TPS2372-3RGWR
Product Specifications
TPS2372-3RGWR
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS2372-3RGWR 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
TPS2372-3RGWR IEEE 802.3bt Type 3 powered device interface with a 100 V hot swap FET and autoclass
The TPS2372-3 is the Type 3 member of the TPS2372 powered device interface family: a complete PD front end for 60 W operation over four pair Power over Ethernet, built around a 100 V hot swap MOSFET with a 0.3 A inrush limit and a 1.85 A operating limit. It handles detection, classification, inrush, maintain power signature and the power good signal, in a twenty pin VQFN body specified to 125 C junction.
The part removes almost all of the discrete circuitry a PD front end used to need. Detection and classification resistors are external, but the classification current sources, the inrush FET, the maintain power signature pulses and the autoclass measurement are inside. That matters on a device that has to work with both a Type 3 and a Type 4 source, because the negotiation is done by the part and the system only has to read the allocated power outputs.
Key parameters taken from the TPS2372-3RGWR document
| Standard | IEEE 802.3bt Type 3 powered device, 60 W at the PD input |
| Hot swap FET | 100 V rating with a 0.3 A inrush current limit and a 1.85 A operating current limit |
| Inrush | 335 mA typical inrush limit and a 87 millisecond inrush delay when the delay pin is left open |
| Undervoltage lockout | Rising threshold 36.3 V minimum, 38.1 V typical and 40 V maximum, falling 30.5 V to 33.6 V, with 6.1 V of hysteresis |
| Classification | Two external resistors set the first and second classification currents, so the same board can present different classes |
| Autoclass | An enable input lets the source measure the actual load during classification and allocate power on that basis |
| Maintain power signature | Automatic MPS pulses, with the amplitude set by a resistor and the option to disable the function by leaving the pin open |
| Power good | Output with an inrush completion delay, so the converter is only started after the bulk capacitor is charged |
| Interfaces | Allocated power indicator outputs, and a delay enable pin referenced to the return rail |
| Thermal range | Minus 40 to 125 C junction temperature, which is printed in the addendum against this order code |
| Package | VQFN RGW twenty pins |
| Packing | 3000 pieces on a large reel, NIPDAU finish, moisture level 1 with unlimited floor life, marking TPS2372-3 |
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
The two current limits do different jobs and the numbers are far apart. The inrush limit is about a third of an amp and it holds while the bulk capacitor charges, which is what keeps the source from seeing a step. The operating limit is 1.85 A, which at 50 V is 92 W, deliberately above the 60 W the Type 3 class allows so that a short duration overload does not trip the port. A board that treats the 1.85 A figure as the working load is reading the wrong number: the load budget comes from the class allocation, and the current limit is a protection threshold.
The undervoltage lockout window is wide and it is the first thing to check against a source. The part turns on between 36.3 V and 40 V and turns off between 30.5 V and 33.6 V, with about 6 V of hysteresis. A source that sits at 37 V under load is inside the turn on window on one unit and outside it on another, which shows up as a port that starts on the bench and fails at the customer site. The IEEE specification sets the PD voltage range, and the part is built to it, but the margin between the source voltage and the lockout threshold is a system question that has to be answered with the cable loss included.
The inrush delay pin is the feature that most boards leave at its default, and the default is not always right. Left open, the pin enables a delay of about 87 milliseconds before the operating current limit takes over, which gives the bulk capacitor time to charge through the inrush limit. A board with a small capacitor bank can shorten or bypass the delay, and a board with a large bank may need the full delay. The delay is referenced to the return rail, not to ground, which matters when the return is not at chassis potential.
The maintain power signature is what keeps a low power device connected. A PD that draws less than the maintain current is disconnected by the source, so the part generates pulses at a duty cycle set by a resistor, and the option to disable the function exists for systems that draw enough current continuously. A board that disables MPS and then drops into a 5 milliwatt standby will lose the port, and the failure looks like a random disconnect rather than a power problem. The autoclass function is the other side of the same coin: it lets a source allocate power from a measurement instead of a class letter, which is how a 60 W port can serve a device that only needs 25 W.
- The lockout window against the source voltage at the end of the cable, since the threshold spread is wide.
- The bulk capacitor value against the inrush limit and the delay, because the capacitor sets the charge time.
- The classification resistors, since they decide which class the source sees and therefore how much power it allocates.
- The maintain power signature setting for any product that can drop below the maintain current.
- The power good delay, because the downstream converter must not start before the capacitor is charged.
- The junction temperature, since the hot swap FET dissipates during inrush and during any fault.
Which projects this part fits
This part fits PoE lighting, security cameras, multiband access points, small cells and pico base stations, and any device that needs more than 30 W from a four pair source. It also fits dual signature and pass through arrangements, where one port powers a device and a second port is passed downstream, because the allocated power outputs tell the system what the source granted.
Questions that come up before a purchase order is released
Does the part work with a Type 4 source?
Yes. It negotiates with a Type 3 or Type 4 source, and the Type 4 member of the same family is the one to use when the device needs the full 90 W. The difference between the members is the current limit point and the power level they support.
What sets the classification?
Two external resistors on the classification pins, one for the first event and one for the second. A board that needs a different class for two product variants can change one resistor rather than the whole front end.
What is autoclass for?
It lets the source measure the actual power draw during the classification phase and allocate on that basis, which can give a device more headroom than its class letter would suggest. The enable pin turns the function on.
Can the maintain power signature be turned off?
Yes, by leaving the control pin open. A product that does this has to guarantee it never draws less than the maintain current, otherwise the source disconnects the port and the device loses power without a fault indication.
What is the inrush delay for?
It holds the operating current limit off for about 87 milliseconds so the bulk capacitor can charge through the lower inrush limit. A board with a large capacitor bank needs the delay, and one with a small bank can shorten it.
How SUPERB handles this line
We supply the TPS2372-3RGWR on original 3000 piece reels with the marking TPS2372-3 recorded at goods in, and we can quote the Type 4 member of the same family when a board needs the higher power level. The power management cross reference page lists the PoE interfaces we carry by type and current limit, component lead time solutions covers the case where the exact member is short and the Type 4 part has to be qualified as a substitute, and one stop BOM sourcing is how a customer sends one list and gets the PD front end, the transformer and the passives quoted together.