Sampling Rate as a Deployment Constraint: Separating Bandwidth from Windowing in Spectral-Shift Evaluation of Power-Quality Waveforms

The sampling rate a feeder monitor runs at is a deployment constraint before it is a signal-processing choice. It fixes how much memory a record occupies, how much telemetry the link carries and how much energy the meter spends per hour of continuous monitoring, and it is fixed when the hardware is specified rather than at inference time. Choosing it is therefore a benchmarking-and-evaluation problem in a deployment-constrained setting, and a minimal-supervision one, because the ranking has to be produced without labelling a test set at any candidate rate. The evidence currently offered is a degradation curve, frozen time-series foundation model accuracy against sampling rate, which bundles bandwidth removal with the evaluation's own windowing and reports only their sum. We propose a protocol that separates the two, pairing a decimation ladder with a band-limiting ladder at fixed rate, both read through a window held fixed in seconds rather than samples, with a linear physical-feature baseline and randomly initialized twins as controls. Under that correction, all three models still fall steeply on both ladders, so the loss survives holding duration and visibility fixed. However, once duration is held fixed, the model-dependent sign reversal an uncorrected ladder comparison reports disappears, so how the window is specified mainly decides the disagreement. A label-free statistic computable from the sampling specification alone, harmonic order retained, ranks monitoring regimes out of sample here and on a second generator, but does not size the loss to the accuracy registered in advance. A spectral-shift evaluation therefore cannot be read as spectral, or used to tell a utility which monitoring rate a feeder needs, unless it reports window placement, window duration in seconds, and event-visibility fraction at every rung.