Dual-Actuator Calibration Test¶
Test or assessment — instantiates Opponent-Channel Regulation
Exercises the activating and restraining channels alone and together to measure each one's gain, timing, and health before they are trusted in service.
A Dual-Actuator Calibration Test is a bounded, deliberate exercise that drives each side of an opposed pair — first in isolation, then together — and measures what comes back: how much output each unit of input produces (gain), how fast the response arrives and decays (timing), and whether each channel still behaves the way its spec says it should (integrity). Its defining move is isolation before opposition: because two large opposed channels can cancel and look fine in aggregate, the only way to know each one's true characteristic is to test it with its partner held out, then re-test the pair. It is an assessment run at a chosen moment, not a live display and not a running controller — it produces a measured, dated calibration record, not a continuous feed.
Example¶
Before an aircraft is returned to service after maintenance, engineers run a controlled test of its two opposed longitudinal-authority channels: engine thrust (the accelerating side) and the speedbrakes and spoilers (the decelerating side). The Dual-Actuator Calibration Test exercises them the way the archetype demands. First thrust alone across its commanded range, logging how much acceleration each throttle position actually delivers and how long spool-up lags the command. Then the speedbrakes alone, logging drag produced and deployment lag. Only then are the two commanded together through a scripted profile, checking that the measured combined response matches the sum of the isolated ones — if it doesn't, a coupling or a fault is hiding in the interaction.
The output is a calibration sheet: thrust gain and lag, brake gain and lag, and a pass/fail on whether each channel stayed inside its integrity band. That sheet is what lets the controller and the limit-setters be trusted — you cannot tune a push-pull pair you have never separately measured. If the brake's lag came back long, the finding is not "adjust in flight" but "this channel is out of spec; do not release."
How it works¶
- Isolate each channel. Hold one side inert and sweep the other across its input range, recording output. This is the step a live monitor can never take, and it is what makes the numbers trustworthy.
- Fit gain and latency. From the sweep, extract each channel's gain curve and its onset/decay timing — including where it saturates and stops responding linearly.
- Run the joint profile. Command both channels through a scripted co-activation and check that combined output matches the composed isolated characteristics within tolerance; a mismatch is evidence of coupling or fault.
- Score integrity. Compare every measured value against its expected band and issue an explicit pass/fail per channel, so a silently degraded side cannot hide inside an acceptable net result.
Tuning parameters¶
- Sweep coverage — how much of each channel's input range the test exercises. Full-range sweeps catch saturation and end-stop faults but cost time and stress the hardware; narrow sweeps are quick but miss the edges where failures live.
- Isolation fidelity — how completely the partner channel is held out. Cleaner isolation gives truer single-channel numbers but is harder to arrange safely in a coupled system.
- Timing resolution — how finely onset and decay are sampled. Fine sampling exposes latency mismatch but generates far more data to reduce.
- Pass/fail tolerance — how far a measurement may drift before the channel fails. Tight tolerance catches early degradation but flags benign variation; loose tolerance passes marginal channels.
- Test cadence — how often the calibration is repeated. Frequent testing catches drift but consumes downtime; rare testing lets a channel age out of spec between checks.
When it helps, and when it misleads¶
Its strength is that it produces ground truth no aggregate view can: the actual, separately measured characteristic of each channel, which is the prerequisite for setting gains, limits, and timing rules honestly. It is the mechanism that catches a channel which is present but weak, or fast but out of spec, before it is loaded into service.
Its failure mode is that a calibration is a snapshot. It certifies the channels at test time under test conditions and says nothing about how they behave an hour later under real load — the phenomenon of actuator saturation[n1] under sustained real-world demand may never appear in a bounded sweep. The classic misuse is treating a green calibration sheet as ongoing assurance and skipping live monitoring, so a channel that degrades in service passes unnoticed until it fails. The guarding discipline is to date every calibration, re-run it on a cadence matched to how fast the channels drift, and hand the running-health question to a continuous monitor rather than to the test.
How it implements the components¶
channel_gain_parameters— the isolated sweeps measure each channel's gain, sensitivity, latency, and saturation point; producing those numbers is the test's core output.coactivation_timing_rule— the joint profile checks that the two channels' onset and decay line up within the window the pair needs to shape output correctly.independent_channel_integrity_monitor— the per-channel pass/fail against expected bands is an integrity check that certifies each side separately rather than trusting the net.
It measures at a moment and does not watch continuously — the running imbalance_observability_signal belongs to the Excitation–Inhibition Ratio Dashboard — and it certifies rather than reacts; the degraded_mode_or_escape_path invoked when a channel fails in service is the High-Gain Degraded-Mode Trigger.
Related¶
- Instantiates: Opponent-Channel Regulation — supplies the separately measured characteristics the rest of the machinery is tuned against.
- Sibling mechanisms: Excitation–Inhibition Ratio Dashboard · Push–Pull Controller Pair · High-Gain Degraded-Mode Trigger · Complementary Cap-and-Floor Rule
Editorial Notes¶
Form Classification¶
Form family: Experiment, Test & Rehearsal
Rationale: Dual-Actuator Calibration Test operates as a bounded trial, probe, simulation, or rehearsal that generates evidence from performance because it exercises the activating and restraining channels alone and together to measure each one's gain, timing, and health before they are trusted in service.
Independent corroboration: The frozen evidence defines Dual-Actuator Calibration Test as 'Exercises the activating and restraining channels alone and together to measure each one's gain, timing, and health before they are trusted in service', so its operative form is Experiment, Test & Rehearsal.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Engineering & Design
Origin pattern: Single lineage
Present-day reach: Specialized
Rationale: Control engineering cohered calibration sweeps of opposing actuators individually and jointly to estimate gain, timing, saturation, dead zones, and interaction.
Related originating lineages:
- Systems Thinking & Cybernetics — Feedback theory supplied the paired activation-restraint model and stability interpretation.
Review resolution: Engineering design is primary because paired-actuator calibration and joint exercise are control-system verification practices; cybernetics supplies the gain-and-feedback model.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] Actuator saturation is the point at which an actuator can no longer increase its output no matter how large the command — the throttle is wide open, the brake fully deployed. Past saturation a paired relation quietly becomes one-sided control, which is exactly why a calibration test sweeps to the end-stops rather than only near the operating point. ↩