
PRELIMINARY TECHNICAL DATA
REV. PrF 09/2003
AD5380
–17–
FUNCTIONAL DESCRIPTION
DAC Architecture — General
The AD5380 is a complete single supply, 40-channel,
voltage output DAC offering 14-bit resolution, available
in a 100 lead LQFP package and features both a parallel
and serial interfaces. This family includes an internal
1.25/2.5V, 10ppm/°C reference that can be used to drive
the buffered reference inputs, alternatively an external
reference can be used to drive these inputs. Reference
selection is via a bit in the control register. All channels
have an on-chip output amplifier with rail-to-rail output
capable of driving a 5k
ohm in parallel with a 200pf
load.
The architecture of a single DAC channel consists of a
14-bit resistor-string DAC followed by an output buffer
amplifier operating at a gain of two. This resistor-string
architecture guarantees DAC monotonicity. The 14-bit
binary digital code loaded to the DAC register determines
at what node on the string the voltage is tapped off before
being fed to the output amplifier. Each channel on these
devices contains independant offset and gain control
registers allowing the user to digitally trim offset and gain.
The inclusion of these registers allows the user the ability
to calibrate out errors in the complete signal chain
including the DAC using the internal M and C registers
which hold the correction factors. All channels are double
buffered allowing synchronous updating of all channels
using the
LDAC
pin. Figure 7 shows a block diagram of
a single channel on the AD5380.
The digital input transfer function for each DAC can be
represented as:
x2 = [(m + 1 )/8192 × x1] + (c-2
n-1
)
x2
is the Dataword loaded to the resistor string DAC
x1
is the 14-bit Dataword written to the DAC input
register.
m
is the13-bit Gain Coefficient (default is all 1FFF Hex
on the AD5380. The gain coefficient is written to the 13
most significant bits. If a 14 bit data word is provided to
the m register the lsb of the data word will be a zero.
n=DAC resolution (n=14 for AD5380)
c
is the14-bit Offset Coefficient (default is 2000Hex on
the AD5380)
The complete transfer function for these devices can be
represented as:
VOUT
= 2 × V
REF
× x2/2
n
x2
is the Dataword loaded to the resistor string DAC
V
REF
is the reference voltage applied to the DAC, 2.5V for
specified performance.
Data Decoding
The AD5380 contains a 14-bit data bus, DB13-DB0.
Depending on the value of REG1 and REG0 outlined in
Table 1, this data is loaded into the addressed DAC input
register(s), Offset (c) register(s), or Gain (m) register(s).
The format data, Offset (c) and gain (m) register contents
are outlined in tables II to IV.
x1 INPUT
REG
m REG
c REG
x2
DAC
REG
14-BIT
DAC
INPUT
DATA
R
R
+
-
AVDD
VOUT
VREF
Figure 7. Single Channel Architecture
DB13 to DB0
DAC Output
11 1111 1111 1111
11 1111 1111 1110
10 0000 0000 0001
10 0000 0000 0000
01 1111 1111 1111
00 0000 0000 0001
00 0000 0000 0000
2 V
REF
×
(16383/16384) V
2 V
REF
×
(16382/16384)V
2 V
REF
×
(8193/16384) V
2 V
REF
×
(8192/16384) V
2 V
REF
×
(8191/16384) V
2 V
REF
×
(1/16384) V
0
V
Table II. DAC Data format (REG1 = 1, REG0 = 1)
REG1 REG0
Register Selected
1
1
0
0
1
0
1
0
Input Data Register (x1)
Offset Register (c)
Gain Register (m)
Special Function Registers (SFRs)
Table I. Register Selection
DB13 to DB1
Gain Factor
1 1111 1111 1111
1 0111 1111 1111
0 1111 1111 1111
0 0111 1111 1111
0 0000 0000 0000
1
0.75
0.5
0.25
0
Table IV. Gain Data format (REG1 = 0, REG0 = 1)
DB13 to DB0
Offset
11 11111111 1111
11 11111111 1110
10 00000000 0001
10 00000000 0000
01 11111111 1111
00 00000000 0001
00 00000000 0000
+8191LSB
+8190LSB
+1
+0
-1
-8191
-8192
LSB
LSB
LSB
LSB
LSB
Table III. Offset Data format (REG1 = 1, REG0 = 0)