INA253A2IPWR
Product Specifications
INA253A2IPWR
Texas Instruments — Power Management & Signal Chain
Product Overview
The INA253A2IPWR 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
INA253A2IPWR: 2 milliohm shunt inside the package, 200 mV per amp
The INA253A2IPWR is a bidirectional current sense amplifier with the shunt resistor built into the package: 2 milliohm, 0.1 percent tolerance, 3 nH of inductance and 10 ppm per degree of temperature coefficient between 0 and 125 C. The A2 gain is 200 mV per amp, so 10 A produces 2 V at the output and the 3 dB bandwidth is 350 kHz across all gains. Offset is specified as a current, 15 mA maximum, which is the honest way to state it when the shunt is fixed and every error is already referred to amps.
Integrating the shunt removes the two errors that dominate a discrete current sense circuit: the tolerance of the resistor and the error in the Kelvin connection to it. A 1 percent shunt with a 4 wire connection on a 2 oz board can easily be 2 percent off once the copper's own temperature coefficient and the tap placement are counted. A 0.1 percent shunt inside the package, trimmed and measured with the amplifier, removes both from the budget at the cost of losing the ability to choose the shunt value.
Key data for the INA253A2IPWR
| Shunt | 2 milliohm integrated, 0.1 percent maximum tolerance, 3 nH inductance, 10 ppm per degree from 0 C to 125 C |
| Gain | 200 mV per amp on the A2 row; the family offers 100 and 400 mV per amp in the same body |
| Current | 15 A continuous from minus 40 C to 85 C with 1 oz input planes, derated to 10 A at 125 C ambient |
| Accuracy | Gain error 0.4 percent maximum, gain drift 45 ppm per degree, offset current 15 mA maximum, offset drift 125 uA per degree |
| Rejection | Better than 120 dB DC common mode rejection, 90 dB at 50 kHz, with PWM rejection circuitry fitted |
| Supply | 2.7 V to 5.5 V, 3 mA maximum, common mode 4 V to 80 V, minus 40 C to 125 C |
Body, marking and reel data
| Body | PW, 20 pin TSSOP carrying the shunt and the amplifier, with the current path through the wide pins |
| Reel | 2000 pieces on a large reel for the R suffix |
| Finish | NIPDAU at moisture level 2 with a one year floor life, bake required after floor time |
| Package resistance | 4.5 milliohm total through the package, of which 2 milliohm is the shunt and the rest is lead and bond path |
| Thermal limit | The 15 A figure assumes no airflow, 85 C ambient and 1 oz copper planes on the input pins, and it falls away above that |
Where an integrated shunt changes the board
Motor drives and solenoid control are the listed uses, and both have the same problem: the shunt has to sit in the current path and the current path wants to be short and wide, while the measurement wants a clean differential tap. Putting both inside a TSSOP lets the vendor do the tapping, and the placement question becomes a thermal one instead of a signal integrity one.
Three consequences follow. The first is that the current range is fixed by the part number, so a board that later needs a different range needs a different gain row rather than a different resistor. The second is dissipation: at 15 A through 4.5 milliohm the package dissipates about 1 W, which is why the 15 A rating is stated at 85 C with 1 oz planes and falls to 10 A at 125 C. The third is that the package inductance, 3 nH of shunt plus the lead frame, still limits the edge rate the part can follow, so a 350 kHz bandwidth is what the whole current path supports, not just the amplifier. Where a board is comparing an integrated shunt against a discrete one, the amplifier cross reference lists both approaches by gain and bandwidth, and our medical device components note covers the traceability that a clinical build needs on the shunt tolerance.
Buyer questions on the INA253A2IPWR
Does 15 A mean 15 A continuous at any temperature? No. The figure holds to 85 C ambient with 1 oz copper, and the vendor derates it to about 10 A at 125 C.
Why is the offset given in amps? Because the shunt is fixed, an input offset voltage would be an abstract number. Fifteen milliamps is the error a customer actually sees at the low end of the range.
Can two of these be placed in parallel for a higher current? Not sensibly. The shunts would share current by their own tolerance and the two outputs would not add in a way the host can calibrate, so a higher current range is a different part number rather than a parallel pair.
Availability and ordering notes
INA253A2IPWR is active at 2000 pieces on a large reel with NIPDAU at moisture level 2. Two rows in this addendum are printed differently and worth reading before a purchase order. INA253A1IPW, the tube row of the A1 gain, is Obsolete while the reel row is active. INA253A3IPWR, the plain row of the A3 gain, is also printed Obsolete while its G4 and .A rows are active. In both cases the gain is still available, but the plain string a legacy bill of materials calls out may no longer be published. Because the gain is a resistor inside the package, a substitution between gains is a real change in the current range and not a calibration tweak, which is why we confirm the gain row in writing before scheduling.