ECSS Power — LCL Class Selection (space-systems/ecss/e2020-lcl-class-selection)
Use when the task is the latching-current-limiter class choice of ECSS-E-ST-20-20C clause 5.2.1.1.1 — taking one class out of the standard's class table for a given protected load and demonstrating that the performance figures attached to that class are actually met by the load it is being asked to feed.
Domain quick reference
- An LCL sits between a regulated bus and one load. In normal operation it is a closed switch; when the load draws too much it holds its output inside a limiting current band, and after a trip-off delay it opens and stays open until commanded back on. That is why the class is a table entry and not a free parameter: each class ships with a continuous rating, a limiting band and a trip-off window that were qualified together.
- The four figures a class selection is graded against are the highest steady current the class carries without entering limitation, the lower and upper edges of the limiting current band, and the shortest and longest trip-off delay. All four enter the decision; picking on the continuous rating alone is the common defect.
- Energising a capacitive load is a limitation event by construction. While the limiter is limiting, the output is a current source, so the load capacitance charges linearly and the charge time is C*V/I. The binding case is the slowest charge against the earliest trip: the LOWER edge of the limiting band paired with the SHORTEST trip-off delay. Using the band midpoint or the nominal trip time hides a unit that will refuse to start.
- The UPPER edge of the limiting band is the fault current the harness, the connector pins and the load see for the whole trip-off delay, so the class only protects the hardware downstream when that upper edge sits under the harness rating. A class that carries the load comfortably can still fail this check.
- The smallest adequate class wins. Going up a class raises the held fault current and usually lengthens the trip-off delay, so an oversized limiter delivers more fault energy into the same harness while buying nothing.
- Derating factor, start-up time margin and the low-utilisation advisory floor are declared project policy rather than physical constants; the defaults in the logic module are a starting point a project substitutes its own values into.
Workflow
- Validate the class table: ascending continuous ratings, unique names, a limiting band that is the right way up and that sits above the class's own continuous rating, and a trip-off window that is not inverted. A table that trips at its own rating is a data error, not a conservative entry.
- Validate the load: steady-state current, load capacitance, bus voltage and the rating of the harness being protected. A load with no declared capacitance cannot be assessed for start-up, so it is refused rather than assumed to be resistive.
- Derate each class continuous rating by the project factor and keep the classes whose derated capability covers the steady load.
- For each class, charge the load capacitance at the LOWER limiting current, apply the start-up time margin, and compare against the SHORTEST trip-off delay. A class that cannot finish the charge in time is rejected however well it carries the steady load.
- Compare the UPPER limiting current against the harness rating and reject any class whose fault current the harness cannot survive.
- Take the smallest class that passes all three checks, and report its utilisation, its start-up slack in seconds and its harness slack in amperes so a reviewer sees which check is nearest its edge.
- Raise an advisory, not a finding, when the retained class is heavily oversized for the load; and when no class passes, report every reason each class was rejected rather than a bare refusal.
Pitfalls
- Picking the class on the continuous rating alone. The rating answers only whether the load can be carried; the limiting band and the trip-off window decide whether the load can be started and whether the harness survives a fault, and either can force a different class.
- Charging the load capacitance at the limiting band midpoint or at its upper edge. The limiter is allowed to limit anywhere in the band, so the start-up case is the lower edge; a midpoint calculation reports a start-up that a compliant unit will not deliver.
- Comparing the charge time with the longest trip-off delay. The unit may open at the shortest delay in the window, so the comparison is against that value, with the margin applied to the charge time rather than by relaxing the trip-off figure.
- Treating a bigger class as the safe answer. The upper limiting current and the trip-off delay both grow with the class, so oversizing pushes more fault energy through the same harness and can turn a passing harness check into a failing one.
- Comparing a charge time or a derated capability by bare arithmetic. A case meant to sit exactly on the derated capability or exactly on the trip-off delay can land a few units in the last place the wrong side of the bound; the comparison absorbs that representation error while the derating and the trip-off figure stay as specified.
Behavior contract (gate 3)
The class-table validation, load validation, derating, inrush charge time, harness comparison, per-class assessment and smallest-adequate selection are exercised by the gate 3 contract test: scripts/test_e2020_lcl_class_selection.py against scripts/e2020_lcl_class_selection_logic.py (stdlib unittest, offline). Run: python3 scripts/test_e2020_lcl_class_selection.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
