{"id":"c11b11a9-e06d-41c7-98c4-40e730d5124e","arxiv_id":"2604.11645","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Scalability of frequency-selective RF wireless power for multi-robot magnetic actuation is governed by resonator Q-factor; three cm-scale actuators were selectively triggered at 734, 785, and 855 kHz without cross-actuation.","lead":"This paper shows that how many tiny wireless magnetic robots you can control independently in one space is set mainly by each robot's resonator Q-factor, and demonstrates three centimeter-scale devices that fire only at their own RF frequencies. It matters for anyone building swarms of battery-free magnetic robots that must share one transmitter without cross-talk.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.5","headline":"Loaded trigger Q, not datasheet Q, sets usable packing density; large-N claim is an unloaded upper bound.","rationale":"The reader correctly isolates the load-bearing gap: the packing analysis that underwrites the scalability claim uses inductor datasheet Q, while the function that must remain selective (triggering) operates at substantially lower loaded Q. That mismatch is not a peripheral implementation detail; it directly scales the half-power bandwidth that enters Eqs. 7–8 and therefore the headline number of addressable robots. The three-device force and selectivity experiments still support the qualitative claim that frequency-selective WPT can drive independent magnetic actuation without cross-triggering, so the paper remains accept-shaped as a proof of concept once large-N statements are framed as upper bounds under high loaded Q (exactly the reader’s CONDITIONAL). No stronger internal inconsistency appears: the theory is standard resonator packing, the force and beam-deflection results are consistent with the circuit, and the authors already flag MOSFET de-loading and gate damping as next steps. The concrete re-packing check settles whether the numerical claim needs revision without requiring new hardware.","tokens_in":10919,"tokens_out":736,"duration_ms":8851,"concrete_test":"Re-run the discrete packing of Eqs. 7–8 over 100 kHz–1 MHz with Δf(f) taken from the measured trigger Q values in Table II (or from a constant Q_eff ≈ 13) instead of the datasheet Q(f) of Fig. 3; report the resulting N and the fractional reduction relative to the 177-resonator figure. If N drops by more than ~3×, the large-N claim must be restated as an unloaded upper bound contingent on MOSFET (or equivalent) de-loading.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that N is set by packing half-power bandwidths Δf ≈ f0/Q (Eqs. 4–8) so that adjacent resonators plus guard band do not overlap, and that scalability therefore depends primarily on Q. Section III and Fig. 2 compute ~177 addressable resonators over 100 kHz–1 MHz using a fixed 10 µH inductor and datasheet-derived Q(f)/series-loss model (Eqs. 5–6, 9; Fig. 3). The experimental trigger path, however, is resistively loaded by the BJT base: Table II reports trigger Q ≈ 12–14 (Δf ≈ 55–60 kHz) versus charger Q ≈ 55–57 (Δf ≈ 20 kHz). Fig. 7 confirms measured activation bands of ~60 kHz with 10–20 kHz overlaps among only three devices. Because the selective addressability that the paper claims to scale is the trigger function, not the charger, the governing Q is the loaded trigger Q. Using unloaded datasheet Q therefore overstates packing density by roughly the ratio of those Qs (order 4×). The three-device demo remains valid as a proof of concept, but the quantitative scalability result in Section III is an optimistic upper bound that does not yet apply to the demonstrated architecture.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper claims that the scalability of frequency-selective wireless power transfer for independent untethered magnetic actuation is set primarily by resonator Q-factor. It derives a packing rule (Eqs. 4–8) that places LC center frequencies so that adjacent half-power bandwidths Δf ≈ f0/Q plus a guard band do not overlap, and uses a fixed-inductor, datasheet-based Rs(f)/Q(f) model to estimate how many resonators fit in 100 kHz–1 MHz (Fig. 2, ~177 with L = 10 µH). Three centimeter-scale prototypes with shared charging resonators and distinct trigger resonators (measured f0 ≈ 734, 785, 855 kHz) convert harvested RF energy into EM-coil actuation of magnetic cantilever beams, with measured force peaks of 60–70 mN, quantified activation bands, and limited spectral overlap without unintended cross-triggering.","tokens_in":11262,"tokens_out":816,"duration_ms":7898,"significance":"If the Q-limited packing argument holds for the actual selective path, the work supplies a concrete design equation and experimental template for scaling multi-robot RF magnetic actuation in a shared workspace—an open problem relative to prior spatial-field and step-out methods. Strengths include independent experimental characterization (Table II, Figs. 4, 6, 7 with n=3 error bars), a falsifiable packing criterion, and explicit discussion of loaded-Q and miniaturization limits. The three-device selective beam actuation is a solid proof of concept. The quantitative large-N claim is currently an unloaded upper bound rather than a validated system capacity, so the main contribution is the framework plus demonstration rather than a settled count of addressable robots.","major_comments":[{"comment":"Section III / Fig. 2 vs. Section IV-B / Table II: The central scalability claim (N set by packing Δf ≈ f0/Q) is computed with datasheet-derived unloaded Q(f) for a fixed 10 µH inductor (~177 resonators). The demonstrated selective path is the trigger resonator, which is resistively loaded by the BJT base and measures Q ≈ 12–14 (Δf ≈ 55–60 kHz) versus charger Q ≈ 55–57. Fig. 7’s ~60 kHz activation bands and 10–20 kHz overlaps among only three devices are consistent with loaded Q. Because addressability is the trigger function, the governing Q is loaded trigger Q; the Section III count is an optimistic upper bound (roughly 4× overstated). Either recompute N with measured/loaded Q (or a MOSFET high-Z model) or clearly reframe Fig. 2 as an unloaded ceiling and give a loaded-Q packing estimate for the demonstrated architecture.","section":null},{"comment":"Eq. (7)–(8) and experimental selectivity: The packing rule treats non-overlap of half-power bandwidths plus guard ef as sufficient for independent addressability. Fig. 7 shows finite 10–20 kHz trigger overlaps yet reports no unintended cross-triggering under sequential one-second charge/trigger cycles. The manuscript does not define the decision threshold (e.g., C4 discharge onset vs. force threshold), simultaneous multi-tone excitation, or how ef should be chosen from measured overlap. Without that link, the claim that the packing rule quantifies “reliable” addressability for large N remains incompletely validated by the three-device demo.","section":null}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The useful core is the hardware: three centimeter-scale charge-then-trigger units that harvest RF, store it, and fire an onboard EM coil to deflect magnetic cantilevers at 734, 785, and 855 kHz with measured forces of 60–70 mN and only 10–20 kHz activation overlap. That is a clean, reproducible proof that frequency-selective WPT can produce independent magnetic actuation in a shared workspace. Parts, measured f0/Δf/Q/energy (Table II, Fig. 4 with n=3), force traces, and selectivity bands are all there.\n\nWhat is new is the packaging of that architecture with an explicit packing rule (Eqs. 4–8) that says N is set by half-power bandwidths plus a guard. The underlying Δf ≈ f0/Q relation and discrete spectral packing are textbook RF; the paper’s own citations already show frequency-selective wireless actuation (Boyvat SMA, Song LCE, Takeuchi electrostatic, etc.). The contribution is applying the packing language to this magnetic-robot setting and shipping a working three-unit demo, not inventing the packing idea.\n\nThe soft spot is real but proportional. Section III and Fig. 2 claim ~177 resonators using a fixed 10 µH inductor and datasheet Q(f). The actual trigger path is BJT-base loaded: measured trigger Q ≈ 12–14 versus charger Q ≈ 55. Because selectivity lives on the trigger, the governing Q is the loaded one, so the large-N number is an optimistic unloaded upper bound. The authors themselves flag MOSFET replacement and loading reduction in the discussion, so they are not hiding it; they just still lead with the high-N figure. Force ringing/retriggering is a minor engineering note they already plan to damp.\n\nMath and citations are fine for an eess.SY systems paper; circularity is low. This is for people building multi-robot magnetic or RF-powered actuators who need a concrete charge/trigger reference and a reminder that loaded Q, not datasheet Q, sets density. It deserves a serious referee. I would engage, cite the demo and the measured bandwidths, and treat the 177-resonator plot as an upper-bound design target rather than a claim about the present circuit.","headline":"Solid three-device RF-triggered magnetic actuator demo; the Q-packing math is standard RF and the large-N figure is an unloaded upper bound, not the loaded architecture they built.","tokens_in":11912,"tokens_out":563,"would_cite":true,"duration_ms":5094,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"How many untethered magnetic robots you can address with one RF field is set mainly by resonator Q-factor.","keywords":["frequency-selective wireless power transfer","LC resonators","Q-factor","untethered magnetic actuation","scalability","RF energy harvesting","cross-triggering"],"falsifier":"Build or simulate a larger array using the paper’s packing rule but with the measured loaded trigger Q (≈12–14) instead of datasheet Q; if adjacent devices then show substantial cross-triggering or the measured N falls far below the predicted count, the primary-Q claim is overstated.","tokens_in":11741,"feed_emoji":"📡","tokens_out":801,"duration_ms":8068,"temperature":0.7,"pith_summary":"This paper asks how many independent untethered magnetic actuators can share one radio-frequency workspace without cross-triggering. The authors show that the answer is governed by the quality factor of each LC resonator: sharper resonance (higher Q) means narrower half-power bandwidth, so more center frequencies can be packed into a fixed band without overlap. They write design equations that convert frequency-dependent series resistance into bandwidths and an addressable count N, analyze packing from 100 kHz to 1 MHz, and build three centimeter-scale devices that harvest energy, then fire an electromagnetic pulse that bends a magnetic cantilever at 734, 785, and 855 kHz. Force and bandwidth measurements confirm selective mechanical motion with only small spectral overlap. The result matters because it turns an open scaling question into a concrete Q-optimization problem for fleets of battery-free magnetic robots.","feed_headline":"Q-factor sets how many robots one RF field can address","feed_subtitle":"Three selective magnetic actuators prove packing by half-power bandwidth works in 100 kHz–1 MHz.","key_machinery":"The discrete packing rule (Eqs. 4–8): fi+1 − fi ≥ Δf(fi)/2 + Δf(fi+1)/2 + ef, with Δf drawn from frequency-dependent series resistance Rs and effective Q, which converts Q into the maximum addressable count N inside a fixed spectrum.","core_discovery":"Scalability of frequency-selective wireless power transfer for untethered magnetic actuation depends primarily on resonator Q-factor: the maximum number of individually addressable LC harvesters inside a fixed RF band is set by spacing center frequencies so that adjacent half-power bandwidths (Δf ≈ f0/Q) plus a guard band do not overlap, and three fabricated actuators convert that selectively harvested energy into mechanical beam motion at distinct frequencies with no unintended cross-triggering.","pith_inferences":["The same packing rule could be applied to other frequency-selective WPT loads (SMA heaters, electrostatic actuators) once their loaded Q is measured.","Onboard energy storage that decouples charging from triggering would free the high-Q path for pure frequency selection and ease further size reduction.","If loaded Q stays near 12–14, the practical fleet size in this band is closer to a few tens than the 177-resonator computational upper bound."],"forward_implications":["Raising loaded Q (lower Rs, less circuit loading) directly increases the number of robots addressable in 100 kHz–1 MHz.","Miniaturization that shrinks inductance without preserving Q will widen bandwidths and shrink the addressable population.","Design equations give a concrete target for component choice and guard-band allocation before fabrication.","Replacing the BJT with a high-impedance MOSFET gate would raise trigger Q and improve both selectivity and range."],"fun_headline_variants":["Q-factor sets packing density of untethered magnetic actuators","Half-power bandwidths limit addressable LC harvesters per RF band","Three selective actuators prove Q rules multi-robot RF magnetic drive","Resonator Q dictates max robots one frequency-selective field can address","Scalability of selective wireless magnetic actuation hinges on Q-factor"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The large packing numbers rest on datasheet inductor Q, while the actual trigger circuits are heavily loaded by the transistor base and measure much lower Q, so the usable bandwidths may be wider than the scaling equations assume.","fun_headline_variants_meta":{"raw":{"variants":["Q-factor sets packing density of untethered magnetic actuators","Half-power bandwidths limit addressable LC harvesters per RF band","Three selective actuators prove Q rules multi-robot RF magnetic drive","Resonator Q dictates max robots one frequency-selective field can address","Scalability of selective wireless magnetic actuation hinges on Q-factor"]},"model":"grok-4.5","effort":"low","cost_usd":0.004626,"raw_usage":{"total_tokens":1379,"prompt_tokens":816,"num_sources_used":0,"completion_tokens":74,"cost_in_usd_ticks":46260000,"prompt_tokens_details":{"text_tokens":816,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":489,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":816,"tokens_out":74,"duration_ms":4658,"temperature":1.0,"reasoning_tokens":489,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T19:43:39.755349+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Build or simulate a larger array using the paper’s packing rule but with the measured loaded trigger Q (≈12–14) instead of datasheet Q; if adjacent devices then show substantial cross-triggering or the measured N falls far below the predicted count, the primary-Q claim is overstated.","supporting_citations":[],"review_version":2}