Another Moog Enters the Conversation

To our surprise, it is now two engineers named “Moog” that inform this framework’s narrative in some way or other. Robert Moog built the synthesizers that carry the family name into the music industry. Houston Haynes, designer of this framework, its language, and its type system, began his career as Bob’s student and later helped him restart “Moog Music” as an international brand. And now, it also gains some perspective from William Moog, Bob’s cousin, who founded the motion-control company in East Aurora, New York in 1951. His electrohydraulic servo valve and by extension that company became a fixture of precision control in aerospace. The two sides of that family story connected only recently. Houston encountered the company’s published engineering in the course of separate research, after this framework’s design had taken shape. This entry is about control systems and precision: one lesson taught in a classroom, and a later recognition found in print.

Houston’s time in Bob’s shop covered a wide variety of hardware and software engineering, including high-frequency analog circuit work and, later, wired cascading interrupt structures across Zilog processors. But before that Houston was one of Bob’s students at the University of North Carolina, Asheville. One seminal lesson that started in those classes, with Bob as research professor, stands out above others. It points to where engineering answers can be found in sound mathematics.

The Ring Modulator Lesson

A ring modulator’s signal path runs through transformers, filters, and lines whose behavior over time is governed by the Telegrapher’s equations. For a voltage $v(x,t)$:

The equation is precise and, at a workbench, unusable: in the time domain there is no practical route from that PDE to the capacitor rating the circuit needs.

Euler’s formula changes the domain rather than the physics. In the case of a ring modulator, the opening assumption is that carrier and modulator signals are sinusoidal, so substitute $e^{j\omega t} = \cos(\omega t) + j\sin(\omega t)$. The derivative of $e^{j\omega t}$ is $j\omega \, e^{j\omega t}$, so differentiation becomes multiplication by $j\omega$, the calculus collapses into algebra, and the capacitor’s full behavior reduces to a complex impedance:

In this domain the component values fall out of the equations. The series capacitor that blocks DC at the input follows from $f_c = \frac{1}{2\pi R C}$: with a $100\text{ k}\Omega$ input impedance and a $15\text{ Hz}$ cutoff,

and a workbench with a decent supply of components would provide a standard $100\text{ nF}$ film part. The parallel capacitor that shorts a $25\text{ kHz}$ carrier leak to ground follows from the same relation with a cutoff near $15\text{ kHz}$, which puts the part between $1$ and $4.7\text{ nF}$ ceramic.

The principle is larger than the blackboard lesson, and is the reason Houston retells the story. The PDE and the algebra describe the same circuit. In one representation the answer is unreachable at any reasonable cost. In the other it falls out figuratively speaking, carrying a derivation any engineer can use for component selection.

The Other Moog

In somewhat a parallel history, William Moog’s servo valve turned milliamps of electrical signal into precisely metered hydraulic force, with feedback closing the loop inside the device itself, and that component made high-authority flight control practical. The company grew into a fixture of precision motion control across aerospace: aircraft flight surfaces, launch vehicle steering, spacecraft actuation. What Houston found, decades after the bench years, spans two generations of the company’s published engineering: the servo-valve transfer-function bulletins, and the digital-thread program it documents today.

The first generation is the servo-valve literature, and its centerpiece is Transfer Functions for Moog Servovalves, Technical Bulletin 103, written by W. J. Thayer in 1958 and revised in 1965. The bulletin taught control engineers to work with the valve as a frequency-domain transfer function, and for most system design a first-order approximation served:

Flow $Q$ per unit of drive current $I$, a flow gain $K_v$, and a single time constant $\tau$: enough algebra to size a control loop the way the bench algebra sized a capacitor. Where a design pressed closer to the valve’s dynamics, the bulletins supplied the second-order form, with damping ratio and natural frequency stated per valve family:

Substitute $s = j\omega$ and this is the bench lesson again: differential equations exchanged for algebra, with each approximation’s range of validity stated so a customer’s engineer could check every derivation before an aircraft depended on the part. We take the deeper precedent from the practice itself: a component vendor publishing checkable mathematical models of its own products, so that the model is part of the product and verification is the customer’s right rather than the vendor’s favor.

The lesson here runs deep. Moog is old enough to have run every generation of manufacturing software: paper travelers, then MRP in the 1970s and 80s, then ERP through the 1990s and 2000s, then product-lifecycle management, then digitized shop-floor execution. Each system holds authority over something real: PLM is the design, MES is the build record, ERP is the digital ledger. Run separately, the three systems assign the same physical part three different numbers. The divergence stays invisible until a mismatch puts the wrong part in an assembly. A seventy-year-old company cannot replace all of this in one motion. It can only thread the systems together. This is the object lesson that we found sympathetic to the Fidelity Framework’s posture.

Moog’s aerospace business publicly documents how it threaded them. Teamcenter, the as-designed authority, synchronizes to Solumina execution systems in the US and UK and to SAP S/4HANA on the enterprise leg. What crosses between systems is identity and linkage only: parts and EBOM references, change orders, quality clauses, document links, and trade-compliance attribution. The identities are held in an Object-Relationship Store, the bulk data never moves, and each system remains authoritative for its own records. The integration deployed in months with no components installed into the endpoint systems, and because the identifiers survive, the thread is rebuildable. Over such a spine, configuration states can line up as effectivity-dated baselines: as-designed, as-planned, as-built, as-maintained, as-flown.

  flowchart TB
    TC["Teamcenter (PLM)<br/>as-designed authority"]
    SOL["Solumina (MES, US & UK)<br/>as-built authority"]
    SAP["SAP S/4HANA (ERP)<br/>enterprise authority"]
    ORS["Object-Relationship Store<br/>identities and linkages only"]
    TC --- ORS
    SOL --- ORS
    SAP --- ORS

The director of product-lifecycle management for Moog Aircraft Group presented its framing, end-to-end traceability as an element of company transformation, at CIMdata’s PLM Road Map in May 2023, and a Moog PLM architect presents the Aerospace & Defense PLM Action Group’s digital-twin and digital-thread benchmark on the group’s behalf, whose catalog of eighty use cases published in 2026 with twenty-eight demonstrated against commercially available software.

Note what rides the thread alongside the part numbers: trade-compliance attribution, the export-control dimension carried as first-class identity as opposed to being bolted-on after the fact.

From valve dynamics in $j\omega$ to configuration identity spanning as-designed to as-flown, the discipline is continuous across seventy years: choose the representation in which the answer is checkable, keep one authoritative identity for every fact, and publish enough of the model that an independent engineer can verify it. From the workbench, Houston found Bob’s regard for his cousin easy to understand. It is the same standard, held at two scales.

The Tractable Domain

One of these lessons shows up in our Fidelity Framework at its foundation. Clef’s dimensional types descend from Andrew Kennedy’s units of measure, and our published type system work keeps that algebra in fragments with the phasor domain’s character: dimensional consistency is exact integer arithmetic over an abelian group, value ranges propagate as interval algebra through the compiler’s program graph, and the verification tiers are drawn so a solver decides the everyday obligations quickly and unattended.

That is the Euler substitution made permanent.

The mathematics was placed, deliberately, where answers fall out, and the tooling is being built so they arrive at the point of writing. Our commitments were set before Houston ever read Moog’s digital-thread documentation: BAREWire holds both sides of every boundary to one checked contract, and the certificate design attaches a labeled derivation to every discharged obligation, so what the compiler asserts stays checkable by engineers, by toolchains, and in time by auditors.

Finding the same commitments, independently, in the published practice of the other Moog’s company is the kind of confirmation an engineer trusts precisely because it emerges from sympathetic principles. We are gratified that our own path found the adjacency, and the companion entry on our Braidpoint site, The French Connection, follows the surrounding evidence tradition through the certification tooling regulators already trust.

Houston took the standard from one Moog in person and, decades later, found it again in the other’s company in print. We are building this framework to hold that standard: models in tractable domains, answers that carry integrity, and identity kept precise enough that the next engineer, or the next system, can verify and trust the work.