AD5684RBRUZ
Product Specifications
AD5684RBRUZ
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The AD5684RBRUZ from Analog Devices is a high-performance analog / mixed-signal IC providing precision signal acquisition, conversion, and conditioning. Analog Devices (ADI) is a world leader in high-performance analog, mixed-signal, and DSP ICs for instrumentation, healthcare, automotive, and communications.
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-resolution data conversion with excellent SNR and THD
- Wide dynamic range and high linearity across operating conditions
- Integrated voltage reference with low temperature drift
- Flexible digital interface — SPI, I2C, or parallel options
- Industrial/automotive temperature range with guaranteed specs
- Low power consumption with power-down modes
Key Specifications
| Resolution | bits |
| Channels | — |
| Sample Rate | kSPS |
| Interface | — |
| Supply Voltage | V |
| Operating Temp | °C |
| Package | — |
| INL | LSB |
| SNR | dB |
| THD | dB |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
AD5684R: Four 12-Bit Channels, One 2 ppm Reference, Four Wires of Bus
The AD5684R is a quad, 12-bit, voltage output digital to analog converter from the nanoDAC+ family, with an internal 2.5 V reference that drifts 2 ppm per degree C typical and 5 ppm per degree C maximum, a gain select pin that sets full scale to 2.5 V or 5 V, and a 50 MHz SPI port that also reads back. Integral nonlinearity is plus or minus 1 LSB at 12 bits, total unadjusted error is 0.1 percent of full scale maximum, and the whole part runs from 2.7 V to 5.5 V while dissipating 3.3 mW at 3 V. It comes in a 3 mm by 3 mm 16-lead LFCSP or a 16-lead TSSOP and is specified from minus 40 C to plus 105 C.
The reference is the reason to buy this one
Four DAC channels in one body save board area, but what separates the AD5684R from a plain quad converter is the reference inside it. A 5 ppm per degree C reference is a part that would otherwise have to be bought separately, place close to the converter, decouple and route, and its drift lands directly on every output. Integrating it means the reference and the converter share a die and a temperature, so the drift error is common to all four channels and is removed by a single system calibration instead of one per channel. Where the application needs better than 5 ppm, the reference input can be driven externally and the internal one switched off, which is the same part used a different way rather than a different order code.
Values at a glance
| Parameter | Value | Note |
|---|---|---|
| Channels, resolution | 4 channels, 12 bits | AD5685R is 14 bits, AD5686R is 16 bits, same footprint |
| Integral nonlinearity | Plus or minus 1 LSB maximum | Guaranteed monotonic |
| Total unadjusted error | 0.1 percent of full scale maximum | Offset 1.5 mV maximum, gain 0.1 percent |
| Internal reference | 2.5 V, 2 ppm per degree C typical | 5 ppm per degree C maximum |
| Output range | 2.5 V or 5 V, set by the GAIN pin | Reset to zero scale or midscale by the RSTSEL pin |
| Output drive | 20 mA within 0.5 V of either rail | Enough for a 4 mA to 20 mA loop stage ahead of it |
| Interface | SPI at up to 50 MHz with readback or daisy chain | 1.8 V logic compatible |
| Glitch | 0.5 nV seconds | Matters when the output feeds a control loop |
| Supply, power | 2.7 V to 5.5 V, 3.3 mW at 3 V | 16-lead LFCSP 3 mm by 3 mm or TSSOP |
Where a 12-bit quad part is the right answer
Laser bias, optical module setting, base station power amplifier bias and programmable reference levels in a PLC I/O card are all cases where four channels of 12 bits with a good reference beat one channel of 16 bits with a trimmed external reference. The reason is that the four outputs usually share a common scaling, so one calibration covers them, and the loop that consumes each output has its own gain that dominates the error. Where 12 bits is not enough is where the DAC output is the measurement, not a set point: a calibrator or a source meter needs 16 bits or better, and the AD5686R in the same footprint is the part to specify rather than adding a second converter.
The interface details that cost a day at bring up
The SPI port runs at 50 MHz, which is fast enough that a slow processor cannot overrun it and is also fast enough that a long ribbon cable will ring, so series resistors at the source are the usual fix. Readback means the host can verify what was written without trusting its own shadow register, and daisy chain means four channels cost one chip select if the bus is already crowded. The RSTSEL pin decides what the outputs do at power up, zero scale or midscale, and that choice decides whether the amplifier behind the DAC is driven into a rail at power up or held at a defined midpoint. The GAIN pin doubles the span, and it is a strapping, not a register, so the range is fixed by the board and a range change needs a resistor move and a new calibration.
Procurement and second source
Pin compatibility across the family is the useful property for supply: the 12, 14 and 16-bit members share the package and the pinout, so a shortage of one resolution can be covered by the next one up if the firmware scales the code. The ADC DAC cross reference lists the family and its neighbours, the pin to pin cross reference page covers the substitutions that need no board change, and lead time options for the LFCSP body are described on the component lead time solutions page.