Hand-considered hypotheses, 3101–3200
Aerospace and space systems: structures, propulsion, thermal protection, attitude control, orbital hardware, testing. Each entry was written on its own and given its own logical form, printed beside it, so that the entries can be seen to differ structurally and not only in subject. Where a form appears twice the second instance had to earn its place with a different mechanism. None has been tested. No entry rests on a citation or on a number presented as measured: where a magnitude appears it is an explicit estimate, and the smallest discriminating experiment is named instead.
Structures, joints and dimensionally stable hardware
Misattribution
H3101. A composite overwrap pressure vessel's stiffness begins to roll off well below burst pressure, and the roll-off is routinely charged to fibre damage. At those pressures no fibre has failed; the loss is the liner yielding and shedding load into the overwrap at a rate set by the liner's own strain hardening. The test is two vessels with identical overwrap and liners of different yield strength, where the knee should move with the liner's yield pressure and not with the fibre's strain to failure.
Bound
H3102. Facesheet wrinkling of a sandwich panel cannot be raised indefinitely by thickening the core, because past a core thickness set by the core's shear modulus against the facesheet's bending stiffness the governing failure switches to core shear at a load nearly independent of facesheet thickness. Two cores of equal density and different shear modulus, at three thicknesses each, put the switch where the bound says it is.
Reversal
H3103. Raising a launch lock's preload should make the mounted unit quieter, and it does until the clamp's contact stiffness overtakes the isolator's; past that point the lock shorts the isolation path and transmitted microvibration rises with preload. Sweeping preload on one unit while measuring interface force above the lock finds a minimum rather than a monotone curve. The minimum's position should track the isolator's stiffness rather than the lock's.
Hidden term
H3104. Every damping ratio measured on a spacecraft panel in air carries a term from acoustic radiation into the surrounding gas, and for a thin facesheet that term can rival the material's own damping. The vacuum value should therefore be lower by a factor computable from radiation efficiency and gas density, of order two for a thin aluminium skin at a few hundred hertz (estimate). A modal survey in air and then in a vacuum chamber, on the same suspension, separates the two without touching the panel.
Threshold
H3105. Below a corner radius of roughly five to eight ply thicknesses (estimate) a composite angle bracket's failure load stops scaling with its laminate and becomes a property of the radius filler, which carries through-thickness load in a direction the plies do not. A radius series at fixed laminate, with the filler's own modulus varied, shows whether strength follows the radius or the noodle.
Count
H3106. A deployed truss has as many independent load paths as it has redundant bays, so at most that many joint stiffnesses can be identified from its global modes, and a model updating more joints than there are paths is fitting sensor placement. The tell is that the fitted stiffnesses change when the accelerometer set changes while the modes do not.
Scale argument
H3107. Below a wall of roughly ten ply thicknesses, a composite tube's bending neutral axis migrates toward the stiffer side of an unbalanced layup by an amount growing as the inverse of wall thickness. Thin tubes are therefore not described by laminate theory's stated bending stiffness even where the layup looks symmetric in extent. A thickness series at constant diameter and layup puts the migration on a curve, and the curve's knee is the real limit of the design rule.
Confound
H3108. Drift of an optical bench is normally attributed to thermal gradients, but on a composite bench much of the first weeks' drift is moisture desorption, which is slow, monotone, and largely indifferent to the thermal cycling that gets the blame. The two are separable because desorption continues under isothermal dry nitrogen and stops when the same bench is thermally cycled inside a sealed dry enclosure.
Path dependence
H3109. With clearance-fit bolts the load a joint carries is set by which bolt seats first, so measured stiffness depends on assembly sequence and not only on geometry. Two joints from identical parts in opposite bolt orders should differ by more than the part scatter, and re-torquing in reverse order should move the stiffness without changing anything else.
Aliasing
H3110. High-frequency peaks in a pressurized module's acoustic response are frequently reported as panel modes, but a closed volume's cavity modes fall in the same band. Both families are excited by the same source, which is why the attribution is so often made. The discriminator is the fill gas: cavity modes move with sound speed and panel modes do not, so an air-to-nitrogen swap at constant pressure shifts one family and leaves the other.
No-go claim
H3111. A single-stage passive isolator cannot be soft enough for on-orbit jitter and stiff enough for launch, because the launch case sets a minimum stiffness that caps isolation at a frequency usually above the wheel's disturbance band. Any design claiming both is either relying on a launch lock that changes the boundary condition, or reporting isolation where the wheel does not excite. Testing one isolator with and without its lock, across the same band, shows the cap.
Conservation audit
H3112. A pyroshock's high-frequency content is the strain energy stored in the preloaded bolt, released into the structure in a time set by the bolt's own wave transit. Spectrum amplitude above a few kilohertz should therefore scale with preload rather than with charge size. Firing one charge against three preloads tests it, and a preload-independent shock would falsify the account.
Degeneracy
H3113. Two laminates can share an extensional stiffness matrix exactly and differ in their coupling and bending matrices, so a coupon programme reporting only in-plane moduli cannot distinguish them. The pair is degenerate for every in-plane test and separated by a flatness check after cure, which is why the discriminating measurement is warpage rather than another tensile coupon.
Symmetry argument
H3114. A symmetric reflector's thermal distortion vanishes only if its mount shares the dish's expansion, and the mount's presence breaks the symmetry the layup was designed to preserve, so distortion follows the mount's coefficient rather than the dish's. One dish on two mount materials, with the dish's own gradients held fixed, shows the figure moving with the mount and not with the laminate.
Order of limits
H3115. With unilateral contacts, applying preload before thermal load and thermal load before preload give different final stress states, because the contact set differs along the two paths; standard practice computes them separately and superposes, which is the commuted answer. A structure with a gapped joint, loaded in both orders to the same final temperature and preload, distinguishes them by the gap's final opening.
Two-clock race
H3116. A sealed cavity vented through a small orifice sees its worst differential pressure after maximum aerodynamic pressure rather than at it, because internal pressure lags the external drop by the vent's diffusion time. The lag is the race between the vent's time constant and the ascent profile, and the panel's design case is the product of the two. Two vent areas on one cavity move the peak's time and amplitude as that product predicts.
Dimensional argument
H3117. A scale model preserving modal frequency cannot also preserve acoustic coincidence, because one scales with length and the other with length times the gas sound speed, so a one-third model misses coincidence by a known factor unless the gas is changed. Testing the model in a heavy-gas mixture chosen to restore the ratio, with the same article in air as control, is the discriminating experiment.
Null claim
H3118. Reverberant acoustic tests in two chambers should agree above the chamber's Schroeder frequency and disagree below it, because below that frequency the field is a few modes rather than diffuse. Testing it needs one article in both chambers on the same suspension, and that comparison would also expose any facility-dependent margin.
Memory claim
H3119. A composite tube returned to its reference temperature after a thermal cycle does not return to its reference strain. The residual set is a function of the peak gradient rather than the peak temperature, because the gradient loads the matrix while temperature only expands it. Cycling to one peak temperature at two gradients, and measuring the set after each return to reference, separates the two.
Non-identifiability
H3120. From a modal survey alone a joint's stiffness and its damping cannot both be identified, because a softer joint with light damping and a stiffer one with heavy damping produce nearly the same transfer function at the measured points. The pair becomes identifiable only when the excitation level changes, since interface damping is amplitude-dependent and stiffness is not; a two-level sweep is the test.
Reciprocity
H3121. Reciprocity should fail measurably across a slipping bolted interface, because the slip dissipates differently depending on which side is driven. The asymmetry of the transfer function is then a measure of the slip rather than an inference from damping. Driving the same structure from both sides at increasing force, and plotting asymmetry against force, puts the microslip threshold where the asymmetry leaves zero.
Saturation
H3122. A tensioned membrane's wrinkle pattern does not disappear as tension rises, it coarsens, and the count saturates at a value set by the ratio of tension to bending stiffness. A membrane at four tensions, with the pattern read optically, shows the count flattening while flatness keeps improving.
Propulsion: feed, combustion and nozzles
Phase lag
H3123. A thruster's specific impulse measured at short pulse widths falls below its steady value, and the deficit is charged to combustion inefficiency at low duty cycle. The actual cause is the chamber pressure's lag behind the valve command, since at these widths the pressure never arrives before the valve closes. The same thruster at two feed pressures and one pulse width should show the deficit scaling with fill time rather than with mixture ratio.
Floor
H3124. A monopropellant thruster's minimum impulse bit has a floor set by the catalyst bed's thermal recovery rather than by the valve, because pulses shorter than the bed's cooling time borrow heat the next pulse must repay, so impulse per pulse stops falling with width. A pulse-width series at fixed total propellant, with bed temperature logged between pulses, exhibits the floor and locates it.
Inversion
H3125. Raising chamber pressure normally raises delivered specific impulse, but in a small pump-fed engine with the turbine exhaust dumped overboard the dump's share grows faster than the chamber's efficiency. Delivered impulse can therefore fall as pressure rises, which is the opposite of the usual expectation. The crossover pressure follows from the dump fraction and the nozzle's area ratio, and a pressure series on one engine with the dump area fixed finds it.
Hidden variable
H3126. Vacuum thrust measured in a finite facility carries a term from the plume's back-flow pressurizing the nozzle's exterior, and that term changes sign when the flow separates because the separated plume expands into a different volume. Two facility pressures at one chamber condition should move measured thrust by an amount that reverses at separation onset, which is the test.
Regime boundary
H3127. The expansion ratio at which a vacuum nozzle separates at sea level is set by the contour's inflection rather than by the usual pressure-ratio rule, so the safe area ratio is a property of the contour family and not of the ambient pressure alone. Two families at one area ratio, fired at one ambient pressure, separate at different chamber pressures if the claim holds.
Overconstraint
H3128. A regeneratively cooled chamber must meet wall temperature, coolant pressure drop and chamber pressure, and once the material is fixed only two geometric variables are free. A design claiming all three at a novel channel geometry is therefore relaxing one of them silently. A channel series in which pressure drop is measured rather than computed shows which requirement gave way.
Similarity failure
H3129. Below a chamber diameter of a few inches, wall heat loss removes enough enthalpy that measured characteristic velocity becomes a function of the chamber's surface-to-volume ratio rather than of the propellant, so small thrusters are not similar to large ones. Three diameters at matched geometry, with heat loss measured calorimetrically, test the similarity rather than the propellant.
Identity
H3130. Acoustic and feed-coupled combustion instability are one feedback loop labelled by which element's lag dominates, not two mechanisms, with the chug and screech names marking where the loop's phase crosses. A single sweep of injected pressure perturbation through the feed line's resonance should trace one continuous stability boundary if this is right.
Artefact mimics effect
H3131. A thrust oscillation at the turbopump's blade-pass frequency is reported as pump forcing, but when it fails to shift with pump speed while shifting with the stand's mass it is the stand resonating under the pump's broadband noise. The pump supplies noise and the stand supplies the resonance, so neither alone explains the line. Two pump speeds on two stand masses separate forcing from resonance in an afternoon.
Combinatorial bound
H3132. A solid motor's thrust curve can have no more inflections than its grain has features, because each inflection is a change in burning area and changes in area are produced by the features. Casting two grains with the features in different axial order shows whether the inflections travel with the features or with the area.
Crossover
H3133. Ignition of a torch igniter is most reliable away from stoichiometric, because a rich kernel starves and a lean one is quenched at the injector's cold wall before it reaches the main chamber. Ignition probability against mixture ratio should be a hump whose peak moves with wall temperature, and two wall temperatures test it.
Hysteresis
H3134. A pintle injector's spray angle differs between increasing and decreasing throttle because the pintle's wake attaches at one of two positions, and the two branches bound a band a throttle profile can enter and not leave. Sweeping throttle up and down while imaging the spray, rather than inferring angle from chamber pressure, maps the band.
Conservation argument
H3135. A resistojet's specific impulse at fixed electrical power is capped by the heat it must reject to the spacecraft rather than by the heater's temperature, since the rejected fraction never reaches the jet. The cap is the power divided by the mass flow and the gas's specific heat, reachable only at a flow the thermal design permits. A power series at fixed flow, with rejection measured by calorimeter, should flatten at the value the cap predicts.
Competing rates
H3136. Cold-start performance of a green monopropellant is set by the race between the bed's thermal diffusion and the propellant's residence time, so past the diffusion length a longer bed buys nothing. Two bed lengths at two flow rates give four starts whose times should collapse on diffusion time rather than residence time.
Non-monotonic
H3137. Thrust coefficient against area ratio is not monotonic in a real nozzle, because divergence loss grows while the pressure-ratio gain saturates, so the maximum sits below the ideal-gas optimum. Four area ratios at one chamber pressure put the maximum where the two terms cross.
Latent variable
H3138. Two engines built to one drawing that differ in measured specific impulse usually differ in injector discharge coefficient rather than in combustion efficiency. The coefficient sets chamber pressure at fixed flow, and pressure sets the expansion, so a small geometric difference is amplified into an impulse difference. Inferring the coefficient from steady-state pressure at fixed flow, for both engines, separates a hardware difference from a combustion one.
Null space
H3139. The geometries giving the same cold-flow spray angle form a large set, so angle cannot specify an injector for combustion performance; angle's null space is the design freedom that matters. Two injectors matched in angle but differing in orifice count should differ in stability while matching in cold flow.
Ceiling
H3140. A regeneratively cooled engine's pressure is capped by the coolant's critical pressure rather than by the wall's temperature limit, because past critical the coolant cannot absorb the heat at the flow the wall allows. A hydrocarbon-cooled and a hydrogen-cooled chamber of one geometry should reach very different ceilings, and a pressure series to each ceiling tests it.
Order dependence
H3141. In a staged-combustion cycle the start sequence decides whether the bootstrap sustains itself, since the first turbine to reach speed sets the other's inlet pressure. The same hardware can start in one order and stall in another. Two sequences on one engine, all else fixed, is the test.
Boundary-condition ownership
H3142. Above some chamber pressure the pressure measured at the nozzle exit belongs to the cell's diffuser rather than to the nozzle, so the altitude performance inferred from it is a facility property. One nozzle with two diffuser geometries, at one chamber condition, should move the inferred exit pressure while thrust stays put.
Sign argument
H3143. A lateral thrust component measured on a gimballed engine reverses when the gimbal's offset is reversed, so a misalignment trend across a lot that fails to reverse belongs to the adapter. A trend that does reverse belongs to the engine. Two engines re-mounted in swapped adapters test it.
Feedback
H3144. Erosive burning is a positive feedback between port velocity and burn rate, so below a port-to-throat area ratio its loop gain falls under one and the effect vanishes rather than weakening. Four area ratios at one pressure should show a threshold, not a gradual onset.
Unobservable
H3145. Catalyst bed degradation cannot be seen in steady-state thrust, because the controller raises pressure to hold thrust and absorbs the loss. It appears only in valve duty cycle at fixed thrust, so a life test run at constant duty rather than constant thrust reveals it.
Existence claim
H3146. A minimum bed length exists below which a monopropellant cannot be decomposed at any preheat, because the flame's quench distance in the channels exceeds the bed. Four lengths at one preheat should show a sharp cutoff rather than a gradual fall.
Thermal protection and entry hardware
Misattribution
H3147. Post-flight recession of a charring ablator is usually attributed to oxidation, but at low surface shear most of the lost mass is mechanically removed weak char. That leaves a density profile the oxidation account does not predict. Sectioning a flown or arc-tested article and reading char density against depth distinguishes them, since mechanical removal truncates the profile while oxidation thins it uniformly.
Bound
H3148. A radiative heat shield's equilibrium temperature at a given flux cannot fall below the value set by its coating's ratio of solar absorptance to infrared emissivity, and every material class has a floor for that ratio near one. A coating series spanning the ratio, measured at one flux in vacuum, should approach the blackbody value from above and never cross it.
Reversal
H3149. Adding a high-emissivity coating to a metallic panel usually lowers its temperature, but when the panel faces a warm structure the coating also raises absorptance of that structure's infrared, and the net can be a higher bondline temperature. Two panels, coated and bare, facing a temperature-controlled plate at two plate temperatures separate the terms by their sign.
Hidden term
H3150. An arc-jet's measured heat flux includes the probe's catalytic correction, and for a fully catalytic surface the value exceeds the flight value by a factor set by the probe's material rather than by the flow. The correction is usually assumed rather than measured. Testing one condition with catalytic and non-catalytic probes of identical geometry gives two fluxes whose ratio is the correction.
Threshold
H3151. There is a surface temperature below which a reusable silica tile's coating does not devitrify and above which its emissivity changes permanently, and that threshold marks the boundary between the tile's reusable and recession regimes. A tile heated in steps, with emittance measured after each step at room temperature, should show a step change rather than a gradual drift.
Scale argument
H3152. Subscale ablator tests cannot reproduce flight recession because char thickness grows with the square root of time while mechanical erosion grows linearly, so a one-third-scale test over-weights erosion by about the square root of the scale factor (estimate). Two scales at matched enthalpy and duration, compared through the recession-to-char-depth ratio, test the scaling rather than the material.
Identity
H3153. Suture tear and coating crack in a flexible thermal blanket are one stress state seen at two temperatures: the crack is the low-temperature form and the tear the high-temperature one, where the coating yields instead of fracturing. A blanket cycled through both ranges under one mechanical load should show the two forms meeting at a crossover temperature.
Conservation audit
H3154. A heat shield's radiative equilibrium in low Earth orbit is not the blackbody balance, because Earth's albedo and infrared supply a term of a few per cent that is usually dropped. The inferred emissivity then absorbs the dropped term. A shield suspended between a cold shroud and a controlled warm plate, at two plate temperatures, separates the environmental term from the material's emissivity.
Hysteresis
H3155. A ceramic tile's first heating produces a larger expansion than later cycles because microcracks close irreversibly, so a thermal-vacuum distortion test on a never-cycled article measures a transient flight will not see. Pre-cycling one article three times and comparing its distortion with a virgin twin's at one gradient isolates the one-time set.
Path dependence
H3156. An ablator's virgin permeability depends on the cure's pressure history, so two lots of equal density can vent pyrolysis gas differently. The difference shows up as a bondline pressure difference rather than as a recession difference, which is why it survives material acceptance. Two lots cured at two pressures, tested at one heating rate with internal pressure taps, should differ in peak pressure while matching in density and in recession.
Null claim
H3157. A spallation-resistant coating should make no difference to recession in a pure-shear environment, because it resists compressive char loading and a pure-shear environment supplies none. Coated and uncoated samples at four shear levels should converge at the high-shear end, and that convergence is the test.
Crossover
H3158. An inflatable decelerator's drag benefit crosses over with its mass penalty at a ballistic coefficient: below it the inflatable wins and above it the rigid aeroshell does, for the same entry mass. Two decelerators sized to one entry mass, tested in one facility at two ballistic coefficients, should reverse their ranking across the crossover.
Two-clock race
H3159. A heat shield's bondline temperature is set by the race between the TPS's diffusion time and the entry's heat pulse, so for pulses shorter than the diffusion time added thickness buys almost nothing. Three thicknesses at one pulse, and one thickness at three pulse lengths, collapse onto a single curve only if the diffusion time is the right variable.
Saturation
H3160. A heat-pipe-cooled leading edge's effective conductivity saturates at the wick's capillary limit, so beyond that flux the tip's temperature rises at the wall material's conductivity. The heat pipe stops helping rather than degrading gracefully. A flux series on one leading edge, with the wick's limit computed rather than assumed, shows the knee and the slope change.
Impossibility
H3161. Mass loss of a carbon-phenolic in a nitrogen arc cannot be oxidation, since there is no oxygen in the flow; the loss is pyrolysis gas mechanically eroding its own char, and the label has been carried over from air tests. Running the same material in nitrogen and in air at matched enthalpy gives two rates whose difference is the oxidative part, and whose ratio should approach one at low shear.
Dimensional argument
H3162. The recession-to-char-depth ratio is the only dimensionless group that should govern an ablator's mechanical loss, so two tests with that ratio matched but durations different must agree on it. A disagreement indicates a time-dependent mechanism the group cannot capture.
No-go claim
H3163. No single-layer TPS can be both the radiative outer surface and the insulating layer above a certain flux, because the outer surface wants high conductivity to spread heat and the insulator wants low conductivity to block it, so a two-layer design is forced rather than chosen. A single-layer article tested to that flux should fail at the bondline before its surface reaches the radiative limit.
Causal ordering
H3164. In a failed tile field the gap filler is usually judged to have failed after the tile because the tile's debris lands on it. The filler's own erosion precedes the tile's loss, since the gap is where hot gas first penetrates. A post-mortem mapping filler erosion depth against tile loss on one panel inverts the apparent order and is the test.
Attitude control and momentum hardware
Misplaced transfer
H3165. Bearing friction is inferred from a wheel's spin-down and charged to the bearing, but at low speed most of the deceleration is the harness's torsional stiffness and the rotor's residual imbalance. A spin-down in two harness configurations at the same vacuum separates them by the difference in decay.
Separability
H3166. A wheel's stored momentum is bounded by its rotor's burst margin and not by the motor's torque, so a larger motor raises slew authority and leaves momentum where it was. Two wheels sharing a rotor and differing in motor, run to one speed limit, show equal momentum and unequal torque.
Non-monotonic
H3167. Raising a magnetorquer's core permeability raises its dipole at low current and lowers the achievable dipole once the core saturates. The ranking of two designs therefore reverses between low-disturbance and high-disturbance orbits, which is a nuisance for a single design review. Driving both over a current range with dipole measured by fluxgate puts the crossing where the core saturates.
Provenance gap
H3168. A star tracker's accuracy is quoted as an instrument property, but over a decade the star catalogue's proper motion contributes a term larger than the tracker's noise for the brightest stars. Part of the quoted number therefore has a provenance outside the box. Comparing the tracker against a radio ephemeris early and late in a long mission separates the two.
Regime boundary
H3169. There is a wheel speed at which bearing lubrication passes from elastohydrodynamic to boundary, and it shows as a change in the disturbance spectrum's slope rather than in its amplitude. A speed sweep at constant temperature, with slope computed in bands, puts the knee where the transition is.
Common-mode
H3170. Wheel disturbance at once per revolution and the structure's response at that frequency are common-mode: the quantity measured on the coupled assembly already contains the structure's amplification, so calling one the disturbance and the other the response double-counts. Measuring the wheel on a rigid mount and again on the flight structure at one speed gives the amplification directly.
Null space
H3171. A four-wheel pyramid's fourth wheel adds no controllable axis; it spans a null space that buys redundancy and not authority, so a pointing budget derived from four wheels is really derived from three. Disabling one wheel and repeating a slew should leave the torque envelope unchanged and halve the redundancy.
Conservation argument
H3172. Magnetic dumping does not dispose of momentum: the torque on the torquer is matched by a torque on the field and appears as a change in the orbit's angular momentum. The change is small but not zero, so the accounting matters more as missions lengthen. Comparing a long magnetically dumped mission's orbit against a drag-only prediction should leave a residual along the dumping direction.
Sign argument
H3173. An apparent pitch misalignment measured on an air bearing reverses with tank fill, because propellant motion adds a delayed torque; a trend that tracks fill is a propellant effect, not a thruster one. Firing at three fill levels in one orientation separates them by the sign of the reversal.
Aliasing
H3174. A wheel tachometer sampled at the controller's rate aliases the bearing's cage frequency into a low-frequency line that has been reported as a structural mode. Changing the sampling rate moves the line and leaves the modes, which is the whole test.
Asymmetry
H3175. A control moment gyroscope's gimbal torque shows hysteresis from its slip ring, and the width of the loop is independent of gimbal rate, which separates it from bearing friction that grows with rate. A rate series with the loop closed at each rate gives a width that either holds or grows.
Non-identifiability
H3176. One slew cannot separate inertia from a constant disturbance torque, because a small inertia error and a small torque give the same angular acceleration profile. The two are degenerate for that measurement. A slew in the opposite direction flips the torque's sign relative to the inertia error and makes the pair identifiable.
Degeneracy
H3177. Two inertia matrices differing only in the product of inertia about the spin axis give identical spin response and different nutation, so a spin test cannot tell them apart. Nutation decay after a small transverse impulse resolves the pair, since their predicted nutation frequencies differ by a few per cent (estimate).
Ceiling
H3178. A reaction wheel's torque falls with speed because the bus voltage less the back-EMF shrinks, so the maximum torque in a specification is a zero-speed value no slew above half speed can use. A torque-versus-speed curve at constant bus voltage draws the line and locates the speed at which torque reaches zero.
Memory claim
H3179. A star tracker's focal plane keeps a bias after a bright object blooms and decays slowly enough to contaminate the next frames. The bias is a function of the previous exposure and not the current one. One bright frame followed by ten dark ones shows the decay and its dependence on the first frame's brightness.
Feedback
H3180. In thermal vacuum a sun sensor's field of view and the blanket's specular reflection close a loop: the blanket reflects the sensor's heater, the sensor responds, the heater changes, and the blanket's shape follows. Replacing the blanket with a diffuse witness surface breaks the loop and should change the apparent sun angle.
Reciprocity
H3181. A thrust stand's thrust and a spacecraft's acceleration are not reciprocal measurements of one quantity, because the stand absorbs plume impingement the spacecraft does not. Two standoffs give two apparent specific impulses whose difference is the impingement term.
Boundary-condition ownership
H3182. A misalignment measured during a static fire belongs to the adapter's thermal soak rather than to the thruster, since the adapter's gradient tilts the mount while the thruster's own gradient is symmetric. The measurement is of the interface, not of the engine. Repeating the fire with the adapter at two temperatures, and the thruster at one, moves the apparent misalignment without touching the hardware.
Separation, deployment and orbital hardware
Load-path misassignment
H3183. Tip-off at separation is charged to the clamp band's release, but the rate follows the adapter's stored strain energy at release. It therefore tracks adapter stiffness rather than band tension, which changes which drawing controls the specification. Two adapters of different stiffness released with one band give tip-off rates that follow the stiffness.
Plateau
H3184. A solar array's deployed first mode cannot exceed what its hinges give at their preload, and preload past the point where the hinge's own structure yields buys nothing, so the mode saturates. A preload series with the mode measured at each level shows the plateau.
Cold-start threshold
H3185. Below a minimum deployment torque a hinge stalls at cold temperature, and the threshold is the grease's yield stress acting at the hinge radius rather than the motor's stall torque. A torque series at two temperatures, with stall observed directly, puts the threshold where the yield stress predicts.
Hidden variable
H3186. A docking mechanism's capture envelope is quoted for zero closing rate, but the effective envelope shrinks by the product of the rate and the mechanism's compliance time. At realistic rates that product is a substantial fraction of the envelope. Capture tests at three rates, with the envelope mapped by probe position, show the shrinkage and its linearity.
Turning point
H3187. Increasing a drag sail's area can slow a tumbling satellite's deorbit, because the sail raises the tumble amplitude as well as the drag and the mean projected area is non-monotonic in sail size. Two sail sizes on one tumbling body, with tumble decay measured optically, separate the terms by their sign.
Limiting-case collapse
H3188. A Whipple shield's hole and its rear-surface spall are one shock process seen at two standoffs, so threshold thickness plotted against standoff should be one curve for both. Two damage modes would give two curves. A standoff series with hole diameter and spall area both measured puts them on one curve or does not.
Similarity failure
H3189. Hypervelocity impact cannot be scaled by velocity alone, because pulse duration scales with projectile diameter while the target's response time does not, so the ratio of the two must be matched. Two scales at matched velocity but unmatched ratio should give different damage morphology.
Count
H3190. A separation system needs at most two independent inhibits, because a third adds failure probability faster than it removes it once the inhibit's own failure rate exceeds the release's. A fault-tree count with the inhibit's rate measured rather than assumed puts the crossover where the tree says it should be.
Cap
H3191. A momentum-exchange tether's gain per pass is capped by its tensile strength at tip speed, so tether performance is a material property and not a design choice. The cap is the strength-to-density ratio times the square of tip speed over the gravity term. Two materials at one tip speed show the ratio of gains that the cap predicts.
Order dependence
H3192. A hold-down and release mechanism's shock depends on the order its bolts release, not only on the total preload, because the first release changes the load path for the rest. Sequential and simultaneous release at one preload give different spectra, and only the simultaneous case matches the stored-energy estimate.
Ground testing and qualification
Artefact mimics effect
H3193. A vibration test's input notching is credited to the article's damping, but the notch tracks the amplifier's current limit, so the shaker's force limit is shaping the input. Two amplifier gains at one notched level move the notch's depth while the response stays put.
Bottleneck bound
H3194. A thermal-vacuum test's fidelity is bounded by the shroud's ability to pump the article's outgassing, so a small shroud's vacuum is a partial pressure of the article's own contaminants. The article sits in its own atmosphere regardless of the gauge reading. Two shroud sizes at one nominal pressure, with a residual gas analyser, show different contaminant pressures and different deposition.
Diffuse-field threshold
H3195. Below a certain level a reverberant chamber's modal overlap is too low for a diffuse field, so the article sees a few modes and the test stops simulating launch. A level series with the Schroeder frequency computed, and response statistics measured at each level, shows where the statistics stop being Gaussian.
Instrument term
H3196. A pyroshock spectrum contains the accelerometer's own resonance and its mount's, so the high-frequency knee belongs to the sensor rather than to the structure. The spectrum is therefore partly a measurement of the accelerometer. Two mounts of different stiffness under one source move the knee and leave the low-frequency spectrum unchanged.
Provenance gap
H3197. A radiated emission attributed to a satellite panel is often the facility's ground plane resonating, which is why it does not move when the article rotates but does move when the plane is resized. Rotating the article and resizing the plane in two steps settles it.
Statistical argument
H3198. Testing a qualification unit and a flight unit at one level demonstrates margin only if the two units' failure distributions coincide, and without the unit-to-unit scatter a test three decibels above acceptance means nothing. Two qualification units tested to failure measure the scatter the margin claim assumes.
Irreversibility
H3199. Environmental test order changes the outcome: vibration before thermal vacuum opens connectors that thermal vacuum alone leaves closed. The reverse order produces fewer failures for the same levels, which is why test order is a variable and not a formality. Two identical units in the two orders at one level give the order's contribution directly.
Energy budget
H3200. The energy a shaker delivers should equal what the article dissipates plus what the fixture dissipates. The residual is the fixture's loss, usually the largest term and almost never measured. Driving the fixture with a dummy mass of known damping and then with the article separates them by subtraction.
Status and plan
100 of 1,000. The forms were tracked while writing, so that a repeated form is visible rather than hidden; sixty-five distinct labels appear across the hundred entries. Every claim here is about hardware or about a test, not about a programme or a budget.
Most of these are cheap to check with equipment a space systems laboratory already has: a shaker and an amplifier, a vacuum chamber with two shrouds, a spin stand, a thrust stand with a movable standoff, a modal hammer. That is the point of writing them this way. None has been tested; a hypothesis list is worth the ideas in it, and a test is worth something else.
If one is worth an experiment, it is H3111 — that a single-stage passive isolator cannot be simultaneously soft enough for wheel-disturbance isolation and stiff enough for launch, so every design claiming both is hiding a launch lock that changes the boundary condition. It needs one isolator, two configurations, and a modal survey across the wheel band.
Incoprea Research · hand-considered hypotheses 3101–3200 · 3 November 2026 · ~/Desktop/Nov3/