{"document":{"slug":"filing-20260709-5023","title":"Second Supplemental Technical Comment of L.M. Marlowe on the Frequency-Response Absorption Standard and Coupling-Discrimination Protocol","issued":"2026-07-09","accession":"20260709-5023","docket":"EL26-67-000","group":"PJM, SPP, NYISO, MISO, CAISO, ISO-NE filings","filer":"Individual No Affiliation","kind":"filing","file":"sources/originals/20260709-5023_01.pdf","url":"https://elibrary.ferc.gov/eLibrary/filelist?accession_Number=20260709-5023","sha256":"a6f3babf59b0d2d1c7f92f8a31a21d890d468b0a589b8640e7c3a78a7e2fa727","pages":4,"tokens":1621,"emptyPages":[]},"pages":[{"page":1,"startToken":0,"endToken":384,"text":"UNITED STATES OF AMERICA \nBEFORE THE \nFEDERAL ENERGY REGULATORY COMMISSION \n \nIntegration of Large Loads and Associated Reliability and Cost-\nAllocation Methodologies; \nShow-Cause Proceedings and Related Dockets \n) Docket No. EL26-67-000 \n) Docket No. EL26-68-000 \n) Docket No. EL26-69-000 \n) Docket No. EL26-70-000 \n) Docket No. EL26-71-000 \n) Docket No. EL26-72-000 \n \nSECOND SUPPLEMENTAL TECHNICAL COMMENT OF L.M. MARLOWE \n(Frequency-Response Absorption Standard and Coupling-Discrimination Protocol) \nDate: July 8, 2026 \nThis submission is made in good faith by a private citizen and independent researcher, supported by extensive \noperational documentation and prior art. It is offered for the public benefit — to improve grid reliability, \ntransparency in large-load cost allocation, and the protection of ratepayers from systemic waste and undue cost-\nshifting. It provides additional technical detail responsive to the reliability and observability questions raised in \nthese proceedings, and asks that all submissions be evaluated against the full requirements of the Commission’s \nShow Cause Orders. \nLisa Melton, filing as L.M. Marlowe (“Commenter”), a party whose Motion to Intervene in Docket No. \nEL26-67-000, et al. was accepted for filing (Accession No. 202607025022, July 2, 2026), respectfully \nsubmits this Second Supplemental Technical Comment. It supplements Commenter’s prior Supplemental \nTechnical Comment with additional technical specification responsive to the reliability, observability, and \ncost-causation questions presented, and respectfully addresses the standard by which the Commission \nmay wish to evaluate the responses before it. \nI. The Standard of Evaluation Respectfully Urged \nCommenter respectfully urges the Commission, in evaluating the responses to its Show Cause Orders, to \nhold each proposed framework, methodology, and tariff to the full rigor of what those Orders require. The \nCommission’s directives ask not merely whether a proposal addresses large-load integration in some \nrespect, but whether it satisfies the just-and-reasonable standard across the complete set of concerns the \nrecord has raised: \n(1) physical reliability and dynamic stability; (2) observability at the temporal resolution the underlying \nphenomena demand; (3) cost-causation that assigns costs to those who cause them rather than socializing \nthem onto ratepayers; and (4) durability under credible worst-case operating conditions, not representative \nones. \nA submission that resolves the commercial or market-access dimension of large-load integration without \ndemonstrably addressing the physical stability and observability questions that gave rise to these \nproceedings has not, in Commenter’s respectful submission, fulfilled the full requirement of the Show \nCause Orders. Commenter asks that each submission — including Commenter’s own — be measured"},{"page":2,"startToken":384,"endToken":884,"text":"against that complete standard, so that the resolution the Commission adopts rests on frameworks that \nfulfill the whole of what reliability and consumer protection require, not a part. \nII. The Binding Contingency Has Shifted \nReserve adequacy and frequency-response planning have historically been sized against an enumerated \nlargest credible contingency — typically the loss of the single largest generating unit — treated as \nphysical and independent. Large computational load introduces a contingency of a different character: a \nset of geographically dispersed loads that, through a shared orchestration or control plane, can reduce \ndemand in near-simultaneity, behaving as a single correlated block. \nThis block is invisible to topology-based N-1 analysis, which scores dispersed loads as independent low-\nprobability events. Measured at the interconnection boundary, the same loads exhibit correlated ramp \nevents at machine speed. Where the coordinated-drop magnitude of the block exceeds the loss-of-largest-\nunit value, the largest credible contingency is no longer a generator — it is the load block. A reserve \nadequacy computed against the former, at representative system conditions, is measuring the wrong event \nat the wrong condition. \nIII. The Frequency-Response Absorption Standard (Minimum-Inertia Condition) \nA load drop pushes system frequency upward. The immediate rate of change of frequency (RoCoF), \nbefore primary frequency response (PFR) engages, is governed by the swing equation: \ndf/dt  =  − ΔP / (2 H · S_base) \nwhere ΔP is the net imbalance from the coupled-block drop, H is the system inertia constant, and S_base \nis the system base. Stability requires two conditions simultaneously: the initial RoCoF must remain below \nthe protection threshold at which relays trip generation, and the frequency excursion before PFR arrests it \nmust remain within the trip band. Both depend on H. \nSystem inertia is not constant. As synchronous generation retires and is displaced by inverter-based \nresources, H declines, reaching its lowest values under high non-synchronous output and low \nsynchronous commitment — conditions that recur on a dispatch-driven schedule. The consequence is a \ncompounding exposure: available inertia (H) shrinks while the coordinated-drop magnitude (ΔP) grows, \nboth driving RoCoF upward at once. Evaluation at average or representative inertia systematically \nunderstates the risk, because the binding failure occurs at the minimum, not the mean. \nCommenter therefore respectfully submits that the absorption of any behaviorally-coupled load block \nmust be evaluated at a credible minimum-inertia condition — the lowest inertia state expected to recur \nwithin the planning horizon at a stated low exceedance probability, adjusted for scheduled synchronous \nretirements and forecast resource mix, and crediting only firmly committed grid-forming or synchronous-\ncondenser contribution. Expressed as a checkable interconnection requirement, a coupled block is \nabsorbable only if, at that minimum-inertia condition, its resulting RoCoF remains below the protection \nthreshold and deliverable primary frequency response within the arrest window is at least equal to the \nblock magnitude. Where neither holds, there exists a coupled-load penetration beyond which further \ninterconnection is not physically feasible without defined, committed grid-strengthening measures. This is \na limit set by physics, not by policy; the standard’s function is to locate it before it is discovered by a \ncoordinated-drop event."},{"page":3,"startToken":884,"endToken":1433,"text":"IV. The Coupling-Discrimination Protocol \nA rigorous standard must distinguish the true common-mode risk from legitimate distributed flexibility, \nbecause both produce correlated load movement. Two loads may move together because a shared control \nplane commanded them (orchestration-coupling — the contingency), or because each independently \nresponded to the same public signal, such as price or system frequency (environmental correlation — the \ndemand-response behavior the system relies upon). A metric based on aggregate correlation alone cannot \nseparate them and would erroneously penalize legitimate response. \nThe protocol discriminates on signatures observable at the boundary, without knowledge of internal \ncause: the dispersion of ramp-event onset times (a shared command arrives within a narrow machine-\nspeed window; independent responses disperse over each participant’s own reaction time); the \ncongruence of ramp waveform shape (a command replicates one instruction; independent responses vary \nin shape); and, decisively, the correlation that survives conditioning on common environmental signals. \nOnly co-movement that remains after removing the price-, frequency-, dispatch-, and weather-explained \ncomponents counts toward the coupling coefficient: \nρ_coupling  =  corr( residualᵢ , residualⱼ )  after conditioning on common environmental signals \nA set of loads all responding to a price signal shows high raw correlation but near-zero residual \ncorrelation — correctly cleared as environmental. A set sharing a scheduler shows correlation that \nsurvives conditioning — correctly identified as orchestrated. Only genuinely orchestration-coupled load \nenters the block magnitude tested under Section III. This resolves each participant into a measured \nposition — coupled and subject to the absorption test, environmentally correlated and cleared, or \nindependent — so that no party is swept into a contingency obligation by mere correlation, and none \nescapes one by disputing intent. The classification is reproducible from boundary data. \nV. Remediation Is Intrinsic to the Measurement, Not Appended to It \nThis section states an engineering property of the architecture, not a policy position. In the settlement \nmeasure, the quantity that identifies unaccounted-for load and the quantity that reassigns its cost are the \nsame quantity. The measure computes the difference between declared and delivered load; that difference \nis, simultaneously and by the same arithmetic, the measured extraction and the cost-causation basis that \nreassigns it to the load that caused it. There is no intermediate discretionary step — no point at which an \noperator elects whether to reallocate — because the reallocation is not a subsequent decision applied to \nthe measurement; it is the measurement, read as a ledger entry. Detecting the extraction and correcting it \nare one operation. Value recovered is thereby directed, by the arithmetic itself, to those who bore its cost, \nto the grid infrastructure its presence stressed, and to the structural correction of the conditions that \npermitted the misallocation, so that it does not recur. \nBecause the two are one operation, accountability is not a discretionary layer that any institution is asked \nto honor by declaration; it is a property of the invariants, such that the measurement and the reassignment \ncannot be performed independently of one another. This removes the discretion through which \nunaccounted-for costs are ordinarily socialized: a system that performs the measurement has, in the same \nstep, performed the reallocation. Stated in the Commission’s terms, the protection of the ratepaying public \nhere is not an aim asserted alongside the technical work — it is a defined output of the equations, present \nor absent as a measurable quantity rather than as a declared posture. The specific mechanism for"},{"page":4,"startToken":1433,"endToken":1621,"text":"administering the resulting allocation is a matter Commenter is prepared to address in the appropriate \nforum and is not placed at issue here. \nVI. Relationship to the Record and Reservation \nThis Comment supplements and does not supersede Commenter’s prior filings. The complete \nspecifications of the frequency-response absorption standard and the coupling-discrimination protocol are \nheld by Commenter and are described here at the functional level; Commenter is prepared to describe \nthem in full in response to the Commission’s inquiries. Commenter reserves all rights previously stated, \nincluding under 18 U.S.C. § 1833(b) and the referenced USPTO filings, and asserts no legal conclusion \nherein. \nVII. Conclusion \nCommenter respectfully requests that the Commission consider these technical considerations, and that it \nhold each framework, methodology, and tariff before it — including Commenter’s own — to the full \nrigor of what the Show Cause Orders require. Commenter offers these considerations in good faith and in \nthe shared interest of a reliable Bulk Power System and of just and reasonable rates for the public the \nCommission was established to protect. \nRespectfully submitted, \n_______________________________ \nLisa Melton aka L.M. Marlowe \nlm.marlowe@pm.me \nIntervenor, pro se"}],"sections":[{"p":1,"at":0,"title":"Opening text","method":"opening text","endPage":1,"endAt":1669,"startToken":0,"endToken":219},{"p":1,"at":1669,"title":"I. The Standard of Evaluation Respectfully Urged","method":"printed heading","endPage":2,"endAt":187,"startToken":219,"endToken":412},{"p":2,"at":187,"title":"II. The Binding Contingency Has Shifted","method":"printed heading","endPage":2,"endAt":1256,"startToken":412,"endToken":560},{"p":2,"at":1256,"title":"III. The Frequency-Response Absorption Standard (Minimum-Inertia Condition)","method":"printed heading","endPage":3,"endAt":0,"startToken":560,"endToken":884},{"p":3,"at":0,"title":"IV. The Coupling-Discrimination Protocol","method":"printed heading","endPage":3,"endAt":2066,"startToken":884,"endToken":1154},{"p":3,"at":2066,"title":"V. Remediation Is Intrinsic to the Measurement, Not Appended to It","method":"printed heading","endPage":4,"endAt":144,"startToken":1154,"endToken":1456},{"p":4,"at":144,"title":"VI. Relationship to the Record and Reservation","method":"printed heading","endPage":4,"endAt":707,"startToken":1456,"endToken":1536},{"p":4,"at":707,"title":"VII. Conclusion","method":"printed heading","endPage":4,"endAt":1302,"startToken":1536,"endToken":1621}]}