APD Questions & Answers

What is excess noise in APDs?

Excess noise is the additional noise created by a detector with gain. A detector with excess noise of 1 means that the intrinsic noise is multiplied by 1—therefore, the noise is not made any worse—but any value over 1 means the noise is increased due to the gain. The magnitude of excess noise can be measured by back calculating from the anode output current the equivalent average and standard deviation of incident flux. By using a sufficiently high intensity of incident light, we can fairly assume that the photon shot noise is the primary source of noise; therefore, without excess noise, the standard deviation of incident flux should be equal to the square root of the average incident flux, i.e., the photon shot noise. The fraction of the standard deviation of incident flux above photon shot noise is the excess noise.

APDs have a specification called excess noise figure. What does it mean?

Excess noise figure, also called excess noise index, is a value that can be used to estimate excess noise. The formula for calculating excess noise from the excess noise figure is gain raised to the excess noise figure. Excess noise figure is an empirically measured value that allows the sensor’s user to estimate the excess noise at a given voltage or gain setting. The excess noise figure is determined by sweeping the gain setting then measuring the excess noise. The excess noise figure is then determined to generate a linear slope that approximates the slope of the excess noise vs. gain plot.

What is the typical excess noise of Si APDs?

Si APDs’ excess noise is typically between 2 and 6. This is much higher than other detectors like PMTs (between 1.2 and 1.3) and MPPCs/SiPMs (between 1.1 and 1.2). One of the reasons why Si APDs’ excess noise values are relatively high is the ability for both electrons and holes to cause impact ionization with varied efficiency.

Why does the excess noise change depending on the wavelength?

The absorption depth of the photon varies depending on the wavelength, with shorter wavelengths absorbed closer to the front surface. Where the photon is absorbed and the avalanche is initiated affects how the carrier multiplication progresses in the avalanche region. The charge carriers must also traverse a minimum distance to generate sufficient kinetic energy to initiate impact ionization. As a result, the stability of the avalanche multiplication depends on the location where the first charge carrier is generated.

If you have a technical question you’d like to see answered on this page, email us.

Meet the engineer

Dino Butron is an Applications Engineer, specializing in all of our point detectors and signal-to-noise simulations. He and his photonic sidekick (his cat Bash) tackle technical issues daily.