TPS3840PL25DBVR
Product Specifications
TPS3840PL25DBVR
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS3840PL25DBVR 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
TPS3840PL25DBVR 2.5 V supervisor for a 3.3 V rail with a push-pull output
The TPS3840PL25DBVR watches a rail against a fixed 2.5 V falling threshold and drives a push-pull active-low reset. It is the member of the TPS3840 family that sits closest to a 3.3 V supply, because 2.5 V leaves 0.8 V of margin, which is enough room for a 5 % rail tolerance, the 1.5 % threshold window and the drop across a series diode all at once.
Everything around the comparator is built for a product that sleeps. The supply current is 300 nA typical, the start-up delay is 220 microseconds, and the reset delay is set by one capacitor between 50 microseconds and 6.2 s. The body is the 5 pin SOT-23, 2.90 mm by 1.60 mm, with the same pinout as the 2.0 V and 4.3 V members of the family.
Key parameters taken from the SNVSB03E document
| Supply range | 1.5 V to 10 V on VDD, with the inputs allowed to 0.3 V above VDD |
| Falling threshold | 2.5 V nominal, 1 % typical and 1.5 % maximum over the full temperature range |
| Worst case trip | 2.4625 V at the low end of the 1.5 % window, which is the number to use against a 3.0 V minimum rail |
| Hysteresis | 100 mV typical, so the release point is 2.6 V and a rail that recovers to 2.55 V stays in reset |
| Supply current | 300 nA typical and 700 nA maximum over temperature |
| Output drive | 0.8 times VDD minimum high level at 2 mA source, 200 mV maximum low level at 2 mA sink |
| Start-up delay | 220 microseconds typical with CT open, and a capacitor on CT adds to this first event |
| Reset delay | 50 microseconds with CT open, 6.2 ms at 10 nF, 619 ms at 1 microfarad, 6.2 s at 10 microfarad |
| Timing accuracy | The delay is set by a 350 to 650 kOhm internal resistor, so the part to part spread is wider than the capacitor tolerance |
| Propagation | 15 microseconds typical and 30 microseconds maximum from threshold crossing to output |
| Manual reset | MR pin, 600 mV maximum low, 0.7 times VDD minimum high, 100 kOhm pull-up |
| Power-on reset | Defined output from 300 mV of VDD on the push-pull member |
| Temperature | Minus 40 to plus 125 degrees, junction |
Reading the threshold against a 3.3 V rail
A 3.3 V rail at minus 5 % is 3.135 V and at minus 10 % is 2.97 V. A 2.5 V threshold sits well below both, so the supervisor does not trip on a normal load step. That is the point of this member: it is a fault detector and not a precision monitor. Where the trip has to land inside a narrow band around 3.0 V, the family has 2.9 V, 3.0 V and 3.1 V options and the 2.5 V part is the wrong choice.
The 100 mV of hysteresis is the number that decides whether the release is clean. After the rail falls below 2.5 V the output stays asserted until the rail passes 2.6 V, and then the delay starts. On a supply that recovers slowly, the reset therefore ends later than the trip point suggests, which is usually what a customer wants and occasionally what they complain about when the boot time is measured from the fault rather than from the recovery.
The 1.5 V floor of the operating range is below the trip point on purpose, so the output stays valid while the rail collapses. The exception is the first 300 microamperes of a slow ramp, below which the output is undefined, and a pull-down on the enable line is the usual answer.
Choosing the CT capacitor for a 3.3 V processor
Most processors need the reset held for at least a few milliseconds after the rail is valid, and a few need tens of milliseconds before the clock is stable. The published points cover both ends without calculation.
| No capacitor | 50 microseconds, enough when the reset only has to reach a logic gate |
| 1 nF to 10 nF | Roughly 0.6 ms to 6.2 ms, the range used for a microcontroller with an internal oscillator |
| 100 nF | Around 60 ms by the published scaling, a common figure where a crystal has to start |
| 1 microfarad | 619 ms, the setting used when a flash write or a safe state has to complete before reboot |
| 10 microfarad | 6.2 s, the top of the characterised range and only sensible with a low leakage ceramic part |
The delay is measured from the moment the rail passes 2.6 V, so a supply that ramps through the threshold slowly adds its own rise time to the total. On a 3.3 V rail with a 5 ms ramp, a 10 nF capacitor gives a reset that ends about 11 ms after the rail started moving.
Where the PL25 belongs
- 3.3 V processor boards that need a fault detector with a wide margin and a defined release point
- Smart meters, circuit breakers and protection relays, where the supervisor has to run from the same rail it watches
- Field transmitters and PLC cards with a 3.3 V rail and a battery backup that takes over below 2.5 V
- Wireless modules that must be held in reset until the reference and the crystal have settled
- Boards where the MR pin doubles as a service button, with no external pull-up and no debounce circuit
A second use that comes up in enquiries is supervising a 2.5 V rail directly. The part works there too, but the 100 mV hysteresis is then 4 % of the rail and the release point sits at 2.6 V, above nominal, which means the rail has to overshoot before the reset is released. That is usually a sign the 2.4 V or 2.3 V member is the better choice.
Order code, marking and reel
| TPS3840PL25DBVR | Active, production, SOT-23 DBV 5 pin, 3000 pieces on a large reel |
| Lead finish | NIPDAU, moisture level 1 at 260 degrees, unlimited floor life |
| Part marking | PL20 on the body, one code below the threshold option, so the marking alone does not confirm the 2.5 V trip |
| Siblings | TPS3840DL25DBVR open-drain and TPS3840PH25 active-high, same body and pinout |
| Other thresholds | Every 0.1 V step from 1.6 V to 4.9 V exists in the family table, with minimum order quantities for the less common ones |
Sourcing notes for this line
The 2.5 V option is one of the two or three threshold values that move first when a programme ramps, and it is also the one that is easiest to substitute wrongly, because a 2.4 V part looks identical on a reel and moves the release point by 100 mV. The power management cross reference lists the equivalents by pinout and states where the threshold and the hysteresis differ.
Where a build needs the part before the vendor lead time clears, the practical routes are a distributor stock check by marking code and a documented alternate from a second source. Both are on the component lead time page, and both need a trip point measurement on the first article rather than a visual check.
For programmes that run for years, the same family with a 2.0 V or 2.9 V threshold is worth keeping on the approved list, because the pinout is shared and only the divider or the code threshold changes. That consolidation is what the BOM consolidation page sets out in more detail.
Questions buyers ask about this part
Does the output need a pull-up? No on the PL member. It sources current, so a pull-up only adds load and leakage. The DL member is the one that needs the resistor.
Can it drive a relay or a load switch gate directly? At 2 mA of source current and 0.8 times VDD, it drives a logic level gate. A power MOSFET gate with more than a few nanocoulombs of charge needs a buffer, because the part is not specified for a capacitive load above 10 pF in the timing tables.
What is the current draw with the output asserted? The 300 nA figure is the quiescent current with the output released. With the output low and sinking current, the supply current rises by what the load draws through the pin.
Is the delay repeatable unit to unit? Within the 350 to 650 kOhm window, no. A 10 nF capacitor gives a delay that varies by close to a factor of two across the window, so a specification that needs a tight delay needs a larger capacitor and a wider tolerance allowance.