OPA388_17 Datasheet, PDF (19/34 Pages) Texas Instruments – Precision, Zero-Drift, Zero-Crossover, True Rail-to-Rail Input/Output, Operational Amplifiers

www.ti.com
OPA388, OPA2388, OPA4388
SBOS777 – DECEMBER 2016
Typical Application (continued)
8.2.1 Design Requirements
This solution has the following requirements:
• Supply voltage: 3.3 V
• Input: –1 A to 1 A
• Output: 1.65 V ±1.54 V (110 mV to 3.19 V)
8.2.2 Detailed Design Procedure
The load current, I_LOAD, flows through the shunt resistor (R_SHUNT) to develop the shunt voltage, V_SHUNT. The shunt
voltage is then amplified by the difference amplifier consisting of U1A and R₁ through R₄. The gain of the
difference amplifier is set by the ratio of R₄ to R₃. To minimize errors, set R₂ = R₄ and R₁ = R₃. The reference
voltage, V_REF, is supplied by buffering a resistor divider using U1B. The transfer function is given by Equation 1.
V_OUT = V_SHUNT ´ Gain_{Diff_Amp} + V_REF
where
• ^VSHUNT ^{= I}LOAD ^{´ R}SHUNT
•
Gain_{Diff_Amp}
=
R₄
R₃
V
REF
=
V
CC
´
R₆
R +R
•
5
6
(1)
There are two types of errors in this design: offset and gain. Gain errors are introduced by the tolerance of the
shunt resistor and the ratios of R₄ to R₃ and, similarly, R₂ to R₁. Offset errors are introduced by the voltage
divider (R₅ and R₆) and how closely the ratio of R₄ / R₃ matches R₂ / R₁. The latter value affects the CMRR of the
difference amplifier, ultimately translating to an offset error.
The value of V_SHUNT is the ground potential for the system load because V_SHUNT is a low-side measurement.
Therefore, a maximum value must be placed on V_SHUNT. In this design, the maximum value for V_SHUNT is set to
100 mV. Equation 2 calculates the maximum value of the shunt resistor given a maximum shunt voltage of
100 mV and maximum load current of 1 A.
R = _SHUNT(Max)
V_SHUNT(Max)
I_LOAD(Max)
= ^{100 mV}
1A
= 100 mW
(2)
The tolerance of R_SHUNT is directly proportional to cost. For this design, a shunt resistor with a tolerance of 0.5%
was selected. If greater accuracy is required, select a 0.1% resistor or better.
The load current is bidirectional; therefore, the shunt voltage range is –100 mV to 100 mV. This voltage is divided
down by R₁ and R₂ before reaching the operational amplifier, U1A. Take care to ensure that the voltage present
at the noninverting node of U1A is within the common-mode range of the device. Therefore, use an operational
amplifier, such as the OPA388, that has a common-mode range that extends below the negative supply voltage.
Finally, to minimize offset error, note that the OPA388 has a typical offset voltage of merely ±0.25 µV (±5 µV
maximum).
Given a symmetric load current of –1 A to 1 A, the voltage divider resistors (R₅ and R₆) must be equal. To be
consistent with the shunt resistor, a tolerance of 0.5% was selected. To minimize power consumption,
10-kΩ resistors were used.
To set the gain of the difference amplifier, the common-mode range and output swing of the OPA388 must be
considered. Equation 3 and Equation 4 depict the typical common-mode range and maximum output swing,
respectively, of the OPA388 given a 3.3-V supply.
–100 mV < V_CM < 3.4 V
(3)
100 mV < V_OUT < 3.2 V
(4)
The gain of the difference amplifier can now be calculated as shown in Equation 5.
Gain_{Diff_Amp}
=
V_{OUT_Max} - V_{OUT_Min}
R_SHUNT ´ (I_MAX - I_MIN)
=
3.2 V - 100 mV
100 mW ´ [1 A - (- 1A)]
=
15.5
V
V
(5)
Copyright © 2016, Texas Instruments Incorporated
Submit Documentation Feedback
19
Product Folder Links: OPA388 OPA2388 OPA4388