ADS62C17
Product Specifications
ADS62C17
Texas Instruments — Power Management & Signal Chain
Product Overview
The ADS62C17 from Texas Instruments is a power management / signal chain IC delivering high efficiency, precision regulation, and robust protection. Texas Instruments (TI) is a global leader in analog ICs, power management solutions, and embedded processors.
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-efficiency power conversion with low quiescent current
- Wide input voltage range with precision output regulation
- Integrated over-current, over-temperature, and short-circuit protection
- Small footprint packaging for space-constrained designs
- AEC-Q100 qualified options available for automotive applications
- Flexible sequencing and soft-start control
Key Specifications
| Input Voltage | V |
| Output Voltage | V |
| Output Current | A |
| Switching Frequency | kHz |
| Efficiency | % |
| Quiescent Current | µA |
| Topology | — |
| Package | — |
| Operating Temp | °C |
| Protection | — |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
ADS62C17: dual 11 bit 200 MSPS converter with SNRBoost for wideband receivers
ADS62C17 is a two channel 11 bit pipeline converter in a 64 pin QFN body, built for multi carrier base station and wideband instrument receivers where the signal sits in a band far below Nyquist in a QFN body of 64 pins. The ADS62C17 samples both channels at up to 200 MSPS and puts the samples out on either a double data rate LVDS bus or a parallel CMOS bus, so the same part serves a narrowband FPGA capture and a wide parallel interface. The block that makes it different from a plain 11 bit converter is SNRBoost, which shapes the quantisation noise away from a chosen band. The vendor states 79.8 dBFS of SNR at a 125 MHz intermediate frequency inside a 20 MHz band, and 90 dBc of spurious free dynamic range at a 10 MHz input. Total power is 1.1 W at 200 MSPS on both channels.
Order code and package
| Order code | ADS62C17 |
| Channels | two, sampled simultaneously on one die |
| Resolution | 11 bit, no missing codes |
| Maximum sample rate | 200 MSPS per channel |
| Output interface | DDR LVDS or parallel CMOS, selected in the serial register |
| Input clock | differential, amplitude down to 400 mV peak to peak |
| Reference | internal or external |
| Package | 64 pin QFN, 9 mm by 9 mm |
| Analog supply | 3.3 V |
| Total power | 1.1 W at 200 MSPS |
| Crosstalk | 90 dB between channels |
| Document | SLAS631A, April 2009, revised July 2009 |
Electrical parameters that decide the application
| SNR in band | 79.8 dBFS at a 125 MHz intermediate frequency inside a 20 MHz band |
| SFDR | 90 dBc at a 10 MHz input |
| Quantisation floor | about 68 dBFS for 11 bit over the full Nyquist zone |
| Programmable gain | up to 6 dB, traded against SNR and SFDR |
| Gain step | 0.001 dB, used to match the two channels |
| DC offset correction | on chip, per channel |
| Power per sample | about 2.75 nJ for both channels at 200 MSPS |
| Channel matching | both channels share one sample clock buffer and one die |
| Clock duty | 50 percent, 400 mV peak to peak minimum amplitude |
| Temperature | minus 40 to 85 C |
| Standby | register controlled power down |
| Interface loading | the 200 MSPS, 11 bit stream is 2.2 Gbps per channel |
Points to check before this part goes on a BOM
SNRBoost is a band limited promise, not a Nyquist one. The 79.8 dBFS figure is quoted for a 20 MHz band around a 125 MHz intermediate frequency, and the technique moves quantisation noise out of that window. An application that needs the full first Nyquist zone keeps the plain 11 bit floor, which is 6.02 times 11 plus 1.76, or about 68 dBFS, and no register setting changes that.
Eleven bits is the ceiling on dynamic range and the reason this part is cheaper than a 14 bit converter. Set beside a 14 bit part on the same board, the difference in the quantisation floor is 18 dB, so a receiver that must resolve a small signal next to a large blocker at a different frequency should be checked against the SFDR number rather than the SNR number.
The data rate per channel is the interface problem. 200 MSPS at 11 bit is 2.2 Gbps, and the capture side has to take that on a source synchronous DDR bus or on a parallel CMOS bus running at 200 MHz. The CMOS option moves 22 lines plus clock, which is a package and routing decision that belongs in the board review before the part is committed.
Gain of up to 6 dB is a trade, not a fix. The coarse and fine gain blocks exist to improve SFDR when the full scale input is smaller than the converter range, and the fine step of 0.001 dB is there to match the two channels against each other. Turning gain up to improve a small signal also brings the noise floor up with it.
The two channels share one die, one clock buffer and one reference, which is what makes the part useful for a phase coherent pair. Crosstalk of 90 dB is the isolation between them, so a 1 V signal in one channel couples about 32 uV into the other, which matters when one channel watches a strong local transmitter and the other watches a weak return.
Power of 1.1 W in a 9 mm by 9 mm body is 2.75 nJ for each pair of samples. That number is what decides whether the part fits a passive cooled enclosure, and it also sets the thermal pad requirement on the board beneath the body.
The serial register holds the output format, the gain and the offset correction. On a build where the register is loaded by firmware at start up, a failed load shows up as an interface that never locks rather than as a data error, so a second source check should include the register sequence, not only the package.
Where this order code fits
This order code belongs on multi carrier cellular receive paths, software defined radio front ends, satellite receivers and spectrum monitoring equipment where a 20 MHz to 40 MHz slice of spectrum sits at an intermediate frequency and the capture converter does not need to resolve the whole first Nyquist zone. Instrument makers use it where two channels must be phase matched, since both channels sit on one die and share one clock buffer. Boards comparing it with a neighbouring part start from our ADC DAC Cross Reference page, and boards replacing a part that has gone end of life start from the Pin to Pin Cross Reference page.
Questions buyers ask
Is this part a drop in for a 14 bit converter in the same package family? No. The pin count and the body may match a neighbouring family member, but the resolution, the sample rate and the register map have to be compared field by field before a substitution is claimed.
Does SNRBoost work at any input frequency? It works where the wanted signal occupies less than the first Nyquist zone. The published figure assumes a 20 MHz band, and a full band application sees the plain 11 bit floor instead.
What decides the capture side? The 2.2 Gbps per channel figure. Either the FPGA takes a DDR LVDS bus at that rate or it takes 22 CMOS lines plus clock, and the choice is a pin count and timing question rather than a software setting.
SUPERB supplies this order code against the exact ordering string, with date codes and packing stated on the offer, and pairs it with the Obsolete RF Part Replacement page when a board needs a second source that keeps the same interface timing.