SAR vs Delta-Sigma vs Pipeline vs Flash ADC: Architectures Compared

An analog-to-digital converter (ADC) can be built in several ways, and each architecture trades speed against resolution. Flash ADCs are the fastest but only a few bits. SAR ADCs are small, low-power and the general-purpose choice for medium speed. Pipeline ADCs reach high speed with medium-to-high resolution. Delta-sigma ADCs trade speed for the highest resolution, which is ideal for audio and precision sensors.

AspectFlashSARPipelineDelta-sigma
How it worksCompares the input with all thresholds at once using 2ᴺ − 1 comparatorsBinary search: one comparator and an internal DAC decide one bit per clock cycleA cascade of stages, each resolving a few bits and passing an amplified residue to the nextOversamples with a coarse modulator that shapes noise, then a digital filter produces the output
Typical resolution4 to 8 bits8 to 18 bits8 to 16 bits16 to 24 bits or more
Typical speedFastest: up to several GS/skS/s up to tens or hundreds of MS/s10 MS/s up to hundreds of MS/sSlowest: up to about 1 MS/s
LatencyOne clockN clocks for N bitsSeveral clocksLong, because of the digital filter
Power and areaGrows rapidly with resolution; large and power-hungryVery low power and smallModerate to highModerate; the digital filter takes area
Anti-aliasing filterNeeds a sharp filterNeeds a sharp filterNeeds a sharp filterRelaxed, thanks to oversampling
Typical usesHigh-speed oscilloscopes, optical and wireless receivers, sub-ADC inside pipelinesMicrocontroller ADCs, IoT sensors, general data acquisitionVideo, communications and imaging systemsAudio codecs, weigh scales, precision measurement

Figures are typical orders of magnitude and vary by process generation and product.

When to use which

Flash
When you need extreme speed and only a few bits, or a fast building block inside another architecture.
SAR
The default for medium speed and medium-to-high resolution, especially when power and size matter.
Pipeline
When you need both high speed and good resolution, as in video and radio receivers.
Delta-sigma
When resolution matters more than speed: audio, scales and low-frequency sensors.

FAQ

Which ADC architecture is the fastest?

The flash ADC. It compares the input with all thresholds at once and can reach several gigasamples per second, but only with a few bits of resolution.

Why is a delta-sigma ADC so accurate?

It oversamples with a coarse modulator and pushes quantization noise to high frequencies, where a digital filter removes it. This gives 16 to 24 bits of resolution at low speed.

What is a SAR ADC?

A successive-approximation ADC finds the input value by binary search, using one comparator and an internal DAC and deciding one bit per clock. It is small, low-power and common in microcontrollers.

Why does a flash ADC need so many comparators?

An N-bit flash ADC needs 2ᴺ − 1 comparators, for example 255 for 8 bits, so its size and power grow rapidly with resolution.

Check your understanding

Which is a characteristic of a flash ADC?

  1. AIt converts in one step using many comparators in parallel, so it is fast, but circuit size grows rapidly with resolution
  2. BIt compares one bit at a time, so it is slow but small
  3. CIt integrates the input voltage for a long time, so it is accurate but slow
Show answer and explanation

Answer: A. It converts in one step using many comparators in parallel, so it is fast, but circuit size grows rapidly with resolution

A flash ADC compares the input against every threshold at once using 2ᴺ−1 comparators, so conversion finishes in a single comparison—the fastest architecture (several GS/s). But N bits need 2ᴺ−1 comparators: 255 at 8 bits and 1023 at 10, so area, power and input capacitance grow rapidly. Flash converters are therefore used for ultra-high-speed applications at about 6–8 bits and as building blocks in pipelined and subranging ADCs.

Which correctly describes the operation of a successive-approximation (SAR) ADC?

  1. AIt integrates the input voltage for a fixed time and measures the discharge time
  2. BStarting from the most significant bit, it compares the input voltage with the output of an internal DAC and decides one bit at a time
  3. CIt achieves high resolution through oversampling and noise shaping
Show answer and explanation

Answer: B. Starting from the most significant bit, it compares the input voltage with the output of an internal DAC and decides one bit at a time

A successive-approximation (SAR) ADC compares the input with an internal DAC output starting from the most significant bit, settling one bit per step like a binary search. An N-bit conversion takes N comparisons and needs only one comparator. With a good balance of speed (kS/s to tens of MS/s), resolution (8–18 bits) and power, it dominates microcontroller ADCs. Integration is used by dual-slope converters, and oversampling by ΔΣ converters.

Which is a characteristic of a delta-sigma (ΔΣ) ADC?

  1. AIt readily achieves high resolution through oversampling and noise shaping
  2. BIt is the fastest of all architectures
  3. CIt requires 2ᴺ−1 comparators for N-bit conversion
Show answer and explanation

Answer: A. It readily achieves high resolution through oversampling and noise shaping

A ΔΣ ADC samples far above the signal bandwidth (oversampling), uses a feedback loop to push quantization noise to high frequencies (noise shaping), and removes it with a digital filter. Even a coarse 1-bit quantizer can yield over 20 bits of resolution at low frequencies, with little dependence on analog component precision. It is used in audio, instrumentation and sensor readout, but its latency makes it unsuitable for high-speed uses.

  1. How many comparators does a 3-bit flash ADC need?
  2. How many comparisons does an N-bit successive-approximation ADC basically perform per conversion?
  3. In a delta-sigma ADC, which digital circuit converts the fast, coarse quantized output into high-resolution data at a lower sampling rate?

Practice A/D and D/A conversion questions

Key terms explained

Successive-approximation ADC
An ADC that uses an internal DAC and comparator to determine one bit at a time, starting from the most significant. It balances speed, accuracy and power well.
Delta-sigma ADC
An ADC that oversamples at a high rate, pushes quantization noise to high frequencies and then filters it out. Used in high-resolution applications such as audio and measurement.
Flash ADC
The fastest type of ADC, which compares the input against many comparators at once. N bits need 2ᴺ − 1 comparators, so the circuit grows rapidly with resolution.
Pipelined ADC
An ADC that splits conversion into stages, each resolving a few bits and passing the residue to the next, like an assembly line. It combines high speed with medium to high resolution.

Related comparisons