Design and characterization of W-band and D-band calibration sources for the AliCPT-1 experiment

Ali Cosmic Microwave Background Polarization Telescope (AliCPT-1) is the first Chinese cosmic microwave background experiment aiming to make sensitive polarization maps of the potential B-mode signal from inflationary gravitational waves. The telescope was deployed on the Tibet Ali site at 5250 m above sea level in early 2025. Before and after each observation season, the instrument performance must be carefully calibrated, including the far field beam performance, far sidelobe, spectral response, polarization angle, and cross-polar beam response. To characterize these optical performances, several calibrators have been developed. We developed a W-band source and a D-band source for the AliCPT-1 telescope’s beam characterizations. We present the design and performance of the two calibration sources.

💡 Research Summary

The paper presents the design, construction, and laboratory validation of two millimeter‑wave calibration sources—one operating in the W‑band (75–110 GHz) and the other in the D‑band (120–170 GHz)—developed for the AliCPT‑1 cosmic‑microwave‑background (CMB) polarization telescope. AliCPT‑1, deployed at the 5,250 m high Ali astronomical observatory in Tibet in early 2025, uses a 72 cm refracting optics system to observe at 90 GHz and 150 GHz with angular resolutions of 19′ and 11′ respectively. Precise calibration of the far‑field beam (FFBM) and far‑sidelobe (FSLM) response is essential because any beam mismatch between the orthogonal detector pairs can leak temperature anisotropy into polarization, contaminating the sought‑after B‑mode signal.

Traditional calibration with a large black‑body source is impractical for AliCPT‑1 due to the long far‑field distance (≈1.4 km) and the large aperture. Consequently, the authors adopted compact, electronically generated millimeter‑wave sources. The design requirements derived from Friis transmission calculations and detector saturation limits are: output power >10 mW (≈10 dBm), dynamic range >60 dB, power stability within ±1 % over several hours, linear polarization, a rotatable polarization angle covering 0–360°, sweep time shorter than the TES thermal time constant (≤8 µs), and GPS‑synchronized timing for demodulation.

Each source contains two excitation paths selectable by a single‑pole‑double‑throw (SPDT) switch: (1) an amplified thermal‑noise channel using a 50 Ω resistor and low‑noise amplifiers (LNAs) covering 12.5–18.33 GHz (W‑band) or 10.8–14.7 GHz (D‑band), and (2) a voltage‑controlled oscillator (VCO) channel capable of single‑frequency or rapid frequency‑sweep operation. A PIN diode switch chops the signal at ~1 kHz. Frequency multiplication stages and high‑power amplifiers raise the signal into the target band, after which two waveguide tunable attenuators (WT‑A) provide >60 dB of controllable attenuation. The final radiation is emitted from a custom horn antenna (≈22 dBi gain).

Control electronics consist of an STM32F103 microcontroller, a MAXII CPLD, a high‑speed DAC, and a GPS receiver. The MCU communicates via Ethernet, programs the VCO, and timestamps each chopping event with the GPS pulse‑per‑second (PPS) signal. An internal algorithm determines the optimal DAC step size (minimum 20 ns) and number of frequency points to meet a user‑defined sweep bandwidth and period, achieving up to 125 k sweeps per second (8 µs per sweep) and typically >160 frequency steps per sweep. This ensures the TES detectors see a quasi‑steady power despite the rapid sweep.

Mechanical integration includes a high‑precision rotary stage (resolution 0.0005°, repeatability ≤±0.003°) for polarization angle adjustment, a tiltmeter for monitoring source orientation, and a weather‑proof enclosure. The entire transmitter is housed in an aluminum box lined with Eccosorb microwave absorbers to suppress stray reflections.

Laboratory measurements confirm the design goals. Using a Keysight E4416A power meter and waveguide power sensors, the sources achieve maximum output powers of +14 dBm (W‑band) and +11 dBm (D‑band) in both thermal‑noise and VCO modes. Background noise floors are –25 dBm (W‑band) and –17 dBm (D‑band). S‑parameter measurements of the two‑stage WT‑A attenuators demonstrate a usable attenuation range exceeding 60 dB. Power stability tests over four hours show fluctuations within ±1 % for both the continuous thermal‑noise channel and the rapid VCO sweep mode, satisfying the stringent stability requirement for repeated raster scans of the detector array.

For the far‑field beam calibration, the sources will be mounted on a 28 m mast at the C1 site, 1.4 km from the telescope, with a low‑deformation aluminum mirror redirecting the beam toward the focal plane. The required output power for the FFBM campaign (–35 dB beam level) translates to –16 dBm to 0 dBm, comfortably covered by the sources. For far‑sidelobe mapping at 18 m distance, the sources must deliver >–3 dBm at the –90 dB point, which the measured >10 mW output also satisfies.

In summary, the authors have successfully delivered two broadband, high‑power, highly stable, and precisely rotatable millimeter‑wave calibration sources that meet all the optical and electrical requirements for AliCPT‑1 beam and sidelobe characterization. The presented architecture—combining dual excitation paths, fast frequency sweeping, wide dynamic range attenuation, and GPS‑locked timing—offers a practical solution for future high‑altitude CMB polarization experiments that demand accurate, repeatable calibration of large‑aperture telescopes.