INA156EA-2K5
Product Specifications
INA156EA-2K5
Texas Instruments — Power Management & Signal Chain
Product Overview
The INA156EA-2K5 from Texas Instruments is a power management / signal chain IC delivering high efficiency, precision regulation, and robust protection. Texas Instruments (TI) is a global leader in analog ICs, power management solutions, and embedded processors.
Superb Automation partners with leading brands including Qorvo, TI, ADI, XILINX, NVIDIA, Intel, ST, MACOM, Renesas, and Coilcraft to provide complete BOM and supply-chain solutions.
Key Features
- High-efficiency power conversion with low quiescent current
- Wide input voltage range with precision output regulation
- Integrated over-current, over-temperature, and short-circuit protection
- Small footprint packaging for space-constrained designs
- AEC-Q100 qualified options available for automotive applications
- Flexible sequencing and soft-start control
Key Specifications
| Input Voltage | V |
| Output Voltage | V |
| Output Current | A |
| Switching Frequency | kHz |
| Efficiency | % |
| Quiescent Current | µA |
| Topology | — |
| Package | — |
| Operating Temp | °C |
| Protection | — |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
INA156EA-2K5: the low cost CMOS member on a 2500 piece reel
INA156EA-2K5 is the low cost member of the CMOS instrumentation family on a 2500 piece reel, in the eight pin MSOP that is the only body it comes in. The gain strap is the same 10 or 50, the bias current the same picoamp, the bandwidth the same 550 kHz, and the supply the same 2.7 to 5.5 volts. What changes is the offset window, 8 millivolts maximum against 1 millivolt, and the gain error, 0.4 per cent against 0.1.
A reel of these is what a cost sensitive board buys when the offset is removed by the measurement rather than by the part: an AC coupled channel, a fault detector, a trim at test. The reel matters more here than on the tight families, because there is no tube and no SO-8, so the parts arrive on a carrier the line has to be able to run and a prototype has to be able to break.
INA156EA-2K5 as the vendor document publishes it
| Gain | 10 V/V with pin 1 open, 50 V/V with pin 1 linked |
| Gain law with an external resistor | 10 plus 400 kohm divided by 10 kohm plus RG |
| Gain error, G = 10 | 0.08 per cent typical, 0.4 maximum |
| Gain error, G = 50 | 0.1 per cent typical, 0.8 maximum |
| Gain drift, G = 10 | 2 ppm per degree typical, 10 maximum |
| Offset | 2.5 millivolts typical, 8 millivolts maximum |
| Offset over temperature | 9 millivolts maximum |
| Offset drift | 5 microvolts per degree |
| Common mode rejection, G = 10 | 66 dB minimum, 78 typical |
| Common mode rejection, G = 50 | 74 dB minimum, 87 typical |
| Input bias current | 1 picoamp typical, 10 picoamps maximum |
| Input impedance | 10 to the 13 ohms in parallel with 3 picofarads |
| Voltage noise at 1 kHz | 40 nanovolts per root hertz |
| Bandwidth, G = 10 | 550 kHz |
| Bandwidth, G = 50 | 110 kHz |
| Slew rate | 6.5 volts per microsecond |
| Settling to 0.01 per cent | 8 microseconds at G = 10, 15 at G = 50 |
| Output swing from rail | 5 millivolts typical, 20 maximum into 10 kohm |
| Quiescent current | 1.8 milliamps typical, 2.5 maximum, 3.2 over temperature |
| Supply range | 2.7 to 5.5 volts specified, 2.5 to 6 volts operating |
| Body | eight pin MSOP only, marking A56 |
| Temperature range | minus 55 to plus 125 degrees |
| Catalogue code | INA156EA-2K5 |
| Vendor string | INA156EA/2K5 |
| Carrier | 2500 piece reel, MSOP only |
| Grade | EA, 8 millivolts maximum offset |
The same CMOS front end with the offset window opened to eight millivolts
INA156EA-2K5 is the low cost member of the same CMOS instrumentation family as the INA155, and the difference is not only the offset bin. Offset is 2.5 millivolts typical and 8 millivolts maximum, 9 over temperature, against 200 microvolts and 1 millivolt on the other part, and the gain error rows move with it: 0.08 per cent typical and 0.4 maximum at a gain of 10, 0.1 and 0.8 at 50, where the INA155 prints 0.02 and 0.1.
That matters because it means INA156EA-2K5 is not a sorted INA155 and cannot be substituted for one on a calibrated channel without a deviation. What is unchanged is the architecture: the same pin strap for a gain of 10 or 50, the same 10 to the 13 ohm input, the same 1 picoamp bias current, the same 550 kHz and the same 2.7 to 5.5 volt supply. The quiescent current is slightly higher, 1.8 milliamps typical and 2.5 maximum against 1.7 and 2.1, which is the only row that moves in the wrong direction for a part sold on cost.
MSOP only, and the marking A56
INA156EA-2K5 exists in one body, the eight pin MSOP, and in one grade, the EA, on a 250 piece reel or a 2500 piece reel. There is no SO-8 and no tube, so a prototype that has to be hand assembled needs an adapter, and a line that runs the part needs the reel and the feeder rather than a stick of parts. The marking is A56, which is the only visible difference from the A55 on the INA155 MSOP and is what an inspector reads when the labels are gone.
The single body is a sourcing fact rather than an electrical one, and it is worth stating because the two families look interchangeable on a data sheet. A board laid out for the SO-8 INA155 cannot take INA156EA-2K5 at all, and a board laid out for the MSOP can take either, at which point the choice is the offset window and the price. The moisture level on this family is the row to keep on the order, because the MSOP is a small body with a thin package and the floor life after the bag is opened is what the line has to plan around.
550 kHz, 6.5 volts per microsecond and five microseconds to settle
INA156EA-2K5 keeps the AC behaviour of the family exactly: 550 kHz of small signal bandwidth at a gain of 10 and 110 kHz at 50, a slew rate of 6.5 volts per microsecond, 5 and 11 microseconds to settle to 0.1 per cent on a 2 volt step, 8 and 15 to 0.01 per cent, and 0.2 microseconds of overload recovery. The output is stable into the capacitive load in the vendor curve and the short circuit current is 50 milliamps into ground.
What changes is the output swing specification, which is written as 5 millivolts typical and 20 maximum rather than 5 and 10, and the gain error window, which is wide enough that a 1 per cent resistor in the strap is no longer the dominant term. For a channel that digitises at 12 bits and is trimmed once at test, none of that is visible. For a channel that expects 16 bits from a single uncalibrated pass, the 8 millivolts of offset at a gain of 10 is 80 millivolts at the output and the part is simply the wrong one.
One picoamp of bias and a 10 to the 13 ohm input
The input stage of INA156EA-2K5 is the same CMOS pair as the INA155: 1 picoamp of bias current typical, 10 maximum, offset current the same, and an input impedance of 10 to the 13 ohms in parallel with 3 picofarads on both the differential and the common mode path. On a pH electrode, a photodiode with a transimpedance stage, or a 350 ohm bridge with a long cable, that is the specification that decides whether the channel works at all.
The same clamp applies. The inputs are diode protected to both rails, and anything beyond 0.5 volts outside the supply has to be current limited to 10 milliamps, so a cable that can be plugged in while the board is live needs a series resistor. On a high impedance source the board itself is the other half of the problem: 10 to the 13 ohms in parallel with a contaminated surface is a leakage path, and a guard ring driven to the common mode voltage is worth more than any change of grade.
Sixty six decibels of rejection at a gain of 10
Common mode rejection on INA156EA-2K5 is 66 dB minimum and 78 typical at a gain of 10 over a common mode from 0.6 to 3.7 volts, falling to 65 dB minimum over temperature; at a gain of 50 it is 74 and 87. Against the 92 dB of the INA155 at the same gain that is a real difference, and it shows up as an error that moves with the sensor's common mode rather than with its signal.
The arithmetic is worth doing before the part is chosen. A bridge sitting at 1.5 volts of common mode on a 5 volt supply, with 2 volts of common mode swing from a supply that is not clean, sees 2 volts divided by 10 to the 3.3, about 1 millivolt, which is the same order as the maximum offset and eight times the typical. On a ratiometric bridge the supply moves both the bridge and the amplifier, so part of that swing cancels; on an absolute measurement it does not, and the rejection row becomes the binding specification rather than the offset row.
Where the cheap member is the right member
INA156EA-2K5 is the right part when the offset is removed by the measurement rather than by the part: an AC coupled channel, a lock-in amplifier, a chopper front end, a digital potentiometer that trims at test, or a monitor whose job is to detect a fault rather than to measure a value. In those channels the 8 millivolts is a number the system never sees, and the picoamp bias, the rail to rail output and the 2.7 volt floor are what the channel needs.
It is the wrong part when a customer's specification quotes an offset budget, when the channel is calibrated once and expected to hold, or when the bill of materials says INA155 and an inspector compares labels. Our drop in replacement check page sets out the rows to compare before any substitution is written, and on this family there are three of them: the offset window, the gain error window and the quiescent current.
One body and one grade means one order line, and the reel is the only practical carrier above prototype quantity. Our BOM shortage recovery page covers the route when the reel is short, and the pin to pin page covers the candidates that share the eight pin MSOP before a second source is qualified.
Can an INA156 replace an INA155 on an existing board? Only the MSOP versions are pin compatible, and only against a deviation, because the offset and gain error rows are four to eight times wider.
Why is the quiescent current higher on the cheaper part? The two dies are trimmed differently and the cheaper one draws 1.8 milliamps typical against 1.7. It is the only row that moves the wrong way, and it does not change the application.