TPS2491DGSR
Product Specifications
TPS2491DGSR
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS2491DGSR 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
TPS2491DGSR positive high voltage hot swap controller with power limiting and automatic retry
The TPS2491 is a ten pin hot swap controller for positive rails up to 80 V that limits both the current and the power in the external MOSFET. The power limit is the feature that separates it from a plain current limit controller: a fixed current limit into a discharged load capacitance delivers the full current at the full rail voltage, which is the worst case for the FET, while a power limit folds the current back as the drain to source voltage rises. The retry behaviour on this member restarts the load after a fault without an operator action.
The part provides a 4 V reference that sets the power threshold through a resistor on PROG, a current sense input that takes the drop across an external sense resistor, a timer capacitor for the fault window, a power good output, and an enable input that can be driven by a logic signal or by an undervoltage divider. Its sibling member latches off after a fault, and the two share a document and a pinout.
Key parameters taken from the TPS2491DGSR document
| Input range | 9 V to 80 V on the supply pin, which covers 12 V, 24 V and 48 V telecom rails |
| Reference | 4.0 V output with a 3.9 to 4.1 V window, used to set the power limit through the PROG pin |
| Power limit threshold | A 25 mV sense threshold with the power limit active, 17 to 33 mV window, set by the divider on PROG |
| Current limit threshold | A 50 mV sense threshold with the power limit inactive, 45 to 55 mV window, set when PROG is held at 4 V |
| Retry | 0.75 percent duty cycle typical with a 0.5 to 1 percent window, which is the fraction of time the part spends retrying |
| Timer reset threshold | 1.0 V typical with a 0.96 to 1.04 V window on the retry member |
| Gate drive | Charge pump output that holds the gate above the source, with 125 mA of sink current to turn the FET off |
| Gate regulation | The gate is held so that the drop across the sense resistor follows the PROG setting, which is the power limit loop |
| Power good | Thresholds on the output rising and falling, open drain output |
| Enable | Input that both enables the part and can be used as an undervoltage lockout with a divider |
| Package | VSSOP DGS, 10 pins, 2500 pieces on a large reel and 80 pieces in a tube |
| Moisture and temperature | Level 2 at 260 C with a one year floor life, minus 40 to 85 C |
| Addendum status | TPS2491DGS and TPS2491DGSR are Active and in Production, marking BIX on the reel rows |
Every row above is read from the vendor document, not from a distributor summary. Where a window is printed, the window is what a production test should use, and the typical value is only a starting point for a first article.
Where the numbers matter in a real build
Power limiting is the answer to the inrush problem on a 48 V rail. A 48 V board with 470 microfarad of input capacitance and a 5 A current limit would dissipate 240 W in the MOSFET for the whole ramp, which no single device survives. With the power limit set at 20 W the part folds the current back to 0.4 A while the drain to source voltage is high and lets it rise as the output charges, so the same ramp happens inside the safe operating area.
The 25 mV and 50 mV thresholds are two different mechanisms. The 25 mV row applies when the power limit loop is active, which is the normal case during a start up or a fault. The 50 mV row applies when the loop is inactive, which happens with PROG held at 4 V, and that is the setting a product uses when it wants a plain current limit and no fold back. Reading the two rows as one number is the most common mistake on this part.
The retry duty cycle of 0.75 percent is what keeps a permanent fault from becoming a heater. The part tries for a fraction of a percent of the time, which on a 48 V rail with a 470 microfarad load means the average power in the MOSFET stays low even though each attempt is at the full power limit. The timer components set that fraction, and the document gives the retry duty as a specified row rather than a note, which means it is a number a production test can check.
The reference output is a 4 V source with a 1 mA capability, and it is what makes the power setting a resistor calculation rather than a lookup. The divider on PROG compares a fraction of the drain to source voltage against that reference, and the loop adjusts the gate until the product of current and voltage matches. Because the reference is specified over temperature and over the operating range, the power limit is a real limit and not a nominal one.
The gate sink current of 125 mA typical is what turns the MOSFET off in a fault. A 47 nF gate capacitor discharges in about 20 microseconds at that current from a 15 V gate drive, which is inside the time a 48 V rail takes to collapse. A larger gate capacitor slows the turn off and is the item to check when a product shows a large negative spike on the rail during a hot pull.
- Power limit setting against the safe operating area of the MOSFET, since the whole purpose is to keep the FET inside it during the ramp.
- Sense resistor value and tolerance, because the 25 mV and 50 mV thresholds both scale with it.
- Retry duty cycle against the upstream supply and the thermal time constant of the FET.
- Gate capacitor value against the turn off time the rail needs during a hot pull.
- Enable divider against the undervoltage point the product needs, since the same pin serves both functions.
- Timer capacitor against the longest legitimate start up, because a short timer turns a slow ramp into a fault.
Which projects this part fits
This part fits 48 V telecom line cards, 24 V industrial backplanes, blade servers, storage arrays and any product that has to hot plug into a live backplane with a large input capacitance. It also fits 12 V systems that need a power limit rather than a plain current limit, because the fold back is what keeps a small MOSFET inside its safe area when the load capacitance is large.
Questions that come up before a purchase order is released
What is the difference between this part and the current monitor member?
This controller sets a power limit and a current limit and has no current monitor output. The sibling part in the same document adds an analog current monitor so a system can read the load current, at the cost of one more pin function.
How is the power limit calculated?
By the divider on PROG against the 4 V reference and the sense resistor value. The document gives the equation, and the sense threshold with the loop active is 25 mV, so the power limit is the product of that threshold and the rail voltage scaled by the divider.
Can the part be used without power limiting?
Yes, by holding PROG at 4 V, which makes the current limit threshold the 50 mV row. That is the configuration for a product that wants a plain current limit and no fold back.
What happens during a hot pull?
The gate is pulled down with 125 mA of sink current and the part enters its retry cycle. The negative spike on the rail depends on the gate capacitance and the parasitic inductance of the connector, which is why the gate capacitor is part of the review.
Is the part suitable for a 48 V rail with a 100 V transient?
The supply pin is rated to 80 V in the recommended range, so a rail that can reach 100 V needs a clamp in front of the controller. The document's application section shows a TVS in that position.
What is the retry duty cycle measured as?
As the fraction of time the part spends attempting to restart after a fault. The document specifies 0.75 percent typical with a 0.5 to 1 percent window, which is a number a test can verify on a retry event.
How SUPERB handles this line
We supply the TPS2491DGSR on original 2500 piece reels and as 80 piece tubes, with the date code and marking recorded on the packing list. The power management cross reference page lists the hot swap controllers we carry by rail voltage, pin to pin cross reference covers the sense resistors and the MOSFETs that sit next to them, and one stop BOM sourcing is how we deliver the controller, the FET and the passives against one build.