Skip to content

Meredith effect

Aerodynamic recovery of some cooling-system drag when waste heat added to a shaped duct's airflow increases downstream momentum, subject to pressure and installation losses.

Version
v1 · 2026-10-07 · History
Domain-specific #
13941
Domain group
Applied Sciences & Engineering
Origin domain
Aviation & Aeronautics
Subdomains
Aircraft Cooling, Aerothermal Design → Aviation & Aeronautics

Core Idea

The Meredith effect is the recovery of some of an aircraft cooling installation's aerodynamic drag when waste heat transferred into a moving duct airflow contributes to downstream momentum. A shaped duct admits and slows air for a heat exchanger, then lets the heated flow leave with a different pressure and velocity balance. The exchanger and duct also impose losses. The effect therefore names a heat-driven recovery mechanism, not a promise that every cooling installation produces net thrust.[1][2]

F. W. Meredith analyzed ducted aircraft radiators as a way to reduce cooling drag. Later tests and design studies distinguish a possible energy-recovery mechanism from its practical magnitude. Silverstein's 1939 model experiments found partial recovery but judged the recovery not of practical importance in the tested configurations up to 400 mph. A modern computational study found a larger modeled benefit under its own design conditions; neither result is a universal performance law.[1][3][2]

Structural Signature

  • Moving cooling airflow and duct. A flow path admits air and shapes its speed and pressure through the cooling installation. Without the flow path, the specified duct-momentum recovery is absent.
  • Heat exchanger receiving rejected heat. Heat from a propulsion or other onboard system transfers to the air. If no heat is added, the heat-driven part of the recovery mechanism cannot occur.
  • Pressure loss and downstream momentum. The exchanger's resistance and the duct's external and internal losses compete with the momentum benefit of heating the flow. The net balance must count both sides rather than label all heating as thrust.
  • Operating and design conditions. Duct shape, flow rate, heat transfer, exchanger resistance, and flight condition govern whether the observed result is negligible, partial recovery, or potentially positive net thrust. No single outcome defines membership.[1][3][2]

What It Is Not

The effect is not simply an engine making thrust from fuel. Its heat input is rejected heat that the aircraft must remove; using fuel heat in a ramjet is a different energy role. It is also not any radiator carried by an aircraft. An exposed radiator with no designed duct flow and momentum recovery lacks the defining flow relation.[1][2]

The effect does not guarantee zero cooling drag or positive thrust. Silverstein found some heat recovery but concluded that its practical importance was limited for the geometries and speeds tested. Adler and colleagues' modeled drag reduction compares one fixed duct with and without heat transfer, rather than proving that any flown aircraft gains propulsive thrust.[3][2]

Scope of Application

The literal habitat is aircraft thermal management where a moving airflow passes through a heat exchanger in a designed duct and exits with a measurable aerodynamic balance. Meredith's original analysis concerned an enclosed cooling radiator. Silverstein experimentally examined expanded wing ducts using an electrical resistance heater to simulate a radiator's heat addition; that surrogate tests the mechanism but is not an operating engine-waste-heat installation.[1][3]

A contemporary habitat is aircraft design simulation. Adler, Lamkin, and Martins modeled a ducted heat exchanger for a fuel-cell propeller-aircraft concept and optimized its flow path. Their high-speed cruise result is a computed comparison at a design point, not a flight-test result. Similar hardware in another vehicle would require its own heat-source, airflow, and loss accounting before the named effect could be asserted literally.[2]

Clarity

The abstraction separates heat-driven recovery from net installation performance. A warm outlet may contribute momentum while the whole cooling system still produces drag. Conversely, a drag reduction due to heat transfer need not mean the aircraft has net forward thrust from its radiator. The correct comparison fixes the duct and flight condition, then asks what changes when heat transfer is present.[3][2]

It also separates kinds of evidence. Meredith's analysis supplies the design proposition, Silverstein's electrical surrogate provides a limited physical test, and Adler and colleagues provide a modern computational design comparison. Each can support the mechanism within its own scope; none should be silently promoted to fleet-wide flight performance.[1][3][2]

Manages Complexity

A cooling installation involves inlet losses, heat-exchanger resistance, heat-transfer rate, duct shape, external drag, and outlet momentum. The concept compresses the first analysis to an energy and momentum balance: how much rejected heat reaches the airflow, how much useful downstream momentum results, and what pressure and installation losses offset it? This common accounting makes the 1939 experiment and the 2024 simulation comparable without treating their aircraft, geometries, or evidence types as interchangeable.[3][2]

The compression is deliberately bounded. Adler and colleagues optimized a particular design and evaluated specific flight points. Their roughly 71% reduction at modeled high-speed cruise is a within-design heat-transfer comparison, not a universal coefficient to apply to a different exchanger or duct.[2]

Abstract Reasoning

To assess a Meredith-effect claim, first locate the rejected heat source and the moving ducted air that receives it. Then identify where the flow slows, gains heat, and exits. Compare outlet momentum and pressure recovery against exchanger resistance and the other duct losses at the stated flight condition. If the heated case improves the balance over the same unheated duct, that supports heat-driven recovery; a separate claim of net thrust requires the full installation balance.[1][2]

Evidence should be typed before generalization. An electrical heater in a model can probe the heat-addition mechanism without demonstrating actual engine integration. A CFD design can reveal sensitivities and candidate geometry without establishing flight performance. Different speed or cooling demand may change the sign and value of the benefit.[3][2]

Knowledge Transfer

Within aircraft cooling, the same role map can guide analysis of historical radiator ducts and proposed electric-aircraft heat-exchanger installations: identify rejected heat, shaped airflow, pressure loss, and downstream momentum. The mechanisms can be compared literally even when the cooling source and geometry differ.[1][2]

The live Heat Engine names a cyclic heat-to-work system; this cooling-duct effect does not presuppose that full cycle. Generic waste-heat recovery is broader. The Meredith effect retains its aircraft cooling-duct flow relation. Applying its name to unrelated heat recovery without that aerodynamic balance would be analogy rather than demonstrated transfer.

Examples

Canonical test: Silverstein's heated wing-duct models

Silverstein tested expanded wing cooling ducts with an electrical resistance unit that simulated a radiator's heat input. The experiments found some recovery from heat addition, while the report judged that recovery not of practical importance in the tested arrangements up to 400 mph. This is an experimental surrogate of the mechanism, not a demonstration of a flying aircraft's rejected engine heat.[3]

Mapped back: expanded wing ducts supply the moving cooling airflow and duct; the resistance heater is a surrogate for rejected heat transfer; measured drag and recovery address pressure loss and downstream momentum; the tested geometries and speeds are the operating and design conditions. The surrogate role is explicit so the test is not mistaken for an installed radiator case.

Applied design: simulated fuel-cell-aircraft heat exchanger

Adler, Lamkin, and Martins simulated a ducted heat exchanger for a fuel-cell propeller-aircraft concept. At their modeled high-speed cruise point, the same fixed optimized duct had computed drag of 265 N without heat transfer and 77 N with heat transfer, about a 71% reduction. Their paper also compares separately optimized designs; that is a different comparator from this within-duct result.[2]

Mapped back: the optimized inlet and downstream flow path are the moving cooling airflow and duct; the modeled thermal-management exchanger adds rejected heat; computed drag with and without heat transfer tests the pressure-loss and momentum balance; the chosen aircraft and cruise point set the operating and design conditions. The result is simulation evidence, not flight confirmation.

Structural Tensions

T1: Reject enough heat vs limit aerodynamic losses. Moving more cooling air or imposing a larger heat-exchanger duty can help remove heat, while exchanger resistance and duct flow can raise drag. A design favoring cooling alone may waste aerodynamic performance; one favoring minimal drag alone may fail to cool the system. The Meredith mechanism can offset part of the cost only after heat transfer and all installation losses are accounted for. Diagnostic: At the required cooling load and flight point, how much drag does heat addition recover after exchanger and duct losses?[2]

Structural–Framed Character

The entry is mixed, leaning structural. Heat transfer, pressure loss, and airflow momentum are physical relations; the effect is not created by institutional designation. Evaluative weight enters when an engineer judges a recovery practically useful: cooling duty, accepted drag, and operating point are design choices. Human practice chooses the geometry and comparison, while physics constrains the outcome. The term began in aircraft design analysis, not a formal certification category.[1][3]

Its vocabulary travels literally among ducted aircraft cooling designs when the same heat-addition and momentum roles can be shown. Importing it to any reuse of waste heat would mistake an analogy for this aerodynamic mechanism. A broader energy-and-momentum accounting relation is a future-prime question, not an approved parent or a claim that this named effect transfers outside cooling ducts. Its character: a physically grounded aircraft-cooling mechanism whose practical value is conditional on a complete installation balance.

Structural Core vs. Domain Accent

The skeletal relation is heat entering a moving fluid and changing its momentum and loss balance. That broader relation is a future-prime question because no existing live parent passed the typed test. General thermodynamics and flow analysis can describe parts of it. The domain accent is an aircraft cooling duct, rejected onboard heat, exchanger pressure drop, and the aerodynamic question of cooling drag recovery. The live Heat Engine, Dissipation, and Transformation entries are relevant neighbors, but their typed definitions fail the necessary-parent test for this named effect.[1][2]

This entry does not clear the Prime bar. Its defining test requires the aerothermal cooling installation and its outlet momentum, not merely conversion of energy in any substrate. The current catalog lacks an exact necessary genus, so the typed challenge approves a provisional unparented root rather than forcing a broad parent into the frontmatter.

No structured parent is asserted after typed DAG review; this entry is a provisional unparented root. Heat Engine may look similar because heat can contribute to useful work, but its live identity requires a cyclic heat-to-work system absent from some cooling-duct recovery cases. Dissipation concerns energy loss, which is part of the balance rather than the genus of recovery. Transformation can describe a state change too broadly to be an informative necessary parent. These are comparisons, not approved edges. Thermodynamic Process and Energy Transformation likewise require stronger path or accounting conditions than every Meredith-effect case establishes; Flow and Cooling name a carrier or activity without subsuming the effect.

Neighborhood in Abstraction Space

Meredith effect sits in a sparse region of the domain-specific corpus (99th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2551 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With

  • Net radiator thrust: recovery can be partial and leave positive drag; positive net thrust is a conditional possible balance.
  • An exposed radiator: without shaped airflow and downstream momentum recovery, heat rejection alone is not this effect.
  • A fuel-heated ramjet: fuel is the energy source rather than rejected cooling heat.
  • A flight-proven result: Silverstein used an electrical radiator surrogate; Adler and colleagues report computational design comparisons.[3][2]

References

[1] F. W. Meredith, “Cooling of Aircraft Engines with Special Reference to Ethylene Glycol Radiators Enclosed in Ducts,” Aeronautical Research Committee Reports and Memoranda No. 1683 (1935), original report and abstract. https://reports.aerade.cranfield.ac.uk/handle/1826.2/1425 registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j

[2] Eytan J. Adler, Andrew H. R. Lamkin, and Joaquim R. R. A. Martins, “Ducted Heat Exchanger Thermal and Aerodynamic Shape Optimization,” ICAS 2024, paper 0080 (2024), especially §3.1 and Fig. 18, printed pp. 16–17, and p. 19 conclusions; Fig. 19 compares separately optimized designs. https://www.icas.org/icas_archive/icas2024/data/papers/icas2024_0080_paper.pdf registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q

[3] Abe Silverstein, “Experiments on the Recovery of Waste Heat in Cooling Ducts,” NACA Special Report 111 (May 1939), original report abstract and introduction; the introduction describes the electrical resistance unit used to simulate a radiator. https://ntrs.nasa.gov/citations/20090015245 registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k