TPS2813DR
Product Specifications
TPS2813DR
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS2813DR 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
TPS2813DR dual noninverting high speed MOSFET driver
The TPS2813DR is the noninverting twin of the TPS2812, and it is the member most often used on a card where the controller output idles low and the gate has to follow it. Two channels, 2 A peak into a capacitive load, 25 nanoseconds maximum rise and fall at 1 nF with a 40 nanosecond propagation delay, and a 4 V to 14 V supply extended to 40 V by an internal regulator whose output can feed other circuits.
It shares a package, a pinout and a die with the inverting member, so the substitution risk runs in one direction only: putting a 2812 where a 2813 belongs inverts every gate signal on the card. On a single ended converter that produces a supply that never starts. On a half bridge it produces a shoot through event in the first switching cycle. There is no marking on the top of the package that tells the two apart at a glance.
Key parameters taken from the TPS2811 to TPS2815 document
| Channels | Two, independent |
| Logic | Dual noninverting on this member |
| Peak output current | 2 A at a 14 V supply |
| Rise and fall | 25 nanoseconds maximum into a 1 nF load at 14 V |
| Propagation delay | 40 nanoseconds maximum |
| Supply range | 4 V to 14 V direct, up to 40 V through the internal regulator |
| Supply current | Kept low by an output stage that holds shoot through current down |
| Regulator pins | REG_IN and REG_OUT, either usable to power other circuits or left open |
| Temperature | Minus 40 to plus 125 degrees ambient |
| Package | SOIC, eight pins, D suffix, and the same die in PDIP and TSSOP |
| Family fit | One inverting and one noninverting driver is the TPS2811, two input logic gates is the TPS2814 and TPS2815 |
| Order code note | D is the SOIC body, R is the reel tape |
These figures are taken from the vendor document rather than from a catalogue page. The timing numbers are maximum values at the stated load, which is the figure a timing budget should use.
Where the numbers matter in a real build
A 40 nanosecond propagation delay with a 25 nanosecond edge is fast enough that the driver, not the controller, sets the dead time on a bridge unless the controller adds it deliberately. On a card where the two channels drive the high and low side of a half bridge, the difference between the rising and falling propagation delays is the number that matters, and it is smaller than the absolute delay. A dead time budget built from the absolute figure wastes switching loss; one built from nothing at all produces shoot through.
The regulator is the second consideration. On a 24 V input the internal regulator carries the difference between 24 V and the gate drive level, and that voltage times the average gate current becomes heat inside an eight pin body. Two channels driving 2 nF at 200 kHz draw enough average current that the regulator dissipation dominates, and the package thermal resistance then sets the maximum ambient. The document gives the dissipation rating table for the three package options, and the TSSOP is the one with the smaller thermal resistance of the three.
The third point is the input threshold. The family sits at roughly two thirds and one third of the supply, which is high enough that a 3.3 V controller driving a 12 V supplied driver may not cross the positive threshold at all. A level shift, or a lower supply for the driver, is the answer the document points to.
- The logic polarity, since this part and the TPS2812 are visually identical.
- The input threshold against the controller swing, because two thirds of the supply is a high bar at a 12 V rail.
- The regulator dissipation at the switching frequency in use, which shares the package budget with the output stage.
- The package option, since the thermal resistance differs between SOIC, PDIP and TSSOP.
- The gate loop and the ground return for each channel, kept separate down to the pin.
- The date code and revision letter on a part with a long production history.
Which projects this part fits
The document presents the family as an industry standard driver replacement. The noninverting member suits controllers whose output idles low, which covers most PWM controllers and many DSP outputs. Boards in switched mode supplies, motor drives, telecom power and lamp ballasts use this part where the board already exists.
Questions that come up before a purchase order is released
Is the TPS2813 a direct replacement for the TPS2812?
Electrical and mechanical yes, functionally no. The logic is inverted between them, so a substitution changes the gate signal on every channel.
Can the two channels drive a half bridge directly?
They can drive the two gates, but the dead time has to come from the controller. The part does not generate dead time and the two channels are independent.
What happens if REG_OUT is left open?
Nothing. The regulator is bypassed when REG_IN is tied to VCC and REG_OUT is left open, and the part then runs directly from the supply within the 4 V to 14 V range.
How do I tell a 2812 from a 2813 on a finished board?
By the reel label or the purchase record. The package marking does not carry the logic variant, so a mixed bin on the line is a real risk and a functional test at the gate is the only reliable check.
How SUPERB handles this line
Parts that are one character apart and identical on the outside are the reason we keep the reel label and the date code with the shipment rather than only on the invoice. Our pin to pin cross reference covers the families with this kind of variant, how to verify a drop in replacement sets out the bench checks, and our incoming inspection plan covers the label and marking checks on arrival.