TPS259472ARPWR
Product Specifications
TPS259472ARPWR
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS259472ARPWR 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
TPS259472ARPWR: 2.7 V to 23 V, 5.5 A eFuse, 28 milliohm, overvoltage clamp, auto retry
TPS259472ARPWR is an electronic fuse that puts two back to back FETs in the input path of a board, limits inrush with a capacitor, limits current with a resistor and clamps or cuts an overvoltage before it reaches the load in a VQFN-HR body of 10 pins. The two FETs measure 28 milliohm typical together, so at 5 A the drop is about 140 mV and the loss about 0.7 W. The overvoltage clamp threshold is chosen on the OVCSEL pin from 3.8 V, 5.7 V and 13.8 V, and the clamp responds in about 5 microseconds by regulating the FET linearly rather than switching it off. Current limit is set by a resistor on the ILM pin across 0.5 A to 6 A, with a fast trip at twice that value.
Order code and package
| TPS259472ARPWR | Active, VQFN-HR (RPW) body, 10 pins, 3000 pieces on a 180 mm reel, marking 2ABH |
| TPS259472ARPWR.A | Active, same body and same 3000 piece reel, material row at MSL Level-2-260C-1YEAR |
| TPS259472LRPWR | the latch off variant, same body, holds the output off after a fault until the part is re-enabled |
| TPS259470ARPWR | adjustable overvoltage lockout instead of a clamped threshold, same body |
| Temperature range | minus 40 to 125 degrees junction, 28 V absolute maximum input, negative input withstand to minus 15 V |
Electrical parameters that decide the application
| Input voltage | 2.7 V to 23 V operating, 28 V absolute maximum |
| Continuous current | 5.5 A |
| On resistance | 28.3 milliohm typical for the back to back pair |
| Current limit | 0.5 A to 6 A set by the resistor on the ILM pin, plus or minus 10 per cent above 1 A |
| Fast trip | twice the set current limit, 1 microsecond settling, 12 microseconds to fully interrupt |
| Overvoltage clamp | 3.8 V, 5.7 V or 13.8 V chosen on the OVCSEL pin, about 5 microseconds of response |
| Overvoltage lockout | 1.2 microseconds of response on the adjustable variants |
| Slew rate | 28.1 V per ms with the dVdt pin open, reduced by a capacitor to ground |
Family variants and how they differ
| TPS259472A | clamped overvoltage with automatic retry after a fault, the string on this page |
| TPS259472L | the same clamp with latch off behaviour |
| TPS259470A | adjustable overvoltage lockout, auto retry, the choice for a power MUX |
| TPS259474A | circuit breaker behaviour with a 110 ms interrupt delay instead of active current limiting |
| Status | the vendor states the clamped overvoltage variants are not recommended for new ORing or power MUX use |
Points to check before this part goes on a BOM
This variant clamps instead of switching off, and that changes the thermal story. During a clamp the FET is held in its linear region and the excess voltage appears across it. A 19 V surge clamped to 13.8 V at 1 A dissipates 5 W inside a ten pin body for as long as the surge lasts. That is the intended behaviour and it is why the clamp is specified with a response time and a load derating curve rather than a simple threshold.
The vendor does not recommend the clamped variants for ORing. That statement is in the data sheet for this family, and it is a status note rather than a limitation of the silicon. A board that ORs two supplies needs the adjustable lockout variants and their behaviour during a fast role swap. Quoting a clamped part into an ORing socket is the kind of substitution that passes a bench test and fails in the field.
Inrush is set by a capacitor, and the energy is the real limit. With the dVdt pin open the output rises at 28.1 V per ms, which is fast enough that a 470 microfarad load looks like a short for a few hundred microseconds. Adding capacitance slows the ramp and cuts the peak current, but the part then spends longer in its linear region. The worked example in the data sheet charges 470 microfarad to 12 V with an inrush of 282 mA and 1.69 W of dissipation, and the thermal plot for that case is the one to check against a hot enclosure.
A blanking timer separates a transient from a fault. A load step that exceeds the current limit for a short time is allowed through, because the timer capacitor has to discharge before the limit engages. That prevents a converter start or a disk spin up from tripping the fuse, and it means the response to a real overload is delayed by the timer value. Setting the timer is a compromise between nuisance trips and protection speed.
Reverse current is blocked at all times, not only when the part is off. The back to back arrangement means the body diodes do not conduct in either direction, so a load held above the input cannot feed the input rail. That is what makes the part usable on a board with a battery and an adapter on the same node, and it is the difference between an eFuse and a load switch with a single FET.
The current limit accuracy is quoted above 1 A for a reason. At the low end of the range the offset of the sense circuit dominates and the tolerance is wider than ten per cent. A board that needs a 0.6 A limit should budget more margin than one that needs 3 A, and the resistor value table in the data sheet is the place to read the intended point rather than an interpolation.
A negative input transient is part of the specification. The part withstands minus 15 V on the input, which covers a reversed adapter or an inductive kick from a long cable. That is a claim most load switches cannot make, and it is worth stating on an offer to a customer who has field returns from hot plug events on a barrel jack.
Auto retry versus latch off is a system decision, not a part decision. The A variant restarts after a fault on its own, which suits an unattended box that has to recover. The L variant stays off until the host re-enables it, which suits a system that has to log the fault and hold the load down. The two are the same silicon with different behaviour after a fault, and mixing them in a bill of materials produces field behaviour that nobody can reproduce on a bench.
The ten pin body has no thermal pad to spare. At 5 A the pair dissipates about 0.7 W, and during any fault or clamp event far more. The exposed pad is the only path, and the vias under it are part of the thermal solution. A board that connects the pad to a small island will reach thermal shutdown at a current the part is rated for, and the shutdown is a retry loop rather than a clean latch.
IEC 62368-1 certification is stated on the front page. That matters for a product going through safety approval, because the eFuse can be part of the argument instead of an additional discrete network. The certificate applies to the part in its specified configuration, so the surrounding components, especially the resistor on the ILM pin, have to match what the certificate assumes.
What to confirm before this part goes on a purchase order
State the input bus, the load current and the load capacitance together. The dVdt capacitor and the current limit resistor are both chosen from those three numbers, and a part quoted without them cannot be checked for inrush energy.
Say whether the board ORs two supplies. If it does, this clamped variant is the wrong one in the words of the vendor, and the adjustable lockout variant is the part to quote.
Where this order code fits
It sits on the input of a server card, an add in card or a USB PD port where a hot plug event, a surge or a reversed adapter has to be contained before it reaches the DC/DC stages behind it. Teams comparing it with a neighbouring part start from our Power Management Cross Reference page, and teams replacing a part that has gone end of life start from the Pin to Pin Cross Reference page.
Questions buyers ask
Can it replace a fuse in a 12 V, 5 A rail? It can, and it does more than a fuse because it limits inrush and clamps overvoltage. The differences are that it needs a current limit resistor and a dVdt capacitor, that it has a quiescent current, and that it can retry rather than stay open. A safety argument that relies on a fuse breaking needs to be rewritten around the eFuse behaviour.
What happens during a short on the output? The current rises to twice the set limit and the fast trip comparator interrupts within about 12 microseconds, faster than the blanking timer would allow. The part then retries on this variant, so the output pulses until the short is removed. A latch off variant stays down until the host re-enables it.
Does the clamp protect a 5 V load from a 24 V adapter? Only within the energy the part can absorb. The clamp regulates the output to the selected threshold and the FET dissipates the difference, so a sustained 24 V input on a 5.7 V clamp with any load current will drive the junction into thermal shutdown. It is transient protection, not a regulator.
SUPERB holds and sources this line for production and repair demand, quoted against the exact ordering string with date codes, packing and country of origin stated on the offer. When a part is moving, the BOM Shortage Recovery page sets out the checks that keep a substitute from failing on the bench.