ADS5481
Product Specifications
ADS5481
Texas Instruments — Power Management & Signal Chain
Product Overview
The ADS5481 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
ADS5481: 16 bit 80 MSPS converter with LVDS outputs on a 5 V and 3.3 V supply pair
ADS5481 is the 80 MSPS member of a three part 16 bit family that runs from a 5 V analog rail and a 3.3 V digital rail and drives LVDS outputs from a QFN body with an exposed pad in a QFN body of 64 pins. The ADS5481 is a 16 bit converter that samples at 80 MSPS, with an integrated analog input buffer, an internal reference generator and LVDS compatible outputs. It runs from a 5 V analog supply and a 3.3 V analog and digital supply and dissipates 2.2 W, with a power down mode that drops to 70 mW. The body is a 64 pin QFN with an exposed pad, 9 mm by 9 mm, and the die is built on a complementary bipolar process, which is why the noise floor is low enough for 16 bit at this rate.
Order code and package
| Order code | ADS5481 |
| Channels | one |
| Resolution | 16 bit |
| Maximum sample rate | 80 MSPS |
| Family | ADS5481 at 80 MSPS, ADS5482 at 105 MSPS, ADS5483 at 135 MSPS |
| Analog supply | 5 V and 3.3 V |
| Digital supply | 3.3 V |
| Output interface | LVDS compatible |
| Input buffer | integrated, high impedance |
| Reference | internal generator, external option |
| Package | 64 pin QFN with exposed pad, 9 mm by 9 mm |
| Document | SLAS565C, June 2008, revised October 2009 |
Electrical parameters that decide the application
| Total power | 2.2 W at rated sample rate |
| Power down | 70 mW |
| Power per sample | 27.5 nJ at 80 MSPS |
| SFDR, family headline | 95 dBc at a 70 MHz input and 135 MSPS |
| SNR, family headline | 78.6 dBFS at a 70 MHz input and 135 MSPS |
| Quantisation floor | about 98 dBFS for an ideal 16 bit converter |
| Thermal path | the exposed pad, 49 vias in a 7 by 7 array in the vendor note |
| Clock | 3 V peak to peak differential |
| Input common mode | set by the internal buffer |
| Temperature | minus 40 to 85 C |
| Process | complementary bipolar, BiCom3 |
| Power up | supplies sequenced per the vendor table |
Points to check before this part goes on a BOM
The two headline dynamic figures belong to the 135 MSPS member of the family, not to this order code. The document states 95 dBc of SFDR and 78.6 dBFS of SNR at a 70 MHz input and 135 MSPS, and the 80 MSPS part has its own column in the same tables. A specification written from the headline line for an 80 MSPS build is a specification for a different part.
2.2 W in a 9 mm by 9 mm body is the first constraint on the board, not the converter performance. The vendor note calls for a 7 by 7 array of thermal vias under the pad, and the power down mode at 70 mW exists so a channel that is not in use stops heating the ones beside it, which matters when eight channels share one board.
Three supplies mean three sequencing rules. The document states that with one analog rail off the analog inputs must stay below 0.5 V, otherwise the protection diode between the input and that rail conducts. A hot plug event or a rail that comes up late is therefore a failure mode that damages the part rather than one that merely produces bad data.
An integrated buffer at 16 bit is what keeps the drive requirement sane. The sampling network is isolated from the source, so the amplifier in front sees a high impedance load, and the internal reference removes one more error term. On a multichannel board that repetition is the difference between a working channel count and a thermal problem.
Power per sample at 27.5 nJ is high against CMOS converters, and it is the price of 16 bit at 80 MSPS on a bipolar process. A 16 channel board at this figure is 35 W of converter power before the amplifiers, which drives the enclosure decision as much as the converter choice.
The 3 V peak to peak differential clock requirement is a clock distribution problem on a multichannel board. The clock must arrive with the stated amplitude at each converter, and a fan out that droops at the last channel shows up as an SNR difference between channels that no register setting can fix.
For a second source exercise the family structure matters. The three members differ in rated rate, and the package is shared, so a board built for the 80 MSPS part can take the faster one if the clock and the thermal path allow it. That is an upgrade path, not a substitution path, and the clock source is the item that usually decides it.
The document is SLAS565C, revised in October 2009, and the electrical tables separate typical from maximum columns. For a production limit the maximum column is the one that binds, and on a 16 bit converter the difference between typical and maximum offset or gain error is often the whole margin on a channel that has no calibration step.
Where this order code fits
This order code belongs on multi carrier GSM, WCDMA and LTE receive paths, test and measurement digitisers, medical imaging receive channels and radar baseband capture where 16 bit is needed to see a small signal beside a large one. The buffered input and the internal reference reduce the parts around it, which matters on a channel that is repeated eight or sixteen times on one board. Boards comparing it with a neighbouring part start from our ADC DAC Cross Reference page, and boards replacing a part that has gone end of life start from the Medical Device Components page.
Questions buyers ask
Are the SNR and SFDR figures on the front page valid for the 80 MSPS part? They are quoted at 135 MSPS, which is the fastest member of the family. The 80 MSPS order code has its own column in the electrical tables.
What is the heat load of this converter? 2.2 W per channel at rated rate and 70 mW in power down. On a multichannel board the power down mode is what keeps an unused channel from heating its neighbours.
Can the input be driven directly from a passive filter? The integrated buffer presents a high impedance load, so a passive network can drive it. The signal still has to arrive with the distortion and noise the 16 bit resolution expects.
SUPERB supplies this order code with date codes and packing on the offer, and the Obsolete RF Part Replacement page covers what to verify when a 16 bit channel has to be kept alive past the vendor life cycle.