AD1580
Rev. F | Page 6 of 12
THEORY OF OPERATION
The AD1580 uses the band gap concept to produce a stable,
low temperature coefficient voltage reference suitable for high
accuracy data acquisition components and systems. The device
makes use of the underlying physical nature of a silicon tran-
sistor base emitter voltage in the forward biased operating
region. All such transistors have an approximately −2 mV/°C
temperature coefficient (TC), which is unsuitable for use
directly as a low TC reference; however, extrapolation of
the temperature characteristic of any one of these devices to
absolute zero (with collector current proportional to absolute
temperature) reveals that its V
BE
goes to approximately the
silicon band gap voltage. Thus, if a voltage could be developed
with an opposing temperature coefficient to sum with V
BE
, a
zero TC reference would result. The AD1580 circuit in Figure 10
provides such a compensating voltage, V1, by driving two
transistors at different current densities and amplifying the
resultant V
BE
difference (∆V
BE
, which has a positive TC).
The sum of V
BE
and V1 provides a stable voltage reference.
V+
V–
V1
ΔV
BE
V
BE
00700-010
Figure 10. Schematic Diagram
APPLYING THE AD1580
The AD1580 is simple to use in virtually all applications. To
operate the AD1580 as a conventional shunt regulator (see
Figure 11), an external series resistor is connected between the
supply voltage and the AD1580. For a given supply voltage, the
series resistor, R
S
, determines the reverse current flowing through
the AD1580. The value of R
S
must be chosen to accommodate
the expected variations of the supply voltage, V
S
; load current,
I
L
; and the AD1580 reverse voltage, V
R
; while maintaining an
acceptable reverse current, I
R
, through the AD1580.
The minimum value for R
S
should be chosen when V
S
is at
its minimum and I
L
and V
R
are at their maximum, while
maintaining the minimum acceptable reverse current.
The value of R
S
should be large enough to limit I
R
to 10 mA
when V
S
is at its maximum and I
L
and V
R
are at their minimum.
The equation for selecting R
S
is as follows:
R
S
= (V
S
V
R
)/(I
R
+ I
L
)
Figure 12 shows a typical connection of the AD1580BRT
operating at a minimum of 100 µA. This connection can
provide ±1 mA to the load while accommodating ±10%
power supply variations.
V
S
I
R
+ I
L
R
S
V
OUT
I
L
V
R
I
R
00700-011
Figure 11. Typical Connection Diagram
+5V(+3V) ±10%
2.94k
(1.30k)
R
S
V
R
V
OUT
00700-012
Figure 12. Typical Connection Diagram
TEMPERATURE PERFORMANCE
The AD1580 is designed for reference applications where stable
temperature performance is important. Extensive temperature
testing and characterization ensure that the devices performance is
maintained over the specified temperature range.
Some confusion exists in the area of defining and specifying
reference voltage error over temperature. Historically, references
have been characterized using a maximum deviation per degree
Celsius, for example, 50 ppm/°C. However, because of nonlinear-
ities in temperature characteristics that originated in standard
Zener references (such as S type characteristics), most manufac-
turers now use a maximum limit error band approach to specify
devices. This technique involves the measurement of the output
at three or more different temperatures to guarantee that the
voltage falls within the given error band. The proprietary
curvature correction design techniques used to minimize the
AD1580 nonlinearities allow the temperature performance to
be guaranteed using the maximum deviation method. This
method is of more use to a designer than the one that simply
guarantees the maximum error band over the entire temper-
ature change.
Figure 13 shows a typical output voltage drift for the AD1580
and illustrates the methodology. The maximum slope of the two
diagonals drawn from the initial output value at +25°C to the
output values at +85°C and −40°C determines the performance
grade of the device. For a given grade of the AD1580, the designer
can easily determine the maximum total error from the initial
tolerance plus temperature variation.
AD1580
Rev. F | Page 7 of 12
OUTPUT VOLTAGE (V)
1.2238
1.2248
1.2250
1.2252
1.2254
1.2256
1.2258
1.2244
1.2246
1.2240
1
.2242
V
MAX
V
MIN
SLOPE = TC =
(V
MAX
– V
O
)
(+85°C – +25
°C) × 1.225 × 10
–6
SLOPE = TC =
(V
MIN
– V
O
)
(–40°C – +25°C) × 1.225 × 10
–6
V
O
–55 –35 –15 5 25 45 65 85 105 125
TEMPERATURE (°C)
00700-013
Figure 13. Output Voltage vs. Temperature
For example, the AD1580BRT initial tolerance is ±1 mV;
a ±50 ppm/°C temperature coefficient corresponds to an
error band of ±4 mV (50 × 10
−6
× 1.225 V × 65°C). Thus, the
unit is guaranteed to be 1.225 V ± 5 mV over the operating
temperature range.
Duplication of these results requires a combination of high
accuracy and stable temperature control in a test system.
Evaluation of the AD1580 produces a curve similar to that
in Figure 5 and Figure 13.
VOLTAGE OUTPUT NONLINEARITY vs.
TEMPERATURE
When a reference is used with data converters, it is important to
understand how temperature drift affects the overall converter
performance. The nonlinearity of the reference output drift
represents an additional error that is not easily calibrated out of
the system. This characteristic (see Figure 14) is generated by
normalizing the measured drift characteristic to the end point
average drift. The residual drift error of approximately 500 ppm
shows that the AD1580 is compatible with systems that require
10-bit accurate temperature performance.
600
300
0
RESIDUAL DRIFT ERROR (ppm)
500
400
200
100
–55 –35 –15 5 25 45 65 85 105 125
TEMPERATURE (°C)
00700-014
Figure 14. Residual Drift Error
REVERSE VOLTAGE HYSTERESIS
A major requirement for high performance industrial
equipment manufacturers is a consistent output voltage at
nominal temperature following operation over the operating
temperature range. This characteristic is generated by measur-
ing the difference between the output voltage at +25°C after
operation at +85°C and the output, at +25°C after operation
at −40°C. Figure 15 displays the hysteresis associated with the
AD1580. This characteristic exists in all references and has been
minimized in the AD1580.
QUANTITY
0
15
20
25
30
35
40
5
10
HYSTERESIS VOLT
AGE (µV)
–400 –300 –200 –100 0 100 200 300 400
00700-015
Figure 15. Reverse Voltage Hysteresis Distribution
OUTPUT IMPEDANCE vs. FREQUENCY
Understanding the effect of the reverse dynamic output imped-
ance in a practical application may be important to successfully
apply the AD1580. A voltage divider is formed by the AD1580
output impedance and the external source impedance. When
an external source resistor of about 30 kΩ (I
R
= 100 μA) is used,
1% of the noise from a 100 kHz switching power supply is devel-
oped at the output of the AD1580. Figure 16 shows how a 1 µF
load capacitor connected directly across the AD1580 reduces
the effect of power supply noise to less than 0.01%.
1k
10
0.1
1
100
FREQUENCY (Hz)
C
L
= 0
C
L
= 1µF
ΔI
R
= 0.1I
R
I
R
= 100µA
I
R
= 1mA
OUTPUT IMPEDANCE (Ω)
10 100 1k 10k 100k 1M
00700-016
Figure 16. Output Impedance vs. Frequency
AD1580
Rev. F | Page 8 of 12
NOISE PERFORMANCE AND REDUCTION
The noise generated by the AD1580 is typically less than
5 µV p-p over the 0.1 Hz to 10 Hz band. Figure 17 shows the
0.1 Hz to 10 Hz noise of a typical AD1580. Noise in a 10 Hz to
10 kHz bandwidth is approximately 20 μV rms (see Figure 18a).
If further noise reduction is desired, a 1-pole low-pass filter can
be added between the output pin and ground. A time constant
of 0.2 ms has a −3 dB point at about 800 Hz and reduces the
high frequency noise to about 6.5 μV rms (see Figure 18b).
A time constant of 960 ms has a −3 dB point at 165 Hz and
reduces the high frequency noise to about 2.9 μV rms (see
Figure 18c).
1s/DIV1µV/DIV
4.5µV p-p
00700-017
Figure 17. 0.1 Hz to 10 Hz Voltage Noise
40µV/DIV
21µV rms
20µV/DIV
10µV/DIV
10ms/DIV
6.5µV rms, τ = 0.2ms
(a)
(b)
(c)
2.90µV rms, τ = 960ms
00700-018
Figure 18. Total RMS Noise
TURN-ON TIME
Many low power instrument manufacturers are becoming
increasingly concerned with the turn-on characteristics of
components being used in their systems. Fast turn-on compo-
nents often enable the end user to keep power off when not
needed, and yet those components respond quickly when
the power is turned on for operation. Figure 19 displays the
turn-on characteristic of the AD1580.
Upon application of power (cold start), the time required for
the output voltage to reach its final value within a specified
error is the turn-on settling time. Two components normally
associated with this are time for active circuits to settle and time
for thermal gradients on the chip to stabilize. This characteristic
is generated from cold start operation and represents the true
turn-on waveform after power-up. Figure 21 shows both the
coarse and fine turn-on settling characteristics of the device;
the total settling time to within 1.0 mV is about 6 µs, and there
is no long thermal tail when the horizontal scale is expanded to
2 ms/div.
250mV/DIV
5µs/DIV
C
L
= 200pF
V
IN
0V
2.4V
00700-019
Figure 19. Turn-On Response Time
+
R
S
= 11.5k
R
L
C
L
V
OUT
V
R
V
IN
00700-020
Figure 20. Turn-On, Settling, and Transient Test Circuit
Output turn-on time is modified when an external noise
reduction filter is used. When present, the time constant
of the filter dominates overall settling.
0V
V
IN
2.4V
OUTPUT ERROR
1mV/DIV, 2µs/DIV
OUTPUT
0.5mV/DIV, 2ms/DIV
00700-021
Figure 21. Turn-On Settling

AD1580ARTZ-REEL7

Mfr. #:
Manufacturer:
Analog Devices Inc.
Description:
Voltage References 12V MICROPOWER REF IC
Lifecycle:
New from this manufacturer.
Delivery:
DHL FedEx Ups TNT EMS
Payment:
T/T Paypal Visa MoneyGram Western Union