AD7689BCPZ
Product Specifications
AD7689BCPZ
Analog Devices — Analog & Mixed-Signal ICs
Product Overview
The AD7689BCPZ 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
AD7689BCPZ: Eight 16-Bit Channels at 250 kSPS, and the Scan Timing That Comes With Them
The AD7689BCPZ is the eight-channel member of the same 16-bit, 250 kSPS family as the AD7682, and the extra channels are the whole reason to choose it. Both parts share the die architecture, the internal 2.5 V and 4.096 V reference options, the SPI compatible interface and the LFCSP body, so the difference on a drawing is four more analogue inputs and four more pins of routing. What changes for the system is not the accuracy, it is the time budget: eight channels at 250 kSPS means a complete scan every 32 microseconds, and everything downstream of the converter has to keep up with that rate or the channels stop being simultaneous in any practical sense.
Eight channels, one converter, and what a scan costs
A sequencer inside the part walks the enabled channels in order, so the host reads eight results without addressing each one, and the conversion for each channel happens in its own 4 microsecond window. Eight channels therefore occupy 32 microseconds of the frame, which is exactly the 250 kSPS throughput, and any host that wants to do something else with the bus in that window is fighting the converter. The practical consequence is that the serial clock has to be fast enough to move 16 bits per channel inside the window, and that an interrupt driven host needs to service the frame at 31.25 kHz. Where the host cannot sustain that, the honest answer is to slow the converter down rather than to drop channels, because the filter that follows will average the missing samples into a distortion term.
The reference question is identical to the four-channel part
The internal reference is selectable at 2.5 V or 4.096 V, and the minimum supply follows the selection: at least 3 V for the 2.5 V setting and at least 4.5 V for the 4.096 V setting. A 3.3 V system therefore runs a 2.5 V full scale, where one code is 38.1 uV, and a 5 V system can run 4.096 V, where one code is 62.5 uV. The signal to noise and distortion figure of about 91 dB at a 5 V reference and a 20 kHz input is the same for both channel counts, because it is the same converter core, and it corresponds to 14.8 effective bits. The 16-bit resolution is a resolution statement and not an accuracy statement, and a system that needs 16 bits of real accuracy has to bring its own reference, its own driver and its own grounding.
| Parameter | Value on this order code | What it decides |
|---|---|---|
| Resolution | 16 bits, no missing codes | One code is 38.1 uV at 2.5 V full scale |
| Throughput | 250 kSPS | A full eight-channel scan every 32 microseconds |
| Channels | 8 single ended, AD7689 | Four more inputs than the AD7682 on the same core |
| Conversion window | 4 microseconds per channel | The serial frame has to fit inside the window |
| Internal reference | 2.5 V or 4.096 V | 3 V and 4.5 V minimum supply respectively |
| SINAD | About 91 dB at 5 V reference | 14.8 effective bits at a 20 kHz input |
| Standby current | 50 nA at 5 V | Scans can be gated to the measurement window |
| Package | LFCSP, 4 mm by 4 mm | Same body as the four-channel part |
| Interface | SPI compatible with daisy chain | Two parts give sixteen channels on one bus |
Where eight channels earn their place
Three-phase power monitoring with neutral and a spare, eight channel strain or pressure arrays, multi-cell battery measurement in a rack, and instrumentation front ends that need a common time base across many slow signals all fit this part. The common thread is that the signals change slowly compared with the conversion rate and the value comes from sampling them at known instants rather than from bandwidth. A system that needs 250 kSPS on one channel and nothing else should not pay for eight channels, and a system that needs eight channels at 250 kSPS each has to accept that the total data rate is 4 Mbps and plan the host around it.
Board level notes that apply to both channel counts
The reference bypass capacitor belongs at the pin and not at the edge of the board, because the charge taken on each conversion comes out of that capacitor. The analogue inputs need a driver that settles inside the 4 microsecond window from the source impedance that will actually be connected, not from the impedance on the bench, and a series resistor added to protect the input lengthens the settling and shows up as gain error. Where this converter has to replace another part on an existing board, the ADC and DAC cross reference lists the family pairs by resolution and channel count, the ADI part second source page sets out what has to be compared before a substitution, and the one stop BOM sourcing page covers buying the converter and its driver on one order.