Casper Farret Jentink
Career Stage
Student (undergraduate)
Poster Abstract
KISS stands for Kapteyn Interferometer for Short-baseline Solar Observations. It is a two-element radio interferometer that was designed and built on a limited budget as part of my BSc. Thesis. The project consisted of three theses, the other two involved simulations of measurements and the design of the setup including dish and receiver characterization. My share in the project was the modifications that needed to be done to the receivers for it to work and doing observations of the Sun including data processing. KISS is currently used by Astronomy and mechanical engineering students to learn about interferometry and do observations. With help of KISS we were able to determine the angular size of the Sun.
Plain text summary
Slide 1/2:
At the very beginnings of the project some back-of-the-envelope calculations were made regarding required sensitivity and resolution. From these calculations we derived that the Sun could be easily observed with help of camping satellite dishes (the cheapest you can buy) and standard Low-Noise-Block (LNB) receivers, usually used to receive TV-signals. This time, we applied them for other purposes. LNB-receivers operate in the 10.75-12.75 GHz band. This band is split into two 1 GHz bandwidth parts. We chose the 10.75-11.75 GHz band. Furthermore, we required mounts, a power meter, a frequency generator, 50 meters of coax cable, a Raspberry Pi, a voltage supply, a soldering iron and wires, and two antenna splitters.
Slide 3:
The LNB receiver works by converting the incoming signal at 10.75-11.75 GHz down to a 1-2 GHz signal by applying a 9.75 GHz local oscillator signal to the antenna. The outgoing 1-2 GHz signal is easier to sample for, for example, a TV-decoder. In our case, this signal was easier to correlate. Correlation is the process of mixing signals of multiple dishes/receivers in such a way that they interfere. For big observatories like A.L.M.A., the Atacama Large Millimeter Array, this is done with supercomputers that sample the signals and multiply them at incredibly high rates. In our case we used an inverted antenna splitter that allowed the signals of both dishes to interfere (A.L.M.A.’s computer cost $11 million, our splitter < $5,-). To properly let signals of both dishes to interfere, the receivers need to be in phase. This implies equal length cables but also a local oscillator signal in both LNB’s that runs in the same phase. The local oscillator is fed by a 25MHz quartz crystal in each LNB. We decided to solder it out and replace it by an external 25MHz signal provided by a frequency generator. This was done for both LNB’s and they were consecutively hooked up to the same frequency generator with help of coax cable and another antenna splitter. One other simple modification was also done to the LNB: soldering on an external power supply.
Slide 4:
Mounting everything together we were now ready to do observations of the Sun. We decided to do drift scans: letting the Sun pass through the beam of both dishes thanks to Earth’s rotation while keeping the dishes still and pointed in the same direction. As the Sun passes through the beam, both dishes receive a (more or less equal) signal from the Sun. This signal travels through coax cables to the inverted antenna splitter where it is correlated. However, the dishes are placed some distance apart, so the Solar radiation will be received at different times for both receivers throughout the measurement. Given that the Sun ‘drifts’, the delay between these times changes throughout the measurement. Every time the delay changes with one wavelength we get constructive and negative interference. The diagram at the left bottom of the slide explains this method.
An interferometer measures the Fourier transform of the sky, these values are called visibilities. If we determine these values for multiple measurements at various baselines (distances between the telescopes) we can reconstruct the Fourier transform of the Sun, which tells us its shape and size (The plot in the right bottom part of the slide displays visibility values for various baselines). From 13 measurements we were able to determine that the Sun has an angular size of 0.54±0.02 degrees, very close to its actual value! Its shape is likely circular but square still remains a possibility.
At the very beginnings of the project some back-of-the-envelope calculations were made regarding required sensitivity and resolution. From these calculations we derived that the Sun could be easily observed with help of camping satellite dishes (the cheapest you can buy) and standard Low-Noise-Block (LNB) receivers, usually used to receive TV-signals. This time, we applied them for other purposes. LNB-receivers operate in the 10.75-12.75 GHz band. This band is split into two 1 GHz bandwidth parts. We chose the 10.75-11.75 GHz band. Furthermore, we required mounts, a power meter, a frequency generator, 50 meters of coax cable, a Raspberry Pi, a voltage supply, a soldering iron and wires, and two antenna splitters.
Slide 3:
The LNB receiver works by converting the incoming signal at 10.75-11.75 GHz down to a 1-2 GHz signal by applying a 9.75 GHz local oscillator signal to the antenna. The outgoing 1-2 GHz signal is easier to sample for, for example, a TV-decoder. In our case, this signal was easier to correlate. Correlation is the process of mixing signals of multiple dishes/receivers in such a way that they interfere. For big observatories like A.L.M.A., the Atacama Large Millimeter Array, this is done with supercomputers that sample the signals and multiply them at incredibly high rates. In our case we used an inverted antenna splitter that allowed the signals of both dishes to interfere (A.L.M.A.’s computer cost $11 million, our splitter < $5,-). To properly let signals of both dishes to interfere, the receivers need to be in phase. This implies equal length cables but also a local oscillator signal in both LNB’s that runs in the same phase. The local oscillator is fed by a 25MHz quartz crystal in each LNB. We decided to solder it out and replace it by an external 25MHz signal provided by a frequency generator. This was done for both LNB’s and they were consecutively hooked up to the same frequency generator with help of coax cable and another antenna splitter. One other simple modification was also done to the LNB: soldering on an external power supply.
Slide 4:
Mounting everything together we were now ready to do observations of the Sun. We decided to do drift scans: letting the Sun pass through the beam of both dishes thanks to Earth’s rotation while keeping the dishes still and pointed in the same direction. As the Sun passes through the beam, both dishes receive a (more or less equal) signal from the Sun. This signal travels through coax cables to the inverted antenna splitter where it is correlated. However, the dishes are placed some distance apart, so the Solar radiation will be received at different times for both receivers throughout the measurement. Given that the Sun ‘drifts’, the delay between these times changes throughout the measurement. Every time the delay changes with one wavelength we get constructive and negative interference. The diagram at the left bottom of the slide explains this method.
An interferometer measures the Fourier transform of the sky, these values are called visibilities. If we determine these values for multiple measurements at various baselines (distances between the telescopes) we can reconstruct the Fourier transform of the Sun, which tells us its shape and size (The plot in the right bottom part of the slide displays visibility values for various baselines). From 13 measurements we were able to determine that the Sun has an angular size of 0.54±0.02 degrees, very close to its actual value! Its shape is likely circular but square still remains a possibility.
Poster file
Poster Title
KISS - the Kapteyn Interferometer for Short-baseline Solar observations
Url
https://www.linkedin.com/in/casper-farret-jentink/