In a wide range of applications, semi-conductor photo sensors (SiPMs) are increasingly replacing classical photo multiplier tubes (PMT). They have the advantage of an easier handling due to their significantly lower bias voltage and a long life time without aging. Usually, detectors need an adapted design for the application of SiPMs due to their smaller size compared to PMTs.

While the linear dynamic range of a PMT is inherently limited and usually depends strongly on the individual PMTs, SiPMs promise a dynamic range which only depends on the SiPM type applied and not on the individual sensor.

SiPMs are compiled from individual Avalanche Photo Diodes operated in Geiger-mode (G-APD). Every of these diodes is only capable of the detection of a single photon at a time. Thus, the number of G-APDs inherently limits the dynamic range of a SiPM. Strictly speaking, a SiPM is non-linear starting from the first detected photon. If this non-linearity is taken into account, the dynamic range for today's sensors can reach 10^6 for coincident photons. A complication arises for extended pulses from the fact that typical re-charge times of individual cells are in the order of several nano-second.

With a 3.8 sqm scintillator detector developed for the upgrade of the Pierre-Auger Observatory, it has been shown that SiPMs can nowadays act as an ideal replacement even in applications which require a high dynamic range. This has been successfully proven by operating two identical detectors on top of each other, one read out with SiPMs and one by a PMT. It is demonstrated that even at very strong illumination the SiPM response is still understood. Furthermore, laboratory measurements confirm that individual sensors are, within the systematic errors, exhibiting identical response. Given the precision of the devices and their advantages in operation, including the possibility of characterizing their response during measurement without any additional calibration device, the application of SiPMs will be a revolution for high dynamic-range applications, significantly reducing systematic uncertainties due to improved stability.