TPS3899DL30DSER
Product Specifications
TPS3899DL30DSER
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS3899DL30DSER 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
TPS3899DL30DSER nano power 3.0 V supervisor with an open-drain reset
The TPS3899DL30DSER watches a rail against a fixed 3.00 V falling threshold and pulls an open-drain reset line low when the rail drops below it. The supply current is 125 nA typical, the body is a 6 pin WSON of 1.50 mm by 1.50 mm, and the sense input is separate from the supply, so the part can monitor one rail while running from another.
The DL field is the open-drain output and the 30 field is the 3.00 V option. The same silicon in the PL field is push-pull active-low and in the PH field is push-pull active-high, which means three parts share one land pattern and differ only in how the reset line is driven. This member is the one to pick when several supervisors have to share one reset net.
Key parameters taken from the SLVSFM0C document
| Supply range | 0.85 V to 6 V on VDD for the open-drain member, so a rail can be watched from a 1 V housekeeping supply |
| Falling threshold | 3.00 V nominal, one of the fixed steps from 0.8 V to 5.4 V in 100 mV increments |
| Threshold accuracy | 0.5 % typical and 2 % maximum for fixed thresholds from 1.8 V to 5.4 V, and 2.5 % maximum below that |
| Hysteresis | 5 % typical of the trip point, printed as a 3 to 8 % window, so the release point is around 3.15 V |
| Supply current | 125 nA typical for a supervisor that includes programmable sense and reset delays |
| Sense input current | 25 nA typical and 0.1 microamp maximum, which is what makes an external divider practical |
| Sense delay | Set by the CTS capacitor: 654666 times the capacitance in farads plus 50 microseconds |
| Reset delay | Set by the CTR capacitor: 654666 times the capacitance in farads plus 80 microseconds |
| Capacitor ceiling | 10 microfarad on either pin, because the capacitor has to discharge during the fault for the next delay to be correct |
| Push-button use | The CTS input doubles as a debounce and manual reset path, holding the output asserted while the pin is pulled low |
| Output | Open drain active-low, 10 nA typical and 100 nA maximum leakage, 300 mV maximum low at 2 mA |
| Power-on reset | Output defined from 700 mV of VDD |
| Body | WSON DSE 6 pin, 1.50 mm by 1.50 mm |
Why the open-drain member matters on a shared reset net
A board with four supervisors and one reset line needs every driver to be able to pull low without fighting another driver that is sitting high. A push-pull output in that arrangement is a short circuit waiting for the first fault. An open-drain output on all four parts makes the net a wired-or, and the pull-up sets the high level for the whole chain.
The pull-up value is then a real decision. The leakage on this member is 100 nA maximum, so four parts on one net leak 400 nA, which a 100 kOhm pull-up to 3.3 V absorbs with a 33 microamp standing current. A 1 MOhm pull-up drops the standing current to 3 microamp and pushes the rise time up, and on a reset net the rise time is what a slow Schmitt trigger input will notice.
The second reason to use the open-drain member is level translation. The output can be pulled to a rail above VDD, which is how a 1.8 V supervisor drives a 3.3 V reset input without a translator. The same is true in reverse, and the data sheet prints the leakage at 6 V for that case.
Two delays, and the debounce that comes free
This part has two capacitors and two independent delays, which is unusual at this current level. CTS sets how long a fault has to last before the output asserts, and CTR sets how long the output stays asserted after the rail recovers.
| Both open | 50 microseconds to assert and 80 microseconds to release, the fastest configuration |
| CTS 1 nF | About 0.7 ms of debounce before the fault is accepted |
| CTS 100 nF | About 65 ms of debounce, which covers a push-button press and the bounce that follows it |
| CTR 100 nF | About 65 ms of hold after recovery, the figure used to let a reference settle |
| CTR 1 microfarad | About 0.65 s of hold, used where a converter has to discharge before restart |
| Both 10 microfarad | About 6.5 s each, the top of the range and the point where leakage in the capacitor dominates the timing |
The discharge condition is the detail that catches people out. If the fault is shorter than about a tenth of the programmed delay, the capacitor does not discharge fully and the next delay is shorter than the calculation says. On a rail that dips repeatedly that shows up as a reset that seems to arrive early, and the fix is either a smaller capacitor or a longer fault.
Where the DL30 is used
- Multi-rail boards where several supervisors drive one reset net
- Battery powered and portable instruments that need a supervisor drawing less than a microamp
- Solid state drives and enterprise boards where a 3.3 V rail has to be watched with a defined release point
- Body control modules and building automation controllers with a 3.3 V rail and a button on the same pin
- Boards that monitor a divided battery voltage as well as a logic rail, using the separate sense pin
Electricity meters, motor drives, smart thermostats and set top boxes are the applications the vendor prints, and all of them share a 3.3 V rail, a long standby life and a reset net that has more than one driver on it.
Order code, marking and reel
| TPS3899DL30DSER | Active, production, WSON DSE 6 pin, 3000 pieces on a large reel, marking LM |
| Lead finish | NIPDAU, moisture level 1 at 260 degrees with unlimited floor life |
| Reel drawing | 210 mm reel, 8.4 mm tape, 4 mm pitch, which is the same carrier as the rest of the family |
| Same threshold, other outputs | TPS3899PL30DSER push-pull active-low and TPS3899PH30DSE push-pull active-high |
| Adjacent thresholds | 2.90 V and 3.10 V sit on either side, and the 3.10 V option leaves very little margin on a 3.3 V rail at minus 5 % |
Sourcing notes for this line
A nanoamp supervisor with two programmable delays is a narrow category, and the parts that claim to match it often drop one of the two capacitors. The power management cross reference lists the candidates by pinout and marks whether the sense delay is programmable or fixed.
Where a customer needs the 3.0 V option and the reel is short, the push-pull member of the same threshold is the closest substitute and it changes the reset net rather than the threshold, which is a smaller change than moving to a 2.9 V part. The pin to pin cross reference page sets out what has to be re-checked when the output topology changes.
The 3.3 V rail supervisor is also one of the first parts to be pulled forward when a programme ramps, so the standing stock position is worth agreeing early. The safety stock planning page covers how that buffer is sized against the consumption rate.
Questions buyers ask about this part
Can two of these drive one reset line? Yes, that is what the open-drain output is for, and the pull-up has to be sized for the combined leakage.
Does the 5 % hysteresis mean a 150 mV release on a 3.0 V trip? Yes, about that, and on a 3.3 V rail it means the rail has to pass 3.15 V before the delay starts.
Is the delay accurate across units? It follows a 500 kOhm internal resistance with its own tolerance, so a factor of about two across the window is realistic for small capacitors and narrows as the capacitance rises.
What happens if the CTS capacitor is left off? The part asserts the output 50 microseconds after the threshold is crossed, which is fine on a clean rail and too fast on a noisy one.