AD5754BREZ
Product Specifications
AD5754BREZ
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The AD5754BREZ 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
AD5754BREZ: Four 16-Bit Bipolar Outputs and a 16 LSB Linearity Figure to Read Carefully
The AD5754 is the 16-bit member of the quad voltage output family that also contains a 12-bit and a 14-bit part. It runs from a single supply of 4.5 V to 16.5 V or from dual supplies of plus and minus 4.5 V to 16.5 V, and the output range is chosen in software from plus 5 V, plus 10 V, plus 10.8 V, and the same three spans bipolar. Each channel has an integrated output amplifier and a reference buffer, the four can be updated together from a LDAC pulse, an asynchronous CLR takes them to zero or midscale, and the SPI port runs at 30 MHz. Integral nonlinearity is 16 LSB maximum, differential nonlinearity 1 LSB, total unadjusted error 0.1 percent of full scale, and settling time 10 us typical in a 24-lead TSSOP over minus 40 to plus 85 degrees C.
16 LSB of INL is 4.88 mV on a 20 V span, and 0.1 percent TUE is 20 mV
On the plus and minus 10 V range the span is 20 V and one code is 305 uV, so 16 LSB of integral nonlinearity is 4.88 mV, or 0.024 percent of full scale. The total unadjusted error of 0.1 percent is 20 mV, four times larger. Reading those two numbers in the right order matters: the linearity term describes the shape of the transfer function, the unadjusted error describes where the endpoints actually are, and for a set point application the second one is the one that has to be trimmed. A calibration that removes the offset and gain error at two points leaves the 4.88 mV bow of the INL curve, which is 0.024 percent and is the real accuracy of a trimmed channel.
10 us of settling against 0.8 us of SPI frame time
A 24-bit frame at 30 MHz takes 0.8 us, so the interface can load a code far faster than the output can settle: 10 us of settling sets the fastest useful update rate at about 100 kHz, and it is the settling and not the bus that decides how fast a waveform can be generated. For dc set points and for a servo loop running at a few kilohertz this is irrelevant. For a function generator or an ATE source that has to step between codes at a megahertz, the part is the wrong choice and a lower resolution converter with a 1 us settling time is the right one.
| Parameter | Value | What it decides |
|---|---|---|
| Resolution | 16 bits, four channels | One code is 305 uV on a 20 V span |
| Output ranges | Plus and minus 5 V, 10 V, 10.8 V and 0 V to 5 V, 10 V, 10.8 V | Six spans, register selected |
| Integral nonlinearity | 16 LSB maximum | 4.88 mV on a 20 V span, the trimmed accuracy |
| Differential nonlinearity | 1 LSB maximum | Monotonic, no missing codes |
| Total unadjusted error | 0.1 percent of full scale | 20 mV, the untrimmed error |
| Settling time | 10 us typical | About 100 kHz of useful update rate |
| Supplies | Single 4.5 to 16.5 V, or dual 4.5 to 16.5 V | Plus and minus 15 V for a plus and minus 10 V span |
| Reference | External, with integrated buffers | One reference for four channels |
| Simultaneous update | LDAC pulse | Four channels change on the same edge |
| Asynchronous clear | Zero scale or midscale | Hardware state that a host fault cannot override |
| Input coding | Twos complement or offset binary | Pin selected, matches the host arithmetic |
| Interface | SPI to 30 MHz, double buffered | 0.8 us per frame, load ahead of the update |
| Package and range | 24-lead TSSOP | Minus 40 to plus 85 degrees C |
Why plus and minus 10.8 V exists as a range
The 10.8 V span is not a round number by accident: it matches a 10 V full scale instrument plus 8 percent of over range, which is the convention in process control and in some aerospace signal chains. Having it as a register setting means the same board can serve a plus and minus 10 V product and a plus and minus 10.8 V product without a component change, which is the point of a quad converter with software ranges. The cost is that the plus and minus 10.8 V setting needs rails that leave headroom for 10.8 V, so a plus and minus 12 V supply is only 1.2 V above the output, and the amplifier's output stage needs to be checked against that at full load current.
What an asynchronous clear protects
The CLR pin takes all four outputs to zero or midscale without waiting for the serial port, which is the state a board needs when a watchdog fires or when a host resets mid frame. Without it, the outputs hold their last code until the host writes again, and in a system where a valve or a bias current is the load, holding the last code is the wrong answer. The pin is asynchronous, so it works even if the SPI clock is stopped, and the state it selects has to be the fail safe one for the load. It is not a substitute for a hardware interlock on anything that can hurt a person.
Bipolar output from a single supply
The part can produce a plus and minus 10 V span from a single supply only if the supply is above 10.8 V and the output stage can swing to both extremes, which for a single supply means the internal amplifier generates its own negative rail or the output is referenced to a mid rail. In practice the dual supply arrangement is the one that gives a clean plus and minus 10 V with low distortion, and the single supply arrangement is used for unipolar spans where 0 V to 10 V from a 12 V or 15 V rail is all that is needed. Getting that choice wrong shows up as a compressed negative half of the span, not as a failure.
Buying a quad channel source
Four matched channels in one package are used where the channels have to track, such as a four quadrant bias or a multi axis driver, and the tracking is a property of the package rather than of individual parts. A substitute has to be compared on the output ranges, the reference arrangement and the settling time, and the four channel parts that can be compared on those terms are listed in ADC DAC cross reference. Where a board is being built around several converter types, the consolidation route is in one stop BOM sourcing, and the cost comparison between package options is covered in how to reduce component cost.