AD8021ARZ
Product Specifications
AD8021ARZ
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The AD8021ARZ 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
AD8021ARZ: 2.1 nV Noise With a Compensation Pin You Choose
The AD8021ARZ is a voltage feedback amplifier for 16 bit signal chains, and its unusual feature is a compensation pin: a single capacitor from CCOMP to ground sets the gain bandwidth product, so the same part can be compensated for a gain of minus one at 200 MHz or a gain of minus ten at 190 MHz with almost no loss of bandwidth. Voltage noise is 2.1 nV per root hertz and current noise is 2.1 pA per root hertz, offset is 1 mV maximum with 0.5 uV per degree Celsius of drift, the supply range is plus and minus 2.25 V to plus or minus 12 V, and the quiescent current is 6.7 mA at 5 V. It settles to 0.01 percent in 23 ns on a 1 V step and recovers from overload in 50 ns.
Published values
| Parameter | Value |
|---|---|
| Voltage noise | 2.1 nV per root hertz |
| Current noise | 2.1 pA per root hertz |
| Bandwidth | 200 MHz at a gain of minus one, 190 MHz at a gain of minus ten |
| Settling time | 23 ns to 0.01 percent on a 1 V step |
| Overload recovery | 50 ns |
| Offset voltage | 1 mV maximum, 0.5 uV per degree Celsius drift |
| Bias current drift | 10 nA per degree Celsius |
| Distortion | Minus 93 dBc second harmonic, minus 108 dBc third harmonic at 1 MHz |
| Supply range | Plus and minus 2.25 V to plus and minus 12 V |
| Quiescent current | 6.7 mA at 5 V, 1.3 mA disabled |
| Bodies | 8-lead SOIC and 8-lead MSOP, minus 40 to plus 85 degrees |
Two noise terms, and the one that matters
Voltage noise of 2.1 nV per root hertz over a 1 MHz bandwidth is 2.1 uV RMS, which is 1.4 percent of a 16 bit code on a 10 V range, so the amplifier is not the limit at that bandwidth. Current noise of 2.1 pA per root hertz through a 10 kohm source impedance integrates to 21 uV over the same bandwidth, ten times the voltage noise term, and it is the reason a low impedance source is preferred in front of this part. Above roughly 15 kohm of source resistance the current noise dominates, and the low voltage noise specification stops being the reason to buy it. That crossover is the single most useful number when deciding whether a 16 bit front end can use this part at all.
The compensation pin is not optional decoration
The external capacitor sets the phase margin, and the datasheet gives the value for each gain. Leaving the pin open gives a part that is fast and marginally stable, which shows up as a few hundred millivolts of ringing on a step that should settle flat, and adding too much capacitance gives a part that is stable and much slower than the labels suggest. Boards that have been copied from a reference circuit without the capacitor, or that used a different value because the specified one was not fitted, are a common source of intermittent instability in 16 bit acquisition channels. The intended use is one capacitor per gain setting, so a programmable gain stage needs either a switched capacitor bank or a fixed gain.
The 50 ns overload recovery in a multiplexed channel
Overload recovery matters when the input multiplexer switches from a large signal channel to a small one. A 50 ns recovery means the amplifier is usable again well within the acquisition window of a 1 MSPS converter, so the first conversion after a channel change is not corrupted. An amplifier without a specified recovery figure can take microseconds, and the error appears as a fixed offset on the first sample of every channel change, which is easy to mistake for a multiplexer fault. For a 16 bit system this is the second specification to check after noise, and it is rarely on the front page of competing parts.
Six point seven milliamps and the 24 V option
At 6.7 mA and a 5 V supply the part dissipates 34 mW, which is low for this bandwidth and is the reason it is described as the lowest quiescent current in its class. The supply range up to plus and minus 12 V means it can sit in an industrial front end with a 20 V signal, and at 24 V total the same 6.7 mA becomes 160 mW in an 8-lead package, which is still manageable but changes the thermal calculation for a dense board. The MSOP body has a higher thermal resistance than the SOIC, so the same dissipation produces a larger junction rise. The amplifier cross reference page groups the low noise high speed parts by noise and supply range, the pin to pin cross reference page lists the bodies that swap, and the lead time page shows the availability position.