AD5313WBRUZ
Product Specifications
AD5313WBRUZ
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The AD5313WBRUZ 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
AD5313WBRUZ: Two 10-Bit DACs With a 2 ppm Reference and 20 mA of Drive
The AD5313R is a dual 10-bit voltage output converter from the nanoDAC family with an on chip 2.5 V reference specified at 2 ppm per degree C typical, a total unadjusted error of 0.1 percent of full scale maximum, and output amplifiers that source or sink 20 mA while staying within 0.5 V of either rail. It has a pin that selects a gain of 1 or 2, another that selects reset to zero scale or midscale, and a SPI port that runs to 50 MHz with readback and daisy chain options. Power consumption is 3.3 mW at 3 V, the supply range is 2.7 V to 5.5 V, logic inputs are compatible with 1.8 V, and the package is a 3 mm by 3 mm 16-lead LFCSP over minus 40 to plus 105 degrees C.
A 10-bit step is 2.44 mV, and the 0.1 percent error is the same size
With the internal 2.5 V reference and a gain of 1, one least significant bit is 2.44 mV, and a total unadjusted error of 0.1 percent of full scale is 2.5 mV, so the part's guaranteed error is one code wide. Switching the gain pin to 2 doubles the span to 5 V and the step to 4.88 mV, and the error scales with it to 5 mV, which is again one code. That is the correct way to read the specification: the absolute accuracy of this converter is about one code, and any application that needs better than that has to calibrate the channel against a reference, at which point the 2 ppm reference drift becomes the term that matters rather than the initial error.
20 mA of drive inside half a volt of the rail
An output amplifier that holds 20 mA within 0.5 V of the rail is delivering 10 mW into its own pass device at that current, and it is doing so inside a 3 mm body. That capability is what allows the part to drive a 4 mA to 20 mA loop shunt, a LED bias or a relay coil directly instead of adding an external buffer. It is not a short circuit rating, and the dissipation has to be checked: at 20 mA and 0.5 V the amplifier dissipates 10 mW, which is small, but driving 20 mA into a 100 ohm load from a 5 V rail puts 4 V across the load and 1 V across the amplifier, which is 20 mW and no longer negligible in a 3 mm package.
| Parameter | Value | What it decides |
|---|---|---|
| Resolution | 10 bits, two channels | One code is 2.44 mV on a 2.5 V span |
| Reference | 2.5 V, 2 ppm per degree C typical | Drift is the limit after the channel is calibrated |
| Total unadjusted error | 0.1 percent of full scale maximum | About one code, before any calibration |
| Offset error | 1.5 mV maximum | Most of the error budget at the bottom of the range |
| Gain error | 0.1 percent of full scale maximum | Scales with the span selected by the gain pin |
| Output drive | 20 mA within 0.5 V of the rails | Drives a loop shunt or a coil without a buffer |
| Gain selection | 1 or 2, by pin | 2.5 V or 5 V span from the same reference |
| Reset state | Zero scale or midscale, by pin | Decides what the output does at power up |
| Interface | SPI to 50 MHz, with readback | Readback proves the register, not the output |
| Glitch energy | 0.5 nV per second | Small enough for a level set point, not for a waveform |
| Logic levels | 1.8 V compatible | Direct connection to a low voltage host |
| Supply and power | 2.7 V to 5.5 V, 3.3 mW at 3 V | 1.1 mA at 3 V, two channels included |
| Package and range | 16-lead LFCSP, 3 mm by 3 mm | Minus 40 to plus 105 degrees C |
Why a 2 ppm reference changes the error budget
The internal reference is specified at 2 ppm per degree C typical, which over 145 degrees of ambient range is about 290 ppm, or 0.725 mV on 2.5 V. That is smaller than the 2.5 mV total unadjusted error, so a channel that is calibrated at room temperature stays inside one code over temperature as far as the reference is concerned, and the remaining drift comes from the output amplifier and the resistor string. This is the reason the part appears in optical module bias controls and base station power amplifier bias networks, where the set point has to hold to a fraction of a percent across a rack that runs at 70 degrees C inside and minus 40 outside.
What the readback does and does not prove
The SPI port can read the input register back, and on a board that is being brought up that is the difference between knowing the host wrote the value and hoping it did. It does not prove the output voltage, because the readback path is digital and stops before the output amplifier. A production test that relies on readback alone will pass a part with a shorted output. The correct check is one measurement per channel at two codes, which catches both a stuck output and a reference fault, and costs two ADC readings per board.
Sourcing a dual set point converter
The order code carries the package and the temperature grade, and the two channels in one package are matched by construction, which is why the part is used where two set points have to track. A substitute has to be compared on the reference drift, the drive current and the reset state, because a part that powers up at midscale instead of zero will drive an amplifier into its limit before the host is out of reset. The converters that can be compared on resolution, reference and interface are listed in ADC DAC cross reference, the cost side of moving to a smaller package is covered in how to reduce component cost, and availability of the exact code is handled through component lead time solutions.