{"id":"83c0baa7-c399-4e1f-9a3e-39e143867f70","arxiv_id":"2510.12162","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Cold-load tests of 280 GHz TiN and Al KIDs for Prime-Cam show NEP degrades up to 2x with only ~0.2 pW of optical loading drift and depends strongly on readout tone power and placement.","lead":"This paper reports lab measurements of the tiny superconducting light sensors planned for Prime-Cam, a camera for the Fred Young Submillimeter Telescope, testing how they respond to light and readout settings. The results are meant to guide calibration and operation of the telescope's 100,000-plus sensors when it starts observing in 2026.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stray light self-flagged in Sec. III.C is load-bearing for the 0.2 pW mapping, but not for the relative NEP degradation factor; a light-leak check would settle which numbers survive.","rationale":"The reader's weakest assumption identifies exactly the same issue: the optical power calibration depends on a light-tight enclosure, which Sec. III.C itself calls into question. This is the most load-bearing concern because it directly affects the quantitative claim of '0.2 pW' and the absolute NEP numbers, while leaving the relative degradation factor and the qualitative operational recommendation intact. I considered whether the NEP degradation factor itself might be an artifact of the phase-to-frequency conversion at detuned positions, but the paper's method of fitting the phase–frequency relation and computing frequency noise is standard, and the degradation factor is likely correct in relative terms. I also considered the edge-pixel selection and lack of error bars, but these are secondary to the stray light issue because they would mostly affect the magnitude of the effect, not the existence of the detuning penalty. The paper's own language is appropriately cautious, and the CONDITIONAL verdict from the reader is well matched to the evidence. A concrete light-leak test would settle whether the absolute numbers can be trusted; until then, the conditional status remains appropriate.","tokens_in":7004,"tokens_out":7238,"duration_ms":63890,"concrete_test":"Perform a direct light-leak check: with the cold load at 4 K (or 18 K), measure the detector resonance frequency and noise timestreams with the optical path between the cold load and the detector enclosure deliberately blocked by a light-tight blank-off plate (or a known cryogenic absorber) at the enclosure temperature. If the resonance frequency or noise floor changes when the blank is inserted, stray light is present. Then, with the blank removed, repeat the measurement and use the difference to estimate the stray-light power as a function of cold load temperature. Re-derive the responsivity R and the NEP using the corrected optical power. If the ΔP_opt corresponding to 0.5 linewidth detuning changes by more than ~30% relative to the paper's quoted 0.2 pW, the central quantitative claim is compromised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III.A estimates incident optical power from the cold load temperature assuming emissivity 1, known filter transmission, and a light-tight enclosure. Section III.C explicitly states that preliminary optical efficiency analysis indicates an inflation of responsivity and a reduction in NEP, consistent with excess stray light entering the setup, possibly from the custom aluminum box not being fully light-tight. If stray light is present, the estimated change in optical power ΔP_opt for a given cold load temperature step is incorrect, because the stray light adds an unknown (and possibly varying) component to the detected power. Consequently, the central quantitative claim that 0.5 linewidth detuning corresponds to approximately 0.2 pW is called into question: with stray light, the true power change could be smaller or larger, which would shift the atmospheric-loading-change threshold that motivates the tone-reset cadence. Additionally, the responsivity R is extracted from Δf/f0 vs ΔP_opt, so an error in ΔP_opt propagates directly to R and hence to the absolute NEP (NEP = sqrt(Sxx)/R). The relative NEP degradation factor of 1.5–2 at a given detuning is a ratio and is less sensitive to a constant calibration error, so the operational conclusion that tones must be optimized and periodically reset is not overturned. But the paper's headline numbers—especially the 0.2 pW mapping and the absolute NEP values—depend on the stray-light assumption. The paper itself acknowledges this uncertainty and says further measurements are underway, but no quantitative bound is provided.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents laboratory characterization of sample 280 GHz TiN and Al kinetic inductance detectors (KIDs) for the Prime-Cam instrument on CCAT/FYST. Using a temperature-controlled cold load, the authors measure optical responsivity (Eq. 1), frequency noise PSDs (Eq. 3), and NEP (Eq. 4) as functions of optical loading, readout tone power, and tone placement. A separate drift test fixes the tone at the 9 K resonance and steps the cold load from 8 K to 10 K, yielding NEP increases of 1.5–2× for approximately 0.2 pW optical-power changes. The paper concludes that readout tone placement and power must be optimized and that periodic tone resetting is needed to avoid noise penalties under varying atmospheric loading. The authors explicitly acknowledge in Sec. III.C that preliminary optical efficiency analysis indicates excess stray light, which may inflate responsivity and reduce NEP.","tokens_in":7351,"tokens_out":2799,"duration_ms":29286,"significance":"If the absolute calibration is correct, the paper provides useful quantitative guidance for operating >100,000 KIDs on Prime-Cam, especially the relationship between detuning and NEP degradation and the need for tone-reset cadence. The comparison of TiN versus Al nonlinearity and noise behavior is valuable for array design. The paper is honest about its main limitation, which is a strength: Sec. III.C openly flags possible stray light and states that further measurements are underway. However, because the central quantitative claims—the 0.2 pW mapping and the absolute NEP values—rest on an optical-power model that the authors themselves suspect is contaminated, the significance of those specific numbers is currently conditional. The relative degradation factor (1.5–2× at 0.5 linewidth detuning) is less sensitive to a constant calibration error and supports the operational conclusion even if the absolute power scale shifts.","major_comments":[{"comment":"The absolute responsivity R and NEP scale directly with the assumed incident optical power ΔP_opt. Sec. III.C states that preliminary optical efficiency analysis indicates excess stray light, possibly from the custom aluminum box not being light-tight. If stray light is present, the change in optical power for a given cold-load temperature step is not known, so R from Eq. (1) is overestimated and NEP from Eq. (4) is underestimated. The quoted \"0.2 pW\" mapping in Sec. IV is therefore not established. The authors should either perform a light-leak check and report a corrected optical-power model, or explicitly reframe the quantitative claims as relative (linewidth detuning) and remove or strongly caveat the pW conversions and absolute NEP values. At minimum, a quantitative error budget for the optical-power estimate is needed.","section":"Sec. III.C and Eqs. (1), (4)"},{"comment":"No error bars or uncertainty estimates are provided for R, NEP, or the quoted detuning-to-power mapping. The figures show representative pixels without statistical scatter or pixel-to-pixel variation. The claim \"NEP degraded by a factor of 1.5 to 2\" needs a statement of how many pixels were measured, the spread across pixels, and how the uncertainty in the cubic phase-to-frequency fit (Sec. III.A) and the Welch PSD estimate propagate into NEP. Without this, the precision of the central result cannot be assessed.","section":"Sec. III.A, III.B; Figs. 3, 4"},{"comment":"The discrepancy with previous measurements [6,7] is attributed to a combination of wafer/edge effects and possible stray light, but the preliminary optical efficiency analysis is not described. The reader cannot tell how large the inferred inflation is or how it would change R and NEP. Please specify the analysis that indicates stray light, the estimated magnitude of the effect, and how it affects the conclusions. This is load-bearing for the absolute numbers, not merely a caveat.","section":"Sec. III.C"}],"minor_comments":[{"comment":"Typo: \"demonsrate\" should be \"demonstrate\".","section":"Sec. IV"},{"comment":"Typo: \"bifuraction\" should be \"bifurcation\" (also in Fig. 2 caption).","section":"Fig. 3 caption"},{"comment":"The assumption of emissivity 1 for the Eccosorb cold load is stated but not justified with a reference or uncertainty. A brief justification or citation would help.","section":"Sec. III.A"},{"comment":"The caption cites \"0.2 pW optical loading changes\" without indicating that this value depends on the optical-power model flagged in Sec. III.C. Add a qualifier such as \"estimated optical power.\"","section":"Fig. 4 caption"},{"comment":"The phrase \"photon-noise-limited performance\" in the introduction cites [6]–[8], but the text does not specify which of the presented measurements are photon-noise-limited. Clarify whether the NEP values in this work are limited by photon noise or by readout noise.","section":"Sec. II"}],"recommendation":"major_revision","confidential_remarks":"The paper's honest self-flagging of the stray-light issue in Sec. III.C is commendable, but it means the headline quantitative claims are currently conditional. The relative effect on NEP with detuning is robust, and the operational advice about tone optimization is well supported. I would like to see either a definitive light-leak check or a reframing of the paper around relative measurements before publication. The paper fits the journal's scope as an instrument-focused experimental study."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid instrument paper with a clear honest limitation. The authors measure how NEP for 280 GHz TiN/Al KID witness pixels depends on optical loading, readout tone power, and tone placement, using a cold load. The new bit is the NEP-vs-detuning and NEP-vs-readout-power data for these specific pixels. The main operational claim—that 0.5 linewidth detuning, roughly 0.2 pW of loading change, degrades NEP by a factor 1.5–2—is plausible as a relative effect and directly relevant for Prime-Cam's readout strategy. The paper does well to flag in Sec. III.C that their measured responsivities are higher than previous arrays and that stray light in the custom aluminum box may be inflating R and suppressing NEP. That is exactly the right thing to do, and it makes the paper more useful, not less.\n\nThe soft spots are the absolute numbers. The optical power estimate assumes emissivity 1, known filter transmission, and a light-tight enclosure; the stray-light caveat means the 0.2 pW mapping and the absolute NEP values are not yet trustworthy. There are no error bars anywhere, so you cannot tell how much the calibration uncertainty matters. Also, the pixels are edge witnesses from a pre-edit wafer, so they may not represent the production arrays. The relative NEP degradation factor is less sensitive to a constant calibration error, so the operational conclusion about tone optimization and periodic retuning survives. But the quantitative loading threshold and the absolute NEP should be treated as provisional until the stray-light issue is bounded.\n\nThe method itself is standard cold-load KID characterization; there is no circularity in computing NEP from independently measured frequency noise divided by a responsivity fit. The phase-to-frequency cubic is an instrument calibration, not the claim under test. Citation pattern looks fine, with prior array papers [6],[7] appropriately cited.\n\nWho is this for? The CCAT collaboration and anyone working on KID readout optimization. It does not change detector physics, but it is exactly the kind of measured input needed to commission a >100k-pixel focal plane. I would bring it to a reading group focused on instrumentation and would accept it for peer review, with the request that the authors quantify the stray-light bound and report uncertainties. The relative trends are likely to hold; the absolute calibration needs another iteration.","headline":"Useful, honest KID characterization for Prime-Cam; the relative readout trends are solid, but the absolute power scale needs a light-leak check before the calibration numbers are trusted.","tokens_in":7955,"tokens_out":3906,"would_cite":true,"duration_ms":49142,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A fixed readout tone on a kinetic inductance detector loses 1.5–2x in sensitivity from a 0.2 pW optical-power drift, so Prime-Cam must periodically reset its tones.","keywords":["kinetic inductance detectors","submillimeter astronomy","readout optimization","noise equivalent power","optical responsivity","tone detuning","detector calibration","Prime-Cam"],"falsifier":"At fixed cold-load temperature, shift a readout tone by half a linewidth and measure NEP; the paper's claim predicts a 1.5–2x degradation, so a substantially smaller penalty would indicate overestimated detuning sensitivity. Separately, resealing the detector enclosure to eliminate stray light and remeasuring responsivity would settle whether the absolute optical calibration is inflated.","tokens_in":6898,"feed_emoji":"🔭","tokens_out":10368,"duration_ms":82702,"temperature":0.7,"pith_summary":"This paper reports laboratory measurements of 280 GHz superconducting microwave resonators (kinetic inductance detectors) destined for the Prime-Cam receiver, using a temperature-controlled blackbody cold load to simulate the sky. The goal is to establish how readout tone power and tone placement, together with varying optical loading, set the detectors' noise equivalent power (NEP), and to use that understanding to plan calibration and retuning for an array of more than 100,000 detectors. The central quantitative finding is that a shift in optical loading of only about 0.2 pW, which moves a fixed readout tone by half a linewidth off resonance, degrades NEP by a factor of 1.5–2. That means the readout system must periodically reset tone positions as atmospheric loading drifts, or the array spends much of its time in a degraded-noise state. The measurements also show that aluminum KIDs have a nonlinear response and that suboptimal readout power can inflate NEP by as much as 200%, reinforcing the need for per-pixel calibration and careful tone optimization.","feed_headline":"Optical drift of 0.2 pW doubles Prime-Cam detector noise","feed_subtitle":"Lab tests of 280 GHz kinetic-inductance detectors show a 0.2 pW drift doubles detector noise unless tones are reset.","key_machinery":"The load-bearing mechanism is the fixed microwave readout tone parked on a KID resonance. Incoming optical power changes the resonator's kinetic inductance, shifting its frequency and therefore the tone's position relative to the resonance; the resulting amplitude and phase modulation is the signal. The paper's metric is NEP(f) = sqrt(S_xx(f))/R, where S_xx is the power spectral density of fractional frequency fluctuations and R is the optical responsivity (fractional frequency shift per unit optical power). Because R and the noise level both depend on where the tone sits on the resonance, the detuning distance in linewidths becomes the controlling variable, and the paper calibrates how much","core_discovery":"The paper's claim, stated on its own terms, is that for 280 GHz TiN and Al KIDs the attainable noise-equivalent power depends strongly on both the power and the frequency placement of the fixed readout tone, and that this dependence is sharp enough to matter under real observing conditions. Using a cold load as a blackbody, the authors measure optical responsivity as the slope of fractional frequency shift versus incident optical power, map phase timestreams to frequency fluctuations, and compute NEP as the ratio of fractional-frequency noise spectral density to responsivity. They find that a tone detuned by half a linewidth, corresponding to roughly 0.2 pW of optical-power change, shows a 1","pith_inferences":["Beyond the paper, the 0.2 pW / 1.5–2x threshold provides a concrete input for a tone-reset control loop: any detected loading change of that size warrants a retune. The paper documents the penalty but does not propose a control algorithm.","A testable extension: if the readout continuously re-centers tones using a resonance-tracking loop, the drift penalty should disappear; the lab data here give the target performance such a loop must meet.","If the stray-light interpretation is correct, the reported absolute responsivities are upper limits and the NEPs are lower limits, which may change how the TiN and Al arrays compare and how the array sensitivity is projected."],"forward_implications":["Prime-Cam will need a tone-retuning plan: a 0.2 pW optical loading drift multiplies NEP by about 1.5–2 at half-linewidth detuning.","Operating readout tones at lower power is costly: 10 dB below the optimum degrades NEP by up to 200%.","Aluminum KIDs require a nonlinear responsivity model (lookup tables or fitted curves); a single linear calibration will bias photometry.","TiN and Al arrays will need distinct calibration and optimal tone settings because their responsivity and nonlinearity differ.","The cold-load characterization procedure will be used to set tone powers and reset cadence during Prime-Cam commissioning."],"fun_headline_variants":["0.2 pW tone shift doubles Prime-Cam detector noise","Half-linewidth detuning doubles KID noise at 280 GHz","Readout tone drift doubles noise in CCAT KID arrays","Small optical drift doubles noise in Prime-Cam detectors"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The results rest on the assumption that the optical power incident on the detectors is known from the cold-load temperature, an assumed emissivity of one, known filter transmission, and a light-tight detector enclosure; if stray light enters, the inferred responsivities are too high and NEPs too low.","fun_headline_variants_meta":{"raw":{"variants":["0.2 pW tone shift doubles Prime-Cam detector noise","Half-linewidth detuning doubles KID noise at 280 GHz","Readout tone drift doubles noise in CCAT KID arrays","Small optical drift doubles noise in Prime-Cam detectors"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000617,"raw_usage":{"total_tokens":2751,"prompt_tokens":843,"completion_tokens":1908,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":587,"completion_tokens_details":{"reasoning_tokens":1837}},"tokens_in":587,"tokens_out":1908,"duration_ms":10695,"temperature":1.0,"reasoning_tokens":1837,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T09:59:11.241636+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"At fixed cold-load temperature, shift a readout tone by half a linewidth and measure NEP; the paper's claim predicts a 1.5–2x degradation, so a substantially smaller penalty would indicate overestimated detuning sensitivity. Separately, resealing the detector enclosure to eliminate stray light and remeasuring responsivity would settle whether the absolute optical calibration is inflated.","supporting_citations":[],"review_version":1}