UCC28911DR
Product Specifications
UCC28911DR
Texas Instruments — Power Management & Signal Chain
Product Overview
The UCC28911DR 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
UCC28911DR: a 700 V flyback switcher with primary side regulated voltage and current loops, 115 kHz maximum switching, a 700 mA peak current and under 30 mW of standby input power, in the seven pin SOIC on a 2500 piece reel
UCC28911DR is a complete flyback power stage in one body, with the 700 V FET, the high-voltage start-up current source and the control loop on the same die, for adapters and bias supplies up to about 10 W in a SOIC body of 7 pins. The D body is a seven pin SOIC that leaves out the eighth pin because the drain sits on its own isolated tab, and the same outline carries both members of the family.
Order code and package
| Catalogue string | UCC28911DR, where D is the seven pin SOIC and R is the 2500 piece large tape and reel |
| Body | SOIC, D, seven pins, with the drain on pins 5, 7 and 8 of the thermal tab arrangement the family uses |
| Tube | UCC28911D ships 75 pieces in a tube for a prototype, and the G4 row UCC28911DRG4 carries the same 2500 piece reel with a different material declaration |
| Finish and moisture | NIPDAU plating, moisture level 2 at 260 C with one year of floor life |
| Temperature | minus 40 to 125 C operating, with the electrical table specified over the same ambient range |
| Marking | U28911 on the body |
| Document | SLUS769D, July 2013, revised December 2016 |
| Status | Active and in production, with the 600 mA sibling UCC28910 also active on the same outline |
| Output power | 8 W in an enclosed adapter and 12 W on an open frame at 175 VAC to 265 VAC, or 7.5 W and 10 W at 90 VAC to 265 VAC, with the junction held at 125 C |
Electrical parameters that decide the application
| Power FET | 700 V drain to source breakdown, integrated on the die, with 6.25 Ohm typical and 7.2 Ohm maximum on resistance at 25 C rising to 11.4 Ohm and 13.4 Ohm at 125 C |
| Peak current | 700 mA with the IPK pin shorted to ground, window 680 to 720 mA, against 600 mA on the UCC28910 whose on resistance is 10.5 Ohm typical |
| Over-current | 980 mA typical, window 880 to 1090 mA, on the UCC28911, which is the hard limit rather than the programmed peak |
| Bias lockout | VDD turns on at 9.5 V typical, 9.0 V minimum and 10.0 V maximum, turns off at 6.5 V, and the 3.0 V of hysteresis is what lets the VDD capacitor ride through a line dip |
| Run current | 2.9 mA typical in the run state, with a wait current near 270 uA and a start current of a few microamps before the converter runs |
| VS regulating level | 4.05 V at no load and 25 C, window 4.01 V to 4.09 V, which is the constant voltage reference read from the auxiliary winding |
| Over-voltage | 4.60 V at the VS pin, window 4.45 V to 4.75 V, so a 5 V output trips at about 5.6 V |
| Switching frequency | 115 kHz maximum, window 105 to 125 kHz, and about 420 Hz minimum with 360 to 500 Hz across the two devices |
| Valley switching | the controller waits for a valley in the drain ring before it turns the FET on, and skips valleys at light load, which is where the frequency dithering and the low standby power come from |
| Thermal | 102.2 C per W junction to ambient and 39.1 C per W junction to case in the seven pin body, with the shutdown at 150 C and 50 C of hysteresis |
| Protections | output over-voltage, low line, over-current, output short circuit and thermal shutdown, all of which end in a restart rather than a latch |
| Standby | under 30 mW of input power at no load is the headline, achieved by switching off the high-voltage current source during operation and dropping the control current with load |
What the seven pins do
| DRAIN | the drain of the internal 700 V FET, which also feeds the high-voltage current source that charges VDD at start |
| VDD | the bias rail, charged from the drain at start and held by the auxiliary winding in run, with a 28 V clamp that sinks the excess current |
| VS | the auxiliary winding sense input, which sets the output regulation point, the over-voltage threshold, the line sense and the demagnetisation timing |
| IPK | the peak current programming pin, where a resistor to ground sets the peak, and shorting the pin to ground selects the maximum peak of 700 mA |
| GND | the return for the control and the FET source |
| Thermal tab | the drain connection is also the heat path, so the copper under the body carries both the current and the heat |
Points to check before this part goes on a BOM
The 700 V FET inside the body sets the whole application. A flyback running from a 265 VAC mains sees a drain voltage of about 375 V plus the reflected output voltage plus the leakage spike, and a 700 V device with a 600 V reflected plateau leaves very little room. The usual arrangement is a 1:8 turns ratio for a 5 V output and a clamp that holds the spike at 550 V, which keeps the drain under 650 V at the highest line. Raising the output to 12 V changes the reflected voltage and eats the margin, and that is the point at which the board either moves to a 900 V device or accepts a lower maximum line.
The 700 mA peak current and the 6.25 Ohm on resistance are the two numbers that decide the output power. Conduction loss at 700 mA through 6.25 Ohm is 3 W, which is more than the converter delivers, so the part spends most of its life at a much lower current and the peak is reached only during a transient or at low line. The equation in the data sheet ties the peak to the transformer primary inductance and the target frequency, and the 8 W to 12 W power figures come from a 125 C junction rather than from the current rating alone. Below 90 VAC the same transformer delivers less, which is why the 90 VAC to 265 VAC column is lower.
Primary side regulation from an auxiliary winding is the same compromise the rest of this family makes. The output voltage is inferred from the winding voltage during demagnetisation, so the leakage inductance spike has to have decayed before the sample is taken, and the 2.1 us zero crossing timeout is what the controller waits before it decides there was no valley. A transformer with poor coupling gives a supply that is 5 per cent high at no load and 5 per cent low at full load, and the vendor's 5 per cent guarantee assumes a winding built to the recommended order with the auxiliary winding on top of the secondary.
The start-up current source is switched off during operation, and that is where the standby power comes from. At start the drain feeds a current source that charges VDD to 9.5 V, then the source turns off and the auxiliary winding takes over. At no load the controller drops into a wait state where the current is a few hundred microamps and the switching frequency falls to a few hundred hertz, so the input power stays under 30 mW. A board that leaves a bleeder resistor across the output, or a feedback divider that draws 100 uA, loses that number regardless of what the controller does, and the measurement then blames the part.
The ISENSE arrangement is internal and the IPK pin is the only adjustment. A resistor from IPK to ground sets the peak current, and shorting the pin to ground selects the 700 mA maximum, which is the setting for a board that wants the full power and accepts the transformer that comes with it. The data sheet notes that a resistor between the short threshold and the minimum usable value is not allowed, because the part cannot tell which of the two sense paths is in use, and a value in that window gives an unpredictable peak. That window is the one detail on this page most likely to be missed on a first build.
Thermal performance is the limit on the output power, not the current rating. With 102.2 C per W and a 125 C junction limit in a 50 C ambient, the package can dissipate about 0.73 W, and at 85 per cent efficiency a 10 W output means 1.8 W of loss, which has to be shared between the FET, the rectifier and the clamp. The data sheet's power table assumes a copper area of more than 560 mm2 on the ground pin for the enclosed adapter case and a 50 C per W path for the open frame case. A two layer board with a small pour on the drain tab meets neither, and the supply that passes at 25 C derates to 6 W at 50 C.
UCC28910 or UCC28911 on the same board
The two devices share a pinout, a document and a control loop, and differ in the FET. The 28910 has 10.5 Ohm on resistance and a 600 mA peak, the 28911 has 6.25 Ohm and 700 mA, and the difference is worth about 2 W of output at the top of the range. A board that is laid out for one accepts the other, and that is useful when a build has to ship at a lower power because a transformer is late.
The choice is not free at the bottom of the range. The 28911 has a higher peak current and therefore a larger transformer, and at 5 W the extra copper and the larger core cost more than the FET saves. The 28910 with the smaller peak is the better part below about 6 W, and the 28911 above it.
Where the 30 mW claim comes from
The number is an input power measured at the plug with the converter in burst mode and no load, and it depends on the VDD capacitor, the bleed resistor across it and the leakage of the output capacitor. The controller contributes the active part: the high-voltage current source is off, the wait current is a few hundred microamps, and the switching frequency falls to a few hundred hertz. Everything else on the board contributes the rest, and a 10 uF VDD capacitor with a 1 MOhm bleeder alone burns 0.4 mW at 20 V, which is small but not zero.
Any measurement of standby power has to be taken with a power meter that resolves 1 mW and after the converter has been at no load for several minutes, because the burst period is long and a short capture window catches a switching burst and reports a number several times too high.
Where this order code fits
Wall adapters and chargers up to 10 W, bias supplies for metering, smoke and fire alarms and thermostats, building automation nodes and white goods control boards that have to hold standby power under 30 mW Teams comparing it with a neighbouring part start from our Power Management Cross Reference page, and teams replacing a part that has gone end of life start from the Component Lead Time Solutions page.
Questions buyers ask
Does it need an optocoupler? No. The output voltage and current are read from the auxiliary winding, so the secondary carries a rectifier and a capacitor.
What is the difference from the UCC28910? The same body and control loop with a lower on resistance and a higher peak current, 700 mA against 600 mA, which buys about 2 W of output.
How is the peak current set? A resistor from the IPK pin to ground, or a short to ground for the 700 mA maximum. Values between the short threshold and the minimum usable resistance are not allowed.
What limits the output power? The thermal path. The power table assumes a 125 C junction and a copper area on the ground pin that a small board does not have.
What happens on a short circuit? The over-current and output short circuit protections stop switching and the part restarts, with the over-voltage and low line protections sharing the same restart behaviour.
SUPERB holds and sources this line for production and repair demand, quoted against the exact ordering string with date codes, packing and country of origin stated on the offer. When a part is moving, the How to Verify a Drop in Replacement page sets out the checks that keep a substitute from failing on the bench.