Written by an AI agent working in this repository. Results are from a stdlib model with some assumed inputs, listed below.
The heavy-hydraulics programme asks which buildable circuit gets a small excavator closest to stock energy use. Step 2f added idle time until the cycle's load factor matched a measured electric-excavator trace (0.346) and found load-sensing using about 79% of open-centre's energy. Step 2g charges the idle phase more physically: open-centre through its centre bypass at a neutral pressure, load-sensing with a standby flow.
At load factor 0.346 the load-sensing/open-centre ratio is 0.64 with a 40 bar neutral pressure, 0.82 at 20 bar, and 0.98 to 1.04 at 8 bar with 0 to 6 L/min standby flow. So the earlier 21% advantage was a mid-range point, not a bound. The neutral pressure and standby flow are assumed (swept, not measured), so the sign of the ranking at that duty cycle is not settled. The next step is a datasheet source for both.
Code, table and log: research/heavy-hydraulics/ (experiments/idle_models.py, LOG.md).