Design and Performance of 220 and 270 GHz Bandpass Filters for BICEP Array

Imagine the universe as a giant, glowing fog that has been expanding since the very beginning of time. This fog is called the Cosmic Microwave Background (CMB), and it's the oldest light we can see, a baby picture of the universe taken just 380,000 years after the Big Bang. But here's the tricky part: this light isn't just a simple glow; it's polarized, meaning the light waves are vibrating in specific directions. Scientists believe that if they can measure these tiny vibrations with extreme precision, they might find the "fingerprint" of cosmic inflation—a theory that says the universe expanded faster than the speed of light in a split second after it was born.

However, looking at this ancient light is like trying to hear a whisper in a hurricane. Our own atmosphere, the dust in our galaxy, and even the instruments we use can create "noise" that drowns out the signal. To solve this, scientists build telescopes that act like super-sensitive ears, but they need to be incredibly selective about what sounds they listen to. They need to tune out the static of the atmosphere and only listen to very specific "notes" of light. This is where the story of the BICEP Array comes in, a team of scientists building a fleet of these super-telescopes at the South Pole to catch that whisper of the Big Bang.

---

Tuning the Cosmic Radio: The BICEP Array's New Filters

The BICEP Array is a collection of high-tech telescopes sitting at the South Pole, designed to listen to the polarization of the Cosmic Microwave Background. Think of the telescope as a giant radio, but instead of listening to music, it's listening to the universe's birth cry. The problem is, the universe is noisy. To hear the specific signal they want, the detectors inside these telescopes need to be like a very picky bouncer at a club, letting only the right guests (frequencies of light) in and turning away the rest.

This paper is all about the design and performance of those "bouncers," which are called Bandpass Filters (BPFs). Specifically, the team built and tested filters for two very high-frequency channels: 220 GHz and 270 GHz. Why these numbers? The 220/270 GHz range is the sweet spot for spotting "polarized dust" in our galaxy. If the scientists can measure how much dust is interfering with their signal, they can subtract it out and get a clearer picture of the cosmic inflation they are hunting for.

The Filter Design: A Lego Puzzle Without the Hard Pieces

The engineers needed a filter that could be built easily and consistently. They chose a design called a pi-network. Imagine you are building a circuit with tiny electronic Lego pieces. Most designs require you to use a specific piece called a "shunt inductor," but these are notoriously difficult to build perfectly every time—like trying to stack a wobbly tower of Jenga blocks.

The team's clever solution was to use a pi-network design that doesn't use shunt inductors at all. Instead, they used a mix of capacitors (which store electrical energy like a sponge) and series inductors (which resist changes in current like a heavy flywheel). By arranging these in a specific "pi" shape (looking like the Greek letter π), they could achieve the same filtering effect but with pieces that were much easier to manufacture consistently. They simulated these designs using a computer program called Sonnet, essentially building a virtual version of the filter to see how it would behave before cutting any real metal.

Tuning for the Perfect Signal

Once the design was set, the team had to tune it. They didn't just pick random numbers; they ran a complex calculation to find the "noise minimum." Imagine you are trying to listen to a radio station, but there is static from the sky (the atmosphere) and static from the cosmic background. The team calculated exactly which frequency range would let in the most signal from the universe while letting in the least amount of noise from the atmosphere.

For the 220 GHz filter, they found a "sweet spot" centered at 229 GHz with a bandwidth (the width of the frequency range it accepts) of 0.256. This setup gave them a "Noise Equivalent Temperature" (NET) of 187.9 µKcmb/√Hz. This number is a measure of how sensitive the detector is; the lower, the better. They got within 1% of the theoretical best possible performance.

For the 270 GHz filter, the math was a bit trickier because the atmosphere is less transparent at that frequency. They settled on a center of 271 GHz and a bandwidth of 0.273. While this resulted in a slightly higher noise level (358.3 µKcmb/√Hz) compared to the theoretical minimum, the team decided it was worth it. Why? Because having data at this specific frequency helps them understand the dust models better, which is crucial for their main science goals.

Testing the Real Thing

Theory is great, but does it work in the real world? The team took three 220 GHz detector modules to the South Pole and tested them during the 2024-25 winter. They used a special instrument called a Martin–Puplett Fourier Transform Spectrometer (FTS) to measure the actual light coming through the filters.

The results were impressive but showed a small gap between the computer simulations and reality.

• The Expectation: Based on their simulations and measurements of the telescope's antenna, they expected the filter to center around 229 GHz.
• The Reality: The actual measurements showed the center was at 235.6 GHz, and the bandwidth was 0.252.

This means the real filters were shifted slightly higher in frequency than the computer predicted. The authors suspect this might be because the computer simulation didn't perfectly guess the electrical properties of the materials (like the "relative permittivity" or how the metal conducts electricity at high speeds) or because the antenna measurement had a small error. However, they emphasize that this shift is small—just over one standard deviation—and it does not stop the telescope from achieving its science goals. The filters are still doing their job perfectly well.

As for the 270 GHz filters, they have only been tested in a lab so far, not at the South Pole. They measured a center frequency of 274.0 GHz and a bandwidth of 0.207. Because they couldn't measure the antenna performance for these yet (due to a lack of "loss test" detectors), they couldn't compare the lab results to the full system performance like they did for the 220 GHz. But, just like the 220 GHz filters, the 270 GHz filters look good enough to do the job.

The Bottom Line

The BICEP Array team successfully designed, built, and tested a new type of filter that avoids the tricky-to-build components of older designs. While there is a tiny difference between what their computer models predicted and what they measured in the real world, the filters are performing exactly as needed to help scientists listen to the faint whispers of the Big Bang. They have proven that you can build a highly sensitive, noise-optimized filter without the difficult "shunt inductors," paving the way for clearer views of the universe's earliest moments.