Contact Us

UCC27282DR

Brand: Texas Instruments
Texas Instruments UCC27282DR 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

UCC27282DR — Power Management & Signal Chain | Superb Automation

UCC27282DR

Texas Instruments — Power Management & Signal Chain

Product Overview

The UCC27282DR 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

UCC27282DR: a 120 V half-bridge gate driver with 3 A of peak output, 16 ns of typical propagation delay and a bootstrap diode on the die, in the eight pin SOIC on a 2500 piece reel

UCC27282DR is a high-side and low-side driver for two N-channel FETs in a half bridge or a synchronous buck, with independent inputs, an enable pin and an input interlock that stops both outputs being high at once in a SOIC body of 8 pins. The D body is the eight pin SOIC that the industry already routes, and the part exists because a controller with a 5 V rail cannot swing a high-side gate that sits on a 100 V switch node without a level shift and a bootstrap path of its own.

Order code and package

Catalogue stringUCC27282DR, where D is the eight pin SOIC and R is the 2500 piece large reel
BodySOIC, D, eight pins, 4.9 mm by 3.9 mm of moulding; the same die is also offered as UCC27282DRCR in a ten pin VSON for boards that need the smaller outline
Reel2500 pieces on a large tape and reel, printed against the D body on the addendum dated 12 August 2026
Tubethe same die ships 75 pieces in a tube as UCC27282D, which is the route for a first article build
Finish and moistureNIPDAU plating, moisture level 2 at 260 C with one year of floor life; the ten pin VSON rows print level 1 with no floor life limit
Temperatureminus 40 to 125 C on the reel row, with the electrical table specified to 140 C junction
MarkingU282 on the body, which is short enough that the lot code sits on the second line
DocumentSNVSAQ5B, November 2018, revised May 2022
StatusActive and in production, not a last time buy and not a not recommended for new builds part

Electrical parameters that decide the application

Bias rangethe driver runs from a VDD rail that the vendor specifies at 12 V for the electrical table, with 20 V as the absolute maximum on the supply pin and on the enable and input pins
Switch node100 V on HS as a recommended ceiling, with 120 V as the absolute maximum bootstrap voltage and minus 14 V of negative pulse tolerance on the switch node
VDD lockoutrising at 4.7 V minimum, 5.0 V typical and 5.4 V maximum, falling at 4.2 V, 4.5 V and 4.9 V, so a logic level gate can be driven from a 5 V rail with no separate bias
Bootstrap lockoutrising at 3.3 V, 3.7 V and 4.4 V, falling at 3.0 V, 3.3 V and 4.1 V, measured with respect to the switch node rather than to ground
Peak current3 A source and 3 A sink, which is the number that decides how fast a 100 nC gate can move, not the number that decides steady state loss
Propagation delay16 ns typical and 30 ns maximum from input to output in both directions, with 1 ns of typical delay matching between the two channels
Edges12 ns rise and 10 ns fall into a 1.8 nF load, which is the load that stands in for two paralleled FETs of moderate size
Minimum pulse20 ns, so a controller that produces narrow pulses on purpose at light load does not lose them in the driver
Bootstrap diodeon the die, with 0.55 V to 0.85 V of forward drop at 100 uA and 0.88 V to 1.0 V at 80 mA, and 1.5 to 2.5 Ohm of dynamic resistance between those points
Quiescent current0.3 mA typical and 0.4 mA maximum on VDD, 2.2 to 4.5 mA on the bootstrap rail, and 0.2 to 0.4 mA from the bootstrap rail to the switch node at 100 V of blocking
Disabled currentthe enable pin pulls the part down to a few microamps, which is the reason an enable pin exists on a driver that already stops when its input stops
Thermal118.3 C per W junction to ambient in the eight pin SOIC, and 53.6 C per W in the ten pin VSON with its exposed pad

Family options and how they differ

The input interlockthe two inputs cannot both turn their output on, so a controller that asserts both during a fault leaves the bridge in a defined state instead of crowbarring the rail
The enable pina separate logic input that disables both outputs and drops the bias current, used where a product has to hold the bridge off during a self test
Independent inputsthe inputs are referenced to the logic rail rather than to VDD, so a 3.3 V controller can drive a driver that runs from 12 V without a level shifter
The negative ratingsminus 5 V on the inputs and minus 14 V on the switch node for pulses under 100 ns are what keep a half bridge alive when the parasitic inductance of the source connection pushes the switch node below ground at turn off
The ten pin VSON optionUCC27282DRCR carries the same die with 53.6 C per W instead of 118.3, which matters when the driver sits next to the FETs inside a sealed enclosure
Against a controller with an integrated driverthe integrated part usually offers a fraction of 3 A and a slower edge, which is fine at 20 kHz and not fine at 300 kHz

Points to check before this part goes on a BOM

The bootstrap diode on the die removes a component and adds a constraint. A discrete bootstrap diode has to be fast, has to hold the full switch node voltage, and has to survive the reverse recovery that comes with charging the bootstrap capacitor from a 12 V rail while the switch node swings 100 V in 10 ns. Integrating it fixes all three and leaves one limit in its place: the diode drop rises with current, from about 0.6 V at 100 uA to about 0.95 V at 80 mA, so the bootstrap capacitor charges to VDD minus that drop and the high-side gate never quite reaches the low-side gate voltage. At 12 V of bias the loss is under one per cent, and at 5 V of bias it is nearly twenty per cent of the gate drive, which is the reason a 5 V rail suits a logic level FET and not a standard one.

The 100 V switch node rating is the number that decides whether the part fits a 48 V bus and not whether it fits a 400 V bus. A 48 V rail with a 20 V margin of overshoot sits comfortably inside 100 V. A 60 V rail with the same margin does too. A 90 V rail with a snubber that lets the node ring to 110 V does not, and the failure is a driver that works on the bench and dies on the first hot plug event. The 120 V figure on the bootstrap pin is an absolute maximum for the HB to HS differential, not a switch node rating, and reading the two as one number is a common mistake on a bill of materials review.

Sixteen nanoseconds of propagation delay and one nanosecond of matching are what make a synchronous buck at high frequency possible. The dead time has to cover the driver delay plus the FET turn off time, and every nanosecond of that dead time is a nanosecond of body diode conduction. At 500 kHz a 20 ns dead time is one per cent of the period, and the body diode loss that comes with it is a real fraction of the total. The matching figure matters more than the absolute delay, because a mismatch between the high-side and low-side paths is what produces shoot through on one edge and excess body diode time on the other.

The quiescent current on the bootstrap rail is larger than most people expect. 2.2 mA typical and 4.5 mA maximum of HB current, plus 0.2 to 0.4 mA from HB to the switch node at 100 V of blocking, means the bootstrap capacitor has to be sized for a load that never stops. A 100 nF capacitor with 4.9 mA of drain loses 0.49 V per microsecond, so at a 100 kHz switching frequency with 50 per cent duty the charge delivered in one on time is 25 uS times 4.9 mA, which is far more than the capacitor holds. The arithmetic is the reason the bootstrap capacitor on these parts is usually 100 nF to 1 uF and the reason a bootstrap resistor is often fitted to limit the recharge current rather than to slow anything down.

The eight pin SOIC at 118.3 C per W is the thermal limit that decides whether the driver can be run at full current continuously. Driving a 100 nC gate at 15 V and 200 kHz is 300 mW of gate charge, and with the quiescent current the total is around 350 mW. At 118.3 C per W that is 41 C of rise inside the package, which is comfortable at 25 C ambient and marginal at 85 C ambient inside a sealed box. The ten pin VSON option at 53.6 C per W exists for exactly that case, and a board that changes package has to change the copper under the part as well, because the thermal number assumes the exposed pad is soldered to a plane.

What the negative ratings are really for

Minus 14 V of pulse tolerance on the switch node and minus 5 V on the inputs are not marketing numbers. Every half bridge with a source connection that carries di/dt produces a negative excursion at the switch node when the low-side FET turns off, because the current that was flowing in the source inductance has to keep flowing somewhere. A 20 nH source connection and 20 A of current switched in 20 ns produce 20 V of inductive kick, and the driver sees it on HS. A part rated to minus 5 V will latch or destroy itself on the first such event, and the repair is a driver rated to minus 14 V rather than a bigger snubber.

The inputs carry a similar but smaller exposure. A gate drive transformer or a long gate trace couples the switch node event back into the controller output, and the input pin sees a negative spike. The minus 5 V rating means the part survives a few volts of that without a clamp diode, which is what keeps a two layer control board working in a product that was not laid out by a specialist.

Reading a 4.7 V lockout against a 5 V rail

The VDD lockout rises at 5.0 V typical and 5.4 V maximum, and a 5 V rail with five per cent tolerance sits at 5.25 V at the top of its window. That is inside the lockout maximum, which means some parts will not start on a rail that is nominally fine. The vendor expects the rail to be a regulated 5 V with the tolerance stated, and a board that feeds the driver from a 5 V logic rail through a series resistor, which is common, will see the rail drop below 5.4 V under load and the driver will sit in lockout while the rest of the board runs.

The bootstrap lockout at 3.7 V typical is the second half of the same problem. The bootstrap capacitor charges to VDD minus a diode drop, so 5 V of bias gives about 4.3 V of bootstrap rail and 12 V gives about 11.3 V. At 5 V of bias the margin above the 4.4 V maximum lockout is 0.1 V, and the first cycle after a long off period is the one that fails, because the capacitor has had time to leak down. The fix is a bootstrap capacitor large enough to hold the charge and a bias rail that stays above 5.2 V.

Where this order code fits

This is the driver stage of a synchronous buck, a half bridge in a motor drive or a Class D amplifier, an active clamp forward converter, or a 48 V to 12 V converter in a telecom or server supply 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 Component Lead Time Solutions page.

Questions buyers ask

Can it be driven from a 3.3 V controller? Yes. The inputs are referenced to the logic rail and rated independently of VDD, so a 3.3 V signal drives a driver whose bias is 12 V.

Does it need an external bootstrap diode? No. The diode is on the die, with about 0.85 V of drop at 100 uA and 1.0 V at 80 mA, so the bootstrap capacitor charges to VDD minus that drop.

What happens if both inputs go high? The interlock holds both outputs off. That is deliberate, and it is the difference between a defined bridge state and a shoot through during a fault.

How much gate charge can it drive? 3 A of peak current, with 12 ns rise and 10 ns fall into 1.8 nF. Two large FETs in parallel need a boost stage rather than a bigger driver.

Is it the same part in the ten pin package? The same die with the same ratings, in a VSON body with an exposed pad, and the thermal resistance falls from 118.3 to 53.6 C per W.

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 How to Verify a Drop in Replacement page sets out the checks that keep a substitute from failing on the bench.