AD7924BRUZ
Product Specifications
AD7924BRUZ
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The AD7924BRUZ 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
AD7924BRUZ: Twelve Bits for a Microsecond per Channel, and 70 dB of Signal to Noise at 50 kHz
The AD7924 is the 12-bit member of the four channel 1 MSPS family, sharing its 16-lead TSSOP outline, its 2.7 V to 5.25 V single supply, its sequencer and its separate VDRIVE pin with the 10-bit AD7914 and the 8-bit AD7904. It is the member to pick when monitoring has to become measurement: 4096 codes instead of 1024, with the same conversion timing, the same power per conversion and the same interface. The signal to noise ratio is specified at 70 dB with a 50 kHz input, and the input bandwidth is quoted as greater than 8 MHz.
Seventy decibels is eleven point three effective bits
A 12-bit converter has a theoretical signal to noise ratio of 6.02 times 12 plus 1.76, which is 74 dB. The specified 70 dB at 50 kHz converts to an effective number of bits of 11.3, so at the top of the useful band the part is about 0.7 bits short of its word length. That is a normal and honest result for a 1 MSPS successive approximation converter, and it is the number to compare against a requirement rather than the 12-bit word. On a 2.5 V reference the code size is 610 uV, so a signal of 5 mV occupies eight codes and 70 dB of dynamic range is what allows a small signal to be seen on top of a large one.
One microsecond per channel, four channels, no pipeline
At 1 MSPS a conversion takes one microsecond, so a four channel sequence completes in 4 microseconds and repeats without a host write. There is no pipeline delay, which means the result returned on chip select belongs to the channel just addressed rather than to the previous one, and a control loop that changes channel between conversions does not have to track an offset. The conversion time is set by the serial clock, so the same part runs slower and cooler with a slower port, and the power figures of 6 mW at 3 V and 13.5 mW at 5 V are the maximum at the full rate.
| Parameter | Value | What it decides |
|---|---|---|
| Resolution | 12 bits | 610 uV per code with a 2.5 V reference |
| Channels | Four single ended with sequencer | Repeat scan without host writes |
| Throughput | 1 MSPS maximum | Four channels in 4 microseconds |
| Signal to noise | 70 dB at 50 kHz input | 11.3 effective bits at the top of the band |
| Input bandwidth | Greater than 8 MHz | Undersampling is possible with care |
| Supply | 2.7 V to 5.25 V single | One code covers 3 V and 5 V systems |
| Power at 1 MSPS | 6 mW at 3 V, 13.5 mW at 5 V | 2 mA and 2.7 mA maximum |
| Shutdown | 0.5 uA maximum | Idle current below the leakage of most boards |
| Aperture | 8.2 ns delay, 1.6 ns jitter | Sampling instant defined for a fast signal |
| Package | 16-lead TSSOP | Same footprint as the 10-bit and 8-bit members |
Between 10 and 12 bits, the socket does not change
The three members of this family are pin compatible and differ in resolution, so a board can be populated with the 10-bit part for an early build and the 12-bit part for the production one without a board change. The reverse move is the one to watch: a board qualified with the 12-bit part and then populated with the 10-bit sibling loses a factor of four in resolution and returns codes that look plausible but are coarser, which is exactly the kind of substitution that passes a functional test and fails a field requirement. Any such change should go through the substitution checks rather than being treated as a like for like part.
Where the part fits
Four channel monitoring in motor control, power supplies and instrumentation, where a megasample per second and 12 bits are both needed inside one control period. The sequencer makes the host a receiver, the VDRIVE pin removes level shifters, and the shutdown modes make a burst sampling instrument practical. Where 10 bits is enough, the AD7914 in the same socket is cheaper; where more channels are needed, the eight channel members of the same converter generation are the next step. For sourcing, the ADC and DAC cross reference page lists the family, the how to verify a drop in replacement page covers the resolution substitution checks, and the one stop BOM sourcing page covers consolidating the family across products.
Six hundred microvolts per code in a current sense application
The 12-bit code size of 610 uV at a 2.5 V reference looks small until it is referred to a shunt. Across a 10 milliohm shunt one code is 61 mA, so a 20 A load produces 200 mV and uses 328 of the 4096 codes, which wastes nine of the twelve bits. Across a 50 milliohm shunt the same 20 A produces 1 V and uses 1639 codes, which is a usable measurement with 12 bits of resolution and 1.2 W of shunt loss. That trade, between shunt loss and code utilisation, is the real selection exercise in a current sense channel, and the converter resolution is the constraint that sets it rather than the other way round.