TPS546B24ARVFR
Product Specifications
TPS546B24ARVFR
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS546B24ARVFR 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
TPS546B24ARVFR: 20 A PMBus buck in a 7 by 5 mm QFN, stackable to 80 A
TPS546B24ARVFR is a 20 A synchronous step-down converter in the 40 pin LQFN-CLIP body that runs from 2.95 V to 18 V on the power stage, carries 5.5 milliohm and 1.8 milliohm integrated switches and reports output current and die temperature over PMBus in a LQFN-CLIP body of 40 pins. The TPS546B24A is the 20 A member of the TPS546x24A family, sharing the 7 mm by 5 mm by 1.5 mm QFN and the PMBus interface with the 40 A TPS546D24A. Output is 0.5 V to 5.5 V from a pin strap or 0.25 V to 5.5 V from a command, switching is selectable over twelve frequencies from 225 kHz to 1.5 MHz, and up to four devices stack for a single 80 A output. The feature list quotes 18 V on both input rails while the electrical table in the same document stops at 16 V, which is a point worth reading in full before a 17 V bus is connected.
Order code and package
| Order code | TPS546B24ARVFR |
| Power stage input voltage | 2.95 V to 18 V in the feature list, 16 V in the electrical table |
| Analogue input voltage | 2.95 V to 18 V in the feature list |
| Output voltage from pin strap | 0.5 V to 5.5 V |
| Output voltage from PMBus | 0.25 V to 5.5 V |
| Rated output current | 20 A |
| Stacking | two, three or four devices, up to 80 A shared |
| Switching frequency | 225 kHz to 1.5 MHz, twelve settings |
| Package | 40 pin LQFN-CLIP, 7 mm by 5 mm by 1.5 mm |
| Document | TPS546x24A family data sheet |
Electrical parameters that decide the application
| High side switch resistance | 5.5 milliohm |
| Low side switch resistance | 1.8 milliohm |
| Output accuracy over the junction range | better than 1 percent from minus 40 C to 150 C |
| Junction to board characterization parameter | 9.3 C per W |
| Junction to case bottom resistance | 1.0 C per W |
| Junction to top characterization parameter | 2.0 C per W |
| Analogue supply current, not switching | 12.5 mA typical at 12 V |
| Power input turn on, programmable range | 2.75 V to 15.75 V in 0.25 V steps |
| Pin strap frequency options | eight of the twelve settings |
| Reference board | eight layer, 2 oz copper per layer |
| Junction temperature range | minus 40 C to 150 C |
Points to check before this part goes on a BOM
Twenty amps through the 1.8 milliohm low side switch is 0.72 W of conduction loss, and through the 5.5 milliohm high side switch at a 10 percent duty it is a further 0.22 W. That is under a watt, which the package moves comfortably when the pad is soldered to a board with copper: the junction to board parameter of 9.3 C per W gives a 9 C rise for it, while the junction to case bottom figure of 1.0 C per W describes the die to pad path only. The difference between those two numbers is the board, and it is the only part of the thermal path that a board change can affect.
The 18 V and 16 V inputs in the same document are not a typographical slip to be argued with. The feature list on the first page quotes 2.95 V to 18 V for both rails, and the electrical characteristics table later in the same file gives 2.95 V to 16 V. A 17 V bus sits in the gap, and the right response is to specify the converter at 16 V or to ask the vendor in writing, not to read the feature line as the guarantee. This is the kind of discrepancy that a document review catches and a rushed bill of materials does not.
The relationship with the 40 A part in the same family is worth understanding before choosing. Both devices have 40 pins in the same 7 by 5 mm body, both take the same PMBus commands and both stack four deep, so the difference is the switch size, the current rating and the stacking total: 80 A here against 160 A there. A board that carries the 20 A part can move to the 40 A part without changing copper, provided the inductor, the output capacitance and the input capacitance were sized for the higher current from the start.
Current and temperature telemetry come from inside the device, which removes the sense resistor and its loss. The output voltage accuracy of better than 1 percent over the full junction range is the voltage loop specification; the current reading is a separate term with its own tolerance, and it is the one that matters when the PMBus current is used for power capping rather than for a display. A 20 A rail capped by a reading that is 3 percent low runs 0.6 A hotter than the firmware believes, which is a thermal margin question rather than an accuracy one.
Pin straps and PMBus defaults are two layers of configuration. Eight of the twelve switching frequencies are reachable from a strap resistor, the output voltage comes from a strap in 0.5 V steps or from a command in fine steps, and the MSEL pins choose which set of PMBus defaults is loaded at power up. The consequence is that a resistor change can alter the under voltage lockout and the fault response as well as the frequency, so the strap network belongs in the controlled documentation rather than in a note on the assembly drawing.
A prebiased output is supported, which matters in a sequencing scheme where a rail is pulled up through a body diode before its converter starts. The converter will not discharge the output when it begins switching, so the rail rises from wherever it was rather than being pulled to zero first. That behaviour has to match the sequencer: a controller that expects every rail to start from zero will see a shorter ramp on the prebiased one and may report a fault that never happened.
Where this order code fits
This order code belongs on intermediate rails in switches and storage, on 12 V to 1.8 V and 12 V to 1.2 V point of load stages at 15 A to 20 A, and on any rail that needs margining or voltage scaling at run time without an external trim network. It is also the natural step when a 40 A rail is over specified: the same footprint and the same interface carry the 20 A part, so a family can be standardised on one board and two current levels. 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 OEM Procurement page.
Questions buyers ask
Can this part replace the 40 A version on the same board? The package and the pin out are the same within the family, so the copper can stay. What changes is the current the rail can deliver and the loss at a given load, which is lower here. The reverse move, from 20 A to 40 A, needs the inductor, the input capacitance and the output capacitance to have been sized for 40 A, and it needs the current limit and the PMBus configuration checked. A footprint match is not a specification match.
Why does the document say 18 V on one page and 16 V on another? The feature list and the electrical characteristics table in the same file disagree, and the table is the specification. A 17 V rail therefore needs written confirmation from the vendor rather than a reading of the marketing page. This is worth flagging in a bill of materials review because a 48 V telecom bus stepped down to 17 V is exactly where the question arises.
Is the internal current reading accurate enough to protect the rail? For protection and for capping, yes, and that is what the telemetry is for. For billing or for a calibrated measurement, no: the internal sense has an offset and a gain tolerance that a shunt plus a dedicated monitor would beat. The practical approach is to use the internal reading for the control loop and for fault thresholds, and a separate monitor when the current is a recorded quantity.
How many phases does a 60 A rail need? Three devices give 60 A of headroom on a 20 A part, and the phases interleave so the input ripple appears at three times the switching frequency. Four devices give 80 A and more thermal spread, at the cost of one more inductor and one more set of decoupling. The decision is usually thermal rather than electrical: at 60 A the loss is spread over three packages instead of two, which is what keeps the junction temperature inside the range at 70 C ambient.
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 Component Lead Time Solutions page sets out the checks that keep a substitute from failing on the bench.