TLV5638
www.ti.com
SLAS225C–JUNE 1999–REVISED JANUARY 2004
ELECTRICAL CHARACTERISTICS (Continued)
over recommended operating conditions, Vref = 2.048 V, Vref= 1.024 V (unless otherwise noted)
STATIC DAC SPECIFICATIONS
PARAMETER
Resolution
TEST CONDITIONS
MIN
TYP
MAX
UNIT
12
bits
C and I
suffixes
±1.7
±4
LSB
Integral nonlinearity, end point ad-
justed
(1)
INL
See
Q and M
suffixes
±1.7
±0.4
±6
±1
LSB
LSB
mV
(2)
(3)
DNL
EZS
Differential nonlinearity
See
See
Zero-scale error (offset error at zero
scale)
±24
Zero-scale-error temperature coef-
ficient
(4)
EZSTC
See
10
10
ppm/°C
% full
scale V
EG
Gain error
See(5)
±0.6
(6)
EGTC
Gain error temperature coefficient
See
ppm/°C
V
OUTPUT SPECIFICATIONS
VO
Output voltage
RL= 10 kΩ
0
VDD-0.4
±0.25
% full
scale V
Output load regulation accuracy
VO = 4.096 V, 2.048 V, RL= 2 kΩ
REFERENCE PIN CONFIGURED AS OUTPUT (REF)
Vref(OUTL) Low reference voltage
Vref(OUTH) High reference voltage
1.003 1.024
2.027 2.048
1.045
2.069
1
V
VDD > 4.75 V
V
Iref(source)
Iref(sink)
Output source current
Output sink current
mA
mA
pF
dB
-1
Load capacitance
100
PSRR
Power supply rejection ratio
-65
REFERENCE PIN CONFIGURED AS INPUT (REF)
VI
RI
CI
Input voltage
0
VDD-1.5
V
Input resistance
Input capacitance
10
5
MΩ
pF
Fast
1.3
525
-80
MHz
kHz
dB
Reference input bandwidth
Reference feedthrough
REF = 0.2 Vpp + 1.024 V dc
REF = 1 Vpp at 1.024 V dc(7)
Slow
DIGITAL INPUTS
IIH
IIL
Ci
HIgh-level digital input current
VI = VDD
VI = 0 V
1
µA
µA
pF
Low-level digital input current
Input capacitance
-1
8
(1) The relative accuracy or integral nonlinearity (INL) sometimes referred to as linearity error, is the maximum deviation of the output from
the line between zero and full scale excluding the effects of zero code and full-scale errors. Tested from code 32 to 4095.
(2) The differential nonlinearity (DNL) sometimes referred to as differential error, is the difference between the measured and ideal 1 LSB
amplitude change of any two adjacent codes. Monotonic means the output voltage changes in the same direction (or remains constant)
as a change in the digital input code.
(3) Zero-scale error is the deviation from zero voltage output when the digital input code is zero.
(4) Zero-scale-error temperature coefficient is given by: EZS TC = [EZS (Tmax) - EZS (Tmin)]/Vref× 106/(Tmax - Tmin).
(5) Gain error is the deviation from the ideal output (2Vref - 1 LSB) with an output load of 10 kΩ excluding the effects of the zero-error.
(6) Gain temperature coefficient is given by: EG TC = [EG(Tmax) - EG (Tmin)]/Vref× 106/(Tmax - Tmin).
(7) Reference feedthrough is measured at the DAC output with an input code = 0x000.
5