Fast Radio Burst project
About the photo
This is our first 6-meter reflector erected across the street at the East Asian Observatory base office. The prime focus receiver at the end of the quadrapod legs houses a dual linear polarization clover leaf antenna along with a pair of LNAs and optical transmitters. The optical receivers and backend digital hardware are located in the bay seen to the right of the dish. The reflector provides 30 dB gain and a 3 dB beamwidth of 5.8 degrees or about 11 times the angular size of the moon. The photo was taken with my iPhone XS Max on September 12, 2022 while performing a drift scan of the sun.
Our frequency of interest is 400 to 800 MHz and requires an RFI quiet site which we have identified in the Kau district of the Big Island. We plan to build a 10 antenna array to serve as an outrigger to the CHIME telescope located in British Columbia, Canada. The Hawaii site will be one of several outrigger locations that include Green Bank, West Virginia, and Algonquin Provincial Park, Ontario Canada. Once online, these outriggers will increase the angular precision of FRB detections and provide further clues as to the source of these energetic radio bursts.
My involvement
I started working on the FRB project during the latter part of 2020 and have slowly ramped up to nearly 50% level of effort as of 2023.
engineering activities
LTE Yagi Antenna Characterization
Log Periodic Antenna Characterization
2023-Jun - Characterization of ZDA Communications, model ZDADJ750-16-YG, 746 to 806 MHz, +16 dBi. Here we perform gain measurements using a pair of identical antennas at 776 MHz. Freespace loss theoretically follows the 1/r^2 rule where r is the distance between antenas, however, groundwave propagation loss is more accurately modeled as 1/r^(2.x), where x ranges from 0 to 5, depending on the height from the ground. In this case signal strength appears to diminishes as 1/r^(2.35).
We plan to use this antenna to capture realtime high SNR LTE signals at our site in Pahala and perform on the fly RFI cancellation. Sounds easy but is probably quite difficult in practice (D. Kubo, P. Oshiro).
Log Periodic Antenna Characterization
Log Periodic Antenna Characterization
2023-Apr - Characterization of MobileMark model Y42400WB, 400to 800 MHz, +9 dBi. Here we are peforming gain measurements using a pair of identical antennas at 600 MHz. The FRB project will utilize ten 6-meter dishes, each with dual polarization receivers, along with eight of these log periodic antennas to match the Algonquin array.
In addition, the BURSTT project will intially consist of eight log periodic antennas and will eventually be expanded to 256 antennas in a 16 x 16 rectangular array (D. Kubo, P. Oshiro).
System Characterization in Lab
2023-May - The cloverleaf antenna is integrated with the LNAs and therefore it is not convenient to perform system noise temperature tests due to RFI pickup. Instead, we used the connectorized LNAs to perform these tests.
The 4-element system was characterized to have receiver noise temperatures ranging from 80 to 115 K, however, we believe much of this was caused by large attenuator values prior to the optical transmitter. The ADC input drive level of 32 counts RMS (of 128) has been confirmed to be ideal. The first LNA stage produces non-trivial slope which is largely mitigated by the digital slope equalization. Channel delays for the four ADCs have been characterized and can now be applied to astronomical sources (D. Kubo, P. Oshiro).
4-element Tests in Pahala
4-element Tests in Pahala
2023-Apr - This image is from an early attempt at duplicating the 8-element array at Algonquin Radio Observatory in Ontario in Canada. We oriented the elements to intercept the suns path and expected to see clean linear phase fringes but something wasn't working correctly in the digital backend. We saw strong fringes but they were not linear across the 400 to 800 MHz band. However, we did manage to capture the RFI environment at our Pahala site (D. Kubo, P. Oshiro, C. Jeschke).
6-meter Parabolic Dish
4-element Tests in Pahala
EMI Chassis for ICE Board
2022-Sep - ZLE Technology, model GL-DYS600AM12PM, 6 meter diameter, 231 cm focal length, 0.38 f/d ratio. We didn't measure gain but calculated it to be approximately 30 dB at 600 MHz. We have purchased a total of ten dishes and erected one across the street at EAO. Unfortunately the RFI in Hilo is quite poor and saturates our front-end preventing us from performing astronomical measurements. More about this dish can be found in the URSI Talk document listed (M.T. Chen, D. Kubo, P. Oshiro, A. Mills).
EMI Chassis for ICE Board
4-element Tests in Pahala
EMI Chassis for ICE Board
2023-May - Our digital backend consists of a custom ICE board running a single Xilinx Kintex 7 XC7K420T FPGA. There are two ADC daughter boards each with a pair of E2V EV8AQ160 5 Gsps 8-bit ADCs. The ICE board can digitize up to 16 analog inputs each with input bandwidths of 800 MHz. We plan to utilize two ICE boards to digitize and process the analog input signals from the ten 6-meter dishes and eight log periodic antennas.
One problem we have encounted is the EMI generated by the ICE board falls within our 400 to 800 MHz observing band. Thus we have designed an EMI chassis based on a COTS Pi-Metal chassis. The above photo shows the ICE board located within this EMI chassis with the lid removed. More on this can be found in the EMI for ICE Board document below (D. Kubo, P. Oshiro, R. Chilson).