Contact Us

TPS62867RQYR

Brand: Texas Instruments
Texas Instruments TPS62867RQYR is a power management component designed for AI Servers & Low-Altitude Economy (Drone and UAV) and related applications. In stock at Superb Automation — request a quote for competitive pricing and fast delivery.
quote now

Product Specifications

TPS62867RQYR — Power Management & Signal Chain | Superb Automation

TPS62867RQYR

Texas Instruments — Power Management & Signal Chain

Product Overview

The TPS62867RQYR 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 VoltageV
Output VoltageV
Output CurrentA
Switching FrequencykHz
Efficiency%
Quiescent CurrentµA
Topology—
Package—
Operating Temp°C
Protection—

Applications

Industrial Automation Automotive Electronics Telecom Infrastructure Server & Data Center Power Medical Devices Consumer Electronics

Request Quote — In Stock, 100% New & Original, Full Warranty

TPS62867RQYR: 6 A synchronous buck converter with 4 uA quiescent current in a 1.5 by 2.5 mm QFN

TPS62867RQYR is a 6 A step-down converter in the nine pin RQY body that runs from 2.4 V to 5.5 V, switches at 2.4 MHz and uses DCS-Control with 11 milliohm and 10.5 milliohm integrated switches in a RQY QFN body of 9 pins. The TPS62867 is the 6 A member of the TPS6286x pair, alongside the 4 A TPS62865. Input is 2.4 V to 5.5 V, output spans 0.6 V to the input rail, and the reference is 600 mV. Order codes come in a fixed output version whose level is chosen by a resistor from the VSET/MODE pin to ground and in an adjustable version set by a divider on FB. The converter is a current mode DCS-Control regulator with power save and forced PWM modes, a power good window comparator and hiccup short circuit protection.

Order code and package

Order codeTPS62867RQYR
Input voltage2.4 V to 5.5 V
Output voltage, adjustable0.6 V to VIN
Output voltage, fixed versionset by a resistor on VSET/MODE
Rated output current6 A
Reference voltage600 mV
Switching frequency2.4 MHz
Packagenine pin RQY QFN, 1.5 mm by 2.5 mm
ControlDCS-Control, power save or forced PWM
DocumentSLUSDN8, March 2021

Electrical parameters that decide the application

Quiescent current, not switching4 uA typical, 10 uA maximum
Shutdown current, EN low0.24 uA typical, 1 uA maximum
Input UVLO, rising2.2 V to 2.4 V, 2.3 V typical
Input UVLO, falling2.1 V to 2.3 V, 2.2 V typical
High side switch resistance11 milliohm
Low side switch resistance10.5 milliohm
High side current limit, TPS628677 A minimum, 7.7 A typical, 8.5 A maximum
Low side current limit, TPS628676.5 A typical
Negative current limitminus 3 A
Feedback voltage594 mV to 606 mV, 600 mV typical
Output accuracy, 0 C to 85 Cplus or minus 1 percent
Output accuracy, full rangeplus or minus 2 percent
Power good, lower window85 to 96 percent of nominal
Power good, upper window103 to 120 percent of nominal
Output ramp time0.8 ms to 1.5 ms, 1 ms typical
Enable delay with a 249 kohm VSET resistor420 us to 1100 us
Output discharge resistance3.5 ohm
Thermal resistance, junction to ambient90.9 C per W on a JEDEC board, 60.3 C per W on the EVM
Logic high threshold at EN and VSET/MODE0.84 V
Logic low threshold at EN and VSET/MODE0.4 V

Points to check before this part goes on a BOM

Six amps in a 1.5 by 2.5 mm body makes the copper the heat sink. Conduction loss at full load is 36 times 11 milliohm in the high side switch and 36 times 10.5 milliohm in the low side, which is 0.40 W and 0.38 W before the duty cycle is applied, so a 5 V to 1.8 V rail carries about 0.39 W of switch loss. On the 90.9 C per W JEDEC figure that is a 35 C rise, and on the 60.3 C per W figure measured on the EVM it is 23 C. The gap between the two numbers is the whole thermal argument: the package alone cannot move the heat, the pad and the two copper planes under it do.

The current limit leaves less headroom than the rating suggests. The high side limit is 7 A minimum and 8.5 A maximum on the 6 A part, and the low side valley limit is 6.5 A typical, so a load that sits at 6 A runs within 500 mA of the valley threshold. A converter feeding a motor or a capacitive inrush will reach that threshold on a transient long before it reaches it on a steady load, and the hiccup protection then turns the rail off rather than folding back, which is a behaviour to know before a restart counter is written into firmware.

The fixed output version encodes the rail in a resistor from VSET/MODE to ground, which means the rail is a component value and not a software setting. The enable delay specification is given with a 249 kohm resistor in that position, at 420 us to 1100 us, so the same pin carries the code and the timing. A 1 percent E96 part is the right choice, and a node with flux residue or a contaminated stencil opening shifts the code, which is why boards that use the resistor code should check the rail at first power up rather than assume it.

The negative current limit of minus 3 A matters in two situations that are easy to overlook. In forced PWM mode the low side switch sinks current when the load is removed, and in a stack of rails the same behaviour appears when one rail comes up first. A pre biased output, where the rail is already charged through a body diode from a neighbouring supply, is pulled down instead of being left alone, so a sequencer that depends on the second rail staying at zero until its turn will see it move.

Four microamps of quiescent current is 35 mAh over a year, which is the whole standby budget of a coin cell product and about a tenth of it on a 400 mAh lithium cell. The number belongs to the non switching condition with the enable pin high, so it describes the state a battery product sits in for most of its life. In power save mode the converter skips pulses to hold that current, and the output ripple rises to whatever the comparator window allows, which is acceptable on a sensor rail and not acceptable on a reference or an analogue front end.

The 2.4 MHz switching frequency sets both the inductor size and the minimum on time. With the period at 417 ns, a 5 V to 0.9 V conversion runs at 18 percent duty, so the on time is 75 ns and the converter stays in continuous conduction. At the other end, a 2.5 V input to a 2.3 V output is 92 percent duty, and the 100 percent duty mode takes over before dropout, which is what keeps a 3.3 V rail alive on a cell that has fallen to 3.0 V.

The power good comparator watches a window rather than a single threshold: the lower edge releases between 85 and 96 percent of nominal and the upper edge trips between 103 and 120 percent. At a 1.8 V rail that is a release band of 198 mV, wide enough that a board which gates a reset on power good sees it assert somewhere inside that band. With a 34 us deglitch the comparator ignores the transient of a load step, so a rail that dips under the window for less than that does not drop the flag.

Two order codes sit close enough to confuse. The 4 A TPS62865 shares the same document and the same nine pin body, and its high side limit is 5 A to 6 A against the 7 A to 8.5 A of the TPS62867. A bill of materials copied from the 4 A part and re-suffixed to the 6 A part keeps the same footprint, so the swap looks free on the drawing, but the inductor and the output capacitance were sized for 5 A of ripple current and the 6 A part will push more through them. The package marking and the reel label are the only places where the difference shows.

Thermal shutdown trips at 150 C with 20 C of hysteresis, which is above the 125 C junction limit of most industrial specifications. In practice the rail decides its own current long before that: a 1 W dissipation in the RQY body is already a 60 C rise on the EVM figure and 91 C on the JEDEC one, so an ambient of 70 C leaves no room at all. The honest way to size this part is to compute conduction loss at the real duty cycle, add the switching term, and check the result against the plane area the board actually has.

Where this order code fits

This order code belongs on point of load rails in handheld instruments, battery powered data loggers, field radios and portable medical monitors, where a 3.7 V cell or a 5 V bus has to become 1.8 V, 1.2 V or 0.9 V at several amps inside a square of about four millimetres. It is also the part chosen when a rail has to stay regulated while the cell falls to 2.5 V, because the 100 percent duty mode passes the input through instead of dropping out. 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 BOM Shortage Recovery page.

Questions buyers ask

Does the 6 A rating hold at a 2.5 V input? At a 2.5 V input the duty cycle is high and the high side switch conducts for most of the period, so the 11 milliohm resistance dominates and the conduction loss is close to 36 times 11 milliohm, about 0.40 W. That is a 36 C rise on the EVM figure, which a small board can carry only if the pad has vias into at least two inner layers. The current limit does not fall with input voltage, so the electrical side is unchanged and the thermal side is what limits the rail.

What does the resistor on VSET/MODE actually do? On the fixed output version it selects the output voltage from a table, and the same pin sets the enable delay, which the data sheet specifies with a 249 kohm resistor fitted. The resistor therefore has to be 1 percent and the node has to be clean. On the adjustable version the pin is a mode select and the rail comes from a divider on FB, so the two versions are not interchangeable on one footprint without a change of resistor network.

Is 4 uA enough for a coin cell product? It is 35 mAh over a year, against roughly 225 mAh for a CR2032, so the converter alone would take about six years to flatten the cell. The load and the leakage of the rest of the board are almost always larger terms. The number describes the non switching state, so it applies to a product that sleeps and wakes rather than one that regulates continuously.

How much margin is there above 6 A? The high side limit starts at 7 A, a 17 percent margin, and reaches 8.5 A at the top of the window, while the low side valley limit is 6.5 A typical. The narrow term is the valley limit, because it acts on the bottom of the ripple. A load that steps from 4 A to 7 A will hit the valley limit within a few cycles and the hiccup protection will respond, so a rail with a 6 A steady load needs its transient measured rather than reasoned about.

SUPERB holds and sources this converter line against the exact order code, with the reel, date code and country of origin stated on the offer. When a rail has to move to a second part, the Component Lead Time Solutions page sets out the checks that separate a working substitute from one that fails on the bench.