INA132U2K5
Product Specifications
INA132U2K5
Texas Instruments — Power Management & Signal Chain
Product Overview
The INA132U2K5 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
INA132U2K5: the unity gain difference amplifier on a reel after the tube row was withdrawn
The INA132U2K5 is the plain grade of the unity gain difference amplifier on a 2500 piece large reel, and it is now the only way to buy the plain grade: the addendum marks INA132U in the tube as obsolete with no quantity and no carrier, while the reel row stays active at level 3 with 168 hours of floor life. The part is not discontinued; the carrier changed.
The function is one op amp and four laser trimmed resistors in an eight pin SOIC, with a gain of 1 V/V, 0.075 per cent maximum gain error, 0.001 per cent of full scale nonlinearity, 175 µA of supply current and a supply window from 2.7 V single to plus or minus 18 V dual. The common mode range reaches the negative rail, which is what makes it usable on a single supply shunt in the return leg.
The INA132U2K5 rows as the addendum publishes them
| Part class | Unity gain difference amplifier, trimmed resistor network |
| Vendor status | Active on the reel row, the tube row is obsolete |
| Data sheet | SBOS059B, revised January 2026 |
| Addendum row | INA132U/2K5, 2500 pieces on a large reel |
| Grade | U bin, 250 µV maximum offset |
| Gain | 1 V/V fixed |
| Gain error | 0.075 per cent maximum |
| Gain nonlinearity | 0.001 per cent of full scale |
| Input offset voltage | 250 µV maximum referred to the output |
| Offset drift | 5 µV per degree maximum |
| Common mode rejection | 90 dB typical, 76 dB minimum |
| Input impedance | 80 kΩ differential and common mode |
| Bandwidth | 300 kHz small signal |
| Slew rate | 0.1 V per microsecond on one chip origin, 0.25 on the other |
| Settling | 88 µs to 0.01 per cent after a 10 V step |
| Output | Within 0.8 V of the positive rail into 100 kΩ, 10 nF stable |
| Supply range | 2.7 V to 36 V single, plus or minus 1.35 V to plus or minus 18 V dual |
| Quiescent current | 175 µA typical, 230 µA maximum |
| Body | SOIC, D outline, 8 pins, 4.9 by 6 mm |
| Moisture level | Level 3, 260 degrees, 168 hour floor life |
One trimmed network instead of four resistors
A difference amplifier is one op amp and four resistors, and the four resistors have to be ratio matched to better than the amplifier's own gain error or the common mode rejection collapses. INA132U2K5 puts the network on the die, laser trimmed, with the ratios tracking over temperature because they sit on the same piece of silicon. The published result is 0.075 per cent maximum gain error and 0.001 per cent of full scale nonlinearity on the U bin, with a gain drift of 10 ppm per degree.
The alternative on a discrete build is four 0.1 per cent resistors with a 25 ppm coefficient, which costs more board area and more test time than the integrated network costs in unit price, and still does not track as well because four separate parts do not sit at the same temperature. That is the argument for buying the function rather than building it, and it is the reason this part exists in the catalogue at 175 microamps of supply current.
A 2.7 V to 36 V window and a common mode that reaches the negative rail
INA132U2K5 runs from a single supply between 2.7 V and 36 V, or from a dual supply between plus or minus 1.35 V and plus or minus 18 V, and the internal amplifier common mode range extends to the negative supply. That last row is what makes the part usable on a single rail: a shunt in the return leg sits at a few tens of millivolts above ground, which is inside the window on a single supply and outside it on many difference amplifiers.
The common mode range at the top is 2 times the positive rail minus 2 V, so on a 5 V single supply the window runs from ground to about 8 V, and on plus or minus 15 V it runs from minus 15 V to plus 28 V. Input voltage beyond that does not damage the part, because the absolute maximum input is 80 V, but the output stops following. On a current sense application the window is therefore never the limit; on a high side measurement it can be.
175 microamps and the 80 kohm network across the source
The supply current is 175 microamps typical, which is what lets INA132U2K5 sit in a battery instrument or a loop powered transmitter. The cost of that current is the resistor network: the differential and common mode input impedance are both 80 kohm, and 80 kohm across a source is a load. A 1 kohm source sees a 1.2 per cent gain loss from the divider, which at 0.075 per cent gain error is twenty times the amplifier contribution and has to be calibrated or buffered.
On a bridge or a shunt the loading also unbalances the source, and an unbalanced source converts common mode into differential. That is why the four source resistances around a difference amplifier should match, and why a series resistor added to one input for protection has to be added to the other. The network is ratio matched but its absolute value carries 25 per cent tolerance on the older family sheets, so the divider error is predictable but not small.
250 microvolts on the U and 500 on the UA
The offset of INA132U2K5 is referred to the output and quoted at 250 microvolts maximum on the U bin and 500 microvolts on the UA bin at a plus or minus 15 V supply, with a drift of 5 and 10 microvolts per degree. At unity gain the input referred and output referred figures are the same number, which makes this part easy to budget: the offset is the offset.
The second point about the bins is availability. In the current addendum the tube rows for both INA132U and INA132UA are marked obsolete with no quantity and no carrier, and only the 2500 piece reel rows are active. That is a real constraint on a prototype build: the part is not discontinued, but it is sold in reels, so a ten board build either buys a reel or buys from stock that was cut from one. The reel row carries moisture level 3 with a 168 hour floor life.
90 dB of rejection and the source imbalance that removes it
Common mode rejection on INA132U2K5 is 90 dB typical and 76 dB minimum on the U bin, measured with the common mode swept from minus 15 to plus 28 V. The rejection depends entirely on the ratio match between the two input paths, so a 1 kohm imbalance in the source, which is one precision resistor in series with one input, converts common mode into a differential error at roughly the reciprocal of that match, and the 90 dB becomes 60 dB or worse.
Two pins decide the same thing. The reference pin has to be driven from a low impedance, and the sense pin has to see the load; the vendor states that interchanging the reference and the positive input, or the negative input and the sense, degrades rejection. On a board those are adjacent pins, so a footprint error shows up as a part that works and measures badly rather than a part that fails. The check is a common mode sweep with the inputs tied together.
300 kHz, 10 nF of load and a slew rate of 0.1 to 0.25 V per microsecond
The small signal bandwidth of INA132U2K5 is 300 kHz at unity gain, but the large signal behaviour is set by a slew rate of 0.1 volts per microsecond on one chip origin and 0.25 on the other, with a settling time of 88 microseconds to 0.01 per cent after a 10 V step. A 300 kHz small signal figure next to a 0.1 V per microsecond slew is the classic trap: the bandwidth describes a millivolt signal, and a 10 V step takes 100 microseconds. The output is stable into 10 nanofarads, which is unusual and useful for a cable or a multiplexer input.
Output swing is to within 0.8 V of the positive rail into 100 kohm and 0.25 V of the negative rail into 10 kohm.
Short circuit current is quoted by chip origin: 6 mA source and 15 mA sink on one flow, 26 and 22 mA on the other, so a circuit that relies on a specific fault current has to know which chip it received. The quiescent current is the same 175 microamps on both. On a single 3.3 V supply the output swing leaves about 2.5 V of usable range, which is the number to check before putting a 3 V full scale signal through it.
Buying INA132U2K5
With the tube row obsolete, a prototype build either takes a reel or takes cut tape from one, and cut tape on a level 3 part needs a bake if the exposure is not recorded. That is the practical cost of the carrier change and it is worth planning for on a first article rather than discovering it the week of the build. Our BOM shortage recovery page covers the options when only a reel exists.
The grade question is separate. The UA row is active with the same 2500 piece quantity and a wider 500 µV offset window, so a shortage of the U grade can be covered by the UA grade if the offset budget allows, and the price step is small. What cannot be covered is the tube: both tube rows are obsolete, so a through hole substitute or a different package is the only answer for a line that cannot take a reel.
Questions that arrive with this part number
Whether the part is discontinued is the first question, and it is not: the reel row is active and in production, and only the tube carrier was withdrawn. Whether the UA grade can stand in is the second, and it can where the offset budget has 250 µV of room, because that is the whole difference between the bins.
The third is the load the network puts on the source. At 80 kΩ differential, a 1 kΩ source loses 1.2 per cent of gain, which is sixteen times the amplifier gain error and has to be calibrated out. The fourth is the slew rate: 0.1 V per microsecond on one chip origin means a 10 V step takes 100 microseconds, so a 300 kHz bandwidth figure should not be read as a 300 kHz large signal capability.