DAC ICs · Resolution and channel count, linearity (INL, DNL) and total error, reference, output range, gain and drive, settling time and glitch, interface, supply and power, temperature and package
A digital-to-analog converter IC turns digital codes from a processor into precise voltages or currents for setting bias levels, controlling actuators, generating test signals and calibrating systems. Settle three things before choosing one: the resolution, channel count and accuracy the outputs need, the reference, output range, drive and speed, and the interface, supply, power, temperature range and package.
Precision voltage- and current-output DAC ICs, single and multichannel, for industrial automation, data acquisition, communications and test equipment. Audio DACs and codecs, ADCs, digital potentiometers and DAC output modules are outside this page; ADC chips and audio codecs have their own pages.
01
TI's DACx0508 datasheet describes octal (eight-channel) voltage-output DACs at 16 bits (DAC80508), 14 bits (DAC70508) and 12 bits (DAC60508), each with INL and DNL of ±0.5 LSB typical and ±1 LSB maximum and specified monotonic at full resolution, and total unadjusted error of ±0.05% typical and ±0.1% of full-scale range maximum. [1]
Output settling from ¼ to ¾ scale and back takes 5 µs to ±2 LSB at 5.5 V supply and gain 2, with a 25 mV power-up glitch at zero scale. Outputs can reset to zero scale or midscale, and the DACx0508C versions add a clear function. [1]
TI's DAC7568/DAC8168/DAC8568 family is another octal voltage-output line at 12, 14 and 16 bits, listing relative accuracy of 0.3 LSB, 1 LSB and 4 LSB INL respectively, glitch energy of 0.1 nV-s and monotonic operation over the full temperature range. [2]
For 4–20 mA loops, TI's DACx760 single-channel 12- and 16-bit DACs program a current output of 4–20 mA, 0–20 mA or 0–24 mA, or a voltage output of 0–5 V, 0–10 V, ±5 V or ±10 V with 10% overrange, both able to run simultaneously from one data register, with ±0.1% FSR total unadjusted error maximum and ±1 LSB DNL maximum. [3]
Editorial calculation: with the 2.5 V internal reference at gain 2 the full-scale range is 5 V, so one LSB at 16 bits is 5 ÷ 65,536 ≈ 76 µV, and ±0.1% FSR total unadjusted error is ±5 mV. [1]
Give the supplier the resolution and channel count, required INL, DNL and total error, reference (internal or external) and drift, output range and gain, load current, settling time, interface and logic voltage, supply and power budget, temperature range, package and quantity.
02
The DACx0508 integrates a 2.5 V precision reference with ±5 mV initial accuracy and 2 ppm/°C typical drift (DAC80508), offers user-selectable gain of 2, 1 or ½, and drives up to 20 mA with 0.5 V from the supply rails. It runs from 2.7–5.5 V at 0.6 mA per channel at 5.5 V over −40 to 125 °C, with a 50 MHz SPI-compatible interface for 1.7–5.5 V logic, daisy-chaining and CRC error checking. [1]
It is available in a 3 × 3 mm 16-pin WQFN or 2.4 × 2.4 mm 16-pin DSBGA, rated ±3,000 V HBM and ±1,000 V CDM ESD, for optical networking, wireless infrastructure, industrial automation and data acquisition. [1]
The DAC8568 family's 2.5 V internal reference is disabled by default, with 0.004% typical initial accuracy and 2 ppm/°C typical (5 ppm/°C maximum) drift; it draws 1.25 mA at 5 V including the reference, sinks or sources 20 mA and runs from 2.7–5.5 V over −40 to 125 °C with a Schmitt-triggered serial interface up to 50 MHz. [2]
The DACx760 runs from a single 10–36 V supply or dual supplies up to ±18 V, has an internal 5 V reference (10 ppm/°C maximum) and a 4.6 V supply output, adds CRC check, watchdog timer, thermal, open-circuit and short-current alarms, can superimpose a HART signal on the current output, and comes in 6 × 6 mm VQFN-40 and HTSSOP-24 packages for analog output modules, PLCs and transmitters. [3]
Editorial calculation: on a 4–20 mA range, the 16-bit DAC8760's 16 mA span divides into 65,536 codes, about 0.24 µA per code (16 ÷ 65,536 mA). [3]
Editorial calculation: eight channels at 0.6 mA each from 5.5 V draw about 8 × 0.6 × 5.5 ≈ 26 mW before load current. [1]
03
Microchip's MCP4728 data sheet (DS22187E) describes a quad 12-bit voltage-output DAC with on-board EEPROM that stores DAC codes, configuration and I²C address bits, so outputs come up at saved settings after power-up. It has an internal 2.048V reference (output 0–2.048V at gain 1, 0–4.096V at gain 2) or uses VDD as reference, runs from 2.7V to 5.5V over −40°C to +125°C, settles in 6 µs typical, and talks I²C at 100 kbps, 400 kbps or 3.4 Mbps in a 10-lead MSOP. [4]
The two parts sit at different points. The MCP4728's EEPROM keeps set points through power cycles and its I²C address is programmable [4], which suits calibration and bias adjustment; the TI DACx0508 gives much tighter linearity and a lower-drift reference [1]. Choose on the error budget: ask each supplier for INL, gain error and reference drift over your temperature range, not only resolution.
Microchip lists its applications as set point or offset adjustment, sensor calibration, closed-loop servo control and bias adjustment for power amplifiers [4]. For 4–20 mA loop outputs, the TI DACx760 integrates the current output [3], which a voltage-output DAC like the MCP4728 does not; ask whether an external current stage is needed.
Editorial calculation: one LSB of a 12-bit DAC on a 4.096V range is 1 mV, so ±13 LSB maximum INL is about ±13 mV; one LSB of a 16-bit DAC on a 5 V range is about 76 µV, so ±1 LSB is about ±0.076 mV. [4] [1]
| Item | Microchip MCP4728 | TI DACx0508 (already cited) |
|---|---|---|
| Channels / resolution | 4 × 12-bit | 8 × 16/14/12-bit |
| INL | ±2 LSB typ., ±13 LSB max | ±0.5 LSB typ., ±1 LSB max |
| DNL | ±0.2 LSB typ., ±0.75 LSB max | ±0.5 LSB typ., ±1 LSB max |
| Gain error | 0.4% typ., ±1.25% of FSR max | Total unadjusted error ±0.1% FSR max |
| Internal reference | 2.048V (2.007–2.089V); 45 ppm/°C typ. (0 to +125°C) | 2.5 V ±5 mV; 2 ppm/°C typ. |
| Interface | I²C up to 3.4 Mbps | SPI 50 MHz |
| Non-volatile memory | EEPROM; 25 ms typ. write; 200 years retention at +25°C | — |
| Supply current | 800 µA typ., 1400 µA max (whole device) | 0.6 mA per channel at 5.5 V |
Values are from Microchip's MCP4728 data sheet DS22187E (text extracted from the PDF).
04
The companies in the manufacturer list below show this product on their own websites. Confirm the legal seller, factory and scope for your order.
The reviewed data sheets cover three kinds: compact octal voltage-output DACs with integrated references and SPI (DACx0508, DAC8568 family), and single-channel industrial DACs with programmable 4–20 mA current and bipolar voltage outputs plus diagnostics (DACx760). [1] [2] [3]
Ask each supplier for INL, DNL, total error, reference drift, settling time, output ranges and drive, supply range and current, diagnostics and package options, and for evaluation boards or software. [1]
05
This is a functional reference built from the source cited; it is not a full specification.
| Block | What it controls | Fix in the order |
|---|---|---|
| Resistive DAC network | Resolution; linearity. [1] | Bits; INL/DNL. |
| Internal reference | Accuracy; drift. [1] | V; ppm/°C. |
| Output buffer | Drive; settling. [1] | mA; µs. |
| Gain stage | Output range. [1] | ×2; ×1; ×½. |
| Serial interface | Host link. [1] | SPI MHz; CRC. |
| Power-down / reset | Start-up state. [1] | Zero; midscale. |
06
A DAC quotation needs values for your design. The table explains each field, gives published values where the reviewed source states them, and says what to obtain from the supplier.
| Field | What it means and published values | Ask the supplier for |
|---|---|---|
| Resolution | 16, 14 or 12 bits [1]; 12-bit quad with EEPROM in one I²C part. [4] | bits. |
| Sampling Rate | Update set by 5 µs settling and SPI. [1] | Updates/s. |
| Output Voltage Range | Reference × gain (2, 1, ½). [1] | V. |
| Output Current Range | Up to 20 mA drive. [1] Industrial DACs program 4–20, 0–20 or 0–24 mA. [3] | mA. |
| Integral Nonlinearity (INL) | ±1 LSB max [1]; ±13 LSB max on a low-cost 12-bit part. [4] | LSB. |
| Differential Nonlinearity (DNL) | ±1 LSB max; monotonic. [1] | LSB. |
| Supply Voltage | 2.7–5.5 V. [1] | V. |
| Power Consumption | 0.6 mA/channel at 5.5 V. [1] | mW. |
| Operating Temperature Range | −40 to 125 °C. [1] | Range. |
| Storage Temperature Range | Absolute maximum. | Range. |
| Relative Humidity | Board-level IC. | Per system. |
| Ingress Protection (IP) Rating | Board-level IC. | N/A. |
| Package Type | 16-pin WQFN 3 × 3; DSBGA 2.4 × 2.4 mm. [1] | Package. |
| Weight | Package mass. | g. |
07
Paste it into your enquiry below or into an email to a maker, then replace every bracket.
Channels/bits: [count; bits] Accuracy: [INL; DNL; TUE] Reference: [internal; external; ppm/°C] Output: [V range; gain; mA] Speed: [settling µs] Interface: [SPI MHz; logic V] Supply/power/temp: [V; mW; °C] Package/quantity: [type; units]
These companies list this product on their own websites.
Its website lists digital-to-analog converters, including the SGM5347 and SGM5348.
Product scope checked 2026-10-12
Its website lists precision DACs, including the TPC2161.
Product scope checked 2026-10-12
Explore company information and manufacturing capabilities, then discuss the scope of your project with a relevant supplier.
TI's DAC80508 lists ±1 LSB maximum INL and DNL at 16 bits and ±0.1% FSR maximum total unadjusted error. [1]
The DACx0508 integrates a 2.5 V reference with ±5 mV initial accuracy and 2 ppm/°C typical drift. [1]
The DACx0508 settles from ¼ to ¾ scale in 5 µs to ±2 LSB. [1]
The DACx0508 uses a 50 MHz SPI-compatible interface with daisy-chaining and CRC. [1]
Industrial DACs such as the DACx760 program 4–20 mA, 0–20 mA or 0–24 mA current outputs, with loop diagnostics and optional HART. [3]
Yes. Microchip's MCP4728 quad 12-bit DAC stores its DAC codes, configuration and I²C address in on-board EEPROM, with 200 years data retention at 25 °C, so outputs return to the saved settings at power-up. [4]
Sources checked 2026-10-12. Texas Instruments' DACx0508 datasheet is used to show what a DAC specification covers; values for your order must come from the supplier's current datasheet.
Ask for use case, specification boundaries, supplier type, and RFQ preparation information for this product.
Compare manufacturer profiles with relevant product and process capability.