Prime-Cam, a receiver being prepared for the Fred Young Submillimeter Telescope, met its stated deployment requirements in laboratory testing and was shipped to Chile with its 280 and 350 GHz modules by June 2026. Integration and commissioning were scheduled for 2026, followed by early-science observations.
The report describes engineering tests and deployment plans rather than completed telescope observations. Its main test setup used a fully populated 280 GHz module with all three detector arrays and the complete optics set, allowing the team to assess cryogenic, optical and readout deployment criteria. No completed on-sky science measurements are reported.
A large design, still partly on paper
Prime-Cam is projected to provide more than ten times the mapping speed of current facilities across wavelengths from 1.4 to 0.3 millimetres, or 220 to 850 GHz. Mapping speed is the rate at which an instrument could build a survey. When fully populated, the design calls for more than 100,000 kinetic inductance detectors across seven independently optimised modules.
That scale remains a projection. The report contains no on-sky measurement of the claimed mapping-speed advantage, and it does not describe the full seven-module configuration as installed or operating. The module plan lists 10,332 detectors in the 280 GHz module and 10,448 in the 350 GHz module, both shipped for 2026.
Other modules were at earlier stages. EoR-Spec, with 6,528 detectors, and the 850 GHz module, with 38,000, were in development for 2027. The 410 GHz module, with 21,000 detectors, was in development for 2028, while Spec-on-Chip was in design for that year and another module was still listed as to be decided.
Cold enough for the next test
The receiver's cryostat, the enclosure and cooling system for its detectors, is built around thermal stages at 300 K, 80 K, 40 K and 4 K. Selected parts at the 80 K and 40 K stages use Al 1100 rather than the Al 6061-T6 used for general metal parts.
With the fully assembled 280 GHz module installed, the system took approximately seven days, or 160 hours, to cool. The focal plane reached about 80 millikelvin, while the dilution refrigerator's mixing chamber reached about 45 millikelvin. Turning on all 18 low-noise amplifiers raised the focal-plane temperature to 88 millikelvin.
Thermal tests estimated about 20 microwatts of radiative loading from inside the enclosed 4 K shell. The 280 GHz module contributed about 70 microwatts at the mixing chamber and about 12 milliwatts at the still plate. Loads equivalent to two and four modules produced focal-plane temperatures of 98 mK and 110 mK, respectively.
These measurements describe laboratory behaviour under applied loads, not a demonstration that every planned module will operate at the same temperature. Some thermal values are estimates or projections for similar modules, and the complete configuration still requires testing.
Optics and readout pass the bench test
For the optical test, the team averaged measurements from the 30 detectors with the highest signal-to-noise ratios in each of the three arrays. All three arrays showed an approximately 80 GHz-wide bandpass, the frequency range passed by the optics, centred near 270 GHz and confirmed to within about 5 GHz.
The result came with a qualification. Alignment limitations in the Fourier-transform spectrometer meant that differences among the arrays could not be assessed. The test therefore supports the reported band shape for the sampled detectors, but it does not provide a full comparison among arrays.
The readout architecture could handle more than 1,000 kinetic inductance detectors on a single radio-frequency transmission line. Each RFSoC could read four RF channels, with up to 1,000 detectors over 512 MHz per channel using current firmware.
The next stage is on the telescope
Commissioning plans begin with cooldowns of the 280 and 350 GHz modules. They also call for stationary dark tests to assess detector yields and stable base temperatures, followed by scanning tests for mechanical and cryogenic stability.
The observing plan allocates 200 hours of early science to WFS, 100 hours to CIB and 160 hours to Galactic polarization. Over the planned five-year survey, the corresponding allocations rise to 2,000, 500 and 525 hours. These are prospective time allocations, not completed observations.
For time-domain work, the plan schedules revisits to Galactic star-forming regions at around 20 minutes per epoch on a monthly cadence. It also describes repeated observations at a two-week cadence and target-of-opportunity follow-up at one hour per epoch.
Observing data were to be registered with OpsDB, transferred to on-site computers for quick-look quality assessment, archived at the CCAT Data Center and processed by developing pipelines into calibrated science-ready products.
The report identifies the laboratory results as validation for deployment, while telescope integration, commissioning and early-science observations remain scheduled or planned. The document is an arXiv version 1 preprint dated 25 August 2026, and it reports no completed on-sky science measurements.
Paper data and sources
Original title: CCAT: The Prime-Cam Instrument for the Fred Young Submillimeter Telescope -- Overview and Status
Authors: Eve M. Vavagiakis, Yuhan Wang, Lawrence T. Lin et al.
Journal/Repository: Proc. SPIE 14156, Millimeter, Submillimeter, and Far-Infrared Detectors and Instrumentation for Astronomy XIII, 1415604 (20 Aug 2026)
Status: Preprint, not yet peer-reviewed
First online: 2026-08-25
DOI: 10.1117/12.3105297
Original paper · Full text