 2.5 VREFHI – 0.5  
AVCC  
VREFLO  
V
VREFHI Input Resistance  
VREFLO Bias Current  
VREFHI Input Current3  
to VREFLO  
VFS = 5 V  
3.3  
0.2  
750  
kΩ  
µA  
µA  
RESOLUTION  
Coding  
Binary  
10  
Bits  
DIGITAL INPUT CHARACTERISTICS  
Input Data Update Rate  
Clock to Data Setup Times: t1  
Clock to STSQ Setup Times: t3  
Clock to XFR Setup Times: t5  
Maximum CLK Rise and Fall Time, t7  
Clock to A[0:2] Hold Times: t9  
Clock to Data Hold Times: t2  
Clock to STSQ Hold Times: t4  
Clock to XFR Hold Times: t6  
Clock to A[0:2] Setup Times: t8  
CIN  
75  
Ms/s  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
ns  
pF  
µA  
V
1
1
1
4
4
4
4
4
1
3
IIN  
VIH  
0.6  
2.0  
VIL  
0.8  
V
VTH  
Threshold Voltage  
1.4  
V
VIDEO OUTPUT CHARACTERISTICS  
Output Voltage Swing  
AVCC – VOH, VOL – AVEE  
50% of VIDx  
1.1  
15.5  
1.3  
17.5  
V
ns  
mA  
CLK to VID Delay4  
13.5  
30  
Output Current  
VIDEO OUTPUT DYNAMIC PERFORMANCE  
TMIN to TMAX, VO = 5 V Step,  
CL = 150 pF, RS = 25 Ω  
Data Switching Slew Rate  
Invert Switching Slew Rate  
Data Switching Settling Time to 1%5  
Data Switching Settling Time to 0.25%  
Invert Switching Settling Time to 1%5  
Invert Switching Settling Time to 0.25%  
CLK Feedthrough6  
270  
625  
26  
35  
30  
85  
2
V/µs  
V/µs  
ns  
ns  
ns  
32  
65  
40  
100  
5
ns  
mV p-p  
All-Hostile Crosstalk7  
Amplitude  
Glitch Duration  
95  
40  
mV p-p  
ns  
POWER SUPPLY  
Supply Rejection (VDE)  
DVCC, Operating Range  
DVCC, Quiescent Current  
AVCC, Operating Range  
Total AVCC Quiescent Current  
STBY AVCC Current  
+VS = 15 V 1 V  
1
mV/V  
V
mA  
V
mA  
mA  
mA  
3
9
5.5  
35  
24  
44  
5
22  
33  
0.5  
0.1  
STBY = H  
STBY = H  
STBY DVCC Current  
5
OPERATING TEMPERATURE RANGE  
0
85  
°C  
NOTES  
1For definitions of VDE and VCME, see the Transfer Function section. Scale factor error is expressed as percentage of VFS.  
2See Figure 1 for valid ranges of VMID.  
3VREFHI Input Current = (VREFHI – VREFLO)/(VREFHI Input Resistance) = 2.5 V/3.3 k.  
4Delay time from 50% of falling CLK edge to 50% of output change. Measurement is made for both states of INV.  
5For best settling time results, use minimum series output resistance, RS of 25 .  
6An output channel is selected, and glitch is monitored as CLK is driven. STSQ and XFR are set to logic low.  
7Input data is loaded such that any five output channels change by VFS (i.e., 5 V), and the sixth unselected channel is monitored. Measurement is made for both states of INV.  
Specifications subject to change without notice.  
–2–  
REV. B  
AD8380  
PIN FUNCTION DESCRIPTIONS  
Description  
Pin No.  
Mnemonic  
1
NC  
No Connect.  
2–11  
12  
DB[0:9]  
E/O  
Video Data Inputs. DB9 is the MSB.  
Even/Odd data select, input latches are loaded at the falling edge of CLK if E/O is low or  
rising edge if E/O is high.  
13  
R/L  
Determines starting point of internally generated channel-loading sequence.  
R/L Low (when address = 111) loads from Channel 0 up to Channel 5.  
14  
15, 16  
INV  
DVEE, DVCC  
When high, analog video outputs are above the VMID setpoint. See Figure 3.  
Digital Supplies. Nominally 3.3 V and 0 V, respectively.  
17, 20, 22, 24,  
26, 28, 30, 32,  
34, 37, 38  
AVCCxxx, AVEExxx Analog Supplies. Nominally 15 V and 0 V, respectively.  
18  
STBY  
Stand By. When high, all digital and analog circuits are “debiased” and the power dissipation  
drops to a minimum.  
19  
21  
BYP  
VMID  
An external capacitor connected from here to VEE will help to ensure rapid DAC settling time.  
Externally supplied voltage applied here sets the midpoint reference for the video output.  
23, 25, 27, 29,  
31, 33  
VID5–VID0  
Analog Video Outputs.  
36, 35  
VREFHI, VREFLO Voltage between these pins sets DAC full-scale range. An external reference must be applied  
and should be common to all devices to ensure best tracking.  
39–41  
42  
A[0:2]  
STSQ/CS  
3-bit channel address for addressable loading of the digital input latches.  
STSQ to start internal sequencing or Chip Select to enable addressable channel addressing.  
See functional description. Used in conjunction with A[0:2].  
43  
44  
XFR  
CLK  
If XFR = HIGH at the rising edge of CLK, data is transferred to the DACs on the next falling  
edge of CLK. See Figures 4, 6, 7, and 8.  
Master Clock Input.  
CHANNEL SELECTION FUNCTIONALITY  
There are two channel selection modes, addressed channel  
loading, (in which the user directly controls which DAC is  
loaded), and internally sequenced loading (in which the user  
controls the direction and clock phase in which the loading  
proceeds).  
MAXIMUM OUTPUT VOLTAGE  
The maximum output signal swing is constrained by the output  
voltage compliance of the DACs and the output dynamic range  
of the output amplifiers. The minimum voltage allowed at the  
outputs of the DACs is about 6 V. This constrains the minimum  
value of VMID to be 6 V. The output amplifiers will swing and  
settle cleanly, as described on the specification page, for output  
voltages within 1.5 V from each supply voltage rail.  
ADDRESSED CHANNEL LOADING:  
When channel address (A0, A1, A2) = 000 through 101, the  
video data is loaded into Channels 0 through 5. (STSQ/CS  
functions as “Chip Selection” this case.)  
For a given value of VMID, the voltage required to saturate the  
video output voltages defines the maximum usable full-scale  
voltage. For example, if VMID is less than AVCC/2, the maxi-  
mum value of VFS is (VMID – 1.5 V). If VMID is greater than  
AVCC/2, the maximum useful VFS is (AVCC – 1.5 – VMID).  
Figure 1 graphically describes these limiting factors.  
INTERNALLY SEQUENCED LOADING:  
When channel address = 111 the video data is loaded in a  
sequence determined internally. The sequencing is initiated by  
a pulse applied to STSQ/CS input. The count proceeds from  
0 to 5 if R/L is LOW or from 5 to 0 if R/L is HIGH.  
6
DAC TRANSFER FUNCTION  
VOUT = VMID + VFS × (1 – N/1023); if INV is HIGH,  
4.5  
VOUT = VMID – VFS × (1 – N/1023); if INV is LOW  
where VFS = 2 × (VREFHI – VREFLO)  
6
7.5  
VMID – Volts  
Figure 1. Valid Range for VMID with Respect to VFS  
(AVCC = 15 V)  
REV. B  
–3–  
AD8380  
ABSOLUTE MAXIMUM RATINGS1  
MAXIMUM POWER DISSIPATION  
Supply Voltage AVCC–AVEE . . . . . . . . . . . . . . . . . . . . . 26 V  
The maximum power that can be safely dissipated by the  
AD8380 is limited by the associated rise in 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 parametric performance due to a change in  
the stresses exerted on the die by the package. Exceeding a junc-  
tion temperature of 175°C for an extended period can result in  
device failure.  
Internal Power Dissipation2  
Quad Flat Package (S) . . . . . . . . . . . . . . . . . . . . . . . 1.7 W  
Output Short Circuit Duration  
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . See Associated Text  
Storage Temperature Range . . . . . . . . . . . . –65°C to +125°C  
Operating Temperature Range . . . . . . . . . . . . . . 0°C to 85°C  
Lead Temperature Range (Soldering 10 sec) . . . . . . . . . 300°C  
NOTES  
1Stresses above those listed under Absolute Maximum Ratings may cause perma-  
nent 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 absolute maximum rating  
conditions for extended periods may affect device reliability.  
2Specification is for device in free air:  
Output Short Circuit Limit  
The AD8380’s internal short circuit limitation is not sufficient to  
protect the device in the event of a direct short circuit between a  
video output and a power supply voltage rail (VCC or VEE). Tem-  
porary short circuits can reduce an output’s ability to source or  
sink current and, therefore, impact the device’s ability to drive a  
load. Short circuits of extended duration can cause metal lines to  
fuse open, rendering the device nonfunctional.  
44-Lead MQFP Package: θJA = 73°C/W (Still Air), where PD = (TJ – TA)/θJA  
.
θ
JC = 22°C/W.  
PIN CONFIGURATION  
To prevent these problems, it is recommended that a series  
resistor of 25 or greater be placed as close as possible to the  
AD8380’s video outputs. This will serve to substantially reduce  
the magnitude of the fault currents and protect the outputs from  
damage caused by intermittent short circuits. This may not be  
enough to guarantee that the maximum junction temperature  
(150°C) is not exceeded under all conditions. To ensure proper  
operation, it is necessary to observe the maximum power derat-  
ing curve in Figure 2 below.  
44 43 42 41 40 39 38 37 36 35 34  
1
2
33  
32  
31  
30  
29  
28  
27  
26  
25  
24  
23  
NC  
VID0  
PIN 1  
IDENTIFIER  
DB0  
DB1  
AVCC0,1  
VID1  
3
4
DB2  
AVEE1,2  
VID2  
5
DB3  
3.0  
AD8380  
TOP VIEW  
(Not to Scale)  
6
DB4  
AVCC2,3  
VID3  
T , MAX = 150؇C  
J
7
DB5  
8
DB6  
AVEE3,4  
VID4  
2.5  
2.0  
1.5  
1.0  
0.5  
9
DB7  
10  
11  
DB8  
AVCC4,5  
VID5  
(MSB) DB9  
12 13 14 15 16 17 18 19 20 21 22  
NC = NO CONNECT  
0
10  
20  
30  
40  
50  
60  
70  
80  
90  
AMBIENT TEMPERATURE ؇C  
Figure 2. Maximum Power Dissipation vs. Temperature  
ORDERING GUIDE  
Temperature  
Range  
Package  
Description  
Package  
Option  
Model  
AD8380JS  
0°C to 85°C  
44-Lead MQFP  
S-44A  
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 AD8380 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.  
WARNING!  
ESD SENSITIVE DEVICE  
–4–  
REV. B  
Typical Performance CharacteristicsAD8380  
INV = H  
VMID = 7V  
VFS = 5V  
CODE = 0  
VMID = 7V  
VFS = 5V  
VMID + VFS  
VMID  
25  
25  
VIDX  
C
VIDX  
L
C
L
150pF  
150pF  
1.25V/DIV  
1V/DIV  
VMID VFS  
INV = L  
VMID VFS  
20ns/DIV  
20ns/DIV  
TPC 1. Invert Switching 10 V Step Response (Rise) at CL  
TPC 4. Data Switching Full-Scale Step Response (Fall) at  
CL, INV = L  
INV = H  
VMID = 7V  
VFS = 5V  
VMID + VFS  
1.25V/DIV  
CODE = 0  
VMID = 7V  
VFS = 5V  
VMID + VFS  
25  
25⍀  
VIDX  
C
L
VIDX  
C
L
150pF  
150pF  
1V/DIV  
VMID  
INV = L  
VMID VFS  
20ns/DIV  
20ns/DIV  
TPC 2. Invert Switching 10 V Step Response (Fall) at CL  
TPC 5. Data Switching Full-Scale Step Response (Rise) at  
CL, INV = H  
VMID = 7V  
VFS = 5V  
VMID = 7V  
VFS = 5V  
VMID + VFS  
VMID  
25⍀  
VIDX  
C
L
150pF  
1V/DIV  
VMID  
1V/DIV  
25⍀  
VIDX  
C
L
150pF  
VMID VFS  
20ns/DIV  
20ns/DIV  
TPC 3. Data Switching Full-Scale Step Response (Rise) at  
CL, INV = L  
TPC 6. Data Switching Full-Scale Step Response (Fall) at  
CL, INV = H  
–5–  
REV. B  
AD8380  
VMID = 7V  
VFS = 5V  
INV = H  
25  
VMID = 7V  
VFS = 5V  
INV = H  
VIDX  
C
L
VMID+ VFS  
150pF  
VMID +VFS  
VMID  
25  
VIDX  
C
L
150pF  
VMID  
VMID +VFS  
t
= 0  
t
= 0  
VMID  
10ns/DIV  
10ns/DIV  
TPC 10. Output Settling Time Response to 0.25% of Full  
Scale (Falling Edge) at CL  
TPC 7. Output Settling Time Response to 1% of Full  
Scale (Rising Edge) at CL  
7.5  
5.5  
3.5  
VMID = 7V  
VFS = 5V  
INV = H  
CODE 482  
1.5  
0
25⍀  
VIDX  
CODE 738  
C
L
150pF  
1.5  
3.5  
5.5  
7.5  
VMID +VFS  
VMID  
VMID  
t = 0  
10ns/DIV  
0
10  
20  
30  
40  
50  
60  
70  
80  
90  
TEMPERATURE ؇C  
TPC 8. Output Settling Time Response to 1% of Full  
Scale (Falling Edge) at CL  
TPC 11. Differential Error Voltage (VDE) vs. Temperature  
3.5  
2.5  
VMID = 7V  
VFS = 5V  
INV = H  
1.5  
CODE 738  
VMID + VFS  
0.5  
0
0.5  
CODE 482  
25⍀  
1.5  
2.5  
3.5  
VIDX  
C
L
150pF  
VMID +VFS  
t = 0  
VMID  
0
10  
20  
30  
40  
50  
60  
70  
80  
90  
10ns/DIV  
TEMPERATURE ؇C  
TPC 12. Common-Mode Error Voltage (VCME) vs.  
Temperature  
TPC 9. Output Settling Time Response to 0.25% of Full  
Scale (Rising Edge) at CL  
–6–  
REV. B  
AD8380  
0.5  
0.4  
0.5  
0.4  
0.3  
0.3  
0.2  
0.2  
0.1  
0.1  
0
0
0.1  
0.2  
0.3  
0.4  
0.5  
0.1  
0.2  
0.3  
0.4  
0.5  
0
128  
256  
384  
512  
CODE  
640  
768  
896  
1024  
0
128  
256  
384  
512  
CODE  
640  
768  
896  
1024  
TPC 13. Differential Nonlinearity (DNL) vs. Code, INV = H  
TPC 16. Integral Nonlinearity (INL) vs. Code, INV = H  
0.5  
0.4  
0.5  
0.4  
0.3  
0.3  
0.2  
0.2  
0.1  
0.1  
0
0
0.1  
0.2  
0.3  
0.4  
0.5  
0.1  
0.2  
0.3  
0.4  
0.5  
0
128  
256  
384  
512  
640  
768  
896  
1024  
1024  
0
128  
256  
384  
512  
640  
768  
896  
CODE  
CODE  
TPC 14. Differential Nonlinearity (DNL) vs. Code, INV = L  
TPC 17. Integral Nonlinearity (INL) vs. Code, INV = L  
7.5  
5.0  
3.5  
2.5  
2.5  
0
0
2.5  
3.5  
2.5  
5.0  
7.5  
1024  
0
128  
256  
384  
512  
640  
768  
896  
1024  
0
128  
256  
384  
512  
CODE  
640  
768  
896  
CODE  
TPC 15. Differential Error Voltage (VDE) vs. Code  
TPC 18. Common-Mode Error Voltage (VCME) vs. Code  
REV. B  
–7–  
AD8380  
VFS = 5V  
VMID = 7V  
INV = L  
R
C
= 25⍀  
S
L
= 25⍀  
VID0,1,2,3,4  
OUTPUT @ CODE 0  
1mV/DIV  
= 150pF  
2.5V/DIV  
VID5  
(QUIET)  
20mV/DIV  
CLK  
2V/DIV  
20ns/DIV  
10ns/DIV  
TPC 19. Clock Switching Transient (Feedthrough) at CL  
TPC 21. All-Hostile Crosstalk at CL  
60  
40  
VFS = 5V  
VMID = 7V  
INV = L  
20  
R
C
= 25⍀  
S
L
5mV/DIV  
2V/DIV  
= 150pF  
0
OUTPUT  
@ CODE 1023  
VOUTP (INV = L)  
20  
40  
DATA  
VOUTN (INV = H)  
60  
20ns/DIV  
80  
10k  
100k  
FREQUENCY Hz  
1M  
5M  
TPC 20. Data Switching Transient (Feedthrough) at CL  
TPC 22. AVCC Power Supply Rejection vs. Frequency  
–8–  
REV. B  
AD8380  
THEORY OF OPERATION  
The region over which the output voltage varies with input code  
is defined by the status of the INV input. When INV is low, the  
video output voltages rise 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 voltages drop  
from (VMID + VFS) to VMID with increasing code (see  
Figure 4).  
The AD8380 is a system building block designed to directly drive  
the columns of poly-silicon LCD panels of the type popular-  
ized 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 parallel input. Precision current  
feedback amplifiers providing well-damped pulse responses  
and rapid voltage settling into large capacitive loads buffer the  
six outputs. Excellent linearity performance and laser trimming  
of scale factors and output offsets at the wafer level ensure low  
absolute output errors over all input codes. Tight channel-to-  
channel matching in high channel count systems is guaranteed  
by reliance on an externally-applied voltage reference.  
For each value of input code there are then two possible values  
for the output voltage, depending on the status of INV. When  
INV is low the output is defined as VOUTP(N) where N refers  
to the input code, and the P refers to the positive slope of the  
voltage variation with code. When INV is high, the output is  
defined as VOUTN(N).  
To best correlate transfer function errors to image artifacts, the  
overall accuracy of the AD8380 is defined by comparing the output  
voltages, VOUTP(N) and VOUTN(N), to each other and to  
their ideal values. Two parameters are defined, one dependent  
on the difference between the signal amplitudes at a particular  
code, and one dependent on their average value. These are VDE  
and VCME. Their defining expressions are:  
AD8380  
10  
10  
10  
10  
10  
10  
10  
2-STAGE  
LATCH  
DB [0:9]  
VID0  
VID1  
VID2  
DAC  
DAC  
DAC  
2-STAGE  
LATCH  
CLK  
STSQ/CS  
XFR  
CHANNEL  
SELECTOR  
2-STAGE  
LATCH  
VDE = [VOUTN(N) – VOUTP(N)]/2 – [(1 – N/1023) × VFS]  
where  
E/O  
R/L  
10  
10  
10  
10  
10  
10  
2-STAGE  
LATCH  
VID3  
VID4  
VID5  
DAC  
DAC  
DAC  
A[0:2]  
3
N = input code, and VFS = 2 × (VREFHI VREFLO)  
VCME = [[VOUTN(N) +VOUTP(N)]/2 – VMID] × (1/2)  
2-STAGE  
LATCH  
STBY  
BYP  
BIAS  
where  
2-STAGE  
LATCH  
VMID = midpoint reference voltage for the video outputs.  
VREFHI  
SCALING  
CONTROL  
Setting the Full-Scale Output  
VREFLO  
The full-scale output voltage (VFS), which defines the maxi-  
mum output voltage excursion for a full code input transition, is  
defined as twice the voltage difference between the VREFHI and  
VREFLO inputs.  
INV  
VMID  
Figure 3. Top Level Block Diagram  
Transfer Function  
Operating Modes, Control Logic and DAC Latches  
The transfer function of the AD8380 is made up of two regions  
of operation, in which the video output voltages are either above  
or below an output reference voltage externally applied at the  
VMID input.  
Control logic included on the AD8380 chip facilitates channel  
loading in ascending or descending order (for image mirroring),  
data loading on rising or falling clock edges (for even/odd word  
loading), and addressing and loading individual channels (for  
system testing or debugging). The on-chip logic makes it easy to  
build systems requiring more than six drive channels per color.  
(VMID + VFS)  
INV = H  
DAC latches are of a two-stage master-slave design that guaran-  
tees all channel outputs are updated simultaneously.  
VOUTN  
VMID  
VOUTP  
INV = L  
(VMID VFS)  
0
1023  
INPUT CODE  
Figure 4. Definition of Output Transfer Function  
REV. B  
–9–  
AD8380  
SVGA System Operation  
1 COLOR OF EVEN/ODDXGA  
DVCC  
3
An SVGA system is characterized by the requirement of six  
channels of panel drive for each displayed color. Such a system  
would use a single AD8380 per color.  
STSQ/CS  
XFR  
A[0:2]  
STSQ_A  
STSQ_B  
XFR  
AD8380  
DEVICE A”  
E/O  
R/L  
INV  
E/O_A  
R/L  
INV  
PANEL  
With E/O and all address bits A[0:2] set high, channel loading  
commences on the first rising edge of CLK following a valid  
assertion of the Start Sequence (STSQ) input. The second stage  
latches, and therefore the video outputs, are updated on the  
first falling edge of the clock following a valid Transfer (XFR)  
signal. (See Figure 5 for signal timing details.)  
6
VIDEO  
OUT  
CONTROLLER  
DB[0:9]  
CLK  
E/O_B  
CLKIN  
DVCC  
3
STSQ/CS  
XFR  
A[0:2]  
IMAGE  
AD8380  
DEVICE B”  
PROCESSOR  
E/O  
R/L  
INV  
6
DB[0:9]  
5
0
5
0
VIDEO  
OUT  
DB[0:9]  
VIDEO  
t1  
t2  
DB[0:9]  
10  
DCLK/2  
t7  
CLK  
t7  
2.0V  
2.0V  
0.8V  
CLK  
0.8V  
STSQ/CS  
Figure 6. Even/Odd: Outputs of Devices A and B are  
Configured as Even and Odd Data Channels and Loading  
Sequence Is Defined by Status of E/O and R/L Inputs  
t3  
t4  
XFR  
t5  
t6  
Figure 5. Sequenced SVGA Timing (A[0:2] = HIGH, E/O =  
HIGH, See Table I)  
10  
11  
0
9
10  
11  
DB[0:9]  
0
t1  
t2  
Table I. Sequenced SVGA Data Byte to Channel Assignment  
CLK  
(EVEN  
CHIP)  
Channel Number  
Data Byte Number  
t4  
t3  
E/O = HIGH VID0  
0
1
2
3
4
5
STSQ/CS  
(EVEN  
CHIP)  
R/L = LOW  
VID1  
VID2  
VID3  
VID4  
VID5  
t1  
t2  
CLK  
(ODD  
CHIP)  
t4  
t3  
STSQ/CS  
(ODD  
CHIP)  
t5  
t6  
Load Sequence Switching (Right/Left Control)  
To facilitate image mirroring, the order in which channels are  
loaded can be easily switched. When the voltage on the right/left  
control input (R/L) is low, the internal sequencer will load data  
starting with Channel 0 and counting up to Channel 5. When  
this voltage is high, channel loading will be in reverse order, from  
Channel 5 down to Channel 0.  
XFR  
A0:A2 = HIGH  
Figure 7. Sequenced Even/Odd XGA Timing, A[0:2] =  
HIGH (See Table II)  
XGA System Operation  
Table II. Sequenced Even/Odd XGA Data Byte to  
Channel Assignment  
In an XGA system, twelve column drivers (two AD8380s) are  
required for each color (refer to Figure 6). An “even/odd”  
system, in which one AD8380 drives even numbered columns  
and another drives odd numbered columns, can be easily imple-  
mented as detailed in Figures 7 and 8. A clock at one-half the  
pixel rate is applied to the CLK input. Even bytes are loaded on  
the rising edge of the clock, while odd bytes are loaded on the  
falling edge. Identifying whether a chip is to load on rising or falling  
edges is done by setting the proper level on the E/O input.  
Data Byte Number  
Channel Number  
R/L = LOW  
R/L = HIGH  
E/O = HIGH  
VID0  
0
2
4
6
8
10  
10  
8
6
4
2
VID1  
VID2  
VID3  
VID4  
VID5  
0
E/O = LOW  
VID0  
VID1  
VID2  
VID3  
VID4  
VID5  
1
3
5
7
9
11  
11  
9
7
5
3
1
–10–  
REV. B  
AD8380  
DCLK  
AD8380 INPUT DATA  
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24  
CLK (AD8380)  
STSQ/CS (EVEN)  
STSQ/CS (ODD)  
XFR  
12  
24  
0
14  
2
16  
4
18  
6
20  
8
AD8380 #1  
22  
10  
EVEN CHANNEL  
(E/O = HIGH)  
(R/L = LOW)  
12  
0
14  
16  
18  
20  
22  
2
4
6
8
10  
13  
1
15  
3
17  
5
19  
7
21  
9
AD8380 #2  
23  
13  
15  
17  
19  
21  
23  
11  
1
ODD CHANNEL  
(E/O = LOW)  
(R/L = LOW)  
3
5
7
9
11  
Figure 8. Operation of Even/Odd XGA System  
REV. B  
–11–  
AD8380  
SXGA and Beyond  
APPLICATION  
Very high resolution display systems can be built using the E/O  
XGA system as a model. By using four AD8380s, twenty-four  
columns can be driven together for an SXGA display. Two would  
be designated for even columns and two for odd. Four separate  
STSQ signals would be used to coordinate data loading with a  
single XFR to synchronize updating of output voltages.  
The AD8380 is a mixed-signal, high speed, very accurate device  
with multiple channels. In order to realize its specifications, it is  
essential to use a properly designed circuit board.  
Layout and Grounding  
The analog and digital sections of the AD8380 are pinned  
out on approximately opposite sides of the package. When  
laying out a circuit board, please keep these sections separate  
from each other to minimize crosstalk and noise coupling of  
the digital input signals into the analog outputs.  
Using a single external voltage source to drive the VREF inputs  
on all drivers for a particular color and a single voltage source  
for all their VMID inputs, will guarantee matching for all channels.  
The exceptional accuracy of the AD8380’s transfer function will  
ensure that high channel count systems can be built without fear of  
image artifacts resulting from channel-to-channel matching errors.  
All signal trace lengths should be made as short and direct as  
possible to prevent signal degradation due to parasitic effects.  
Please note that digital signals should not cross or be routed  
near analog signals.  
Direct Channel Loading  
For debug or characterization purposes, it may be desirable to  
load data directly into a single channel without requiring exercise  
of the STSQ and XFR inputs. This can be done by applying  
dc logic high levels to the STSQ and XFR inputs, and addressing  
the desired channel through the A[0:2] inputs. Data will then  
be loaded into the selected channel on each falling edge of the  
CLK signal.  
It is imperative to provide a solid ground plane under and  
around the part. All of the ground pins of the part should be  
directly connected to the ground plane with no extra signal  
path length. For conventional operation, this includes the  
pins DVEE, AVEEDAC, AVEEBIAS, AVEE0, AVEE1,2;  
AVEE3,4; and AVEE5. The return currents for any of the sig-  
nals for the part should be routed close to the ground pin for  
that section to prevent stray signals from appearing on other  
ground pins.  
The maximum rate at which a channel can be updated will be  
limited by the settling time of the output amplifiers.  
Addressed Channel Loading  
Power Supply Bypassing  
The direct channel loading method can be extended. Channels  
may be loaded in an arbitrary sequence through the use of an  
active XFR signal with STSQ set to a high level. Use the A[0:2]  
inputs to define the desired channel sequence. Data will be loaded  
on the falling edge of CLK into the channel whose address was  
valid on the preceding rising edge of CLK. All channel outputs  
are then updated together by qualification of a valid XFR signal.  
See Figure 9 for timing details.  
The AD8380 has several power supply and reference voltages  
that must be properly bypassed to the ground plane for opti-  
mum performance. The bypass capacitor for each supply pin, as  
well as VREFHI, VREFLO, and VMID, should be connected as  
close as possible to the IC pins and directly to the ground plane.  
A 0.1 µF capacitor, preferably a ceramic chip, should be used to  
minimize lead length.  
To provide low frequency, high current bypassing, larger value  
tantalum capacitors should also be used. These should be con-  
nected from the supply to ground, but it is not necessary to place  
these close to the IC pins. Stray inductance will not greatly affect  
their performance. The high current outputs should be bypassed  
with these capacitors. It is recommended that two 22 µF tantalum  
capacitors be placed from the AVCC supply to ground at either  
end of the output side of the IC. AVCCBIAS and AVCCDAC  
should each have a 10 µF tantalum bypass capacitor to ground.  
See Figure 10.  
DB[0:9]  
5
0
1
5
0
t1  
t2  
CLK  
t5  
t6  
XFR  
t8  
t9  
VREFHI Reference Distribution  
In a system that uses more than one AD8380 per color, it is  
important that all of the AD8380 devices operate from equal  
reference voltages to ensure that the video outputs are well  
matched. VREFLO is not a concern due to its high input resis-  
tance and very low bias current. Therefore, it is not likely that  
there will be significant dc voltage drops in the circuit traces to that  
supply. It is recommended to have good local supply bypassing  
at each AD8380 from their respective VREFLOs to ground.  
A[0:2]  
IN THIS CASE, INPUT LATCHES ARE LOADED IN THE ORDER SELECTED  
BY A[0:2], THEN DACs ARE UPDATED TOGETHER BY XFR.  
Figure 9. Addressed Channel Timing (E/O = HIGH,  
STSQ/CS = HIGH)  
Standby Mode  
A high level applied to the standby (STBY) input will turn off  
most of 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 a low level, normal operation is restored.  
The higher input current that flows in the VREFHI circuit  
requires that this be laid out more carefully. VREFHI connects  
internally to a 20 kresistor for each of the six channels to provide  
an input resistance of about 3.3 k. Thus with a (VREFHI –  
VREFLO) voltage of 2.5 V (to yield a VFS of 5.0 V), about  
750 µA will flow into each VREFHI circuit.  
–12–  
REV. B  
AD8380  
In order to obtain the best matching, the traces to each of the  
VREFHI pins of the AD8380s should be connected by an  
approximately same length and same width circuit trace in a  
“star” configuration. The source of the VREFHI voltage should  
be at the center of the “star.” Therefore, the VREFHI currents  
for two devices will not share a significant length of circuit trace,  
and each trace will provide an approximately equal voltage drop.  
For example, if a trace length is 5 in. long (13 cm.), then the  
trace width for a 1 oz. copper foil should be wider than 0.025 in.  
(0.7 mm) in order to keep the trace impedance below 100 m.  
Driving a Capacitive Load  
A purely capacitive load can react with output impedance of the  
AD8380 resulting in overshoot and ringing in its step response.  
To minimize this effect, and optimize settling time, it is recom-  
mended that a 25 resistor be placed in series with each of the  
driver’s outputs as shown in Figure 10.  
In addition, if the VREFHI traces must be long, then the traces  
should be widened to minimize differences in the voltage drops  
due to differences in the VREFHI input currents of different  
AD8380s. The dc resistance of these traces should be less than  
100 m. If the VREFHI input current is about 1 mA, then the  
voltage drop will be about 100 µV.  
10V  
22F  
10F  
10F  
22F  
2.500V  
15V  
3.3V  
7V  
0.1F  
0.1F  
0.1F  
0.1F  
0.1F  
0.1F  
0.1F  
0.1F  
10F  
0.1F  
10F  
DVCC  
VREFHI  
VREFLO AVCCDAC  
AVCCBIAS  
AVCC0,1  
AVCC2,3  
AVCC4,5  
VMID  
CLK  
10  
DB[0:9]  
XFR  
R
24.9⍀  
(TYP FOR 6)  
S
E/O  
R/L  
6
LOGIC  
DAC  
BIAS  
DRIVERS  
VID0 VID5  
STSQ/CS  
3
A[0:2]  
INV  
STBY  
AVEE1,2  
AVEE3,4  
DVEE  
AVEEDAC  
BYP AVEEBIAS  
AVEE0  
AVEE5  
0.1F  
Figure 10. Interface Drawing  
REV. B  
–13–  
AD8380  
OUTLINE DIMENSIONS  
Dimensions shown in inches and (mm).  
44-Lead MQFP  
(S-44A)  
0.530 (13.45)  
0.510 (12.95)  
SQ  
0.096 (2.45)  
MAX  
0.398 (10.10)  
0.390 (9.90)  
SQ  
0.041 (1.03)  
0.029 (0.73)  
44  
34  
1
33  
SEATING  
PLANE  
0.315 (8.00)  
REF  
TOP VIEW  
(PINS DOWN)  
11  
23  
0.010 (0.25)  
MAX  
22  
12  
0.009 (0.23)  
0.005 (0.13)  
0.018 (0.45)  
0.012 (0.30)  
0.031 (0.80)  
BSC  
0.083 (2.10)  
0.077 (1.95)  
AD8380Revision History  
Location  
Page  
Data sheet changed from REV. A to REV B.  
Single-Channel Block Diagram section and graphic removed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9  
–14–  
REV. B  
–15–  
–16–  

相关型号:

AD8380JS

Fast, High-Voltage Drive, 6-Channel Output DecDriver Decimating LCD Panel Driver
ADI

AD8381

Fast, High Voltage Drive, 6-Channel Output DecDriverTM Decimating LCD Panel Driver
ADI

AD8381-EB

Fast, High Voltage Drive, 6-Channel Output DecDriver Decimating LCD Panel Driver
ADI

AD8381AST

Fast, High Voltage Drive, 6-Channel Output DecDriverTM Decimating LCD Panel Driver
ADI

AD8381AST-REEL

Fast, High Voltage Drive, 6-Channel Output DecDriverTM Decimating LCD Panel Driver
ADI

AD8381JST

Fast, High Voltage Drive, 6-Channel Output DecDriver Decimating LCD Panel Driver
ADI

AD8381JSTZ

Fast, High Voltage Drive, 6-Channel Output DecDriver Decimating LCD Panel Driver
ADI

AD8382

High Performance 12-Bit, 6-Channel Output, Decimating LCD DecDriver
ADI

AD8382ACP

High Performance 12-Bit, 6-Channel Output, Decimating LCD DecDriver
ADI

AD8382ACPZ

暂无描述
ADI

AD8383

Low Cost 10-Bit, 6-Channel Output Decimating LCD DecDriver
ADI

AD8383ACPZ

Low Cost 10-Bit, 6-Channel Output Decimating LCD DecDriver
ADI