TPS54160ADGQ
Product Specifications
TPS54160ADGQ
Texas Instruments — Power Management & Signal Chain
Product Overview
The TPS54160ADGQ 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
TPS54160ADGQ: a 1.5 A step down converter that takes 60 V at the input and draws 116 uA when it runs
TPS54160ADGQ is a wide input buck regulator with the high side FET inside, built to hold a low voltage rail from a 12 V, 24 V or 48 V bus and to keep its own consumption small enough for a load that sleeps in a HVSSOP body of 10 pins. The DGQ body is ten pins on a 3 mm by 3 mm pad with an exposed thermal pad underneath, and the A suffix in the code marks the silicon with the tighter enable threshold rather than a different power stage.
Order code and package
| Catalogue string | TPS54160ADGQ, the A silicon in the DGQ body |
| Body | HVSSOP, ten pins, DGQ, 3.00 mm by 3.00 mm, with an exposed pad that has to be soldered to the ground plane |
| Reel | 80 pieces in a tube, which is a handling format rather than a reel and has to be planned for on an automated line |
| Finish and moisture | NIPDAUAG plating, moisture level 1, 260 C, unlimited floor life |
| Temperature | minus 40 to 125 C junction |
| Marking | 5416A, and the A in the marking is the only visible difference from the earlier silicon |
| Document | SLVSB56C, revised February 2014 |
| Alternative body | the DRC VSON body at the same 3 mm by 3 mm and the same pinout, with a better thermal path |
Electrical parameters that decide the application
| Input range | 3.5 V to 60 V, which covers a 48 V telecom bus with margin for a transient |
| Output current | 1.5 A continuous with the high side FET inside the package |
| High side resistance | 200 mOhm, which is what sets the conduction loss and therefore the practical current at low duty cycles |
| Output range | 0.8 V up to the input minus the dropout of the internal switch, set by a divider from a 0.8 V reference |
| Reference | 0.792 V minimum, 0.800 V typical, 0.808 V maximum at 25 C junction, so 1 per cent at room temperature |
| Quiescent current | 116 uA typical and 136 uA maximum in the non switching state at 12 V in |
| Shutdown current | 1.3 uA, which is what a battery fed industrial product needs from a regulator that is switched off |
| Switching frequency | 100 kHz to 2.5 MHz set by a resistor, with synchronisation to an external clock available |
| Current limit | 1.8 A minimum and 2.7 A typical, measured at 12 V in and 25 C, which is a peak switch limit rather than an output limit |
| Lockout | 2.5 V internally, raised by a divider on the enable pin, and the A silicon has the tighter enable threshold of the two |
| Power good | open drain, asserting when the output is inside 94 to 107 per cent of nominal, with the window widened from the earlier 93 per cent |
| Soft start | a capacitor on the slow start pin sets the ramp, and the same pin can be driven for tracking and sequencing |
| Eco-mode | pulse skipping at light load, which is where the 116 uA figure comes from and what keeps a standby rail efficient |
| Thermal | 62.5 C per W junction to ambient in the DGQ body on a standard board and 40 C per W in the DRC body |
The A silicon against the earlier part, and the two bodies
| TPS54160ADGQ | the A silicon in the HVSSOP body, 80 pieces in a tube |
| TPS54160ADRC | the same silicon in the VSON body, with the lower thermal resistance |
| TPS54160DGQ | the earlier silicon in the same body, with the wider enable threshold and the 93 per cent power good window |
| TPS54161 | a relative with a different current limit and a lower switching floor |
| TPS54260 | the 2.5 A member of the same family, for a load that outgrows 1.5 A |
| What the A suffix changes | the enable threshold and the power good window, nothing in the power stage |
Points to check before this part goes on a BOM
The 80 piece tube is the detail that stops a line. Almost every other part on a modern board arrives on a reel, and this row is a tube, so a pick and place machine needs a tube feeder or the parts have to be re reeled. The DRC body is the same silicon on a reel, which makes it the practical choice for volume even though the DGQ body is the one in the title of this page.
1.5 A from a 200 mOhm high side FET at 48 V in and 5 V out is a duty cycle of about 10 per cent, so the conduction loss is small and the switching loss dominates. At 2.5 MHz the switching loss can be larger than the conduction loss at full load, which is why the frequency range goes down to 100 kHz. A build that picks the highest frequency for a small inductor and then finds the part running hot has picked the wrong end of the range.
The internal lockout is 2.5 V and the enable pin raises it. That matters on a 48 V bus that ramps slowly, because the converter would otherwise start at 2.5 V, draw a large current at a low input voltage, and stop again as the bus continues to rise. A divider on the enable pin with a defined hysteresis holds the part off until the bus is genuinely up, and the A silicon makes that threshold more repeatable than the earlier part.
The power good window is 94 to 107 per cent and it changed from 93 per cent in this revision. Six per cent above nominal is a wide window, and it is wide on purpose: the pin is there to tell a system controller that the rail exists rather than to certify it. A rail that has to be inside 2 per cent before a memory device is released needs an external supervisor, and the PWRGD pin is not that.
Eco-mode is the reason the quiescent current is 116 uA. At light load the part skips pulses and runs at a low effective frequency, which keeps the switching loss down and the efficiency up. The trade is ripple and noise at light load, which is larger and less regular than in forced continuous mode. An application that measures the rail at no load and expects a clean waveform will be surprised by pulse skipping, and that is the mode working as intended.
The exposed pad is not optional. 62.5 C per W junction to ambient in the DGQ body assumes the pad is soldered to a copper area, and a pad that is left floating changes the thermal path completely. On a 24 V to 5 V conversion at 1 A the loss is around 0.5 W, which is 31 C of rise on the quoted figure and considerably more without the pad. The pad also carries the ground return, so the switching loop depends on it as much as the heat does.
Choosing the frequency against the inductor and the loss
The frequency range runs from 100 kHz to 2.5 MHz and each end costs something different. At 100 kHz the inductor is large and the switching loss is small, which suits a 48 V input where the switching loss dominates. At 2.5 MHz the inductor is small and the switching loss is large, which suits a 12 V input with a tight board area. The useful starting point is the input voltage, because the switching loss scales with it.
The second constraint is the minimum on time. At 2.5 MHz the period is 400 ns, and a 48 V to 3.3 V conversion needs a duty cycle of 7 per cent, which is 28 ns of on time. The part can produce that, and the pulse becomes short enough that the current limit comparator has to be fast to see it. At the low end of the frequency range the same conversion has 700 ns of on time and no such concern, which is why high input voltage rails stay below 500 kHz.
What the enable divider is really doing
The divider on the enable pin is a programmable lockout with hysteresis, and it is the difference between a converter that starts cleanly and one that chatters on a slow bus. The pin has a 1.25 V threshold and a defined hysteresis, and a divider from the input sets the voltage at which it crosses. Two resistors therefore decide the turn on point and, through the hysteresis, the turn off point below it.
The reason it matters is the start up current. A converter that begins switching at 2.5 V input draws its full current limit from a source that is still ramping, which drags the bus down, which drops the part below its lockout, which restarts it. The result is a sawtooth on the bus that can last for hundreds of milliseconds and that looks like a failed supply. Setting the lockout above the point where the source can supply the start up current removes the whole behaviour.
Where this order code fits
This is the low voltage rail on a 12 V, 24 V or 48 V industrial board, a telecom line card, an aftermarket automotive accessory or a building controller, where the bus is high and the load is modest 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 One Stop BOM Sourcing page.
Questions buyers ask
Can it run from a 48 V bus? Yes, the input range is 3.5 V to 60 V. At that input the duty cycle is low, so the switching frequency should be kept near or below 500 kHz to keep the switching loss and the minimum on time under control.
What is the difference between the A and the non A part? The A silicon has a tighter enable threshold and a power good window of 94 to 107 per cent instead of 93 to 107. The power stage, the current limit and the package are the same.
Why is this row in a tube? The DGQ body is offered in an 80 piece tube. The same silicon in the DRC VSON body is available on a reel, and that is the practical choice for an automated line.
Is 1.5 A available at 48 V in? The current limit is 1.8 A minimum, so the peak switch current is above the load. The thermal calculation at 48 V is dominated by switching loss, and the DRC body has the lower junction to ambient resistance.
What sets the output voltage? A divider from the 0.8 V reference, which is 1 per cent at 25 C. The reference window is 0.792 V to 0.808 V, so the divider tolerance has to be added to that.
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 Component Lead Time Solutions page sets out the checks that keep a substitute from failing on the bench.