HCA-S Silicon and InGaAs Photoreceivers

The detector should be selected primarily from the operating wavelength and required responsivity. Si versions cover 320–1060 nm at 20 MHz or 320–1000 nm at 200 and 400 MHz, with typical maximum sensitivities of 0.6 A/W at 920 nm or 0.55 A/W at 800 nm. InGaAs models cover 900–1700 nm and provide 0.95 A/W typical sensitivity at 1550 nm, making them applicable to common near-infrared and telecommunications wavelengths. The HCA-S-20M-BSI-FST extends coverage to 200–1100 nm with 0.73 A/W typical sensitivity at 1010 nm. In practice, choose the detector material first and then compare bandwidth, active area, NEP and optical power limits.

The 20 MHz models suit applications where higher transimpedance gain is more important than nanosecond-scale response, providing 1 × 10⁵ V/A with an 18 or 19 ns rise/fall time. The 200 MHz variants reduce the gain to 2 × 10⁴ V/A and provide a 1.8 ns response. The 400 MHz models offer a 1.0 ns response with 5 × 10³ V/A gain. Increasing the bandwidth raises the specified NEP but also increases the available linear optical saturation power, reaching 400 µW for the 400 MHz Si model. Selection should therefore consider pulse-edge speed, optical power, required voltage amplitude and the analogue bandwidth of the connected instrument.

The FST input is appropriate for free-space beams and optical-bench arrangements that may need lenses, tubes, cage-system components or interchangeable fibre adaptors. It uses a 1.035″-40 threaded flange and can accept optional FC or FSMA adaptors. Coupling through an adaptor is less critical with larger detectors, such as the 3 mm Si detector and 0.5 mm InGaAs detector used in the 20 MHz models. The 0.3 mm InGaAs detectors in the 200 and 400 MHz FST versions require more careful alignment, while the corresponding fixed FC models use an integrated ball lens suitable for fibre cores up to 62.5 µm. Choose fixed FC when repeatable fibre coupling is the priority and FST when optical configurability is more important.

The BNC output is designed for a 50 Ω load and should be terminated with 50 Ω at the receiving instrument. The stated gain, output range and noise values are based on this termination, so a different load may not reproduce the specified electrical behaviour. Depending on the model, the linear output range is ±1.0, ±1.2 or ±1.5 V, which should be checked against the input range of the oscilloscope or A/D converter. A 50 Ω RF cable supports impedance matching and helps limit reflections and electromagnetic interference at higher frequencies. Cable length should be kept appropriate for the required bandwidth, particularly with the 200 and 400 MHz models.

Optical saturation power defines the approximate input level for linear amplification and is not the same as the absolute detector damage limit. Linear saturation values range from 20 µW for the 20 MHz InGaAs model to 400 µW for the 400 MHz Si model, with each figure applying at its stated reference wavelength. Absolute CW optical input limits are 6 mW for HCA-S-20M-SI, 20 mW for the 200 and 400 MHz Si models, and 10 mW for the InGaAs models. The black-silicon version uses an irradiance limit of 80 mW/mm² rather than a total-power value. Optical attenuation should be selected to keep normal and transient signals within both the linear output range and the applicable absolute limit.

The photoreceivers require a dual ±15 V supply, with an accepted operating range from ±14.5 to ±16.5 V. Model-dependent supply current is between ±50 and ±60 mA, while a supply capability of at least ±150 mA is recommended. Power is connected through a three-pin Series 1S socket, and the mating connector is included with the unit. A 20-minute warm-up is used for the specified test conditions, although at least 10 minutes is recommended before measurements. Operation is specified from 0 to +60 °C, and the measurement environment should be free from condensing moisture, dust, oil, smoke and other contaminants that could affect optical or electrical performance.

NEP indicates the optical power noise density at a stated wavelength and measurement frequency, so the values should not be compared without their test conditions. The lowest listed figures are 3.3 pW/√Hz for the black-silicon model at 1010 nm and 3.5 pW/√Hz for the 20 MHz InGaAs model at 1550 nm. The 400 MHz versions specify 40 pW/√Hz for Si and 24 pW/√Hz for InGaAs, measured at 100 MHz. As a first-order estimate, integrated noise rises with the square root of measurement bandwidth where the noise density remains approximately flat. Engineers should therefore compare NEP at the intended wavelength and then limit downstream bandwidth to what the signal actually requires.