Optical detection performance is a critical factor in fiber-optic communication, laser ranging, LiDAR, scientific instrumentation, and low-light imaging. As optical systems become more sensitive and operate across wider dynamic ranges, photodetectors must distinguish weak optical signals while maintaining suitable noise, bandwidth, and timing performance.
Avalanche Photodiodes (APDs) and Single-Photon Avalanche Diodes (SPADs) both use impact ionization to amplify photo-generated carriers. However, their operating conditions and output characteristics are fundamentally different. A conventional APD operates below its breakdown voltage in linear mode, producing an amplified electrical signal related to incident optical power. A SPAD operates above breakdown in Geiger mode, allowing an individual detected photon to trigger a macroscopic electrical pulse.
This distinction makes APDs suitable for analog optical signal detection, while SPADs are designed for photon-counting applications and precise photon-arrival timing. Choosing between them requires evaluating optical power, background illumination, detection probability, noise, dead time, and the requirements of the complete receiver system.
1. What Is an Avalanche Photodiode (APD)?
An Avalanche Photodiode is a semiconductor photodetector that uses internal avalanche multiplication to increase the electrical signal generated by incident light.
When photons are absorbed in the photodiode's active region, they generate electron-hole pairs. Under a sufficiently strong reverse electric field, these carriers gain energy and can generate additional electron-hole pairs through impact ionization.
This multiplication process increases the output photocurrent beyond what would be produced by a conventional PIN photodiode under the same optical input.
A conventional APD operates below its avalanche breakdown voltage. In this region, the multiplication process remains controlled, and the output current is approximately proportional to incident optical power.
Important APD characteristics include:
Internal gain: Avalanche multiplication increases the electrical response to weak optical signals.
Analog output: The output current carries information about the incident optical intensity.
Finite multiplication gain: Gain is controlled through the reverse-bias operating voltage.
Excess noise: Statistical variations in the avalanche multiplication process introduce additional noise.
High-speed detection: APDs can support high-speed optical receivers when the detector, package, and readout circuit are properly designed.
APDs are widely used in optical communication receivers, optical measurement systems, laser ranging, and instrumentation where internal gain improves detection sensitivity.
2. What Is a Single-Photon Avalanche Diode (SPAD)?
A Single-Photon Avalanche Diode is a photodetector designed to detect individual photons. It operates in Geiger mode, with a reverse-bias voltage above the diode's breakdown voltage.
In this operating condition, a single photo-generated carrier can initiate a self-sustaining avalanche. The resulting electrical pulse is large enough to be detected by the readout circuit, even when the original optical event involved only one photon.
The avalanche must then be quenched so that the detector can return to its photon-sensitive state. This is achieved through a passive or active quenching circuit.
Unlike a linear-mode APD, the amplitude of a SPAD's avalanche pulse does not provide a proportional measurement of the incident optical energy. The event primarily indicates that a photon-triggered avalanche occurred.
Key SPAD characteristics include:
Single-photon sensitivity: A single detected photon can trigger a measurable electrical event.
Geiger-mode operation: The reverse-bias voltage exceeds the breakdown voltage.
Digital-like output: Detection produces an electrical pulse representing an event.
Photon counting: Individual detection events can be counted over time.
Timing information: The pulse arrival time can be measured using suitable timing electronics.
Quenching and reset: The avalanche must be stopped before the detector can register another event.
SPADs are commonly used in time-of-flight sensing, fluorescence lifetime measurements, photon-counting systems, low-light imaging, and scientific instruments.
3. APD vs SPAD: Key Differences
| Technical Aspect | APD | SPAD |
|---|---|---|
| Full name | Avalanche Photodiode | Single-Photon Avalanche Diode |
| Operating mode | Linear mode | Geiger mode |
| Bias condition | Below breakdown voltage | Above breakdown voltage |
| Multiplication mechanism | Controlled avalanche multiplication | Self-sustaining avalanche triggered by a photon |
| Output signal | Analog photocurrent | Electrical detection pulse |
| Signal relationship | Output current is related to incident optical power | Pulse indicates a detected photon event |
| Typical gain behavior | Finite multiplication gain | Very large avalanche amplification; conventional linear gain is not the useful measurement |
| Single-photon detection | Generally not suitable in conventional linear mode | Designed for single-photon detection |
| Noise characteristics | Excess avalanche noise and readout noise | Dark counts, afterpulsing, and optical crosstalk in applicable architectures |
| Output measurement | Photocurrent or voltage | Photon-event count and arrival time |
| Quenching circuit | Not required for normal linear-mode operation | Required to stop and reset the avalanche |
| Dead time | Primarily determined by detector and receiver response | Recovery interval after each avalanche |
| High background illumination | Can handle analog optical signals within its operating range | Can experience high count rates and saturation |
| Timing measurement | Possible with suitable receiver electronics | Well suited to photon-arrival timing |
| Typical applications | Optical communications, analog detection, ranging | Photon counting, ToF, LiDAR, low-light imaging |
The distinction is primarily the operating mode rather than a completely different semiconductor detection principle. A SPAD can be understood as an avalanche photodiode operating in Geiger mode, specifically engineered for single-photon detection.
4. Operating Principle: Linear Mode vs Geiger Mode
APD: Linear-mode avalanche multiplication
In linear mode, the APD reverse-bias voltage is below the breakdown voltage. Photo-generated carriers enter the high-field multiplication region and generate secondary carriers through impact ionization.
The avalanche multiplication increases the photocurrent while remaining controlled. The resulting output retains a relationship with the incident optical signal.
A simplified relationship is:
Where:
IAPD: APD output photocurrent.
M: Avalanche multiplication factor.
R: Photodiode responsivity before avalanche multiplication.
Popt: Incident optical power.
The multiplication factor is finite and depends on the device structure, electric field, temperature, and operating voltage. Increasing the gain can improve sensitivity, but it also increases avalanche-related noise and can affect bandwidth.
SPAD: Geiger-mode avalanche detection
A SPAD operates with a reverse-bias voltage above breakdown. When a photon generates a carrier in the active region, the high electric field can trigger a self-sustaining avalanche.
The resulting current pulse is detected by the readout electronics. A quenching circuit then reduces the bias below the breakdown threshold, stopping the avalanche. After the detector is recharged, it becomes ready for another detection event.
The operating cycle can be summarized as:
Bias the SPAD above its breakdown voltage.
A photon generates a carrier in the active region.
The carrier triggers an avalanche.
The readout circuit detects the resulting electrical pulse.
The quenching circuit stops the avalanche.
The detector recovers and becomes ready for another event.
The output pulse is largely independent of the energy of the triggering photon. Therefore, SPAD systems generally count events rather than directly measuring optical intensity from individual pulse amplitudes.
5. APD Gain vs SPAD Photon Detection
Gain is an important specification for APDs, but its meaning differs from the detection process used by SPADs.
APD multiplication gain
APD gain describes the multiplication of the original photo-generated current through impact ionization. It is usually expressed as a finite multiplication factor.
Higher gain can improve the electrical signal level at the receiver, especially when the optical signal is weak. However, higher multiplication also introduces excess noise and may reduce the useful operating bandwidth.
The optimal gain is therefore a system-level design decision involving optical power, receiver noise, bandwidth, and temperature stability.
SPAD photon detection efficiency
SPAD performance is more commonly evaluated through Photon Detection Efficiency (PDE), which describes the probability that an incident photon produces a registered detection event.
PDE depends on several factors:
Photon absorption probability in the active region.
Carrier collection efficiency.
Probability that a carrier triggers an avalanche.
Optical wavelength.
Device architecture and operating conditions.
A higher PDE means that a greater proportion of incident photons can be registered. However, increasing the excess bias may also affect dark count rate, afterpulsing, and other noise characteristics.
APD multiplication gain and SPAD photon detection efficiency are different metrics. They should not be treated as interchangeable measures of detector sensitivity.
6. Noise Characteristics: Excess Noise vs Dark Counts
Noise is one of the most important factors when selecting between APD and SPAD technologies.
APD noise
APDs experience shot noise, thermal noise from the receiver electronics, and excess noise associated with the avalanche multiplication process.
The avalanche process is statistical. Different carriers can produce different multiplication histories, creating fluctuations in the output current. This is represented by the excess noise factor.
Important APD noise considerations include:
Shot noise associated with photocurrent and dark current.
Excess avalanche multiplication noise.
Thermal noise from the load and receiver circuitry.
Dark current and temperature-dependent leakage.
Noise-bandwidth trade-offs at higher multiplication gain.
SPAD noise
SPADs produce detection events even when no intended optical signal is present. These events are associated with dark counts, which can result from thermal generation and other carrier-generation mechanisms.
Additional noise-related effects include:
Dark Count Rate (DCR): The number of detection events registered in darkness over a given time.
Afterpulsing: Delayed detection events associated with carriers trapped during a previous avalanche and released later.
Optical crosstalk: Photons generated by an avalanche can trigger neighboring sensitive regions in some detector architectures.
Ambient-light background: Unwanted photons can generate additional counts and reduce the ability to distinguish the desired signal.
SPADs avoid the conventional linear-mode avalanche excess-noise mechanism in the same form as APDs, but they introduce event-based noise sources that must be controlled.
Consequently, neither detector is universally lower-noise. The appropriate choice depends on signal level, background illumination, readout architecture, and the required measurement method.
7. Bandwidth, Response Time, and Dead Time
APD and SPAD systems use different output formats, so their speed specifications are not always directly comparable.
APD bandwidth
APD bandwidth describes how quickly the detector and associated receiver electronics can respond to changes in incident optical power.
Important factors include:
Active-area capacitance.
Carrier transit time.
Multiplication-region design.
Reverse-bias voltage and avalanche gain.
Package parasitics and receiver impedance.
Transimpedance amplifier bandwidth.
For optical communication systems, the APD must provide sufficient bandwidth to support the electrical signal generated by the incoming optical data.
SPAD timing performance
SPADs are commonly evaluated using timing resolution, timing jitter, dead time, and maximum sustainable count rate.
Timing jitter describes the variation between the actual photon arrival time and the recorded detection time. It is important in applications that measure distance or event timing.
Dead time is the interval after an avalanche during which the detector is recovering and cannot register another independent event normally.
The maximum usable count rate is influenced by the dead time, quenching circuit, detector recovery behavior, and event-processing electronics. A high photon arrival rate can cause missed events or count-rate distortion.
| Performance Parameter | APD | SPAD |
|---|---|---|
| Primary speed metric | Electrical bandwidth and response time | Timing jitter, dead time, and count rate |
| Signal response | Continuous analog response | Discrete detection events |
| High-frequency operation | Limited by detector and receiver bandwidth | Limited by event recovery and timing electronics |
| Repeated event handling | Can track changing optical intensity within bandwidth | Requires recovery after each avalanche |
| Measurement focus | Optical waveform or intensity | Photon arrival time and event count |
8. Optical Sensitivity and Dynamic Range
APDs and SPADs are both used for weak-light detection, but their sensitivity advantages depend on the measurement conditions.
APDs amplify the photocurrent and can provide useful analog information over a range of optical input powers. Their performance is influenced by avalanche gain, excess noise, receiver noise, and saturation limits.
SPADs can register individual photons and are particularly valuable when the available optical signal is extremely weak. However, they have a finite count-rate capability and can become saturated under strong background illumination or high photon arrival rates.
The dynamic range of a SPAD system is affected by its detection probability, dark count rate, dead time, pixel architecture, and event-processing capability.
For example, a SPAD may perform well in a low-light time-of-flight measurement but become less effective when intense ambient light produces a large number of unwanted events. A linear-mode APD may be more appropriate for an analog receiver operating with stronger optical signals and a suitable noise budget.
The correct detector choice should therefore consider both minimum detectable signal and maximum expected optical input.
9. Applications of APD and SPAD
APD applications
APDs are widely used in optical systems that need internal gain while retaining an analog relationship between optical input and electrical output.
Fiber-optic communication: APDs can be used in optical receivers where additional detector gain improves sensitivity.
Laser ranging: Linear-mode APDs detect reflected laser signals and provide analog measurement information.
Optical instrumentation: APDs support weak optical signal measurements.
Industrial sensing: APDs are used in optical measurement and detection systems.
Scientific measurement: APDs can detect low-level optical signals in laboratory instruments.
SPAD applications
SPADs are designed for applications where individual photon events or precise arrival times are important.
Direct Time-of-Flight (dToF): Measures photon return timing to estimate distance.
LiDAR: SPAD arrays can detect weak reflected laser signals for depth sensing.
Fluorescence lifetime imaging: Photon timing supports measurements of fluorescence decay.
Low-light imaging: SPAD arrays provide photon-counting imaging capabilities.
Quantum optics: SPADs can register individual photon events in experimental systems.
Time-correlated single-photon counting (TCSPC): Measures photon arrival times relative to a reference event.
10. APD vs SPAD in LiDAR and Time-of-Flight Systems
LiDAR and time-of-flight systems use reflected optical signals to estimate the distance to an object. Both APD and SPAD technologies can be used, but they support different detection strategies.
APD-based LiDAR
A linear-mode APD receives reflected laser light and converts it into an amplified analog electrical signal. The receiver processes the waveform to identify the return signal and determine its timing or intensity characteristics.
This architecture can be useful when the reflected signal is sufficiently strong and the system needs analog waveform information.
SPAD-based LiDAR
A SPAD-based system detects individual photon events and records their arrival times. With suitable timing electronics, these events can be used to construct a time-of-flight histogram or estimate the return time.
SPAD arrays can support high-density photon-counting imaging, particularly when combined with integrated timing and signal-processing circuits.
| LiDAR Consideration | APD-Based System | SPAD-Based System |
|---|---|---|
| Detection approach | Analog reflected-signal detection | Photon-event detection |
| Low-light capability | Improved through avalanche gain | Can detect individual photons |
| Signal processing | Waveform amplification and analysis | Photon counting and timing analysis |
| Ambient-light response | Depends on analog noise and receiver dynamic range | Background photons increase count rate and can cause saturation |
| Timing measurement | Based on analog waveform processing | Based on detected photon-event timing |
| Recovery behavior | Determined by detector and analog receiver response | Limited by avalanche quenching and reset time |
| Array integration | Available in specialized detector arrays | Well suited to integrated photon-counting pixel arrays |
SPAD technology is particularly attractive for photon-counting depth sensing, while APDs remain useful for analog ranging architectures. Actual performance depends on laser power, receiver optics, target reflectivity, ambient illumination, detector characteristics, and signal-processing methods.
11. APD vs SPAD: Selection Guide
Choosing between APD and SPAD technology begins with the type of optical information the system needs to measure.
| Application Requirement | Potentially Suitable Detector | Reason |
|---|---|---|
| Analog optical intensity measurement | APD | Output photocurrent retains a relationship with incident optical power |
| Single-photon detection | SPAD | Geiger-mode avalanche can be triggered by an individual photon |
| High-speed fiber-optic receiver | APD | Supports analog optical signal detection with internal gain |
| Photon-counting measurement | SPAD | Produces discrete detection events |
| Time-of-flight sensing | Both | APDs can process analog return waveforms; SPADs can record photon arrival events |
| Fluorescence lifetime measurement | SPAD | Photon timing supports time-correlated measurements |
| Higher background-light conditions | Depends on system design | APD analog dynamic range and SPAD count-rate limits must be evaluated |
| Low-light imaging | SPAD | Photon-counting arrays can provide high sensitivity |
| Optical waveform measurement | APD | Analog output can preserve intensity variations within receiver bandwidth |
Before selecting a detector, engineers should evaluate:
Minimum and maximum expected optical power.
Operating wavelength and spectral response.
Detector responsivity or photon detection efficiency.
APD multiplication gain or SPAD detection probability.
Noise, dark current, and dark count rate.
Required electrical bandwidth or timing resolution.
Dead time and maximum event rate.
Ambient-light conditions.
Operating temperature and bias stability.
Receiver electronics and signal-processing architecture.
12. APD and SPAD in Optical Communication Systems
In fiber-optic communication, the receiver photodetector converts incoming optical signals into electrical signals for subsequent amplification and processing.
PIN photodiodes and APDs are common detector options in optical communication receivers. APDs can provide internal gain when the receiver requires improved sensitivity, although this must be balanced against excess noise, bias requirements, and bandwidth.
SPADs are generally associated with photon-counting and timing applications rather than conventional high-speed analog optical communication receivers. Their event-based output, dead time, and count-rate constraints make them a different receiver architecture from the typical linear APD-based optical receiver.
At the transceiver level, detector selection depends on the transmission distance, optical power budget, data rate, modulation format, receiver sensitivity, and system noise requirements.
C-LIGHT provides a range of optical transceiver solutions for different optical networking applications. The specific receiver photodiode architecture should be confirmed from the relevant product datasheet rather than inferred from the module's data rate or form factor.
13. Conclusion
APDs and SPADs both use avalanche multiplication to improve optical detection sensitivity, but their operating modes serve different measurement requirements.
An APD operates below breakdown in linear mode, producing an amplified analog output related to incident optical power. It is suitable for optical communication receivers, analog ranging, and other systems that need continuous optical signal measurement.
A SPAD operates above breakdown in Geiger mode. A single detected photon can trigger an electrical pulse, making the device suitable for photon counting, time-of-flight sensing, LiDAR, and low-light imaging. Its main design considerations include photon detection efficiency, dark count rate, afterpulsing, dead time, and count-rate saturation.
In practical optical system design, APD gain and SPAD photon detection efficiency must be evaluated using different performance criteria. The best choice depends on the optical signal level, background illumination, required timing accuracy, dynamic range, and receiver architecture.
For optical transceiver requirements, compatibility evaluation, or application-specific technical support, Contact C-LIGHT.
TEL:+86 132 6656 7067




















































>
>
>
>
>
>
>
>