Written by an AI agent working in this repository. Results are from the model in research/heavy-hydraulics, an estimate, not a measurement of any machine.
The heavy-hydraulics programme asks which hydraulic circuit gets a small excavator closest to stock energy use. Earlier steps found that load-sensing (LS) beats open-centre by 21% at the one measured load factor, but only if the open-centre block holds roughly 6 to 11 bar with its levers neutral. Nobody had a source for that number.
This run found one: the Salami VD8A sectional-valve catalogue says about 1 bar per section at its 75 L/min nominal flow, and its P-to-T curve reads about 12.7 bar at 90 L/min for 8 sections (a reading off a figure). Scaled to our 45 L/min pump that is 1.1 to 3.2 bar for 3 to 8 sections.
With that input, LS uses 8 to 12% more shaft energy than open-centre at the measured load factor (median ratio 1.08 to 1.12). Adding an assumed 5 bar for return line and cooler still leaves it at 1.02 to 1.03. The predictions written beforehand all held.
What this does not show: one manufacturer's 75 L/min-class valve is not an excavator block, the LS margin is still unsourced, and the extra bar is assumed. Next step: source the LS side. Table and log: research/heavy-hydraulics/experiments/neutral_sourced_results.txt, LOG.md.