Broadband Detector

A Low-Cost Broadband Detector for Observing Weak Continuum Sources in Amateur Radio Astronomy

In radio astronomy, there are two principal approaches to signal processing: analog and digital. Each has its advantages and limitations, particularly for amateur radio astronomers working with modest budgets.

Digital signal processing (DSP) is based on converting analog input signals into digital data that can be analyzed by a processor. This approach is flexible, powerful, and allows extensive post-processing. Software-defined radios (SDRs) are typical representatives of this class and have become increasingly popular compared to purely analog signal processing.

While SDRs offer many advantages — such as high precision, adaptability, and access to a broad software ecosystem — they also have a key limitation: bandwidth. Affordable SDRs often process only a few megahertz at a time. Achieving larger bandwidths requires expensive high-end hardware, along with a powerful computer capable of handling the necessary computations.

This is where simple broadband receivers — such as logarithmic detectors or true power detectors — demonstrate their strength. Their major advantage is a wide usable bandwidth, often several hundred megahertz.

The radiometer equation tells us that the sensitivity of a radio telescope depends on four factors:

Antenna aperture (diameter)
System noise temperature
Receiver bandwidth
Integration time

For amateur astronomers using relatively small antennas, bandwidth is often the only practical parameter that can be increased when system noise temperature cannot be further improved. Increased sensitivity is essential for detecting weak continuum sources such as supernova remnants.

For comparison: the flux density of the supernova remnant Taurus A in the Ku band is roughly 100 times lower than that of the Moon — and the Moon’s flux density is about 100 times lower than that of the Sun. The challenge is substantial, but not insurmountable.

Amateur radio astronomers benefit from two important circumstances. First, for the Ku band — the frequency range used by TV satellites — antennas and LNBs are available at very low cost. Satellite dishes with LNBs are sometimes even given away online. Modern LNBs operating around 11 GHz achieve remarkably low noise figures, enabling high system sensitivity.

Second, supernova remnants exhibit a relatively flat spectral index, meaning their flux density does not decrease as rapidly with increasing frequency compared to many other continuum sources. Apart from the Sun and solar system objects, most continuum sources are better observed at lower frequencies (for example, in the L band around 1420 MHz). However, the supernova remnant Taurus A (the Crab Nebula, M1) remains detectable even at 11 GHz.

System Architecture: AD8362 + ADS1115 + Arduino

To develop a highly sensitive broadband detector for our radio telescope, we tested several approaches, including logarithmic detectors, true power detectors, and simple diode detectors. Many of these components are now inexpensive and available as convenient breakout modules online.

Among the tested detector boards, the AD8362 delivered the best results. Unlike logarithmic detectors such as the AD8317, which produce an inverted output signal, the AD8362 provides a non-inverted output. It also includes an integrated amplifier and can reliably measure input signals in the range from –55 dBm to +5 dBm. Its slope is approximately 51 mV/dB, which is significantly higher than that of many other logarithmic detectors.

Thanks to its integrated gain stage, the AD8362 enables a simple and compact system design. This reduces susceptibility to interference and lowers the probability of failure — both important advantages for sensitive radio astronomical measurements.

The output signal from the AD8362 is fed directly into a 16-bit ADC (ADS1115) and then transmitted via an Arduino microcontroller to a PC for digital processing. The total hardware cost is approximately 30 euros.

Circuit Design

The detector schematic is attached. The signal is first sampled by the ADS1115. Within the Arduino, the samples are integrated.

The improved version of the detector includes an LCD display showing the signal strength directly in dBm, as well as a buzzer that converts signal strength into an audible tone. This feature has proven particularly useful for aligning a satellite dish with the Sun.

For optimal performance in the Ku band, we recommend filtering the input signal to suppress interference from Starlink satellites. The best results were achieved using a 915 MHz bandpass filter with a bandwidth of 50 MHz from DYKB, housed in a metal enclosure and available at low cost from online suppliers in China.

To compensate for the filter’s insertion loss and raise the signal into the detector’s dynamic range, a low-noise amplifier (LNA) must be placed before the filter. A schematic for a power supply unit for the LNB, including integrated LNA and filtering, is also provided.

The detector can also be used in other frequency bands, such as the Ka band. In this range, filtering is typically unnecessary, since Starlink satellites do not transmit there. As a result, the full receiver bandwidth can be utilized.

Arduino Code

The Arduino code for the receiver is included. Communication with the ADS1115 is handled using the open-source library by Andre McKay.

The software implements a minimal main loop that continuously reads measurement values and stores them in a ring buffer. Once the buffer is filled, a moving average is calculated to reduce noise. The oldest value in the buffer is replaced by the averaged value while new samples are asynchronously added.

Rotary Encoder Functions

Since the current software version does not include a menu system, the integrated push button of the rotary encoder serves as both a marker input and a buzzer on/off switch.

The rotary encoder itself is used to adjust the integration time. A single button press inserts a timestamp or trigger marker into the data stream, simplifying later event analysis. Two button presses toggle the buzzer on or off. Rotating the encoder adjusts the integration time in steps from 10 ms to 1000 ms.

This averaging function effectively reduces system noise, which is particularly important for detecting very small changes in signal level and, consequently, very weak continuum sources.

Integration with Radio Astronomy Software

For visualization and further data processing, the system uses SkyRadio Pipe, a well-established software package in radio astronomy.

To ensure compatibility, a dedicated protocol parser was implemented to handle both push and poll requests according to the SkyRadio Pipe specification. The Arduino responds with either the current measurement value or the averaged value from the ring buffer.

The project is available as an open-source solution in our public repository and can be compiled and uploaded to compatible Arduino boards using PlatformIO. The modular structure of the code simplifies future extensions, such as implementing an RTOS or migrating to more powerful microcontrollers.